BMS Caffeine Adjuvancy Study Acetaminophen-Only Arm Unnecessary – FDA

An acetaminophen-only treatment arm in Bristol-Myers Squibb's planned clinical study on the efficacy of different caffeine doses in analgesic adjuvancy is unnecessary, FDA says in a recent letter.

The Feb. 5 letter responds to a July 30 submission by BMS outlining a proposed study intended primarily to persuade FDA to recognize the increased analgesic effect provided by a higher dose of caffeine - 130 mg - than is currently accepted (¹ (Also see "Caffeine Dose Response Study To Support Category I Status Proposed By BMS" - Pink Sheet, 13 Aug, 2001.), p. 3).

BMS proposed comparing acetaminophen 1,000 mg with different combinations of aspirin, acetaminophen and caffeine. However, FDA states, "for the purpose of establishing the caffeine dose response, the comparison of efficacy between the aspirin/acetaminophen/caffeine (AAC) combinations and acetaminophen 1,000 mg is not very informative."

"The primary objective of the study should be the evaluation of the relative efficacy of the AAC combinations to placebo and with each other," the agency maintains.

BMS had proposed that results from the study could be extrapolated to support the combination of acetaminophen 1,000 mg and caffeine 130 mg. FDA has recognized the adjuvancy of caffeine 65 mg in combination with aspirin or with aspirin and acetaminophen together but not with acetaminophen alone (² (Also see "OTC analgesic 65 mg caffeine limit "not supported" by the record, Bristol-Myers tells FDA." - Pink Sheet, 22 May, 1995.), p. 7).

Inclusion of the acetaminophen-only arm also could provide BMS with fuel for claiming its caffeine combinations are superior to competing acetaminophen-only analgesics.

The firm's Excedrin Extra Strength contains a per-dose combination of aspirin 500 mg, acetaminophen 500 mg and caffeine 130 mg, while Aspirin Free Excedrin contains acetaminophen 1,000 mg and caffeine 130 mg per dose.

FDA recommended the BMS study include another pain model in addition to headache so as not to limit the applicability of results.

"It is not clear that the results from a headache study can be used to support the other general claims available for internal analgesics," the agency says. "Caffeine may have unique benefits in a headache model that may not be apparent in other pain models (e.g. dental pain models)."

FDA further questioned the number of subjects and primary efficacy variables planned for the study. The agency notes in analgesic trials for combination products, "the sample size of the treatment groups... usually contain 80-90 subjects per study arm," adding: "Please explain why 400 subjects per arm are needed."

Stating "for an acute analgesic claim, a drug should work within one hour of ingestion," the agency also criticized BMS' proposal to assess one designated "summary" primary efficacy parameter at four hours.

Instead, FDA states, "you should use other primary efficacy variables such as pain intensity and pain relief that need to be evaluated every 20 minutes or the first two hours and then at three hours post dosing in addition to the other parameters that you are proposing to study," such as time to onset of meaningful pain relief and time to rescue medication.

In addition, "the two-stopwatch method is better than the one-stopwatch method to measure the onset of pain relief since it provides information to calculate the time to perceptible pain relief and the time to meaningful pain relief."

FDA also requests clarification of how evaluation of "other measures" such as "muscle stiffness, psychic tension, degree of relaxation and interference with daily activities" will be handled.

Although BMS had requested a meeting on the study protocol, FDA notes it "does not believe that a meeting is warranted at this time until major revisions to the design of the study are made." A meeting to discuss unresolved issues will be considered once the agency has received a revised and more complete protocol, the letter states.

The agency also explained its comments on the submission were limited to the efficacy evaluation planned because it had not yet reviewed the safety information provided by BMS.