TA870 · CDF_review

Ixazomib with lenalidomide and dexamethasone for treating relapsed or refractory multiple myeloma

Recommended with restrictionsNovember 2022

Only if they have had 2 or 3 lines of therapy and the company provides ixazomib according to the commercial arrangement (simple discount PAS)

Source documents

Intervention

ixazomib with lenalidomide and dexamethasone
· proteasome inhibitor with immunomodulatory agents · oral

Condition

relapsed or refractory multiple myelomahaematology · relapsed_refractory

Comparators

NameType Established Committee preferred
lenalidomide and dexamethasoneactive drugYesYes

Clinical trials

TrialDesignPhasePivotal
TMM1RCT3Yes

Economic model

partitioned survival (company)
Time horizon: lifetime

ICER

£20,000–£30,000 (ixazomib combination vs lenalidomide and dexamethasone) · high uncertainty

Methodological decisions (5)

mortality assumption

Adjustment of overall survival to remove effect of subsequent treatments using 2-stage method with recensoring

Company: 2-stage method with recensoring appropriate; minimal difference between adjusted analysis with recensoring and unadjusted analysis on cost-effectiveness

ERG: Noted that adjustment departed from clinical plausibility; more people had survival-extending treatments in lenalidomide arm yet saw greater reduction in life years; questioned why post-progression life-year gain changed from 7.4% to 30.5% of total

Committee: Although uncertainty exists, company's approach was appropriate; bias from recensoring likely less than bias from not censoring; true cost-effectiveness could be higher than presented

ICER impact: decreases

population generalisability

Representativeness of TMM1 trial population versus NHS eligible population

Company: Trial population clinically appropriate for decision making

ERG: Not stated as separate position

Committee: TMM1 evidence robust and most appropriate; population differences likely contributed to differences versus SACT dataset; people in SACT dataset older and poorer prognosis; TMM1 follow-up longer and more mature

ICER impact: uncertain_direction

survival extrapolation

Choice between generalised gamma and Weibull curves for extrapolating overall survival

Company: Generalised gamma model chosen because clinicians stated it provided reasonable estimation of long-term outcomes

ERG: Weibull model was as valid on grounds of clinical plausibility and statistically better fitting than generalised gamma for both arms

Committee: Generalised gamma acceptable but Weibull should also be considered; clinical experts acknowledged both estimated similar overall-survival outcomes

ICER impact: negligible

treatment effect waning

Whether to model waning of relative treatment benefit after stopping ixazomib combination

Company: No treatment waning effect included in base case; discontinuation and waning almost completely captured within 8-year observation time

ERG: Discontinuation of treatment captured but separate from waning of treatment effect; over 90% of people only observed for around 2 years post-discontinuation, insufficient to capture waning; conducted scenarios exploring impact

Committee: Long follow-up of TMM1 largely captured treatment waning; no additional adjustments needed

ICER impact: increases

utility source

Source of health-related quality of life values for model health states

Company: Updated EQ-5D data from final analysis of TMM1

ERG: Agreed that utilities from TMM1 final analysis better aligned with literature

Committee: Utility values acceptable; better aligned with literature than prior myeloma appraisals

ICER impact: negligible

Evidence gaps

short follow upOver 90% of patients only observed for around 2 years following discontinuation of ixazomib combination, insufficient to assess potential waning of treatment effect

Commercial arrangement

simple discount pas · confidential

Special considerations

End of life considered (not met) Cancer Drugs Fund eligible Equality issues raised

Cross-references

TA505precedent — Previous appraisal of ixazomib combination; committee reconsidered treatment waning assumption and end-of-life criteria from TA505