Recurrent differentiated thyroid carcinoma: biological implications of age, method of detection, and site and extent of recurrence.

Abstract

We identified factors predictive of outcome for recurrent differentiated thyroid carcinoma (DTC). Fifty-seven patients with local (LR), regional (RRec), and/or distant recurrence (DR) of 431 recurrent DTCs were studied. Disease-specific survival (DSS) rate was estimated with the Kaplan-Meier method. Univariate and multivariate comparisons were conducted by log-rank and Cox regression analysis. The median follow-up was 13 years. Distribution of the first relapse was LR only (35%), LR and RRec (23%), LR and DR (30%), and LR, RRec, and DR (12%). Factors predictive of resectability were a long (>or=5-year) disease-free interval (DFI) and subclinical and thyroid remnant recurrence. Only 26% of symptomatic and 45% of thyroid bed LR, and 43% with DFI <5 years, could be resected completely. No isolated thyroid remnant and 75% of thyroid bed LR resulted in tumor-related mortality. Age <45 years, subclinical recurrence, isolated LR, and the ability to render the patient disease free independently predicted DSS. Fifteen-year DSS for LR only; LR and RRec; LR and DR; and LR, RRec, and DR were 49%, 28%, 15%, and 0%, respectively. Isolated thyroid remnant recurrence defines a benign phenotype. Age, method of detection, site and extent of recurrence, and the ability to render the patient disease free predict outcome for recurrent DTC. Multimodality long-term follow-up is warranted to detect recurrence at a subclinical potentially curative stage.