Metabolic response during preoperative chemotherapy can predict prognosis in pancreatic cancer.

Abstract

Background: With the evolving treatment paradigms for pancreatic cancer, there is an increasing need for reliable markers to evaluate the effectiveness of preoperative chemotherapy. Due to the limitations of current indicators, this study aimed to evaluate the prognostic value of metabolic response based on the changes in the maximum standardized uptake value (SUV_max) on fluorine-18-fluorodeoxyglucose positron emission tomography.

Methods: This study included 210 patients with pancreatic cancer who underwent post-chemotherapy curative surgery between 2013 and 2022. Using maximally selected rank statistics for survival, the metabolic response was defined as follows: metabolic responder (%ΔSUV_max >75%), metabolic stable disease (15%-75%), and metabolic non-responder (≤15%).

Results: Among patients, 15.7%, 66.7%, and 11.6% were categorized into the metabolic responder, metabolic stable disease, and metabolic non-responder groups. The metabolic responder group (83.0 months) had longer median overall survival than the metabolic stable disease (51.0 months, p = .013) and metabolic non-responder (32.0 months, p = .002) groups. In addition, being metabolic responders (vs. non-responders) was an independent predictor of low recurrence rates (hazard ratio [95% confidence interval]: 0.46 [0.23-0.91]; p = .026) and achieving pathologic complete response (odds ratio [95% confidence interval]: 13.39 [1.61-300.77]; p = .035).

Conclusions: Metabolic response during preoperative chemotherapy has predictive power for post-resection prognosis and residual tumor status.