A FinnGen pilot clinical recall study for Alzheimer's disease

Abstract

Successful development of novel therapies requires that clinical trials are conducted in patient cohorts with the highest benefit-to-risk ratio. Population-based biobanks with comprehensive health and genetic data from large numbers of individuals hold promise to facilitate identification of trial participants, particularly when interventions need to start while symptoms are still mild, such as for Alzheimer's disease (AD). This study describes a process for clinical recall studies from FinnGen. We demonstrate the feasibility to systematically ascertain customized clinical data from FinnGen participants with ICD10 diagnosis of AD or mild cognitive disorder (MCD) in a single-center cross-sectional study testing blood-based biomarkers and cognitive functioning in-person, computer-based and remote. As a result, 19% (27/140) of a pre-specified FinnGen subcohort were successfully recalled and completed the study. Hospital records largely validated registry entries. For 8/12 MCD patients, other reasons than AD were identified as underlying diagnosis. Cognitive measures correlated across platforms, with highest consistencies for dementia screening (r = 0.818) and semantic fluency (r = 0.764), respectively, for in-person versus telephone-administered tests. Glial fibrillary acidic protein (GFAP) (p < 0.002) and phosphorylated-tau 181 (pTau-181) (p < 0.020) most reliably differentiated AD from MCD participants. We conclude that informative, customized clinical recall studies from FinnGen are feasible.

Figure 1

Operationalization of this clinical recall study from FinnGen. DPIA data protection impact assessment, BB biobank, MTA material transfer agreement, StC steering committee.

Figure 2

(a) Screening and enrollment process for study. AD Alzheimer’s disease, MCD mild-cognitive disorder, UEF University of Eastern Finland. (b) New data acquired during study. CERAD-NB Consortium to Establish a Registry for Alzheimer’s Disease Neuropsychological Battery, WMS Wechsler Logical Memory Scale, TMT-A/B trail making test parts A and B, CDR clinical dementia rating scale, CDR-SB CDR sum of boxes, ADCS-ADL Alzheimer’s Disease Cooperative Study-Activities of Daily Living scale, TELE telephone assessment for dementia, TICS-m modified telephone interview for cognitive status, Aβ-42/40 amyloid beta (Aβ)1–42/1–40, p-Tau181 phosphorylated-tau181, NfL neurofilament light chain, GFAP glial fibrillary acidic protein;

Figure 3

Comparison of cognitive testing results across the three platforms utilized in the study. Correlations of (a) cognitive screening, (b) semantic fluency, (c) immediate and delayed recall, and (d) executive function are presented between in-person, computerized, and telephone-based assessments. Scatterplots with 95% confidence intervals (shaded areas) display correlation coefficient r and p-value from Spearman’s rank partial correlation analyses. cCOG computerized cognitive testing, CERAD-NB Consortium to Establish a Registry for Alzheimer’s Disease Neuropsychological Battery, MMSE Mini Mental State Examination, TELE telephone assessment for dementia, TICS-m modified telephone interview for cognitive status with three learning trials of the 10-word list, TMT-A/B trail making test parts A and B.