Washington Editor

Acadia Pharmaceuticals Inc. saw its market cap double Monday following positive Phase IIb findings showing that its investigational compound ACP-103 improved the efficacy of a widely prescribed schizophrenia drug and also minimized its side effects.

The San Diego company's shares (NASDAQ:ACAD) rocketed $6.92 to $13.61, a 103.4 percent gain, on top-line results showing that 20 mg of ACP-103 in combination with 2 mg of Risperdal (risperidone, Johnson & Johnson) was statistically significantly better than 2 mg of the atypical antipsychotic plus placebo (p=0.01) and as effective as 6 mg. Specifically, there was a 23 point change in the Positive and Negative Syndrome Scale (PANSS) among patients treated with ACP-103 and 2 mg of Risperdal, a 16.6 point change in those on 2 mg of Risperdal and placebo and a 23.2 point change in those on 6 mg of Risperdal plus placebo. In addition, patients in the co-therapy arm also had 50 percent less weight gain than those receiving the higher dose.

Those data were better than expected by Eun Yang, an analyst with Jeffries & Co. in New York, a belief she said most of her colleagues also held, possibly explaining Acadia's stock surge. In addition, she said the run-up may have resulted from a short squeeze, which happens when short sellers rush to buy shares to cover their short position and cut their losses. The stock had been down about 25 percent year to date, Yang said.

Regardless of the reason, there's clearly excitement over the potential for ACP-103 to be used as co-therapy with atypical antipsychotics, which Yang said collectively generated $17 billion in worldwide sales last year. The half-dozen of them on the market account for the lion's share of schizophrenia drug sales these days.

"If one can reduce the doses of Risperdal, or the other atypical antipsychotic drugs, one will reduce many of [their] side-effect burdens," Acadia CEO Uli Hacksell explained during a conference call. He noted that dose-related issues such as weight gain, diabetes and cardiovascular problems impede schizophrenics from achieving the full benefit of those products because lower doses to minimize side effects also dampen their efficacy. In fact, previous studies have indicated that most patients stop taking atypical antipsychotics because of those side effects.

But as the Phase IIb study showed, ACP-103's ability to enhance the effectiveness of low doses would seem to obviate that problem because better side effect profiles "could improve compliance" and produce "better clinical outcomes," Yang told BioWorld Today. However, because ACP-103 has only been compared to Risperdal, which generated more than $4 billion in worldwide sales last year, she said it would be premature to infer its potential impact on other atypical antipsychotics.

Even more skeptical was Jonathan Aschoff, an analyst with Brean Murray, Carret & Co. LLC in New York, who questioned the strength of the study's findings.

"The hurdle was set too low," he told BioWorld Today, noting that the results aren't necessarily relevant to a Phase III study because of limitations in what can be concluded from Phase IIb. He said the next stage of testing would not include "such subtherapeutic doses of approved therapies" as were used in this study, but rather a higher dose. "That's a higher bar, because now the placebo group should do better."

Importantly, Aschoff said, the improvement in the results in the 2 mg Risperdal plus placebo arm raises questions in the context of pivotal trials done about 15 years ago that demonstrated 1.8 point and 11.1 point changes in PANSS from baseline with 2 mg of Risperdal. "Something about this patient population is not looking too representative of the treatment population as a whole," he said. "The comparison to baseline is certainly never going to be as believable in a patient population that's looking a little more treatable than the average patient."

In the trial's primary endpoint, 20 mg of ACP-103 in combination with 2 mg of Risperdal or the generic typical antipsychotic drug haloperidol produced a statistically significant antipsychotic effect as measured by the reduction in PANSS. Specifically, there was a 27.4 percent improvement in the ACP-103 plus Risperdal arm compared to baseline after six weeks of treatment, as well as a 25.6 percent improvement in those treated with ACP-103 and haloperidol in the same time frame (p<0.0001).

Aschoff questioned the importance of those data, saying that combination therapy "always reduces the relevance" of positive changes from baseline. "Phase III would never be that way," he said.

Other detailed findings from the randomized, double-blinded, five-arm study, which enrolled 423 patients in the U.S. and Brazil, showed that co-therapy with ACP-103 also led to a faster onset of antipsychotic action. After only two weeks of therapy, about 50 percent more patients in the ACP-103/risperidone arm responded to treatment compared to each of the low-dose risperidone (p<0.008) and high-dose risperidone (p<0.03) arms. That could reduce hospital stays for patients awaiting stabilization with their therapies, Hacksell said, potentially lowering health care system costs.

However, haloperidol plus placebo outperformed ACP-103 and haloperidol in improving patients' PANSS scores, producing a 29.2 percent improvement after six weeks, a more potent response than expected. But the ACP-103/haloperidol arm appeared to result in a faster onset of antipsychotic action after only two weeks of treatment compared to haloperidol plus placebo, and patients in the co-therapy arm had less weight gain compared to those in the haloperidol arm.

Each of the treatments was generally safe and well tolerated, adverse events were comparable among the five study arms and were generally mild to moderate in severity.

Going forward, Hacksell foresees "a range of exciting commercial opportunities" for co-therapy with ACP-103, an internally discovered selective 5-HT2A inverse agonist. More immediately, the company plans to further analyze this study and present full data at an undetermined venue later this year, explore potential partnerships and eventually begin Phase III testing. Yang expects a larger study to begin in the first half of next year, likely in tandem with a partner, though Aschoff remains unconvinced that the data will elicit a partnership.

If an agreement is reached, New Brunswick, N.J.-based J&J would seem to be the most likely partner.

On its own, Acadia also is developing ACP-103 for psychosis in Parkinson's disease, with a Phase III study scheduled to begin next quarter, and sleep maintenance insomnia. In addition, the company plans to begin a Phase IIb study next quarter of another drug for schizophrenia, ACP-104, with antipsychotic efficacy measured by PANSS as its primary endpoint.