Identifying disease-modifying potential in myelofibrosis clinical trials

The ultimate goal of bringing most new drugs to the clinic in hematological malignancy is to improve overall survival. However, the use of surrogate endpoints for overall survival is increasingly considered standard practice, since a well validated surrogate endpoint can accelerate the outcome assessment and facilitate better clinical trial design. Established examples include monitoring minimal residual disease in chronic myeloid leukemia and acute leukemia, and metabolic response assessment in lymphoma. However, what happens when a clinical trial endpoint that is not a good surrogate for disease-modifying potential becomes ingrained as an expected outcome, and new agents are expected or required to meet this endpoint to demonstrate "efficacy"? JAK inhibitors for myelofibrosis have a specific impact on reducing symptom burden and splenomegaly, but limited impact on the natural history of the disease. Since the introduction of ruxolitinib more than a decade ago there has been modest incremental success in clinical trials for myelofibrosis, but no major leap forward to alter the natural history of the disease. We argue that the clinical development of novel agents for myelofibrosis will be accelerated by moving away from using endpoints that are specifically tailored to measure the beneficial effects of JAK inhibitors. We propose that specific measures of relevant disease burden, such as reduction in mutation burden as determined by molecular endpoints, should replace established endpoints. Careful re-analysis of existing data and trials in progress is needed to identify the most useful surrogate endpoints for future MF trials and better serve patient interest.