Tree-Based Model for Thyroid Cancer Prognostication

Background:Deathisuncommoninthyroidcancerpatientsandthefactorsimportantinpredicting survival remain inadequately studied. The objective of this study was to assess prognostic effects of patient, tumor and treatment factors and to determine prognostic groups for thyroid cancer survival. Methods: Using data from the Surveillance, Epidemiology and End Results Program (SEER), we evaluatedoverallanddisease-specificsurvivalin43,392well-differentiatedthyroidcancerpatients diagnosed between 1998–2005. Multivariable analyses were performed using Cox proportional hazards regression, survival trees, and random survival forest. Similar analyses were performed using National Cancer Database (NCDB) data, with overall survival evaluated in 131,484 thyroid cancer patients diagnosed between 1998–2005. Relative importance of factors important to survival was assessed based on the random survival forest analyses. Results:Usingsurvivaltreeanalyses,weidentifiedfourdistinctprognosticgroupsbasedondiseasespecific survival (p0.0001). The 5-year disease-specific survival of these prognostic groups were 100%, 98%, 91%, 64%, while the 10-year survival were 100%, 96%, 85%, 50%. Based on random survivalforestanalyses,themostimportantfactorsfordisease-specificsurvivalwereSEERstageand age at diagnosis. For overall survival, important prognostic factors were similar, except age at diagnosis demonstrated marked importance relative to SEER stage. Similar results for overall survival were found using NCDB data. Conclusion: This study identifies distinct prognostic groups for thyroid cancer and illustrates the importance of patient age to both disease-specific and overall survival. These findings have implications for patient education and thyroid cancer treatment.