Genomic insights into stroke recovery: cross-phenotype associations

Abstract

Abstract Stroke is a major cause of long-term disability with variable recovery. While clinical factors such as initial severity play a role, genetic factors are increasingly recognized as important contributors to stroke recovery. Genotype studies are generally focused on a single post-stroke behavioural domain, but some genes might relate to broad mechanisms of plasticity. This study therefore aimed to identify cross-phenotypic genetic variants associated across two or more stroke recovery domains. DNA from Stroke, Stress, Rehabilitation, and Genetics study participants was genotyped, resulting in 9 814 610 variants. In order to examine cross-phenotypic results, we first conducted genome-wide association studies on the six recovery domains: motor (grip force), cognition (Telephone Montreal Cognitive Assessment), depression (Patient Health Questionnaire-8), stress (Primary Care Post-Traumatic Stress Disorder Screen), functional status (Stroke Impact Scale-Activities of Daily Living), and disability (modified Rankin Scale 0–2 versus 3–6), some of which were tested longitudinally, yielding nine phenotypes. Models were adjusted for age, sex, initial severity (NIH Stroke Scale score), and ancestry. Cross-phenotype associations were identified by evaluating single nucleotide polymorphisms (SNPs) associated (< 5e-5) with multiple phenotypes. To determine how these genetic variants may relate to biological mechanisms of recovery, we conducted gene enrichment analyses. Participants (= 565, 59% male) had mild-moderate initial stroke severity (median acute NIH Stroke Scale score = 4). After accounting for the correlation structure among the nine phenotypes, we observed 319 cross-phenotypic SNPs, 3.45 times the expected number. Five of the cross-phenotypic SNPs were linked to genes relevant to neural development, function and plasticity, e.g.(rs11778883-C),(rs55726768-G),(rs76401391-T),(rs113518460-C) and(rs147150392-C), as were enrichments in,,andcomplexes. Multiple gene enrichments were found, e.g. Stroke Impact Scale-Activities of Daily Living and Patient Health Questionnaire 8 at 3 months were enriched forphosphorylation, which is important for long-term potentiation. We identified cross-phenotypic SNPs associated with multiple behavioural domains of stroke recovery. Some of these genes encode, or regulate, druggable proteins. These genetic factors are not well captured by clinical or neuroimaging assessments and so provide a unique window into stroke recovery. These findings, if validated, suggest that some genes may be broadly important to stroke recovery. Aldridge. report that 319 genetic variants associated with multiple stroke recovery domains occur more than expected by chance. Genome-wide analyses across six recovery domains in 565 participants identify variants linked to neural plasticity genes, suggesting some genetic factors broadly influence stroke recovery and may encode druggable therapeutic targets. teaser Graphical Abstract Graphical Abstract Graphical abstract showing the three-step stroke genomics workflow. At left, blue panel depicts a genome-wide association study of DNA from 565 stroke survivors and 9.8 million variants. The blue arrow leads to a central panel listing the six recovery domains examined: motor, cognition, depression, stress, function, and disability. The orange second arrow leads to an orange panel, at right, labeled “Cross-Phenotype Discovery,” reporting 319 variants and key genes—ERICH1, FOX3, LIFR-AS1, RSS6KA2, and TUBGCP2—showing enrichment for neuronal plasticity. http://www.w3.org/1999/xlink float portrait fcag322_ga.webp anchor fcag322_ga portrait graphical