LAUSR

OBJECTIVE Despite recent advances, it is not clear whether the various genes/genetic variants related to ALS interact in modifying patients phenotype. The aim of this study was to determine if the co-presence of genetic variants related to ALS has interactive effects on the course of the disease. METHODS The study population includes 1,245 ALS patients identified through the Piemonte Register for ALS between 2007 and 2016 and not carrying SOD1, TARDBP and FUS pathogenic variants. Controls were 766 Italian subjects age-, sex-, and geographically-matched to cases. We considered UNC13A (rs12608932), CAMTA1 (rs2412208), SLC11A2 (rs407135) and ZNF512B (rs2275294) variants, as well as ATXN2 polyQ intermediate repeats (≥31) and C9orf72 GGGGCC intronic expansions( ≥30). RESULTS The median survival time of the whole cohort was 2.67 years (IQR 1.67-5.25). In univariate analysis only C9orf72 (2.51 years, IQR 1.74-3.82; p=0.016), ATXN2 (1.82 years, IQR 1.08-2.33; p<0.001) and UNC13A C/C (2.3 years, IQR 1.3-3.9; p<0.001) significantly reduced survival. In Cox multivariable analysis, also CAMTA1 emerged to be independently related to survival (HR 1.13, 95% c.i. 1.001-1.30, p=0.048). The co-presence of two detrimental alleles/expansions was correlated with shorter survival. In particular, the median survival of patients with CAMTA1 G/G+G/T and UNC13A C/C alleles was 1.67 years (1.16-3.08) years compared to 2.75 years (1.67-5.26) of the patients not carrying these variants (p<0.001); the survival of patients with CAMTA1 G/G+G/T alleles and ATXN2 ≥31 intermediate polyQ repeats was 1.75 years (0.84-2.18) (p<0.001); the survival of patients with ATXN2 ≥31 polyQ repeats and UNC13A C/C allele was 1.33 years (0.84-1.75) (p<0.001); the survival of patients with C9ORF72 ≥30 and UNC13A C/C allele was 1.66 years (1.41-2.16). Each pair of detrimental alleles/expansions was associated to specific clinical phenotypes. CONCLUSIONS We showed that gene variants acting as modifiers of ALS survival or phenotype can act on their own or in unison. Overall, 54% of patients carried at least one detrimental common variant or repeat expansion, emphasizing the clinical impact of our findings. In addition, the identification of the interactive effects of modifier genes represents a crucial clue for explaining ALS clinical heterogeneity and should be considered when designing and interpreting clinical trials results.