An investigational drug has shown promise in improving memory and cognition in patients with mild to moderate Alzheimer’s disease in a 6-month, randomized, placebo-controlled trial that was presented at the Alzheimer’s Association International Conference 2012.

The fact that the patients were still continuing to improve on statistically significant measurements of those aspects of the disease provided a hint, at least, that the drug EVP-6124, an alpha-7 nicotinic acetylcholine receptor partial agonist, could change the course of Alzheimer’s.

But it’s too soon to place any long-term bets, Dr. Dana Hilt said in an interview.

“This is being investigated as a procognitive and symptomatic drug,” said Dr. Hilt, senior vice president of clinical development and chief medical officer of EnVivo Pharmaceuticals, which funded the trial. “It certainly appears to improve cognition in a clinically beneficial way at 6 months. But if the curves continue to diverge, is it possible that something more fundamental could be going on? Could this be maintained at 12 months? That’s what we need to know.”

Dr. Paul S. Aisen, director of the Alzheimer’s Disease Cooperative Study and a professor of neurosciences at the University of California, San Diego, said that EVP-6124’s phase II data looked very good.

“Reviewing the results, I would consider this to be a solidly positive trial,” he said in an interview. “While we cannot rely on a relatively small phase II trial to accurately indicate the size of a treatment effect, the consistency of the findings here are encouraging. If efficacy is confirmed in additional trials, this may prove to be an important therapeutic option.”

The study group comprised 409 patients who were randomized to one of three drug doses (0.3 mg, 1 mg, or 2 mg daily), or to placebo.

Half of the group were on stable doses of donepezil (Aricept) or rivastigmine (Exelon) and half were not taking any acetylcholinesterase inhibitor (AChEI), at least at the time of the trial, Dr. Hilt said. “Some of them had never taken an AChEI, and some had taken them in the past but were now off, for a number of reasons,” including insurance coverage and lack of efficacy.

The subjects’ mean MMSE (Mini-Mental State Exam) at baseline score was 20, but the range was wide, from 12 to 29; the mean score on the CDR (Clinical Dementia Rating) scale was 6.

The primary end points were changes in the ADAS Cog-13 (Alzheimer’s Disease Assessment Scale Cognitive-13) and the CDR sum of boxes scores. Additional prespecified end points were the ADAS Cog-11 and composite measures of cognition and memory.

In each finding, Dr. Hilt noted effect size as well as statistical significance. Effect size is important in clinical trials because it suggests clinical efficacy better than statistical findings, he said in an interview.

“The effect size of approved drugs is somewhere around 0.15-0.28. In terms of clinical change, an effect size of 0.2 is something a trained clinician could detect. And effect size of 0.4 is something the next-door neighbor could detect.”

During the first month of the study, the lower-dose groups and the placebo groups all improved their ADAS Cog-13 scores, a finding consistent with a placebo effect often seen in such studies. Interestingly, the 2-mg group remained at baseline for the first 4 months of the trial; after 4 months, cognition scores increased rapidly and outstripped all of the other groups. By the end of the study, the 2-mg group had a mean improvement of 1.5 points over baseline, whereas the placebo group had dropped below baseline, giving a total group separation of 2.2 points (P = .0189; effect size 0.39). This effect size is higher than the 0.28 estimated for high doses of donepezil, rivastigmine, or memantine (Namenda), Dr. Hilt said.

At the end of the treatment period, the curves were still diverging, suggesting that longer treatment might confer more benefit.

On the CDR sum of boxes, the 2-mg dose also performed significantly better than placebo. Again, there was the early upward trend for all groups, including placebo. But after 4 months, the placebo group and the active groups of 0.3 mg and 1 mg began losing those gains. The 2-mg group maintained its gain (P = .0253; effect size 0.31).

Findings on the secondary end points were similarly encouraging, although all were not statistically significant, Dr. Hilt said.

On the ADAS Cog-11, those taking the 2-mg dose faired significantly better than did those in any of the other groups. By the end of the study period, the lower-dose groups and placebo group had declined below baseline to the same degree. The 2-mg group had improved significantly and was still on an upward trend (P = .0151; effect size 0.34).

The ADCS-ADL (Alzheimer’s Disease Cooperative Study Activities of Daily Living) scale showed the pattern of early improvement observed with other scales. The 2-mg group maintained its gains, but the between-group difference was not significant at the end of the study (P = .0925; effect size 0.20). Nor was there a significant 24-month difference in the MMSE (P = 0.0955; effect size 0.21).

A composite measure of cognition that included word recall on the ADAS-Cog, word recognition, word association, and orientation significantly favored the 2-mg group (P = .0037; effect size 0.42), as did a composite measure of memory (P = 0.0088; effect size 0.37).

Besides being richly distributed in the hippocampus and frontal cortex, alpha-7 nicotinic acetylcholine receptors are found in many other tissues, including the gut and skeletal muscles. A compound that attaches nonselectively could be loaded with adverse effects, including cardiovascular changes and unpleasant GI reactions. These adverse effects have proved difficult to overcome, and in fact are one reason that a number of nicotinic acetylcholine receptor studies have been discontinued.

“A number of companies tried to develop these, but at high dose levels that had indiscriminate stimulation of the receptors,” Dr. Hilt said. “When you give a neurotransmitter to affect receptors in the brain, you have to live with the consequences of stimulating that receptor everywhere. The way we have dealt with this is to use a drug with a very long half-life, in very low concentrations.”

The long half-life allows the drug to build up slowly, and keeps levels at a steady concentration without extreme peaks or nadirs. “It turns out this receptor is very finicky. Too high a level overstimulates it and too low a level doesn’t stimulate it at all. I’m convinced this is the major trouble with some of these other drugs.”

Adverse events in the trial were mild gastrointestinal issues, including nausea and constipation. No patients of the active-drug study groups withdrew because of an adverse effect. The rate of adverse events in the 2-mg group also varied by baseline medication status, with more among those already taking an AChEI at baseline than among those not on the drugs (64% vs. 42%). Similar differences in adverse events rates occurred in the other treatment groups, although the rates were lower than in the 2-mg group. These differences reflect the expected problems with AChEI drugs, with the additional increase due to EVP-6124, Dr. Hilt said.

All told, there were 23 serious adverse events in the study: 4 in the placebo group, and 19 in the active groups. The investigators deemed that only two were treatment related; two additional events were considered possibly related to treatment.

Dr. Hilt said that EnVivo plans to go forward with a phase III trial early next year.

Dr. Aisen serves as a consultant to Eisai, Pfizer, Novartis, and other companies. He also receives research support from Pfizer and others.