Immunosuppressive medication exposure and risk of invasive fungal and nontuberculous mycobacterial infections following TNF-α inhibitor initiation: a time-varying analysis

PURPOSE: Invasive fungal and nontuberculous mycobacterial (NTM) infections are important complications following TNF-α inhibitor initiation, but the contribution of additional immunosuppressive medication exposure remains incompletely defined. We evaluated their incidence and spectrum and associations with time-varying immunosuppressive medication exposure. METHODS: We conducted a retrospective cohort study using Merative MarketScan data (2018-2022) among adults initiating TNF-α inhibitors. Invasive fungal and NTM infections were identified using ICD-10-CM codes. Systemic glucocorticoid and non-glucocorticoid immunosuppressant exposures were modeled as time-varying variables. Cox proportional hazards models evaluated associations between immunosuppressive medication exposure and infection risk. RESULTS: Among 62,772 adults, 238 (0.4%) developed an invasive fungal or NTM infection after a median of 14.6 months, corresponding to 146 per 100,000 person-years. Histoplasmosis (26%), NTM infections (22%), and coccidioidomycosis (16%) were most common. In time-varying analyses, glucocorticoid exposure alone (aHR 1.77, 95% CI 1.24-2.52) and non-glucocorticoid immunosuppressant exposure alone (aHR 2.25, 95% CI 1.43-3.52) were associated with higher infection risk compared with periods without either exposure. Combined exposure was not associated with infection risk (aHR 1.31, 95% CI 0.92-1.85). CONCLUSIONS: Invasive fungal and NTM infections were uncommon following TNF-α inhibitor initiation. Time-varying glucocorticoid and non-glucocorticoid immunosuppressant exposures were associated with higher infection risk, illustrating the importance of dynamic immunosuppressive medication exposure when assessing infection risk during longitudinal care.