Immune checkpoint inhibitor induced thyroid dysfunction is a frequent event post-treatment in NSCLC

Abbreviations: ICI; Immune checkpoint inhibitors. γClinically Acted Upon Thyroid Dysfunction is defined as thyroid diagnosis documentation and/or thyroid medication administration. *1 set of thyroid labs was defined as a TSH with reflex total triiodothyronine (T3), total thyroxine (T4), and/or free T4. ΔDenominator is different because of restricted analysis to the 53 patients that had at least one set of thyroid labs evaluated after ICI and no prior history of clinically acted upon dysfunction. [Display omitted] •Thyroid dysfunction after immune checkpoint inhibitors is a frequent event in NSCLC.•NSCLC patients are not adequately monitored for thyroid toxicity after ICI therapy.•Thyroid toxicity develops more than 6 months after immunotherapy discontinuation.•Close monitoring of thyroid function after ICI therapy is imperative in NSCLC. Thyroid dysfunction is the most frequent endocrine immune related adverse event (irAE) in non-small cell lung cancer (NSCLC), typically arising 3–6 months into immune checkpoint inhibitor (ICI) therapy, but arising after ICI cessation, in some cases. Due to limited post-treatment adverse event reporting requirements on ICI trials, the incidence of ICI-induced thyroid dysfunction arising after therapy is unclear. We investigated ICI-induced thyroid dysfunction in a cohort of 294 NSCLC patients, with a specific focus on the post-treatment setting. Retrospective analysis of ICI-induced thyroid dysfunction (clinically acted upon or laboratory only) was performed in 294 UCLA NSCLC patients treated 2012–2018. Clinically acted upon thyroid dysfunction was defined as thyroid diagnosis documentation and/or thyroid medication administration. Laboratory only dysfunction was defined as abnormal thyroid labs in the absence of clinical action. Timing of thyroid dysfunction relative to ICI treatment and thyroid monitoring patterns were also assessed. 82% (241/294) of ICI treated NSCLC patients had thyroid labs during treatment. Of these 241 patients, 13% (31/241) had clinically acted upon thyroid dysfunction prior to, 8% (18/241) during, and 4% (9/241) after ICI. Most patients, 66% (159/241), did not have thyroid labs after ICI, but in the 53 patients with labs and no prior clinical dysfunction, 17% (9/53) developed clinical dysfunction after ICI. In these 9 patients, median time from ICI initiation to dysfunction was 253 days. Two patients with post-treatment laboratory only dysfunction were observed. ICI-induced thyroid dysfunction ari