XOMA's XOMAZYME-CD5+ (H65-RTA) RECOMMENDED FOR APPROVAL BY FDA ADVISORY COMMITTEE DESPITE CBER REVIEW GROUP's VIEW THAT DATA DOES NOT SUPPORT LICENSURE

Xoma's XomaZyme-CD5+ was recommended for approval for use in treating steroid resistant acute graft-versus-host disease in bone marrow patients by FDA's Biological Response Modifiers Advisory Committee on June 10. The data presented by Xoma "provide sufficient information for us to recommend that H65-RTA be approved for the therapy of steroid resistant acute graft-versus-host disease" in patients having undergone allogeneic bone marrow transplants, Committee Chairman Jerome Groopman, MD, New England Deaconess Hospital, concluded. Xoma's first product license application, XomaZyme-CD5+, also called OrthoZyme-CD5+ (the product will be marketed by Ortho Biotech), is the first immunoconjugate to be reviewed by an FDA advisory committee. The PLA for H65-RTA, which destroys the T- cells that mediate GvHD, was filed in December 1988. Xoma's second monoclonal antibody-based immunotoxin product, E5, and Centocor's Centoxin are awaiting review by the committee for gram negative sepsis. The committee voted seven-to-one in favor of approval despite reservations about the significance of the H65-RTA data when compared to historical controls. "From a pure regulatory standpoint, no I don't think that we have shown that this data is conclusive," remarked Frederick Appelbaum, MD, Fred Hutchinson Cancer Research Center. However, Appelbaum noted that "as a physician . . . I want this to be available." The advisory committee recommended approval for XomaZyme-CD5+ over the objections of FDA's reviewing committee, which disliked the company's use of historical controls for an efficacy comparison. The agency argued that the treatment of bone marrow transplant patients has evolved since Xoma's collection of the historical data from 1975 to 1987. FDA also asserted that the historical database Xoma used in the comparison was missing data that could have biased the study results in favor of XomaZyme- CD5+. While agreeing that Xoma's comparison of CD5 treatment to historical controls was not ideal, the advisory committee was unable to suggest a better efficacy protocol because of the lack of an effective standard treatment for the refractory condition. Groopman acknowledged that "there doesn't seem to be really a valid active control group" for comparison. He commented that the committee has "a sense that there is activity here despite all the discomfort or difficulties in terms of using a historical database." The committee urged the sponsor to seek out more timely control data, but did not condition the approval of the drug on the new data. Appelbaum observed that the committee has been in the situation before of evaluating drugs on the basis of "uncontrolled data." Referring to the first pass of Cetus' Proleukin (interleukin-2) and Immunex' Leukine (GM-CSF) before the committee last July ("The Pink Sheet" Aug. 6, p. 6) Appelbaum observed that "in some cases" the advisory committee has withheld approval recommendations "because the appropriate controls were not there." Possibly referring to GM-CSF, Appelbaum noted that "unlike one of the agents that [the panel] dealt with in the past, this is an agent which is unlikely to be abused." Alluding to Proleukin's side effect profile, he noted that "given the toxicity profile we have seen, unlike another agent which we failed to approve, this does not by itself need people to be in the ICU." Of the 4,000 allogeneic bone marrow transplants that will be performed this year, approximately 1,500 patients will develop steroid refractory acute GvHD. "We are talking about a very limited group of patients for whom this would be useful," Appelbaum remarked. Xoma VP-Clinical Research Samuel Saks, MD, presented data from two Phase III trials: studies 8705 and 8901, which included 69 and 75 patients, respectively. Allogeneic bone marrow transplant patients who had failed steroid therapy were treated intravenously with 0.1 mg/kg/day of H65-RTA over 14 days. The treatment groups were compared to data from 157 historical control patients gathered by the Fred Hutchinson Cancer Research Center in Seattle. Of the patients treated with H65-RTA, 17% in the 8705 study and 21% in the 8901 study had complete responses compared to 2% of historical control patients at 45 days following the development of acute GvHD. A total of 31% of the patients receiving H65-RTA showed some response to treatment at day 45 in the 8705 study and a total of 40% of the patients in the 8901 study showed at least some response compared to 10% of the patients in the control group. Of the 27 patients who achieved a complete response in the trials, 22 continued to show no sign of disease after 100 days. In addition, time to treatment failure (defined as progression of disease or death with acute GvHD as a contributing cause) was reduced with H65-RTA therapy. Fifty-one percent of patients in the 8705 and 50% of patients in 8901 experienced treatment failure compared to 70% of control patients. Saks also noted that there was an improvement of one grade in disease stage in the H65-RTA group compared to a minimal improvement in the control group. Although survival was not a primary endpoint of the studies, the 8901 study showed a statistically significant difference in survival between the H65-RTA group and the historical control -- 44% versus 37%. The greatest survival benefit was seen in patients with disease grades 1-3. Those patients showed a 61% survival rate compared to 39% in the historical control. Patients with grade four disease showed no survival benefit. The 8705 study showed no statistically significant difference in survival with H65-RTA compared to the historical control. The major toxicity seen with H65-RTA was renal dysfunction in 11.5% of patients. The treatment was also associated with peripheral edema and a decrease in serum albumin. The side effects were reversed upon discontinuation of therapy. Eleven patients had antibody responses to the murine monoclonal antibody-Ricin A product. The advisory committee arrived at an approval recommendation despite the FDA review team's problems with Xoma's historical control data. FDA Review Committee Chairman Ezio Bovini, MD, told the committee that "after a review of the data submitted . . . and an independent statistical and clinical analysis performed by the Center for Biologics, the FDA review committee has reached the conclusion that the evidence available do not support licensure of H65-RTA for the treatment of steroid resistant acute GvHD." Bovini also told the committee that "the product has not been satisfactorily characterized with respect to potency, specificity and stability." He noted that FDA is working closely with Xoma to resolve those manufacturing issues. The manufacturing problems were the subject of an hour-long closed session during the meeting. Bovini said that Xoma's comparisons of H65-RTA to historical controls had ignored the fact that almost 21% of patients from the historical control group were excluded because of missing data at day 45. Bovini also pointed out that the survival benefit seen in 8901 was not reproduced in 8705. When FDA adjusted for the missing data by assigning those patients a stage 0 designation, or complete response, the statistical significance between the groups disappeared. At day 45, the complete response rates were 16% for study 8705, 21% for 8901, and 15% for control. The overall response rates changed to 28% in study 8705, 41% in study 8901 and 31% in the historical control. FDA's viewpoint was countered by bone marrow transplant experts from M.D. Anderson, Fred Hutchinson, and Memorial Sloan- Kettering. "Having an alternative therapy and hearing the testimony of many of the people in the transplant community convinces me [given] the low toxicity that this [product] should be available for patients with this condition," committee member Jordan Gutterman, MD, M.D. Anderson Cancer Center, said.