Abstract
The C9orf72 hexanucleotide repeat expansion (HRE) is the most common monogenetic cause of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). Neurodegenerative disease incidence in C9orf72 HRE carriers has been studied using cohorts from disease-affected families or by extrapolating from population disease incidence, potentially introducing bias. Age-specific cumulative incidence of ALS and dementia was estimated using Kaplan-Meier and competing risk models in C9orf72 HRE carriers compared to matched controls in UK Biobank. Risk modification by UNC13A genotype was examined. Of 490 331 individuals with valid genetic data, 701 had >100 repeats in C9orf72 [median age 55 (interquartile range 48-62), follow-up 13.4 years (12.3-14.1)]. The cumulative incidence of ALS or dementia was 66% (95% confidence interval 57%-73%) by age 80 in C9orf72 HRE carriers versus 5.8% (4.5%-7.0%) in controls, or 58% (50%-64%) versus 5.1% (4.1%-6.4%), accounting for the competing risk of other-cause mortality. Forty-one per cent of dementia incidence accrued between age 75-80. C-allele homozygosity at rs12608932 in UNC13A increased ALS or dementia risk in C9orf72 HRE carriers [hazard ratio 1.81 (1.18-2.78)]. C9orf72 HRE disease was incompletely penetrant in this population-based cohort, with risk modified by UNC13A genotype. This has implications for counselling at-risk individuals and modelling expected phenoconversion for prevention trials.