Four-Tier Pathologic Tumor Regression Grading System Predicts the Clinical Outcome in Patients Who Undergo Surgical Resection for Locally Advanced Pancreatic Cancer after Neoadjuvant Chemotherapy.

Background/aims Neoadjuvant chemotherapy is increasingly utilized in patients with borderline or locally advanced pancreatic cancer (LAPC). However, the pathologic evaluation of tumor regression is not routinely performed or well established. We aimed to evaluate the prognostic value of three tumor regression grading systems frequently used in LAPC and to determine the correlation between pathologic and clinical response. Methods We included a total of 38 patients with LAPC who were treated with neoadjuvant chemotherapy and subsequent resection. Pathologic tumor regression was graded based on the College of American Pathologists (CAP), Evans, and MD Anderson grading systems. Results One out of 38 patients (2.6%) achieved a pathologic complete response. Unlike other grading systems (Evans, p=0.063; MD Anderson, p=0.110), the CAP grading system was a significant prognostic factor for overall survival (p=0.043). Pathologic N stage (p=0.023), margin status (p=0.044), and radiologic response (p=0.016) correlated with overall survival. In the multivariate analysis, CAP 3 was an independent predictor of shorter overall survival (p=0.026). The CAP grading system correlated with the radiologic response (p=0.007) but not the carbohydrate antigen 19-9 level (p=0.333). Conclusions The four-tier CAP pathologic tumor regression grading system predicted the clinical outcome in LAPC patients who underwent resection after neoadjuvant chemotherapy. Therefore, a more comprehensive pathologic evaluation is warranted in these patients.