(Adds expert, analyst comments, details from presentation, share move)
By Julie Steenhuysen and Ankur Banerjee
Oct 25 (Reuters) - Eisai Co Ltd and partner Biogen Inc said on Thursday that the highest dose of their experimental Alzheimer's disease drug showed effectiveness in a new analysis, but investors remained skeptical about the reliability of the data and Biogen shares fell about 2 percent.
The companies have been trying to put to rest concerns about trial results of the drug, BAN2401, which in July showed that patients in early stages of Alzheimer's experienced 30 percent less cognitive decline after 18 months of treatment than those on a placebo.
Those results were muddied by concerns that the data was tainted by a decision from European regulators to remove patients with an Alzheimer's genetic mutation called APOE4 from the trial group that received the highest dose, potentially lowering the bar for success.
"I think that the data is promising and intriguing but continues to beg more questions that we don't have all the answers for," Jefferies analyst Michael Yee told Reuters, adding that the drug needs a much larger study to confirm any benefit.
There is a desperate need for an Alzheimer’s treatment that works after dozens of failures of experimental drugs. The most common form of dementia affects nearly 50 million people worldwide and is expected to rise to more than 131 million by 2050, according to Alzheimer’s Disease International.
Eisai scientists who presented the data at the Clinical Trials on Alzheimer’s Disease conference in Barcelona stressed that patients with the APOE4 mutation who completed 18 months of treatment at the highest dose experienced 63 percent less cognitive decline than the placebo group, while non-carriers did only 7 percent better than placebo.
Dr. Jeffrey Cummings of the Cleveland Clinic, a consultant to the companies and an Alzheimer's expert, said on Thursday the APOE4 genotype had "very little effect on the rate of decline," suggesting that it was not a factor in the results.
But other experts pointed out that because of those who were removed from the study, the data included only 10 patients with the APOE4 mutation who received the highest dose.
And when the highest and second-highest dose groups were combined, the overall 30 percent reduction in cognitive decline shrank to 21 percent, skewed by APOE4 carriers who had a 25 percent benefit versus only a 6 percent benefit in those who did not carry the mutation.
Leerink analyst Geoffrey Porges called the data unconvincing and confusing. He said the small number of patients remaining on the drug after 18 months and a lack of clear dose responses diminish the reliability of Thursday's data.
"You have to be careful with small numbers," Maria Carrillo, chief science officer for the Alzheimer's Association, said in a telephone interview.
She stressed that the findings come from a secondary analysis of 18-month data after the trial failed to meet its primary goal at 12 months using a more complicated method of statistical analysis. "We should interpret these with caution," she said.
Carrillo noted that the company's enthusiasm for the findings was "more positive than the reaction we had in the room about the study."
Laurie Ryan, chief of the Dementias of Aging Branch at the National Institute on Aging, said the presentation supported the initial argument of a positive signal and offered data suggesting the imbalance in patients with APOE4 may not have had a big impact.
"Clearly, that still needs to be tested," Ryan said. "You can't say this is ready to rock and roll at all."
Reporting by Manas Mishra and Ankur Banerjee in Bengaluru and Julie Steenhuysen in Chicago; Editing by Shounak Dasgupta and Bill Berkrot