GLAXO’s ZOFRAN RECOMMENDED FOR APPROVAL IN PREVENTION AND TREATMENT OF POST-OP NAUSEA AND VOMITING; ZOFRAN USE NOT ADVISED FOR ALL PROCEDURES

Glaxo's Zofran (ondansetron) should be approved by FDA for use in 4 mg. injections pre-operatively or post-operatively for the prevention and treatment of post-operative nausea and vomiting. FDA's Gastrointestinal Drugs Advisory Committee concluded Sept. 3. The new indication would significantly broaden the clinical potential for Zofran under approved labeling. First cleared by FDA in January 1991, the serotonin 5HT[3] receptor antagonist currently is approved for the prevention of chemotherapy-induced emesis, which has an occurrence rate of about 2.5 mil. episodes annually. Glaxo estimates that there will be approximately 30 mil. surgical procedures performed annually by 1995. A prime post-op submarket for Zofran is women undergoing laparoscopy and gynecological surgery, since there is about a 35% incidence of emesis among these patients. In general, there is a higher rate of post-operative nausea and vomiting in women compared to men. Data submitted by Glaxo in support of the ondansetron supplemental NDA was generated from clinical trials in which almost 90% of the patients were women, most of whom were undergoing either laparascopy or gynecological surgery. The committee was unanimous in its opposition to approval of ondansetron for "all" surgical procedures. One concern expressed by committee members is that the 5HT[3] antagonist may reduce bowel motility and could cause intestinal obstruction in patients undergoing abdominal surgery. FDA Gastrointestinal and Coagulation Drug Products Division Director Stephen Fredd, MD, indicated that ondansetron will probably not be contraindicated for any procedure, but that labeling will "detail what is known and what is not known" about ondansetron relative to certain procedures. Addressing an FDA question about how Zofran might be used in the post-operative setting, Glaxo consultant Bernard Wetchler, MD, University of Illinois, commented: "The significant problems -- multiple retching, long periods of nausea, protracted vomiting -- those are the patients that we feel should be treated." Wetchler said that with regard to "every single patient undergoing surgery and anesthesia, should they be given a drug for emesis? No, not every single patient." He added that "clinicians are looking for a drug that is relatively clean, that we can treat patients with, and [that] doesn't provide [certain] symptoms -- allowing our patient to go home and feel satisfied." Currently marketed therapies used for the treatment of post-op nausea and vomiting -- e.g. Janssen's Inapsine (droperidol) and metoclopramide -- cause significant side effects, committee members agreed. Adverse events associated with these drugs include drowsiness, extrapyramidal symptoms, hypotension, blurred vision and dysphoria (restlessness), which often can delay a patient's hospital release. In the clinical trials presented by Glaxo to the committee, Zofran demonstrated a "wide margin of safety," committee reviewer Burks remarked. The most frequent adverse events occurring in patients receiving ondansetron were headache, dizziness and muscle pain. However, the frequency of adverse events in the ondansetron group was not significantly different from the placebo group. Glaxo Associate Clinical Research Director Alan Joslyn, PhD, presented efficacy data from four Phase III trials conducted with ondansetron. In the two prevention studies, ondansetron or placebo was given pre-operatively to 1,169 women undergoing laparoscopy or general surgery. Doses of 1 mg, 4 mg or 8 mg of ondansetron or placebo were diluted to 20 ml with normal saline and infused over a period of two to five minutes prior to the induction of anesthesia. Patients received a general anesthesia regimen of an induction agent (thiopental, methohexital or thiamylal), an opioid (alfentanil or fentanyl) and nitrous oxide (use of supplemental isoflurane was allowed). Ondansetron's efficacy was assessed for 24 hours following administration -- during the first two hours by a nurse observer and from two to 24 hours via patient diary. In a combined analysis of the two prevention studies, ondansetron showed a complete response rate of 69% for the 4 mg dose, which was statistically significantly superior to the 45% complete response rate for placebo. A complete response was defined as no episode of vomiting. In addition, 40% of patients receiving 4 mg ondansetron preventively experienced no nausea compared to 30% of patients on placebo. The two ondansetron treatment trials enrolled a total of 1,022 women and men undergoing various surgeries. However, 90% of the patients in the first treatment trial were women undergoing predominantly gynecological procedures. Women accounted for 80% of the patients in second treatment trial, with most of them also having gynecological surgeries. In the trials, ondansetron-treated patients were twice as likely to have a complete response than patients receiving placebo, 45% compared to 21%. The committee recommended that Glaxo conduct post-marketing studies in certain populations not well represented in the clinical trials. The committee suggested that the company gather information on children and the elderly, obese patients, patients undergoing other types of procedures such as abdominal surgeries, and patients receiving repeat doses of ondansetron. Since a vast majority of the patients in the ondansetron post- operative nausea and vomiting trials were women, especially in the prevention trials, FDA asked the committee whether it thought that the efficacy data of the drug in this setting may be extrapolated to men. "While males were not included in the prevention studies, there is no a priori reason to expect males to differ materially from females in their response to ondansetron," Burks said. The committee members agreed that the data in the NDA under review was not conclusive concerning efficacy in men. However, they were reassured by information presented by Glaxo showing that ondansetron's control of chemotherapy-induced emesis is superior in men compared to women. While women predominated in the clinical trials, pharmacokinetic data presented by Glaxo was exclusively conducted in healthy male volunteers, FDA Biopharmaceutics Division reviewer Lydia Kaus, PhD, noted. Echoing Kaus' observation, Committee Chairman Eugene Schiff, MD, University of Miami School of Medicine, remarked that although vomiting and nausea are "disproportionately present among women, we have women under- represented in these volunteer trials." He told the company that "in the future you want to correct this." Another concern raised early in the committee discussions addressed the discrepancy between the timing of ondansetron administration in the clinical trials (over a two-to-five minute period) and the sponsor's requested wording in labeling for dosing of "not less than 30 seconds." Committee member David Earnest, MD, University of Arizona College of Medicine, commented: "I was concerned that the recommendation might be implied that it can be given in 30 seconds, which is what an anesthesiologist will likely do. If that is the case, we have inadequate safety data to support that." The committee, however, agreed that data presented by Glaxo from four healthy male volunteers, who were given 8 mg ondansetron over a 30-second period, demonstrated the safety of the dosing time period. None of the volunteers experienced unusual adverse reactions from the drug dosing. Committee consultant James Eisenbach, MD, Bowman Grey School of Medicine, suggested that FDA "include a brief description of the fact that [ondansetron] has little or no respiratory or cardiovascular effect in the healthy volunteers." While the worldwide clinical safety experience in over 15,000 patients with Zofran appears to raise little concern, earlier this year Glaxo was asked by FDA to address reports from the Royal Prince Alfred Hospital in Sydney, Australia that linked the occurrence of vascular occlusive events, including seven deaths, to ondansetron therapy in cancer patients. After a thorough review of its safety database and a literature, the company concluded that the overall incidence of adverse events in patients receiving ondansetron was not substantially different than those for patients receiving placebo or comparator agents. The firm attributed the reports of vascular occlusive events to cancer or cisplatin therapy. Mark Kris, MD, Memorial Sloan-Kettering Cancer Institute, who first reported the possible association of ondansetron therapy and vascular occlusive events in 1987, told the committee that he came to the same conclusion that "ondansetron is not involved."