Use of Volumetric Analysis to Predict Pathologic Response in Esophageal Cancer: A Single Institution Cohort Study

Background: Complete pathologic response (pCR) after neoadjuvant chemoradiotherapy (CRT) is associated with improved survival in esophageal cancer. We used volumetric analysis of hypermetabolic esophageal tissue to assess if a decrease in volume following CRT was associated with pCR in esophageal cancer patients who underwent neoadjuvant therapy and esophagectomy. We aimed to determine if a decrease in the hypermetabolic esophageal volume could be a potential marker for better response to neoadjuvant CRT.

Methods: Patients who received CRT followed by esophagectomy between 2013 and 2023 at a single institution were retrospectively reviewed. Patients were grouped by pathological response grade. The hypermetabolic tumor volume, tumor size, and maximum standardized uptake value (mSUV) pre- and post-CRT were compared.

Results: Of 309 patients, 69 had positron emission tomography scans available. Most patients were white (88.4%), but black patients were more likely to have pCR on univariate analysis (50% vs. 24.6%, p=0.038). Similarly, most patients had adenocarcinoma (81.2%), but pCR was more likely in squamous cell carcinoma (72.7% vs. 17.9%, p=0.002). The pCR group had a significantly larger decrease in tumor size (5.6 vs. 3.2 vs. 2.9 vs. 2.8 cm, p=0.013) and esophageal volume (34.1 vs. 8.0 vs. 29.7 vs. 27.1 cm³, p=0.015) compared to the grade 1, grade 2, and grade 3 response groups, respectively. On multivariable logistic regression, change in tumor size [odds ratio 0.67, 95% confidence interval (0.52,0.88), p=0.003] was a significant predictor of better pathologic response with every 1 cm decrease in tumor size representing a 33% increase in the odds of pCR, while change in esophageal volume was not.

Conclusions: Patients with pCR had a significantly larger decrease in tumor size and esophageal volume compared to other grade responders on univariate analysis, but on multivariable analysis only change in tumor size remained as an independent predictor of pCR In this limited cohort, these findings encourage further investigation of volumetric analysis as a tool to predict pathologic response and indicate that it may potentially be a useful adjunct when counseling high-risk surgical patients deciding between resection and active surveillance.