Identifying a combination of biomarkers to predict treatment response to nabilone for agitation in Alzheimer's disease - an exploratory post hoc analysis.

BACKGROUND

To identify if a combination of blood-based biomarkers related to inflammation and oxidative stress predict treatment response to nabilone for Alzheimer's disease (AD)-associated agitation.

RESEARCH DESIGN AND

METHODS

Agitation was assessed using the Cohen-Mansfield Agitation Inventory (CMAI). Serum concentrations of 13 markers were quantified. Univariable and multivariable regression were used to determine differences in CMAI change given nabilone and placebo. A model combining biomarkers with clinical predictors was also evaluated.

RESULTS

Overall, 38 participants enrolled in the original trial (76% male, mean ± SD age 87 ± 10). Nabilone was more efficacious in participants with higher IL-6, higher 8-ISO, higher 24S-OHC, and lower clusterin. Participants in the first tertile (T1) of index scores demonstrated better response to nabilone compared to placebo with a mean difference in CMAI change of -20.6 (95%CI: -30.3, -10.4). During the nabilone phase, 83% of participants in T1 were responders versus 38% in T2 + 3 (Fisher's p = .01). In the combined model, T1 showed better response to nabilone with a mean difference in CMAI change of -26.4 (95%CI: -34.0, -19.6). The proportion of responders was significantly higher in T1 (91%, n = 11) compared to T2 + 3 (32%, n = 19) (Fisher's p = .002).

CONCLUSION

A combination of biomarkers could help characterize responders and non-responders to nabilone.