Type: Poster Presentation
Presentation during EHA21: On Friday, June 10, 2016 from 17:15 - 18:45
Location: Poster area (Hall H)
Chronic lymphocytic leukemia (CLL) is incurable disease and patients eventually relapse after lines of therapies. Ofatumumab (OFA) is the first therapy approved by the Food Drug Agency as extended treatment in relapsed CLL. PROLONG was phase III trial where 474 subjects in remission after second- or third-line therapy for CLL were randomized 1:1 to OFA or observation (OBS). OFA showed lengthened progression free survival (PFS) by 14.2 months (median 29.4 vs. 15.2; hazard ratio [HR] 0.49, p<0.0001) and longer time to next treatment (median 38.0 vs. 31.1 months; HR 0.623, p=0.005). The estimated median overall survival (OS) was not reached in either arm (HR 0.85; p=0.517). In addition to clinical efficacy, economic evaluation of novel health care technologies is important for decision making in both Europe and North America.
To evaluate the cost-effectiveness of OFA compared to OBS as maintenance treatment in patients with relapsed CLL who are in remission after induction therapy.
A partitioned survival model, commonly adopted in economic evaluations based on randomized clinical trials in oncology, was developed with three health states: progression free, post-progression, and death. Parametric survival functions for PFS, OS, and time to treatment discontinuation were developed based on the PROLONG individual patient data. Model fit was assessed by Akaike Information Criteria and Cox-Snell residual plots.Resource use was modelled adopting a Canadian public health care system perspective. Costs (2015 CAD) included drug acquisition and administration, medical care for adverse events, and pre- and post-progression routine care including follow up therapies (a variety of mono and combo therapies) based on the PROLONG data. Utilities were sourced from the literature, and used to calculate quality-adjusted life years (QALYs) based on modeled PFS, OS, and adverse events (grade 3 or 4 events occurring in at least 2% of the PROLONG patients). Cost-effectiveness was evaluated as the incremental cost per QALY gained, and uncertainty was explored through deterministic and probabilistic sensitivity analyses.
Lognormal functions were selected to model PFS and OS. The former included a treatment effect (acceleration factor) that was applied for two years (whilst patients remained progression free and on-therapy), consistent with the trial; due to the immaturity of the OS data and the availability of effective salvage treatments following relapse, no impact of OFA on OS was assumed. Over patients’ modelled lifetime, OFA provided an additional 1.26 progression free life years. After applying utilities and discounting (at 5% per annum), this translated into 0.44 QALYs gained. OFA was associated with approximately an additional cost of CAD $30,000 over patients’ lifetime. Incremental cost/QALY was CAD $68,600 (95% confidence interval 49,900 – 107,600), with a 96% probability of being cost-effective at a threshold of 100,000 CAD/QALY. Deterministic sensitivity analyses suggested OFA’s cost-effectiveness was most sensitive to shorter time horizons and quality of life weights for post-progression, but was largely insensitive to alternative scenarios regarding PFS and OS functions, including direct use of Kaplan-Meier survival estimates.
In the PROLONG trial OFA demonstrated substantial gains in terms of PFS and time to next therapy. OFA may represent a cost-effective addition to the management of relapsed CLL patients who have responded to induction therapy.
Session topic: Quality of life, palliative care, ethics and health economics 1
Keyword(s): Chronic lymphocytic leukemia, Cost effectiveness