CHICAGO—Randomized placebo-controlled clinical trials are considered by most to be the gold standard for proving hypotheses in cancer treatment and are often called “practice changing” when the results are highly statistically significant compared with earlier treatments. But don't put too much emphasis on any single trial, says Ian Tannock, MD, PhD, Emeritus Professor of Medical Oncology at Princess Margaret Cancer Centre and the University of Toronto.
Speaking here at the American Society of Clinical Oncology as Discussant for RTOG 0521 (Abstract LBA5002) and STAMPEDE (Abstract 5001), both Phase III trials of chemotherapy for men with prostate cancer, Tannock reviewed and commented on those trials, but then spoke about “correcting old mistakes and not making new ones: The trials we discussed are good trials and they give an estimate of the truth, but we need repetition rather than looking at a single detail,” he said. “Repetition of important results is essential before changing practice.”
Laboratory-based scientists must repeat experiments and they often give different results, he said, and the same will be true for “clinical experiments.”
Tannock noted, for example, that the GETUG-AFU 15 (Lancet Oncology 2013;14:149-158) and CHAARTED (ASCO 2014 Abstract LBA2) trials, while excellent studies, gave different results: “And the inclusion of STAMPEDE indicates no statistical heterogeneity among the three studies—the true outcome lies between the extremes.”
Tannock also warned against overextending the results of trials to patients who are not represented by the trial's participants: “These current results do not support adding docetaxel to androgen-deprivation therapy in men who develop metastases several years after diagnosis,” he said.
His last bit of advice: “Clinicians should expect less benefit and more toxicity when applying the results of clinical trials in routine practice.” To illustrate, he described a 2013 study for which he was senior author (Templeton et al: Ann Oncol 2013;12:2972-2977) that compared the outcomes of patients in the TAX327 trial with those of patients in routine practice, all with castrate-resistant prostate cancer receiving docetaxel:
- Median survival in TAX327 was 19.3 months, and the rate of septic neutropenia was three percent;
- Median survival in 43 additional on-trial patients was quite similar—20.4 months—and the rate of septic neutropenia was zero percent; but
- Median survival for 314 patients in routine practice, treated by the same doctors and at the same hospital, was significantly less—13.6 months—and the rate of septic neutropenia was higher: 9.6 percent.
“As you move from selected trial patients to clinical practice, the effects are going to be less and the toxicities greater,” Tannock said.
How Times Have Changed
Tannock concluded with some revealing history: “In 1985, Mario Eisenberger and I published separate articles in JCO questioning whether there was any role for chemotherapy in prostate cancer,” he said, with those articles concluding:
- “There is no evidence that chemotherapy causes a meaningful prolongation of survival... Chemotherapy adds considerable toxicity” (Tannock et al: JCO 1985;7:1013-1021); and
- “The palliative role of nonhormonal cytotoxic chemotherapy in the treatment of endocrine-resistant prostatic carcinoma has not been established” (Eisenberger et al: JCO 1985;6:827-841).
“And now, 30 years later, we're seeing the evolution of medical oncology for prostate cancer, and there definitely is a role,” Tannock said. “How times have changed.”