FDA CLINICAL GUIDELINES FOR OVARIAN CANCER: COMPLETE RESPONSE COULD BE

FDA CLINICAL GUIDELINES FOR OVARIAN CANCER: COMPLETE RESPONSE COULD BE a surrogate endpoint for survival in evaluating the effectiveness of chemotherapies, according to a general consensus reached by FDA's Oncologic Drugs Advisory Committee at a recent meeting. Summing up the committee view, committee Chairman Martin Abeloff, MD, Johns Hopkins Oncology Center, stated, "certainly for the drug that really looks very promising, that gives your a significant increase in complete response rate to this disease, with some data on time to relapse one might be able to move toward approval." However, Abeloff noted that for investigational drugs with response rates equivalent to normal therapies, "we are going to have to [consider efficacy endpoints] on a case-by-case look." Committee member Charles Moertel, MD, Mayo Clinic, suggested that complete response might be a valid endpoint for a confirmatory study for a drug in which an initial study has identified a strong correlation between complete response and survival. "If you have one study that does in fact show a striking correlation between complete response and survival [and] you are seeing the same complete response difference initially on that confirmatory study for the same drug," waiting for the survival data from the second study before approval might not be necessary, Moertel said. The committee was asked by FDA at the Dec. 8 meeting for input on agency guidelines for approval of new drug regimens for ovarian cancer. The agency plans to solicit input from the committee on appropriate efficacy endpoints for evaluating drugs for other types of tumors on a case-by-case basis ("The Pink Sheet" Oct. 5, T&G-9). Committee Chairman Abeloff asserted that based on ovarian cancer treatment data, "complete response does impact on survival." However, in other types of cancer, committee members noted, the association may not be as clear. Use of complete response as an efficacy endpoint "has to be very disease specific," Moertel commented. In cancer treatment, equivalency in complete response "may be zilch as far as equivalence in ultimate survival." The drug may have "dreadful long term consequences," or "have a very transient complete response," Moertel added. George Canellos, MD, Dana-Farber Cancer Institute, emphasized the importance of a determination of the disease-free interval following a complete response. "You have already heard that a complete remission is a foot in the door. A pathologic complete remission is another foot in the door, but not reassuring enough. But it seems to me a durable, pathologic complete remission, which you may be able to show us, it may be only 12 months, may be an important indicator of an impact on the biology of the disease," Canellos stated.