Introduction The C9orf72 repeat expansion has been reported as a negative prognostic factor in amyotrophic lateral sclerosis (ALS). We have examined the prognostic impact of the C9orf72 repeat expansion in European subgroups based on gender and site of onset.
Methods C9orf72 status and demographic/clinical data from 4925 patients with ALS drawn from 3 prospective ALS registers (Ireland, Italy and the Netherlands), and clinical data sets in the UK and Belgium. Flexible parametric survival models were built including known prognostic factors (age, diagnostic delay and site of onset), gender and the presence of an expanded repeat in C9orf72. These were used to explore the effects of C9orf72 on survival by gender and site of onset. Individual patient data (IPD) meta-analysis was used to estimate HRs for results of particular importance.
Results 457 (8.95%) of 4925 ALS cases carried the C9orf72 repeat expansion. A meta-analysis of C9orf72 estimated a survival HR of 1.36 (1.18 to 1.57) for those carrying the expansion. Models evaluating interaction between gender and C9orf72 repeat expansions demonstrated that the reduced survival due to C9orf72 expansion was being driven by spinal onset males (HR 1.56 (95% CI 1.25 to 1.96).
Conclusions This study represents the largest combined analysis of the prognostic characteristics of the C9orf72 expansion. We have shown for the first time that the negative prognostic implication of this variant is driven by males with spinal onset disease, indicating a hitherto unrecognised gender-mediated effect of the variant that requires further exploration.
Statistics from Altmetric.com
Amyotrophic lateral sclerosis (ALS) is a debilitating disease with a poor prognosis. Progress towards developing new treatments has been limited both by disease heterogeneity, and by the likely interaction between genetic and environmental factors in disease pathogenesis.1 A pathological expansion of a hexanucleotide repeat in the C9orf72 gene2 ,3 accounts for up to 10% of those with ALS in populations of European extraction, and is associated with a distinctive clinical phenotype that includes frontotemporal dementia (FTD) in some instances.4–10 Although the C9orf72 repeat expansion has been shown to be an important negative prognostic factor in survival analyses,4–10 until now no study has been sufficiently large to permit robust analysis of interactions between the variant and demographic features including age, gender and site of onset. Here, we used our combined clinical data sets to determine whether the presence of the expanded variant differentially modulates survival based on gender and site of onset.
Data sources/case capture
Clinical data from ALS cases incident from January 2000 to April 2015 were collected from Belgium, Ireland, Italy, Netherlands and the UK. All patients fulfilled the diagnostic criteria for ALS, and core data elements as defined by the European Network for the Cure of ALS (ENCALS consortium) were harmonised across data sets for consistency based on existing consortia agreements.11 Owing to the lack of an agreed international definition of ‘familial ALS’, and given that a previous population-based familial aggregation analysis from Ireland demonstrated a much higher familial ALS occurrence (16%) than usually recognised,12 we did not exclude cases based on family history. The Belgian and UK cases were collected from clinical research centre cohorts, while the Dutch, Irish and Italian cases were sourced from the prospective population-based national registers.13–19
In accordance with existing Consortia agreements, data were collated using the following variables: age of onset, date of onset, date of diagnosis, date of death/last known follow-up date, site of onset, revised El-Escorial diagnostic category (except Belgium) and C9orf72 status (normal or expanded). For all study participants, C9orf72 status was determined by repeat primed PCR as described previously (with individual laboratory-based validation and quality control by Southern blot analyses).3
Survival analysis strategy
Initially exploratory models were constructed using Cox proportional hazards regression to explore the effect of different time of entry to the studies. Cox models were generated including known important survival covariates including age of onset, site of onset, diagnostic delay and C9orf72. Cox models were compared using a likelihood ratio test, and by testing the validity of the proportional hazards assumption of each covariate at each timescale.
A base model using Royston-Parmar flexible parametric regression20 was built on the preferred timescale, with a proportional hazards scale and a number of degrees of freedom selected by comparison of the Akaike information criterion and Bayesian information criterion from models with increasing degrees of freedom, and the variance–covariance matrix clustered by country. Survival follow-up was limited to 5 years from entry. Models were then built to explore the effect of C9orf72 status in sex and site of onset subgroups. The stpm221 and ipdmetan22 commands from Stata MP V.14.0 were used to perform the survival analysis and produce the meta-analysis graphs, while the ggplot223 package in R V.3.1.1 was used to generate selected final graphs.
Descriptive statistics and basic survival model
In total, 5106 ALS cases met the inclusion criteria, of which 457 (8.95%) carried the C9orf72 repeat expansion. Breakdown of the demographics of the overall cohort by country is shown in table 1. Missing values were minimal affecting 181 cases (3.5%). Online supplementary information 1 summarises the basic survival model. An individual patient data (IPD) meta-analysis of C9orf72 status in the base model estimated an HR of 1.36 (1.18 to 1.57) for those carrying the expansion versus those not.
C9orf72, gender and site of onset subgroup analysis
Survival curves were generated to evaluate the effect of C9orf72 status on gender and site of onset (figure 1) which suggested a three-way interaction. Therefore, gender, site of onset and C9orf72 status were categorised into one variable with eight levels as demonstrated in Online supplementary information 2 table 1. Through comparison of survival curves, redundant subgroups were combined (see online supplementary information 2), leaving three groups: spinal onset males with the C9orf72 expansion, other spinal onset patients and all bulbar onset patients. Survival curves for these groups showed that male spinal onset patients with the C9orf72 repeat expansion had a prognosis distinct from other spinal onset patients and similar to bulbar onset patients (figure 2). Meta-analysis calculated a survival HR of 1.56 (95% CI 1.25 to 1.96) for male spinal onset patients with the C9orf72 repeat expansion (figure 3). The finding was in the same direction in each country, although only the pooled estimate was statistically significant (figure 3).
The median ages and distribution of diagnostic delay across the final subgroups is shown in table 2. While age of onset was oldest in the bulbar onset group and youngest in the male spinal onset C9orf72 expanded group, the male spinal onset C9orf72 expanded group also had the highest proportion in the ‘short’ diagnostic delay category, consistent with the finding that the C9orf72 expansion differentially affects disease course in a gender-specific manner. Adjustment for the El-Escorial category (table 3) did not substantially alter the HR (1.57 CI 1.26 to 1.97).
Previously, studies have shown that people with ALS carrying a C9orf72 repeat expansion in blood present at a younger age and have reduced survival when compared to patients without the expanded variant (table 4). However, studies until now have not been sufficiently powered to determine whether the expanded variant differentially affects outcome in subgroups based on gender and site of onset. Our findings demonstrate an intriguing and previously unrecognised interaction between the expanded variant and male patients with spinal onset disease, which appears to drive the overall survival effect. Within this cohort, the median age of onset was 59.3 and the median survival was 2.29 years. This compared to a median age of onset of 62.3 and median survival of 2.77 years in all other spinal onset disease, and a median age of onset of 65 and median survival of 2.38 years in all bulbar onset disease. Moreover, and contrary to the usual pattern in young onset disease, male spinal onset C9orf72 expanded cases were also more likely to have experienced a shorter diagnostic delay, suggesting rapidly progressing disease.
Female gender has previously been reported as an independent predictor of faster functional decline in ALS;24 however, our observation of an interaction between site of onset, gender and C9orf72 has not been previously noted, possibly due to limitations in the power of previous studies due to lower numbers (table 4). Taken together, our findings and those of previous studies imply that distinct processes may operate in differing subgroups of ALS even when a known genetic factor is present as the underlying cause, and demonstrate that male gender is likely to be an important interacting factor in the biology of C9orf72-related disease.
A number of pathogenic mechanisms have been proposed to explain the role of the C9orf72 repeat expansion in ALS. These include haploinsufficiency, toxic RNA interfering with the function of RNA-binding proteins or other cellular factors, and the presence of toxic dipeptide repeat proteins through Repeat-associated non-ATG (RAN) translation.25 ,26 Recent work has also pointed towards C9orf72-induced pathology of nucleocytoplasmic transport processes.26–29 However, the pathobiology of the observed interaction between the C9orf72 variant and gender remains unclear, but it is congruent with observations in the SOD1 mouse model, in which transgenic mutant males have shorter survival compared to their transgenic female littermates with similar copy numbers.30 The mechanism for this gender effect in animals, although well recognised, has not been characterised, but can be attenuated when mice are bred on a different genetic background.30
A potential weakness of our study is that it did not include clinical scores for the presence of cognitive change, which is a known prognostic indicator in ALS. We and others have shown that those with C9orf72 repeat expansions are more likely to experience cognitive and behavioural change; however, to the best of our knowledge, until now no gender-mediated effect has been demonstrated in C9orf72-related cognitive profiling. Moreover, since C9orf72 is part of the causal pathway for some forms of FTD, inclusion of cognitive status as a variable would have introduced a selection bias based on ‘conditioning on a common effect’.31 A further limitation to this study is that our analysis does not include C9orf72 repeat expansion analysis by Southern blot, although individual definition of pathological expansion performed by each centre using repeat primed PCR was validated by Southern blot. While the length of expansion varies from tissue to tissue,32–34 diagnostic testing within a clinical setting uses blood samples from which all previous prognostic and clinical correlative studies have been performed. Finally, we did not account for riluzole use in the analysis, as this is routinely prescribed in all patients from the participating countries, and it was not possible to determine the level of compliance using available data.
In conclusion, we have performed an analysis of the effect of the C9orf72 expansion on survival in almost 5000 European patients with ALS. We have shown, for the first time, that C9orf72 repeat expansion is a significant negative prognostic indicator in males with spinal onset disease only. These findings suggest a hitherto unrecognised interaction between the C9orf repeat expansion, site of onset and gender. This has important implications in the understanding both the pathobiology of C9orf72-mediated disease, and in the development of future disease-related prognostic models.
The authors would like to thank people with MND and their families for their participation in this project. They also acknowledge sample management undertaken by Biobanking Solutions funded by the Medical Research Council at the Centre for Integrated Genomic Medical Research, University of Manchester.
Contributors JR had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. JR, IF and OH were involved in study concept and design. H-JW, AV, RML, MH, AJ, RvE, AC, LM, CS, KM, PJS, WR, PVD, AA-C; LvdB; AC; JV and OH were involved in acquisition of data. JR; H-JW and JV contributed to statistical analysis. JR and OH contributed to drafting of the manuscript. All authors were involved in critical revision of the manuscript for important intellectual content.
Funding This work was supported by the Interuniversity Attraction Poles (IUAP) programme P7/16 of the Belgian Federal Science Policy Office, by the FWO-Vlaanderen, and ZonMw under the frame of E-Rare-2, the ERA-Net for Research on Rare Diseases (PYRAMID). The research leading to these results has received funding from the Health Research Board Interdisciplinary Capacity Enhancement Programme, the European Community’s Seventh Framework Programme (FP7/2007-2013) under the Health Cooperation Programme and the project EUROMOTOR (number 259867). This work was funded in part by Fondazione Vialli e Mauro per la Sclerosi Laterale Amiotrofica onlus, the Italian Ministry of Health (Ricerca Sanitaria Finalizzata, 2010, grant RF-2010-2309849). This project was supported by the charity Research Motor Neurone, and the Irish Health Research Board, SOPHIA and STRENGTH which are European Joint Programme - Neurodegenerative Disease Research (JPND) projects. The projects are supported through the following funding organisations under the aegis of JPND-http://www. jpnd.eu (UK, Medical Research Council and Economic and Social Research Council; Netherlands, ZonMW; Italy, MUIR; Belgium FWO-Vlaanderen) and the Health Research Board of Ireland. JR was funded by the Health Research Programme Clinical Fellowship Programme (grant number HPF-2014-527). IF was supported by funds from Motor Neurone Disease Association of Great Britain and Northern Ireland (grant number 905-793, 6058). PVD holds a clinical investigatorship from FWO-Vlaanderen and is supported by the ALS liga Belgium. WR is supported by the European Community’s Health Seventh Framework Programme (FP7/2007-2013) under grant numbers 259867 & 340429, by funding from Opening the Future Fund, Hart voorALS fund and by the E von Behring Chair for Neuromuscular and NeurodegenerativeDisorders, all at the University of Leuven, and by funding from the ALS LeagueBelgium, the Thierry Latran Foundation and the Geneeskundige Stichting KoninginElisabeth (G.S.K.E.). PJS is supported as an NIHR Senior Investigator. CES and AAC receive salary support from the National Institute for Health Research (NIHR) Dementia Biomedical Research Unit at South London and Maudsley NHS Foundation Trust and King’s College London. PJS is supported as an NIHR Senior Investigator. LvB received a grant from the Netherlands Organization for Health Research and Development (Vici scheme).
Samples used in this research were in part obtained from the UK National DNA Bank for MND Research, funded by the MND Association and the Wellcome Trust.
Disclaimer The funders played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data and preparation, review or approval of the manuscript.
Competing interests AA-C is chief investigator of a clinical trial by OrionPharma, served on the scientific advisory board of the Motor Neurone Disease Association, is on the Editorial Board of the journals ‘F1000’ and ‘Amyotrophic Lateral Sclerosis and Frontotemporal Dementia’, has consulted for GSK, OrionPharma, Biogen Inc, Cytokinetics, and Treeway, and receives royalties from the books ‘The Brain’ (Oneworld Publications) and ‘The Genetics of Complex Human Diseases” (Cold Spring Harbor Laboratory Press). PJS is Principal Investigator of a clinical trial by OrionPharma, served on the scientific advisory board of the UK Medical Research Council, is on the Editorial Board of the ‘Amyotrophic Lateral Sclerosis and Frontotemporal Dementia Journal’, has consulted for, OrionPharma, Biogen, Cytokinetics, Treeway and Sanofi Aventis and receives royalties from the books ‘Oxford Textbook of Neurology. She has grants from the Motor Neurone Disease Association, National Institute for Health Research, European Commission, Medical Research Council and ALS Worldwide. AC serves on the editorial advisory board of the journal ‘Amyotrophic Lateral Sclerosis and Frontotemporal Dementia’ and has been a member of advisory panels Biogen Idec, Cytokinetics, Italfarmaco and Neuraltus. OH is funded by the Health Research Board Clinician Scientist Programme. Professor Hardiman has received speaking honoraria from Novarits, Biogen Idec, Sanofi Aventis and Merck-Serono. She has been a member of advisory panels for Biogen Idec, Allergen, Ono pharmaceuticals, Novartis, Cytokinetics and Sanofi Aventis. She serves as Editor-in-Chief of the journal ‘Amyotrophic Lateral Sclerosis and Frontotemporal Dementia’. LvB received travel grants and consultancy fees from Baxalta; serves on scientific advisory boards for Prinses Beatrix Spierfonds, Thierry Latran Foundation, Cytokinetics and Biogen. He serves on the editorial boards of ‘Amyotrophic Lateral Sclerosis and Frontotemporal Dementia’ and ‘Journal of Neurology, Neurosurgery and Psychiatry’.
Patient consent Obtained.
Ethics approval Ethical approval for this study and data sharing was obtained at each participating centre, and data shared through a legal agreement under the auspices of the EU JPND STRENGTH consortium, administered through King’s College London. For the purposes of this analysis, overarching approval was through the Irish Centre (Trinity College Dublin and Beaumont Hospital Research Ethics Committee (02/28; 05/49).
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.