FDA Panel Recommends Approval of Therapeutic Prostate Cancer Vaccine

ROCKVILLE, Md., March 30 -- FDA advisers have recommended that the agency approve a therapeutic personalized vaccine for treatment of metastatic prostate cancer that may increase survival by several months.

The panel of advisers voted 13 to 4 that there was "substantial evidence" that treatment with Provenge (APC8015) was effective and 17 to 0 that the treatment was safe.

But the efficacy vote was influenced by a median survival benefit of 4.5 months that only emerged in a post hoc analysis of data from a trial that failed to meet its prespecified endpoint of increase in time to disease progression.

The deadline for the FDA to act on the panel's recommendation is May 15, and while the FDA usually follows recommendations of its advisory panels, it is not obligated to do so.

APC8015 is an active cellular immunotherapy product, known as a dendritic-cell vaccine, proposed for the treatment of men with asymptomatic metastatic androgen independent prostate cancer.

The product consists of peripheral blood mononuclear cells, which are obtained from patients by leukopheresis and activated in vitro with a recombinant fusion protein (prostatic acid phosphatase fused with GM-CSF). These activated cells including antigen presenting cells are then re-infused intravenously into the autologous patients.

In two trials submitted to the FDA the primary endpoint was a 3.7 month increase in time to disease progression.

Patients were stratified by study center and bisphosphonate use, centrally randomized in a 2:1 ratio of APC8015 to APC-placebo. They were scheduled to receive three intravenous infusions of either APC8015 or APC-placebo preceded by leukopheresis two to three days prior to the infusion date on weeks 0, two, and four.

Patients were evaluated at weeks two, four, 12, and clinical evaluations were combined with radiographic tumor staging at baseline, weeks eight, 16, 24, and 32, and every 12 weeks afterward until disease progression.

One trial enrolled 127 patients and the second trial-which was terminated prematurely because the initial study did not meet its primary endpoint of improvement in time to disease progression-enrolled 98 patients.

The estimated median time to disease progression was 11.0 weeks for the 82 patients randomized to APC8015 arm in the first trial compared with 9.1 weeks in the APC-placebo arm. This 1.9-week delay in the time to objective disease progression did not reach statistical significance (P = 0.085).

In the smaller trial, D9902A, the estimated median time to disease progression in D9902A was 10.9 weeks for the 65 men in the APC8015 arm compared with 9.9 weeks for 33 patients in the APC- placebo arm (P=0.72).

Neither trial was powered to show a survival advantage, but a post hoc analysis of 36-month survival of the first trial determined that median survival in the vaccine arm was 25.9 months versus 21.4 months in the placebo arm (P=0.010).

"Therefore, study D9901 failed in achieving its primary objective, but a post hoc analysis demonstrated an apparent survival increase in APC8015-treated subjects, the basis for the efficacy claim in this BLA submission," according to the FDA briefing document.

The same analysis found that there were 20% fewer prostate-cancer specific deaths in the APC8015-treated patients, but 13% more deaths from other causes. Moreover, 5.4% of the vaccine-treated patients experience cardiovascular adverse events versus no such events in the placebo patients.

Panel member Richard J. Chappell, Ph.D., a professor of biostatistics at the University of Wisconsin, who was one of the nay votes on efficacy said the approval would provide a false sense of hope for patients.