Patients who received the PROSTVAC vaccine, with or without granulocyte-macrophage colony-stimulating factor (GM-CSF), or placebo had a median overall survival of 33-34 months. The key secondary endpoint of patients alive without events (AWE) at 6 months also did not differ significantly among the three treatment arms.
The phase III international trial failed to corroborate results of a phase II trial, which showed almost a 10-month improvement in survival with the vaccine versus placebo, as reported online in the Journal of Clinical Oncology.
“What happened in the interim [since the phase II trial]? We’re not 100 percent sure,” said James L. Gulley, MD, PhD, of the National Cancer Institute (NCI) in Bethesda, Maryland.
One possible explanation lies in the approval of five new drugs for mCRPC. Treatment with one or more of those agents could have diluted the survival benefit observed in the phase II trial. A more likely scenario — and one that is being actively pursued in multiple studies — involves blunting of the vaccine‘s biologic activity.
“It’s one thing to get the T-cells into the tumor microenvironment, but it’s another thing to allow those T-cells to be functional in the tumor microenvironment,” Gulley told MedPage Today. “There are a number of things in the microenvironment that potentially could prevent the T-cells from doing what they’re supposed to do, including PD-L1 or other regulators of the immune system that can downregulate the function of T-cells.”
Multiple ongoing investigations are evaluating strategies to overcome the downregulating effects, such as combining the vaccine with a PD-1/PD-L1 inhibitor, he added.
PROSTVAC comprises two poxvirus vectors encapsulating transgenes for human prostate-specific antigen and three costimulatory molecules for T-cells. A phase II randomized, placebo-controlled trial of the vaccine failed to meet the primary endpoint of improved progression-free survival or result in any objective response. However, patients treated with the vaccine lived longer.
GM-CSF has immunomodulatory activity, which provided a rationale for evaluating the agent in combination with PROSTVAC, but previous trials did not definitively demonstrate a benefit from the necessity of adding GM-CSF, the authors noted in their introduction to the study.
The collective clinical experience led to the design of a phase III, randomized, placebo-controlled trial to evaluate PROSTVAC, alone or in combination with GM-CSF, in patients with asymptomatic or minimally symptomatic mCRPC and documented progressive disease. The patients were randomized to PROSTVAC, PROSTVAC plus GM-CSF, or placebo. The trial had a primary endpoint of overall survival.
At the third interim data analysis, the trial met prespecified criteria for futility. The final analysis included 1,197 randomized patients. The data showed a median overall survival of 34.4 months with PROSTVAC alone, 33.2 months for PROSTVAC plus GM-CSF, and 34.3 months for the placebo group. AWE at 6 months also did not differ significantly among the three treatment groups, ranging from 28.0% with PROSTVAC/GM-CSF to 30.3% with placebo.
Rates of adverse events (all grades) did not differ across the three treatment groups. The most commonly reported adverse events were injection-site reaction (62-72%) and fatigue (21-24%). The most common cardiac events were arrhythmias, occurring in 1.4-3.5% of patients in each group. No cases of myocarditis or pericarditis were reported. Serous treatment-related adverse events occurred in fewer than 1% of all patients.
The data did not make a case for or against the use of GM-CSF. “Therefore, use of GM-CSF as a vaccine adjuvant must still be considered investigational,” the authors stated.
The trial’s failure to show a survival benefit does not represent an investigational death knell for the vaccine. Gulley remains optimistic about the prospects for finding strategies to unlock the therapeutic potential of PROSTVAC and cancer vaccines in general. In a recent review article, he and NCI colleague Jeffrey Schlom, PhD, made a case for vaccines as an “integral component of cancer immunotherapy,” either by themselves or in combination with other approaches that target the immune system.
The study was supported by Bavarian Nordic.
Gulley disclosed relationships with EMD Serono, Bavarian Nordic, Astellas Medivation, Pfizer, NantBioScience, Bristol-Myers Squibb, and Merck; multiple co-authors disclosed relationships with the pharmaceutical industry and other commercial entities.