Modeling the age-specific incidence of mild cognitive impairment incorporating the time-varying relationship of Alzheimer's disease biomarkers over 28 years

Abstract

Background

As life expectancy increases, the prevalence of mild cognitive impairment (MCI) due to Alzheimer's disease is expected to rise. Biomarkers from cerebrospinal fluid (CSF) and magnetic resonance imaging (MRI), combined with age-specific modeling, may enhance understanding of the timing and risk factors of MCI onset.

Objective

To estimate age-specific incidence rates of MCI onset and examine the impact of biomarkers and demographic factors on risk.

Methods

We analyzed data from 307 cognitively unimpaired BIOCARD participants with one or more CSF and MRI measurements. To account for competing risk of death, we applied age-specific cumulative incidence functions and cause-specific Cox proportional hazard models. Time-varying biomarker values were incorporated to assess their influence on MCI risk.

Results

The cumulative incidence rate of MCI onset estimated in age brackets from 50 to 80 was 2.33% at age 55, 5.03% at age 60, 12.38% at age 65, 17.91% at age 70, 28.84% at age 75, and 46.91% at age 80. APOE4 carriers had a significantly increased risk before age 70 (HR = 9.55, p = 0.005). Female APOE4 carriers showed a non-significant trend toward reduced risk compared to males (HR = 0.35, p = 0.127). Faster decline in entorhinal cortex volume (HR = 1.24, p = 0.028) and elevated p-tau₁₈₁ levels (HR = 1.51, p = 0.012) were both associated with increased MCI risk.

Conclusions

This study presents age-specific incidence rates of MCI and highlights the age-dependent influence of genetic and biomarker risk factors. Incorporating time-varying biomarkers provides valuable insights into dynamic risk prediction for MCI onset.