CIBA-GEIGY COMPARISON OF VIOFORM-HC TO DESI EFFECTIVE STELAZINE
Executive Summary
CIBA-GEIGY COMPARISON OF VIOFORM-HC TO DESI EFFECTIVE STELAZINE is inappropriate, FDA counsel stated in a March 10 submission to FDA Administrative Law Judge Daniel Davidson. The agency's filing was in response to a Ciba-Geigy motion requesting the Administrative Law Judge to consider facts pertaining to the approval of SmithKline & French's Stelazine (trifluoperazine HCl) in its review of Vioform-HC (iodochlorhydroxyquin with hydrocortisone). "Ciba-Geigy's request for official notice should be denied because the Stelazine facts are totally irrelevant to the issues concerning Vioform-Hydrocortisone," FDA asserted. In its motion, Ciba-Geigy stated that FDA approved Stelazine as effective for the treatment of non-psychotic anxiety "on the basis of a single multicenter clinical study of 415 patients with a primary diagnosis of generalized anxiety disorder." The company said that in the Stelazine study, "a statistically significant result for treatment comparisons was defined as a one-tailed p-value of 0.05 or less." In addition, the 11 investigators in the study "were randomized by FDA into two substudies prior to the sponsor's analysis," Ciba-Geigy said. A DESI hearing on Vioform-HC was held through November. In a post-hearing brief, Ciba-Geigy argued that one of the studies it submitted in support of Vioform-HC efficacy, the Brecker study, is similar to a study on which FDA based its approval of Stelazine. Both studies, Ciba-Geigy said, involved split analyses of a single multicenter study that confirmed the pooled results. FDA announced that Stelazine was upgraded to effective in the Dec. 13 Federal Register ("The Pink Sheet" Dec. 16, "In Brief"). FDA maintained in its filing with the Administrative Law Judge that "there is no evidence whatsoever that Vioform-Hydrocortisone, the product at issue here, and Stelazine are even remotely similar; in fact, the drugs are highly dissimilar." Noting that Stelazine is a single-ingredient, orally-administered drug for treatment of non-psychotic anxiety and Vioform-HC is a two-way combination, topically-administered product for treatment of dermatologic conditions, FDA said the standards it applied in upgrading Stelazine "have no bearing whatsoever on the question of what is necessary to find Vioform-Hydrocortisone effective. More specifically, that one-tailed p-values and creation of substudies were appropriate in the context of Stelazine does not mean that either or both are appropriate in the Vioform-Hydrocortisone efficacy studies." Addressing the Brecker study in a March 3 post-hearing brief, FDA said the results "are at best marginal . . . certainly not spectacular . . . and, given the small superior effect in only a narrow part of the study, are not enough to establish the steroid's contribution to the combination's effectiveness. FDA said Ciba-Geigy's decision to divide the Brecker data into two substudies was not provided for in the protocol and thus the reanalysis (ITALICS)post hoc "violates the basic principles of the scientific method." Furthermore, the agency said results of the substudies "are even less impressive than the study analyzed as a whole." In its post-hearing brief, Ciba-Geigy also cited FDA's decision on the efficacy of Syntex' Neo-Synalar (neomycin/fluocinolone). The company maintained that in the Neo-Synalar decision, FDA "approved a whole range of anti-infective/corticosteroid combinations for the treatment of secondarily infected steroid-sensitive dermatoses on the basis of studies of neomycin (the shared anti-infective) in combination with a single corticosteroid." One of the studies Ciba-Geigy submitted for DESI evaluation of Vioform-HC was a study of Vioform in combination with the corticosteroid Locorten (flumethasone pivalate), which the company says is relevant to assessing the efficacy of Vioform in combination with another corticosteroid (hydrocortisone). FDA stated in its post-hearing brief that "the data from the study of Locorten-Vioform are not relevant to an evaluation of the effectiveness of Vioform-Hydrocortisone because the two products contain different steroids of different potencies." The agency added that effectiveness cannot be established without data "showing that there is no negative (or positive) interaction between the different steroids and Vioform." FDA said that even if the data were considered, it does not establish superiority of the combination.