Serum albumin is associated with immune-related hypothyroidism and treatment continuity in non-small cell lung cancer

Background Immune checkpoint inhibitors (ICIs) are a standard treatment for advanced non-small cell lung cancer (NSCLC), but clinically practical predictors of immune-related adverse events (irAEs) and treatment continuity remain limited. Methods We conducted a retrospective cohort study using two nationwide Japanese administrative claims databases: Medical Data Vision (MDV) and the Japan Medical Data Center (JMDC). Patients with advanced NSCLC treated with ICIs were identified. Those with available thyroid func…

By the Truvace deskOctober 5, 20261 min read1 sourcePeer-reviewedTRV-2026-1285· cite & verify

Serum albumin is associated with immune-related hypothyroidism and treatment continuity in non-small cell lung cancer — Small cell carcinoma - CT scan (6327961286) by Yale Rosen from USA. CC BY-SA 2.0 · https://creativecommons.org/licenses/by-sa/2.0

In brief

On 2026-10-03, a peer-reviewed study reported development and external validation of machine learning models, including Elastic Net, to predict immune checkpoint inhibitor-induced hypothyroidism in advanced non-small cell lung cancer using two nationwide Japanese claims databases. Baseline TSH was the strongest predictor, while serum albumin consistently emerged as an important predictor across models.

The finding matters because albumin is a routinely available lab that could help clinicians anticipate both an immune-related adverse event and longer treatment durability on first-line immunotherapy, potentially informing monitoring and counseling. Uncertainty remains about causality, generalizability outside Japan, and whether albumin reflects nutritional status, inflammation, or other confounders not captured in claims data.

Main points

Retrospective cohort used two Japanese claims databases MDV and JMDC with 1786 and 1083 patients for hypothyroidism analysis and 1088 and 1007 for treatment continuity.
Elastic Net model achieved AUC 0.72 training, 0.73 validation, 0.71 test and was externally validated across cohorts.
Baseline TSH was the strongest predictor of ICI-induced hypothyroidism, with serum albumin emerging as consistent predictor across all models.

The gain

In patients with advanced NSCLC receiving first-line immune checkpoint inhibitors, higher baseline serum albumin predicted by machine learning models was associated with significantly prolonged time to next treatment or death.

The rundown

Researchers identified advanced NSCLC patients treated with ICIs in MDV and JMDC databases and built Elastic Net and other machine learning models to predict hypothyroidism, reporting consistent AUCs around 0.71-0.73 across training, validation and test sets.

Analysis of first-line ICI recipients showed higher serum albumin linked to both higher incidence of hypothyroidism and longer TTNT-D with hazard ratios 0.75 in MDV and 0.79 in JMDC, supporting albumin as a practical biomarker for risk stratification.