Department of Hepatobiliary Pancreatic Surgery, Changhai Hospital, Navy Military Medical University (Second Military Medical University), 200433 Shanghai, China.
Department of Hepatobiliary Pancreatic Surgery, Changhai Hospital, Navy Military Medical University (Second Military Medical University), 200433 Shanghai, China; Department of General Surgery, No.971 Hospital of Navy, 266071, Qingdao, Shandong, China.
Surgery. 2020 Dec;168(6):1003-1014. doi: 10.1016/j.surg.2020.02.013. Epub 2020 Apr 19.
Neoadjuvant chemotherapy may benefit patients with pancreatic ductal adenocarcinoma with resectable and borderline disease. Inappropriate use of neoadjuvant therapy, however, may lead to the loss of therapeutic opportunities. Until an effective prediction model of individual drug sensitivity is established, no accurate model exists to help surgeons decide on the appropriate use of neoadjuvant chemotherapy. We hypothesized that early recurrence in patients undergoing upfront, early resection may be an indication for neoadjuvant chemotherapy. Therefore, we aimed to use preoperative clinical parameters to establish a model of early recurrence to select patients at high risk for neoadjuvant chemotherapy.
Patients who underwent resection for pancreatic ductal adenocarcinoma between January 2014 and November 2017 were analyzed retrospectively. After the minimum P-value approach, the patients were divided into three groups: early recurrence, middle recurrence, and late/non-recurrence. Preoperative clinicopathologic factors that could predict early recurrence were included in a Cox proportional hazards regression model for univariate and multivariate analyses. The factors related to early recurrence were included to establish nomogram and decision tree models, which were then validated in 68 patients.
We found that 235 (72.5%) of 324 patients had recurrence with a median recurrence-free survival of 210 days. The early recurrence, middle recurrence, and late/non-recurrence groups differed in preoperative carbohydrate antigen 19-9 and carcinoembryonic antigen levels, "resectability" on cross-sectional imaging, resection requiring a vascular resection, T stage, tumor size, and adjuvant chemotherapy. The best cutoff value of early recurrence was the first 162 days postoperatively. Univariate and multivariate analyses showed that selected preoperative chief complaints, lymph node enlargement and resectability on cross-sectional imaging, preoperative carbohydrate antigen 19-9 levels >210 kU/L, and a neutrophil/lymphocyte ratio >4.2 were independent predictors for early recurrence.
We have successfully built a prediction model of early recurrence of patients with pancreatic ductal adenocarcinoma with the optimal cutoff early-recurrence value of 162 days. Our nomogram and decision tree models may be used to select those at high risk for early recurrence to guide preoperative decision-making concerning the use of neoadjuvant therapy in those patients who have "resectable" disease and not only the more classic criteria of borderline resectability.
新辅助化疗可能使可切除和交界性疾病的胰腺导管腺癌患者受益。然而，新辅助治疗的不当使用可能导致错失治疗机会。在建立有效的个体药物敏感性预测模型之前，没有准确的模型可以帮助外科医生决定新辅助化疗的使用。我们假设，接受 upfront，早期切除的患者早期复发可能是新辅助化疗的指征。因此，我们旨在使用术前临床参数建立早期复发模型，以选择新辅助化疗高危患者。
回顾性分析 2014 年 1 月至 2017 年 11 月期间接受胰腺导管腺癌切除术的患者。采用最小 P 值法，将患者分为三组：早期复发、中期复发和晚期/无复发。纳入单因素和多因素分析的 Cox 比例风险回归模型的术前临床病理因素包括预测早期复发的因素。将与早期复发相关的因素纳入建立列线图和决策树模型，并在 68 例患者中进行验证。
我们发现，324 例患者中有 235 例（72.5%）复发，中位无复发生存期为 210 天。早期复发、中期复发和晚期/无复发组在术前肿瘤标志物 CA19-9 和癌胚抗原水平、影像学上的“可切除性”、需要血管切除的手术、T 分期、肿瘤大小和辅助化疗方面存在差异。早期复发的最佳截断值为术后第 162 天。单因素和多因素分析表明，术前主要症状、淋巴结肿大、影像学上的可切除性、术前 CA19-9 水平>210 kU/L、中性粒细胞/淋巴细胞比值>4.2 是早期复发的独立预测因素。
我们成功建立了一个预测胰腺导管腺癌患者早期复发的模型，最佳早期复发截断值为 162 天。我们的列线图和决策树模型可用于选择那些有早期复发高风险的患者，以指导那些有“可切除”疾病的患者在术前决策中使用新辅助治疗，而不仅仅是更经典的边界可切除性标准。
