PARKE-DAVIS' NIPENT GETS FDA CMTE. NOD AS FIRST-LINE HAIRY CELL LEUKEMIA

PARKE-DAVIS' NIPENT GETS FDA CMTE. NOD AS FIRST-LINE HAIRY CELL LEUKEMIA treatment based on the results of a 356-patient comparison study with alpha interferon conducted by the Southwest Oncology Group. FDA's Oncology Drugs Advisory Committee reviewed the study at its March 1 meeting and voted six to two in favor of recommending approval of the supplemental NDA. Commenting on the study results, Harold Harvey, MD, University Hospital, Hershey, Penn., said: "The data from the trial convince me that there is clear superiority for pentostatin over alpha interferon." Nipent (pentostatin) was approved by FDA Oct. 11, 1991, as a second-line treatment for adult patients with hairy cell leukemia refractory to alpha interferon. Warner-Lambert's Parke-Davis unit submitted a supplemental NDA on Nov. 11, 1992 for Nipent as a single-agent treatment for adult patients with both untreated and refractory hairy cell leukemia. To support the first-line indication, the company provided FDA with data from the SWOG study comparing Nipent with alpha interferon in previously untreated hairy cell leukemia patients. Results from that study were not yet complete at the time that Nipent received second-line approval ("The Pink Sheet" Oct. 21, 1991, T&G-1). Parke-Davis' last interim analysis of the SWOG study dates from Oct. 31, 1991. Patients were initially randomized to either pentostatin or alpha interferon and were evaluated at six months with a bone marrow biopsy and a complete clinical assessment. Pentostatin was initially given to 171 patients with confirmed hairy cell leukemia, at a dose of 4 mg/m once every two weeks. Interferon was given to 170 patients with hairy cell leukemia at 3 mil. IU three times a week. At the conclusion of the trial, there were 138 evaluable pentostatin patients and 130 evaluable interferon patients. Parke-Davis Senior Director of Clinical Oncology Charles Kowal, MD, reported that among those evaluable, 84% initially receiving Nipent had a complete response to treatment, and 6% had a partial response, for a total response rate of 90%. This compared to a complete response rate of 18% for patients initially receiving alpha interferon, and a total response rate of 42%. Of 75 patients who were crossed over from interferon to pentostatin, 85% had a complete response on pentostatin and 4% had a partial response. "Interferon failure doesn't diminish the ability of pentostatin to produce a complete response," Kowal pointed out. The median duration of response to interferon in the SWOG study was eight months for the 18% of interferon patients who had achieved complete responses, Kowal said. The median duration of response for pentostatin has not yet been reached. At two years, 16% of the interferon patients and 76% of the pentostatin patients who had complete responses remained in complete remission. Kowal added that some patients on pentostatin were still in complete remission at four years. While committee members were enthusiastic about the Nipent response rates, the group "strongly suggested" that FDA require follow-up from Parke-Davis in light of some uncertainty regarding survival data and infection rates. The analysis of survival rates for Nipent and alpha interferon is made difficult by the crossover design of the SWOG study as well as the low number of deaths which occurred -- 19 among evaluable patients. The matter is further complicated by an April 1992 analysis of the data by study designer Michael Grever, MD, National Cancer Institute, which, in contrast to the findings of Parke-Davis' earlier analysis, found a statistically significantly lower survival rate among pentostatin patients. There were 33 deaths among the 356 patients in the study, 20 in patients who had initially received pentostatin and 13 in patients initially on interferon, for respective mortality rates of 11.2% and 7.4%, Kowal reported. Two deaths occurred among the 11 patients who crossed over from pentostatin to interferon, and three deaths among the 99 patients who crossed from interferon to pentostatin, giving mortality rates among crossover patients of 18% for interferon and 3% for pentostatin. While Kowal said the survival data submitted with the supplemental NDA did not show a statistically significant difference between the two drugs, committee members were surprised to learn that the data he presented were from before Grever's most recent analysis. Some of the group's members expressed concern that survival for patients on pentostatin might turn out to be worse than for interferon. Committee member Paul Bunn, MD, University of Colorado Cancer Center, acknowledged that he would be "happy to approve [Nipent] now" with the reassurance that the advisory committee would address the mortality issue again if a later analysis were to show significantly worse survival. Grever reported that he is currently performing an up-to-date analysis, though he emphasized that the study was not designed with mortality as a primary endpoint. The possibility of a greater risk of infection among patients receiving Nipent was raised by data showing that 38% of Nipent patients had documented infections after treatment began versus 34% for alpha interferon. At baseline, 21% of pentostatin patients had infections versus 13% of those on interferon.