An analysis of a large, phase III trial revealed time interval to biochemical failure as a potential surrogate endpoint for survival outcomes in patients with prostate cancer receiving androgen deprivation therapies (ADTs). The study results were recently published in the Journal of Clinical Oncology.
“One problem we face in prostate cancer clinical trials is that, overall, prostate cancer is a slow-growing cancer,” Guilherme Godoy, MD, medical oncologist at the Dan L Duncan Comprehensive Cancer Center at Baylor College of Medicine, told Cancer Network. Conducting large, phase III trials requires thousands of patients that need to be followed for years, which becomes complicated and expensive. “Having surrogate endpoints for prostate cancer research and clinical trials is a real need,” he said.
To determine whether time interval to biochemical failure could be a surrogate endpoint, study researchers analyzed data from the NRG Oncology/Radiation Therapy Oncology Group (NRG/RTOG) 9202 trial. The NRG/RTOG 9202 trial included approximately 1,500 men with localized high-risk prostate cancer, with the goal of comparing two treatment regimens: short-term androgen deprivation (radiation therapy and 4 months of ADT) versus long-term ADT (radiation therapy and 4 months of ADT plus an additional 24 months of ADT).
There were four landmark time points for biochemical failure status: 12 months, 24 months, 36 months, and 48 months after radiotherapy completion (or 24 months after ADT completion for those in the long-term ADT group). Time interval to biochemical failure was assessed in relation to overall survival (OS) and prostate cancer–specific survival (PCSS). The median follow-up was 20 years.
Long-term ADT conferred improved PCSS (HR, 0.70; 95% CI, 0.55–0.88) and OS (HR, 0.88; 95% CI, 0.79–0.98) compared to short-term ADT. In addition, long-term ADT conferred a reduction in biochemical failure at all four landmark time points assessed.
Overall, biochemical failure status at as early as 24 months after the completion of radiotherapy predicted the clinical outcome of long-term vs short-term ADT, implicating biochemical failure as a possible surrogate endpoint for prostate cancer. At 36 months and 48 months, biochemical failure status was even more predictive. Also, the proportion of prostate cancer–specific deaths was lower among men who did not reach biochemical failure at 36 months compared with those who did (19% vs 50%).
In addition, biochemical failure status was shown to mediate treatment effect. At 36 months after radiotherapy completion, biochemical failure status mediated 85% of PCSS, 47% of metastasis-free survival, and 52% of OS. At 48 months after radiotherapy completion, the proportions rose, approaching 100%, suggesting that nearly all of the treatment effect is explained by biochemical failure.
“Overall it was a well-planned, well-conducted study,” Godoy said. “It showed that using those intervals to biochemical failure times might be a very good, reliable, potential end point that would save a lot of time and resources on prostate cancer clinical trials.”
“Of course, I don’t think this study is definitive enough to say, ‘Okay, this is it. We’re going to start designing trials with that surrogate endpoint as our primary endpoint.’ I don’t think we’re quite there yet,” he said. The time interval to biochemical failure can be implemented into clinical trials, and if validated in future studies, it could be a “very good” marker for response so doctors can determine whether the treatment is working a few years sooner rather than having to wait for an actual clinical response.