Patterns of relapse following radiotherapy for differentiated thyroid cancer: implication for target volume delineation.

Introduction: Post-operative residual disease in differentiated thyroid cancer is an indication for external beam radiotherapy (EBRT) especially if there is poor radioiodine uptake by the residual disease. There are no standardized guidelines or consensus in target delineation for radiotherapy in thyroid cancer.

Aims: To determine the pattern of recurrence in patients with well differentiated thyroid cancer who received adjuvant or definitive radiotherapy as well as radioiodine ablation following surgery or biopsy with a view to better defining future target volume delineation for radiotherapy.

Materials And Methods: Forty-nine patients with differentiated thyroid cancer received radical external beam radiotherapy and radioiodine ablation (3.5GBq) following thyroidectomy or biopsy between 1990 and 2000. Nineteen patients had macroscopic residual (11) or inoperable disease (8), whilst 30 patients had clear (5) or microscopic positive resection margin (24), and 1 patient the resection margin status was unknown. All the patients were deemed high risk for local recurrence or progressive disease. The thyroid bed and regional nodes were irradiated using two radiotherapy techniques: (1) non co-planar lateral fields (NCLF) in coronal plane using 6MV photons to a dose of 45-50Gy in 16 fractions over 22 days and (2) anterior-posterior parallel pair of 6MV photons to a dose of 40-42.5Gy in 16 fractions over 22 days. There was no attempt to irradiate the lymph nodes in that part of the anterior and posterior mediastinum extending from the brachiocephalic veins to the carina.

Results: The median follow-up was 5.4 years (range 0.9-12.4 years). The actuarial 5-year cause-specific survival and local control for the whole group was 75.7% and 81.4%, respectively. Of the 4 patients with mediastinal recurrence, all had neck recurrences and two had distant metastases. All the medisastinal recurrences occurred in superior mediastinum (level VII) and all were treated with NCLF in coronal plane radiotherapy technique. Furthermore, mediastinal recurrences did not occur in isolation. The 5-years loco-regional control rate was 89.1% for those with clear or microscopic positive margins and 69.2% for those with macroscopic residual or inoperable disease. Five-year cause specific survival was 58.3% for patients with macroscopic residual or inoperable disease and 91.4% for those with clear or microscopic positive margins.

Conclusion: The status of postoperative margin relating to bulk of disease influences local control and cause specific survival. Surgical resection in locally advanced thyroid cancer should be performed by an experienced surgeon to achieve macroscopic clearance where possible. The majority of recurrences were loco-regional. The few superior mediastinal recurrences did not occur in isolation. All the mediastinal recurrences occurred in the superior mediastinum (level VII). We recommend the target volume should encompass the thyroid bed and regional neck nodes and the superior mediastinum level VII excluding the lymph nodes on both sides of the trachea within the anterior and posterior mediastinum extending from the brachiocephalic veins to the carina (compartment 4). Thus, this should facilitate dose escalation to improve loco-regional control and avoiding radiation induced mediastinal toxicity.