Multiple myeloma: The paradox and the challenge

Abstract

Cancer Research, Statistics, and Treatment / Volume 2 / Issue 2 / July-December 2019 Multiple myeloma constitutes 10% of hematological malignancies. In the 90s, the median survival was 2–3 years. The outcomes have improved, such that 20%–30% of patients live beyond 10 years.[1] Bagal and Bonda have comprehensively summarized the principles of treatment, the current treatment options and given a glimpse of what lies in store in this field of myeloma. We would like to highlight a couple of points. The design of clinical trials in myeloma and the choice of surrogate endpoints are a topic of debate. Overall survival (OS) is the most meaningful and objective endpoint. An improvement in OS is what matters the most to the patient, provided the quality of life is preserved.[2] None of the recent clinical trials with the newer drugs have used OS as a primary endpoint. OS in myeloma is affected by the numerous salvage therapy options and the presence of competing causes of death such as comorbidities. An interesting paradox is that the availability of newer drugs has made it difficult to conduct a clinical trial with OS as an endpoint. Since the survival has improved progressively, any trial to test an improvement in OS needs to run for at least 8–10 years to detect meaningful differences, for instance, CALGB 100104 trial. By the time the trial is complete, there is a possibility that the research question is irrelevant. This brings us to the question of what is an ideal surrogate endpoint? For most of the trials, progression‐free survival (PFS) is used as the endpoint.[3] This is based on a meta‐analysis that demonstrated a moderate to strong correlation between the PFS and OS. Nevertheless, this analysis was not based on individual patient‐level data. From our earlier experience, we know that an improvement in PFS does not always translate into an OS benefit, the maintenance trial with thalidomide being a case in point. There is an attempt to use the minimal residual disease as an endpoint. As highlighted by Bagal et al. in the review, at present, this is at best a prognostic marker, with several issues such as the need for standardization and validation before it is used as an endpoint to test new drugs.