This article considers when a single study presents both: 1. The same strength of evidence for a drug or biologic effects as the usual Food and Drug Administration (FDA) two-study paradigm, and 2. Also gives sufficient evidence of replicability as in two studies done at disjoint sets of institutions. It is argued that 1. holds if the two-sided significance level is 0.00125 and 2. holds if the clinical sites are divided into two subsets that maximize the smaller sample size that each is statistically significant at the 0.05 significance level. Other approaches to satisfying 2. are also addressed. The approach is illustrated with data from the Condylox® Gel 0.5% (podofilox gel) versus vehicle control.
Food and Drug Administration.Statement Regarding the Demonstrations of Effectiveness of Human Drug Products and Devices. Federal Register.Vol. 60, No. 147, August 1, 1995, Docket No. 95–0230, pp. 39180–39181.
2.
Guidance for Industry: Providing Clinical Evidence of Effectiveness for Human Drug and Biological Products.Draft Guidance—not for implementation. Released for comment on March 13, 1997. U.S. Department of Health and Human Services, FDA, CDER, CBER, March 1997.
3.
TyringSDEdwardsLCherryLKRamsdellWMKotnerSGreenbergMDVanceJCBarnumGDromgooleSKilleyFPToterT.Safety and efficacy of 0.5% Podofilox Gel in the treatment of anogenital warts: A double-blind, vehicle-controlled study. Arch Dermatol.1998.
4.
FisherLDThe use of one-sided tests in drug trials: An FDA advisory committee member's perspective. J Pharm Stat.1991;1:151–156.
5.
HusterWJLouvWCDemonstration of the reproducibility of treatment efficacy from a single multicenter trial. J Biopharm Stat.1992;2:219–238.
