Imaging and fluid biomarkers predictive of progression from mild cognitive impairment to dementia

Background Dementia is a growing socioeconomic and medical problem that affects increasing portions of the population, with a prevalence that doubles every 5 years after the age of 65. Alzheimer’s disease (AD; 60–80%) and frontotemporal lobar degeneration (FTLD; 5–10%) constitute two of the most common causes of dementia. The first symptoms appear during an initial clinical phase referred to as mild cognitive impairment (MCI). Each year, 8–15% of patients with MCI develop clinical dementia. Identifying biomarkers predictive of progression from MCI to dementia will enable initiation of disease‐modifying treatments at a stage where clinical symptoms of dementia are still limited. We investigated the predictive utility of imaging and fluid biomarkers of AD and FTLD including ß‐amyloid plaques and neurofibrillary tangles composed of hyperphosphorylated tau in positron emission tomography (PET), cerebrospinal fluid (CSF), and MRI for predicting progression of MCI to dementia. Method Mixed‐effects logistic regressions were conducted in the National Alzheimer's Coordinating Center (NACC) dataset (n=222–2,462) to investigate the predictive utility of abnormally elevated amyloid or tau on PET (PET‐amyloid, PET‐tau), abnormally low amyloid in CSF (CSF‐amyloid), abnormally elevated CSF tau (CSF‐tau), AD or FTLD pattern in FDG‐PET (FDG‐AD, FDG‐FTLD), hippocampal atrophy in MRI, and MRI evidence for FTLD (MRI‐FTLD). Potential covariates including age at baseline, gender, education, apolipoprotein E genotype, and number of follow‐up visits were included in the regression models. Results PET‐amyloid (OR=5.80), MRI‐FTLD (OR=4.46), hippocampal atrophy (OR=4.22), CSF‐tau (OR=3.49), and FDG‐AD (OR=2.91) were the strongest predictors (p‐values<=0.001), followed by FDG‐FTLD (OR=2.71), and CSF‐amyloid (OR=2.21) (p‐values<0.027). Among patients with available imaging AD biomarkers (n=500), PET‐amyloid (OR = 3.31) and hippocampal atrophy (OR = 3.12) were independent and comparably strong predictors. Among patients with available fluid AD biomarkers (n=214), CSF‐tau was a significant predictor of MCI progression (OR=2.48, p=0.035), whereas CSF‐amyloid was not (OR=1.71, p=0.21). Conclusion PET‐amyloid, hippocampal atrophy, and CSF‐tau were the strongest predictors of progression of MCI to AD. MRI evidence was a stronger predictor than FDG‐PET for progression of MCI to FTLD. Imaging and fluid biomarkers provide complementary prognostic factors for forecasting the trajectory of clinical outcome among patients with MCI.