Abstract ID: 26-515

Sociodemographic, Clinical, and Biomarker Predictors of Systemic Therapy in Thyroid Eye Disease: A Retrospective Cohort Study Using the All of Us Research Program

Author: Asma Alhazmi
Base Hospital / Institution: Stanford Univeristy

Presentation Type: Oral Presentation
Session: Orbit II
Date: 12th September
Time: 15.20PM

Purpose

We aimed to characterize the sociodemographic, clinical, and biomarker predictors of systemic therapy among patients with TED.


Methods

Retrospective cohort study using the All of Us Research Program electronic health record (EHR) dataset. Adults with incident TED were identified using a two-tier phenotype based on diagnosis codes and clinical features. Treatment-requiring TED was defined as receipt of systemic therapy within 24 months of diagnosis. Sociodemographic and clinical characteristics and laboratory biomarkers (TSH, FT4, NLR, PLR, MLR, MPV) were compared using standardized mean differences (SMDs); multivariable logistic regression evaluated predictors of systemic therapy.


Results

Among 388 adults with incident TED, 92 (23.7%) received systemic therapy within 24 months. Baseline characteristics were largely similar between groups, with moderate imbalances for index year (SMD = 0.59) and cardiovascular disease (SMD = 0.25). No clinical variable was significantly associated with treatment (C-statistic = 0.59). Among biomarkers, NLR (median 2.50 vs 2.17; SMD = 0.38) and FT4 (1.13 vs 1.17 ng/dL; SMD = 0.38) showed the largest group differences. Adding NLR to the clinical model improved discrimination by 0.03 AUC (n=99). Longitudinal NLR rose markedly after diagnosis in treated patients but remained stable in untreated patients.


Conclusion

Baseline clinical characteristics showed little ability to predict which patients would require systemic therapy. NLR and FT4, both routinely available at diagnosis, showed meaningful differences between treated and untreated patients and small improvements in predictive value. Longitudinal NLR elevation in systemically treated patients is consistent with a role for systemic inflammatory burden in TED severity. Prospective validation is needed, but these findings suggest routine laboratory markers warrant further evaluation as potential adjuncts to clinical risk stratification in TED.


Additional Authors

First name Last name Base Hospital / Institution
Rita Popat Stanford University

↑ Back to top