There has been turmoil in the world of Alzheimer's disease (AD) clinical drug trials over the past couple of years. Not too long ago we believed we had a good handle on the pathology that underlies brain cell death in this devastating disease – and better yet, drugs that could remove the pathological hallmarks. As well as compounds intended to modify the course of the disease, we also had a crop of drugs that might offer symptomatic relief. Unfortunately the promise of these new compounds has not been realised and we have instead witnessed a succession of failures, of which Neurochem's ‘Alzhemed', Lilly's ‘Semagacestat' and Pfizer's ‘Dimebon' have been just the most high profile examples. Alzheimer's disease remains a major area of unmet need, and there is still plenty of enthusiasm amongst drug developers to continue research in this indication. The failures of the past two years have focused attention on not just the disease mechanisms, but also the instruments used to measure drug effects, particularly with respect to the assessment of cognition, the hallmark deficit of AD.
For the past 20 years the Alzheimer's Disease Assessment Scale – cognitive subscale (ADAS-cog) has been a feature of almost all late-phase development trials, and often included in exploratory studies where the ambition has been to establish Proof of Concept (PoC). However, it has long been recognised that the ADAS-cog, like many of the tests borrowed from clinical psychology, is an imperfect means of measuring cognition in clinical drug trials. A key issue in this context is the instrument's tendency to confounding factors such as practice, familiarity and learning effects. Further, a lack of placebo decline, likely the result of practice, has long confounded our best intentions to test for beneficial drug effects. In spite of this, the drug development community has persevered with the ADAS-cog. In fairness the practice effect is not solely an issue for the ADAS-cog, but also for other measures used in AD trials, and more generally across central nervous system (CNS) drug research. Two reasons are usually proffered to account for continued use of the ADAS-cog. The first is a dogmatic belief that regulators require use of the ADAS-cog. However, recent statements from both the FDA and EMA confirm that it is not a requirement. Second, the ADAS-cog has a reputation of being a ‘gold standard', largely, it would seem, based on the fact that marketing approval for many of the so far registered AD drugs was gained on the basis of significant differences between treatment and placebo patients on this measure. However, many recent study results have undermined the utility of this measure, and in the following section we will discuss its inadequacy with particular relevance to issues of patient capacity and compliance.