Survival outcomes in oropharyngeal cancer: a decision tree analysis
Session type: Poster / e-Poster / Silent Theatre session
Our aim was to evaluate overall survival (OS) using decision tree algorithms in oropharygeal cancer patients.
In total 273 patients with newly diagnosed oropharyngeal cancer were identified from March 2010 to December 2016. All patients were treated with definitive intensity-modulated radiotherapy (IMRT). The open-source R software was used. OS was estimated by Kaplan-Meier method. Nine predictor variables, including gender, age, primary tumor site, alcohol, tobacco smoking, HPV status, clinical T classification, clinical N classification and early responders, were investigated. Important explanatory variables were selected using the random forest approach. A classification tree that optimally partitioned patients with different OS rates was then built.
The 5-year OS for the entire population was 78.1%. The main important variables were HPV status, N stage and early complete response to treatment. Patients were partitioned in five groups: i) patients with HPV-related oropharyngeal cancer (12% probability of death); ii) patients who had HPV-negative disease without nodal involvement at diagnosis (32% probability of death); iii) patients who had HPV-negative cancer, N stage < 2c and were early responders (36% death probability); iv) patients who had HPV-negative cancer, N stage ≥ 2c, and were early responders (83% death probability); v) patients who had HPV-negative cancer, with nodal involvement at diagnosis and were not early responders (71% death probability).
This classification tree could help to guide future research in oropharyngeal cancer field. Further analysis on a validation cohort is required to confirm our results.