Provenge Fails To Meet Interim Study Endpoint, But Dendreon Will Press On

Prostate cancer immunotherapy reduces risk of death by 20 percent compared with placebo, but falls shy of target.

Dendreon will have to wait until mid-2009 at the earliest for final study results of its prostate cancer vaccine Provenge after it failed to meet the primary endpoint in an interim look at data.

The interim data, announced Oct. 6, were highly anticipated by investors hoping the results would suffice for FDA to reconsider approval of Provenge well in advance of the full study's completion. It was the third time the cancer vaccine missed its primary endpoint.

The Phase III IMPACT (9902B) study is an ongoing, double-blind, placebo-controlled Special Protocol Assessment trial testing Provenge (sipuleucel-T) in 512 men with metastatic, androgen-independent prostate cancer. Although full data remain blinded, the Seattle biotech reported a 20 percent reduction in the risk of death in the treatment arm compared to the placebo arm (hazard ratio=0.80).

To meet the primary endpoint of overall survival, the vaccine must demonstrate a 22 percent reduction in risk of death or greater. The trial's independent data monitoring committee has observed no safety concerns to date. It recommended that the trial continue on to the pre-specified completion point of 304 events. Dendreon anticipates final IMPACT data in mid-2009.

During a same-day investor call, Dendreon CEO Mitchell H. Gold attempted to put a positive gloss on 24-month data, asserting that the trial is powered to produce stronger results over time.

Had Provenge met the primary endpoint at the interim look, Dendreon would have met FDA's request for additional clinical data to support efficacy. The first therapeutic cancer vaccine to be reviewed by FDA, Provenge has traveled a winding path the past two years, including positive recommendation by an advisory committee, followed by a "complete response" letter seeking further efficacy data.

FDA accepted the biologics licensing application for Provenge for priority review early in 2007, setting an action date for that May (¹ (Also see "Dendreon's Provenge Therapeutic Cancer Vaccine Has A May 15 PDUFA Date" - Pink Sheet, 16 Jan, 2007.) Indicative of the application's future path was the reality that the vaccine did not meet its primary endpoint - overall improvement in time to progression - in its Phase III registration trial, D9901. Dendreon based its argument on improvement in overall survival in that trial.

When CBER's Cellular, Tissue and Gene Therapies Advisory Committee considered the BLA last March, the vaccine had failed its second Phase III trial. The committee, however, voted 14-3 that it saw substantial evidence of efficacy based on a post hoc analysis showing a 4.5 month increase in median survival in pooled data from the two studies (² (Also see "Astellas’ Prograf “Approvable” For Use In Combination With Roche’s CellCept" - Pink Sheet, 20 Mar, 2007.)).

FDA quickly followed that vote with a "complete response" letter saying it would need additional efficacy data in order to approve Provenge (³ (Also see "Dendreon’s Therapeutic Vaccine Provenge Gets “Complete Response” From FDA" - Pink Sheet, 9 May, 2007.)).

The data, however, were inadequate to sway reviewers in CDER's office of oncology drug products, particularly Richard Pazdur, director of the cancer office with a reputation for being a very tough reviewer (⁴ (Also see "The Provenge Precedent" - Pink Sheet, 1 May, 2007.)).

Provenge must at least meet IMPACT's primary endpoint when the full results are vetted in the middle of next year, upon the trial's completion.

IMPACT powered to produce stronger final data

During the Oct. 6 call, Dendreon's Gold suggested his company is poised to do just that. The IMPACT statistical plan was based on integrated results from the prior two Phase III trials, he explained, which produced a 22 percent reduction in risk of death at 24 months.

Even though IMPACT had just a 20 percent reduction of risk of death at 24 months, Gold asserted Provenge is on pace to hit or exceed the 22 percent mark in the final data. By design, the interim look has less power than the final analysis, he said.

"The final analysis from our integrated studies showed approximately a 33 percent reduction in the risk of death, a hazard ratio of 0.67, supporting the hypothesis that there is a delayed treatment effect of Provenge, which tends to get larger over time," he concluded.

In an Oct. 6 note, analyst Reni Benjamin of Rodman & Renshaw offered guarded agreement for Gold's prediction.

"We remain cautiously optimistic regarding the results of the final analysis," he wrote. "In our opinion, the data presented today is similar to previous Phase III Provenge trials. In addition, we believe that the statistical significance could improve as the number of events approaches 304."

Earlier this year, Dendreon and FDA amended the SPA, agreeing to conclude the IMPACT trial at 304 events, rather than the previously agreed upon 360, accelerating the anticipated timing of final analysis from mid-2010 to mid-2009 (⁵ (Also see "Dendreon Catches A Break For Provenge" - Pink Sheet, 12 Mar, 2008.)).

While Dendreon is attempting to become the first company to bring a therapeutic cancer vaccine to the U.S. market, Oxford BioMedica and Sanofi-Aventis have reached agreement with FDA to develop a clinical path for their vaccine, TroVax , in renal cancer and possibly colorectal cancer, as well ( ⁶ (Also see "Oxford BioMedica/Sanofi Make Headway With FDA On TroVax Cancer Vaccine" - Pink Sheet, 7 Oct, 2008.) ).

-Joseph Haas ([email protected])