For the second time in as many months, shares of Corcept Therapeutics Inc. plummeted on news of a Phase III miss with its antipsychotic drug, Corlux.

The Menlo Park, Calif.-based firm's stock (NASDAQ:CORT) fell 29 percent, or 38 cents, Friday to close at 93 cents, after reporting that the second of three Phase III trials, Study 09, failed to show statistical significance of Corlux (mifepristone), a GR-II antagonist, over placebo in treating the psychotic features of psychotic major depression (PMD). As with the first Phase III trial results released last month, Corcept attributed the failure to an unexpectedly high placebo response rate.

"Again, the real surprise was the extraordinary response rate by the study-end of all participants, both the comparator group and the medicine group," Joseph Belanoff, Corcept CEO, said during a conference call.

Study 09, conducted in Eastern Europe, enrolled 247 patients to receive either 600 mg of Corlux or placebo once daily for seven days, along with antidepressant therapy from day one through day 56. The primary endpoint was defined as a 50 percent improvement in the Brief Psychiatric Rating Scale Positive Symptom Subscale (BPRS PSS) at day seven and day 28, with a secondary endpoint measured by improvement at day seven and day 56.

Results determined that there was no meaningful separation in response between the two study arms and, in fact, overall response was "exceptionally, exceptionally high at over 90 percent" in both the treatment and placebo arms, Belanoff said.

Those data mirrored the results from the first Phase III trial, reported late last month. That study, which involved 258 patients in the U.S., failed to show statistical significance over placebo on the BRPR PSS at day seven and day 56. Corcept's stock last more than half its value on that news, closing Aug. 25 at $1.54. (See BioWorld Today, Aug. 28, 2006.)

The company's next step is to try to figure out why the trials demonstrated higher than expected response rates.

"We're just beginning to sift through the data," Belanoff said. "We've produced two studies that have extra-high response rates" and "we're not really sure why that might be the case."

Researchers also will investigate a trend in Study 09 showing a statistically significant separation between the Corlux and placebo groups in efficacy measured by baseline to study-end. But, while that might sound promising, Belanoff said that finding "is separate from the primary efficacy endpoint," and might not prove clinically relevant enough to pursue a regulatory pathway.

In the meantime, Corcept is continuing with its third Phase III trial of Corlux in PMD. That trial, Study 06, is the largest of the three PMD studies with a planned enrollment of 440 patients, and is testing three doses of Corlux - 300 mg, 600 mg, or 1,200 mg - vs. placebo given once daily for seven days, with antidepressant therapy from day one to day 56. The primary endpoint again will be measured using the BPRS PSS, an 18-item rating instrument used to assess psychopathology plus a subset of four items designed to specifically measure psychosis.

Results from Study 06 are expected after the beginning of next year, Belanoff said, and depending on those data, Corcept will decide how to move forward.

If the trial yields positive results, the company might consider whether to conduct another pivotal trial, since two randomized, placebo-controlled studies would be needed to support a new drug application. And if results are negative, Corcept plans to analyze the data further before reaching any conclusions.

"There's still an important card to be turned over," Belanoff said. Once that data is in hand, "we'll have to make a decision as to how to proceed."

The results in PMD are not expected to affect Corcept's collaboration with Indianapolis-based Eli Lilly and Co., which aims to test Corlux as an adjunctive treatment to prevent weight gain with Lilly's antipsychotic drug Zyprexa (olanzapine). The companies agreed to conduct a proof-of-concept study, but have no plans to take Corlux forward in combination with Zyprexa.

If the study demonstrates proof of concept, Belanoff said, the firms might look at taking a second-generation GR-II antagonist forward in that indication.

Corcept had about $17.5 million in cash as of June 30, and the company anticipates those funds to sustain operations beyond the Phase III Corlux program.