==> picture [215 x 22] intentionally omitted <==

Single Technology Appraisal

Ixazomib citrate for treating relapsed or refractory multiple myeloma

Committee Papers

==> picture [215 x 22] intentionally omitted <==

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

SINGLE TECHNOLOGY APPRAISAL

Ixazomib citrate for treating relapsed or refractory multiple myeloma [ID807]

Contents:

1. Pre-Meeting Briefing

2. Final Scope and Final Matrix of Consultees and Commentators

3. Company submission from Takeda

4. Clarification letters

• NICE request to the company for clarification on their submission

• Company response to NICE’s request for clarification

5. Patient group, professional group and NHS organisation submission from:

• Myeloma UK

• Royal College of Physicians

6. Expert statements from:

• David Lobb – patient expert, nominated by Myeloma UK

• Dr Karthik Ramasamy – clinical expert, nominated by UK Myeloma Forum

• Professor Kwee Yong - clinical expert, nominated by UK Myeloma Forum

7. Evidence Review Group report prepared by Warwick Evidence

• ERG appendix

• ERG addendum

• ERG erratum

8. Evidence Review Group report – factual accuracy check

Any information supplied to NICE which has been marked as confidential, has been redacted. All personal information has also been redacted.

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

1

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

2

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

3

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

4

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

5

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

6

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

7

Confidential

==> picture [458 x 344] intentionally omitted <==

Proposed positioning of ixazomib in combination with lenalidomide and dexamethasone (IXA+LEN+DEX)

• The company propose that IXA+LEN+DEX will be used as a second line or third line treatment. They cite clinician feedback that the predominant use of ixazomib is expected to be in a third line agent prior to panobinostat and other later line agents.

First line treatment options

• TA228 recommends thalidomide (or bortezomib) in combination with an alkylating agent (melphalan) and a corticosteroid (prednisone) for the first-line treatment of multiple myeloma in people for whom high-dose chemotherapy with stem cell transplantation is considered inappropriate.

• TA311 recommends bortezomib in combination with dexamethasone, or with dexamethasone and thalidomide, for the induction treatment of adults with previously untreated multiple myeloma, who are eligible for high-dose chemotherapy with haematopoietic stem cell transplantation.

Second line treatment options:

• NICE recommends bortezomib monotherapy as an option for treating multiple myeloma at first relapse (TA129). In clinical practice in England, bortezomib is often given in combination with dexamethasone.

• For people who have had first-line treatment with bortezomib (that is, people who have bortezomib as part of induction therapy before ASCT or people who are ineligible for ASCT and cannot have thalidomide), bortezomib re-treatment is not reimbursed under TA129.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

8

Confidential

Bortezomib re-treatment was previously available on the Cancer Drugs Fund, but was de-listed in March 2015. Lenalidomide in combination with dexamethasone was another second line option previously available on the Cancer Drugs Fund (regardless of first-line treatment), but this was de-listed in November 2015. With these treatments no longer available, people in England may be offered cytotoxic chemotherapy (such as melphalan) or

cyclophosphamide in combination with thalidomide and dexamethasone (CTd).

• Note that second line use of lenalidomide in combination with dexamethasone is currently being appraised by NICE (in a part review of TA171); publication data to be confirmed.

Fourth line treatment options:

• NICE recommends panobinostat in combination with bortezomib and dexamethasone as an option for treating multiple myeloma, that is, for 'adult patients with relapsed and/or refractory multiple myeloma who have received at least 2 prior regimens including bortezomib and an immunomodulatory agent‘ (TA380). In clinical practice in England, panobinostat regimens are predominantly used at fourth line.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

8

Confidential

==> picture [458 x 344] intentionally omitted <==

The marketing authorisation for ixazomib is conditional on the company providing the following:

• Final OS results from the China continuation study, a multicentre, placebo-controlled RCT in relapsed/refractory multiple myeloma (by December 2016)

• Primary endpoint PFS results from RCTs in ASCT-eligible and ASCT-ineligible patients with newly diagnosed multiple myeloma (by December2017 and December 2018)

• Descriptive data from a global, prospective, non-interventional, observational study in multiple myeloma (by December 2019)

The company is also obligated to provide updated OS analyses from TMM1 (the third interim analysis and final report) by December 2019

Details can be found on page 148 of the European Public Assessment Report (EPAR): http://www.ema.europa.eu/docs/en_GB/document_library/EPAR__Public_assessment_report/human/003844/WC500217623.pdf

Note that the CHMP initially recommended against granting a marketing authorisation for ixazomib, stating that “The data are currently immature, especially for overall survival which is not yet evaluable. Efficacy data in the overall ITT population from the first and second interim analyses do not provide the statistically compelling evidence expected for an application based on a single pivotal trial. Point estimates for efficacy measures are not sufficiently outstanding in the context of other available treatment options.”

After re-examination of the marketing authorisation application, the CHMP’s overall conclusions were as follows (pages 133-147 of the EPAR):

• “the CHMP considered that the primary endpoint, PFS was met and the result was robust in

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

9

Confidential

the first planned interim analysis … However, updated efficacy data from a second interim analysis representing the most up-to date data, showed a reduced difference in effect between arms. The CHMP acknowledged the applicant’s argumentations that the second interim analysis is only a sensitivity analysis, and should be interpreted as such and concluded that it is difficult not to acknowledge the positive result of a large clinical trial (n=722) in this clinical setting”

• “the results in the heavily pre-treated population (new proposed indication)

• contrast with what is observed in patients treated with one previous line only and there does not seem to be a biological rationale or statistical evidence to exclude a significant effect also in patients with 1 prior line of treatment … the CHMP acknowledged the several mechanisms that could explain the increased sensitivity to ixazomib in the subgroup of patients with at least two prior therapies, compared to one prior therapy however it is not possible confirm them by observations in clinical practice.”

• “recently approved drugs for the treatment of multiple myeloma have shown improvements in median in PFS in the range of 4 to 6 months; therefore the 5.9 months improvement observed in the ITT population is considered clinically relevant.”

• “there is some uncertainty about the magnitude of the treatment effect.”

• “In the pivotal trial, the median OS is not evaluable yet and the data is considered immature in this respect. The efficacy evaluation is primarily based on assessment of progression free survival and requires verification of the effect on overall survival.”

• “The delay in disease progression observed with ixazomib is clinically relevant. Concerning the possible uncertainty about the magnitude of the effect, this uncertainty seems acceptable given the favourable toxicity profile, and considering that ixazomib is the first agent to allow oral triple combination therapy in this patient population, which represents a therapeutic innovation in terms of convenience for patients. Therefore, the benefit risk for ixazomib in combination with lenalidomide and dexamethasone for the treatment of adult patients with multiple myeloma who have received at least one prior therapy is considered positive, albeit the efficacy evidence is not as comprehensive as normally required.”

The CHMP heard from its scientific advisory group that (pages 131-133 of the EPAR):

• “The number of events is considered in line with what is generally expected in the field for trials of this size in this population. Hence the ITT analysis is therefore considered mature and the observed effect (about 6 month difference in median PFS) as clinically relevant. Although subsequent analyses showed slightly less statistical significance, this slight fluctuation is not considered to invalidate the conclusions of the primary analysis, based on the totality of the data.”

• “The OS analysis is not considered sufficiently mature and it is also possible that in view of new agents with impact on OS a clear difference in terms of OS might be difficult to observe in the long term. In multiple myeloma in this

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

9

Confidential

treatment setting, however, PFS is considered a clinically relevant endpoint and the magnitude of the effect quite significant. In addition, the favourable trend in OS in the ITT populations is promising and one can rule out a detrimental effect on OS with reasonable certainty.”

• “If any opportunistic subgroup would need to be selected on the basis of available data, it would possibly be for 3 prior treatments which is the subgroup that stands out most in the Forest plot, while 2 prior lines is not very different to 1 prior line. In conclusion, there does not seem to be a strong rationale or statistical evidence to exclude a significant effect also in patients with 1 prior line of treatment.”

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

9

Confidential

==> picture [458 x 344] intentionally omitted <==

Notes about the comparators in the scope

• The interpretation of ‘people who have had at least 1 therapy’ is that it relates to people who have had only 1 prior therapy (that is, the comparators reflect second line therapy). This was discussed and agreed by NICE at the decision problem meeting.

• The interpretation of ‘people who have had at least 2 therapies’ is that it relates to people who have had only 2 prior therapies (that is, the comparators reflect third line therapy). This reflects the anticipated use of ixazomib in the treatment pathway. NICE heard from clinical experts during the scoping process that it is unlikely that people will have ixazomib (which is given in combination with lenalidomide and dexamethasone) as a fourth line therapy because this would mean re-treating the disease with lenalidomide – the current standard of care third line. Lenalidomide re-treatment is not common practice because it is not clinically effective.

• Exclusion of comparators:

  • Bortezomib re-treatment for multiple myeloma was de-listed from the CDF in March 2015. Commissioning experts and clinical advisors to NICE (and the ERG) agree that bortezomib re-treatment is not currently established practice.

  • Lenalidomide as a second line treatment for multiple myeloma was de-listed from the CDF in November 2015. Commissioning experts and clinical advisors to NICE agree that second-line lenalidomide is not currently established practice.

  • Panobinostat: The exclusion of panobinostat as a comparator in people who have had 2 prior therapies was judged appropriate by the committees who appraised carfilzomib (ID934) and pomalidomide (TA427) in multiple myeloma, based on advice

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

10

Confidential

from clinical experts in at the October 2016 committee meetings that, in clinical practice in England, panobinostat regimens are predominantly used at fourth line. The company for this appraisal of ixazomib submitted market research data from IMS (October/November 2016) which suggested that lenalidomide is the most commonly used treatment at 3[rd] line (69% market share) and that panobinostat has a low market share at both 3[rd] and 4[th] line (7% and 19% respectively).

Further detail can be found in the company submission: pages 15-20

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

10

Confidential

==> picture [458 x 344] intentionally omitted <==

To note:

• 57% of people in TMM1 had received an ASCT prior to study entry (TA311 recommends induction treatment with bortezomib and dexamethasone, with or without thalidomide, before ASCT).

• 69% of people in TMM1 had received bortezomib prior to study entry.

• Preliminary guidance for using LEN+DEX second line (in the part review of TA171) is negative.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

11

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

12

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

13

Confidential

==> picture [458 x 344] intentionally omitted <==

TMM1 is ongoing. There have been 2 interim analyses:

• The first interim analysis (IA1) was after a median ~15 month follow up (30 October 2014).

• The second interim analysis (IA2) was after a median ~23 month follow up (12 July 2015).

A third interim analysis of overall survival is planned for Q2 2017 and the final analysis of overall survival is planned for Q3 2019.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

14

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

15

Confidential

==> picture [458 x 344] intentionally omitted <==

Source: table 41 company submission

The primary endpoint of improving PFS was met at the first interim analysis and this was the final statistical analysis of this endpoint.

• At the 15 month follow-up there was a 26% reduction in risk of progression or death with IXA+LEN+DEX compared with LEN+DEX in the ITT population This difference between treatment arms was statistically significant: hazard ratio 0.74 (95% confidence interval 0.59 to 0.94), p=0.012. The company stated that the ~6 month improvement in median PFS (20.6 compared with 14.7 months) in favour of the IXA+LEN+DEX arm is clinically meaningful.

• A non-inferential PFS analysis was conducted at a median follow up of 23 months with 372 PFS events. The hazard ratio of PFS for IXA+LEN+DEX compared with LEN+DEX was 0.82 (95% confidence interval 0.67 to 1.0) in the ITT population. Estimated median PFS was 20 months in the IXA+LEN+DEX group and 15.9 months in the LEN+DEX group.

• Results from the second interim analysis showed a reduced difference in effect between arms in the ITT population for PFS, response rates and time to progression compared to the first interim analysis. That is, the benefit of IXA+LEN+DEX compared with LEN+DEX appeared to reduce in the second interim analysis.

At both the first and second interim analyses, OS data were not yet mature; the median OS was not reached in either trial arm.

• At the first interim analysis, 107 (22%) of the pre-specified 486 deaths required for the final OS analysis had occurred (51 in the IXA+LEN+DEX arm and 56 in the LEN+DEX arm).

• At the second interim analysis, 171 (35%) of the pre-specified deaths had occurred (81 in the

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

16

Confidential

IXA+LEN+DEX arm and 90 in the LEN+DEX arm).

• The company suggested that the data indicate a non-significant survival benefit in favour of IXA+LEN+DEX

  • First interim analysis: hazard ratio of OS for IXA+LEN+DEX compared with LEN+DEX was 0.900 (95% confidence interval 0.62 to 1.32) in the ITT population. The 18-month survival rates were 83% in the IXA+LEN+DEX arm and 80% in the LEN+DEX arm.

  • Second interim analysis: : hazard ratio of OS for IXA+LEN+DEX compared with LEN+DEX was 0.87 (95% confidence interval 0.64 to 1.18) in the ITT population.

The company used the results from the first interim analysis to inform the efficacy outcomes its base case cost-effectiveness analysis (OS, PFS, ORR). It used the results of the second interim analysis in a scenario analysis. Adverse events in the model were based on data from the second interim analysis.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

16

Confidential

==> picture [458 x 344] intentionally omitted <==

Source: table 41 company submission and response to clarification question A4 (95% CI for overall response rate odds ratio)

To note:

• Post hoc analyses by the company showed a regional difference in PFS in patients from North Asia who were enrolled later into the study and who had a disproportionate effect on the second analysis compared with the first analysis. In the study population excluding patients from North Asia (“non-North Asia”), the median PFS in the ixazomib and placebo regimens was 20.5 and 15.6 months, respectively (HR=0.785)

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

17

Confidential

==> picture [458 x 344] intentionally omitted <==

Source : table 54 company submission.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

18

Confidential

==> picture [458 x 344] intentionally omitted <==

Source : table 55 company submission.

Common adverse events were defined as reported in ≥20% patients in either arm.

Data for ‘Rash SMQ’ based on a standardised MedDRA query (SMQ) pooling 27 preferred terms; data for ‘Rash HLT’ taken from the high-level term (HLT) of Rashes, eruptions and exanthems NEC, per the data on rash reported in the United States prescribing information

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

19

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

20

Confidential

==> picture [458 x 344] intentionally omitted <==

Source : table 52 company submission.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

21

Confidential

==> picture [458 x 344] intentionally omitted <==

The company noted that the NMA results were largely consistent across several scenario analyses and summarised the results of the NMA as follows:

• There is a statistically significant difference between IXA+LEN+DEX and BORT+DEX for OS and BoR.

• The difference between IXA+LEN+DEX and BORT+DEX was not statistically significant for PFS or ORR, but indicated a trend in favour of IXA+LEN+DEX.

• The analysis of treatment discontinuations due to AEs indicated higher discontinuations with IXA+LEN+DEX than with BORT+DEX, but this difference was not statistically significant.

• The credible intervals (CrI) were wider for the ORR, BoR and discontinuations due to AEs networks, indicating a potential higher level of uncertainty in these results.

Table footnotes:

a The NMA results for the comparison between IXA+LEN+DEX and LEN+DEX are not presented because they are nearly identical to the results from TMM1; the results of the NMA for this comparison were informed by only the TMM1 trial, and there is no indirect evidence in any of the networks for each population and outcome. See section 4.10 of the company submission for more detail.

b The hazard ratio for PFS and odds ratio for BoR was obtained from a network considering all doses of bortezomib observed in the literature (most studies consider a dose of 1.0mg/m[2] ); due to a lack of data no network could be formed considering the dose specified in the bortezomib marketing authorisation in the UK (1.3mg/m[2] ).

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

22

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

23

Confidential

==> picture [458 x 344] intentionally omitted <==

The ERG made the following comments on the quality of the company’s clinical evidence and data collection:

• Overall, the quality of the systematic review was deemed to be reasonable.

• The company’s assessment of the risk of bias of the pivotal RCT (TMM1) was generally appropriate.

• The quality of TMM1 was good with a low risk of bias.

• The population in the trial appear to be relevant to those treated in the NHS and the ERG does not have any reason to consider the results of the trial to be significantly biased. However, clinical effectiveness data are characterised by a high degree of immaturity since the benefit of ixazomib on OS cannot yet be determined.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

24

Confidential

==> picture [458 x 344] intentionally omitted <==

The ERG suggest that the benefit in the 2–3 prior therapies subgroup appears to be driven by favourable results in the subgroup of patients who have had 3 prior therapies (that is, fourth line treatment). They noted that, although the confidence intervals of the 3 prior subgroup and the 2 prior subgroup cross each other, there is a consistent trend for the point estimates for the 3 prior subgroup to be better than the point estimates for the 2 prior subgroup. The company has not positioned IXA+LEN+DEX as a fourth line treatment; fourth line comparators were not included in the scope following advice that IXA+LEN+DEX would not be used fourth line.

For more detail, see section 5.3.4.3 of the ERG report and the company's response to clarification question A2.

National Institute for Health and Care Excellence Pre, meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

25

Confidential

==> picture [458 x 344] intentionally omitted <==

To note:

• The company also acknowledged in its submission that the Montefusco study was of poor quality.

• The definitions of TPP and PFS are as follows:

  • TTP results only count progression as an event (death are censored)

  • PFS results count both progressions and deaths as events, whichever occurs first.

• The ERG provided further evidence to support their suggestion that TTP can be considered as a good proxy for PFS: the hazard ratio for TPP was similar to the hazard ratio for PFS in the TMM1 trial of ixazomib (for both the first and second interim analyses) and in the MM-009 and MM-010 trials of lenalidomide.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

26

Confidential

==> picture [458 x 344] intentionally omitted <==

For more information refer to:

• Pages 98-102 of the ERG report (section 4.10.2) which details further evidence supporting the ERG’s consideration that the survival benefit associated with IXA+LEN+DEX should be similar to that of BORT+DEX.

• Pages 108-125 of the ERG report (section 4.11.1) which details the methods and results of the ERG’s exploratory NMA.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

27

Confidential

==> picture [458 x 344] intentionally omitted <==

Section 4.11.2 (pages 126-137) of the ERG report describes further exploratory analysis supporting the findings of the ERG’s NMA.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

28

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

29

Confidential

==> picture [458 x 344] intentionally omitted <==

The company developed a partitioned-survival model based on 3 health states: pre-progression, post-progression and death. Death can occur from either progression-free or progressed disease and death is an absorbing health state. A partitioned-survival model considers estimates for each clinical endpoint separately (that is, overall survival, progression-free survival and time on treatment are modelled independently) and therefore maintains consistency with the published clinical data. As a result, although the marketing authorisation for ixazomib states that treatment should stop when the disease progresses, the model allowed treatment to continue after disease progression (that is, time on treatment could exceed progression-free survival), reflecting the observed trial data.

Assumptions for health states

• Pre-progression : During this stage it is assumed that a patient’s disease is in a stable or responding state, and not actively progressing. Patients in this state are assumed to incur costs associated with treatment, including drug acquisition costs, costs of drug administration, and costs associated with medical management of the condition and the management of adverse events. Given that treatment-related adverse events can result in stopping treatment, some patients in the progression-free health state will not be receiving any treatment. Patients in this health state were further subdivided by their Best Overall Response (BoR), which is treatment-specific, into:

  • VGPR+ (which included stringent complete response, complete response and very good partial response [VGPR])

  • partial response (PR)

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

30

Confidential

• stable disease (SD)

• Post-progression : In this stage, a patient’s disease is assumed to have returned or progressed, following which it is reasonable to assume that in clinical practice patients would still be eligible for active therapy, and therefore patients are assumed to move onto next-line treatment and eventually best supportive care before death. The company assumed that the treatment effect of any subsequent line of therapy was captured in TMM1 OS estimates and therefore efficacy associated with post-progression therapy was not explicitly modelled; subsequent lines of therapy following progression have been considered in this economic evaluation only in terms of their costs. Disease progression was defined as the time from the date of randomisation to the date of first documentation of disease progression based on central laboratory results and International Myeloma Working Group (IMWG) criteria as evaluated by an Independent Review Committee (IRC), or death due to any cause, whichever occurred first.

• Death : This is an absorbing health state; once patients experience death they remain in this health state for the rest of the model time horizon.

These health states relate to clinical benefit and do not account for costs; costs are primarily determined by the Time on Treatment (ToT) curves.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

30

Confidential

==> picture [458 x 344] intentionally omitted <==

• To estimate the distribution of patients across subsequent therapies from the TMM1 trial, the company pooled data from the IXA+LEN+DEX and LEN+DEX arms because no significant differences were found between the trial arms.

• The company’s assumption that people who received bortezomib-, carfilzomib- or pomalidomide-based regimens as subsequent treatment in the TMM1 trial would receive panobinostat in UK clinical practice was based on feedback from clinical experts.

• Of the 176 (24.4%) patients in the TMM1 trial who received subsequent treatment after disease progression:

  • 25 (14%) of these received pomalidomide

  • 99 (56%) received bortezomib

  • 13 (7%) received carfilzomib

• The total patients receiving each subsequent therapy sums to more than the initial 176, this is because some patients go on to receive multiple lines of subsequent therapy.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

31

Confidential

==> picture [458 x 344] intentionally omitted <==

The company provided the following justification for using results from the first interim analysis of TMM1 (15 month follow-up) rather than the second interim analysis (23 month follow-up):

• The first interim analysis is the primary analysis data cut for the PFS endpoint

• Although a later interim data cut is available with more mature OS and ToT data (23 months follow up compared with 15 months), these data are still immature (median OS not reached in either arm)

The company provided the following justification for using 2–3 prior therapy data as a proxy for third line positioning:

• It considered that post hoc analysis of the 2 prior therapies only subgroup would not benefit from prior stratification, meaning likely imbalances in important clinical, patient and diseaserelated factors across the IXA+LEN+DEX and LEN+DEX arms which would confound the interpretation of the results.

• The company noted that, of the 281 TMM1 trial patients who had received 2–3 therapies, 208 (74%) had received 2 prior therapies and 73 (26%) had received 3 prior therapies. The company considered that any efficacy benefits seen in the combined 2 or 3 prior therapies subgroup (used as a proxy for third line positioning) would have been driven primarily by the 2 prior therapies patients.

• The company did not explore the impact on the ICERs of using a post hoc analysis of the 2 prior therapies group, but provided the clinical outcomes for this group (see slide 25), which the ERG incorporated in the model as a scenario analysis.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

32

Confidential

==> picture [458 x 344] intentionally omitted <==

To extrapolate PFS, OS and ToT for the duration of the model, the company explored the applicability of 6 parametric distributions (exponential, log-normal, log-logistic, Gompertz, generalised gamma and Weibull). The fit of each parametric model was assessed using covariate-adjusted data. That is, the company applied covariate adjustment to account for potential imbalances between the 2 treatment arms, because they found that there were several patient risk factors that appeared to be associated with differences in clinical endpoints (including ECOG performance score 2, ISS stage III, primary refractory status, age >65 years, and renal dysfunction – see table 64 of the company submission). A scenario analysis, provided in response to clarification question 11d, explored the impact of using unadjusted estimates.

The distributions selected for the company base case are summarised in the table above. Further detail is provided in section 5.3.3 of the company submission and in response to clarification question B3. Appendix 11 to the company submission presents the company’s exploration of the fit of each parametric model to the covariate-adjusted data. Section 5.2.6 of the ERG report provides additional detail on the company‘s methods, that was not provided in the company submission.

The company concluded that the assumption of proportional hazards was supported for PFS, OS and ToT in the 2+ prior therapies population and for PFS in the 1 prior therapy population. For these outcomes, the company fitted a single parametric model to the LEN+DEX data and estimated relative efficacy for IXA+LEN+DEX compared with LEN+DEX. Proportional hazards did not hold for OS or ToT in the 1 prior therapy population. The company explored the impact on the model results of selecting different parametric curves in section 5.3.5 and 5.8 of its submission.

Note that the hazard ratio for OS is based on the company NMA, in which the ERG identified an

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

33

Confidential

error. The ERG also had concerns about the validity of the PFS network.

Table footnotes

The potential for the PFS curve to cross the OS curve was curtailed by applying the minimum of PFS and OS if PFS was greater than OS at a given time point. The same approach was used to avoid the ToT curve crossing the OS curve. This was apparent in only early model cycles and was adjusted to attain clinical validity.

a The company recognised that the Weibull model had slightly better AIC/BIC scores than the generalised gamma model for PFS in the 1 prior therapy population, but it selected the generalised gamma to maintain consistency between the curve fits across the 2 populations. The company also considered that the visual fit of the modelled generalised gamma IXA+LEN+DEX curves with the observed IXA+LEN+DEX PFS data indicates that this method provides a good fit to the data. Using the Weibull had a negligible impact on the ICER.

b The hazard ratio for PFS was obtained from a network considering all doses of bortezomib observed in the literature; due to a lack of data no network could be formed considering the dose specified in the bortezomib marketing authorisation in the UK.

c Because the proportional hazards assumption was violated for OS in the 1 prior therapy population, the company fit a delayed exponential model to the data which satisfied the proportional hazards assumption. Kaplan-Meier hazards observed from the TMM1 clinical trial were applied for the first 5 months. Note that, although the accelerated failure time (AFT) assumption was satisfied, the company considered that the AFT models (the generalised gamma, Weibull, log-logistic and lognormal functions) resulted in clinically implausible estimates, for example the generalised gamma curve estimates 13.62% and 17.97% of patients are alive after a 25-year time horizon (52% of patients are over the age of 65 at baseline). The company’s base-case method predicted that on the IXA+LEN+DEX arm 25.83% of patients would be alive after 10 years, 13.02% at 15 years and 6.57% at 20-years. The company considered that these estimates are clinically valid because although most patients with relapsed/refractory multiple myeloma have a poor prognosis, it is a heterogeneous disease and therefore a small proportion of patients can experience relatively long survival. The company also noted that people receiving second line therapy have a better prognosis than those patients at third line or later.

d The company submission reported a hazard ratio of 0.89 but confirmed at clarification that this was in favour of LEN+DEX. The inverted hazard ratio for IXA+LEN+DEX compared with LEN+DEX is 1.12.

e The hazard ratio for OS (and hazard ratios for overall response rate) was obtained from a network considering only those studies which used the dose specified in the marketing authorisation for bortezomib in the UK. A scenario analysis considers the hazard ratio from a network considering all studies and all doses observed in the literature [HR: 3.05, 95% CI: 1.78 – 5.22].

f The proportional hazards assumption was violated for time on treatment in the 1 prior therapy population.

g The network meta-analysis did not provide ToT for BORT+DEX, so the company assumed that the ToT for BORT+DEX was equivalent to the LEN+DEX arm (which

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

33

Confidential

was based on TMM1 trial data). However, the summary of product characteristics for BORT+DEX state that the maximum number of treatment cycles is eight. Therefore, the company model assumed that after eight 21-day treatment cycles patients in the BORT+DEX arm no longer receive treatment with BORT+DEX.

h The 95% confidence intervals for the treatment effects and hazard ratios span over 1 for overall survival in the 2 prior therapies population, indicating a nonsignificant difference between IXA+LEN+DEX and LEN-DEX . The company suggest that the extrapolation includes a lot of uncertainty because median survival has not yet been reached in the trial data informing the model.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

33

Confidential

==> picture [458 x 344] intentionally omitted <==

To note:

• In in intention-to-treat population in TMM1, 42.18% and 34.75% of patients remain on treatment with IXA+LEN+DEX and LEN+DEX after 24 treatment cycles. Real-world evidence from the UK, cited by the company, showed that after 24 treatment cycles only 17.59% of patients remained on LEN+DEX treatment.

• The ERG report highlighted that the company's approach to modelling treatment costs using the ToT curves actually underestimates treatment costs because ToT curves were consistently below PFS curves (see slide 48).

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

34

Confidential

==> picture [458 x 344] intentionally omitted <==

Administration costs

• Administration cost was calculated per treatment cycle and then multiplied by the average number of treatment cycles sourced from available literature.

• The company did not include administration costs for bortezomib because, although the marketing authorisation for bortezomib includes intravenous and subcutaneous administration, clinical experts advised NICE during the appraisals of carfilzomib and panobinostat in multiple myeloma it is delivered subcutaneously in clinical practice in England.

• The company included drug costs dosing intensities to capture the impact on costs of potential dose reductions and missed doses. The dose intensity of IXA+LEN+DEX was reported to be 93.10% in the TMM1 clinical trial and for LEN+DEX was reported to be 94.90%. Dose intensity was not reported in the BORT+DEX trial, therefore this was assumed to be 100%.

Concomitant medication

• The company included concomitant medications based on the drugs used in TMM1.

• Due to lack of data, the weekly cost of concomitant medications was assumed equal for all comparators.

• Due to lack of data on concomitant medications associated with post-progression therapies, the cost of concomitant medications was applied to patients in the post-progression health state.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

35

Confidential

Resource-use costs

• Resource costs were applied to all patients on treatment. Patients who moved onto active subsequent therapy continued to receive routine care, and thus incurred the routine care costs. Patients who do not move onto active subsequent therapies were assumed to receive an anti-cancer treatment plan.

• The company assumed the same resource costs for all chemotherapy treatments with the exception of panobinostat, which is a more toxic treatment, and therefore patients receive a transthoracic echocardiogram each treatment cycle.

Hospitalisations

• Captured non-routine health care use in the pre- and post-progression health states.

• Obtained from the patient-level data from the TMM1 using the ITT population (because the company did not find any difference between the trial subgroups in the number of hospitalisations) but considered separately by treatment arm and progression status.

Adverse events

• The proportion of patients whose adverse events were treated, and the proportion treated in a primary care versus secondary care setting, was obtained from the company submission to NICE for TA171. Where no data were available, assumptions were reviewed by a UK clinical expert

End of life

• The cost of end-of-life care was sourced from the PSSRU (2015): £10,670 per decedent.

• The company assumed that 20% of patients receive end of life care (consistent with previous submissions in RRMM - TA338 and TA171), resulting in a one-off cost of £2,134 per person.

Lenalidomide patient access scheme (PAS)

• The lenalidomide PAS is operational when lenalidomide, in combination with dexamethasone, is used to treat multiple myeloma in people who have received 2 or more prior therapies (that is, third line or later). This is linked of the positive recommendations in TA171.

• The lenalidomide PAS can be considered when costing the combination of lenalidomide and dexamethasone with ixazomib, but only for the third line use of ixazomib.

• The company included the lenalidomide PAS in the ixazomib arm at both second line and third line; its inclusion at second line (people with 1 prior therapy) was not appropriate. The ERG removed the lenalidomide PAS from its exploratory analyses in people with 1 prior therapy.

See table 91 of the company submission for the cost breakdown for each health state.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

35

Confidential

==> picture [458 x 344] intentionally omitted <==

To note:

• Response status was based on patients best overall response (BOR) during the study induction period and not necessarily the response observed at time of EQ-5D assessment

• The company obtained the utility decrement for intravenous and subcutaneous treatments (0.025) from 2 previous NICE appraisals in small-cell lung cancer; no information specific to multiple is available, but this decrement was also used in the appraisal of pomalidomide in relapsed/refractory multiple myeloma (TA427).

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

36

Confidential

==> picture [458 x 344] intentionally omitted <==

Source : table 79 company submission

To note:

• VGPR+ includes stringent complete response (sCR), complete response and very good partial response

• The company explains that the utility value for progressed disease is better than for stable disease because of the benefits of subsequent lines of treatment.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

37

Confidential

==> picture [458 x 344] intentionally omitted <==

Sources : company patient access scheme submission page 15 (2 prior therapies) and response to clarification question B3 (1 prior therapy). The ERG-corrected ICER for the 1 prior therapy group is reported in the ERG revised addendum, table 8.

Notes about the cost-effectiveness results in the population who have had 1 prior therapy:

• The results for 1 prior therapy population were corrected in response to clarification question B3 from (£69,565 to £73,333) and the corrected ICER is presented above.

• The company’s base case cost-effectiveness analysis uses the results of the company’s NMA, which contained an error in the OS network for people with 1 prior therapy (see the ERG critique on slides 26-27). The company's base case analysis in people with 1 prior therapy also erroneously incorporated the patient access scheme for LEN+DEX in the intervention arm. The ERG provided an addendum which used the company NMA but corrected the error in the OS network, and removed the LEN+DEX PAS. The results are presented above.

• The company made other corrections to the model in response to clarification questions B3 and B13, but these did not affect the base case results in the submission. Any ICERs presented by the company in their response to clarification incorporated corrections based on questions B3, B13 and B19.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

38

Confidential

==> picture [458 x 344] intentionally omitted <==

Source : figure 2 company PAS submission

The parameters with the greatest impact on model outcomes were:

• the hazard ratio applied to the LEN+DEX data to obtain the overall survival for BORT+DEX

• the coefficients for the adjusted time on treatment (ToT) parametric curve.

The hazard ratio for overall survival directly impacts the incremental life years associated with IXA+LEN+DEX and BORT+DEX, and therefore is a driver of the ICER. Modelled ToT has a significant impact on costs within the model (a larger ToT results in larger costs) and so this is also a driver of the results.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

39

Confidential

==> picture [458 x 344] intentionally omitted <==

Source : figure 2 company PAS submission

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

40

Confidential

==> picture [458 x 344] intentionally omitted <==

To note:

• The company also explored the impact of fitting alternative parametric distributions to extrapolate OS, PFS and ToT. The results were sensitive to the distribution chosen for OS and ToT (with the ICER increasing in most scenarios), but less sensitive to the method of extrapolating PFS.

• In the final scenario analysis the company set the price of ixazomib to £0 to explore the paradox that the cost-effectiveness of the regimen is adversely affected by the incremental costs of lenalidomide in the additional PFS time patients experience with ixazomib, such that it is difficult to demonstrate cost-effectiveness of ixazomib even at zero price.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

41

Confidential

==> picture [458 x 344] intentionally omitted <==

Source : company patient access scheme submission page 17-23 (results of scenario analyses), response to clarification question B3 (base case), and response to clarification question A5 (2[nd] interim analysis from TMM1; not reported in PAS submission)

Note: the company corrected its base case ICER in the 1 prior therapy population in response to clarification question B3 (from £69,565 to £73,333) and the corrected base case ICER is presented above. The company did not correct the results of the scenario analyses. These results are based on the company NMA, which contained an error in the OS network for the 1 prior therapy population (identified by the ERG, see slides 26-27); the results above are not corrected for the error in the NMA.

Additional scenario analyses: The company performed additional scenario analyses in the 2 prior therapies population (that were not relevant to the 1 prior therapy population) relating to the cost of LEN+DEX. In the base case, LEN+DEX was costed in the IXA+LEN+DEX regimen as per standard methods using ToT and UK cost references.

• Exploratory scenario A: only additional LEN+DEX in the IXA+LEN+DEX regimen, over and above what is received in the LEN+DEX regimen, was costed. This scenario captures the additional cost of LEN+DEX required due to the increase in ToT associated with IXA+LEN+DEX. This substantially reduced the ICER for IXA+LEN+DEX to £63,675/QALY gained compared with LEN+DEX.

• Exploratory scenario B: additional LEN+DEX over and above what is received in the LEN+DEX regimen was not costed. This scenario only captures the cost of the LEN+DEX that would be received in current practice anyway. This slightly reduced the ICER for IXA+LEN+DEX to £125,672/QALY gained compared with LEN+DEX.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

42

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

43

Confidential

==> picture [458 x 344] intentionally omitted <==

The company base case model estimates that IXA+LEN+DEX results in an increase in progression-free survival of 0.25 years, and an increase in post-progression survival of 4.29 years, compared with BORT+DEX. The balance of the survival gains is reflected in the balance of QALY gains (that is, most of the QALY gain is achieved in the post-progression health state). The ERG noted that the increase in post-progression survival with IXA+LEN+DEX relative to BORT+DEX is much larger than the progression-free survival benefit, resulting in an overall survival gain of 4.54 years. The company did not comment on this. The ERG believes that there is no plausibility for the considerable increase in post-progression survival given that most people will stop treatment at the time of progression and there is nothing in the mechanism of action of proteasome inhibitors like ixazomib that could explain this observation of continued benefit after progression (that is, altering the course of disease subsequent to progression).

The ERG considered that the results of the comparison with LEN+DEX were more plausible, where the benefits of IXA+LEN+DEX were driven by survival gains in the progression-free health state.

The ERG believe that the company model substantially underestimates the survival benefit with BORT+DEX, which is explained by the error in the company NMA (see slide 27). For more detail, see pages 185–7 of the ERG report.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

44

Confidential

==> picture [458 x 344] intentionally omitted <==

The ERG also identified a cost associated with BORT+DEX which the company had omitted from its cost-effectiveness analysis: the resource cost of outpatient visits to have subcutaneous injections administered.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

45

Confidential

==> picture [458 x 344] intentionally omitted <==

Using data from the first interim analysis of TMM1

The ERG consider that a model based on the second interim analysis from TMM1 (together with a revised NMA based on the second interim analysis) would require an entirely new company submission and critique. For more detail of the ERG’s concerns about the company’s scenario analysis using the second interim data cut from TMM1, see pages 205-6 of the ERG report.

The ERG also noted that, in light of the immature survival data, there is a strong argument for exploring a reduction of the treatment effects after the end of the trial period, in line with the NICE methods guide. This has not been done and is not something that is simple to implement within the company model structure.

Proxy data: 2 prior therapies subgroup

• The ERG’s conclusion that using data from people with 2–3 prior therapies as a proxy for third line positioning is likely to overestimate the benefits of IXA+LEN+DEX is based on analysis of the subgroup results from TMM1 - the benefit of IXA+LEN+DEX in the 2–3 prior therapies subgroup appears to be driven by favourable results in the subgroup of patients who have had 3 prior therapies (see slide 25).

• To provide further support for this conclusion, the ERG compared the Kaplan–Meier curves for the 2 prior therapies subgroup with the curves for the 2–3 prior therapies subgroup, which suggested that the OS and PFS gains in the 2 prior subgroup are less than for the 2–3 prior therapies subgroup (see pages 197–202 of the ERG report).

• The ERG explored the effect on the ICER of using data specific to people with only 2 prior therapies in a scenario analysis. This was based on post hoc analysis of the 2 prior therapies

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

46

Confidential

subgroup of the TMM1 trial (the pre-specified subgroups in the trial were less granular: people with 1 prior therapy and people with 2–3 prior therapies).

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

46

Confidential

==> picture [458 x 344] intentionally omitted <==

The ERG consider that cost of subsequent therapy is underestimated

• The company estimated that 24% of patients receive subsequent treatment after disease progression, based on dividing the number of people who progressed and received further therapy (n=176) by the baseline number of patients in the trial (n=722). The ERG note that the denominator should have been number of patients who had progressed.

• The ERG also noted that the company might have underestimated the cost of concomitant medication because it did not include administration costs for subcutaneous (enoxaparin or nadroparin) or intravenous (zoledronic acid or pamidronic acid) treatments. The ERG noted that the cost of concomitant medication is small relative to the myeloma treatments, and has only a small impact on the ICERs.

The ERG were concerned about the company’s utility estimates for several reasons:

• Applying the disutility for sc injection for the entire 3-week duration of each BORT cycle means that patients lose 4% of their quality of life due to sc injections over their entire treatment period. It appears to the ERG that the company attributes the disutility of sc injections to the hospital visits rather than the sc injection itself. But if this is the case, disutilities should be applied to hospital visits for other treatments as well.

• The company assumes a higher utility value for people with progressed disease than for stable disease. In TA171 and TA338, utilities associated with progressed disease (0.610– 0.640) were lower than the utilities associated with progression-free survival or stable disease (0.650–0.810).

• HRQoL in the PFS state may be overestimated because it is based on a patient’s best overall

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

47

Confidential

response and not their response status at the time of the EQ-5D assessment. The analysis uses duration of response from trial, which measures time from the first documented partial response or better until progression; this will overestimate the duration of best overall response because it will take time to attain best overall response and patients wont remain in best overall response for duration of PFS. The ERG consider that the effect of this is unclear, but suggest that utility values in the PFS state might be overestimated because data are extrapolated well beyond the trial period.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

47

Confidential

==> picture [458 x 344] intentionally omitted <==

a The Weibull curve had the lowest Bayesian information criterion (BIC) value and resulted in a more clinically plausible prediction of the number of people alive at the end of the time horizon. With the Weibull distribution, close to zero patients were alive, whereas the delayed exponential predicted 3% of patients in the IXA+LEN+DEX arm and 5% in the LEN+DEX arm would still be alive at the end of the time horizon.

b The choice of distribution for PFS had a negligible impact the company’s ICERs because the costs in the model were determined by the ToT curves. The ERG suggest that PFS curves may be more appropriate for modelling treatment costs than the ToT curves (as described on this slide), and therefore the model will become more sensitive to method for extrapolating PFS.

c The reason that ToT curves were consistently below PFS curves was because patients in TMM1 who were lost to follow-up or withdrew consent were handled differently in the company’s analyses of PFS and ToT. Patients lost to follow-up/withdrawing from the study were censored for the PFS analysis but counted as events for the ToT analysis. Any patients lost to follow up without having disease progression retain the benefits of treatment but without incurring the costs. This reduces the ToT curves below PFS curves and underestimate the treatment costs for PFS.

d In the subgroup with 1 prior therapy: in the IXA+LEN+DEX arm, the area under the ToT curve is 82% of the area under the PFS curve. In the subgroup with 2 prior therapies: in the IXA+LEN+DEX arm the area under the ToT curve is only 65% of the area under the PFS curve, while in the LEN+DEX arm it is 75%. The ERG considered that this rate of stopping treatment before progression was high, and considered it unreasonable that the ToT to PFS ratio would be much lower in the IXA+LEN+DEX arm than in the LEN+DEX arm (see section 5.3.4.8 ERG report for more details).

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma

48

Confidential

==> picture [458 x 344] intentionally omitted <==

SA## refers to sensitivity analyses conducted by the ERG, which were applied to the ERG’s alternative base case and the ERG’s 2 scenario analyses.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

49

Confidential

==> picture [458 x 344] intentionally omitted <==

SA## refers to sensitivity analyses conducted by the ERG, which were applied to the ERG’s alternative base case and the ERG’s 2 scenario analyses.

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

50

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

51

Confidential

==> picture [458 x 344] intentionally omitted <==

Note that the results in the ERG report and appendix do not include the effect of removing the lenalidomide patient access scheme (PAS) from the intervention arm in the 1 prior therapy group. Please refer to the ERG addendum for the results after the lenalidomide PAS was removed.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

52

Confidential

==> picture [458 x 344] intentionally omitted <==

Sources:

• ERG addendum due to lenalidomide PAS: table 1 (1 prior therapy group)

• ERG appendix: table 7 (2 prior therapies subgroup)

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

53

Confidential

==> picture [458 x 344] intentionally omitted <==

Sources:

• ERG addendum due to lenalidomide PAS: table 3 (1 prior therapy group)

• ERG appendix: table 10 (2 prior therapies subgroup)

The ERG did an additional scenario to compare IXA+LEN+DEX with LEN+DEX in the 1 prior therapy population. In this analysis, including the PAS for ixazomib, LEN+DEX dominated IXA+LEN+DEX (that is, LEN+DEX provided more QALYs at a lower cost than IXA+LEN+DEX). LEN+DEX was listed as a comparator in the NICE scope for people with 1 prior therapy, subject to the ongoing NICE appraisal of lenalidomide (part review of TA171). LEN+DEX is not currently available as a second line treatment option for relapsed/refractory multiple myeloma in England.

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

54

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

55

Confidential

==> picture [458 x 344] intentionally omitted <==

Source: ERG addendum due to lenalidomide PAS: tables 4–7

a In SA01-SA03 the ERG explored the following parametric functions: Exponential, Weibull, Log Normal, Log Logistic, Gompertz, Gamma

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

56

Confidential

==> picture [458 x 344] intentionally omitted <==

Source: ERG appendix: tables 11–14

aIn SA01-SA03 the ERG explored the following parametric functions: Exponential, Weibull, Log Normal, Log Logistic, Gompertz, Gamma

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

57

Confidential

==> picture [458 x 344] intentionally omitted <==

The ERG noted that removing the disutility for subcutaneous injections had a minimal impact on the ICERs.

==> picture [426 x 69] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

58

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

59

Confidential

==> picture [458 x 344] intentionally omitted <==

National Institute for Health and Care Excellence

Pre-meeting briefing – ixazomib with lenalidomide and dexamethasone for relapsed/refractory multiple myeloma Issue date: March 2017

60

Appendix B

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE Single Technology Appraisal

Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma

Final scope

Remit/appraisal objective

To appraise the clinical and cost effectiveness of ixazomib citrate within its marketing authorisation for relapsed or refractory multiple myeloma.

Background

Multiple myeloma is a form of cancer that arises from plasma cells (a type of white blood cell) in the bone marrow. Myeloma cells produce large quantities of an abnormal antibody, known as paraprotein. Unlike normal antibodies, paraprotein has no useful function and lacks the capacity to fight infection. Myeloma cells supress the development of normal blood cells that are responsible for fighting infection (white blood cells), carrying oxygen around the body (red blood cells) and blood clotting (platelets). The term multiple myeloma refers to the presence of more than one site of affected bone at the time of diagnosis. People with multiple myeloma can experience bone pain, bone fractures, tiredness (due to anaemia), infections, hypercalcaemia (too much calcium in the blood) and kidney problems.

In 2013, about 4,700 people were diagnosed with multiple myeloma in England.[1] It is most frequently diagnosed in older people, with 59% of people diagnosed aged 70 years and over.[2] Multiple myeloma is more common in men than in women and the incidence is also reported to be higher in people of African and Caribbean family origin. The 5-year survival rate for adults with multiple myeloma in England and Wales is estimated to be 47%.[3]

The main aims of therapy are to prolong survival and maintain a good quality of life by controlling the disease and relieving symptoms. Initial therapy can include induction treatment with bortezomib (given with dexamethasone, or with dexamethasone and thalidomide) before having chemotherapy and stem cell transplantation (NICE technology appraisal 311). If high-dose chemotherapy with stem cell transplantation is inappropriate, NICE technology appraisal guidance 228 recommends thalidomide (or bortezomib if the person is unable to tolerate or has contraindications to thalidomide) in combination with an alkylating agent and a corticosteroid.

Subsequent therapy is influenced by previous treatment and response to it, duration of remission, comorbidities and patient preference. NICE technology appraisal guidance 129 recommends bortezomib monotherapy as an option for treating progressive multiple myeloma in people who are at first relapse having received 1 prior therapy and who have undergone, or are unsuitable for bone marrow transplantation. NICE technology appraisal guidance 171

National Institute for Health and Care Excellence Final scope for the appraisal of ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma Issue Date: March 2016

Appendix B

recommends lenalidomide in combination with dexamethasone as a treatment option for people with multiple myeloma who have received at least 2 prior therapies. NICE technology appraisal guidance 380 recommends panobinostat in combination with bortezomib and dexamethasone as an option for treating relapsed and/or refractory multiple myeloma in adults who have had at least 2 prior regimens including bortezomib and an immunomodulatory agent. Other subsequent treatment options may include repeating high-dose chemotherapy or chemotherapy with alkylating agents and anthracyclines, thalidomide and corticosteroids.

The technology

Ixazomib citrate (Ninlaro, Takeda UK) is an oral small molecule proteasome inhibitor, which acts by inducing apoptosis via the disruption of proliferative tumour cells.

Ixazomib citrate does not currently have a marketing authorisation in the UK for treating multiple myeloma. A randomised controlled trial compared ixazomib citrate with placebo, both in combination with lenalidomide and dexamethasone, in adults with relapsed or refractory multiple myeloma.

• NICE recommends bortezomib monotherapy as an option for treating multiple myeloma at first relapse. In clinical practice, bortezomib is often given in combination with dexamethasone.

National Institute for Health and Care Excellence Final scope for the appraisal of ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma Issue Date: March 2016

Appendix B

National Institute for Health and Care Excellence Final scope for the appraisal of ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma Issue Date: March 2016

Appendix B

Appendix B

References

1. Cancer research UK, Myeloma incidence statistics [accessed March 2016]

2. Cancer research UK, Myeloma survival statistics [accessed March 2016]

National Institute for Health and Care Excellence Final scope for the appraisal of ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma Issue Date: March 2016

Appendix C

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

Single Technology Appraisal

Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma [ID807]

Final matrix of consultees and commentators

National Institute for Health and Care Excellence

Final matrix for the proposed single technology appraisal of ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma [ID807] Issue date: March 2016

Appendix C

NICE is committed to promoting equality, eliminating unlawful discrimination and fostering good relations between people who share a protected characteristic and those who do share it. Please let us know if we have missed any important organisations from the lists in the matrix, and which organisations we should include that have a particular focus on relevant equality issues.

PTO FOR DEFINITIONS OF CONSULTEES AND COMMENTATORS

National Institute for Health and Care Excellence

Final matrix for the proposed single technology appraisal of ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma [ID807] Issue date: March 2016

Appendix C

Definitions:

Consultees

Organisations that accept an invitation to participate in the appraisal; the company that markets the technology; national professional organisations; national patient organisations; the Department of Health and the Welsh Government and relevant NHS organisations in England.

The company that markets the technology is invited to make an evidence submission, respond to consultations, nominate clinical specialists and has the right to appeal against the Final Appraisal Determination (FAD).

All non-company consultees are invited to submit a statement[1] , respond to consultations, nominate clinical specialists or patient experts and have the right to appeal against the Final Appraisal Determination (FAD).

Commentators

Organisations that engage in the appraisal process but that are not asked to prepare an evidence submission or statement, are able to respond to consultations and they receive the FAD for information only, without right of appeal. These organisations are: companies that market comparator technologies;

Healthcare Improvement Scotland;; other related research groups where appropriate (for example, the Medical Research Council [MRC], National Cancer Research Institute); other groups (for example, the NHS Confederation, NHS Alliance and NHS Commercial Medicines Unit, and the British National Formulary .

All non-company commentators are invited to nominate clinical specialists or patient experts.

1Non-company consultees are invited to submit statements relevant to the group they are representing .

National Institute for Health and Care Excellence

Final matrix for the proposed single technology appraisal of ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma [ID807] Issue date: March 2016

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

Single technology appraisal

Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807

Submission by Takeda UK Ltd

15[th ] December 2016

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 1 of 315

Contents

6. Assessment of factors relevant to the NHS and other parties ............................................................ 278

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 2 of 315

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 3 of 315

List of Tables

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 4 of 315

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 5 of 315

Table 63: Features of the de novo analysis ............................................................................................. 166 Table 64: Base case analysis covariate data ........................................................................................... 168 Table 65: NMA results used in the 1 prior therapies (IXA+LEN+DEX vs. BORT+DEX) comparison .............................................................................................................................. 174 Table 66: Calculation of AE rates for IXA+LEN+DEX and BORT+DEX ................................................... 181 Table 67: TRAEs grade 3/4 ..................................................................................................................... 182 Table 68: Comparison of clinical outcomes with model outcomes ........................................................... 184 Table 69: Comparison of the median OS: PFS ratio with those observed in other MM literature ............. 185 Table 70: Clinician validation ................................................................................................................... 186 Table 71: Utility coefficients for parameters obtained using the EQ-5D from the TMM1 trial .................... 189 Table 72: Response status from the TMM1 trial ...................................................................................... 190 Table 73: Response status by population subgroup ............................................................................... 191 Table 74: Number of hospitalisations by treatment .................................................................................. 192 Table 75: Probability of hospitalisation by treatment ............................................................................... 192 Table 76: HRQoL studies ........................................................................................................................ 196 Table 77: HRQoL utility decrements for each TRAE and mean duration of TRAEs ................................. 200 Table 78: Per cycle utility decrements associated with TRAEs ................................................................ 200 Table 79: Utility values by health state .................................................................................................... 201 Table 80: Cost analyses identified in the economic SLR ......................................................................... 204 Table 81: Unit costs for treatment acquisition .......................................................................................... 219 Table 82: Cost per treatment cycle of chemotherapy (comparator treatments) ........................................ 223 Table 83: Cost per treatment cycle of chemotherapy (subsequent treatments) ....................................... 224 Table 84: Unit costs for IV chemotherapy administration ......................................................................... 226 Table 85: Administration costs ................................................................................................................ 226 Table 86: Resource use in routine care .................................................................................................. 228 Table 87: Rate of hospitalisation for pre- and post-progression ............................................................... 229 Table 88: Hospitalisation unit costs ......................................................................................................... 230 Table 89: Overview of concomitant medications ..................................................................................... 231 Table 90: Post-progression resource use for patients receiving active anti-cancer therapy ..................... 234 Table 91: Cost breakdown for each health state...................................................................................... 235 Table 92: Proportion of AEs actively treated, by treatment setting and unit cost ...................................... 237 Table 93: Weight per-cycle costs of AEs ................................................................................................ 237 Table 94: Base case assumptions ........................................................................................................... 239 Table 95: Base-case results – comparison with BORT+DEX (1 prior therapy) ........................................ 243 Table 96 - Comparison of clinical outcomes with model outcomes for 1 prior therapy ............................. 244 Table 97: Summary of life years by health state ...................................................................................... 248 Table 98: Summary of QALYs by health state ......................................................................................... 248 Table 99: Summary of costs by health state ............................................................................................ 249

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 6 of 315

Table 100: Summary of predicted resource use by category of cost ...................................................... 249 Table 101: Base-case results ................................................................................................................. 250 Table 102 Comparison of clinical outcomes with model outcomes for the 2+ prior lines population ................................................................................................................................ 251 Table 103: Summary of life years by health state ................................................................................... 254 Table 104: Summary of QALYs by health state ...................................................................................... 255 Table 105: Summary of costs by health state ......................................................................................... 255 Table 106: Summary of predicted resource use by category of cost ...................................................... 256 Table 107: OWSA: ten most influential parameters for IXA+LEN+DEX compared with BORT+DEX - 1 prior therapy ................................................................................................... 258 Table 108: Scenario analyses for 1 prior therapy ................................................................................... 259 Table 109: Scenario analysis results - 1 prior therapy ............................................................................ 261 Table 110: OWSA: ten most influential parameters for IXA+LEN+DEX compared with LEN+DEX – 2+ prior lines........................................................................................................ 265 Table 111: Scenario analyses for 2+ prior therapies .............................................................................. 268 Table 112: Scenario analysis results for 2+ prior therapies .................................................................... 269 Table 113: Comparison of this submission efficacy estimates with identified literature ........................... 273 Table 114: Estimated market share with ixazomib – 1 prior line ............................................................. 278 Table 115: Patients in England and Wales eligible for treatment – 1 prior line ....................................... 279 Table 116: Patients in England and Wales eligible for treatment with IXA+LEN+DEX – 1 prior line .......................................................................................................................................... 280 Table 117: Estimated market share with ixazomib – 2 prior lines ........................................................... 281 Table 118: Patients in England and Wales eligible for treatment – 2 prior lines ...................................... 281 Table 119: Patients in England and Wales eligible for treatment with IXA+LEN+DEX – 2 prior lines ......................................................................................................................................... 282

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 7 of 315

Li8st of Figures

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 8 of 315

Figure 25: Leverage versus Deviance Residual Plot: Overall Response Rate: 1+ Prior Therapies: RCT and Observational Studies, combined doses, primary publications Dimopoulos et al., 2010 study omitted ..................................................................................... 119 Figure 26: Network for BoR in the 1+ prior therapies population – RCT studies, combined doses, and primary publications. ............................................................................................. 120 Figure 27: Network for Treatment discontinuation due to AEs in the 1+ prior therapies population – RCT and observational studies, specific doses, and primary publications. .......... 121 Figure 28: Network for PFS in the 2+ prior therapies population – RCT only, dose specific, primary publications ................................................................................................................. 124 Figure 29: Network for OS in the 2+ prior therapies population: RCT only, dose specific: primary publications Data ..................................................................................................................... 126 Figure 30: Network for ORR in the 2+ prior therapies population: RCT only, dose specific, and primary publications ................................................................................................................. 127 Figure 31: PRISMA diagram for economic and cost SLR[170] .................................................................. 143 Figure 32: Three state partitioned-survival model structure ..................................................................... 165 Figure 33: Comparison of fitted covariate adjusted PFS curves (generalised gamma) with unadjusted KM curves for IXA+LEN+DEX and LEN+DEX in the 2+ prior lines subgroup ................................................................................................................................. 169 Figure 34: Comparison of fitted OS curves (Weibull) with unadjusted KM curves for IXA+LEN+DEX and LEN+DEX in the 2+ prior lines subgroup ................................................. 171 Figure 35: Comparison of fitted adjusted ToT curves (generalised gamma) with unadjusted KM curves for IXA+LEN+DEX and LEN+DEX in the 2+ prior lines subgroup ................................. 173 Figure 36: Comparison of the fitted covariate-adjusted PFS curve with unadjusted KM curve for IXA+LEN+DEX in the 1 prior therapy population ..................................................................... 175 Figure 37: Comparison of fitted covariate-adjusted PFS curves for IXA+LEN+DEX and BORT+DEX in the 1 prior therapypopulation ........................................................................... 176 Figure 38: Comparison of the fitted OS curve with unadjusted KM curve for IXA+LEN+DEX in the 1 prior therapy population .................................................................................................. 177 Figure 39: Comparison of fitted OS curves for IXA+LEN+DEX and BORT+DEX in the 1 prior therapy population ................................................................................................................... 178 Figure 40: Comparison of fitted covariate-adjusted ToT curve with unadjusted KM curve for IXA+LEN+DEX in the 1 prior line population ............................................................................ 179 Figure 41: Comparison of fitted adjusted ToT curves for IXA+LEN+DEX and BORT+DEX in the 1 prior therapy population ........................................................................................................ 180 Figure 42: PRISMA diagram for economic and cost SLR[170] .................................................................. 194 Figure 43: Patient distribution over time for patients who have received one prior therapy and are receiving IXA+LEN+DEX ................................................................................................... 245 Figure 44: Patient distribution over time for patients who have received one prior therapy and are receiving BORT+DEX ........................................................................................................ 246 Figure 45: Markov trace for QALYs accrued – IXA+LEN+DEX .............................................................. 247 Figure 46: Markov trace for QALYs accrued – BORT+DEX ................................................................... 247 Figure 47: Patient distribution over time for patients who have received 2+ prior therapies and are receiving IXA+LEN+DEX ................................................................................................... 252

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 9 of 315

Figure 48: Patient distribution over time for patients who have received 2+ prior therapies and are receiving LEN+DEX ........................................................................................................... 252 Figure 49: Markov trace for QALYs accrued – IXA+LEN+DEX .............................................................. 253 Figure 50: Markov trace for QALYs accrued –LEN+DEX ....................................................................... 254 Figure 51 Cost-effectiveness plane from 1,000 PSA iterations for IXA+LEN+DEX compared with BORT+DEX – 1 prior line ........................................................................................................ 257 Figure 52: CEAC for IXA+LEN+DEX compared with BORT+DEX – 1 prior line ..................................... 257 Figure 53: Results of the one-way sensitivity analysis for IXA+LEN+DEX compared with BORT+DEX – 1 prior therapy .................................................................................................. 259 Figure 54: Cost-effectiveness plane from 1,000 PSA iterations for IXA+LEN+DEX compared with LEN+DEX – 2 prior therapies ........................................................................................... 264 Figure 55: Cost-effectiveness acceptability curve for IXA+LEN+DEX compared with LEN+DEX – 2 prior therapies ................................................................................................................... 265 Figure 56: Results of one-way sensitivity analysis – 2+ prior therapies – IXA+LEN+DEX vs LEN+DEX ................................................................................................................................ 267

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 10 of 315

Abbreviations

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 11 of 315

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 12 of 315

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 13 of 315

1. Executive summary

1.1 Statement of decision problem

1.1.1 Remit/appraisal objective

The remit/appraisal objective, as defined in the final NICE scope, is to appraise the clinical effectiveness and cost effectiveness of ixazomib citrate within its marketing authorisation for relapsed or refractory multiple myeloma.

On November 21[st] 2016, the European Commission granted conditional marketing authorisation for Ninlaro[®] (ixazomib), indicated in combination with lenalidomide and dexamethasone for the treatment of adult patients with multiple myeloma who have received at least one prior therapy. Ixazomib is the first oral proteasome inhibitor to become available.

Clinical evidence regarding ixazomib is from the TOURMALINE-MM1 study, which is an ongoing Phase III, double blind randomised controlled trial (RCT), examining the efficacy and safety of ixazomib combined with lenalidomide and dexamethasone (IXA+LEN+DEX) versus placebo + lenalidomide and dexamethasone (LEN+DEX) for the treatment of adult patients with relapsed and/or refractory multiple myeloma (RRMM) who have received 1-3 prior therapies.[1]

In the NICE scope the comparators are defined based on line of therapy (Table 1). The interpretation of ‘patients who have had at least one therapy’ is that this relates to those with one prior therapy only – this has clinical practice relevance as the comparators in the scope are those used second line in RRMM, and this fits into the current treatment pathway as specified in NICE clinical guidelines[2] and in the recent carfilzomib ACD[3] (see Figure 2 in the ACD)This issue was discussed at the NICE/ERG decision problem meeting with agreement that this was the appropriate way to interpret the NICE scope.[3]

In the TOURMALINE-MM1 study, in addition to the analysis of the ITT population (all randomised patients who received 1-3 prior therapies), pre-specified subgroup analyses were conducted based on a number of factors including by line of therapy (i.e. patients who had received one prior line and those who had received 2 or 3 prior lines of therapy). Therefore, in addition to the ITT population, clinical and economic evidence is presented based on these subgroups to reflect the NICE scope. The recent carfilzomib ACD clearly specifies the relevant treatment pathway, with 2[nd] line consisting of patients having received one prior therapy, and 3[rd] line based on having received two prior therapies. Like carfilzomib the positioning of ixazomib in combination with lenalidomide and dexamethasone is that it will mainly be used as a 3[rd] line agent, but with potential use as a second line agent also, and so the relevant comparators are lenalidomide + dexamethasone in 3[rd] line, and bortezomib + dexamethasone in 2[nd] line (see figure 2 in the carfilzomib ACD, and also section 1.1.3 in this submission).[3]

1.1.2 Background to relapsed refractory multiple myeloma

Multiple myeloma (MM) is classified as an orphan (rare) disease, defined in the EU as a prevalence not exceeding 5 in 10,000 people. In 2014, there were approximately 4,652 new cases of MM in England, with 2,462 deaths.[4] It is most frequently diagnosed in older people, with 58% of people diagnosed aged 70 years and over.[4]

MM occurs when a malignant transformation results in a population of clonal plasma cells that reproduce uncontrollably[5] and produce large quantities of one antibody (monoclonal protein or M-protein)[6] . The abundance of M-protein, which eventually needs to be degraded via the ubiquitin-proteasome system within the cell, makes MM a

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 14 of 315

particularly good target for a proteasome inhibitor (PI). MM cells have a higher level of proteasome activity than normal cells, meaning that disruption of protein homeostasis by a proteasome inhibitor such as ixazomib results in apoptosis of MM cells more readily than normal cells.

Although outcomes in MM have improved significantly over the past 15 years following the introduction of the first PI Velcade (bortezomib) and immunomodulatory (IMiD) drugs,[7,8] it remains an incurable progressive disease. MM is characterised by multiple relapses; after a successful initial treatment resulting in stable disease or remission, nearly all patients will eventually relapse and will require further therapy. Such patients are difficult to treat as they tend to have more aggressive disease (and thus are less responsive to treatment) and are heavily pre-treated with more pre-existing toxicities.[9] Ultimately, the prognosis of these patients remain poor,[10] with median overall survival (OS) reducing as patients progress through lines of therapy.[11]

Prolongation of progression-free survival (PFS) and OS remain the ultimate goals of treatment, although preserving patients quality of life is also an important consideration. Due to clonal heterogeneity of myeloma, combination therapy has become the standard-of-care in many situations, with the complementary activity of a PI, an IMiD drug (e.g. lenalidomide) and a corticosteroid (dexamethasone) particularly promising.[12-14] Recent Phase III studies have demonstrated superior efficacy with triplet versus doublet combination regimens based on these agents in the frontline[15,16] and relapsed settings,[17,18] without adding any relevant additional toxicity. Additionally, to further improve long-term outcomes, there has been a shift towards a paradigm of extended treatment.[19-21]

In this submission we present evidence of the clinical- and cost-effectiveness of the all-oral triplet regimen of the PI ixazomib in combination with the IMiD lenalidomide and the corticosteroid dexamethasone (i.e. the IXA+LEN+DEX regimen, also known as IRd). Ixazomib is the first in class oral PI that can provide important efficiencies to the NHS, particularly versus parenteral treatments (i.e. intravenous or subcutaneous) that are currently available and widely used. The all-oral triplet combination of ixazomib (a convenient once-weekly, single pill dose) with lenalidomide and dexamethasone can be taken at home, and therefore has the potential to reduce the treatment burden on the NHS, patients and carers. In addition, ixazomib has a favourable toxicity profile which facilitates patient concordance.

1.1.3 Decision problem and NICE scope

The NICE scope for ixazomib issued in March 2016 (Table 1), has specified the comparators based on line of therapy to reflect clinical practice i.e. for people who have received 1 prior therapy: bortezomib (with or without dexamethasone); bortezomib retreatment (with or without dexamethasone); and lenalidomide with dexamethasone (subject to ongoing NICE appraisal [part review of Technology appraisal 171]); and for people who have received at least 2 therapies: lenalidomide with dexamethasone; and panobinostat with bortezomib and dexamethasone.

Of the comparators in the NICE scope, bortezomib monotherapy is the only current NICE recommended treatment in patients who are at first relapse having received one prior therapy and who have undergone, or are unsuitable for, bone marrow transplantation.[2] Of note, both re-treatment with bortezomib, and lenalidomide for the second-line treatment of MM have recently been removed from the Cancer Drugs Fund (CDF).[22-24] The lenalidomide plus dexamethasone doublet combination is currently being appraised by NICE for adults with MM for whom thalidomide is contraindicated and whose disease has progressed after at least 1 prior treatment with bortezomib. Of relevance to this appraisal, on 11[th] November 2016 NICE issued a negative Appraisal Consultation Document (ACD) for lenalidomide plus dexamethasone in this setting.[25]

For those who have received two or more prior treatments, either lenalidomide in combination with dexamethasone; or panobinostat in combination with bortezomib and dexamethasone are recommended by NICE (note that for

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 15 of 315

panobinostat the NICE recommendation stipulates that the prior regimens must include bortezomib and an IMiD).[2] However, in UK clinical practice, lenalidomide plus dexamethasone is mainly used as 3[rd] line therapy (i.e. after 2 prior therapies), while the panobinostat plus bortezomib plus dexamethasone regimen is predominantly used as 4[th] line therapy (i.e. after 3 prior therapies); this has been confirmed in the recent NICE appraisal of carfilzomib and pomalidomide. Hence, we do not consider panobinostat as a relevant comparator for IXA+LEN+DEX in the 3[rd] line setting.

Overall, we are following the treatment pathway specified in the recent carfilzomib ACD (Figure 2 in the ACD[3] , which specifies bortezomib + dexamethasone as the 2[nd] line placement comparator and lenalidomide + dexamethasone as the 3[rd] line placement comparator. This was accepted by the Appraisal Committee, and the same rationale for positioning and comparators applies for ixazomib + lenalidomide + dexamethasone. Clinical feedback has been the predominant use of ixazomib is expected to be in a 3[rd] line agent prior to panobinostat and other later line agents.

What this demonstrates is that for a disease characterised by multiple relapses and where the natural history is that patients will eventually become treatment refractory, there are currently relatively few NICE recommended treatment options available on the NHS in England. In addition, these treatment options have some significant limitations. IXA+LEN+DEX overcomes many of these limitations, providing patients, clinicians and the NHS with the choice and flexibility of an all-oral treatment option that is simple, efficacious and has a favourable toxicity profile.

In contrast to ixazomib, a significant limitation of treatment with the established PI bortezomib (and of the newer PI carfilzomib which is currently undergoing assessment by NICE[26] ) is the requirement for parenteral administration (intravenous (IV) or subcutaneous injection (sc) for bortezomib; IV injection for carfilzomib). The licensed dosing schedule for bortezomib (either alone, with dexamethasone, or in the combination regimen of panobinostatbortezomib-dexamethasone) requires two visits to the clinic per week for weeks 1 and 2 of a 21-day treatment cycle (i.e. 32 visits to the clinic within an 8 cycle treatment schedule)[27,28] (see Table 7). The need for healthcare professionals to administer bortezomib inevitably increases the burden on NHS resources and costs. In addition, this demanding treatment schedule may adversely affect the patient and/or caregiver financially due to travel costs, time away from work[29] as well as impacting their quality of life (e.g. the unpleasantness of having injections; time spent travelling and at the clinic/hospital, as well as the impact on daily activities). Several reports have shown that many patients with cancer prefer oral to parenteral therapy,[30-34] the reasons for which have included convenience, the place of treatment, a dislike of needles, anxiety over an IV line, feeling less ill on oral therapy and reducing the effort in coping with the disease.[30,31,34,35] Having an all-oral combination is also particularly important for patients who find it difficult to attend hospital appointments (e.g. older/frailer patients, those who live far away from hospital or those who are still in employment).[36,37]

Bortezomib is also associated with a number of adverse events; in particular peripheral neuropathy is a doselimiting toxicity which can potentially result in permanent nerve damage to the extremities, while other common side-effects include fatigue, gastrointestinal effects, and modest cytopenias.[38] In addition, panobinostat is associated with severe diarrhoea and severe and fatal cardiac events, arrhythmias and electrocardiogram changes, 39,40 panobinostat has a “Boxed Warning” in the US, alerting patients and healthcare professionals to these risks. 39 Overall, these adverse events can lead to higher costs (i.e. to manage the adverse event), lower patient quality of life, and premature treatment discontinuation which precludes long-term efficacious treatment. By contrast, ixazomib has a favourable toxicity profile which facilitates patient concordance.

Other than the panobinostat-bortezomib-dexamethasone regimen (mainly used 4[th] line in the UK, as explained above), lenalidomide in combination with dexamethasone (mainly used 3[rd] line in the UK) is currently the only other treatment option recommended by NICE for MM patients who have had two or more prior therapies. The

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 16 of 315

TOURMALINE-MM1 Phase III randomised controlled trial (RCT), described in this submission, provides direct head-to-head evidence of the benefits of a triplet regimen of IXA+LEN+DEX (IRd) compared to a LEN+DEX (Rd) doublet. The benefits of the ixazomib regimen include improved progression-free survival and response rates, without any adverse impact on safety/tolerability or patient-reported quality of life.[1] These benefits were observed consistently across key pre-specified subgroups, including patients who generally have a poor prognosis (i.e. heavily pre-treated patients who have received 2 or 3 lines of prior therapy; patients with advanced stage disease (ISS stage III); and patients with high-risk cytogenetic abnormalities).[41]

Overall, IXA+LEN+DEX (IRd) is a simple, efficacious, well tolerated and convenient treatment regimen. Ixazomib is the first and only oral PI to be approved by the EMA, providing a much-needed, all-oral triplet option that patients can take at home, thus reducing the burden on themselves, carers and the NHS.

The final scope issued by NICE in March 2016 and the decision problem addressed in this submission is shown in Table 1.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 17 of 315

Table 1: The decision problem

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 18 of 315

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 19 of 315

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 20 of 315

1.2 Description of the technology being appraised

Ixazomib is the first oral PI to be approved in multiple myeloma. A description of ixazomib is shown in Table 2.

Table 2: Technology being appraised

1.3 Summary of the clinical effectiveness analysis

1.3.1 Identification of studies

A systematic literature review was undertaken to evaluate the relative efficacy and safety of IXA+LEN+DEX versus other selected regimens for the treatment of RRMM.

One randomised clinical trial (RCT) of IXA+LEN+DEX in adult patients with RRMM was identified (TOURMALINE MM-1).[1,42] Evidence versus the other relevant comparators in the NICE scope (i.e. bortezomib with dexamethasone) is derived from a network meta-analysis (NMA) using studies identified in the systematic literature review.

1.3.2 TOURMALINE-MM1 study design

Clinical evidence regarding ixazomib is from the TOURMALINE-MM1 study, which is an ongoing Phase III, double blind RCT, examining the efficacy and safety of IXA+LEN+DEX versus LEN+DEX for the treatment of adult patients with RRMM who have received 1-3 prior therapies.[1] The double-blind design is a particular strength of the TOURMALINE-MM1 study and one that differentiates it from other trials of new agents for RRMM, most of which have an open label design.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 21 of 315

Patients were randomly assigned in a 1:1 ratio to receive either oral ixazomib 4 mg or matching placebo capsule on days 1, 8, and 15, plus oral lenalidomide 25 mg on days 1-21 and oral dexamethasone 40 mg on days 1, 8, 15, and 22, in 28-day cycles. Randomisation was stratified by number of prior therapies (1 vs. 2 or 3), previous proteasome inhibitor exposure (naïve vs. exposed), and International Staging System (ISS) disease stage (I or II vs. III).[1] Treatment continued until disease progression or unacceptable toxicity.

In addition to the outcomes from the entire ITT population (i.e. all patients who were randomised [n=722]), the study also analysed outcomes in pre-specified subgroups according to baseline demographics and disease characteristics, including patients who had received 1 prior therapy (n=425) and who had received 2 or 3 prior lines (n=297). Therefore, the study reports evidence in the patient populations that are of interest for the NICE scope.

The outcomes of interest in the NICE scope (Table 1) that were assessed in this study i.e. progression free survival, overall survival, response rates, adverse effects of treatment and health-related quality of life are presented. However, because the TOURMALINE-MM1 study does not collect data on time to next treatment, the following parameters that may be useful surrogates have been reported: time to progression; time to response; and duration of response.

To date there have been two interim analyses. At the data cut-off for the first interim analysis (30 October 2014), median follow-up was 14.8 months and 14.6 months in the IXA+LEN+DEX and LEN+DEX groups, respectively. As the primary endpoint of improved PFS was met at the first analysis, this was the final statistical analysis of the primary endpoint (primary analysis). Data from the second pre-planned interim analysis (cut-off date: 12[th] July 2015, median follow-up of ~23 months) is also presented – a non-inferential PFS analysis was also conducted at the second interim analysis. Adverse events are reported from the longer duration second interim analysis.

1.3.3 Clinical effectiveness of IXA+LEN+DEX vs. LEN+DEX

The TOURMALINE-MM1 study has shown that the addition of IXA to LEN+DEX significantly improves outcomes including PFS and response rates, with limited additional toxicity.[1]

A statistically significant PFS improvement was demonstrated with IXA+LEN+DEX vs LEN+DEX in patients with RRMM (ITT population at the primary analysis of PFS: median, 20.6 months vs. 14.7 months; hazard ratio (HR) 0.74, p=0.012; median follow-up ~15 months), showing a clinically meaningful ~6 month improvement in median PFS. In addition, median PFS in the 2 or 3 prior therapies stratified subgroup was not estimable (NE) for IXA+LEN+DEX versus 12.9 months for LEN + DEX (HR 0.58; p=0.0033). PFS benefit was also observed consistently across other key pre-specified subgroups, including in poor-prognosis subgroups such as elderly patients, patients with advanced stage disease, and patients with high-risk cytogenetic abnormalities (for whom the LEN+DEX doublet is emerging as sub-optimal treatment)[41]

With median follow-up of ~23 months, overall survival was not yet mature. Follow-up is ongoing, with OS data from the third interim analysis (IA3) expected in Q2 2017 and final OS analysis expected in Q3 2019.

Overall response rates (ORR) in the ITT population were 78.3% in the IXA+LEN+DEX group and 71.5% in the LEN+DEX group (p=0.04). Responses were rapid and durable and deepening responses were noted with increasing treatment duration; median time to response was 1.1 months vs. 1.9 months, and median duration of

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 22 of 315

response was 20.5 months vs. 15.0 months, in the IXA+LEN+DEX and LEN+DEX arms, respectively. ORRs were also 80.4% vs. 67.1% in patients who had received 2 or 3 prior lines of therapy, respectively.

Reflecting the findings of the primary endpoint of PFS, time to progression (TTP) was also significantly longer in the IXA+LEN+DEX vs. LEN+DEX arm (ITT population: median 21.4 months vs. 15.7 months: HR: 0.71, p=0.007). In patients who had received 2 or 3 prior lines of therapy, the median TTP was not estimable vs. 12.9 months at the first interim analysis (median follow-up of 15 months) and was 28.8 months vs. 14.1 months at the second interim analysis (median follow-up of 23 months).

There was no adverse impact on patient-reported quality of life (EORTC-QLQ-C30 and MY-20 questionnaires) from the addition of IXA to LEN+DEX in this double-blind study. This is particularly noteworthy given the tendency to overestimate quality of life benefit in open-label studies.

1.3.4 Safety and tolerability of IXA+LEN+DEX vs. LEN+DEX[1]

Ixazomib triple therapy had a manageable tolerability profile. The frequencies of serious adverse events (47% vs. 49%), discontinuations due to adverse events (17% vs. 14%) and on-study deaths (4% vs. 6%) were similar in both the IXA+LEN+DEX and LEN+DEX groups. Overall, 74% and 69% of patients experienced grade ≥3 adverse events. The only grade ≥3 adverse event for which there was a ≥5% difference between the IXA+LEN+DEX and LEN+DEX groups was thrombocytopenia, a known side effect of bortezomib and carfilzomib[43,44] and for which there were no apparent clinical sequelae. Addition of IXA to LEN+DEX resulted in a slightly increased rate of peripheral neuropathy (27% vs. 22%), with only 2% being grade 3 events (compared with 6% with subcutaneous bortezomib[45] and 3% with carfilzomib in other clinical trials).[18]

Duration of therapy with the IXA+LEN+DEX regimen was prolonged, with half of the patients having received at least 18 cycles at the 23-month analysis. Treatment compliance appeared high and similar between the groups, consistent with the observed tolerability of ixazomib, and suggesting that the all-oral ixazomib triplet regimen was as simple and convenient for patients to take as the lenalidomide-dexamethasone doublet regimen.

The safety data was consistent across the subgroups by prior line of therapy and the overall safety population; despite having received more prior therapy, patients in the subgroup who received 2 or 3 prior therapies did not experience more AEs than patients with 1 prior therapy.

Overall, the favourable tolerability profile of IXA+LEN+DEX is important in RRMM patients who are typically older and less fit. Taken together with its efficacy and convenient oral dosing schedule, the favourable tolerability profile of ixazomib represents a therapeutic innovation and offers a significant benefit for patients with RRMM.

1.3.5 Network meta-analysis: Clinical effectiveness of IXA+LEN+DEX vs BORT+DEX in 1 prior therapy

The results from the NMA show the relative efficacy benefits of IXA+LEN+DEX vs BORT+DEX, the main comparator for a positioning of one prior line of therapy:

• For PFS, a hazard ratio of 0.72 (95% credible interval (CrI) 0.41, 1.19) was estimated, showing a numerical benefit for IXA+LEN+DEX

• For OS, there was a significant benefit estimated for IXA+LEN+DEX with a HR of 0.31 (95%CrI:0.13, 0.65)

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 23 of 315

• For ORR an odds ratio of 0.88 (95%CrI: 0.35, 1.88) was estimated, showing a numerical benefit for LEN+DEX, although not significant and in a scenario analysis was favourable for IXA+LEN+DEX.

• For BoR there was a significant benefit for IXA+LEN+DEX with an OR of 3.82 (95%CrI: 1.32, 8.93)

• Treatment discontinuations due to AEs (safety measure) showed a difference vs BORT+DEX, which was numerically worse for IXA+LEN+DEX but which was not statistically significant (OR= 2.58, 95%CrI: 0.81, 6.32).

There are some limitations to the NMA to support the relative clinical effectiveness of IXA+LEN+DEX vs BORT+DEX in RRMM patients who have received one prior therapy, in particular there is not specific comparator data published or available with which to form a network for this comparison, hence 1+ prior therapy evidence had to be used as a proxy. However, the advantage of this data is that is relatively robust as it is based on a larger dataset and can be considered generalisable for a specific 1 prior treatment RRMM patient population.

Other limits of the NMA were that only fixed effects modelling was possible for each of the networks, and there were limited studies with the approved doses for the comparator. Therefore, for some outcomes (such as PFS), dose specific studies had to be combined with studies with other doses or where none were specified. In general there were no heterogeneity issues in the networks, with the exception of the ORR network. In the base case, evidence networks were based on all evidence, both RCT and observational, and used specific approved doses and evidence from study primary publications. However, in general scenario analysis demonstrated low sensitivity in the HRs and ORs to using only RCT evidence, combining all dose studies, or using later datacuts/evidence (with the exception of the ORR analysis).

Although there is head to head data for IXA+LEN+DEX vs. LEN+DEX from the TOURMALINE MM-1 study, this comparison is still included in the NMA; however, because each patient population and outcome is only really informed by this study, the results from the NMA do not differ from those in this study as reported Section 4.

1.4 Summary of the cost-effectiveness analysis

The cost-effectiveness of IXA+LEN+DEX in RRMM has been evaluated compared with BORT+DEX for 2[nd] line use, and compared with LEN+DEX for 3[rd] line use. The ICERs presented in this submission are above the conventional thresholds of cost-effectiveness adopted by NICE, but are highly uncertain due primarily to the immaturity of the OS and ToT data. There is reason to believe that the ICERs could be much more favourable for IXA, but the economic argument requires further trial follow-up and real world data collection to establish this (see Section 5.11). Clinical opinion received has indicated that the most feasible positioning and area of greatest unmet need is for the use of IXA+LEN+DEX as a 3[rd] line regimen.

The economic model used a partitioned survival model approach based upon progression status and survival to model movement between three independent health states: pre-progression, post-progression and death. Time on treatment was modelled independently and could surpass progression, in line with observed data from the pivotal TOURMALINE-MM1 clinical trial. The comparison of IXA+LEN+DEX with BORT+DEX at a second line positioning used data from the TOURMALINE-MM1 clinical trial for IXA+LEN+DEX and data obtained from an NMA for BORT+DEX relative to LEN+DEX. Due to scarcity of data for a specific one prior therapy population for the comparator from the NMA, hazard/odds ratios for OS, PFS and ORR were obtained from a 1+ prior therapies population instead and assumed generalisable. The comparison of IXA+LEN+DEX with LEN+DEX at a third line positioning used data directly from the TOURMALINE-MM1 clinical trial, with the outcomes from the 2+ prior therapies population used as representative for the outcomes associated with this positioning. This consisted of patients who had received either 2 or 3 prior treatments, and combined represented a pre-specified sub-group in the TOURMALINE MM1 study, hence was robust evidence to use in the model.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 24 of 315

Health related quality of life was captured within the economic model using a regression equation fit to the patient level data from the TOURMALINE-MM1 (TMM1) clinical trial. Explanatory variables included in the regression equation were: response, hospitalisation, adverse events grade 3 or 4 (including new primary malignancies) and whether a patient was in Grade 3 or 4 (including new primary malignancies) and whether a patient was in the final stages of life (≤3 months to death). Therefore, disutilities associated with adverse events were incorporated in the regression equation. To avoid double counting, hospitalisations associated with adverse events were not included. Costs and resource use were included for treatment, administration, adverse events, concomitant mediations, hospitalisations, post-progression therapies and terminal care costs. Estimates of resource use were sourced from the International Myeloma Foundation Patient Handbook or clinical advice. All costs were costed using UK specific sources.

In line with NICE process, results are presented (Table 3 and Table 4) including the PAS for lenalidomide (treatment beyond 26 treatment cycles is refunded). Full incremental analysis was not presented for IXA+LEN+DEX, BORT+DEX and LEN+DEX as the comparators for IXA+LEN+DEX differ depending on positioning (2[nd] or 3[rd] line).

Table 3: Base case results, IXA+LEN+DEX compared with BORT+DEX at a second line positioning (without PAS – see separate PAS template for with PAS results)

Table 4: Base case results, IXA+LEN+DEX compared with LEN+DEX at a third line positioning (without PAS – see separate PAS template for with PAS results)

Treatment with IXA+LEN+DEX is shown to increase the life years and QALYs compared with current UK treatment in both the 1 prior therapy and 2+ prior therapies populations (3.49 and 2.34 for the 1 prior therapy population and 1.39 and 0.97 for the 2+ prior therapies population). Probabilistic analysis found that IXA+LEN+DEX was associated with a QALY gain in 99.0% and 96.6% of the simulations in the 1 prior therapy and 2+ prior therapies populations.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 25 of 315

However, clinically effective and relatively safe treatments often result in patients continuing on treatment for longer. This is shown in the economic analysis as IXA+LEN+DEX is shown to accrue additional costs from treatment, resource use whilst on treatment and TRAEs. Furthermore, increases in survival were associated with increased post-progression costs (+£140 in the 1 prior therapy population and +£10,682 in the 2+ prior therapies population).

Sensitivity and scenario analyses demonstrated that the results were sensitive to key parameters, with the main areas of uncertainty associated with:

• Estimates of relative survival between IXA+LEN+DEX and BORT+DEX in the 1 prior therapies population and between IXA+LEN+DEX and LEN+DEX in the 2+ prior therapies population.

• Absolute and relative duration of treatment estimates for all comparators

The results from this submission provide the first economic analysis using the initial data set from the randomised controlled trial: TMM1. These data show an improvement in progression free survival. However, due to the immaturity of the survival data the impact on overall survival is inconclusive. Ixazomib is the first and currently the only approved oral proteasome inhibitor, offering a convenient, fixed dose, once-weekly, single capsule administration with benefits for patients, carers and the NHS. Ixazomib also has a favourable safety profile compared to some other new RRMM therapies. The impact of AEs on costs and QALYs has been included in the economic analysis. However, the full benefits to clinicians and patients of a favourable AE profile are hard to fully capture within the QALY framework. Alongside being an oral treatment, the favourable AE profile is an important attribute of ixazomib when compared to other therapies that are being considered for a 3[rd] line positioning in the treatment pathway such as carfilzomib (which has a resource intensive IV administration schedule and is associated with rare but serious CV adverse events.)

In performing the economic analysis, a paradox has been identified due to the combination of ixazomib with lenalidomide (+DEX). The high cost of the latter therapy means that only with a very low (or almost zero) price of ixazomib is it possible for ixazomib with lenalidomide (+DEX) to meet conventional thresholds of cost-effectiveness; a cost of £6 per capsule and £175 per capsule to achieve a £30,000 WTP threshold for the 1 prior therapy and the 2+ prior therapies populations respectivly. This is a paradox that needs to be considered in this appraisal when interpreting the cost-effectiveness of ixazomib.

Within the 2+ prior therapies population analysis, another scenario considered only costing the additional LEN+DEX in the IXA+LEN+DEX regimen, over and above what was received in the LEN+DEX regimen; this reduced the ICER from …………to …………. Whilst exploratory for the 2+ prior line population this ICER is potentially more representative of the cost-effectiveness of IXA+LEN+DEX compared to current UK standard of care, as without the introduction of IXA to the UK market LEN+DEX would still be administered. Therefore, tt should be considered that the consequences of introducing IXA+LEN+DEX should only include the costs associated with IXA and the additional LEN+DEX for a fair assessment of the cost-effectiveness of IXA+LEN+DEX vs LEN+DEX.

The current base case ICERs (presented in this submission without a PAS) are not sufficient for IXA+LEN+DEX to be considered as a cost-effective use of resources based on conventional cost-effectiveness thresholds used by NICE. However, we believe the high uncertainty and the potential for improved cost-effectiveness makes IXA a candidate for funding via the CDF alongside a data collection plan (see Section 1.5 and Section 5.11.1).

A PAS consisting of a simple price discount on the list price of ixazomib has been submitted to PASLU, and the results are presented with PAS in the separate PAS template. This PAS is designed to give a price of ixazomib similar to other treatments routinely used in clinical practice for RRMM. The PAS improves the cost-effectiveness of IXA+LEN+DEX vs. the comparators considered.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 26 of 315

1.4.1 Incremental cost-effectiveness results

Incremental analysis was not applicable as there was only one comparator for each of the two patient populations of interest.

1.5 Consideration for Cancer Drugs Fund (CDF)

As of July 2016, NICE is able to recommend a medicine for use within the CDF when it considers there to be plausible potential for the medicine to satisfy the criteria for routine commissioning but there remains uncertainty surrounding the clinical data. Takeda believes ixazomib is a potentially suitable candidate to be recommended for inclusion within the CDF for two years (up to Q3 2019), as this would enable the collection of key clinical data such as more mature overall survival and time on treatment data which would better inform the cost-effectiveness of the IXA+LEN+DEX regimen (see also section 5.11.1).

The TOURMALINE MM-1 study was designed with three sequential interim analyses plus a final analysis for OS. At the first interim analysis (~15 months median follow up), the PFS results crossed the pre-specified O’Brien-Fleming boundary demonstrating a statistically significant benefit of the IXA+LEN+DEX regimen versus LEN+DEX (HR 0.74, 95% CI 0.59-0.94, p=0.012). As per the statistical plan, this was the final statistical analysis of PFS and the study has continued in a double-blind manner in order to obtain more mature OS data.

At time of this NICE submission, there have been two interim analyses (IA1 and IA2 at ~15 months and ~23 months follow up, respectively). At IA1, only 107 (22%) of the pre-specified 486 deaths required for the final OS analysis had occurred (51 in the IXA+LEN+DEX arm and 56 in the LEN+DEX arm). At the second interim analysis, there were 171 deaths (81 in the IXA+LEN+DEX group, 90 in the LEN+DEX group), which represents only 35% of the pre-specified number of deaths required for final OS analysis. Both interim analyses have indicated a trend for an OS advantage with IXA+LEN+DEX versus LEN-DEX. This trend is present within both the ITT population (HRs 0.9 and 0.87) and the stratified 2-3 prior line subgroup (HRs 0.62 and 0.65; see Table 5). Therefore, due to the short follow up and limited number of deaths, OS results are immature and likely underestimate the true OS benefit with IXA+LEN+DEX. The third interim and final OS analyses are due in Q2 2017 and Q3 2019, respectively. With this additional follow up, a statistically significant OS benefit may emerge.

In addition, more mature data would provide increased certainty related to duration of treatment. This would better inform time on treatment, as the current short follow up may result in an overestimation in the modelled duration of time on treatment. Inclusion of ixazomib on the CDF would therefore allow RRMM patients to benefit from interim access to the all oral triplet IXA+LEN+DEX regimen, while OS and time on treatment results mature within the TOURMALINE MM-1 study.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 27 of 315

Table 5: Summary of OS Results at the First and Second Interim Analyses

In addition to more mature OS results and more data on time on treatment from the TOURMALINE MM1 pivotal trial, inclusion in the CDF would allow real-world evidence to be collected on the effectiveness and safety of ixazomib in the UK setting. Since January 2016, ixazomib (for use in combination with LEN and DEX) has been available through a global compassionate use, named patient program (NPP). As of December 2016, a significant number of UK patients have been enrolled into the NPP and have received treatment with IXA+LEN+DEX. Inclusion in the NPP has been restricted to RRMM patients who fulfil the inclusion criteria for TOURMALINE-MM1 and therefore the licenced indication for ixazomib (see Section 2.2.5 for further details). At a recent advisory board, UK clinicians with experience of managing patients on IXA+LEN+DEX within the NPP have confirmed that it would be feasible to retrospectively collect data for PFS, OS, time to next treatment, dose reductions and toxicities. Initiation of a retrospective, non-interventional study using data collected from the NPP could be used to confirm the efficacy and safety of IXA+LEN+DEX in routine UK clinical practice in a population that is reflective of the reimbursed population. However, with a current maximum of only 1 year of follow up, efficacy results will currently be immature and so further, prospective follow-up will be required.

In conclusion, ixazomib is a suitable candidate to be considered for the CDF as the additional two years would allow for data collection to increase the certainty of key clinical parameters with the prospect of more robust overall survival benefit estimates and clarity on the usual duration of treatment that would be expected to lead to improved ICER estimates for IXA+LEN+DEX (see also Section 5.11.1) while allowing timely access to a novel, effective therapy for patients with this incurable disease.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 28 of 315

2. The technology

2.1 Description of the technology

2.1.1 Name and therapeutic class

Brand name: Ninlaro®

Approved name: Ixazomib (formulated as ixazomib citrate)

Takeda research compound name: MLN9708

Chemical name: 1,3,2-dioxaborolane-4,4-diacetic acid, 2-[(1R)-1-[[2-[(2,5-dichlorobenzoyl)amino]acetyl]amino]-3methylbutyl]-5-oxo-

Therapeutic class: Other Antineoplastic Agents, ATC code: L01XX50

2.1.2 Mechanism of action

Ixazomib is a small-molecule, proteasome inhibitor (PI) that reversibly inhibits the 20S proteasome core of the 26S proteasome complex. The ubiquitin-proteasome system (UPS) is the major regulatory system through which protein homeostasis occurs and represents the primary mechanism by which cells degrade proteins, including those involved in growth control, cell cycle regulation, and apoptosis. The 26S proteasome is composed of a catalytic proteolytic core (20S) flanked by two regulatory subunits (19S) (Figure 1). Inhibition of the 20S proteasome pathway has been proven important in the treatment of multiple myeloma (MM). Bortezomib, a drug administered intravenously or subcutaneously, was the first PI in class and has validated the proteasome as an effective anticancer target in MM.[46] The rationale for targeting the proteasome in MM is that malignant plasma cells produce large amounts of immunoglobulin that eventually has to be degraded, leading to a higher level of proteasome activity than in normal cells[47] When protein homeostasis is disrupted by a PI, the MM cells undergo apoptosis more readily than normal cells,[46] thus conferring selectivity to these agents.

Ixazomib refers to the biologically active, boronic acid form of the drug substance (MLN2238) (Figure 1). The drug substance is administered as a stable citrate ester, ixazomib citrate, a prodrug of ixazomib (MLN9708). In physiological conditions, ixazomib citrate rapidly hydrolyses to the biologically active boronic acid, ixazomib, which potently, reversibly, and selectively inhibits the proteasome (Figure 1). Ixazomib (like bortezomib and another PI, carfilzomib) preferentially binds to and inhibits chymotrypsin-like activity of the β5 subunit of the 20S proteasome and, at higher concentrations, inhibits the activity of the β1 and β2 sites (Figure 1). Whereas carfilzomib is an irreversible inhibitor of the 20S proteasome, ixazomib and bortezomib are both reversible inhibitors, and ixazomib has a shorter proteasome dissociation half-life than bortezomib (i.e inhibition is more rapidly reversible). Ixazomib is the first oral PI to be approved for the treatment of MM.[48]

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 29 of 315

Figure 1: Mechanism of action of ixazomib

==> picture [416 x 330] intentionally omitted <==

Ixazomib citrate (MLN9708)

administered orally as a capsule is rapidly absorbed and hydrolysed to the biologically active form (MLN2238) when it comes in contact with aqueous plasma. Ixazomib blocks protein degradation by inhibiting the 20S catalytic subunit of the 26S proteasome. More specifically, at lower concentrations, MLN2238 inhibits the β5 chymotrypsin-like subunit, which cleaves proteins after hydrophobic residues. At high concentrations, MLN2238 inhibits the β1 caspase-like subunit and β2 trypsin-like subunit, which cleave proteins after acidic and basic residues, respectively. Source: Muz et al 2016[38]

Cell-based studies have shown ixazomib to result in the anticipated downstream pathway inhibition arising from proteasome inhibition in MM cells, including increased levels of ubiquitinated proteins and apoptotic markers, activation of the endoplasmic reticulum stress response, and up-regulation of proteins degraded by the proteasome such as the tumour suppressor p53 and the cell cycle inhibitor p21.[49] Effects on the nuclear factor-κB pathway have also been demonstrated.[49] In MM cell lines, ixazomib has been shown to reduce cell viability and to have cytotoxicity, and synergistic activity has been demonstrated with lenalidomide.[49] Additionally, disruption of the bone marrow microenvironment has been seen, with ixazomib inhibiting bone marrow stromal cell-induced proliferation, along with promotion of osteoblastogenesis and inhibition of osteoclast activity.[50] In mouse xenograft models of MM, ixazomib resulted in tumour growth inhibition and prolonged survival.[49] and showed inhibition of bone resorption and alleviation of osteolytic bone disease.[50,51] Marketing authorisation/CE marking and health technology assessment.

2.1.3 UK marketing authorisation

On November 21[st] 2016, the European Commission granted a conditional marketing authorisation for Ninlaro[®] (ixazomib).

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 30 of 315

2.1.4 Indication

The indication for ixazomib is: “ Ninlaro in combination with lenalidomide and dexamethasone is indicated for the treatment of adult patients with multiple myeloma who have received at least one prior therapy .”

The SmPC posology indicates that ixazomib is used in combination with lenalidomide and dexamethasone. The SmPC is shown in Appendix 1.

2.1.5 Restrictions and contraindications

The SmPC for ixazomib states the following contraindications, and special warnings and precautions for use.

Contraindications

Hypersensitivity to the active substance or to any of the excipients listed in

Table 6 .

Table 6: Excipents in ixazomib (NINLARO)

Special warnings and precautions for use

Thrombocytopenia

Thrombocytopenia has been reported with ixazomib with platelet nadirs typically occurring between Days 14-21 of each 28-day cycle and recovery to baseline by the start of the next cycle.

Platelet counts should be monitored at least monthly during ixazomib treatment. More frequent monitoring should be considered during the first three cycles as per the lenalidomide SmPC.[52] Thrombocytopenia can be managed with dose modifications and platelet transfusions as per standard medical guidelines.

Gastrointestinal toxicities

Diarrhoea, constipation, nausea and vomiting have been reported with ixazomib, occasionally requiring use of antiemetic and antidiarrhoeal medicinal products and supportive care. The dose should be adjusted for severe

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 31 of 315

(Grade 3–4) symptoms. In case of severe gastrointestinal events, monitoring of serum potassium level is recommended.

Peripheral neuropathy

Peripheral neuropathy has been reported with ixazomib. The patient should be monitored for symptoms of peripheral neuropathy. Patients experiencing new or worsening peripheral neuropathy may require dose modification.

Peripheral oedema

Peripheral oedema has been reported with ixazomib. The patient should be evaluated for underlying causes and provide supportive care, as necessary. The dose of dexamethasone should be adjusted per its prescribing information or ixazomib for Grade 3 or 4 symptoms.

Cutaneous reactions

Rash has been reported with ixazomib. Rash should be managed with supportive care or with dose modification if Grade 2 or higher.

Hepatotoxicity

Drug-induced liver injury, hepatocellular injury, hepatic steatosis, hepatitis cholestatic and hepatotoxicity have been uncommonly reported with ixazomib. Hepatic enzymes should be monitored regularly and the dose should be adjusted for Grade 3 or 4 symptoms.

Pregnancy

Women should avoid becoming pregnant while being treated with ixazomib. If ixazomib is used during pregnancy or if the patient becomes pregnant while taking ixazomib, the patient should be apprised of the potential hazard to the foetus.

Women of childbearing potential must use highly effective contraception while taking ixazomib and for 90 days after stopping treatment. Women using hormonal contraceptives should additionally use a barrier method of contraception.

Posterior reversible encephalopathy syndrome

Posterior reversible encephalopathy syndrome (PRES) has occurred in patients receiving ixazomib. PRES is a rare, reversible, neurological disorder, which can present with seizure, hypertension, headache, altered consciousness, and visual disturbances. Brain imaging, preferably Magnetic Resonance Imaging, is used to confirm the diagnosis. In patients developing PRES, discontinue ixazomib.

Strong CYP3A inducers

Strong inducers may reduce the efficacy of ixazomib, therefore the concomitant use of strong CYP3A inducers such as carbamazepine, phenytoin, rifampicin and St. John’s Wort (Hypericum perforatum), should be avoided. Closely monitor patients for disease control

if co-administration with a strong CYP3A inducer cannot be avoided.

2.1.6 Regulatory process and timelines

On 30[th] July 2015, Takeda applied for marketing authorisation for ixazomib to the European Medicines Agency (EMA), using the centralised procedure. The Committee for Medicinal Products for Human Use (CHMP) of the EMA

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 32 of 315

granted an accelerated assessment to ixazomib; this is awarded to medicines deemed to be of major public health interest and, in particular, therapeutic innovation.

The EMA dossier was based on the same dataset utilised for the US FDA submission, which led to FDA approval in November 2015. Following data submissions and scientific discussions, the CHMP acknowledged that the TOURMALINE-MM1 study had met its primary endpoint at the first pre-planned interim analysis with a clinically significant median increase of PFS with the ixazomib regimen (IXA+LEN+DEX) versus control (LEN+DEX) (HR 0.74; p=0.012), especially as it showed a consistent benefit in patients with high-risk cytogenetic abnormalities, across subgroups, and other endpoints. In addition, no important uncertainties were identified in relation to ixazomib’s safety profile. As the primary endpoint had been achieved this became the final statistical analysis for PFS. In accordance with the study protocol plan all future statistical analyses (pre-specified interim analyses 2, 3 and Final) would be assessing overall survival. At the time of the second interim analysis (data cut July 2015; ~23 months follow up) an analysis on overall survival was conducted in line with the study protocol. In addition, there was a non-inferential analysis of PFS, overall response rates and TTP for the ITT population and this showed a reduced difference in effect between arms compared to the first interim analysis (hazard ratio for PFS of 0.82 (95% CI 0.67, 1.0; p=0.054). As the p-value for the primary endpoint was higher than that usually required by the EMA for a medicine with a single pivotal trial the CHMP requested that Takeda look to identify a subgroup of patients with a poor prognosis where the risk-benefit of ixazomib was more certain. Takeda and the CHMP were unable agree on such a subgroup within the constraints of the regulatory timeframe. The CHMP therefore issued a negative opinion for ixazomib on 26[th] May 2016.

Following the negative CHMP opinion, Takeda submitted an appeal for a re-examination. As part of the appeal, on September 5[th] 2016, the CHMP consulted with an independent Scientific Advisory Group (SAG) for Oncology which included multiple myeloma experts. The SAG reached the following conclusions:

• The SAG considered unanimously that the data submitted on the basis of the primary ITT analysis of PFS of TOURMALINE-MM1 (HR=0.742; p=0.012), and, importantly, the favourable toxicity profile, established a clear positive benefit-risk balance in the ITT population.

• The SAG discussed the results of the second interim analysis and concerns on data maturity. The SAG considered that on the basis of the primary PFS analysis (i.e. the first interim analysis), which was conducted according to the pre-specified statistical considerations, the trial had met its objective of showing a statistically and clinically significant improvement in PFS.

• The maturity of the dataset was considered adequate and consistent with that of other pivotal trials in this setting. The number of events was considered to be in line with what is generally expected in the field for trials of this size in this patient population. Hence, the ITT analysis was therefore considered mature and the observed effect (an approximate 6-month improvement in median PFS) as clinically relevant. The SAG concluded that the fact that a subsequent exploratory interim analysis showed some uncertainty about the level of statistical significance was not enough to change their conclusions about a clear beneficial effect in terms of PFS on the basis of the pre-planned (and, for PFS, the final) analysis.

• The supportive data from the China Continuation Study (a Phase III, randomised, double-blind, multicentre study comparing oral IXA+LEN+DEX versus LEN+DEX in adult patients with RRMM) provided corroborative findings in terms of a statistical and clinically significant effect in terms of PFS (HR=0.598; p=0.035). The robustness of this conclusion is also supported by internal consistency; namely, the favourable trends in terms of OS in the pivotal and supportive studies (HR=0.868, p=0.36, and HR=0.323, p=0.013, respectively).

• Concerning the benefit-risk balance, given the favourable toxicity profile of ixazomib, the SAG was confident that a positive benefit-risk balance has been established for ixazomib.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 33 of 315

• The SAG also considered that ixazomib is the first agent to allow oral triple combination therapy, which represents a therapeutic innovation in terms of convenience for patients.

• The fact that the results are based on a single pivotal trial is not considered an issue in view of the convincing results, the supportive evidence from the China Continuation Study and the entirety of the evidence.

Following the advice from the SAG and considering all of the detailed grounds for re-examination, including review of data on quality, safety and efficacy, the CHMP concluded that the delay in disease progression observed with ixazomib in the TOURMALINE-MM1 study was clinically relevant. Any remaining uncertainty seemed acceptable given the favourable toxicity profile, and that ixazomib is the first agent to allow oral triple combination therapy in this patient population, which represents a therapeutic innovation in terms of convenience for patients. Therefore, on 15[th] September 2016 the CHMP re-examined its initial opinion and, in its final and binding opinion, recommended the granting of a conditional marketing authorisation for Ninlaro (ixazomib) in the following indication:

‘‘NINLARO in combination with lenalidomide and dexamethasone is indicated for the treatment of adult patients with multiple myeloma who have received at least one prior therapy’’

The CHMP considered that ixazomib fell within the scope of European Commission regulations concerning the granting of a conditional marketing authorisation and fulfilled this on the basis that:

a) The benefit/risk balance of ixazomib is positive:

The 5.9 months gain in PFS data observed in the final statistical analysis of PFS in the ITT population in TOURMALINE-MM1 is significant and clinically relevant in RRMM (as part of its analysis the CHMP concluded that recently approved drugs for the treatment of MM have shown improvements in median in PFS in the range of 4 to 6 months). Together with the low toxicity of ixazomib and the benefit of the oral dosing regimen, the benefit-risk balance is considered positive.

b) As part of the conditional marketing authorisation, Takeda will provide additional data to confirm efficacy and safety of ixazomib from ongoing studies, including:

• The 3rd interim analysis and final OS report from the TOURMALINE-MM1 study.

• Final OS results from C16010 China Continuation study (a Phase III, randomised, double-blind, multicentre study comparing oral ixazomib plus lenalidomide and dexamethasone versus placebo plus lenalidomide and dexamethasone in adult patients with RRMM).

• Primary endpoint PFS results from study C16014 (a Phase III, randomised, double-blind, multicentre study comparing oral ixazomib plus lenalidomide and dexamethasone versus placebo plus lenalidomide and dexamethasone in adult patients with newly diagnosed MM not eligible for ASCT).

• Primary endpoint PFS results from study C16019 (a Phase III, randomised, placebo-controlled, double-blind study of oral ixazomib maintenance therapy in patients with MM following ASCT).

• Descriptive data from study NSMM-5001 (a global, prospective, non-interventional, observational study of presentation, treatment patterns, and outcomes in MM patients).

Data is expected to be provided by Takeda to the EMA for the China Continuation study by December 2016, for the study C16014 by December 2017, for the study C16019 by December 2018, and for study NSMM-5001 and TOURMALINE-MM1 by December 2019.

c) Ixazomib fulfils an unmet medical need:

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 34 of 315

The efficacy benefit of ixazomib is comparable with other therapies, but due to the significantly lower toxicity and the additional benefits of an oral dosing regimen, this product provides a major therapeutic advantage in comparison with available treatments and an important contribution to the care of patients with MM. The oral delivery of the ixazomib regimen addresses multiple myeloma patient needs and overcomes some of the significant burdens they face with currently available intravenous/ injectable therapies.

d) The benefits to the public health of the immediate availability of ixazomib outweighs the risks inherent in the fact that additional data are still required

Considering that ixazomib has favourable safety profile that is superior to that of the available alternatives, and that it is the first agent to allow oral therapy in this patient population (considerably improving convenience for patients), the immediate availability of ixazomib outweighs the risk inherent in the fact that additional data are still required.

Subsequently, on November 21[st] 2016 the European Commission granted conditional marketing authorisation for Ninlaro (ixazomib), indicated in combination with lenalidomide and dexamethasone for adult patients with multiple myeloma who have received at least one prior therapy. In addition at its meeting held on 4-6[th] October 2016 the EMA’s Committee for Orphan Medicinal Products (COMP) concluded that ixazomib’s designation as an orphan medicinal product should be maintained for the treatment of multiple myeloma (Orphan decision number EU/3/11/899, originally granted on 27th September 2011).

The EPAR is publically available.[53]

2.1.7 Expected date of availability in the UK

Takeda UK will not launch commercial stock until confirmed availability of NHS funding, either on receipt of a NICE positive recommendation or via the new Cancer Drugs Fund (CDF). During the period between EC approval (November 2016) and commercial stock availability, ixazomib (for use in combination with lenalidomide and dexamethasone) will continue to remain available through a global compassionate use, named patient program (NPP). This NPP has been open since January 2016 and access has been restricted to RRMM patients who fulfil the inclusion criteria for TOURMALINE-MM1. This is consistent with the European Marketing Authorisation where the positive risk-benefit of IXA+LEN+DEX has been demonstrated. When a final reimbursement decision is made, the NPP will immediately cease to include new patients. At this time, should ongoing therapy for existing patients already enrolled in the NPP not be covered by the reimbursement decision, then ixazomib will continue to be provided to such patients through the existing NPP process.

The availability of ixazomib through the NPP has not been promoted by Takeda and therefore enrolment has been solely restricted to unsolicited, compassionate use requests from physicians. As of 2[nd] December 2016, a substantial number of RRMM patients in the UK and Ireland have been enrolled across 51 sites. The significant uptake of the NPP highlights the unmet need for effective RRMM treatments and also shows the level of clinician demand for ixazomib. The decision to continue the NPP after receipt of Marketing Authorisation shows Takeda’s commitment to support the clinical community and ensure the best outcomes for myeloma patients.

2.1.8 Regulatory approval outside the UK

In November 2015, the US Food and Drug Administration (FDA) approved ixazomib in combination with lenalidomide and dexamethasone for the treatment of patients with multiple myeloma who have received at least

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 35 of 315

one prior therapy.[54] The FDA granted ixazomib priority review (given to drugs that, if approved, would be a significant improvement in safety or effectiveness in the treatment of a serious condition) and orphan drug designation for ixazomib.[55] Ixazomib is also licenced in combination with lenalidomide and dexamethasone for the treatment of patients with multiple myeloma who have received at least one prior therapy in in a number of other countries (Table 7).

Table 7: Ixazomib Worldwide Marketing Authorization Status - APPROVED

2.1.9 Other health technology assessments in the UK

Ixazomib is expected to be appraised by the Scottish Medicines Consortium (SMC), with submission currently scheduled for Q1 2017.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 36 of 315

2.2 Administration and costs of the technology

2.2.1 Administration and costs of ixazomib

Details of the ixazomib treatment regimen and costs are shown in Table 8.

Table 8: Treatment details and costs of ixazomib

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 37 of 315

2.2.2 Patient Access Scheme (PAS)

Takeda UK has submitted a proposed Patient Access Scheme (PAS) for Ninlaro[®] (ixazomib) to PASLU. The proposed PAS would offer a reduction from the basic NHS list price through a discount applied to all original invoices for Ninlaro. The proposed PAS would be made available to all NHS providers including homecare providers, secondary care, NHS patients in a private hospital, outsourced hospital pharmacies, and primary care via homecare arrangements.

2.3 Changes in service provision and management

2.3.1 Additional tests or investigations

Treatment with IXA+LEN+DEX must be initiated and monitored under the supervision of a physician experienced in the management of MM. No additional tests or investigations are required when ixazomib is used in combination with lenalidomide and dexamethasone, other than those that are already required for the lenalidomidedexamethasone regimen.

2.3.2 Resource implications for the NHS

Ixazomib is a convenient fixed-dose once-weekly capsule, and the all-oral combination of IXA+LEN+DEX is expected to have a beneficial impact on healthcare resource use within the NHS, in addition to reducing the treatment burden for both patients and caregivers.

The most relevant comparators in the NICE scope are:

• for people who have had at least 1 prior therapy: bortezomib with dexamethasone

• for people who have had 2 prior therapies: lenalidomide with dexamethasone

The PI bortezomib requires parenteral administration (i.e. subcutaneous [SC] or intravenous [IV]) in a clinic/hospital; the approved dosing schedule (invariably given with dexamethasone) requires two patient visits to the clinic per week for weeks 1 and 2 of the 21-day treatment cycle for cycles 1-8 (i.e. 32 visits over 8 cycles)[27,28] In the UK, it is common practice for bortezomib to be given once weekly and by subcutaneous injection, because this is associated with fewer adverse reactions, and therefore patients are able to receive the full 32 doses.[3]

The need for healthcare professionals to administer SC or IV bortezomib inevitably increases the burden on NHS resources and costs. In addition, this demanding treatment schedule may adversely affect the patient and/or caregiver financially (e.g. travel costs, time away from work[29] ) as well as impacting on their quality of life (e.g. the unpleasantness of having injections; time spent travelling and at the clinic/hospital)[30-35] This is particularly relevant for multiple myeloma patients as bone degradation, fractures and fatigue are common symptoms of the disease, thereby compromising patients’ mobility and increasing the importance of minimising travel. In contrast, the completely oral IXA+LEN+DEX regimen can be taken at home (Table 9), giving patients greater control over their treatment and their lives.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 38 of 315

Table 9: Recommended dosing schedule of BORT+DEX, LEN+DEX, and IXA+LEN+DEX

The IXA+LEN+DEX triplet regimen is not expected to have a significant impact on NHS resource utilisation compared to the existing LEN+DEX doublet regimen. In the TOURMALINE-MM1 study (the pivotal Phase III RCT assessing the efficacy and safety of IXA+LEN+DEX compared to LEN+DEX), an exploratory analysis at the first interim analysis (median follow-up of ~15-months) suggested that healthcare resource utilisation was lower with IXA+LEN+DEX versus LEN+DEX, based on rates per 100-patient-years of treatment exposure, which may potentially translate into lower costs of disease management[42] Rates of hospitalisations (56.0 vs. 63.4), acute care unit stays (45.5 vs. 51.8), outpatient visits (299.3 vs. 343.8), study physician/site visits (72.7 vs. 99.4), other physician/clinic visits (146.7 vs. 168.2), and palliative care unit stays (4.4 vs. 7.2) were lower with IXA+LEN+DEX, whereas the rate of emergency room stays (18.7 vs. 12.1) was higher and the rate of intensive care unit (ICU) stays (3.4 vs. 2.5) was similar. A second interim analysis with a median follow-up of ~23-months also demonstrated that, compared to patients in the LEN+DEX regimen, patients in the IXA+LEN+DEX regimen had lower numbers of hospitalisations (53.0 vs. 56.4), acute care unit stays (41.1 vs. 46.3), study physician/site visits (88.9 vs. 93.9), and other physician/clinic visits (167.7 vs. 181.0); a higher number of emergency room stays (14.3 vs. 10.9); and similarly low numbers of palliative care unit stays 6.9 vs. 6.2) and ICU stays (3.4 vs. 2.4). Because the 95% CIs

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 39 of 315

overlapped, health care utilisation in the two treatment groups would be assessed as similar overall, despite the addition of a third active agent.

2.3.3 Concomitant therapies

Ixazomib is indicated in combination with lenalidomide and dexamethasone (as the IXA+LEN+DEX regimen; see SmPC [Appendix 1]).

Anti-viral prophylaxis should be considered in patients being treated with ixazomib to decrease the risk of herpes zoster reactivation. Patients included in studies with ixazomib who received anti-viral prophylaxis had a lower incidence of herpes zoster infection compared to patients who did not receive prophylaxis. As for the existing LEN+DEX regimen, thromboprophylaxis is recommended in patients being treated with IXA+LEN+DEX (based on the know side-effect profile of lenalidomide), and should be based on an assessment of the patient’s underlying risk and clinical status (see SmPC).

2.4 Innovation

Ixazomib is the first and currently the only approved oral PI. This is a step-change in the management of MM, providing the benefits of a triplet regimen of a PI, an immunomodulatory drug (lenalidomide) and a corticosteroid (dexamethasone) but in an all-oral regimen that can be taken at home. Ixazomib offers a convenient, fixed dose, once-weekly, single capsule administration. The all-oral IXA+LEN+DEX triplet regimen represents a therapeutic innovation by offering practical and logistical benefits to patients, caregivers, healthcare professionals and the broader NHS, particularly versus the parenteral treatments that are currently available and widely used.

The IXA+LEN+DEX triplet is well tolerated and it offers the efficacy benefits of a triplet regimen, without significant increased toxicity versus the doublet LEN+DEX regimen. This favourable toxicity profile is predicted to enable continuation of treatment, leading to a durable response. These factors have been included in the QALY.

However, one factor that has not been included in the QALY is the unmet need for such a regimen. MM is characterised by multiple relapses and, at present, there are relatively few treatment options that are funded in England and Wales in the second and third line setting (see Section 3.3). Therefore, ixazomib would be a welcome addition to the therapeutic armamentarium, with the added advantage of a convenient all-oral regimen that is efficacious with a good tolerability profile. In May 2016,a letter in support of ixazomib was sent to the CHMP by the following organisations: Myeloma UK, the British Society of Haematology, the UK Myeloma Forum, as well as Myeloma Patients Europe and other European organisations. The letter documents the strong clinical support for using ixazomib in RRMM, as evidenced by the widespread uptake of ixazomib through the EMA approved compassionate use programme.[36]

Another factor which is not included in the QALY is the benefit that ixazomib can offer to a patient’s carer by reducing the burden of treatment on them (e.g. reduced need to accompany the patient to hospital visits, freeing up time for other activities). For patients who are still in employment there are potential benefits for the economy/society, for example by allowing patients to continue working or reducing the amount of time they need to take off work for clinic visits to receive injectable therapies.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 40 of 315

3. Health condition and position of the technology in the treatment pathway

3.1 Overview of multiple myeloma and its course

3.1.1 Disease course of MM and relapsed or refractory multiple myeloma (RRMM)

Multiple myeloma (MM) is a malignant plasma cell disorder that accounts for approximately 1.6% of all neoplasms and 16.6% of haematologic malignancies.[56] It is classified as an orphan (rare) disease (defined in the EU as a prevalence not exceeding 5 in 10,000 people), primarily affecting elderly patients (see Section 3.4.1). Ixazomib was granted orphan drug designation in MM in both the U.S. and Europe in 2011.

Plasma cells are an important part of the immune system and are responsible for producing specific antibodies to fight infections.[6] Different clones of plasma cells make numerous types of antibodies (immunoglobulins); MM occurs when a malignant transformation results in a population of clonal plasma cells that divide uncontrollably[5] and produce large quantities of one antibody (monoclonal Ig protein or M-protein)[6] The malignant myeloma cells most commonly produce IgG or IgA, or rarely IgD, IgM, or IgE, M-protein; in approximately a fifth of cases, there is no heavy chain, only the overproduction of κ and λ light chains.[57] The production of large quantities of antibody leads to unfolded immunoglobulins within the myeloma cells, which eventually needs to be degraded via the ubiquitin-proteasome system (UPS). MM cells have a higher level of proteasome activity than normal cells to deal with these unfolded proteins, and disruption of protein homeostasis by a proteasome inhibitor (PI) such as ixazomib results in apoptosis of the MM cells more readily than normal cells.

MM is an incurable progressive disease, characterised by multiple relapses; after a successful initial treatment resulting in stable disease or remission, nearly all patients will eventually relapse and will require further therapy. It typically recurs with a more aggressive disease course after each remission, resulting in shorter duration of response with each successive line of therapy and eventually treatment-refractory disease.[9]

Figure 2 is a schematic representation of the typical disease course in MM. Patients who have relapsed after treatment or who have refractory MM (i.e. RRMM) represent the population of interest in this submission.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 41 of 315

Figure 2: Typical MM disease course

==> picture [403 x 207] intentionally omitted <==

Abbreviations: M = monoclonal; MGUS = monoclonal gammopathy of undetermined significance Source: Adapted from Durie, 2011[58]

Recent advances in genome sequencing have provided evidence of both clonal heterogeneity and shifting clonal dominance over time in MM,[13] which are thought to contribute to drug resistance and relapse over the course of treatment (Figure 3)[13] The clonal evolution occurs over time because of selective pressures from treatment and the bone marrow microenvironment,[13,59] involving either the re-emergence of the dominant clone, linearly acquired mutations within the dominant clone or evolution of a prediagnostic clone with newly acquired mutations. Clonal heterogeneity may at least partly explain the high level of heterogeneity that is observed clinically among RRMM patients, both in terms of disease presentation and their response to treatment.

Combination therapy regimens are frequently used to treat patients with MM; clonal heterogeneity may, at least in part, explain the success of combination therapy in MM as the different therapies can target coexisting sub-clones. 13 The heterogeneity of MM also emphasises the need for physicans to have access to a range of different treatment options so that combination regimens can be tailored to the needs of individual patients.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 42 of 315

Figure 3: Hypothetical and illustrative example of clonal evolution and heterogeneity in patients with MM over the disease course

==> picture [304 x 365] intentionally omitted <==

----- Start of picture text -----
Initial treatment with
combination therapy option
1: remission for 2 years
Relapsed MM: clone 1 remains
most prominent; retreatment
with combination therapy option
1: remission for 6 months
Relapsed and refractory MM:
emergence of clone 3:
treatment with combination
therapy option 2
Abbreviations: BMSC = bone marrow stromal cells; MM = multiple myeloma
Source: Adapted from Cornell and Kassim, 2016 [13]
----- End of picture text -----

3.1.2 Risk assessment: Staging and cytogenetics

The clinical course of MM is highly heterogeneous with survival times ranging from a few weeks in some patients to >20 years for others.[60,61] Many studies have identified prognostic factors capable of predicting at least some of this heterogeneity in survival. For example, the International Staging System (ISS) is a powerful tool that defines 3 risk categories based on the serum concentration of β2-microglobulin and albumin.[60,61]

In addition, certain cytogenetic and molecular genetic abnormalities have been shown to predict outcome in MM. Immunoglobulin heavy chain gene translocations including t(4;14), t(14;16) and t(14;20), as well as copy number changes such as gain of chromosome 1q or deletion of chromosome 17p13 (del[17p]) demonstrated by fluorescence in situ hybridisation (FISH), are associated with adverse outcomes such as more aggressive disease and shorter survival.[37,60,61] The identification of these high-risk subgroups by FISH is recommended by NICE Clinical Guidelines (NG35)[62] and other clinical practice guidelines.[37,60-62]

The International Myeloma Working Group (IMWG) has recently published consensus guidelines on risk categories for MM that incorporate the ISS together with best available data on cytogenetics.[63] In addition, the Mayo Clinic

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 43 of 315

(US) has also published a risk stratification system for MM —the Mayo Stratification of Myeloma and Risk Adapted Therapy (mSMART)[64] (Table 10). As can been seen in Table 10 patients in the high-risk categories have a particularly poor prognosis, with a median overall survival of 2–3 years despite best available treatments, and are therefore in need of new treatments.

Although not a specific focus of this submission, the benefits of IXA+LEN+DEX have been demonstrated consistently across key pre-specified subgroups, including patients with high-risk cytogenetic abnormalities; there is a high unmet need in these patients for whom LEN+DEX is emerging as a sub-optimal treatment.[41]

Table 10: IMWG and Mayo Clinic (mSMART) risk stratification of MM

3.1.3 Symptoms of MM

MM is associated with anaemia, bruising or bleeding (caused by platelet deficiency), and increased risk of infection due to overcrowding of the bone marrow and overproduction of one type of unwanted antibody.[6] These arise as a result of the expanding myeloma cell population in the bone marrow compartment preventing normal reproduction of other blood cell types.[6]

Other features, such as kidney and other end-organ damage, as well as hyperviscosity of the blood, result from the overproduction of M-protein and light chains in particular. The classic diagnostic symptoms of bone pain and the presence of lytic bone lesions are caused by the malignant plasma cells stimulating osteoclasts and suppressing

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 44 of 315

osteoblasts, thereby leading to increased bone resorption. Furthermore, in the bone marrow microenvironment, myeloma cells create a pro-survival supportive signalling network with the stromal cells, producing cytokines and angiogenic factors that promote blood vessel formation and facilitate the survival of the malignant cells. These effects give rise to the classical ‘CRAB’ diagnostic criteria for active MM: hypercalcemia, renal impairment, anaemia, and bone lesions.[57] Patients typically present with at least one of these criteria at diagnosis, as well as a number of associated symptoms such as fatigue, bone pain, and fractures.[66,67]

3.2 The effects of multiple myeloma on patients, carers and society

3.2.1 Impact of symptoms on quality of life

The main signs and symptoms of MM including pain and bone disease, anaemia, fatigue, hypercalcaemia, renal impairment and recurring infections, heavily impact on the quality of life (QoL) of patients, as described in Table 11.

Table 11: Clinical signs and symptoms of MM, causal factor(s), clinical manifestation and frequency, and impact/burden on patients

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 45 of 315

Studies have reported that patients with MM experience a very high symptom burden and low QoL compared with the general population, as assessed using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) and the Quality of Life Questionnaire MM Module 20 (QLQ-MY20).[75] For example pain has been shown to have a significant impact on the QoL of patients with MM.[76,77] Pain can limit mobility, restricting the distance that patients can walk, or walk independently. It can also disrupt sleep and everyday activities, including leisure pursuits and job-related functions. Since pain greatly affects MM patients, dependence on analgesics is high and the side-effects of pain medication also add to the negative QoL of these patients.[78] Likewise, fatigue, which is a common symptom of MM mainly caused by anaemia, is associated with sleep and mood disturbances, reduced ability to complete simple physical activities, and elevated pain levels. 79

3.2.2 Quality of life burden of treatment administration and adverse events

IV and SC administration of drugs is associated with frequent clinic visits for treatment administration, in addition to routine visits for diagnosis and monitoring. For example, a recently approved PI, carfilzomib, requires 6 clinic visits per month, including visits on consecutive days (i.e. it is given on days 1, 2, 8, 9, 15, and 16), followed by a 12-day rest period every 28 days.[80] Bortezomib’s approved dosing schedule is by IV or SC injection twice weekly for 2 weeks on days 1, 4, 8, and 11[27] (in clinical practice in the UK bortezomib is commonly given once weekly subcutaneously for up to 32 doses[3] ). These treatment schedules are illustrated in Figure 4. Due to the impact on

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 46 of 315

patients’ mobility from pain, fatigue, bone degradation and bone fractures, which are symptoms of multiple meyloma, the frequency of travel including hospital visits for treatments is of great relevance.

Figure 4: Administration schedules for bortezomib and carfilzomib

==> picture [441 x 56] intentionally omitted <==

A recent study by Baz et al., 2015[76] developed a conceptual model to illustrate the impact of MM and its treatment on QoL. The authors of the study reported that therapies often cause adverse effects and can have demanding administration and monitoring schedules (Table 12) . Based on feedback from the interviewees, the study authors reported that clinic visits ranged from ‘very quick’ to up to 2 hours. While most patients did not find clinic visits to be inconvenient, one participant had an 80-mile journey to get to the clinic for treatment, which was reported to be the biggest impact on their QoL; others who were receiving bisphosphonates felt that a monthly intravenous treatment had the biggest impact on their life. Participants receiving oral regimens did not report any inconvenience associated with their treatment.[76]

Feedback from clinicians indicates that in many parts of the UK there is significant pressure on hospital infusion capacity and this frequently leads to lengthy waiting times for patients who are being treated with injectable therapies. This has an adverse impact on QoL for patients as it means they must spend additional time at the hospital, thus limiting their time for other life activities. The availability of an all-oral treatment regimen would offer significant QoL benefits for such patients, as well as lessening the pressure on the hospital system.

Table 12: Selected examples of the negative impact of MM treatment on patients’ QoL

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 47 of 315

3.2.3 Quality of life impact on caregivers

The diagnosis of MM can also be a life-changing event for informal caregivers who must face the fear of losing a family member or friend while at the same time providing emotional support to the patient.[81] In addition to emotional support, caregivers often need to provide direct skilled care such as administering medicines (the authors did not specify types of medication, but that the proportion of care-givers who report administering medication as part of their role is 34%), monitoring for AEs and coordinating care between the patient and healthcare providers (e.g. transportation of patients to medical appointments). Increased stress has been observed in caregivers who do not feel adequately prepared for this responsibility (e.g. because of lack of knowledge).[82] A summary of the key elements of the caregiver role is presented in Table 13.

Table 13: Key elements of the caregiver role

A qualitative study by Molassiotis and colleagues found that MM significantly impacts patients’ and caregivers’ emotional, social, and work-related areas of life.[83] Moreover, caregivers often neglect their own needs in order to support patients. In a survey of 132 patients with MM and 93 of their partners, one-third of partners reported unmet supportive care needs; almost half reported signs of anxiety (48.8%) and 13.6% reported signs of depression.[84]

As MM is increasingly experienced as a chronic condition, Stephens and colleagues performed a qualitative study on 10 long-term survivors and their primary carers to understand the experience of long-term survival in MM patients and their caregivers. The carers of these patients had to be mindful of managing the risk to their own wellbeing and balance their own needs along with those of the family and the MM sufferer. Further to this, carers also had to manage their own emotional responses including the feelings of anxiety, loss and uncertainty, particularly when dealing with the patient.[85]

Time commitments, resulting in lost work productivity, can also be considerable for caregivers. Although MM treatment protocols vary between treatment centres, they all require a large time commitment from patients and their caregivers.[86]

3.2.4 Years of potential life lost and indirect costs

The average years of life lost with MM ranges from 36 years in patients younger than 40 years to less than 5 years in patients aged 80 years and older.[87] Using a commonly applied standard value of US$150,000 for 1 year of life

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 48 of 315

for both sexes and all ages,[88,89] Ludwig et al. estimated that the value of life-years lost in a MM patient diagnosed at age 50 amounts to $4,035,000 (= approximately £2,835,700).[87] This figure highlights the significant economic burden that MM places on society.

MM can result in significant lost work productivity because of patients leaving the workforce. This places a considerable financial burden on patients, families and society as a whole. A survey of MM patients (n=762) who had received treatment at a single US centre reported that of 500 (66%) respondents who were employed at the time of diagnosis and treatment onset, only 33% were currently employed, including 41% of patients aged 54 years or younger.[86] The primary explanation given by patients (35%) for why their employment had ended was a physical inability to carry out job-related functions, mainly due to fatigue and/or pain. Additionally, 21% of patients were no longer employed because of the amount of time and/or travel required for treatment (e.g. frequent physician appointments or prolonged treatment stays). Peripheral neuropathy and the potential for exposure to infection in the workplace were other reasons given by patients (14%). Only 11% of patients had retired. Furthermore, some of the patients who were still in employment had reduced their work hours or had used medical leave (paid and/or unpaid). Ensuring the optimal and efficient treatment of patients with MM and improving outcomes for this disease may reduce the personal and wider economic impact of MM.

3.3 Clinical pathway of care and the impact of ixazomib

The current MM treatment pathway in England, reflecting both the current funding position of the different treatment options and their use in clinical practice, along with the proposed positioning of the IXA+LEN+DEX regimen, is shown in Figure 5

Figure 5: Current treatment pathway for multiple myeloma in England and the proposed positioning of the ixazomib regimen

==> picture [399 x 275] intentionally omitted <==

*Panobinostat-bortezomib-dexamethasone is funded in 3[rd] line onwards; however, in clinical practice, patients tend to receive LEN+DEX prior to the panobinostat regimen, meaning that panobinostat is predominantly used as a 4[th] line therapy.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 49 of 315

First line: In newly diagnosed, transplant-eligible patients, the recommended first-line treatment is induction with bortezomib plus either thalidomide and dexamethasone (VTd) or dexamethasone alone (Vd), followed by an autologous stem cell transplant (ASCT).[90] Although ASCT is recommended in newly diagnosed MM patients, with a second transplant recommended for some patients with relapsed myeloma, it should be noted that only a small number of patients are suitable for this treatment,[90] highlighting the importance of systemic drug treatment for the majority of patients. In transplant-ineligible patients, the first-line treatment option is thalidomide in combination with an alkylating agent and a corticosteroid; when thalidomide is unsuitable (due to contraindications or intolerance) bortezomib in combination with an alkylating agent and a corticosteroid is recommended (Figure 5).

First relapse (2[nd] line): The NICE scope has specified a number of potential comparators, including bortezomib ± dexamethasone, bortezomib retreatment and lenalidomide plus dexamethasone. These comparators are based on the NICE clinical pathway and other treatments thought to be of potential relevance in clinical practice in England and Wales. Although bortezomib monotherapy is reimbursed by NICE, in practice bortezomib is usually given in combination with dexamethasone (as agreed in the scope). In the latest multiple myeloma IMS therapy tracker market research (from Oct 2016, based on 37 specialists and 347 patient records across all lines of therapy), bortezomib has a 68% market share at 2[nd] line, as shown in Table 14 and is by far the dominant therapy. Bortezomib plus dexamethasone is therefore the appropriate comparator for the ixazomib regimen at 2[nd] line. According to expert clinical opinion, some of the other comparators in the NICE scope are unlikely to be displaced by the ixazomib regimen in clinical practice for the following reasons:

• Bortezomib retreatment (with or without dexamethasone): this treatment option is not recommended by NICE and is no longer available on the CDF

• Lenalidomide with dexamethasone (subject to ongoing NICE appraisal [part review of Technology appraisal 171]): this treatment option is not currently recommended by NICE, is not available on the CDF and recently received a negative draft opinion from NICE at the appraisal consultation stage (ACD published on 11[th] November 2016).[25]

Second relapse onwards (3[rd] line +): For patients who have received two or more prior treatments, LEN+DEX is recommended by NICE. In addition, in January 2016, panobinostat in combination with bortezomib and dexamethasone received a positive recommendation by NICE within its marketing authorisation, that is, for 'adult patients with relapsed and/or refractory multiple myeloma who have received at least 2 prior regimens including bortezomib and an immunomodulatory agent'. As thalidomide (immunomodulatory agent) and bortezomib are typically used first and second line in England and Wales, this potentially places the panobinostat regimen at 3[rd] line onwards. However, the predominant use of the panobinostat regimen in England and Wales is at 4[th] line, after LEN+DEX; this was confirmed by the clinical experts during the carfilzomib and pomalidomide NICE Appraisal Committee meetings in October 2016 (Figure 5).

Consistent with this proposed care pathway, the latest multiple myeloma IMS therapy tracker market research shows that lenalidomide is dominant in 3[rd] line with a 69% market share. The panobinostat regimen has a low market share at both 3[rd] and 4[th] line (currently 7% and 19% respectively; Table 14)

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 50 of 315

Table 14: IMS multiple myeloma therapy tracker market share data of therapies by line of treatment

In November 2016, pomalidomide in combination with low-dose dexamethasone received a positive recommendation from NICE as an option for treating MM at third or subsequent relapse; that is, after 3 previous treatments including both lenalidomide and bortezomib.[91] As pomalidomide can only be prescribed in routine practice after lenalidomide, the ixazomib regimen will be positioned in an earlier treatment line, excluding pomalidomide as a relevant comparator. For patients who have had four or more relapses, bendamustine is available as a treatment option only through the CDF or otherwise patients are treated with conventional chemotherapy combinations (e.g. melphalan and cyclophosphamide). Of note, and supporting our rationale here, the NICE scope for ixazomib did not include either pomalidomide, bendamustine nor conventional chemotherapy.

Conclusion: Based on the above rationale, the proposed positioning of the ixazomib regimen is as either a 2[nd] or 3[rd] line treatment option. Given where lenalidomide is currently most used we would expect the predominant use of the ixazomib regimen to be in the 3[rd] line setting. Based on the current treatment pathway, the relevant comparator for the ixazomib regimen in patients who have received 1 prior therapy is bortezomib plus dexamethasone, while for patients who have received 2 prior therapies it is lenalidomide plus dexamethasone.

3.4 Epidemiology of multiple myeloma and relapsed/refractory multiple myeloma in the UK

The epidemiology of MM from Cancer Research UK (CRUK), which provides a comprehensive overview of national cancer statistics in the UK using data provided by members of the (UK) and Ireland Association of Cancer Registries, is presented in the following sections.[4] There are no epidemiology data in the published literature on the number of patients with RRMM. Therefore, a study was conducted by Lin and colleagues (Global Outcomes Research, Takeda Oncology, Cambridge, USA) to estimate the number of patients with RRMM (Section 3.4.3).

3.4.1 Incidence and mortality data for MM in the UK

MM is predominantly a disease of the elderly. Incidence rates increase suddenly between ages 55 and 59 with approximately 58% of MM cases being diagnosed in patients aged 70 and over (Figure 6).

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 51 of 315

Figure 6: Average number of new cases of MM per year and age-specific incidence rates per 100,000 population, UK (2012-2014)

==> picture [387 x 251] intentionally omitted <==

Source: CRUK 2016[4]

In 2014, there were approximately 5,501 new cases of MM in the UK, approximately 3,072 (56%) male and 2,429 female (44%) (Table 15) accounting for 2% of all new cancers. Data shows that MM is more common in men than in women and twice as common in black people compared to white and Asian people.

In 2014, there were 2,928 deaths from MM (Table 15), accounting for 2% of all cancer deaths in the UK. Around 6 in 10 (59%) MM deaths in the UK each year are in people aged 75 and over (2012-2014).

Table 15: Incidence and mortality data for MM in the UK (2014)

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 52 of 315

3.4.2 Survival data for MM in the UK

For MM, the 1-year net survival is 76.6% in England and Wales, decreasing to 47%, and 32.5% for 5, and 10 years, respectively (Table 16). Survival for men is slightly higher than women (Table 16). Around three quarters of people in England diagnosed with MM aged 15-49 survive their disease for five years or more, compared with a quarter of people diagnosed aged 80 and over (2009-2013).

Table 16 shows that, at the end of 2006, there were 12,465 people alive in the UK who had been diagnosed with MM within the previous 10 years. Again, a slight gender bias can be seen.

Table 16: Survival (2010–2011) and prevalence (at end of 2006) data for MM in the UK

3.4.3 Estimated incidence of RRMM in the UK

A study was conducted by Lin and colleagues (Global Outcomes Research, Takeda Oncology, Cambridge, USA) to estimate the number of patients with RRMM.[92] This involved a systematic review of MM epidemiology data (the search included electronic databases and online sources) to determine the country-specific incidence of MM and a systematic review of randomised controlled clinical trials in newly diagnosed MM to determine the proportion of patients whose disease progressed after first-line treatment. This is not a country-specific RRMM rate, as response to treatment is assumed to be applicable globally. The results of these two systematic reviews were used to calculate estimates of the incidence (per 100,000 person-years) of RRMM in the UK:

UK-specific incidence of MM × relapse/ refractory rate = number of RRMM in the UK

The relapsed and/ or refractory rate used was either the overall proportion of MM patients estimated to relapse from or be refractory to first-line treatment (65.7%), a 1-year rate (48.6%), or a 2-year rate (63.9%). The detailed methodology is described in Lin et al.,[92] and a summary of the results are reported below .

The sources identified in the systematic review and the incidence rates for MM are shown Table 17.

Table 17: UK MM incidence – sources and rates/100,000 of the general population

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 53 of 315

The UK population size for 2013 (ONS 2013: total: 64,105,654; male: 31,532,900; female: 32,572,800) was used to estimate the number of incident cases of MM in the UK according to incident rates reported in Table 17. Estimated incident cases of MM in the UK are reported in Table 18.

Table 18: UK: MM incidence –rates/100,000 of the general population

The incidence of RRMM in the UK was calculated by applying the proportion of RRMM (global) to the number of incident cases in the UK (Table 18).[92]

Table 19 summarises the estimated number of RRMM patients in the UK. Applying the total rate (65.7%) to the total MM population in the UK indicates that there are approximately 2,854 new cases of RRMM each year, which represents an incidence rate of 4.45 per 100,000 people.

Table 19: Estimated incident cases and incidence rate (cases per 100,000 population) of RRMM in the UK

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 54 of 315

3.4.4 Real-world data: patients by line of therapy

Two recent publications have reported on a pan-European (Belgium, France, Germany, Italy, Spain, Switzerland and the UK) real-world study based on an audit of MM patient charts.[93,94] The chart review consisted of a crosssectional audit (a total of 435 physicians completed 7635 cross-sectional chart reviews[94] ) and a retrospective component (435 physicians retrospectively reviewed 4997 patient charts[93] ).

Figure 7 shows the proportion of patients reaching each subsequent line of therapy (these data were derived partly from the cross-sectional part of the study), together with the duration of treatment and length of the treatment-free interval. Overall, 95% of patients diagnosed with symptomatic MM who were treated by haematologists received at least one line of anti-tumour drug treatment, 61% received ≥2 lines of therapy and 38% received ≥3 lines. Median duration of therapy and median treatment-free interval decreased with increasing lines of therapy, as did time to progression.

Figure 7: Proportion of patients by line of therapy, treatment duration and treatment-free intervals

==> picture [333 x 298] intentionally omitted <==

Source: Yong et al,. 2016[93]

In addition, the depth of response, as assessed by the treating physician, also decreased with each additional line of therapy, with 74% of patients achieving at least a very good partial response at first line, compared with only 11% at fifth line. Deeper responses were associated with longer time to progression, although these were physicianjudged. Toxicities and comorbidities increased with later treatment lines, and were more likely to have led to discontinuation of treatment. Overall, the authors concluded that there was an unmet need for better-tolerated, efficacious treatments in later treatment lines.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 55 of 315

3.5 Clinical guidance

3.5.1 Completed NICE technology appraisal guidance for RRMM

NICE recommendations for technologies in patients with RRMM are shown in Table 20 (see Section 3.3.1 for their position in the clinical pathway).

Table 20: NICE technology appraisals of relevance for patients with RRMM

Other therapies currently under consideration by NICE for patients with RRMM are shown in Table 21. Of note, carfilzomib [ID934] and lenalidomide (post bortezomib) (part rev TA171) [ID667] have recently received initial negative recommendations at the ACD stage by the NICE Appraisal Committee.

Regarding the upcoming NICE appraisals, of interest is the fact that carfilzomib and daratumumab are both administered intravenously. If recommended by NICE, both of these agents have administration schedules that would present logistical challenges to patients and the NHS. In the case of carfilzomib this relates to the frequency of the IV injections while for daratumumab the challenge is the lengthy IV infusion times. By contrast, if recommended by NICE, the IXA+LEN+DEX regimen would be the only all-oral triplet available on the NHS for RRMM, fulfilling an unmet need for additional efficacious treatment options for RRMM patients with a favourable

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 56 of 315

toxicity profile and the added advantages of a regimen that can be taken at home, minimising the travel burden and impact on the daily life of patients and caregivers.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 57 of 315

Table 21: Technologies under NICE review

3.5.2 Cancer Drugs Fund

The Cancer Drugs Fund (CDF) within NHS England is designed to improve the availability of cancer drugs on the NHS. Table 22 shows the therapies for the treatment of RRMM that are available through the CDF (version 1.13; 23[rd] November 2016).

Table 22: Medicines for the treatment of relapsed multiple myeloma available through the CDF

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 58 of 315

Table 23 shows the indications that were removed from the CDF in 2015. As a result of these delistings, there are now fewer treatment options available for MM patients in England in the relapsed setting.

3.5.3 UK clinical guidelines for RRMM

The NICE guidelines and clinical pathway for MM are described in Section 3.1. In addition, the British Committee for Standards in Haematology (BCSH) guidelines for the diagnosis and management of MM were published in 2011[102] and updated in 2014.[61] Table 24 summarises the BCSH clinical practice guidelines for the management of RRMM.

For patients at second and subsequent relapse the BCSH recommended lenalidomide, with patients presenting in renal failure being recommended a bortezomib-containing regimen. They also noted that many new drugs were in development at the time of the guidelines in 2014. The BCSH also stated that “decisions regarding treatment at relapse should be made according to a number of factors including the timing of relapse, efficacy and toxicity of drugs used in prior therapy (e.g. peripheral neuropathy), age, bone marrow and renal function, co-morbidities (e.g. diabetes) and patient preference”.

Table 24: British Committee for Standards in Haematology (BCSH) guidelines for RRMM

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 59 of 315

3.5.4 International and European guidelines

International and European clinical practice guidelines for RRMM are described in Table 25 As described, various factors need to be considered in order to make the optimal treatment choice, including patient-specific factors, tumour characteristics, such as cytogenetics, the type, efficacy and tolerance of the previous treatment, the number of prior treatment lines, the available remaining treatment options and the interval from the last therapy.[37] Of note, in the 2016 National Comprehensive Cancer Network (NCCN) guidelines from the US, the IXA+LEN+DEX regimen is a preferred treatment option for MM patients who have received at least one prior therapy, with category 1 evidence (i.e. based upon high-level evidence, there is uniform NCCN consensus that the intervention is appropriate), and it is also recommended by the US Mayo Clinic guidelines. In addition, the International Myeloma Working Group (IMWG) suggests that an oral regimen may be preferred in patients living far away from their treatment centre (Figure 8).

Table 25: International and European clinical practice guidelines for RRMM

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 60 of 315

Organisation Recommendations for RRMM b Indicated in combination with lenalidomide and dexamethasone for the treatment of patients who have received one to three prior therapies. c Indicated for the treatment of patients who have received at least one prior therapy. d Consider single-agent lenalidomide, pomalidomide, or thalidomide for steroid intolerant individuals. e Indicated for the treatment of patients who have received at least two prior regimens, including bortezomib and an immunomodulatory agent. f Indicated for the treatment of patients who have received at least two prior therapies including bortezomib and an immunomodulatory agent and have demonstrated disease progression on or within 60 days of completion of the last therapy.

Figure 8: Parameters relevant for treatment selection in patients with relapsed/refractory multiple myeloma (IMWG recommendations 2014)

==> picture [396 x 181] intentionally omitted <==

Source: Ludwig et al 2014[37]

3.6 Issues relating to current clinical practice

There is currently no cure for MM and thus prolongation of PFS and OS remain the ultimate goals of treatment. The discovery of novel agents such as immunomodulatory drugs (IMiD) (e.g. thalidomide and lenalidomide) and proteasome inhibitors (PI) (e.g. bortezomib) have contributed to a doubling of survival in MM patients as compared to the 1990s when only conventional chemotherapy was used.[105] However, despite such advances, most patients ultimately relapse.[9] Such patients are difficult to treat as they tend to have more aggressive disease (and thus are less responsive to treatment) and have been heavily pre-treated, thereby having more pre-existing toxicities.[9]

There are several issues relating to current clinical practice, particularly in relation to the logistical burden that existing treatments place on patients and the NHS. The IXA+LEN+DEX regimen overcomes many of these limitations, with the potential to offer meaningful benefits to patients, caregivers and the wider healthcare system, as follows:

Improved treatment choice

For a disease that is characterised by multiple relapses, there are relatively few treatment options available on the NHS in England, especially in light of the recent delistings from the CDF (Section 3.5.2), and negative recommendations from NICE (Section 3.5.1). It is vital for MM patients to have access to a range of treatments and treatment combinations at each stage of the disease pathway, so that their treatment can be tailored to their needs and circumstances.[36]

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 61 of 315

Improved long-term outcomes

Combination therapy has become the standard-of-care in many situations, with the complementary activity of a PI, an IMiD and a corticosteroid (dexamethasone) particularly promising.[12,14] Recent Phase III studies have demonstrated superior efficacy with triplet versus doublet combination regimens based on these agents in the frontline[15,16,106] and relapsed settings.[17,18]

In addition, to further improve long-term outcomes, there has been a shift towards a paradigm of extended treatment.[19-21] However, treatment-related adverse events plus the demanding administration and monitoring schedules of treatments such as bortezomib impose a substantial burden on patients over time, leading to treatment discontinuation in many cases.[38,107]

The triple combination of IXA+LEN+DEX significantly improves progression free survival (PFS) by an average of 6 months compared to LEN+DEX alone, representing a significant and prolonged period for patients and families. In some groups of patients, including third line and high-risk patients, the PFS gain was greater than this; given that these patients currently have very poor outcomes this is a major area of unmet need.[36] In addition, the favourable toxicity profile of ixazomib is predicted to enable continuation of treatment, leading to a durable response.

Improved quality of life

Improving the quality of life of patients is an important treatment goal in MM.[108] In particular, treatment tolerability is important in RRMM patients who are typically older and less fit. Ixazomib is not associated with the troublesome severe peripheral neuropathy or cardiopulmonary toxicities seen with other PIs, such as bortezomib and carfilzomib respectively.[1,18,45] No increase in serious adverse events (SAEs) has been reported with IXA+LEN+DEX compared to LEN+DEX alone.[1]

Furthermore, several reports have shown that many patients with cancer prefer oral to parenteral administration of chemotherapy. The reasons for this are many and include convenience, the place of treatment, a dislike of needles, anxiety over an IV line, feeling less ill and reducing the effort in coping with the disease.[30-35] Oral treatment options can help patients feel more “in control” of their treatment and also to carry on with their everyday lives without having to attend regular hospital appointments.[36] This is particularly important for patients who find it difficult to attend hospital appointments (e.g. if they are older/frailer, live far away from hospital or are still in employment). The all-oral combination of IXA+LEN+DEX can be taken at home, thereby reducing the burden on patients and caregivers (and the NHS – see below).

Reduced burden on the NHS

The need for healthcare professionals to administer bortezomib, either subcutaneously or intravenously, inevitably increases the burden on NHS resources and costs, and the management of treatment toxicity and side-effects also adds to this burden. Given the rising demand for cancer services as a whole (due to the ageing population and increased survival rates), together with limited resources and budgets,[109] the option of an oral PI/ all oral triplet regimen with a favourable toxicity profile could help reduce the burden on the NHS.

Overall, IXA+LEN+DEX would provide patients, clinicians and the NHS with the choice and flexibility of an all-oral treatment option that is efficacious with a favourable toxicity profile, which is predicted to reduce burden on the patient and the NHS. This is particualraly important for future sustainability as the arrival of new injectable

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 62 of 315

treatments, such as carfilzomib and daratumumab, seems set to increase the logistical burden on the NHS even further. In this context, a simple oral product like ixazomib looks like a particularly attractive treatment option for an increasingly resource constrained NHS.

3.7 Equality

We foresee no equality issues with ixazomib.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 63 of 315

4. Clinical effectiveness

4.1 Identification and selection of relevant studies

4.1.1 Systematic literature review

A systematic literature review (SLR) of the clinical evidence relating directly to the appraisal decision problem in Section 1.1 was undertaken in order to perform a decision-focused network meta-analysis (NMA) that is relevant for a NICE single technology appraisal (STA), as well as future submissions to other UK HTA bodies. The primary objective of the SLR was to address the following decision problem:

“What is the clinical efficacy and safety of ixazomib citrate in combination with lenalidomide and dexamethasone (LenDex) for the treatment of patients with relapsed or refractory multiple myeloma (RRMM) who have received at least one prior therapy?”

In line with this, data on the clinical efficacy and safety of ixazomib or comparators were obtained by a systematic literature search and review of published research evidence and conference abstracts (with supporting poster presentations), supplemented by unpublished trial data for ixazomib supplied by Takeda UK.

The search terms were developed specifically for each database. Searches took into account generic and other product names including variations in different countries. Specific search filters for randomised controlled trials were used to retrieve studies of clinical effectiveness. Each abstract was assessed by two independent reviewers.

The search strategy used is reported in Appendix 2. In addition, separate ixazomib SLR reports (firstly an interim analysis utilising the first data-cut date of 30[th] October 2014 [IA1]; followed by an updated analysis by means of the second data of 12[th] July 2015 [IA2]) provide further details.[110,111] The search strategy was executed on 1[st] June 2015 and repeated firstly on the 10[th] April 2016, and then more recently on the 7[th] October 2016 to update the search and fill in gaps in the evidence from the first systematic search in the following databases:

• MEDLINE® with Daily Update

• MEDLINE® In-Process & Other Non-Indexed Citations

• Embase

• The Cochrane library (Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials)

• The Centre for reviews and Dissemination (CRD) Databases: Database of Abstracts of Reviews of Effect (DARE), Health Technology Assessment Database (HTA), NHS Economic Evaluation Database (NHSEED)

• PubMed (for E-publications ahead of print)

Search strategies combined free-text and controlled vocabulary terms (MeSH terms in MEDLINE and CENTRAL and EMTREE terms in EMBASE) for multiple myeloma (MM). To identify relevant studies, the search strategies were designed to identify a MM population that was relapsed or refractory; hence they represent a combination of the terms for the disease of interest –‘Multiple myeloma’ and ‘relapsed or refractory.’

In addition, manual searches of the following conferences were performed to identify relevant abstracts, not yet available as full publications:

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 64 of 315

• American Society of Haematology (ASH)

• American Society of Clinical Oncology (ASCO)

• European Haematology Association (EHA)

• European Society for Medical Oncology (ESMO)

Furthermore, the Clinical trials website (www.clinicaltrials.gov) was searched to identify ongoing studies with interim results not yet published in an abstract/ poster; and lastly references from systematic reviews, indirect comparisons, and network meta-analyses were searched.

4.1.2 Eligibility criteria

The following eligibility criteria were used to determine articles to be included in the systematic review. Eligibility criteria are specified in the table below in terms of patients, interventions, comparators, outcome and study design (PICOS). The same criteria were applied for both the 2016 updates as for the original search carried out in 2015.

Table 26: Eligibility criteria for search strategies conducted in 2015 and 2016

Comparators Patients in the control group receiving placebo or dexamethasone were eligible. Studies without a control group but with two interventions of interest were also eligible for inclusion

Outcomes The following list of outcomes was considered. This list was compiled after reviewing the NICE draft scoping documents for similar regimens (i.e., carfilzomib and panobinostat) for treatment of MM in patients with at least 1 prior therapy:

• Progression Free Survival (PFS)

• Overall survival (OS)

• Overall response rate (ORR)

• Treatment discontinuation

• Adverse events (grades 3-4)

• Time to progression (TTP)

• Duration of response (DOR)

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 65 of 315

• Time to next treatment (TTNT)

• Duration of treatment

• Health-related quality of life (HRQoL)

Study design Peer-reviewed RCTs, observational studies and conference abstracts (both RCTs and observational studies) were eligible. Post-hoc analysis studies were examined only if an updated or new data of an original study was presented. All other types of literature, such as letters, editorials or reviews were not considered for inclusion

4.1.3 Study selection

The process of study selection was made according to specifications in the Protocol.[112] Thus, studies were selected based on the comparators and outcomes listed in the Final NICE scope for ixazomib,[113] as well as for HTA purposes and future submissions. Thus, studies involving comparators bendamustine, chemotherapy, daratumumab, and elotuzumab were not selected from the systematic review for data extraction as they were not to be included in the NMA (see Section 4.10). For the purpose of this STA submission, which has been previously agreed by the NICE Appraisal Committee following the publication of the carfilzomib appraisal consultation document (ACD)[3] the following studies will be reported only:

• Ixazomib + lenalidomide + dexamethasone compared with bortezomib + dexamethasone at second line (i.e. patients that have received 1 prior therapy)

• Ixazomib + lenalidomide + dexamethasone compared with lenalidomide + dexamethasone at third line (i.e. patients that have received 2+ prior therapies)

Due to the limited availability of randomised evidence for RRMM patients who have received 1 or 2+ prior therapies (for the comparators of interest shown above), non-randomised study designs such as prospective interventional studies, prospective observational studies, as well as retrospective studies were considered in this systematic review.

All abstracts were reviewed by two experienced systematic reviewers according to the eligibility criteria outlined in Section 4.1.2; any difference in opinion regarding eligibility was resolved through discussion with a third reviewer. The same process was applied to the subsequent review of full papers.

A PRISMA flow diagram indicating the numbers of studies included and excluded at each stage of the review is provided below (Figure 9). The PRISMA flow diagram from indication includes combined results from the original systematic review (2015) and updates in April 2016 and October 2016.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 66 of 315

Figure 9: PRISMA flow diagram of the study selection process for relapsed or refractory multiple myeloma patients (June 2015 original review, April and October 2016 updates)

Records identified through database searching (n=10,266) Duplicates removed (n=2,428) Records after duplicates removed (n=2,428)

==> picture [454 x 424] intentionally omitted <==

----- Start of picture text -----
Records screened for eligibility (n=7,838)
Full-text articles excluded, with
reasons (n=7,799)
 Animal study (n=10)
 Insufficient information (n=7)
 Intervention type (n=772)
 Outcome type (n=186)
 Population type (n=855)
Full-text article assessed for eligibility
 Protocol (n=8)
(n=39)
 Publication type (n=427)
 Report/ Discussion/ Erratum
(n=30)
 Review (n=120)
 Single-arm study (n=1,346)
 Study type (n=4,038)
Studies identified through other
Studies included for NICE NMA sources (n=2)
network (n=14 studies from 23
publications)  Tourmaline-MM1 trial CSR for
IA2 data cut (n=1)
 SMC DAD (n=1)
4.1.4 Results
SCREENING
ELIGIBILITY
INCLUDED
----- End of picture text -----

A systematic review of electronic databases was conducted on the 1[st] June 2015. An updated systematic review was conducted on 10[th] April 2016, and more recently on 7[th] October 2016. The search strategies used for each database are described in Appendix 2. References were downloaded into dedicated Reference Manager[®] databases. After de-duplication a total of 7,838 articles were identified specifically from these searches. These

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 67 of 315

abstracts were double reviewed for eligibility and 39 full-text articles were ordered for assessment (Figure 9). Following full paper review, 23 of these articles were identified as being eligible for inclusion into an NMA (Figure 8). Two articles (TOURMALINE-MM1 trial CSR[114] and the SMC detailed advice document for panobinostat[3] ) were identified from other sources that were subsequently obtained for full review, met our eligibility criteria and were also included (Figure 8). Reviewing of reference lists is a process which is recommended by the Cochrane Handbook for Systematic Reviews to ensure all relevant studies are identified.[115] In total, 23 full text articles reporting on 14 individual studies were identified for inclusion into the NICE NMA network (Figure 8). In the April 2016 update we identified the abstract from the Tourmaline-MM1 trial (Moreau et al., 2015),[116] along with the full publication (Moreau et al., 2016)[1] which was associated with the first data cut of the TOURMALINE-MM1 trial data (IA1).[42] We identified Hou et al., 2016[117] from the October 2016 update which was a regional expansion of the Tourmaline-MM1 trial conducted in China. As this study relates directly to ixazomib further details of methods and results are provided in Appendix 4. All three articles were included in the review.

Of the 23 articles from 14 individual studies to be included in the NMA, 18 articles were RCTs from 11 individual studies, 3 were observational studies, 1 study was a systematic literature review with 2 analyses (Pano + Bort + Dex versus either Len + Dex or Pom + Dex), and 1 study used data from the ixazomib CSR.

The 23 data sources from 14 studies are listed in Table 27 below.

A complete list of all excluded studies from the systematic review are also included in the systematic review reports.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 68 of 315

Table 27: Data sources of identified RCTs for ixazomib plus lenalidomide – dexamethasone and other treatments for the treatment of RRMM

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 69 of 315

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 70 of 315

4.1.5 Data sources

Table 28 presents the bibliographic details of the references for the one identified study of ixazomib in patients with RRMM. The SLR identified the Clinical Study Report (CSR), an abstract presented at the American Society of Hematology (ASH) in 2015, and a manuscript published in 2016. The study design, procedures and results described in the following sections are from the publication by Moreau et al, 2016,[1] supplemented using the CSR[42] where needed (data from the abstract was not used as it was more comprehensively reported in the manuscript; it was also not included in the NMA to prevent double counting).

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 71 of 315

Table 28: Bibliographic details of the ixazomib RCT in patients with RRMM

4.1.6 Reference list for excluded studies

The China continuation of Study C16010 was a regional extension of the TOURMALINEMM1 study (n=115) that evaluated the safety and efficacy of IXA+LEN+DEX versus LEN+DEX in adult RRMM patients from China.

Compared with the TOURMALINE-MM1 study, patients in the China continuation study had more advanced disease:

 more patients had Durie-Salmon stage IIIa myeloma (63% vs 38%), Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 72 of 315

• more heavily pre-treated (38% and 17% had received 2 or 3 prior lines of treatment vs 29% and 10%),

• more frequently received prior thalidomide (84% vs 45%),

• more frequently had refractory MM (53% vs 11%)

• more frequently had thalidomide-refractory MM (63% vs 12%)

These patient characteristics are consistent with a large retrospective analysis of outcomes of Chinese patients with MM[125] (ref).

The efficacy and safety results seen in the China continuation study are consistent with those of the TOURMALINE-MM1 study, and therefore support the activity of IXA+LEN+DEX across different study populations.

However, due to the divergence in ethnic, disease and treatment characteristics versus the global TOURMALINE-MM1 population, the results of the China continuation study have not been used in the economic model presented in Section 5 of this submission.

An overview of the China continuation study and results is provided in Appendix 3.

4.2 List of relevant randomised controlled trials

4.2.1 List of relevant randomised controlled trials

The TOURMALINE-MM1 study identified in the SLR is presented below in Table 29.

Table 29: List of relevant RCTs

4.3 Summary of methodology of the relevant randomised controlled trials

4.3.1 Trial design

The TOURMALINE-MM1 study (C16010) is an on-going global, Phase III, randomised, double-blind, multicentre clinical trial evaluating the safety and efficacy of an all-oral combination of IXA+LEN+DEX versus LEN+DEX in patients with RRMM who have had 1-3 prior therapies.[1]

Patients were randomly assigned in a 1:1 ratio to receive either oral ixazomib 4 mg or matching placebo capsule on days 1, 8, and 15, plus oral lenalidomide 25 mg on days 1-21 (10 mg for patients with creatinine clearance ≤60 or ≤50 mL/min, depending on local practice) and oral dexamethasone 40 mg on days 1, 8, 15, and 22, in 28-day cycles (Figure 10).

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 73 of 315

Randomisation was stratified by number of prior therapies (1 vs. 2 or 3), previous proteasome inhibitor exposure (naïve vs. exposed), and International Staging System disease stage (ISS I or II vs. III).[1] Treatment continued until disease progression or unacceptable toxicity.

Figure 10: TOURMALINE-MM1 study design with treatment schedule

==> picture [452 x 189] intentionally omitted <==

Source: TOURMALINE-MM1 CSR[42] (ref)

Patients who provided written informed consent received a country-, site-, and patientspecific enrolment code[42] Central randomisation using an interactive voice response system (IVRS) was used, with the randomisation scheme generated by an independent statistician at the sponsor, who was not on the study team. Patients were randomised strictly sequentially at each study centre as they became available for randomisation. Randomisation codes were not re-used from patients who discontinued.

In this double-blind study, all study personnel, including the investigators, site personnel, study clinicians, sponsor, and participants, were blinded to treatment assignments.[42] Only the independent statistical centre and Independent Data Monitoring Committee had, at prespecified interim analysis and interim safety review time points, access to un-blinded individual patient data.

4.3.2 Eligibility criteria

The inclusion criteria were chosen to select patients with RRMM on the basis of standard criteria and measurable disease. The exclusion criteria were chosen to remove from participation any patients who were refractory to lenalidomide or proteasome inhibitor-based therapy at any line, patients who were too ill or otherwise could not potentially benefit from the treatment, and patients with conditions that might confound assessments. The eligibility criteria are described in Table 30

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 74 of 315

Table 30: Eligibility criteria

Note: See table footnotes for abbreviations

Inclusion criteria

• Male or female, age ≥18 years

• MM diagnosed per standard criteria either currently or at initial diagnosis (initial diagnosis must have been symptomatic MM; relapsed disease did not have to be symptomatic)

• Measurable disease, defined as at least one of: serum M-protein ≥1 g/dL, urine M-protein ≥200 mg/24 hrs, serum Free Light Chain (FLC) assay – involved FLC level ≥10 mg/dL, provided that serum FLC ratio was abnormal

• ECOG PS 0–2 (a scale from 0 to 5 where 0 is asymptomatic and increasing numbers indicate increasing tumour-related disability)

• RRMM after 1–3 prior therapies, including:

• Patients who relapsed but were not refractory to previous treatments

• Patients who were refractory to all lines of previous treatment (i.e. patients who had never responded)

• Patients who relapsed from at least one previous treatment and were refractory to at least one previous treatment

• ANC ≥1000/mm[3] , platelet count ≥75,000/mm[3]

• Total bilirubin ≤1.5 x ULN, ALT and AST ≤3 x ULN

• Creatinine clearance ≥30 mL/min

• No active GVHD in patients with prior allo-SCT[†]

• Female patients: post-menopausal for ≥24 months prior to screening; or surgically sterile; or if they were of childbearing potential, negative pregnancy test within 10–14 days and again within 24 hours of start of cycle 1 of lenalidomide – agreed to practice true abstinence or use two reliable methods of birth control, agreed to ongoing pregnancy testing, and adhered to guidelines of RevAssist program (US) or equivalent local programme[†]

• Male patients, even if surgically sterilised: agreed to practice true abstinence; or agreed to practice effective barrier contraception during the entire study treatment period and 90 days after the last dose of study treatment if their partner was of childbearing potential, even if they had a successful vasectomy; and adhered to guidelines of RevAssist program (US) or equivalent local programme[†]

• Able to take concurrent aspirin 81–325 mg daily (or enoxaparin 40 mg subcutaneously daily [or its equivalent] if allergic to aspirin), per published standard or institutional standard of care, as prophylactic anticoagulation[†]

• For patients with prior history of deep vein thrombosis (DVT), low-molecular-weight heparin (LMWH) was mandatory[†]

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 75 of 315

• Voluntary written consent; willing and able to adhere to study visit schedule and other protocol requirements

Exclusion criteria

• Refractory to lenalidomide or PI-based therapy at any line (refractory disease defined as disease progression on treatment or progression within 60 days after last dose)

• Patients refractory to thalidomide-based therapy were eligible

• Evidence of current uncontrolled cardiovascular conditions, including uncontrolled hypertension, uncontrolled cardiac arrhythmias, symptomatic congestive heart failure, unstable angina, or myocardial infarction within 6 months before randomisation in the study

• Failure to have fully recovered (grade ≤1 toxicity) from effects of prior chemotherapy (except alopecia) regardless of interval since last treatment

• Psychiatric illness/social situation that would limit compliance with study requirements.

• Major surgery within 14 days before randomisation

• Radiotherapy within 14 days before randomisation

• Infection requiring systemic antibiotic therapy or other serious infection within 14 days before randomisation

• Central nervous system involvement

• Diagnosis of Waldenstrom’s macroglobulinemia, POEMS syndrome, plasma cell leukaemia, primary amyloidosis, myelodysplastic syndrome, or myeloproliferative syndrome[†]

• Female patients who were breast-feeding or pregnant[†]

• Ongoing or active systemic infection, active hepatitis B or C virus infection, or known human immunodeficiency virus positive

• Comorbid systemic illnesses or other severe concurrent disease which, in the judgment of the investigator, would make the patient inappropriate for entry into this study or interfere significantly with the proper assessment of safety and toxicity of the prescribed regimens (e.g., peripheral neuropathy that is Grade 1 with pain or Grade 2 or higher of any cause).

• Systemic treatment with strong inhibitors of cytochrome P450 (CYP) 1A2 (CYP1A2) (fluvoxamine, enoxacin, ciprofloxacin), strong inhibitors of CYP3A (clarithromycin, telithromycin, itraconazole, voriconazole, ketoconazole, nefazodone, posaconazole) or strong CYP3A inducers (rifampin, rifapentine, rifabutin, carbamazepine, phenytoin, phenobarbital), or use of Ginkgo biloba or St. John’s wort within 14 days before randomisation in the study

• Diagnosed or treated for another malignancy within 2 years before randomisation or previously diagnosed with another malignancy and any evidence of residual disease. Patients with nonmelanoma skin cancer or carcinoma in situ of any type were not excluded if they had undergone complete resection

• Inability to swallow oral medication, inability or unwillingness to comply with the drug administration requirements, or gastrointestinal condition that could interfere with the oral absorption or tolerance of treatment[†]

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 76 of 315

 Known allergy to any of the study medications, their analogues, or excipients in the various formulations of any agent[†]

Abbreviations: Allo-SCT = allogeneic stem cell transplant; ALT = alanine aminotransferase; ANC = absolute neutrophil count; AST = aspartate aminotransferase; ECOG = Eastern Cooperative Oncology Group; FLC = free light chain; GVHD = graft-versus-host disease; PI = proteasome inhibitor; PS = performance status; RRMM = relapsed and/or refractory multiple myeloma; ULN = upper limit of normal

Source: Moreau, 2016[1] ;[† ] TOURMALINE-MM1 CSR[42]

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 77 of 315

4.3.3 Settings and locations where the data were collected

The study was conducted at 147 investigative centres globally in 26 countries shown in Table 31, and included nine centres from across the UK (one was inactive), from which 21 patients were recruited and included in the study.

Table 31: TOURMALINE-MM1 study locations

4.3.4 Trial drugs and concomitant medications

4.3.4.1 Trial drugs

Patients received oral ixazomib 4 mg or matching placebo capsule on days 1, 8, and 15, plus oral lenalidomide 25 mg on days 1-21 and oral dexamethasone 40 mg on days 1, 8, 15, and 22, in 28-day cycles[1] Patients were to continue to receive treatment until disease progression or unacceptable toxicity, whichever occurred first[1] Dose adjustments for toxicities were permitted using established dose-modification guidelines per the protocol/prescribing information for each drug[1]

4.3.4.2 Permitted and disallowed concomitant medications

Permitted and disallowed concomitant medications were as follows[1] . Thromboprophylaxis according to the American Society of Clinical Oncology (ASCO) guidelines or institutional standard of care was required to prevent thromboembolic complications that may occur with lenalidomide-based regimens, e.g., aspirin (81–325 mg orally once daily) or low-molecular weight heparin (equivalent to enoxaparin 40 mg subcutaneous per day) depending on patient risk factors. Prophylactic antiviral therapy was permitted as clinically indicated.

All necessary supportive care consistent with optimal patient care per local standards was available to patients. Use of myeloid growth factors (e.g., granulocyte colony stimulating factor, granulocyte macrophage-colony stimulating factor) and erythropoietin was allowed, with use of erythropoietin to be minimised as much as possible given potential risk of deep vein thrombosis with lenalidomide. Red blood cell and platelet transfusions were given as clinically indicated. Standard anti-emetics including 5-hydroxytryptamine 3 serotonin receptor (5-HT3) antagonists were recommended for emesis if it occurred once treatment was initiated; prophylactic anti-emetics were also permitted at the physician’s discretion. Topical, intravenous, or oral antihistamines or steroids were permitted to manage rash. Concomitant treatment with bisphosphonates was also permitted.

Systemic treatment with strong CYP1A2 inhibitors (fluvoxamine, enoxacin, and ciprofloxacin) or strong CYP3A inhibitors (clarithromycin, telithromycin, itraconazole, voriconazole, ketoconazole, nefazodone, and posaconazole), or use of Ginkgo biloba or St John’s wort

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 78 of 315

was not permitted. Systemic treatment with strong CYP3A inducers (rifampin, rifapentine, rifabutin, carbamazepine, phenytoin, and phenobarbital) was to be avoided. Radiation therapy or any anti-neoplastic treatment with activity against multiple myeloma, other than the study drugs, was not permitted.

4.3.5 Primary and secondary outcomes

The primary, key secondary and other secondary outcomes in the TOURMALINE-MM1 study are described in

Table 32 .

The outcomes of interest in the NICE scope include PFS, overall survival (OS), response rates, time to next treatment, adverse effects of treatment and health-related quality of life . All of these have been reported in the TOURMALINE-MM1 study (shown in bold in

Table 32 ) except for time to next treatment . However, the study did collect data on the following parameters that may be useful surrogates for time to next treatment : time to progression; time to response; and duration of response. The results of all the relevant outcomes have been presented in the clinical effectiveness results (Section 4.7) and adverse events section (Section 4.12).

Table 32: TOURMALINE-MM1 study outcomes

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 79 of 315

Endpoint Association between response or resistance to ixazomib treatment and tumour gene expression patterns including NFκB and protein synthesis signatures Mechanisms of treatment-emergent resistance, such as somatic mutations in proteasome subunits, in tumours that initially respond to therapy and then exhibit progressive disease Development of new or worsening of existing selected skeletal-related events, defined as new fractures (excluding vertebral compression or rib fractures), irradiation of or surgery on bone, or spinal cord compression from baseline through the development of progressive disease Abbreviations: CR = complete response; EORTC QLQ-C30 and MY-20 = the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) and the Quality of Life Questionnaire MM Module 20 (MY-20); EQ-5D = EuroQol-5D; IMWG = International Myeloma Working Group; IRC = independent review committee; OS = overall survival; PFS = progression free survival; VGPR = very good partial response BOLD black = The outcomes of interest in the NICE scope that are reported in the TOURMALINE-MM1 study BOLD grey = The outcomes that may be useful surrogates for the outcome in the NICE scope of time to next treatment a The final statistical analysis plan (SAP) excluded del(13) and 1q (1q21+) from the high-risk evaluations because during the course of the study, the definition of high-risk abnormalities for MM evolved and 1q21+ was considered intermediate-risk. Source: Moreau, 2016[1] ;[† ] TOURMALINE-MM1 CSR[42]

Health-related quality of life was evaluated through patient self-reported instruments including the EORTC QLQ-C30 and MY-20 questionnaires.

Response assessments were performed every cycle (i.e. every 28 days) until disease progression.[1] Response and disease progression assessments were based on central laboratory results and International Myeloma Working Group (IMWG) 2011 criteria (Table 33) as evaluated by the independent review committee (IRC) blinded to both patient assignment and investigator assessment. All patients were followed for survival after disease progression (every 12 weeks until death or termination of the study).

Table 33: International Myeloma Working Group (IMWG) Uniform Response Criteria

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 80 of 315

4.3.6 Summary of the methodology of the TOURMALINE-MM1 RCT

A summary of the methodology of the study is shown in Table 34.

Table 34: Summary of the methodology of the TOURMALINE-MM1 study

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 81 of 315

4.4 Statistical analysis and definition of study groups in the relevant randomised controlled trials

4.4.1 Analysis sets

The intent-to-treat (ITT) population[1] : defined as all patients who were randomised. Patients were analysed according to the treatment they were randomised to receive,

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 82 of 315

regardless of any errors in dosing. The ITT population was used for all primary and secondary efficacy analyses.

The safety population[1] : defined as all patients who received at least 1 dose of any study drug. Patients were analysed according to the treatment actually received, regardless of which treatment they were randomised to receive. Patients who received any dose of ixazomib were included in the IXA+LEN+DEX group and patients who did not receive any dose of ixazomib were included in the LEN+DEX group, regardless of their randomised treatment. The Safety population was used for all safety-related analyses such as AEs, concomitant medications, laboratory tests, and vital signs.

4.4.2 Hypotheses

The study was designed to assess the efficacy and safety of weekly IXA+LEN+DEX versus LEN+DEX in patients with RRMM.[1,42] The primary objective was to determine whether the addition of once-weekly oral ixazomib to the background therapy of lenalidomide and dexamethasone improves PFS in patients with RRMM.[1,42]

4.4.3 Sample size calculations

The total sample size of the study was planned to be approximately 703 patients (actual accrual: 722 patients) in order to provide sufficient power to adequately evaluate both PFS and OS.[42] Total sample size was calculated to provide 80% power (2-sided alpha 0.05) to test for a 30% improvement in OS (assumed hazard ratio of 0.77)[1] . The study was powered to demonstrate PFS superiority (hazard ratio 0.74).[1] An O’Brien-Fleming stopping boundary for efficacy was calculated using a Lan-DeMets alpha-spending function[128] based on the observed number of events at the data cut-off date.[1] An interim analysis was planned when approximately 36% of patients experienced a PFS event[1]

4.4.4 Interim analyses and stopping guidelines

Three sequential interim analyses plus a final analysis were planned (Figure 11)[1] The first interim analysis was planned when approximately 36% of patients experienced a PFS event 1 At the first, pre-planned analysis (data cut-off: 30th October 2014; median follow-up ~15 months), the PFS results crossed the pre-specified O’Brien-Fleming boundary demonstrating a statistically significant benefit of the IXA+LEN+DEX regimen versus LEN+DEX; therefore, consistent with the statistical methodology, this was the final statistical analysis of PFS. Per protocol, the study continued in a double-blind manner to gain more mature OS data; a second pre-planned interim analysis (data cut-off: 12[th] July 2015; median follow-up of ~23 months) was conducted for OS.

The clinical effectiveness results (Section 4.7) presents data from both these interim analyses; safety data from the second interim analysis is presented in Section 4.12. The study continues in a double-blind, placebo-controlled manner to obtain more mature OS data. A third interim analysis is expected in Q2 2017 with a final analysis to follow later.

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 83 of 315

Figure 11: Statistical assumptions in the TOURMALINE-MM1 study schedule

==> picture [387 x 240] intentionally omitted <==

FPI = first patient in; IA = interim analysis; OS = overall survival; PFS = progression-free survival Source: TOURMALINE-MM1 CSR[42]

4.4.5 Statistical methods used to compare groups

Primary and secondary outcomes

The study used a closed sequential testing procedure for the primary endpoint of PFS and the key secondary endpoints OS and OS in patients carrying del(17p) (2-sided alpha 0.05; test for overall survival conducted on its own alpha-spending functions only if PFS was significant)[1] Kaplan-Meier methodology was used to estimate time-to-event distributions, with stratified log-rank tests and Cox models (alpha=0.05, two-sided) used for inter-arm comparisons of time-to-event endpoints.[1]

Safety assessment

Adverse events (AEs) were coded using MedDRA version 16.0.[42] Treatment-emergent AEs (TEAEs) were defined as AEs that occurred after administration of the first dose of any study drug and through 30 days after the last dose of any study drug. AEs were summarised as per the National Cancer Institute’s Common Terminology Criteria for Adverse Events (NCI CTCAE) version 4.03.

4.4.6 Handling of missing data and withdrawals

The ITT population was used for all primary and secondary efficacy analyses.[1] The primary endpoint was PFS, defined as the time from the date of randomisation to the date of first documentation of progressive disease based on central laboratory results and International Myeloma Working Group (IMWG) criteria as evaluated by an independent review committee, or death due to any cause, whichever occurred first. Patients without documentation of progressive disease were censored at the date of the last response assessment that was stable disease or better.[42] The details regarding the handling of missing assessments and censoring for the PFS analysis are presented in Table 35

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 84 of 315

Table 35: Handling of missing assessments and censoring for PFS primary analysis based on FDA guidance

4.4.7 Subgroup analyses

Pre-specified subgroup analyses of PFS and OS were conducted. Subgroup analyses were performed relative to baseline stratification factors, demographics, disease characteristics, and number and types of prior therapy. They included subgroups by prior line of therapy, which are of relevance to the NICE scope (Table 1) i.e. those with 1 versus 2 or 3 prior therapies. Other factors included: cytogenetic abnormalities, ISS stage, age, sex, race, region, Western country, prior therapy, relapsed (and/or) treatment-refractory, ECOG performance status, and baseline creatinine clearance. A stratified Cochran-MantelHaenszel Chi-squared test was used to assess inter-arm differences in response rates.[1]

4.4.8 Summary of the statistical analysis used in the primary analysis

A summary of the statistical analysis is shown in Table 36

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 85 of 315

Table 36: Summary of the statistical analysis

4.5 Participant flow in the relevant randomised controlled trials

4.5.1 Participant flow

Patient disposition in the TOURMALINE-MM1 study is summarised in the CONSORT diagram in Figure 12 There were no unexpected imbalances in drop-outs between treatment groups: at data cut-off for the second interim analysis (July 2015), 136 (38%) and 133 (37%) patients in the IXA+LEN+DEX and LEN+DEX arms, respectively, remained on treatment, and 222 (62%) and 229 (63%), respectively had discontinued. .[1] The two most common reasons for discontinuation were progressive disease (34% and 40% of patients) and adverse events (17% and 14%). Two of the 360 patients randomised to IXA+LEN+DEX did not receive any study treatment, and three of the 362 patients randomised to LEN+DEX accidentally received limited dosing of ixazomib, and were therefore conservatively included within the IXA+LEN+DEX group for analyses of exposure and safety. There were also no imbalances across regions and countries in terms of enrolment by treatment (Asia Pacific: 18% of IXA+LEN+DEX patients; 21% of LEN+DEX patients; Europe: 69% and 65%; North America: 13% and 14%).[42]

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 86 of 315

Figure 12: TOURMALINE-MM1 patient disposition – CONSORT diagram

==> picture [378 x 505] intentionally omitted <==

Source: Moreau, 2016[1]

4.5.2 Patient characteristics at baseline

The demographics and baseline characteristics for patients randomised to the IXA+LEN+DEX and LEN+DEX arms in TOURMALINE-MM1 are summarised in Table 37. Disease characteristics were well balanced between the two arms, notably in terms of the randomisation stratification factors, as would be expected, and other prognostic factors such as age, renal impairment (creatinine clearance of <60 mL/min), cytogenetic risk categories, and refractory disease status.

Of relevance for the NICE scope, 59% (n=425) of patients had received 1 prior line and 41% (n=297) had received 2 or 3 lines of prior therapy (based on stratification factors).Overall, 12% (87) had stage III disease and 19% (137) had high-risk cytogenetics (defined as

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 87 of 315

del[17], t[4;14], t[14;16]). In addition, 70% of the patient population had been exposed to prior PI therapy and 55% had been exposed to prior IMiD therapy. Of the 397 patients who had prior exposure to IMiD therapy, 41 of 193 IXA+LEN+DEX regimen patients (21%) and 50 of 204 LEN+DEX regimen patients (25%) were refractory to any prior IMiD therapy, with all but 1 patient noted as refractory to thalidomide. Of the 502 patients who had prior exposure to PI therapy, 4 of 249 the IXA+LEN+DEX patients (1%) and 8 of 253 the LEN+DEX patients (2%) were refractory to any prior PI therapy (the study had been designed to exclude patients who were refractory to PI-based therapy, however a few were included). Overall, 46 patients (24 in the IXA+LEN+DEX arm [7%] and 22 in the LEN+DEX arm [6%]) were primary refractory (i.e. had never responded to any prior therapy [best response of stable disease or progressive disease]).

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 88 of 315

Table 37: TOURMALINE-MM1 study baseline patient characteristics

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 89 of 315

4.6 Quality assessment of the relevant randomised controlled trials

In order to assess the risk of bias and generalisability of the TOURMALINE-MM1 study, a quality assessment was conducted using guidance from ‘Systematic reviews: CRD's guidance for undertaking reviews in health care (University of York Centre for Reviews and Dissemination)’.[129]

A summary of the quality assessment is provided in Table 38, showing that the TOURMALINE-MM1 study is a high quality RCT with an overall low risk of bias.

Table 38: Quality assessment of the TOURMALINE-MM1 study

4.7 Clinical effectiveness results of the relevant randomised controlled trials

In Section 4.7.1 the outcomes from the entire ITT population (i.e all patients who were randomised [n=722]) are presented. In Section 4.8 we present the results of the subgroup analysis of patients who received 2 or 3 prior lines of therapy, which is of relevance to the NICE scope. Other pre-specified subgroups based on a range of factors including age, cytogenetics, ISS stage and relapsed and/or refractory disease are shown in Section 4.8.3.

To date there have been two interim analyses from the TOURMALINE-MM1 study. At data cut-off for the first analysis (30 October 2014), median follow-up was 14.8 and 14.6 months

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 90 of 315

in the IXA+LEN+DEX and LEN+DEX groups, respectively. As the primary endpoint of improving PFS was met at the first analysis, this was the final statistical analysis of this endpoint. We have presented clinical data from both interim analyses. Adverse events from the longer duration of follow-up second interim analysis , presented for the whole safety population and the subgroups based on prior therapy are presented in Section 4.12.

4.7.1 Clinical effectiveness results in the entire ITT population

4.7.1.1 Primary endpoint: Progression-free survival (PFS)

First interim analysis

At data cut-off for the first analysis there were 129 and 157 IRC-assessed progression or death events in the IXA+LEN+DEX and LEN+DEX groups, respectively. There was a significant 35% improvement in PFS with IXA+LEN+DEX versus LEN+DEX (hazard ratio [HR] 0.74 [95% confidence interval, 0.59 to 0.94], p=0.012) (Figure 13). There was a clinically meaningful ~6 month improvement in median PFS (20.6 versus 14.7 months) in favour of the IXA+LEN+DEX arm. As the primary endpoint of improving PFS was met, this was the final statistical analysis of this endpoint.

Figure 13: Kaplan-Meier distribution of PFS with IXA+LEN+DEX versus LEN+DEX

==> picture [405 x 259] intentionally omitted <==

Source: Moreau, 2016[1]

Second interim analysis

A non-inferential PFS analysis was conducted at a median follow up of 23 months with 372 PFS events. The HR of PFS was 0.82 (95% confidence interval [0.67, 1.0]) for IXA+LEN+DEX versus LEN+DEX, and estimated median PFS was 20 months in the IXA+LEN+DEX group and 15.9 months in the LEN+DEX group.

4.7.1.2 Key secondary endpoints: overall survival/ overall survival in high-risk patients carrying del(17p)

At both the first and second interim analyses, OS data were not yet mature. At the first interim analysis, only 107 (22%) of the pre-specified 486 deaths required for the final OS analysis had occurred (51 in the IXA+LEN+DEX arm and 56 in the LEN+DEX arm). Early data showed a survival trend in favour of IXA+LEN+DEX (HR 0.900; 95% CI 0.62-1.32) with

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 91 of 315

18-month survival rates of 83% versus 80%, respectively.[42] At the second interim analysis, there were 171 deaths (81 in the IXA+LEN+DEX group, 90 in the LEN+DEX group), which represents 35% of the pre-specified number of deaths required for final analysis of OS. The median OS was not reached in either group.[1]

A total of 69 patients had a high-risk del(17p) chromosome abnormality. At the first interim analysis, 13 had died, including 4 of 36 (11%) patients in the IXA+LEN+DEX arm and 9 of 33 (27%) patients in the LEN+DEX arm. There was a 49% reduction in the risk of death with IXA+LEN+DEX (HR 0.506), with 18-month survival rates of 86% and 67%, respectively. At the second interim analysis, 9 of 36 IXA+LEN+DEX regimen patients (25%) and 15 of 33 LEN+DEX regimen patients (45%) had died. OS in patients with del(17p) showed a 51% reduction in the risk of death for patients treated with IXA+LEN+DEX (HR=0.487). The median OS was not reached with the IXA+LEN+DEX regimen and was 30.9 months with the LEN+DEX regimen.

In conclusion, OS data are still immature, although the current data show a survival trend in favour of the IXA+LEN+DEX regimen for both the ITT population and the high-risk subgroup of patients whose tumour harbored del(17p). The study is continuing in a double-blind, placebo-controlled fashion and OS data from the third interim analysis is expected in Q2 2017.

4.7.1.3 Response rates, duration of response and time to progression

First interim analysis

IXA+LEN+DEX resulted in a significantly higher overall response rate (78.3% versus 71.5% [p=0.04]), including significantly higher rates of ≥VGPR and CR, a significantly shorter time to response, and a longer median duration of response versus LEN+DEX (Table 39).

Table 39: Best confirmed treatment responses (blinded IRC assessment) and time to progression in the ITT population (15-month analysis)

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 92 of 315

Reflecting the findings for the primary endpoint of PFS, time to progression (TTP) was also significantly longer in the IXA+LEN+DEX versus LEN+DEX arm (Table 39; Figure 14 )

Figure 14: Kaplan-Meier plot of TTP in the ITT population. (Data from final statistical analysis for PFS)

==> picture [387 x 231] intentionally omitted <==

ITT = intent to treat; PFS = progression-free survival; TTP = time to progression Source: Moreau, 2016[1]

The evolution of response rates over the course of treatment is shown in, which highlights the increasing proportions of higher quality responses seen with increasing time on study treatment. Responses were rapid and durable (Table 39) and deepening responses were noted with increasing treatment duration (Figure 15).

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 93 of 315

Figure 15: Cumulative best responses over time in the ITT population. A. IXA+LEN+DEX; B. LEN+DEX

A

==> picture [372 x 217] intentionally omitted <==

B

==> picture [372 x 217] intentionally omitted <==

Abbreviations: CR = complete response; PR = partial response; VGPR = very good partial response Moreau, 2016[1]

Second interim analysis

Consistent with the first interim analysis, response rates were improved with the IXA+LEN+DEX regimen, with a shorter time to response, and a longer median duration of response and TTP.[42] The ORR was 78.6% in the IXA+LEN+DEX group and 73.2% in the LEN+DEX group. The IXA+LEN+DEX regimen delayed the median TTP by approximately 5 months, a clinically meaningful improvement (Table 40 ).

Company evidence submission for Ixazomib citrate in combination with lenalidomide and dexamethasone for relapsed or refractory multiple myeloma ID 807 Page 94 of 315

Table 40: Best confirmed treatment responses (blinded IRC assessment) and TTP in the ITT population (23-month analysis)