[Prognosis and risk factors of different recurrence and metastasis patterns following pancreatectomy]

Objective: To elucidate the prognostic differences and risk factors associated with different patterns of recurrence and metastasis following pancreatic cancer (PC) resection. Methods: This is a retrospective case series study. Clinicopathological data and follow-up information were retrospectively collected from 210 patients who underwent surgery for PC at the Department of Hepatobiliary and Pancreatic Surgery, Peking University First Hospital, between January 2014 and December 2023. There were 114 males and 96 females; the age was (64.5±10.3) years (range: 29 to 89 years). Survival functions based on different times to recurrence and metastasis and different patterns of recurrence and metastasis were estimated using the Kaplan-Meier method, and survival differences among groups were compared using the Log-rank test. Identifying the optimal cutoff for time to postoperative recurrence/metastasis predicting overall survival (OS) in pancreatic cancer patients via the minimum p-value approach. Univariate and multivariate Cox regression analyses were performed to identify independent risk factors affecting OS following pancreatectomy. Results: A total of 210 patients met the inclusion and exclusion criteria. Among them, 71 patients remained free of recurrence and metastasis, while 139 developed recurrence and metastasis. The patterns included local recurrence (n=34), liver metastasis (n=39), lung metastasis (n=11), peritoneal metastasis (n=13), multiple sites metastasis (n=38), bone metastasis (n=3), and kidney metastasis (n=1). OS was significantly shorter in the group with postoperative recurrence and metastasis compared to the group without recurrence/metastasis (23.07 months vs.not reached, P<0.01). The optimal cut-off time distinguishing early from late recurrence and metastasis was 13 months. There was a significant difference in post-recurrence survival between patients with early and late recurrence and metastasis (16.03 months vs. 52.40 months, P=0.009). The Kaplan-Meier survival curve showed that different postoperative recurrence and metastasis patterns had different impacts on OS, with lung metastasis showing the best prognosis compared to local recurrence, liver metastasis, peritoneal metastasis, and multiple sites metastasis (P<0.01). Multivariate Cox analysis revealed that Eastern Cooperative Oncology Group (ECOG) score 1, postoperative carcinoembryonic antigen (CEA) ≥15 μg/L, poor tumor differentiation, postoperative local recurrence, liver metastasis, peritoneal metastasis, and multiple sites metastases are independent risk factors for postoperative recurrence and metastasis (all P<0.05). Conclusions: Considerable prognostic heterogeneity exists in postoperative PC patients depending on the site and pattern of recurrence or metastasis. Specifically, lung metastasis portends a significantly more favorable prognosis than liver metastasis, peritoneal metastasis, local recurrence, or multiple sites metastases. ECOG score 1, postoperative CEA≥15 μg/L, poor tumor differentiation, postoperative local recurrence, liver metastasis, peritoneal metastasis, and multiple sites metastases are independent risk factors for OS in postoperative PC patients.