Reduction Rate Based on Accurate Area of Tumor as a Predictor of Survival in Esophageal Cancer Patients After Neoadjuvant Chemotherapy

Background In RECIST, esophageal cancer is still classified as a non-measurable lesion, In Japan, the efficacy of neoadjuvant chemotherapy (NAC) for esophageal cancer is evaluated based on the reduction rate of the primary tumor’s short diameter on CT. However, this method does not fully consider the accurate tumor area. Recent advances in software of CT analysis technology have enabled easier measurement of cross-sectional area, enabling us to utilize this method to propose a more appropriate cutoff value. To determine the optimal measurement reflecting survival, we compared primary tumor reduction after NAC in EC patients who underwent NAC followed by surgery. Methods We included 199 consecutive EC patients treated with NAC (triplet/doublet) and surgery (April 2004–March 2020). The percentage decrease in the cross-sectional tumor area, excluding the luminal area, measured by an axial CT slice of the largest part, was termed the accurate area (AA). ROC curves determined optimal cut-off values for AA reduction, shorter diameter (SD), and the product of shorter and longer diameter (SLD) to predict survival. Responders were defined as cases exceeding each cut-off value.Hazard ratios (HRs) for recurrence-free survival (RFS) were estimated, and survival curves were compared between responders and non-responders using the log-rank test. Results The median reduction rates of AA, SD, and SLD were 56.04% (−41.66–91.58%), 34.74% (−24.44–77.47%), and 53.95% (−30.04–87.38%), respectively. The AUC for AA, SD, and SLD were 0.715, 0.663, and 0.695. Optimal cut-offs for predicting 3-year RFS were 30%, 20%, and 30%, with responder proportions of 78.9%, 71.9%, and 79.9%.HRs were 3.47 (P < 0.001) in AA, 2.56 (P < 0.001) in SD, and 3.10 (P < 0.001) in SLD. Multivariate analysis identified age, ypN, clinical response, and AA non-response as independent prognostic factors (HR: 1.99, 95% CI: 1.13–3.49; HR: 3.72, 95% CI: 2.22–6.25; HR: 3.24, 95% CI: 1.83–5.75). Conclusion The reduction rate of primary tumor based on the accurate area of the tumor after NAC was seemed to be one of the optimal measurement reflecting survival in patients with EC who underwent NAC followed by surgery.