MEDIMMUNE RESPIGAM RSV IMMUNE GLOBULIN PIVOTAL TRIAL COMPROMISED BY METHODOLOGY CONCERNS, FDA ADVISORY CMTE. SAYS; APPROVAL NOT RECOMMENDED

MedImmune's pivotal trial for RespiGam respiratory syncytial virus immune globulin (RSVIG) is compromised by methodological problems, FDA's Blood Products Advisory Committee concluded at its Dec. 2 meeting. Committee Chairman Merlin Sayers, MD/PhD, Puget Sound Blood Center, said: "I don't believe the questions about randomization and blinding, particularly at [the clinical site in] Denver... [have] adequately been answered." Sayers added: "I'm also concerned that the follow-up of dropouts has not been addressed adequately." The methodological issues may have been the reason FDA postponed the advisory committee meeting from its initial date of Sept. 23 ("The Pink Sheet" Sept. 6, In Brief). Committee consumer representative Malcolm Schoen, MD, White Plains, New York, added: "I am encouraged about the data I see suggesting a decrease in morbidity, but I don't think it's ready. I think it needs very careful further study." The methodological concerns led the committee to conclude unanimously that the three-year multicenter trial was not adequate to establish the safety and efficacy of RespiGam for prophylaxis of RSV lower respiratory tract infection in infants with prematurity or bronchopulmonary dysplasia. Temporary committee member Paul Meier, PhD, Columbia University, called the committee's position "a very unhappy situation." RespiGam "might be a fine drug, a great drug, but the evidence before us is, in my opinion, quite unable to support such a conclusion," he said. Meier described the study results as a "tragic squandering of three years." MedImmune said Dec. 3 that it is "of course very disappointed by the action taken by [the committee], but we continue to believe that RespiGam can be of significant benefit to high risk infants." The company added: "Our intention is to continue our clinical studies program for RespiGam in accordance with the recommendations of the committee and to submit the results when they become available." Gaithersburg, Md.-based MedImmune licenses marketing rights to RespiGam, except in Massachusetts and Maine, from Massachusetts Public Health Biologic Laboratories. MPHBL submitted the product license application and an amendment to the establishment license application to FDA on Feb. 17. Results of the pivotal trial, conducted by Jessie Groothuis, MD, University of Colorado, et al, were published in the Nov. 18 New England Journal of Medicine ("The Pink Sheet" Nov. 22, T&G-8). The study, which was co-sponsored by the National Institute of Allergy and Infectious Diseases, involved 249 infants and young children with bronchopulmonary dysplasia, prematurity or congenital heart disease, who were at risk of RSV infection. The children were randomized to receive monthly infusions of either RespiGam at a high dose (750 mg/kg), RespiGam low dose (150 mg/kg), or no RespiGam. Groothuis presented the study results to the committee. "High- dose RSVIG proved very effective in reducing both the incidence and the severity of serious RSV lower respiratory tract disease," Groothius said. Seven high-dose RSVIG recipients developed RSV lower respiratory infection as compared with 20 control children. This "represents a 62% reduction in LRI incidence," she stated. "RSV disease severity...was also significantly less in high- dose recipients. Three high-dose recipients had severe disease as compared with 11 control infants." This "represented a 70% reduction in disease severity." Patients receiving RespiGam also had fewer hospitalizations, days in the hospital, ICU admissions, and days in the ICU than the control group, the investigator said. FDA reviewers expressed concern that the data were skewed by results at the Denver center. Dorothy Scott, NM, of FDA's Office of Blood Research and Review, explained that the nurse coordinator who randomized the patients in Denver "was not blinded in the study. She knew the patient's histories and the interim results of the study." Scott noted that the method, which was used for the first two years, "provides less protection against selection bias." Scott added that "treatment comparisons theoretically could have been significantly biased by dropouts in this study." During his review of statistical data, Cornelius Lynch, PhD, Office of Establishment Licensing and Product Surveillance, explained that there were 17 patients who were enrolled and eligible for the study that dropped out. Of the 17, nine were in the high-dose group and eight were in the low-dose group. "There was no followup on the dropouts, so consequently we do not have reliable outcome data on these 17 dropouts," Lynch said. "It is not known for certain which ones experienced RSV LRI and which ones did not." The FDAer pointed out: "The concern is that if the dropouts represent a group that's [at] higher than average risk of RSV LRI, then their removal from the analysis biases the comparison in favor of the treatment." He added that at least one of the dropouts in the high-dose group experienced RSV. Addressing the disparity in treatment effect observed at the Denver study site compared to the four other sites, committee member Susan Leitman, MD, National Institutes of Health, said that "the most severe RSV LRI...was seen at Denver...and the effects of the RSVIG were most pronounced in that group." She thought that the information would be helpful in determining "what patient group might benefit most." Referring to the more favorable effects seen at the Denver site, Leitman said: "It's almost as if there were one major study and then four satellite studies." Committee member Dennis Harpool, Blood Systems, Inc., Scottsdale, Ariz., said: "I feel that the Denver data is the data that's driving the study overall, and with that large disparity... additional data is needed." Providing possible reasons for the different efficacy results, Scott suggested the "altitude theory," which is "that RSV LRI frequency and severity is greater in Denver because of the higher altitude." When the committee was asked what additional data should be gathered to determine RespiGam's safety and efficacy, committee member Jane Piliavin, PhD, University of Wisconsin, suggested: "I think a new data collection should intentionally get patients at different altitudes who are similar kinds of patients. That way you would be able to see to what extent it's related to altitude." Another disparity involved the underlying conditions of patients at the centers. At the Denver site 78% of the infants had bronchopulmonary dysplasia, while 26% of the infants at the other four sites combined had BPD. Leitman said: "I think what is needed is more patients entered with bronchopulmonary dysplasia at non- Denver sites." Meier stated: "I see no choice but to get back together again in a hurry with a capable group to design a study." Some of the committee members were concerned about the six deaths that occurred in the two RespiGam-treated groups, while no deaths occurred in the group that did not get RespiGam. Five of the deaths occurred in infants who had congenital heart disease, and the other death occurred in an infant with bronchopulmonary dysplasia. Presenting a safety overview, MedImmune Exec VP and Medical Director Frank Top, MD, said that an "extensive evaluation demonstrated that RSVIG did not cause the fatalities observed in studied cardiac patients." To confirm that contention, "last winter we began a second comparative study in which we're striving to enroll 160 infants with congenital heart disease in each of a high-dose RSVIG group and a control group." He noted there have been three deaths "all due to heart disease; one in the treatment group and two in the control group." MedImmune expects to finish the study this winter. "Pending the results of this study, we have deferred seeking an indication for prophylactic use of RSVIG in children with heart disease," Top said. The committee voted unanimously that additional studies are needed in infants with congenital heart disease.