Prostate Cancer

NEW YORK CITY—A presentation at the Chemotherapy Foundation Symposium here offered a rare glimpse of the personal perspective of an investigator involved in a major Phase III trial that did not lead to the study drug's approval by the FDA. If the SPARC (Satraplatin and Prednisone Against Refractory Cancer) trial had shown this orally bioavailable platinum agent to be superior in overall survival compared with placebo in men with metastatic prostate cancer with disease progression in whom first-line cytotoxic chemotherapy treatment had failed, the drug may have had a chance as second line. Unfortunately, overall survival was equal in the two trial arms, even though it was associated with longer time to progression and improved symptom control, and the manufacturer subsequently announced it would no longer pursue the drug's development for hormone-resistant prostate cancer. But the FDA's Oncologic Drugs Advisory Committee had already begun to question some of the protocol's endpoints in a review months earlier, such as the use of a composite endpoint for progressive disease. SPARC principal investigator A. Oliver Sartor, MD, Associate Professor at Harvard Medical School and the Lank Center for Urinary Oncology at Dana Farber Cancer Institute, reported at this meeting that overall survival was a disappointing median 61.3 weeks for satraplatin, virtually the same as the 61.4 weeks for the placebo arm—“While satraplatin therapy offered improvements in a variety of endpoints, overall survival was not distinct between the arms,” he said. Besides reporting on the trial outcomes per se, Dr. Sartor spent some time commenting on various aspects of the protocol the FDA found troublesome. These included how the trial was blinded, how bones scans and pain were assessed, and why a composite endpoint for disease progression was used. Lessons Learned “These are among the lessons we've learned,” he said, remarking on how the FDA seemed to view the endpoints initially and how it ended up. For example, the time to progression composite included: Tumor progression, defined by Response Evaluation Criteria in Solid Tumors (RECIST) with two or more new soft tissue lesions on bone scans. Skeletal events, defined as fractures, radiotherapy, bone surgery, or initiation of bisphosphonates. Symptomatic disease progression, defined as an increasing Present Pain Intensity score, increasing analgesic consumption, worsening of Eastern Cooperative Oncology Group performance scores, or a weight loss of more than 10%. The data were collected from patient diaries. In a presentation at the 2006 Chemotherapy Foundation Symposium, Dr. Sartor said that the researchers built upon FDA precedent to define progression. “Everything that we used in this trial had previously been accepted by the FDA as an endpoint of progression, making it much harder for them to argue with us later that our issues were not the relevant issues,” he said (OT, 1/25/07 issue). But here at the most recent Symposium, Dr. Sartor explained that earlier in the trial the composite endpoint issue seemed benign and didn't engender much comment. But then “the composite endpoints we presented to the FDA, utilizing a combination of previously approved endpoints, did not seem to be very well accepted,” he said. Designing a protocol using an endpoint from a similar trial that previously led to another drug's FDA approval seems like a safe way to go. For example, pain was assessed in SPARC by the Present Pain Intensity score on the McGill-Melzack questionnaire, which was based on the protocol in Dr. Ian Tannock's mitoxantrone studies that had led to FDA approval of that drug for palliation of advanced prostate cancer. “But the mitoxantrone trial assessed symptom changes by office visits, not by a diary,” he said. “We thought a diary would be better to capture the data, but that turned out to be an issue for the FDA.” Blind-sided by Toxicity The FDA also had problems with the way the study was blinded, Dr. Sartor said.Figure: A. Oliver Sartor, MD: “Is approval based on endpoints other than survival possible in hormone-refractory prostate cancer? I think there is some clinical benefit here—less pain and progression at a lower rate. But is that going to be recognized? The answer is, I don't know.”“They said, ‘How can you blind a trial if you have toxicity not balanced between the arms?’ They told us ‘if patients have even a little bit of nausea they are going to think they're on the active-treatment arm and they'll think they will have less pain, so what you'll have is a powerful placebo effect.’” And then there were the disagreements between the two blinded radiologists over interpretation of the bone scans, and in 39% of cases the disagreement had to be adjudicated by a third radiologist. This was a very significant point raised by the FDA, Dr. Sartor said. “In the end, the FDA said ‘you don't know how to measure bony lesions; bone scans are not reliable.’ We set it up as what we thought was the right way to do it, but in the end those differences were fatal to the trial.” How to Assess Pain Uniformly in 16 Countries? And the FDA was concerned about analyzing pain uniformly in 16 countries, since pain management differs from country to country. For example, he said, it was pointed out that no study subjects in Romania reached the symptom progression endpoint of starting narcotics for pain, while in the US many patients had. “So if you throw out the blinding, then you start throwing out the patient-reported endpoints such as pain,” Dr. Sartor said. “The endpoints of pain and radiographic tumor progression comprised about 80 percent of the progression events, and if you throw them out and don't have any progression events, you're definitely running into trouble.” At the end of his presentation here, Dr. Sartor asked the question, “Is approval based on endpoints other than survival possible in hormone-refractory prostate cancer?” “I think there is some clinical benefit here—less pain and progression at a lower rate,” Dr. Sartor said. “But is that going to be recognized? The answer is, I don't know.” Multicenter, International The trial, conducted in 16 countries, included 950 men with D2 metastatic hormone-refractory prostate cancer or progressive disease who had failed to respond to one prior chemotherapy regimen. Patients were randomly assigned to receive satraplatin-prednisone or placebo-prednisone. Dr. Sartor and the other key researchers, including Daniel Petrylak, MD, and Cora Sternberg, MD, spoke at major urology and oncology meetings throughout 2006 and 2007 to report on the progress in SPARC. At various times they described superior results for satraplatin in time to progression, pain response, PSA response, and tumor response by RECIST criteria (OT, 4/25/07). They reported a 33% reduction in the risk of disease progression for the 635 satraplatin patients compared with the 315 patients taking placebo. And median time to pain progression improved with satraplatin—66.1 weeks vs 22.3 weeks for placebo. But when the New Drug Application was reviewed by ODAC on July 24, 2007, the committee recommended unanimously to delay its decision until the overall survival data became available (OT, 8/25/07 issue). As of October 31, 2007, shortly after the overall survival data became available, GPC Biotech, the manufacturer and SPARC trial sponsor, had withdrawn its marketing application for this drug after it received a negative response from ODAC. Palliation the Issue The moderator of the Uro-Oncology session that included Dr. Sartor's presentation, Anna C. Ferrari, MD, Co-director of the Genitourinary Cancer Program at New York University Clinical Cancer Center, commented afterwards that she appreciated that the FDA's responses to the satraplatin trial were especially complicated. “It really was not an issue of the survival, but rather an issue that the palliation that they were claiming was not really there, because of flaws in the way that different people treat pain and palliate patients,” Dr. Ferrari said. “In this country we make a lot of emphasis on pain treatment but not everybody is as aware and cautious and follows all the steps.” Since the satraplatin-prednisone regimen could have filled a second-line niche if approved, it might have become a substitute for mitoxantrone for palliation in many patients, she said. “So I could see why the committee had problems with the issue of palliation. That was not the primary endpoint, but a very, very important endpoint, because if it came to the market, that's how it would primarily be used.”

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.