Nomogram and competing risk model to predict recurrence after curative surgical resection of PDAC.

ABSTRACT

BACKGROUND: Surgical resection remains the only potentially curative treatment for pancreatic ductal adenocarcinoma (PDAC). However, a number of patients get disease recurred in a short time post-operation. Few studies have focused on the predictors of different recurrence patterns of PDAC. OBJECTIVE: To try to establish and verify a nomogram to predict recurrence free survival (RFS) in PDAC patients, and to distinguish the risk factors of local recurrence first and distant metastasis first via competing risk model. METHODS: Patients who underwent radical pancreatectomy for PDAC in our center from 2010 to 2018 were reviewed retrospectively. Kaplan-Meier methods and multivariate Cox regression analyses were used to identify the clinicopathological predictors of recurrence post-operation. And then, a nomogram was constructed and validated. Competing risk regression model was used to compare the predictors between local recurrence group and distant metastasis group. RESULTS: A total of 200 patients were included into the final analysis, and 153 patients got disease relapsed post-operation. CA19-9 level, vascular resection, tumor differentiation, lymph node ratio (LNR) and adjuvant chemotherapy were identified as independent risk factors for recurrence free survival (RFS) and incorporated into the nomogram. The C-index of the nomogram was 0.650. Competing risk model indicated that the status of lymph-node metastasis was significantly associated the patterns of first relapse. CONCLUSIONS: Nomogram and competing risk model were constructed to quantify the risk of recurrence following surgery for PDAC. Our findings may be useful for predicting RFS and recurrence pattern in clinical work.