There was ample of evidence of survival benefit conferred by Provenge as a treatment in contrast to Dr. Pazdur’s statement: “Believe me, if there were a clear survival effect, the drug would be approved,”. In fact, Dr. Pazdur’s focus on the statistical flaws showed that the main problem in the FDA drug approval process as practiced by Dr. Pazdur is a slavish following of rigid statistical rules without properly factoring in clinical knowledge required to give meaning to such numerical data.

The irony is that FDA has a process to incorporate clinical knowledge into drug evaluation. Advisory Committees made up of world-class experts are often convened to discuss clinical data and provide advices to the FDA during the evaluation of drugs up for approval. The AC convened to consider Provenge voted 17-0 that the drug was safe and 13-4 that there was substantial evidence that it was effective. When Dr. Pazdur helped rejecting Provenge, he effectively ignored that body of knowledge. He showed himself to be not only a slave to simple statistical calculations but also one who cared little for the scientific process and the needs of desperate patients.

For Provenge, the clinical data were:

1. D9901, a phase 3 trial enrolling 127 patients, whose primary endpoint, Time To Progression, barely missed statistical significance with p-value .052 and Overall Survival was highly significant with p-value .01. The survival rates at three years were 1 in 3 for patients treated with Provenge compared to 1 in 9 for patients treated with placebo. Per several sensitivity analysis performed by the FDA statisticians themselves, this trial was well run, the data self-consistent and the survival benefit unlikely to be spurious.
2. D9902a, a smaller phase 3 trial enrolling 98 patients, missed statistical significance for both TTP and OS. Again, per analysis done by FDA statisticians, this trial unfortunately enrolled much sicker patients on the treatment arm than on the control arm. This imbalance meant that patients on the treatment arm died faster without any intervention and that stipulated the miss of statistical significance. However, after three years, even with such a disadvantage, Provenge stopped the disease in time to again save 1 in 3 patients on the treatment arm compared to 1 in 5 on the placebo arm.
3. The Hazard Ratio is a common measure of comparative risks of death between two treatments. The D9901 trial exhibited an HR of 0.58 meaning that patients on the treatment arm had 42% survival advantage over patients on the placebo arm. The D9902a trial exhibited an HR of 0.76 or a 24% survival advantage for treated patients.
4. Patients in D9902a were much sicker than D9901 and more similar to the patients enrolled in the TAX327 trial that led to the FDA’s approval of the chemotherapy Taxotere as the current Standard Of Care for prostate cancer. The HR for TAX327 was also 0.76 comparable to D9902a.
5. Another Taxotere phase 3 trial, SWOG99-16, showed a worse HR of 0.80. At three year, less than 1 in 5 treated patients were still alive. After four years, all passed away. There are patients treated with Provenge still living today after six to seven years.
6. Placebo patients in the Provenge trials were allowed to cross over to take a form of Provenge made from their cryo-preserved blood. Of the 12 out of 78 patients on the placebo arms of the Provenge trials D9901 and D9902a who survived past three years, only two did not cross over. The cross-over protocol was there for compassionate reason. But without it, the survival benefit of Provenge over non-treatment would have been much clearer in the statistical sense.

Popularity: 21%

Tags: AIPC, Allison Martin, Brachytherapy, Cancer Vaccine, Care To Live, CareToLive, commissioner Andrew Von Eschenbach, Dendreon, FDA, foia, Freedom of Information Act, hormone refractory, Howard Scher, Immunotherapy, Jesse Goodman flubs it, maha hussain, Michael Milken sadly remains silent, Mike Milken, oncologist, Prostate Cancer, Prostate Gland, Prostatectomy, provenge, Richard Pazdur, Rick Pazdur, Robotic Surgery, The Prostate Cancer Foundation stays silent, Thomas R Fleming

This entry was posted on Wednesday, May 28th, 2008 at 2:02 am and is filed under The Cause, The Dysfunctional FDA. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.