Bristol/Otsuka To Submit Abilify Long-Term Efficacy sNDA As Phase IV

Otsuka/Bristol-Myers Squibb are required to submit an NDA supplement before the end of 2002 on long-term efficacy of its schizophrenia drug Abilify .

The atypical antipsychotic, which is being co-marketed by Bristol and Otsuka, was approved Nov. 15 on the basis of short-term trials. FDA asked that the long-term study results be submitted as an efficacy supplement within 30 days of aripiprazole's approval.

FDA's approval letter notes the study "has already been completed and that the safety data were reported as part of the 120 Day Safety Update."

Labeling addresses the lack of long-term efficacy data in the indications section. "Efficacy of Abilify…was established in short-term (4- and 6-week) controlled trials of schizophrenic inpatients," it says, "long-term efficacy has not been established. The physician who elects to use Abilify for extended periods should periodically re-evaluate the long-term usefulness of the drug for the individual patient."

The major atypical antipsychotics on the market have long-term data in labeling: eight months for Lilly's Zyprexa (olanzapine), a year for Pfizer's Geodon (ziprasidone), and one to two years for Janssen's Risperdal (risperidone). Nevertheless, labeling for all three agents includes a similar statement to Abilify's about the need to periodically re-evaluate long-term usefulness in individual patients.

Bristol has been touting Abilify's favorable safety profile over other atypicals with respect to weight gain, serum prolactin levels and QTc interval (Pharmaceutical Approvals Monthly, July 2002, p. 8). Both Zyprexa and Geodon exhibit greater weight gain, and Geodon carries a bolded warning on QT prolongation. Zyprexa, Geodon and Risperdal are associated with hyperprolactinemia.

Bristol and Otsuka also are highlighting Abilify's relatively low level of somnolence: 11% vs. 8% for placebo. Zyprexa labeling reports somnolence as 20%-39% vs. 16% for placebo.

While Lilly has said Abilify's launch will likely have a short-term impact on Zyprexa sales, the firm says it expects Zyprexa to remain the leading antipsychotic and that it plans to conduct a head-to-head efficacy study against Abilify.

The Abilify NDA included a randomized, double-blind, 52-week study comparing aripiprazole to haloperidol, but without a placebo arm. FDA presumably was not satisfied with the study, as results are not included in labeling.

The requested long-term study may refer to a 26-week, placebo-controlled trial that Bristol had hoped to include in the NDA but was not completed in time.

The 26-week study was conducted in 310 patients with chronic schizophrenia evaluated as stable. Endpoints included time to relapse, Positive and Negative Syndrome Scale (PANSS) total score and Clinical Global Impression-Improvement score (CGI-I).

The study showed "significantly less relapses in the aripiprazole group than in the placebo group (34% vs. 57%, p<0.001)," the abstract states. The estimated Kaplan-Meier probability of not relapsing prior to week 26 was 39% for placebo and 63% for aripiprazole.

For PANSS total score, aripiprazole produced a "small but significantly greater improvement." Aripiprazole therapy resulted in an approximately 2 point decrease in PANSS compared to a nearly 5 point increase for placebo (p<0.05).

Safety results from the long-term study appear to support the claims Bristol has been making. No significant increase in QTc interval was observed in the aripiprazole or placebo group. Mean change in QTc from baseline to endpoint was -0.86 msec with placebo and -5.51 msec for aripiprazole. Labeling notes that in the pivotal trials, "the dose range of 10 to 30 mg/day…tended to slightly shorten the QTc interval."

Similarly, there were no clinically important or statistically significant differences in weight effects between placebo and aripiprazole. The mean change in weight in the placebo group was approximately 0.8 kg compared to 1.4 kg in the Abilify group.

"Although mean prolactin levels were elevated at baseline… prolactin levels were within normal limits by week 6 and throughout the remainder of the study," the abstract states. Mean change was approximately 1.2 mg/mL for placebo and nearly 2.0 mg/mL for Abilify (p<0.05).

BMS and Otsuka are also required to conduct Phase IV studies "to determine whether or not doses lower than 10 mg are effective."

The usual starting and target dose of Abilify is 10 mg or 15 mg once daily. However, labeling advises that dosing be halved for patients taking concomitant CYP3A4 (e.g., ketoconazole) or CYP2D6 (e.g., quinidine, fluoxetine, paroxetine) inhibitors. In patients taking CYP3A4 inducers, such as carbamazepine, the usual dose should be doubled.

FDA approved Abilify in 2 mg, 5 mg, 10 mg, 15 mg, 20 mg and 30 mg tablets. The firms do not plan to sell the 2 mg and 5 mg strengths until fulfilling the postmarketing study commitment.

The approval letter specifies the final study report is to be submitted as an NDA supplement by May 2006.

FDA also requested further studies in animal subjects, including a retinal degeneration study in rats. Otsuka is required to conduct "studies to further characterize (e.g., reversibility, functional correlates) and, if possible to determine the mechanism(s) underlying the retinal degeneration observed in the 26-week and 2-year carcinogenicity studies in Sprague-Dawley rat."

The retinal lesion observed in albino SD rats given high doses of aripiprazole "has morphologic features characteristic of light-induced retinopathy," FDA noted. Therefore, "it is critical that the potential for aripiprazole-related ocular changes be investigated in a pigmented rat strain that is less susceptible to light-induced retinal degeneration to rule out a direct effect of drug."

Other postmarketing commitments include a food effect study on the 30 mg dosage strength and submission of an abuse liability study in monkeys that is being conducted in Japan.

Labeling indicates the usual Abilify dose range can be administered "without regard to meals." Regarding abuse, labeling states that "aripiprazole has not been systematically studied in humans for its potential for abuse. In physical dependence studies in monkeys, withdrawal symptoms were observed upon abrupt cessation of dosing."