Azacitidine for Treating Acute Myeloid Leukaemia with More Than 30 % Bone Marrow Blasts: An Evidence Review Group Perspective of a National Institute for Health and Care Excellence Single Technology Appraisal

Abstract

The National Institute for Health and Care Excellence (NICE) invited the manufacturer of azacitidine (Celgene) to submit evidence for the clinical and cost-effectiveness of this drug for the treatment of acute myeloid leukaemia (AML) with more than 30% bone marrow blasts in adults who are not eligible for hematopoietic stem cell transplantation, as part of the Institute's Single Technology Appraisal (STA) process. The Peninsula Technology Assessment Group (PenTAG) was commissioned to act as the Evidence Review Group (ERG). The ERG produced a critical review of the evidence contained within the company's submission to NICE.The clinical effectiveness data used in the company's economic analysis was derived from a single randomised controlled trial, AZA-AML-001. It was an international, multicenter, controlled, phase III study with an openlabel, parallel-group design, conducted to determine the efficacy and safety of azacitidine against a conventional care regimen (CCR). CCR was a composite comparator of AML treatments currently available in the NHS: intensive chemotherapy followed by best supportive care (BSC) upon disease relapse or progression, nonintensive chemotherapy followed by BSC, and BSC only. In AZA-AML-001, the primary endpoint was overall survival (OS). Azacitidine appeared to be superior to the CCR, with median OS of 10.4 and 6.5 months, respectively. However, in the intention-to-treat analysis, survival advantage associated with azacitidine was not statistically significant.The company submitted a de novo economic evaluation based on a partitioned survival model with four health states: "Remission", "Non-remission", "Relapse/Progressive disease" and "Death". The model time horizon was 10 years. The perspective was NHS and Personal Social Services. Costs and health effects were discounted at the rate of 3.5% per year. The base-case incremental cost-effectiveness ratio (ICER) of azacitidine compared with the CCR was £20,648 per quality-adjusted life-year (QALY) gained. In their probabilistic sensitivity analysis (PSA), the mean ICER was £17,423 per QALY. At the willingness-to-pay of £20,000, £30,000 and £50,000 per QALY, the probability of azacitidine being cost-effective was 0.699, 0.908 and 0.996, respectively.The ERG identified a number of errors in Celgene's model and concluded that the results of the company's economic evaluation could not be considered robust. After amendments to Celgene's model, the base-case ICER was £273,308 per QALY gained. In the PSA, the mean ICER was £277,123 per QALY. At a willingness-to-pay of £100,000 per QALY, the probability of azacitidine being cost-effective was less than 5%. In all exploratory analyses conducted by the ERG, the ICER exceeded the NICE's cost-effectiveness threshold range of £20,000 to £30,000 per QALY. Given the evidence provided in the submission, azacitidine did not fulfil NICE's End-ofLife criteria.After considering the analyses conducted by the ERG and submissions from clinician and patient experts, the NICE Appraisal Committee did not r...

Trusting the Results of Model-Based Economic Analyses: Is there a Pragmatic Validation Solution?

“…Based on the experience of one of the authors (MS), who is both an appraisal committee member and director of an ERG, the change to STAs has not altered the typical rankordering of the company-submitted ICER and the ICER deemed most plausible by the ERG. One recent example of a model shown to be particularly favorable was the STA of azacytidine for acute myeloid leukemia [21], where the company's ICER of approximately £21,000/QALY gained increased to approximately £63,000 when undisputed errors in the model construction were corrected; and incorporating further amendments resulted in an ERG-preferred ICER of about £273,000. Another example is regorafenib for previously treated advanced hepatocellular carcinoma [22], where the company's initial submission resulted in a more favorable ICER because it did not pool data from two surveys as believed appropriate by the NICE cancer drugs fund committee and failed to mention one of these two surveys in their submission.…”