National Editor

Cortex Pharmaceuticals Inc.'s CX516 for mild cognitive impairment failed to meet its primary endpoint in a Phase IIb trial, although a subset of patients with the worst memory trouble showed improvement over placebo.

That wasn't enough for Wall Street, which gave Cortex a day investors might rather forget. The company's shares (AMEX:COR) closed Tuesday at $3.15, down 65 cents or 17.1 percent, after falling as low as $2.76 during the day.

Irvine, Calif.-based Cortex's president, chairman and CEO, Roger Stoll, who joined the company a year and a half ago, was not surprised by the trial results.

"If I would have been here, I would never have taken CX516 into the clinic at all," he told BioWorld Today. "You just had to wait for the inevitable to happen."

Stoll called the market's reaction "somewhat muted. We were down 18 or 20 percent, but that's not bad for a small company that once touted this as the be-all and end-all of Ampakine treatment," he said.

Neither Cortex nor Paris-based collaborator Les Laboratoires Servier is giving up on Ampakine technology, which has yielded other drugs with longer half-lives and greater potency - although Servier "put $6 million into [the trials with CX516], so they're probably even less happy than we are."

Ampakine drugs act on AMPA (amino-3-hydroxy-5-methyl-4-isoxazole propionic acid) receptors on neurons to increase the strength of signals at connections between the brain cells. The cross-national study with a fixed dose of CX516 vs. placebo included 175 patients at nine sites in the U.S. and 22 sites across France, Belgium, the UK, Sweden and the Netherlands. Eighty-four patients were given CX516 and the rest placebo.

Data did not show statistically significant improvement in delayed recall of a 15-item word list, but the subset with the most serious baseline memory impairment - that is, the worst 25 percent of the patient population - performed substantially better on the test when given CX516, as compared to placebo.

"If we'd have gone into a patient population like Alzheimer's, we'd have had a much more improved chance of showing something," Stoll said. "This is not a well-studied population, and we were going in with an extremely weak horse."

More patients on CX516 than placebo withdrew from the trial, mainly due to gastrointestinal side effects.

"It's a ridiculous amount of drug you have to give," Stoll said. "With the amount of chemical we were dumping into them, the chances of getting GI irritation were high."

Dosage of CX516 was 12 mg/kg. In a recently completed primate study at Wake Forest School of Medicine in Winston-Salem, N.C., CX516 showed no significant activity until the dose exceeded 20 mg/kg. A clinical study with the drug in healthy adult male volunteers to reverse sleep-deprivation effects indicated no effect at doses below 30 mg/kg.

Though he was open to positive data, Stoll said he would not have advanced CX516 even if the results had been good. "I've said before that I was going to hitch our wagon to another star," he said.

That star may be Cortex's CX717, a much more potent Ampakine compound. In another soon-to-be published preclinical study, also at Wake Forest, a single dose of 0.5 mg/kg to 1.5 mg/kg seemed sufficient for optimal response on short-term memory. That would amount to about 75 mg to 100 mg per day in an elderly patient, most likely as a single dose.

Cortex expects to begin a Phase I trial with CX717 for Alzheimer's disease and mild cognitive impairment (MCI) in the second quarter of this year.

"I'm going to bring a couple of other compounds along that I'm going to kick up to [toxicity studies]," Stoll said. "I'm not going to rely on just one. There's no reason to believe CX717 won't be successful, but I want to have some insurance policies."

Servier has its Ampakine compound, S-18986, aimed for Phase II trials, also for Alzheimer's disease and MCI, by the end of this year.

In January, Cortex got a $2 million milestone payment from another licensee - NV Organon, of Oss, the Netherlands, which has Ampakine Org 24448 in Phase II trials for schizophrenia. Ahead of the bad news about CX516, Cortex last month was able to pull down $19 million in a private placement, its largest financing ever. (See BioWorld Today, Jan. 12, 2004.)

Indianapolis-based Eli Lilly and Co. has another Ampakine drug, LY451395, in Phase II trials for Alzheimer's, Stoll noted.

"The faith that, if you get the right chemical, it's going to work, is pretty high," he said.