Did you know, for example, that the first clinical AD trial, by William Summers (Summers et al., 1986), was of such questionable quality that it embroiled the New England Journal of Medicine in controversy and led to a fraud investigation by the Food and Drug Administration and the respective academic medical center? Or that this initial trial was followed by two negative trials, at which point attempts at cholinergic treatment might have ended? For that matter, few of the earliest trials would pass methodological muster today. Despite those shortcomings, however, the early trials did focus attention on the goal of trying to develop drugs for what was considered an untreatable disease. They sparked larger, controlled, multicenter trials that led to the symptomatic drugs that are the standard of care today.

Beyond recalling historical vignettes, this lecture also offers a serious account of a current dilemma. Aisen discusses why the previously workable Phase 2 model for symptomatic drugs proved ill suited to give reliable efficacy signals for disease-modifying drugs that do not detectably improve symptoms in the short term. He reviews the failure of some of the large, long, expensive Phase 3 trials for nine different drugs in the past decade.

Clearly, better drugs are needed, but the failures have more than one reason. For their part, what can trialists do? Aisen says they can sharpen their tools to test drugs earlier, and in this way, get ready to act on building international consensus that AD starts some 15 years before symptoms. Previous trials in mild cognitive impairment (MCI) have failed, and Aisen dissects the reasons why that might be. He moves on to new trial designs based on a biomarker-supported definition of prodromal AD, which are garnering nods from the FDA and its European counterpart, the European Medicines Agency (EMA). Prodromal AD, unlike MCI, is considered a treatable disorder, Aisen says, and that opens up trial designs of greater statistical power. Several companies have begun using such designs.

But what if prodromal AD proves still to be too late? In the second part of the talk, Aisen speaks about laying the groundwork for secondary prevention trials. Those would be for people who are coming to clinics spurred not by memory problems, but for cognitively normal older people who have been studied and show the signature of pathological biomarkers that is thought to define preclinical AD. Regulators have been encouraging, saying, for example, that such trials could use a combination of a cognitive endpoint plus a panel of biomarkers as outcome measures, doing away with the requirement for a global clinical outcome.

Aisen wraps up with remarks on primary prevention, that is, trials in people in midlife who are normal not only cognitively but also in their biomarker profiles. Primary prevention aimed at staving off a pathological change of those early markers is out of reach at this point in time. Even so, Aisen calls on the field to think ahead and develop the biomarker and aging research necessary to make it feasible.

Play the Webinar. Questions? Comments? Add your perspective in the comment box below.—Gabrielle Strobel.