Nomograms for prognostication of outcome in patients with esophageal and gastroesophageal carcinoma undergoing definitive chemoradiotherapy

Introduction: Level-1 evidence for definitive chemoradiotherapy (bimodality therapy or BM therapy) has been established for patients with esophageal and gastroesophageal junction cancers (EGEJC) who otherwise do not qualify for surgery; however, tools to estimate individual patient prognosis are unavailable. We used a number of clinical pre- and post-treatment parameters to establish two nomograms: for overall survival (OS) and relapse-free survival (RFS). Methods: From 2002 through 2010, 257 consecutive patients with EGEJC who received BM therapy and had pre- and post-treatment positron emission tomography (PET) and post-treatment endoscopic biopsies among other assessments were analyzed from a prospectively maintained database. Standard statistical methods were used to generate the nomograms. Results: None of the 257 patients underwent surgery. Persistent or recurrent cancer was documented in 187 (72.8%) patients. The estimated median survival duration for all 257 patients was 21.1 months (95% CI, 18.9-27.1) and the median RFS duration was 11.6 months (95% CI, 9.43-15.0). After BM therapy, 155 (60.3%) patients achieved a clinical complete response (cCR). In multivariate analyses, maximum initial standardized uptake value and cCR were independent prognostic variables for OS (p = 0.038, p < 0.001). Nomogram concordance indices of 0.70 for OS and 0.77 for RFS were established by 200 cycles of bootstrap resampling for each of the two outcomes. Conclusion: Our data suggest that, in patients with EGEJC, pre- and post-treatment clinical parameters contribute to the establishment of prognostic nomograms of OS and RFS. Upon validation, these nomograms could prove useful in the clinic to individualize therapy.