Abstract 2563

Background:

Results from the phase 3 VISTA trial demonstrated that bortezomib (Velcade®) plus melphalan and prednisone (VMP) has superior efficacy to MP alone in patients with newly diagnosed multiple myeloma (MM) ineligible for ASCT. In a similar patient population, the three-arm phase 3 MM-015 trial compared lenalidomide (Revlimid®) plus MP with or without continuous lenalidomide maintenance treatment after cycle 9 (MPR-R vs MPR) with MP alone. There was no difference in progression-free (PFS) or overall survival (OS) between patients who received MPR or MP during the 9-cycle induction period; MPR-R was associated with superior PFS vs MP during the R maintenance period. To assess the value of R in combination with MP followed by R maintenance therapy we indirectly estimated the incremental cost-effectiveness of VMP vs MPR-R as therapy for MM patients ineligible for ASCT.

Methods:

An Excel-based Markov model from the US payer's perspective was developed. Simulations were performed for hypothetical cohorts of newly diagnosed MM patients ineligible for ASCT with an average age of 70 years at treatment initiation. The model includes seven health states representing periods of treatment response (stable disease/minimal response, partial response, and complete response), treatment-free interval/maintenance treatment, progressive disease, second-line treatment, and death. Monthly transition probabilities were estimated from patient-level data for VISTA for VMP and MP (data cut-off June 15 2007; San Miguel et al NEJM 2008) and published results for MM-015 for MPR-R (data cut-off April 15 2009; Palumbo et al ASH 2009). As previously reported (Wang et al ASH 2009), costs included per-protocol drug and medical costs, treatment-related adverse event costs, second-line treatment costs, and resource utilization during the treatment-free interval and progressive disease; all costs were adjusted to 2010 US dollars. State-specific utility estimates were derived from patient-level EQ-5D data from the VISTA trial using US-specific weights. Health outcomes (as indicators of the effectiveness of therapy) were expressed in life years (LYs) and quality-adjusted LYs (QALYs). Costs and health outcomes were discounted at 3%. Incremental cost-effectiveness ratios (ICERs) were calculated for VMP vs MPR-R over a lifetime horizon (20 years). In the base case, the MPR-R vs MP hazard ratio (HR) for PFS was set to 0.499 and that for OS was set to 1, due to a lack of survival benefit with MPR-R vs MP observed in MM-015. One-way sensitivity analyses were conducted by running the model with upper and lower values of key parameters to assess the general robustness of model findings and identify key drivers.

Results:

Model base case results for the incremental cost-effectiveness of VMP relative to MPR-R are shown in the Table. Estimated OS was 4.187 years with VMP vs 3.409 years with MPR-R over a lifetime horizon. Lifetime direct medical costs were $119,102 with VMP vs $241,247 with MPR-R; the lifetime cost of R maintenance was $107,047. Thus, VMP appears associated with reduced costs and better outcomes vs MPR-R; VMP costs approximately 50% less than MPR-R and seems to provide slightly more QALYs (0.567) on average. One-way sensitivity analyses supported the general robustness of model findings and identified the MPR-R vs MP HR for OS as a key driver; only when this HR was set to ≤0.25 did MPR-R become cost-effective vs VMP at $100,000 per QALY.

Conclusions:

In newly diagnosed MM patients ineligible for ASCT, VMP appears to be associated with lower costs and better health outcomes vs MPR-R. From a cost-effectiveness perspective, R maintenance therapy therefore seems to have little benefit in this patient population. The current comparison was based on published results for MPR-R from MM-015 after a median follow-up of 9.4 months (vs 16.3 months for VMP vs MP in VISTA), at which time no survival benefit was observed for MPR-R vs MP. If longer follow-up in MM-015 indicates a survival benefit for MPR-R vs MP, a re-estimation of the incremental cost-effectiveness of VMP vs MPR-R using the present Markov model would be warranted; however, updated data (Palumbo et al EHA 2010) show no significant difference in OS between the MPR-R and MP arms at a median follow-up of 21 months.

Table.

Cost-effectiveness of VMP vs MPR-R over a lifetime horizon

VMPMPR-RVMP vs MPR-R
Cost $119,102 $241,247 −$122,145 
LY 4.187 3.409 0.778 
QALY 2.994 2.427 0.567 
VMPMPR-RVMP vs MPR-R
Cost $119,102 $241,247 −$122,145 
LY 4.187 3.409 0.778 
QALY 2.994 2.427 0.567 
Disclosure:

Wang:Millennium Pharmaceuticals: Consultancy, Research Funding. Huang:Millennium Pharmaceuticals, Inc: Employment. Ba-Mancini:Millennium Pharmaceuticals, Inc.: Employment. Shi:Millennium Pharmaceuticals, Inc: Employment. Chen:Millennium Pharmaceuticals: Consultancy, Research Funding. Korves:Millennium Pharmaceuticals, Inc: Research Funding. Dhawan:Johnson & Johnson: Employment, Equity Ownership. Cakana:Johnson & Johnson: Employment, Equity Ownership. van de Velde:Johnson & Johnson: Employment, Equity Ownership. Esseltine:Millennium Pharmaceuticals, Inc.: Employment; Johnson & Johnson: Equity Ownership. Duh:Millennium Pharmaceuticals, Inc: Consultancy, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution