PFIZER's NORVASC RECOMMENDED FOR APPROVAL FOR TREATMENT OF ANGINA AND HYPERTENSION; COMMITTEE ASKS FOR BETTER PHARMACOLOGIC STUDIES
Pfizer's calcium channel blocker Norvasc (amlodipine) was recommended for approval for treatment of angina and hypertension by FDA's Cardiovascular and Renal Drugs Advisory Committee at its June 7 meeting. The committee voted 7-0 to recommend that Norvasc be approved as an antihypertensive. Committee member Craig Brater, MD, Indiana University, said that Pfizer had provided at least four adequate and well-controlled studies indicating an antihypertensive effect for the drug. Brater said he saw no contrary data. The committee had more trouble with Pfizer's amlodopine angina data base. The committee supported an indication for chronic stable angina by a 5-2 vote, and recommended approval as a treatment of vasospastic angina 4-3. Pfizer filed the NDA (19-787) for Norvasc in December, 1987. The once-daily dihydropyridine calcium channel blocker is approved in 31 countries and has been launched in 16. Norvasc is Pfizer's follow-up to its successful nifedipine line (Procardia and Procardia XL); at a recent Mabon Nugent research conference, Pfizer's CFO Henry McKinnell projected that amlodipine would be his company's "largest selling cardiovascular" by the mid-1990's ("The Pink Sheet" Dec. 10, In Brief). The committee had no difficulty with the general safety and efficacy of Norvasc as an antihypertensive. Members expressed concern, however, over the design of certain studies intended to address pharmacokinetic aspects of amlodipine. Most protocols incrementally increased dosage at 2-week intervals, and studies showed that amlodipine, while reaching a plateau plasma concentration after 4-6 days, continued to demonstrate increasing antihypertensive effects after 4 weeks. Pfizer's proposed package insert suggests titration of the drug to full dosage over a 7-14 day period. Milton Packer, MD, Mount Sinai Medical School, noted that because amlodipine "is a drug which will take a long time to reach a plateau, and..... a longer time to reach an antihypertensive effect," there is "clearly insufficient information to increase the dose at an interval shorter than 2 weeks." Packer recommended that "the statement be made that the full antihypertensive effects of a dose may not be evident for 4 weeks." Although the panel found that Pfizer's single dose dose- proportionality studies looked "pretty good," they questioned Pfizer's multiple dose-proportionality study. FDA Cardio-Renal Drugs Division Director Raymond Lipicky, MD, asked Pfizer for "any other data..... outside of this poorly designed study." Brater suggested that "perhaps it might be worthwhile to do a really tight study to address this issue and close all the gaps." Brater specifically wanted better information on the oral bioavailability of the drug. Different studies showed bioavailability ranging from 64% to 90%. The committee gave Pfizer "strong motivation" to clarify the issue, Brater said. He also suggested that Pfizer perform drug-drug interaction studies with erythromycin and estradiol. The committee voted that a 2.5 mg initial dose be recommended in some cases, especially for the elderly or patients with liver impairments, while a 5 mg dose was recommended as an acceptable starting dose for younger patients. Pfizer had suggested that 5 mg be the initial dose. The maximum dose recommended by the committee was 10 mg. The recommendation to vary the initial dose according to individual patients' conditions was made pursuant to data which showed that the elderly and patients with renal disease had absorption rates twice those of patients under 65. However, no cognizable risk was associated with this increased sensitivity. "We have to get into the studies and see what happens in old and young people," commented Peter Kowey, MD, Lankenau Hospital. For chronic stable angina treatment, Pfizer presented data on 2,213 patients. The database included five monotherapy trials and three add-on studies that were included in the NDA, as well as two more recent trials. Kowey summarized the committee's frustration with the data: "It's a patchwork. I don't think there was any trial that went all the way across every parameter." Committee Chairman Craig Pratt, MD, Methodist Hospital, added that the results were "consistently wimpy..... even when the changes are significant statistically, they're not massive changes." Both ultimately voted in favor of approval for the indication. Pfizer, in an effort to demonstrate significance in the data, presented a pooled analysis of the studies which showed statistically significant improvement in all categories. Kowey, in particular, supported the pooling. He said he was "comfortable" with it because the individual studies had the same design, similar patient populations, and all showed "trends" in the right direction. Committee member Packer, however, said that "if the only analysis shown were pooled analysis, and there weren't two studies that beat placebo..... I would be skeptical." He added: "I'm not concerned so much about this NDA as about the general concept that by saying that a pooled analysis is good and gives a high degree of comfort, we might encourage future presentations to pool all their data to make all of their p-values astronomically low." Packer also ultimately voted in favor of the indication. Questions prepared by the reviewers suggested that FDA shares Packer's skepticism towards the approach: "It is possible that without the pooled analysis (looking solely at individual trials) one would conclude that the evidence for an amlodipine effect [on angina] is marginal." At the meeting, Lipicky added: "My view is that pooled analysis doesn't change the individual [studies]." The committee's disagreement over the chronic stable angina indication presaged deeper disagreement over the vasospastic angina indication. Pfizer presented one 52-patient trial studying the indication which showed a statistically significant decrease in the angina attack rate in the amlodipine arm compared to placebo. There was also a reduction in the withdrawal rate from the Norvasc arm, although it was not significant. One study would be adequate for approval, FDA's questions noted, if the drug showed "clear anti-anginal and anti-ischemic effects." The committee, however, could not conclude from Pfizer's database that amlodipine had a clear anti-ischemic effect. The committee instead asked that the data be reanalyzed to account for non-responders who were switched from placebo to Norvasc in all protocols, and so might have biased the results against amlodipine. Kowey, before voting to recommend the indication, said that he was "wishy-washy" about whether the data sufficed. Immediately before the vote, Lipicky stressed that, in light of the lukewarm opinion of the strength of the chronic stable angina data, "you really do have to be convinced that you..... have a fairly positive trial." He added: "It's very difficult to know clinically that a drug is or is not working on this condition." The committee voted 4-3 to recommend approval. A final labeling issue was the discussion of Norvasc's effect on patients with congestive heart failure. Pfizer presented data from a 118-patient study that showed a slight improvement in the patient's disease. Other calcium channel blockers have been associated with a worsening of the disease. The committee agreed that this information should be included in the label in "neutral" language. The goal of the study, Packer said, was to demonstrate safety, so no statement should be made about efficacy which could become "an implicit indication." The statement should point out, Packer added, that the trial included only mild to moderate CHF patients. "The drug shouldn't be used" in patients with severe CHF, he said.
Sign in to continue reading.
New to Pink Sheet?
Start a free trial today!
Register for our free email digests: