Original Article
A combination of platelet-to-lymphocyte ratio and carbohydrate antigen 19-9 predict early recurrence after resection of pancreatic ductal adenocarcinoma
Abstract
Background: Early recurrence (ER) after surgical resection is an important factor that impacts the survival of patients with pancreatic ductal adenocarcinoma (PDA). We examined risk factors for ER after PDA resection.
Methods: One hundred and thirteen PDA patients who underwent R0 or R1 resection were retrospectively analyzed. Thirty-four patients (30.1%) received neoadjuvant chemotherapy (NAC) for borderline resectable (BR) (n=13) or initially unresectable (n=21) disease. ER was defined as that diagnosed within 6 months after surgery. Receiver operating characteristic analysis was performed for each variable to determine the optimal cutoff value.
Results: ER occurred in 21 patients (18.6%). In univariate analysis, preoperative platelet-to-lymphocyte ratio (PLR) ≥144, carbohydrate antigen (CA) 19-9 ≥162 U/mL, and pathological tumor size ≥3 cm were significantly associated with ER. High PLR and CA19-9 were independent risk factors for ER by multivariate analysis. Area under the curve (AUC) for predicting ER from a combination of PLR and CA19-9 was 0.765 (95% confidence interval: 0.664–0.866), which increased the AUC compared to that for each risk factor alone. Patients with both risk factors had a significantly worse overall survival than those with one or no risk factors. When 24 patients with BR-PDA were considered, NAC was associated with a reduced likelihood of having risk factors and with a low ER rate.
Conclusions: A combination of PLR and CA19-9 is a useful predictor of ER after macroscopic curative resection for PDA. NAC may reduce the risk of ER in selected patients.
Methods: One hundred and thirteen PDA patients who underwent R0 or R1 resection were retrospectively analyzed. Thirty-four patients (30.1%) received neoadjuvant chemotherapy (NAC) for borderline resectable (BR) (n=13) or initially unresectable (n=21) disease. ER was defined as that diagnosed within 6 months after surgery. Receiver operating characteristic analysis was performed for each variable to determine the optimal cutoff value.
Results: ER occurred in 21 patients (18.6%). In univariate analysis, preoperative platelet-to-lymphocyte ratio (PLR) ≥144, carbohydrate antigen (CA) 19-9 ≥162 U/mL, and pathological tumor size ≥3 cm were significantly associated with ER. High PLR and CA19-9 were independent risk factors for ER by multivariate analysis. Area under the curve (AUC) for predicting ER from a combination of PLR and CA19-9 was 0.765 (95% confidence interval: 0.664–0.866), which increased the AUC compared to that for each risk factor alone. Patients with both risk factors had a significantly worse overall survival than those with one or no risk factors. When 24 patients with BR-PDA were considered, NAC was associated with a reduced likelihood of having risk factors and with a low ER rate.
Conclusions: A combination of PLR and CA19-9 is a useful predictor of ER after macroscopic curative resection for PDA. NAC may reduce the risk of ER in selected patients.
