Formal discussant William K. Oh, MD, Chief of the Division of Hematology and Medical Oncology, Professor of Medicine and Urology, and the Ezra M. Greenspan, MD, Professor in Clinical Cancer Therapeutics at Icahn School of Medicine at Mount Sinai, New York, was not convinced that corticosteroids were responsible for the poor outcomes in patients with on-study use of these drugs. In his opinion, use of corticosteroids appears to be a marker for much sicker patients, which may have accounted for the outcomes in the post hoc analysis of AFFIRM.

“Dr. Scher said that patients taking on-study corticosteroids were generally sicker, but I would argue that they were much sicker. I went online with the Memorial Sloan-Kettering nomogram and found a 4-month difference in survival based on patient characteristics of the two groups. The real question is why patients were on steroids. Are the steroids bad or are they just used in bad situations?” he commented.

In most clinical trials, prednisone is used in both comparator arms, he continued. Three randomized trials in castrate-resistant prostate cancer have been conducted where prednisone was given only in one arm; two out of three studies actually showed that the prednisone arm was associated with improved survival when administered with docetaxel, he noted.

“My conclusion is that corticosteroid use is associated with a bad phenotype. I did not see any data to suggest that prednisone itself was inducing poor response. We would need a prospective clinical trial to study this,” Dr. Oh said. ■

Disclosure: Dr. Oh is a consultant for Medivation, Astellas, and Janssen.