FDA BIOEQUIVALENCE REQUIREMENTS FOR GENERIC DIAZEPAM SHOULD INCLUDE SITE-OF-ACTION PHARMACODYNAMIC STUDIES, ROCHE TELLS FDA; VALIUM OFF PATENT

FDA bioequivalence requirements for generic versions of Roche's Valium (diazepam) should include a pharmacodynamic measurement of the availability of the drug at the site of action, Roche maintained in a Feb. 14 petition to the agency. Roche proposed that "the appropriate bioequivalence study must include a pharmacodynamic study to assess the availability of drug at the site of action, using the methods of QPEEG [quantitative pharmaco-electroencephalography] in human volunteers." The firm maintained that the test is necessary "because of the documented bioequivalence problem, particularly as shown by the QPEEG data, and the inability of standard pharmacokinetic test procedures to detect bionequivalency." Asserting that FDA's diazepam bioavailability guidelines are "inadequate," Roche filed a separate petition requesting FDA to withhold approval of or delay diazepam ANDAs and paper NDAs until adequate criteria for determining bioequivalence of diazepam products are established. Roche's petitions were filed just prior to the Feb. 27 patent expiration of Valium. A number of ANDAs for diazepam are pending at FDA, although reportedly none are close to approval. Roche asserted that FDA's diazepam bioavailability guidelines issued in December ("The Pink Sheet" Dec. 17, T&G-1) are "not adequate for the purpose of ensuring either the bioequivalence or the therapeutic equivalence of different brands of diazepam." In support of that position, Roche cited a recently-completed study which, utilizing QPEEG, found that two brands of diazepam marketed overseas produced different central nervous system responses yet "were found to meet the in vitro tests, and most of the in vivo standards, set forth in the FDA guidelines." Roche contends that QPEEG is a "sensitive, accurate and objective measurement of human brain activity." QPEEG studies "demonstrate that pharmaceutically equivalent diazepam drug products are bioinequivalent. The drugs do not become available at an equal rate to the target site of action, the benzodiazepine receptors in the brain," Roche stated. "These brands, however," Roche continued, "are pharmaceutically equivalent and would probably be deemed bioequivalent if tested in accordance with the FDA guide." The petition states that "in the event FDA determines that a QPEEG pharmacodynamic study is not feasible as a bioequivalence standard," Roche "proposes that the FDA adopt a multipledose steady-state with replicate design, rather than a single-dose study for pharmacokinetic determination." The study, designed to be completed in 24 subjects in a randomized fashion, according to a two treatment, two period crossover design, should be expanded from the FDA guideline recommended 10 days to 14 days for each leg of the crossover, Roche suggested. Roche also recommended that the alternatively proposed pharmacokinetic bioequivalence study should have different plasma sampling times than FDA's protocol. "Plasma samples should be taken at . . . 0, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, and 12 hours on days 1, 13 and 14 of each treatment," Roche stated. The FDA guideline states that blood samples "should be collected at 0, 0.5, 1, 2, 3, 4, 6, 12, 24, 96, 144, 168, 192, 216 and 240 hours." Roche further proposed "that all [diazepam] dosage strengths be evaluated through in vivo bioequivalence testing." FDA's guideline allows 2 mg and 5 mg tabs to be deemed bioequivalent in lieu of in vivo studies if "the 2 mg and 5 mg tabs are composition proportional in their active and inactive ingredients to the 10 mg tab" found bioequivalent in vivo and "the 2 mg and 5 mg tablets have satisfactory dissolution performance compared to the 10 mg tab." Roche explained that "the demonstrated inability of dissolution testing to assure bioequivalence at the site of action, as determined by in vivo QPEEG studies, prevents any reliance for in vitro tests as a substitute for in vivo tests." Asserting that FDA's "dissolution guidelines are adequate for the purpose of determining pharmaceutical equivalence . . . but wholly inadequate for the purpose of determining bioequivalence . . . Roche proposed that pH-dissolution rate profiles be performed on each dosage strength using USP XXI(END ITALICS) method I or II at a minimum of 3 pH conditions [with] a minimum of 12 tabs" at each pH. FDA's methodology provides for the testing of 12 units at a single pH. Should FDA decline to revise the guidelines and proceed to use the guidelines as a basis for approving ANDAs, Roche is requesting "that FDA make no determinations of therapeutic equivalence among different diazepam products based on the tests set forth in the FDA guide. Roche further petitions that FDA establish appropriate standards beyond mere single-dose pharmacokinetic studies prior to listing any diazepam products as therapeutically equivalent to Valium."