West Coast Editor

The pre-panel meeting word from the FDA in online satraplatin briefing documents that walloped GPC Biotech Inc.'s shares - denting licensee Spectrum Pharmaceuticals Inc. and GPC partner Pharmion Corp. as well - could mean a delay of the prostate cancer drug's approval, and for how long is unclear.

GPC's stock (NASDAQ:GPCB) closed Friday at $24, down $7.80, or 24.5 percent. Shares of Waltham, Mass.-based Spectrum (NASDAQ:SPPI) ended the day at $5.45, down $1.45, or 21 percent. Pharmion (NASDAQ:PHRM) finished at $25.39, down $1.76.

Joel Sendek, analyst with Lazard Capital Markets, wrote in a research report that he still sees "a good chance of a satraplatin U.S. launch by early next year."

Satraplatin, to be branded Orplatna if approved, will be debated next Tuesday at a meeting of the FDA's Oncologic Drugs Advisory Committee, and GPC has slated a conference call for the next day. But the panel tipped its hand in documents filed online Friday, bringing up five issues that could lead the agency to decide in favor of waiting for final survival data from the registration trial, due at the end of this year, or - worst case - ask for a new study.

Laurie Doyle, director of investor relations and corporate communications for GPC, said the firm had decided in advance that it would not comment on the briefing documents, no matter what they contained.

She said GPC "remains convinced of the strength of the satraplatin data," and looks forward to its chance in front of the panel tomorrow. Satraplatin's new drug application is supported by results from the Phase III SPARC (Satraplatin and Prednisone Against Refractory Cancer) trial, which compared the oral platinum drug plus prednisone to placebo plus prednisone in 950 patients with hormone-refractory prostate cancer.

At the American Society of Clinical Oncology meeting last month, Martinsried, Germany-based GPC and Pharmion, of Boulder, Colo., said results from SPARC showed satraplatin lowered the risk of disease progression by 33 percent compared to the control group.

The briefing documents noted that the agency has no experience with the progression-free survival endpoint used in the trial, adding that "this was clearly communicated to the applicant during the development phase."

Secondly, two independent radiology readers "disagreed on the progression status in 367 of the 950 patients (39 percent), requiring adjudication by a third independent radiology reader. That raises the question whether PFS could be reliably assessed" in the trial, the documents pointed out.

The third issue has to do with measuring pain progression, which is part of the composite PFS co-primary endpoint and the basis for the endpoint of time to pain progression.

"Because of Orplatna toxicities, it is unlikely that blinding was maintained," the documents asserted. What's more, "based on a review of background materials provided by the applicant describing the methods for assessing pain intensity in the SPARC Study, the FDA has determined that the single item Present Pain Intensity Scale (PPI), derived from the McGill Pain Questionnaire has not been adequately validated for use in this study."

Although the McGill PPI was used a decade ago to win approval of mitoxantrone (marketed as Novantrone) for hormone-refractory prostate cancer, different criteria for pain response and pain progression were used, the documents indicated. "Also, in the mitoxantrone study the primary endpoint was reduction in pain intensity, while in the Orplatna study, the main pain endpoint is time to pain progression." The standards of the FDA's Center for Drug Evaluation and Research have changed since then.

The fourth concern relates to the fact that only 51 percent of trial patients previously got Taxotere (docetaxel), the only drug shown to improve survival in patients with HRPC. "All patients should have had prior docetaxel," the documents noted, though the SPARC trial was started before the FDA approved Taxotere in the indication. So researchers don't know yet whether a survival benefit in this subgroup can be achieved.

"The fifth issue is whether the FDA should wait for the final survival analysis before taking action on the Orplatna application," according to the documents.

"An interim analysis of overall survival after 463 deaths does not show that Orplatna is better than placebo. The final analysis of overall survival will occur when there are 700 deaths, which is estimated to be near the end of 2007," they added.