Recurrence of differentiated thyroid cancer in the elderly.

Objective: Data from the Surveillance Epidemiology and End Results Medicare-linked database were used to estimate the incidence of and risk factors associated with recurrent thyroid cancer, and to assess the impact of recurrence on mortality following diagnosis, controlling for mortality as a competing risk.

Design: We identified 2883 patients over 65 years of age diagnosed with a single, primary well-differentiated thyroid cancer between 1995 and 2007. A recurrence was considered if the patient had evidence of I-131 therapy, imaging for metastatic thyroid carcinoma, or complete thyroidectomy beyond 6 months of diagnosis. Competing risk regressions were performed using Cox proportional hazards models with 1- and 2-year landmarks.

Results: Recurrence was observed in 1117 (39%) of the 2883 patients in the cohort. Age, stage, and treatment status were significant risk factors for developing recurrent disease (P<0.0001). Patients with recurrent disease had a higher risk of all-cause mortality within 10 years of diagnosis than patients with no recurrence at 1- and 2-year landmarks. Patients with follicular histology and a recurrence were less likely to die from cancer (hazard ratio 0.54; P=0.03) than patients with no recurrence.

Conclusions: The rate of recurrence of well-differentiated thyroid carcinomas in this sample of elderly patients was 39%. Extent of disease and older age negatively impacted the risk of recurrence from differentiated thyroid cancer. In these data, patients with follicular histology and a recurrence were less likely to die, suggesting that mortality and recurrence are competing risks. These data should be taken into account with individualized treatment strategies for elderly patients with recurrent malignant thyroid disease.