ADNI has led to remarkable progress in understanding biomarkers in AD and MCI. The course of biomarker change over time is being mapped, the relationship among biomarkers is being defined, and the associations between biomarkers and clinical changes are being demonstrated. Biomarkers are positioned to play a larger role in drug development based on ADNI data. Phase II studies may be shortened, Phase III studies may include biomarkers as part of a disease-modifying NDA, and biomarkers may play key roles in primary prevention trials. Biomarkers will help de-risk Phase III decisions, reduce drug development times and costs, improve safety, and speed the development of urgently needed new treatments.
Review of the ADNI studies reveal several unmet needs in the realm of biomarker development. Most critical are more data on the link between clinical and biomarkers changes in response to treatment. Only a few studies including both outcomes have been reported (Fox, et al., 2005
; Gilman, et al., 2005
; Lannfelt, et al., 2008
; Salloway, et al., 2009
), and only the repeated use of biomarkers in studies of drugs affecting AD pathways will eventually inform the use of these measures as predictors of clinical benefit (Cummings, 2009a
). ADNI is a nonintervential natural history observational study and cannot contribute to this aspect of biomarker development.
Another unmet need in biomarker development pertains to measures of target engagement or drug activity. ADNI biomarkers characterize the natural history of AD. Drug development has been accelerated by combining a target engagement biomarker with natural history outcomes (Wagner, 2008
). For example, in the development of statins, cholesterol lowering can be measured directly as an immediate drug effect and linked to patient outcomes such as death, myocardial infarction, or stroke. Pharmaceutical development in AD would be facilitated by development of drug activity biomarkers that directly measure the effect of the candidate treatment on the target pathways (e.g, Aβ production, inflammation, oxidation) and including such measures together with natural history biomarkers and clinical outcomes in clinical trials. Biomarkers of the drug effect should be sought and characterized during the preclinical phase of drug development and extended into the clinical phases of the development program (Choi, et al., 2009
; Dubois, et al., 2010
; Higuchi, et al., 2010
The populations studied by ADNI anticipate the need to treat patients early in the course of AD and include patients with predementia syndromes and mild dementia. Most AD drugs are tested in patients with mild-to-moderate AD with MMSE scores in the 16–26 range. ADNI data apply only to the more mild end of this range of severity. Biomarker data are needed on patients with more severe disease to assist drug development in this broader AD population.
Clinical trials are increasingly global enterprises. While the US conducts more clinical trials than any other single country, collectively more trials are conducted outside the US than in the US (Glickman, et al., 2009
). Ex-US populations are often more poorly educated and less likely to be Caucasian than the ADNI cohort. The very high educational level of ADNI participants and the low rate of inclusion of non-White subjects limit the generalizability of the clinical and biomarker findings. The global biomarker interest inspired in part by ADNI will assist in characterizing persons with a broader range of educational levels and ethnic backgrounds. Among these world-wide studies are the European ADNI (Buerger, et al., 2009
; Frisoni, et al., 2008
); the AddNeuroMed study (Lovestone, et al., 2009
); the Australian Imaging, Biomarkers, and Lifestyle (AIBL) Study of Aging (Ellis, et al., 2009
); the Swedish Brain Power Initiative; and similar studies in Japan, Korea, and China (Miller, 2009
Together with data from international collaborators, ADNI biomarker data provide information that is increasingly critical to the successful development of new treatments for AD, new therapies to slow the progression of MCI to AD dementia, and agents to prevent cognitive decline in the elderly. Ultimately, ADNI and related biomarkers promise to reduce drug development times, increase success rates, reduce costs, de-risk trials using clinical outcomes, and hasten the development of new treatments for AD.