Frisoni et al. make this interesting point: In HIV and cancer, success in fighting the disease was heralded by biomarkers — CD4 counts and viral load for HIV, and a wide variety of tumor markers in cancer — that allowed researchers to do much faster iterations in clinical trials. The problem is that regulators don’t have good reason to trust these unless a useful agent can be shown to affect that biomarker while also improving clinical symptoms in AD. Once this has been shown — and the effect on that biomarker is fairly specific and/or biologically meaningful — then beyond the obvious clinical use, it will speed up drug development for AD tremendously. This, to me, is a big part of why candidate biomarker research in AD is so important.