KABI’s SPHEREX INJECTION PREMARKET APPLICATION DOES NOT MEET DRUG REVIEW CRITERIA

KABI's SPHEREX INJECTION PREMARKET APPLICATION DOES NOT MEET DRUG REVIEW CRITERIA, FDA Oncology and Pulmonary Drug Products Division medical reviewer Grant Williams, MD, said at a June 17 Oncology Drugs Advisory Committee meeting. The company's data "did not meet our usual criteria unless we change our regulatory stance because it's a device, and...we're not going to do that," Williams said. Company officials, FDAers and committee members agreed that the Spherex Premarket Approval Application was complicated by confusion over the standards by which the product was to be judged. Williams called the meeting, the first time a PMA has been reviewed by the Oncology Drugs Advisory Committee, "precedent setting." No vote was taken by the committee following Williams' presentation of the review team's findings on Kabi's degradable starch microspheres (DSM) product. Kabi was seeking approval of Spherex for co-injection with chemotherapy agents "for the temporary occlusion of the hepatic vasculature in the treatment of primary and secondary liver tumors." Kabi said it plans to have follow-up discussions with FDA on how to proceed with efforts to gain approval of Spherex. Clinical trials of Spherex were conducted prior to the transfer of the review from the Center for Devices and Radiological Health to the Center for Drug Evaluation and Research. The transfer was precipitated by the November 1991 intercenter agreement on the evaluation of products that are not clearly drugs, devices or biologics ("The Pink Sheet" Nov. 25, 1991, p. 14). Kabi officials said they were "surprised" by the switch and had expected the product to be judged as a device, not a drug. Williams told the committee that although the history and regulations surrounding Kabi's PMA were "complicated," FDA's goal remained to determine the safety and efficacy of the product. "You're giving something with the chemotherapy -- it's like giving a second chemotherapy agent. You're looking for patient benefit, and I don't really see the difference to the patient whether somebody classifies it as a drug or a device," Williams argued. "We would treat it essentially the same." Kabi maintained that "the repeated intra-arterial co-injection of Spherex with either doxorubicin or mitomycin has been shown to increase the response rate in liver cancers. In addition, Spherex decreases the peak and total levels of the co-injected drugs in the systemic circulation." The company presentation focused on a trial involving 120 patients with hepatocellular carcinoma or liver metastases conducted in Japan in 1988. Kabi Director of Medical Affairs Bo Nilsson, MD/PhD, said an increase in response rates was seen in the hepatocellular carcinoma and liver metastases patients treated with DSM. Williams called the company's response data "marginal," saying they fell below CDER's normal criteria for approval. He also criticized the company's exclusion of 30% of the patients as "not evaluable" due to an array of serious adverse effects -- including two deaths attributed to Spherex toxicity. The studies presented were also viewed as problematic because the metastatic disease patients were heterogeneous and not stratified by tumor type. Data presented did not include determination of response duration and time to progression, omissions that concerned several on the committee. Members disagreed on the significance of the response rates in the absence of these factors. Committee member Nancy Kemeny, MD, Memorial Sloan-Kettering Cancer Institute, said that "as far as a device, it does do what it is supposed to do. There was an increase in response rate," she said, adding that, "I agree the study is small." Committee member Harold Harvey, MD, University Hospital, Hershey, Penn., appeared to concur with Williams' assessment: "We're being asked to approve a device which we full well know is going to be used essentially as a chemotherapy agent. And I think if that is true, then we have to apply the [same] rules that we do for a chemotherapy agent...If we do that," Harvey continued, "then we have to conclude that...this is not a definitive study." Kabi initially submitted a PMA to CDRH in 1985 but withdrew it later that year after FDA suggested a more limited scope for the product than was initially sought. The company submitted a new PMA to CDRH on Aug. 11, 1992; it was transferred to CDER in January. Standard communication between FDA and the company "really hasn't happened in this case," Williams said, noting that no meeting between the agency and the company was held to design the pivotal study. "Clearly, this isn't the ideal way we would like to see what we review be handled," he remarked.