Reclassifying American Joint Committee on Cancer/Union for International Cancer Control(AJCC/UICC) TNM stage for differentiated thyroid cancer

다국어 초록 (Multilingual Abstract)

Background
The predictability of current AJCC/UICC TNM stage for differentiated thyroid cancer is challenged. TNM staging of thyroid cancer needed to be amended to discriminate better and poorer outcome groups. We aimed to develop an alternative staging groups on patients with differentiated thyroid cancer who were treated with thyroid surgery.

Methods
A total of 6333 patients with differentiated thyroid cancer who underwent surgery at two tertiary referral centers from 1996 to 2005 were included in this study. Recursive partitioning analysis(RPA) was used to determine new stage grouping using our study cohort. The component used as predictive variables were tumor characteristics, nodal involvement, distant metastatic disease and age at diagnosis. Kaplan-Meier curves and log-rank tests were used to detect statistically significant differences in survival. Cox-proportional hazard model was used to compare the predictability of staging system.

Results
In current TNM stage, ten-year disease-specific survival (DSS) rates for stage I, II, III, and IV were 99.4%, 96.1%, 97.7%, and 85.9%, respectively. Reversal of outcome between stage II and III was uncovered. With new staging, 10-year DSS rates were changed to 99.7%, 98.4%, 93.4% and 57.5% for group 1, 2, 3 and 4 respectively. The values of PVE and Harrell’s c-index were 3.0, 0.693 for current TNM stage and 6.3, 0.869 for new staging.

Conclusion
This multicenter cohort study demonstrated that the current TNM staging system has insufficient discrimination ability to predict outcome of thyroid cancer. Our study suggested an alternative classification of thyroid cancer staging with RPA in patient with differentiated thyroid cancer treated with thyroid resection.

목차 (Table of Contents)

• ABSTRACT 1
• I. Introduction 2
• II. Materials and Methods 3
• 1. Study population
• 2. Statistical analysis
• III. Results 4
• 1. Baseline Characteristics
• 2. Changes in the distributions and DSS according to the TNM staging modified by different age cutoffs
• 3. Optimal age cutoff for predicting DSS
• 4. Development of new staging using RPA
• 5. Comparison of current TNM staging and new classification
• IV. Discussion 7
• References
• 논문요약