Propensity-based matching between esophagogastric cancer patients who had surgery and who declined surgery after preoperative chemoradiation

Background: Trimodality therapy (TMT; chemoradiation plus surgery) has level-1 evidence for survival advantage for TMT-eligible esophagogastric cancer patients. Some patients, however, decline surgery after preoperative chemoradiation. The question of which patient should have esophagectomy and which one should not is unlikely to be answered by a prospective comparison; therefore, we matched the clinical covariates of several patients who had surgery with those who declined surgery (DS). Methods: Between 2002 and 2011, we identified 623 patients in our databases. Of 623 patients, 244 patients had TMT and 61 TMT-eligible patients were in the DS group. Using the propensity-score method, we matched 16 covariates between 36 DS patients and 36 TMT patients. Results: Baseline characteristics between the two groups were balanced (p = NS). The median overall survival times were: 57.9 months (95% CI: 27.7 to not applicable, NA) for the DS group and 50.8 months (95% CI: 30.7 to NA) for the TMT group (p = 0.28). The median relapse-free survival times were: 18.5 (95% CI: 11.5-30.4) for the DS group and 26.5 months (95% CI: 15.5-NA) for the TMT group (p = 0.45). Eleven (31%) of 36 patients in the DS group had salvage surgery. Conclusions: Our results are intriguing but skewed by the patients who had salvage surgery in the DS group. Until highly reliable predictive models are developed for esophageal preservation, TMT must be encouraged for all TMT-eligible gastroesophageal cancer patients.