The huge annual Alzheimer’s meeting starts tomorrow in Chicago, and it comes at an interesting time for the field. I mean “interesting” in, of course, the sense of the old curse, “may you live in interesting times.”

The last couple of weeks brought shocks from two different directions. One was the devastating failure of two drugs based on the leading theory of the disease: that Alzheimer’s is caused by an accumulation of sticky brain plaques made of the peptide (a part of a protein) called amyloid-beta, and that if you induce the body to make antibodies against this protein and/or administer a drug to dissolve the plaques, you will be well on your way to treating the disease. The other shock was the surprising success of an antihistamine drug (!) that came out of left field, by which I mean Russia.

Let’s start with the seemingly good news. A drug called dimebon, which years ago was sold in Russia as an antihistamine (manufacturers stopped selling it when a new generation of antihistamines became available) was given to 183 patients there with mild to moderate Alzheimer’s three times a day for six months, followed by another six months of getting the drug or a placebo. As scientists led by Rachelle Doody of Baylor College of Medicine reported in a special edition of The Lancet, the dimebon group showed significant improvements in ability to track dates, understand instructions, follow commands, memorize a list of words, and perform simple tasks such as copying drawings or addressing an envelope. The placebo group declined on these measures.

As the authors write, “patients given dimebon were significantly improved compared with baseline, and compared with those taking placebo.” Considering that there are currently “no approved therapies for mild-to-moderate Alzheimer’s disease [that] have shown increasing improvement over 12 months,” they write, that is no small feat.

In the same issue, scientists reported that although immunizing 80 Alzheimer’s patients against the amyloid-β peptide can clear amyloid plaques in their brains, it does not keep them from getting worse. Hopes that that approach would work were based on positive findings in mice (immunizing them with full-length amyloid-β both reduced plaques and improved brain function), but it doesn’t work in people, Clive Holmes and colleagues at Moorgreen Hospital, in Southampton, England, reported. “There is little evidence to suggest that there is any major effect on cognitive function,” they reported. “All but one of the individuals who died during the follow-up phase had clear end-stage dementia before death, including the two individuals with . . . almost complete elimination of plaques. These findings imply that progressive neurodegeneration can occur in Alzheimer’s disease despite removal of plaques.”

What’s going on? One scientist told me that dissolving plaques may be completely the wrong way to go: you melt these suckers and the gunk they’re made of is free to wash around the brain, somehow exerting mind-killing effects.

But the larger point is that we are more than 20 years into gene-based, rigorous basic research on this horrific disease. And pet hypotheses keep falling, while the totally unexpected keeps happening.

That's worth keeping in mind when the parade of announcements start coming out of the annual Alzheimer’s meeting this weekend and next week. You will be hearing lots of claims for successful therapies, some targeting amyloid-beta and others taking aim at tangles inside brain neurons made of a different protein, called tau. Just remember, we have been down this road before. Some questions to keep in mind:

• Nothing counts in science until it has been replicated.
• If a study says it cleared away plaques, did it also find cognitive improvement? If not, there is no cause for optimism. If so, how long did they follow the subjects, since although stabilizing grandma for a few weeks is not meaningless what really counts is turning back the disease for good.
• Likewise, if a study reports a therapy that destroys tau tangles, how much (if any) cognitive improvement occurred, and for how long?