Association Between Graves' Disease and the Risk of Thyroid Carcinoma: The Role of Clinical, Metabolic, Hormonal, Immunological, and Ultrasound Biomarkers.

BACKGROUND: The relationship between Graves' disease (GD) and thyroid carcinoma (TC) remains controversial, particularly regarding the predictive value of metabolic, hormonal, immunological, and ultrasound biomarkers. We performed a systematic review and meta-analysis to evaluate the prevalence of TC in patients with GD and to identify biomarkers associated with an increased risk of malignancy. METHODS: A systematic literature search of PubMed/MEDLINE, Scopus, Web of Science, Embase, and the Cochrane Library was conducted in accordance with PRISMA 2020. Primary studies providing direct evidence in patients with Graves' disease or indirect evidence concerning thyroid carcinoma-associated biomarkers were considered. Owing to substantial differences in study populations, designs, exposures, and outcomes, quantitative pooling was restricted to sufficiently comparable estimates, while the remaining evidence was synthesized descriptively. RESULTS: Six primary studies were included. Two retrospective cohorts directly evaluated thyroid carcinoma in patients with Graves' disease and included 1051 patients, of whom 317 had thyroid carcinoma. The remaining studies provided indirect evidence from patients with hyperthyroidism, population-based thyroid cancer cohorts, or genetic datasets. Because of substantial clinical and methodological heterogeneity, a single pooled estimate was not considered appropriate. Thyroid nodules and larger nodule size were associated with malignancy in Graves' disease, while elevated triglycerides, reduced HDL cholesterol, and higher body mass index were associated with thyroid carcinoma in a surgically selected Graves' disease cohort. Evidence concerning hormonal, immunological, inflammatory, and genetically determined metabolic markers was heterogeneous and largely indirect. CONCLUSIONS: Thyroid carcinoma risk in Graves' disease is heterogeneous and appears to be concentrated in patients presenting with suspicious ultrasound findings and adverse metabolic profiles. Ultrasound biomarkers remain the most robust predictors of malignancy, while metabolic biomarkers represent promising complementary tools for individualized risk stratification. Current evidence does not support the routine use of hormonal or immunological biomarkers as independent predictors of thyroid carcinoma. Large prospective multicenter studies using standardized biomarker assessment are required to validate these findings and optimize personalized surveillance strategies.