Abstract
INTRODUCTION: Lenalidomide combined with high-dose dexamethasone (Len+Dex), yields improved time-to-progression (TTP) and survival compared to Dex alone in previously treated patients with multiple myeloma (Weber et. al., NEJM2007, 357:21 and Dimopoulos et. al., NEJM2007, 357:21). It is also effective in patients previously treated with thalidomide. This study aimed to estimate long-term health and cost consequences of Len+Dex versus Dex in patients with multiple myeloma (MM) previously treated with at least two prior therapies, including prior thalidomide.
METHODS: A discrete event simulation of a patient’s disease-course following initiation of Len+Dex or Dex was developed. The model uses patient’s response (complete response, partial, stable disease or progression) and estimates corresponding TTP and subsequent survival based on Weibull functions derived from pooled data from two phase III trials (MM-009/MM-010). Long-term results from the UK MRC MM IV, V, VI, and VIII trials were used to estimate Dex survival, as 47% of Dex patients crossed-over to lenalidomide treatment in MM-009/MM-010. Time-dependent adverse event rates were derived from pooled MM-009/MM-010 data and associated management costs reflecting UK NHS practice were applied. Health utilities by response level were obtained from the literature. Patients remained on treatment until disease progression. Disease management costs were reflective of clinical practice in the UK NHS. As recommended by the UK treasury, costs and health outcomes were discounted at 3.5% per annum in order to adjust to present values. A life-time horizon was used to model costs and health outcomes, including survival and quality adjusted life-years (QALYs).
RESULTS: Use of Len+Dex is associated with a substantial improvement in survival and QALYs. Although estimated incremental costs are large, the improvements in health outcomes yield incremental cost-effectiveness ratios (ICERs) below £30,000 per QALY. Len/Dex is similarly cost-effective when used to treat patients previously treated with thalidomide. Undiscounted ICERs are lower because survival benefits are not fully realized until end-of-life and so are subject to a higher degree of compound discounting than the costs which are incurred relatively early. Univariate and probabilistic sensitivity analyses showed that results remain consistent through broad changes in model parameters.
CONCLUSIONS: Lenalidomide in combination with high-dose dexamethasone in patients with at least two prior therapies delivers substantial improvements in quality-adjusted survival in a life-limiting orphan disease and yields an estimated incremental cost per QALY which falls within a cost-effective range.
≥2 prior therapies Undiscounted Discounted Len+Dex Dex Len+Dex Dex Life Years (projected mean) 5.92 1.08 4.76 1.05 Quality Adjusted Life Years (QALYs) 3.98 0.79 3.23 0.77 Average Cost (per patient) £63,809 £707 £61,171 £694 Incremental cost per Life Year Gained £13,038 £16,301 Incremental cost per QALY Gained £19,781 £24,584 ≥2 prior therapies (including thalidomide) Undiscounted Discounted Len+Dex Dex Len+Dex Dex Life Years (projected mean) 5.42 1.03 4.43 1.01 Quality Adjusted Life Years (QALYs) 3.60 0.72 2.96 0.70 Average Cost (per patient) £53,719 £706 51,745 694 Incremental cost per Life Year Gained £9,727 £14,927 Incremental cost per QALY Gained £18,407 £22,589
PCN199 - HOW DO QUALITY-ADJUSTED LIFE YEARS IMPACT REIMBURSEMENT DECISION-MAKING IN THE UK?