Standard Pancreatoduodenectomy Versus Radical Pancreatoduodenectomy With Retroperitoneal Nerve Resection in Patients With Pancreatic Head Cancer (RAPNER): A Multicentre Randomized Controlled Trial

Abstract

Background: Pancreatic cancer is a highly fatal disease and is becoming an increasingly common cause of cancer mortality. The extent of pancreatoduodenectomy for pancreatic head cancer is still controversial, and more high-level clinical evidence is needed. We compared the outcomes of standard pancreatoduodenectomy (SPD) with those of radical pancreatoduodenectomy (RPD) with retroperitoneal nerve resection in pancreatic head cancer patients in this multicentre randomized controlled trial.  Methods: RAPNER was open-label, multicentre randomised trial, patients aged 18–80 years with stage I or II pancreatic head cancer undergoing curative resection were enrolled at 6 centres from China. Patients were allocated (1:1) by use of a computer-based randomisation service to assigned to receive standard pancreatoduodenectomy (PD) or radical PD, with the latter followed by the combined dissection of additional lymph nodes (LNs), nerves and soft tissues 270° on the right side surrounding the superior mesenteric artery and celiac axis. This trial is registered with Chinese Clinical Trial Registry (http://www.chictr.org.cn), ChiCTR-TRC-12002548. Findings: Between October 3, 2012, and February 15, 2017, 468 patients were enrolled; of these, 400 were randomly assigned to the RPD group (n=201) or the SPD group (n=199). The operation time was longer with RPD than with SPD, but the R1 resection rate was slightly higher in the SPD group (12·56%) than in the RPD group (8·46%). Both the total number of retrieved LNs and number of positive LNs were significantly higher in the RPD group (20 (7-63) and 1 (0-29)) than in the SPD group (15 (7-63) and 1 (0-11)). Both the morbidity and inpatient mortality rates were comparable between the two groups. The median overall survival (OS) of the RPD and SPD groups were 23·0 months and 20·2 months (p=0·100), and the disease-free survival (DFS) were 15·6 months and 13·5 months (p=0·060). In the subgroup with a preoperative CA19-9 level less than 200 U/ml, the median OS in the RPD group was 30·8 months, which was significantly longer than that in the SPD group (20·9 months; p=0·009), and the median DFS was significantly longer in the RPD group than in the SPD group (22·3 months vs. 15·0 months, p=0·002). Furthermore, clinical benefit could be found in the RPD group in the subgroup of well & moderate histological differentiation, preoperative CEA level less than 5·00 ng/ml, R0 resection and postoperative adjuvant chemotherapy based on DFS analysis. A multivariate Cox proportional hazards model showed that initial CA19-9≥200 U/ml, N1 stage and poor differentiation suggested poor OS. Radical resection achieved equivalent prognosis regardless of perineural invasion except for aortic nerve plexus invasion. The RPD group exhibited significantly lower locoregional and mesenteric LN recurrence rates but a higher peritoneal seeding frequency. The RPD group also exhibited significantly less back pain intensity 6 months after surgery than the SPD group.  Interpretation: This study suggests that in pancreatic head cancer patients, RPD with retroperitoneal nerve resection does not provide a significant survival benefit compared with SPD. RPD can be performed safely, and be considered a treatment option with improving prognosis for patients with a low chance of systemic metastasis in pancreatic head cancer. Trial Registration: ChiCTR-TRC-12002548. Funding Statement: Sun Yat-sen University Clinical Research 5010 Program (Grant No. 2012007), China; National Natural Science Foundation of China (Grant No. 81871945). Declaration of Interests: We declare no competing interests. Ethics Approval Statement: This study was approved and overseen by the clinical ethics committee of each participating hospital.