The aim of this study was to determine the prognostic factors for local failure and cause specific survival in differentiated thyroid carcinoma and the impact of surgery, radioiodine, and external beam radiation therapy (RT) when used as part of the initial management. The authors reviewed the records of 382 patients with differentiated thyroid carcinoma (papillary, 262; follicular, 120) managed at the Princess Margaret Hospital (PMH) between 1958 and 1985. There were 220 patients with Stage I or II, 102 with Stage III, and 33 with Stage IV disease (according to 1987 staging criteria of the International Union Against Cancer). The median duration of follow-up was 10.8 years. Potentially important prognostic factors for cause specific survival (CSS) and local relapse free rate (LRFR) were tested by multivariate regression analysis with emphasis on treatment factors. For patients with papillary tumors, the 10-year CSS and overall survival were 93% and 85%, respectively; for those with follicular tumors, they were 69% and 56%. The LRFR for both histologic types were 86% at 10 and 15 years. Age >60 years, tumor size >4 cm, poor differentiation, postoperative presence of macroscopic residual disease, and presence of distant metastasis at presentation were identified in multivariate analysis as statistically significant factors for cause specific death. Age >60, tumor size >4 cm, multifocality, postoperative residuum, lymph node involvement, less extensive surgery (less than near-total thyroidectomy), and the lack of use of radioiodine were significant with regard to locoregional failure. The use of external RT was associated with more advanced local disease. There were no statistically significant differences in CSS or LRFR between patients who received RT and those who did not, even after adjustment for identified prognostic factors. In the subgroup of 155 patients with papillary histology and microscopic residuum, both 10-year CSS (100% vs. 95%, P = 0.038) and LRFR (93% vs. 78%, P = 0.01) were higher for patients given RT than for those not given RT. The 33 patients with macroscopic residual disease who received postoperative RT had a 5-year LRFR of 62% and CSS of 65%. The prognostic factors for differentiated thyroid carcinoma have been clearly identified. Initial total thyroid ablation with total thyroidectomy and radioiodine was associated with a lower rate of local relapse, although CSS was unaffected. Many patients at risk for local recurrence received postoperative RT, which did not significantly affect the risk of local relapse or CSS in the entire group. A beneficial effect of RT was demonstrated in the subgroup of patients with papillary tumors and microscopic residuum. Patients with postoperative macroscopic residual disease appeared to benefit from RT (with or without radioiodine). A Phase III trial involving patients at high risk for local recurrence is required to assess the potential benefit of RT.
