TA243 · STA

Rituximab for the first-line treatment of stage III–IV follicular lymphoma (review of NICE technology appraisal guidance 110)

RecommendedOctober 2011

Recommended in combination with CVP, CHOP, MCP, CHVPi, or chlorambucil for symptomatic stage III-IV follicular lymphoma in previously untreated patients

Source documents

Interventions

rituximab (MabThera)
chimeric monoclonal antibody · targets the CD20 surface antigen of mature B-cell lymphocytes · intravenous infusion
rituximab plus cvp
rituximab plus chop
rituximab plus mcp
rituximab plus chvpi
rituximab plus chlorambucil

Conditions

follicular lymphomahaematology · locally_advanced
advanced follicular lymphomahaematology · metastatic
stage iii-iv follicular lymphomahaematology · metastatic
stage iii–iv follicular lymphomahaematology · locally_advanced, metastatic

Comparators

NameType Established Committee preferred
cvp aloneactive drugYes
chop aloneactive drugYes
mcp aloneactive drugYes
chvpi aloneactive drugYes
chlorambucil aloneactive drugYes
cvpactive drugYes
chopactive drugYes
mcpactive drugYes
chvpiactive drugYes
chlorambucilactive drugYes
cvp (cyclophosphamide, vincristine and prednisolone)active drugYes
chop (cyclophosphamide, doxorubicin, vincristine and prednisolone)active drugYes
mcp (mitoxantrone, chlorambucil and prednisolone)active drugYes
chvpi (cyclophosphamide, doxorubicin, etoposide, prednisolone and interferon-α)active drugYes

Clinical trials

TrialDesignPhasePivotal
M39021RCTIIIYes
GLSG-2000RCTIIIYes
OSHO-39RCTIIIYes
FL2000RCTIIIYes
EORTC 20981
Four randomised controlled trials comparing rituximab plus chemotherapy vs chemotherapy aloneRCTYes
Four good-quality RCTsRCTIIIYes

Economic model

markov (manufacturer)
Time horizon: lifetime
markov (company)
Time horizon: 25 years
Cycle length: not specified
markov (assessment_group)
partitioned survival (company, erg, assessment_group)
Time horizon: lifetime

ICER

£20,000–£30,000 (rituximab plus CHOP vs CHOP alone) · high uncertainty
£20,000–£30,000 (rituximab plus CVP/CHOP/MCP vs chemotherapy alone) · moderate uncertainty
Below £20,000 (rituximab plus CVP vs CVP; rituximab plus CHOP vs CHOP; rituximab plus MCP vs MCP) · moderate uncertainty

Methodological decisions (15)

comparator selection

Whether to recommend rituximab with any chemotherapy regimen or to restrict to those with clinical and economic evidence

Committee: Recommending rituximab with any chemotherapy is not appropriate despite likely clinical effectiveness, as it would result in recommending unappraisal combinations and cost-effectiveness cannot be assumed without evidence. However, rituximab plus chlorambucil recommended based on informal cost-effectiveness advice and likely patient benefit

ICER impact: uncertain_direction

comparator selection

Inclusion of rituximab plus chlorambucil despite limited clinical data and no formal cost-effectiveness analysis

Company: Not addressed separately

ERG: Provided informal cost-effectiveness advice that ICER for rituximab plus chlorambucil would remain cost-effective if QALY gain was half that of rituximab plus CHOP

Committee: Committee concluded rituximab plus chlorambucil was appropriate use of NHS resources for older patients or those with lower performance status

ICER impact: uncertain_direction

model structure

Assessment Group developed individual patient model simulating 100,000 patients with separation of responders and non-responders, differing from manufacturer's Markov approach.

Company: Used Markov model estimating costs and benefits from first-line treatment over lifetime

ERG: Developed individual patient model simulating 100,000 patients with responder/non-responder separation

ICER impact: uncertain_direction

population generalisability

Whether trial populations are representative of UK advanced follicular lymphoma patients

Committee: Trial population was younger than median age of UK patients, but results could be considered broadly representative of UK clinical practice outcomes because some rituximab combinations (e.g. rituximab plus CHOP) are given to younger patients and subgroup analysis by age showed benefit regardless of age

ICER impact: negligible

population generalisability

Uncertain relative effect and absolute response rates of rituximab added to chemotherapy regimens other than those studied in clinical trials (CHVPi and chlorambucil)

Company: Submitted cost-effectiveness for rituximab plus CHVPi

ERG: Did not include rituximab plus CHVPi due to trial design issues and limited UK use

Committee: Committee was persuaded that rituximab would provide additional clinical benefit when added to chemotherapy despite limited evidence for some regimens; accepted rituximab plus chlorambucil based on clinical specialist advice despite absence of formal cost-effectiveness analysis

ICER impact: uncertain_direction

proportional hazards

Assessment Group highlighted inconsistencies in manufacturer's model including derivation of transition probability, calculation of post-progression survival, and potential over-estimation of rituximab effect through use of time-to-event data confounded by subsequent treatments.

Company: Used time-to-event data from clinical trials with responders receiving subsequent treatments

ERG: Noted this may have over-estimated the effect of rituximab

ICER impact: decreases

survival extrapolation

For CHOP and MCP comparisons, Assessment Group used M39021 trial data as proxy for progression-free survival curves rather than first-line induction trial data, to avoid confounding by stem-cell transplantation or interferon maintenance therapy.

Company: Not explicitly stated

ERG: Used M39021 trial data as proxy for first-line progression-free survival curves for CHOP and MCP comparisons

ICER impact: uncertain_direction

treatment effect duration

Assessment Group tested assumption of 25% reduction in efficacy of rituximab when used as second-line treatment in patients previously treated with rituximab.

Company: Not explicitly stated

ERG: Explored scenario assuming 25% reduction in rituximab efficacy in second-line for previously-treated patients

ICER impact: increases

treatment effect duration

Manufacturer and Assessment Group did not include first-line rituximab maintenance treatment in base-case analyses

Company: Not included in base case

ERG: Not included in base case

Committee: Committee concluded these factors needed to be considered when making cost-effectiveness decision, and noted that ICERs increased when first-line maintenance was assumed

ICER impact: increases

treatment effect waning

Whether rituximab efficacy is maintained when patients are re-treated with rituximab after relapse

Company: No loss of efficacy on re-treatment

ERG: Uncertainty regarding whether loss of efficacy occurs on re-treatment; sensitivity analysis showed that 25% reduction in re-treatment efficacy would increase ICERs from £7720–£10,800 to £14,900–£26,900

Committee: Efficacy of rituximab after re-treatment is a key uncertainty in the economic modelling but uncertainty was not such that it increased ICERs above the threshold range

ICER impact: increases

treatment effect waning

Whether efficacy of rituximab is maintained when used again as second-line induction treatment after first-line rituximab

Company: Assumed efficacy maintained

ERG: Assumed efficacy maintained

Committee: Committee noted this assumption and concluded ICERs increased if there was a reduction in efficacy on re-treatment

ICER impact: increases

treatment sequencing

Assessment Group considered manufacturer's assumption that patients receive either CHOP or rituximab plus CHOP as second-line treatment did not reflect the range of treatments used in clinical practice.

Company: Assumed patients receive either CHOP or rituximab plus CHOP as second-line treatment

ERG: Did not consider this reflected range of treatments in clinical practice

ICER impact: increases

treatment sequencing

Whether to include rituximab first-line maintenance treatment in the treatment pathway

ERG: Recommended inclusion in scenario analysis; inclusion increased ICERs to £15,000–£21,600

Committee: Appropriate to consider ICERs from scenario analysis including first-line maintenance treatment in light of NICE TA226 recommendations; both guidances should be considered for review together

ICER impact: increases

treatment sequencing

Appropriate subsequent treatment pathways after first-line therapy

Company: Assumed CHOP or CHOP alone after first-line treatment

ERG: Developed model to include CHOP, fludarabine-cyclophosphamide, and stem-cell transplant based on clinical adviser input

Committee: Treatment pathways in Assessment Group's model are appropriate and reflect clinical practice in the UK

ICER impact: uncertain_direction

utility source

Disagreement over utility values: manufacturer used disaggregated values based on degree of response to therapy, Assessment Group used aggregated health-state utilities based on number of lines of treatment.

Company: Used disaggregated utility values from main analysis of Pettengell et al. 2006: PF1 = 0.88; PF2 = 0.79; progressive disease = 0.62, with different values for first-line and second-line progression-free survival

ERG: Used aggregated health-state utilities from additional analysis with unified values for first-line and second-line progression-free survival = 0.805; progressive disease = 0.7363, considered more appropriate as utilities not based on degree of response

ICER impact: uncertain_direction

Evidence gaps

no direct comparisonDifferent chemotherapy regimens (CVP, CHOP, MCP, CHVPi) compared in separate trials with no direct head-to-head comparisons between them
short follow upThe length of follow-up in the four trials was short compared with the natural course of follicular lymphoma
no direct comparisonLimited evidence for rituximab effectiveness when added to chemotherapy regimens other than CVP, CHOP, MCP and CHVPi; one uncontrolled study of rituximab plus chlorambucil in 27 patients
otherLimited evidence regarding loss of efficacy of rituximab after re-treatment in previously treated patients
no direct comparisonRituximab may be combined with chemotherapy regimens not included in the clinical trials (CHVPi, chlorambucil); uncertainty about relative effect and absolute response rates of these unstudied combinations
short follow upLimited clinical data for rituximab plus chlorambucil; absence of formal cost-effectiveness analysis for this combination

Commercial arrangement

Special considerations

Innovation acknowledged Equality issues raised

Cross-references

TA110precedent — This appraisal is a review of NICE technology appraisal guidance 110
TA226utility reuse — Rituximab for the first-line maintenance treatment of follicular non-Hodgkin's lymphoma
TA226same condition — Rituximab maintenance treatment for follicular lymphoma; Committee noted that inclusion of rituximab first-line maintenance treatment (as recommended in TA226) increased ICERs and that both guidances should be considered for review together
TA226same drug — Rituximab for the first-line maintenance treatment of follicular non-Hodgkin's lymphoma
TA137same drug — Rituximab for the treatment of relapsed or refractory stage III or IV follicular non-Hodgkin's lymphoma (review of TA37)