Our purpose as to evaluate the impact of adjuvant radioactive iodine therapy (RAI) on prognosis, as assessed by progression-free survival (PFS) and overall survival (OS), in patients affected by differentiated thyroid carcinoma (DTC) lymph-node relapse and previously treated with lymphadenectomy.Methods
We retrospectively evaluated DTC patients treated with lymph-node dissection for disease relapse. All patients had previously undergone total thyroidectomy and radioiodine remnant ablation (RRA). We used clinical and histological data obtained during follow-up to assess response and outcome. By means of univariate and multivariate time-to-event analyses, we assessed the impact of RAI on outcome (PFS and OS) and the prognostic role of thyroglobulin (Tg) levels under suppression with levothyroxine (Tg-on) measured 1–3 months after lymphadenectomy and of other risk factors.Results
We evaluated 113 patients (age at diagnosis: median 41 years, interquartile range: 31–59), 64 of whom were treated with RAI. Over a median follow-up time of 5.7 years, 27 patients showed disease progression and 13 died. Kaplan-Meier PFS and OS curves showed that age on diagnosis, tumor histology, tumor size, DTC aggressive variant, and Tg-on were associated with prognosis. Patients with Tg-on ≥1 ng/ml treated with RAI showed a better PFS (Log-rank pp 0.001) and OS (p = 0.005) than untreated patients, while no effect of RAI was observed in patients with Tg-on <1 ng/ml. Multivariate models showed that age, Tg-on (≥1 vs. < 1 ng/ml, HR: 18.2, 95% CI: 5.09-64.8, p = 0.001) and RAI (Yes vs. No, HR: 0.36,95%CI: 0.15-0. 9, p = 0.02) remained the only independent factors associated with PFS, but only age and Tg-on remained significantly associated with OS (HR: 8.31, 95%CI:1.56-44.3, p = 0.01). Nonetheless, patients treated with RAI showed a lower risk of mortality (HR: 0.34, 95%CI: 0.1-1.15 p = 0.08) than untreated patients.Conclusions
RAI after lymphadenectomy for DTC relapse is significantly associated with better PFS only in patients with Tg-on ≥1 ng/ml.
To access the full text, please Sign in
If you have institutional access, please click here