OTC TOPICAL NSAID TRIALS SHOULD BE PRECEDED BY PILOT STUDIES

OTC TOPICAL NSAID TRIALS SHOULD BE PRECEDED BY PILOT STUDIES since the effectiveness of the products has not yet been shown, FDA Pilot Drug Evaluation Staff Director John Harter, MD, suggested at the Dec. 6 meeting of the Arthritis Drug Advisory Committee. "Since no one has done a successful study to prove effectiveness...people should do pilot studies," Harter advised. Harter pointed out that pilot studies are essential to determine whether the sponsor has a good study methodology for showing effectiveness of the study drug. Clinical trials with topical nonsteroidal anti-inflammatory agents have been hampered by the lack of a standard therapy that could be used as a comparison. While several OTC topical NSAIDs are currently available on the market, FDA has classified the agents as Category III. Placebo-controlled studies are also difficult to conduct with topical NSAIDs because the application of the products calls for massaging the drug into the skin. The FDAer said that "in studies that I've seen, response rate to placebo and massage has been quite high." Harter also mentioned the possibility of using oral NSAIDS as active controls. However, he stressed that "it may not be the same mechanism and it might lead you astray." Therefore, Harter cautioned, "you may be forced in your assay development to use your drug and placebo and try to find an assay that can tell them apart." The FDAer added that sponsors will have to do more than one study. Harter suggested three study models for OTC topical NSAID efficacy trials: conducting studies in patients with acute painful injuries, which could be studied in a parallel design; conducting trials in patients with chronic painful conditions -- such as headache or dysmenorrhea -- which could be studied in a crossover trial; and the use of a surrogate patient group by inducing muscle pain through overexertion, which could be studied in a parallel or crossover trial. The muscle pain model may have certain benefits. Harter noted that "since you have two arms and legs it could lead to the possibility of bilateral" studies. Both preparations could be used on the same patient "which would lead to a direct comparison in the same individual and even the possibility of sequential analysis," he pointed out. Concerning entrance criteria, Harter emphasized that "whatever the sources of patients and the kind of injury or inflammation...you should use entrance criteria to produce as homogenous a patient poulation as possible." He continued: "Avoid patients that have different etiologies. I've seen trials where they've had OA [osteoarthritis], RA [rheumatoid arthritis], sprains, [and] strains." Trial sponsors should also "make sure that all patients have sufficient disease at baseline [and] each of the primary outcome variables, so that you can check on the clinically meaningful improvement on each patient," Harter said. Sponsors should "focus their proof of effectiveness on three or four primary efficacy outcomes at most and then one or two timepoints."