Context: TSH is the major growth factor for thyrocytes and may have a causative role in thyroid cancer.
Objective: The objective of the study was to systematically assess the association between serum TSH and thyroid cancer.
Data sources: The MEDLINE and EMBASE databases were searched using synonyms for TSH and thyroid cancer, supplemented with reference list searches and author contact.
Study selection: Prospective cohort, case-control, and cross-sectional studies were identified with TSH the exposure and thyroid cancer the outcome.
Data extraction: Three reviewers independently extracted data. Studies reporting odds ratio (OR) for TSH levels and thyroid cancer were analyzed via meta-analysis and generalized least-squares trend estimation for dose-response relationships.
Data synthesis: Data extracted from 28 studies included a total of 42,032 subjects and 5,786 thyroid cancer cases. Dose-response spline analysis revealed a nonlinear relationship (P < 0.001). For TSH levels less than 1 mU/liter, the OR for thyroid cancer was 1.72 (1.42, 2.07) per milliunits per liter. However, the relationship changed for TSH levels 1 mU/liter and greater, with the OR thereafter being 1.16 (1.12, 1.21) per milliunits per liter. Studies controlling for autoimmunity reported the lowest OR [TSH below 2.5 mU/liter, OR 1.23 (1.02-1.47) per milliunits per liter; TSH 2.5 mU/liter or greater, OR 0.98 (0.89-1.09) per milliunits per liter]. Six groups assessed serum TSH in relation to markers of poor thyroid cancer prognosis, with three showing significant positive relationships.
Conclusions: Higher serum TSH concentration is associated with an increased risk of thyroid cancer. Thyroid autoimmunity may partially explain the association, but further epidemiological assessment is required. Future clinical research should investigate the validity of including serum TSH in diagnostic nomograms, its prognostic importance, and the potential for therapeutic TSH suppression in thyroid cancer prevention.