Thyroid hormones and epithelial ovarian cancer risk and survival: results from the European Prospective Investigation into Cancer and Nutrition study

Azam Majidi & Laure Dossus et al.

Abstract

Background

Thyroid-stimulating hormone (TSH) and thyroid hormones (free triiodothyronine [fT3] and free thyroxine [fT4]) may influence cancer outcomes, but evidence for ovarian cancer is limited.

Methods

We conducted a nested case–control study comparing 578 epithelial ovarian cancer (EOC) cases with matched controls within the European Prospective Investigation into Cancer and Nutrition (EPIC). To examine associations between circulating TSH, fT3, and fT4 levels and EOC risk, we estimated risk ratios (RRs) and 95% confidence intervals (CIs) per SD using conditional logistic regression. Among cases, we evaluated all-cause and EOC-specific survival by prediagnostic hormone levels. Hazard ratios (HRs) and 95% confidence intervals were calculated using multivariable Cox regression. We also estimated covariate-adjusted restricted mean survival time (RMST) and survival probabilities at 5 and 10 years.

Results

Thyroid hormones were not associated with EOC risk (RR [95% CI] per SD increase: TSH = 0.99 [0.87 to 1.12], fT3 = 1.12 [0.70 to 1.79], and fT4 = 1.08 [0.56 to 2.07]) levels. However, higher TSH levels were associated with better survival (HR [95% CI] per SD: all-cause death = 0.90 [0.82 to 0.99], EOC-specific = 0.88 [0.79 to 0.97]), whereas higher fT4 levels were associated with worse survival (all-cause = 1.10 [1.00 to 1.22], EOC-specific = 1.17 [1.05 to 1.30]), but no association for fT3. RMST and survival probabilities showed similar patterns: for TSH, 10-year RMST and survival increased from 5.3 years and 42.2% in Quartile 1 (Q1) to 6.4 years and 50.7% in Q4. Conversely, for fT4, 10-year RMST declined from 5.6 years (Q1) to 5.1 years Q4, and survival from 46.3% to 37.8%.

Conclusion

TSH and thyroid hormones might not affect ovarian cancer risk. However, high fT4 and low TSH concentrations may be associated with poorer survival. Further evaluation is suggested in other populations.

Funding
French National Cancer Institute Grant INCA_15874Organisation for Health Research and Development FundingItalian Ministry of Health FundingWorld Cancer Research Fund FundingMedical Research Council Grant MC_UU_00006/1Catalan Institute of Oncology-ICO FundingNational Institute for Health and Care Research FundingInternational Agency for Research on Cancer FundingMedical Research Council Grant MR/Y013662/1Netherlands Cancer Registry FundingCancer Research UK Grant C864/A14136German Cancer Aid, German Cancer Research Center FundingInstituto de Salud Carlos III FundingMedical Research Council Grant MR/N003284/1Regional Governments of Andalucía, Asturias, Basque Country, Murcia FundingCancer Research UK Grant C8221/A29017French Ministry for Higher Education FundingFederal Ministry of Education and Research FundingInstitut National de la Sant? et de la Recherche M?dicale FundingDutch Ministry of Public Health, Welfare and Sports FundingInstitut Gustave-Roussy FundingBiomedical Research Centre FundingMutuelle Générale de l'Education Nationale FundingDepartment of Epidemiology and Biostatistics, School of Public Health, Imperial College London FundingItalian Ministry of University and Research FundingMedical Research Council Grant MC-UU_12015/1French National Research Agency FundingWorld Health Organization Grant 001German Institute of Human Nutrition Potsdam-Rehbruecke FundingItaly: Associazione Italiana per la Ricerca sul Cancro-AIRC-Italy FundingMutuelle Générale de l’Education Nationale FundingCatalan Institute of Oncology—ICO Funding