(548) A locus near ICOS is a genetic modifier for risk of Mycobacterium avium complex airway infection in persons with cystic fibrosis.

Background: Nontuberculous mycobacteria (NTM) airway infection in persons with cystic fibrosis (PwCF) is an increasingly prevalent condition associated with substantial morbidity and mortality. Mycobacterium avium complex (MAC) and Mycobacterium abscessus complex (MABSC) are the most prevalent NTM species complexes in PwCF. Most incident NTM infections in PwCF do not progress to NTM lung disease and published microbiological criteria for NTM disease requires >2 NTM+ sputum cultures. Additional criteria to define NTM lung disease are not available from the Cystic Fibrosis Foundation Patient Registry (CFFPR), so we focus here on microbiological data. We hypothesize there are genetic modifiers of risk for MAC and MABSC infections in PwCF.

Methods: The Cystic Fibrosis Genome Project identified 3,321 PwCF with NTM sputum cultures in the U.S. CFFPR [1]. For homogeneity purposes, we extracted those with pancreatic insufficiency, no QC flags, self-reported non-Hispanic White identity, and a predicted probability for European ancestry >80%. We defined controls as PwCF with no NTM-positive sputum cultures and >2 years of NTM culture data. We defined 2 outcomes—>1 or >2 positive sputum cultures—for MAC and MABSC infections, producing a total of 4 outcomes.

We conducted a genome-wide association study on whole genome sequencing data mapped to GRCh38: cases from each of the 4 outcomes were compared with the shared control set. We used the GENESIS R package for analysis. Relatedness and principal components were estimated using PC-AiR and PC-Relate. We fit a generalized linear mixed model with a binomial family using the GMMAT approach, adjusting for sex, birth cohort (before, during, or after 1995-1999), principal components 1 and 2, and genetic relatedness as a random effect. We performed score tests for each variant with a minor allele count >10 and missingness < 2%, assuming an additive effect.

Results: The case counts for >1 and >2 MAC+ cultures were n = 353 and n = 149, while the case counts for >1 and >2 MABSC+ cultures were n = 257 and n = 120. We identified n = 1,830 controls, shared across all tests. Genomic inflation was absent, as observed in well-behaved QQ plots (1.005 ≤ 𝝀median ≤ 1.014). A signal near ICOS was observed in both the tests of >1 MAC+ cultures (chr2:204111791, p = 2.12×10-7, minor allele frequency (MAF) = 0.031) and >2 MAC+ cultures (chr2:204109873, p = 2.05×10-9, MAF = 0.025) (Figure 1). A modest signal was seen in this region for tests of >1 MABSC+ cultures (chr2:204076672, p = 1.13×10-5, MAF = 0.020) and >2 MABSC+ cultures (chr2:204111791, p = 1.12×10-3, MAF = 0.027).

Conclusions: We identified a genetic modifier of NTM airway infection risk in PwCF. The ICOS locus falls within a lincRNA (lnc-ICOS-1). Underexpression of ICOS has been associated with pulmonary NTM disease, and ICOS signaling is involved in T-cell responses to Mycobacterium tuberculosis infection. Determining risk factors for NTM infection, such as this genetic modifier, could guide targeted surveillance of NTM infection and new therapeutic strategies to minimize the morbidities of NTM lung disease.

Acknowledgements: This work was supported by CFF grants CUTTIN18XX1, BAMSHA18XX0, and KNOWLE18XX0 and is submitted on behalf of the CF Genome Project.

References: [1] Raraigh KS, Aksit MA, Hetrick K, Pace RG, Ling H, O'Neal W et al. Complete CFTR gene sequencing in 5,058 individuals with cystic fibrosis informs variant-specific treatment. Journal of Cystic Fibrosis. 2021. https://doi.org/10.1016/j.jcf.2021.10.011