Sold under the brand Namenda by Forest Laboratories Inc and Germany's Merz Pharma, the drug had US sales of about $1.2 billion (USD) last year.

For the study, researchers culled through published studies, presentations at medical meetings and other sources to find studies that looked at the effectiveness of the drug memantine, Namenda's generic form, in patients with mild Alzheimer's disease.

They found that memantine worked no better than a dummy pill at treating the disease.

Memantine is one of the few approved drugs to treat Alzheimer's disease, which affects 26 million people globally. Alzheimer's can stretch on for years, slowly robbing patients of their mind and memories. And there are currently no drugs that can keep the disease from progressing.

Memantine is intended to treat moderate to severe Alzheimer's, but it is frequently used in mild Alzheimer's patients "off-label" -- for uses other than those approved by the FDA -- either alone or in combination with a drug known as a cholinesterase inhibitor.

Cholinesterase inhibitors increase levels of the brain chemical acetylcholine, which may slow mental decline in Alzheimer's patients.

The team identified three studies that included 431 patients with mild Alzheimer's and 697 patients with moderate levels of the disease.

"There were no significant differences between memantine and placebo on any outcome for patients with mild AD, either within any trial or when data were combined," Dr. Lon Schneider of the University of Southern California and colleagues wrote in the Archives of Neurology.

Among patients with moderate Alzheimer's, there was no significant difference between memantine and the placebo in any individual trial, but there was a significant effect when the three trials were combined.

"Despite its frequent off-label use, evidence is lacking for a benefit of memantine in mild Alzheimer's disease, and there is meager evidence for its efficacy in moderate Alzheimer's disease," the authors wrote.

They said more scientifically rigorous studies are needed to show the drug works.

Bill Thies, chief medical and scientific officer of the Alzheimer's Association, noted that the study was a "meta-analysis" -- a study that combines already published data.

Given that, he said it is not clear how much the analysis will change the way doctors practice medicine.

"All the drugs we have for Alzheimer's disease have modest effects and they are transient because the underlying disease is continuing to get worse while these drugs are being given. Eventually, that overcomes any benefit you get from the drug," Thies, who was not involved with the study, said in a telephone interview.

Thies also said that there is a big range in how patients respond to the drug, with some patients getting quite a good benefit, and others getting none.

"Unfortunately, we don't know how to pick those people out ahead of time," Thies said.

- Reuters