Genetic studies of speech intelligibility in the Million Veteran Program with emphasis on traumatic brain injury

About This Grant

Significance to VA: In 2023, the VA approved 4.4 million hearing-related disabilities, constituting 11.9% of all VA disabilities.1 Service in the Armed Forces universally involves exposure to noise,2 and noise-induced hearing loss accelerates age-related hearing loss, however, only one-third of those who are measured for hearing aids can use them, in large part because of speech understanding in ambient noise.3–5 Further, at least 12% of those with normal audiograms report difficulty in understanding, and this difficulty with speech intelligibility occurs in 87% of those who sustain a traumatic brain injury (TBI), a focus of VA research. Innovation and Impact: This will be the first genome-wide association study to use objective data for analysis of speech intelligibility. The Million Veteran Program (MVP) constitutes the largest collection of detailed audiogram data worldwide, including > 400,000 individuals with sequential data and word testing as of version 24_1. MVP is unique in its large collection of individuals of diverse ancestry for GWAS in non-European populations, representative of the VA populations.6 This unprecedented sample size is pivotal for making the next advances in auditory genomics. Specific Aims: 1. Define a quantitative phenotype using MVP data, based on an individual’s deviation from an age-adjusted norm for the speech intelligibility index (SII). 2. Conduct GWAS based on Veterans in MVP and UKB for comparison and meta-analysis with a measure of speech intelligibility, with and without TBI as a Gene x Environment variable. We will compare these results to our previous GWAS on hearing loss from ICD’s, individual audiogram frequencies, and principal components of audiograms to ascertain unique and shared loci, variants, genes, and biologic pathways relevant to speech intelligibility. 3. Evaluate a polygenic risk score (PRS) based on MVP as a discovery set to predict results from the UK Biobank and the Noise Outcomes in Servicemembers Epidemiology (NOISE) study. The UK Biobank ia a large civilian study with self-report of speech understanding difficulty and digits-in-noise results. NOISE is an ongoing VA longitudinal study with detailed audiologic data, including difficulties with speech intelligibility, tinnitus, etc. Exploratory aim: Study GWAS results for repurposed medication appropriate for noise-induced and age-related hearing loss. We will examine freely available programs such as the Pharos database, DrugBank, and others to identify potential protein targets related to significant GWAS results.7,8 Methodology: After defining an objective measure of speech intelligibility index (SII) using individual audiograms and word-recognition scores as a phenotype, we will perform genome-wide association studies on genotyped MVP participants. We will compare results for those with and without history of TBI. We will then perform functional analysis to identify significant loci, variants, genes, and pathways in both populations. Next, we will compare a polygenic risk score derived from MVP to participants in the United Kingdom Biobank and to Veterans in the NOISE study, to ascertain the best measure of predictability for those who will have difficulty with speech in noise, both with normal and elevated hearing thresholds. Path to Translation/Implementation: An individual PRS will aid in identification of those who would benefit from extended hearing rehabilitation and/or advanced hearing aids.5,9 Examination of the genomic architecture of speech intelligibility compared to our previous GWAS’ on individual audiogram thresholds and audiogram principal components will aid in identifying variants, genes, and pathways amenable to pharmaceutical intervention.