Ethnoracial differences in the prevalence of amyloid abnormality

Abstract

AbstractBackgroundPrevalence estimates of amyloid positivity in persons across the AD clinical spectrum are important to reduce screen failure rates and improve recruitment efficiency in anti‐amyloid trials. We have previously estimated the prevalence according to age, sex, education and APOE genotype. It is however unknown whether these prevalence estimates generalize to diverse ethnoracial groups. We here aim to estimate the prevalence of amyloid abnormality in Black, White and Asian persons without dementia and examine associations with age, sex, educational level and APOE‐e4 carrier status.MethodWe selected 7,419 participants (6,083 with normal cognition; 1,336 with MCI) from 9 Amyloid Biomarker Study cohorts, including 291 Black, 6,310 White and 818 Asian participants. Amyloid positivity was determined with a PET (n = 5,714) or CSF (n = 1,705) biomarker, using data‐driven or study‐specific cutoffs. Generalized estimating equations were used to estimate the age‐related prevalence by ethnoracial group and to evaluate moderation by APOE‐e4 carrier status while adjusting for sex and education.ResultThe mean age was 70.9+‐7.4 years, 3,848 (57%) were women, and 20% were APOE‐e4 carriers. In persons with normal cognition, the age‐related prevalence of amyloid positivity was higher for White (25%) than for Black (17%) and Asian (16%) participants (p = 0.016). Among APOE‐e4 carriers with normal cognition, the prevalence of amyloid abnormality was again higher for White (46%) than for Black (29%) and Asian (30%) participants (Asian vs Black p‐value = 0.046, Asian vs White p‐value<0.001, Black vs White p‐value<0.001). Among APOE‐e4 noncarriers, no differences among the ethnoracial groups were identified (p = 0.56). In persons with MCI, White (59%) and Asian (60%) participants had abnormal amyloid more often than Black participants (43%; p<0.001). Ethnoracial differences were similar for APOE‐e4 carriers and noncarriers with MCI.ConclusionIn persons with normal cognition, the prevalence of amyloid abnormality among APOE‐e4 carriers was ∼15% higher for White than for Black and Asian participants, while among APOE‐e4 noncarriers no ethnoracial differences were found. In persons with MCI, both White and Asian participants ∼15% more often had abnormal amyloid than Black participants, irrespective of APOE‐e4 carrier status. Ethnoracial status may be an important factor to consider in the design of AD clinical trials.