BENZODIAZEPINES WITH SLOWER RATE OF ACTION MAY HAVE LESS ABUSE POTENTIAL; SCHEDULING FOR EXISTING BENZOS RULED APPROPRIATE BY FDA HEARING PARTICIPANTS

Slower acting benzodiazepines may have a lower abuse potential, John Roache, PhD, University of Texas, suggested at a Sept. 11 FDA public hearing on criteria for benzodiazepine scheduling in Washington, D.C. FDA convened a panel to examine whether data support differential scheduling of individual members of the benzodiazepine class, which have been collectively placed in Schedule IV since 1984.

"I would suggest that there are in fact differences in abuse potential within the chemical class of benzodiazepines," Roache said. "And many people would suggest that it's related to pharmacokinetics. That is, the more rapid onset drugs have a higher potential for abuse than do the slow onset compounds."

Roache cited the barbiturate class as a possible model for benzodiazepine differential scheduling. "We see that there is differential scheduling amongst these compounds. And why is that? It's largely because of pharmacokinetic difference. The shorter, more rapidly acting compounds are scheduled higher."

Benzodiazepines are generally regarded as being less likely to lead to dependence, Roache noted, citing a study comparing effects of triazolam (Pharmacia & Upjohn's Halcion and generics) and pentobarbital on drug abusers. The study demonstrated that "it takes a relatively larger amount of triazolam to produce a relatively low dose barbiturate effect, indicating a lower abuse potential for triazolam," Roache said.

Roache cited oxazepam (Wyeth-Ayerst's Serax and generics) and diazepam (Roche's Valium and generics) as examples within the benzodiazepine class where pharmacokinetic differences are significant. Referring to a 1984 Journal of Pharmacology and Experimental Therapeutics article by Griffiths et al., Roache noted that even after extraordinarily high doses of oxazepam (480 mg), subjects generally preferred the lower dose of diazepam (40 mg). "But as you went up in diazepam dose, 80 mg or 160 mg diazepam was clearly preferred over oxazepam."

"Now, these are some of the most convincing data that exist in the human laboratory for differences amongst the benzodiazepine class," Roache contended. These data reflect what is known epidemiologically regarding the greater abuse of diazepam than oxazepam on the street in both the U.S. and Sweden, Roache observed. "We believe that the explanation for this is pharmacokinetic. That is, that oxazepam has a slower onset of action."

While the pharmacokinetic properties of benzodiazepines might serve to explain differences among compounds in the class, "in and of themselves, I don't think we can predict `here's a fast onset compound; it has to be regulated more stringently,'" Roache said.

A 1995 article by Mumford et al. in the Journal of Pharmacology and Experimental Therapeutics further supports the view that individual benzodiazepines can be differentiated, Roache maintained. The study compared alprazolam (Pharmacia & Upjohn's Xanax and generics) to abecarnil, which is currently in Phase III trials sponsored by Novo Nordisk. In ratings predictive of abuse among sedative drug abusers, such as worth on the street and likelihood of taking again, "abecarnil did not even come close to alprazolam," Roache said.

As a new drug, Roache contended, abecarnil "would need to be scheduled if it's going to be approved in the U.S., and this kind of data could inform that scheduling decision."

Addressing the question of whether individual benzodiazepines should be rescheduled, Roache concluded that those compounds "with a higher [abuse] potential could be considered for increased restriction. Whether or not any of the low potential should be deregulated would have to bring together all the issues of physical dependence."

"Even some of our lowest benzodiazepines with a classic agonist profile, like oxazepam, we never said it was a lower potential for abuse than phenobarbital, which is Schedule IV....Now, this might be truer for some of the newer compounds, like the partial agonists that have no real indication of abuse. But for the existing compounds, I'm not sure I can recommend any of those for descheduling," Roache said.

David Greenblatt, MD, Tufts University agreed that there was little danger of abuse liability with partial agonists but stressed that their lack of efficacy makes them "of questionable value" for patients. "Even if you get to full receptor occupancy, you do not get full agonist effects....In fact, many of the studies show that the efficacy is not distinguishable from placebo."

Current regulation of benzodiazepines is "acceptable and reasonable," Greenblatt maintained, and existing evidence does not support eliminating control of the class. A possible alternative to scheduling, however, could be labeling with proper warnings for all benzodiazepines, Greenblatt said.

Greenblatt did suggest that Carter-Wallace's Doral (quazepam) "appears to be a low abusability drug" because it is a prodrug, and that differential scheduling "makes sense from a kinetic standpoint." However, he added that "in the constellation of current controls I feel that they are appropriate and adequate." Carter-Wallace has a petition pending at FDA for descheduling the drug ("The Pink Sheet" Sept. 15, p. 12).

Temazepam was discussed by Greenblatt as an example of the role dosage form can play in abuse liability. Temazepam is marketed in capsule form both generically and by Novartis as Restoril and as a liquid capsule in the U.K. "Liquid capsules are candidates to be very rapidly absorbed, and that's why temazepam in the United Kingdom, which is a very juicy gelatin-filled capsule, hits you very fast, as opposed to the temazepam in the United States. And the abuse issues are vastly different between the two nations for that reason," Greenblatt said.

Appearing for Roche, James Woods, PhD, University of Michigan, argued that further studies are needed to determine the abuse potential of individual benzodiazepines. "Laboratory studies in animals or humans have shown only slight, hair-splitting differences that are not sufficiently robust to support making differential scheduling decisions." The relevance of existing studies "to abuse liability in the real world remains unclear," he said. Reliable interpretation of such a potential "requires a degree of sophistication that we haven't attained."

The factors determining preferences among drug abusers for some benzodiazepines over others have not been systematically investigated, Woods contended. Citing such factors as cost, familiarity and cachet, Woods declared: "We simply don't know the extent to which pharmacologic factors play a role in determining these preferences. However, from the available evidence, it seems clear that pharmacologic factors are not the sole determinants [of abuse]."

In the absence of more conclusive data, Woods proposed the "establishment of an independent panel...charged with a number of interrelated mandates bearing on the assessment of benzodiazepines for purposes of guiding public policy."

The panel should be composed of FDA, the National Institute on Drug Abuse, the Drug Enforcement Agency, and the pharmaceutical industry, Woods suggested. Additional representation should include psychiatric and neurologic epidemiologists, clinical practitioners, primary care physicians in both anxious and drug abuse patient populations, and an FDA attorney. The panel's "central mandate should be the development or refinement of criteria for scheduling benzodiazepines and related drugs which would benefit regulators and industry," Woods said.

DEA's James Tolliver, PhD, reported that in 1996 there were over 4,000 seizures by the agency involving flunitrazepam, which Roche markets internationally as Rohypnol, but is not approved in the U.S. Rohypnol was second to diazepam in numbers of dosage units seized last year, with 66,084 units, compared to 344,364 of diazepam. At a National Wholesale Druggists' Association meeting in Washington, D.C. Sept. 16, DEA's Chief-Liaison and Policy Section, Office of Diversion Control Tom Gitchel reported that benzodiazepines accounted for 19% of emergency room mentions of controlled substances compared to 18% for heroin and 36% for cocaine.