Abstract
BACKGROUND: Prostate and thyroid cancers represent two of the most overdiagnosed tumors in the US. Hypothesizing that patients diagnosed with one of these malignancies were more likely to be diagnosed with the other, we examined the coupling of diagnoses of prostate and thyroid cancer in a large US administrative dataset. METHODS: The surveillance, epidemiology, and end results (SEER) database was used to identify men diagnosed with clinically localized prostate cancer (CaP) or thyroid cancer between 1995 and 2010. SEER*stat software was used to estimate multivariable-adjusted standardized incidence ratios (SIRs) and investigate the rates of subsequent malignancy diagnosis. Additional non-urologic cancer sites were added as control groups. RESULTS: Patients with thyroid cancer were much more likely to be diagnosed with CaP than patients in the SEER control group (SIR 1.28 [95% CI 1.1-1.5]; p < 0.05). Similarly, the observed incidence of thyroid cancer was significantly higher in patients with CaP when compared with SEER controls (SIR 1.30 [95% CI 1.2-1.4]; p < 0.05). When stratified by follow-up interval, the observed thyroid cancer diagnosis rate among men with CaP was significantly higher than expected at 2-11 (SIR 1.83 [95% CI 1.4-2.4]), 12-59 (SIR 1.24 [95% CI 1.0-1.5]), and 60-119 (SIR 1.25 [95% CI 1.0-1.5]) months of follow-up. There was no increased risk of CaP or thyroid cancer diagnosis among patients with non-urologic malignancies. CONCLUSIONS: There is a significant association of diagnoses with prostate and thyroid cancer in the US. In the absence of a known biological link between these tumors, these data suggest that diagnosis patterns for prostate and thyroid malignancies are linked.