Endocrine immune-related adverse events by immune checkpoint inhibitors and potential predictive markers: a prospective study from a single tertiary center

Immune checkpoint inhibitors (ICIs) can trigger immune-related adverse events (irAEs). The appearance pattern of irAEs, who is prone to them, and their mechanisms are still uncertain. In this study, we aimed to monitor patients initiated on ICIs for endocrinological aspects and to investigate the potential predictive markers in the development of endocrine-irAEs. The study prospectively included 43 patients with metastatic disease scheduled for anti-PD-1/L1 therapy. Endocrinological follow-up was conducted at specified intervals as well as in response to any additional reported complaints. Serum concentrations of CXCL10, IL-1beta, and IL-17A were measured prior to ICI and during the endocrine-irAEs. A total of 39.5% of the patients experienced endocrine-irAEs, with a median onset time of 3 months. Among patients, 34.9% developed thyroid-related adverse events, and 4.6% experienced hypophysitis. Thyroid autoantibodies were associated with a higher incidence of thyroid-related irAEs ( P = 0.004). In the irAE group, median pre-ICI CXCL10 and baseline thyroid stimulating hormone (TSH) levels were significantly higher, baseline total testosterone level in men was lower than in the non-irAE group ( P < 0.05), whereas IL-1beta and IL-17A levels did not differ ( P > 0.05). Serum CXCL10, IL-1beta, and IL-17A concentrations did not differ significantly pre-ICI and during adverse events ( P > 0.05). Pre-ICI CXCL10 concentration was correlated positively with anti-TPO levels in patients with at least one thyroid autoantibody positivity ( r = 0.706, P = 0.01) and negatively with baseline total testosterone level of men ( r = 0.509, P = 0.002). Our results suggest that higher pre-ICI serum CXCL10 and TSH levels might have a predictive role in the development of endocrinopathies. Besides, baseline thyroid antibody measurements could be beneficial in predicting thyroid dysfunction.