TA891 · STA

Ibrutinib with venetoclax for untreated chronic lymphocytic leukaemia

Recommended with restrictionsTechnology appraisal committee C

Only if the companies provide both drugs according to the commercial arrangements (simple discount patient access scheme for both ibrutinib and venetoclax, with confidential discounts)

Source documents

Intervention

ibrutinib with venetoclax
· BTK inhibitor with BCL2 inhibitor

Condition

chronic lymphocytic leukaemiahaematology · untreated

Comparators

NameType Established Committee preferred
fludarabine, cyclophosphamide and rituximab (fcr)active drugYes
bendamustine and rituximab (br)active drugYes
acalabrutinib and venetoclaxactive drug
obinutuzumab plus chlorambucilactive drugYesYes
venetoclax plus obinutuzumabactive drugYesYes
acalabrutinibactive drugYesYes
ibrutinibactive drugYesYes

Clinical trials

TrialDesignPhasePivotal
GLOWRCTPhase IIIYes
E1912RCT
CAPTIVATERCTYes
RESONATEobservational
ELEVATE-TNRCT
CLL14RCT

Economic model

markov (company)
Time horizon: 30 to 40 years

Methodological decisions (9)

comparator selection

Committee noted implementing 'FCR or BR suitability' criterion challenging for NHS clinicians; placed greater weight on unsuitable group results where comparators (acalabrutinib, obinutuzumab+chlorambucil, venetoclax+obinutuzumab) more relevant than FCR.

Company: Presented modelling for multiple populations

ERG: Appropriate to focus on FCR/BR unsuitable comparisons

Committee: Greater weight on FCR/BR unsuitable and high-risk group results; comparators more relevant for NHS England

ICER impact: uncertain_direction

cure assumption

Model implied cure of CLL for ibrutinib plus venetoclax and acalabrutinib arms when progression risk fell to 0% due to background mortality exceeding disease hazards.

Company: Age at cap consistent; standard methodology for modelling

ERG: CLL is not curable; treatments aim to maintain remissions; inconsistent across groups

Committee: Acknowledged limitations; scenarios addressing inconsistency acceptable; model outcomes appropriate

ICER impact: negligible

indirect comparison method

For FCR or BR unsuitable group: use of anchored MAICs to compare ibrutinib plus venetoclax against acalabrutinib and venetoclax plus obinutuzumab. EAG noted inconsistency between ATT and ATC results and non-proportional hazards.

Company: Presented scenario analysis with time-varying hazard ratios and anchored MAICs acceptable; clinical equivalence of acalabrutinib to ibrutinib

ERG: Cautious about applying GLOW follow-up (immature) to 30-40 year horizon; concerns about proportional hazards assumption; requested acalabrutinib vs ibrutinib clinical equivalence validation

Committee: ATC approach suitable; anchored MAICs acceptable in absence of direct evidence and more mature data; clinical equivalence between acalabrutinib and ibrutinib plausible and acceptable

ICER impact: uncertain_direction

model structure

Semi-Markov model limited to exponential distributions for transitions from second-line and post-progression states. Tunnel states used only for cost tracking, not health state occupancy.

Company: Semi-Markov model structure appropriate; exponential distribution good fit to RESONATE data

ERG: Appropriate model structure but limitations prevent exploration of alternative survival distributions

Committee: Model structure adequate despite limitations

ICER impact: negligible

mortality assumption

Company estimated first-line progression hazard by subtracting general population mortality from overall hazards, resulting in 0% progression risk after certain age. In FCR/BR unsuitable group earlier than suitable group due to higher background mortality.

Company: Age at progression-free survival cap (~85 years) consistent across groups

ERG: Method created inconsistencies; FCR/BR unsuitable group reached 0% risk much earlier, implying cure; data does not support differential progression risk between groups

Committee: Model structure and outcomes remain appropriate despite limitations; company's scenario with capped progression probability in unsuitable group acceptable

ICER impact: negligible

population generalisability

High-risk CLL group: ELEVATE-TN included high-risk patients but GLOW excluded them. Company assumed efficacy results from non-high-risk could apply to high-risk population.

Company: FCR or BR unsuitable indirect comparison results applicable to high-risk; acalabrutinib clinically equivalent to ibrutinib; referred to previous NICE acceptance of ibrutinib efficacy assumption

ERG: Cautious about applying non-high-risk efficacy to high-risk population; clinical outcomes optimistic

Committee: Clinical equivalence between acalabrutinib and ibrutinib plausible in high-risk CLL population; acceptable for decision making

ICER impact: increases

treatment effect duration

Company assumed treatment effect of ibrutinib plus venetoclax maintained for lifetime (30-40 years). EAG cited immature clinical data and requested waning scenarios.

Company: Treatment effect maintained for lifetime horizon

ERG: Concern about immature data supporting lifetime assumption; requested treatment effect waning scenarios

Committee: Acknowledged waning scenarios; 5-year waning conservative; considered all waning scenarios in decision making

ICER impact: decreases

utility source

Progression-free first-line utility: company mapped EQ-5D-5L from GLOW to EQ-5D-3L.

Company: Value consistent with CLL14 and ELEVATE-TN; ran capping scenario

ERG: Value higher than UK population age-sex matched utility; therefore overestimate

Committee: Prefer capped UK utility values; agreed with clinical experts that quality of life post-CLL treatment lower than general population

ICER impact: increases

utility value choice

Progression-free second-line and progressed state utilities: company used 0.6 from Holzner et al. 2004 for both states.

Company: Used 0.6 from Holzner, age-adjusted to trial populations

ERG: Only 0.6 appropriate for progressed state; for second-line, preferred utility multiplier approach using GLOW EQ-5D ratios to adjust first-line value

Committee: Agreed 0.6 underestimate; EAG multiplier approach appropriate; acknowledged limited impact on final outcomes

ICER impact: increases

Evidence gaps

immature overall survivalGLOW trial had immature clinical data; committee concerned about extrapolating to 30-40 year model horizon
no direct comparisonNo direct head-to-head evidence for ibrutinib plus venetoclax versus most comparators; relied on anchored MAICs
single arm evidence onlyNo additional direct evidence for high-risk CLL population; relied on assumption that FCR/BR unsuitable population results apply

Commercial arrangement

simple discount pas · confidential · critical for recommendation

Special considerations

Innovation acknowledged