Population-based genotype-phenotype correlation to stratify incident cases of motor neurone disease in Scotland from 2015-2017

Abstract

BACKGROUND: Motor neurone disease (MND) refers to a spectrum of rapidly progressive neurodegenerative diseases for which there remains no cure. A recognised and crucial barrier to more accurate diagnosis, prognosis and treatment relates to phenotypic heterogeneity. Recent discoveries in the genetic landscape of MND have resulted in an accelerated research investment exploring aetiology of disease and basis of phenotypic variation. Scotland benefits from a culture of longstanding MND data capture and an integrated healthcare system.

METHODS: I helped to develop Clinical Audit Research and Evaluation of MND (CARE-MND), an evolution of the established Scottish MND Register. CARE-MND is a national electronic platform for prospective, longitudinal monitoring of MND in Scotland. All people with MND (pwMND) diagnosed in Scotland in 2015-17 were included in an epidemiological study of incidence and prevalence of the disease. Patients who consented to sharing their medical records via the Scottish MND Register were included for phenotypic characterisation and prognostic modelling. Patients also donated DNA samples for genetic research to the Scottish Regenerative Neurology Tissue Bank. Two cohorts were genotyped: i) a pilot cohort of patients diagnosed 1989-2014 who were studied using a limited six-gene panel and ii) an incident cohort of patients diagnosed 2015-17 who were genotyped using an extended 49 gene panel. Genotype-phenotype correlations were explored and the impact of genetics included in prognostic models.

RESULTS: By the end of my study period, the CARE-MND electronic platform was fully integrated into routine clinical care across all 14 health boards in NHS Scotland.

Using capture-recapture statistics, coverage of the CARE-MND platform was 99% making it a reliable resource for further study. Direct age-standardised incidence in 2015 was 3.42/100,000 (95% CI 2.99–3.91); in 2016, it was 2.89/100,000 (95% CI 2.50–3.34). This represents a rise in incidence in Scotland by 36.0% over a 25-year period. The standardised incidence was also 66.9% higher than Northern European estimates. Of 619 pwMND diagnosed 2015-17, 437 (70.6%) consented to shared their phenotypic data. The following variables significantly predicted mortality: rapid decline in the ALS Functional Rating Scale Preslope, older age of onset, family history of MND and exposure to heavy metals/pesticides.

Atypical MND phenotypes (PLS, PBP and PMA), a long time to diagnosis and having ever smoked predicted survival. Genetic epidemiology of a historical cohort (diagnosed 1989 2014) using a 6-gene panel revealed pathogenic or loss-of-function variants in 17%. Using an extended panel, up to 22% had pathogenic variants or variants of uncertain significance with pathogenic potential (VUS-P). The cohort was enriched for the Scottish p.I114T SOD1 founder mutation. Gene carrier status was associated with a family history of MND and other neurological conditions (including Parkinson’s disease and multiple sclerosis).

Having a C9orf72 repeat expansion was associated with an increased risk of cognitive impairment.

Having a genetic mutation of any kind did not influence overall survival.

CONCLUSIONS: Through CARE-MND, stratification of the MND population has facilitated participation in observational studies and has established a platform for recruitment into drug trials. The epidemiological data show a changing landscape of MND in Scotland with a marked increase in incidence over 25 years. This is likely attributable to ascertainment in the context of improved neurological services in Scotland. Early disability, older age and a family history of MND are poor prognostic markers in the Scottish population, whereas a delay in time from onset to diagnosis, atypical subtypes of MND and a history of smoking are associated with longer survival. Clinical trial design in Scotland needs to reflect and control for these factors. Using an extended 49-gene panel, 22% of patients have a potentially pathogenic MND-associated variant. Diagnostic genotyping should be considered to inform patients’ prognosis and guide management.