Multicenter randomized phase II trial of prophylactic right‐half dissection of superior mesenteric artery nerve plexus in pancreatoduodenectomy for pancreatic head cancer

Abstract Aim Right‐half dissection of the superior mesenteric artery (SMA) nerve plexus in pancreatoduodenectomy for pancreatic cancer was initiated to accomplish R0 resection; however, subsequent refractory diarrhea was a major concern. This study aimed to evaluate the necessity of this technique. Methods From April 2014 to June 2018, 74 patients with pancreatic head cancer were randomly allocated to either Group A, in which right‐half dissection of the SMA nerve plexus was performed (n = 37), or Group B, in which total preservation of the nerve plexus was performed (n = 37). Short‐term, long‐term, and survival outcomes were prospectively compared between the groups. Results The patient demographics, including the R0 resection rate, were not significantly different between the groups. Postoperative diarrhea occurred in 26 (70.3%) patients in Group A and 18 (48.6%) patients in Group B. There was a tendency for the development of severe diarrhea in Group A within 1 year postoperatively, and the frequency of diarrhea gradually decreased within 2 years, although that did not affect tolerance to adjuvant chemotherapy. There was no difference in either locoregional recurrence (27.0% vs 32.4%) or systemic recurrence (46.0% vs 46.0%). The median overall survival time in Groups A and B was 37.9 and 34.6 months, respectively (P = 0.77). Conclusion We did not demonstrate a clinical impact of right‐half dissection of the SMA nerve plexus on locoregional recurrence or survival. Therefore, the prophylactic dissection of the SMA nerve plexus is unnecessary given that refractory diarrhea could be induced by this technique (UMIN000012241).


| INTRODUC TI ON
Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies worldwide. It currently ranks as the third leading cause of cancer-related death and is estimated to become the second leading cause by 2030. 1 The survival outcome of PDAC is still dismal, with actual 5-year survival rates ranging from 10% to 20%. 2 To conquer this disease, extended lymphadenectomy with dissection of the nerve plexus has been challenged in some randomized controlled trials; however, survival outcomes have not been improved. [3][4][5][6] Among some extended surgery techniques for PDAC, right-half dissection of the superior mesenteric artery (SMA) nerve plexus has been considered a standard technique of pancreatoduodenectomy in Japan. 7,8 Although this concept was initiated to accomplish R0 resection and totally extirpate the tissue that had the potential for nerve plexus invasion, subsequent refractory diarrhea is a major concern. This refractory diarrhea often decreases the relative dose intensity of adjuvant chemotherapy, which has been demonstrated to be a significant prognostic factor in PDAC. 9,10 Therefore, unnecessary right-half dissection of the SMA nerve plexus in pancreatoduodenectomy might induce refractory diarrhea, potentially worsening the prognosis.
This multicenter randomized phase II trial was performed to judge the necessity of right-half dissection of the SMA nerve plexus in the context of the locoregional recurrence rate. We randomly assigned the enrolled patients into those who did and did not undergo right-half dissection of the SMA nerve plexus and statistically analyzed the short-term, long-term, and survival outcomes between the groups.

| Patient recruitment
From April 2014 to June 2018, 80 patients from Japanese regional high-volume centers (Nagoya University Hospital, Kansai Medical University Hospital, and Toyama University Hospital) were enrolled in this phase II study. The eligibility criteria were planned pancreatoduodenectomy for pancreatic head cancer, no invasion of the SMA nerve plexus on preoperative imaging studies and estimated R0 resection by preservation of the nerve plexus, and age of ≥20 to ≤75 years. The exclusion criteria were a malignant pancreatic tumor other than pancreatic cancer, invasion or suspected invasion of the SMA nerve plexus, liver cirrhosis or active hepatitis, interstitial pneumonia or emphysema, artificial dialysis for chronic renal failure, other active concomitant malignancies, and preoperative moderate diarrhea. Written informed consent was obtained from all patients.
This study was conducted in accordance with the Declaration of Helsinki, and the study protocol was approved by the institutional review board of the affiliated hospital. The registration number of this clinical trial is UMIN000012241.

| Study design
In this multicenter randomized phase II study, 80 enrolled patients were randomly allocated to Group A, in which right-half dissection of the SMA nerve plexus was performed (n = 40), and Group B, in which total preservation of the nerve plexus was performed (n = 40).
The modulators of allocation were the institution, portal system invasion, and preoperative radiation therapy. In Group A, the nerve plexus was dissected from the bifurcation of the middle colic artery to the root of the SMA, along with the pancreatic head. 8 The remnant SMA nerve plexus that was the nearest to the presenting part of the tumor was pathologically evaluated intraoperatively in at least two sites. 11 In Group B, the pancreatoduodenectomy was performed with exposure and preservation of the nerve plexus in its entire circumference ( Figure 1). D2 lymph node dissection was conducted in both groups; thus, the only difference was the performance of righthalf nerve plexus dissection. A close-up photograph of the SMA was taken to confirm the accuracy of the procedure ( Figure S1). The CONSORT diagram of this study is shown in Figure 2. Other pathologies were diagnosed in three patients, and another three patients' conditions were deemed unresectable; therefore, these six patients were excluded. Finally, 74 patients (37 in each group) were analyzed.
The primary endpoint was the local recurrence rate at 2 years after surgery. The secondary endpoints were the frequency of diarrhea, accumulation of ascites, peritoneal recurrence rate, initiation ratio and start of adjuvant chemotherapy after surgery, recurrence-free survival (RFS), and overall survival (OS). In the current study, zero mm rule was utilized to evaluate the residual tumor (R) status.

| Postoperative follow-up
Oral S-1 (oral 5-fluorouracil prodrug tegafur with oteracil and gimeracil) was initiated as adjuvant chemotherapy from 3 to 10 weeks postoperatively. S-1 was administered from days 1 to 28 followed by a 2-week rest period. 10 In outpatient clinics, the patients were followed up once per month for 6 months and every 3 months thereafter. Blood examinations, including measurement of serum tumor markers, were performed at every outpatient care visit, and dynamic computed tomography or magnetic resonance imaging was performed every 3 months.

K E Y W O R D S
pancreatic head cancer, pancreatoduodenectomy, phase II trial, superior mesenteric artery nerve plexus

| Statistical analysis
The sample size was calculated as follows. The local recurrence rate at 2 years after surgery in Group A was estimated to be 15% to 20%. Assuming a null hypothesis of 30% and an alternative hypothesis of 10% with a one-sided type I error of 0.05 and power of 0.9, enrollment of 37 patients per arm was required. Continuous variables are expressed as mean ± standard deviation. RFS was defined as the time from surgery to confirmation of any recurrence. OS was defined as the time from surgery to death by any cause. Survival analysis based on the Kaplan-Meier method and log-rank tests were also performed. The last follow-up date was April 2020. All statistical analyses were performed using JMP Pro version 14.2.0 (SAS Institute, Cary, NC, USA) and R version 3.5.3 (http//www.r-proje ct.org/). The level of statistical significance was set at P < 0.05.

| Patient demographics
Patient demographics are compared between the two study groups in Group A and four (10.8%) in Group B, respectively. Among the operative factors, there was no difference in the operation time or blood loss. Pathologically, invasion of the SMA nerve plexus was reported in two (5.4%) and three (8.1%) patients in Groups A and B, respectively, and the R0 resection rate was also comparable between the two groups. As for detailed site of R1, no patient with positive SMA margin was observed in Group A, whereas one was observed in Group B.

| Short-term outcomes
The short-term outcomes after pancreatectomy are compared between the two study groups in Table 2  TA B L E 1 Comparison of demographics between patients with and without righthalf dissection of SMA nerve plexus postoperatively, and the frequency of diarrhea gradually decreased within 2 years in both groups. Also, the postoperative time-course alteration of the incidence for clinically significant grade 2 or 3 diarrhea in each group was depicted in Figure S2, which showed the same pattern. Opioid use was found in four patients for Group A and two in Group B. With respect to other complications, we found no difference in the development of pancreatic fistula, ascites, intraabdominal infection, delayed gastric emptying, or intra-abdominal bleeding. No reoperations or 30-day mortality occurred in this trial.
When the status of adjuvant chemotherapy was compared, we found no differences in the initiation ratio, start of treatment after surgery, or completion ratio between the groups.

| Long-term outcomes
The long-term outcomes are compared between the two study groups in Table 3

| Survival outcomes
Finally, RFS and OS are compared between the two study groups in Figure 4.

| D ISCUSS I ON
In this trial, we randomized the enrolled patients into Group A, in which right-half dissection of the SMA nerve plexus was performed, and Group B, in which the SMA nerve plexus was preserved. We then compared the clinical and survival outcomes between the groups. We found no significant difference in either the total recurrence rate or the locoregional recurrence rate between the groups.
There was no clinical benefit with respect to long-term survival in  reported that extended lymphadenectomy with dissection of the nerve plexus did not provide a significant survival benefit compared with standard resection in pancreatic head cancer; however, they did not genuinely evaluate the benefit of dissection of the SMA nerve plexus.
Generally, refractory postoperative diarrhea is a major concern after dissection of the SMA nerve plexus. Inoue et al 19  Besides lymphadenectomy, dissection of the nerve plexus has been considered an indispensable surgical technique to obtain R0 resection and subsequent reduction of locoregional recurrence.
In particular, when we encounter a tumor with an extrapancreatic nerve plexus and perineural invasion such as borderline resectable cancer, or when conversion surgery for locally advanced cancer is performed, partial or half dissection of the SMA nerve plexus is definitely required to accomplish R0 resection. In this trial, the patients with invasion or suspected invasion of the SMA nerve plexus were excluded from the criteria, and we focused on the necessity of prophylactic dissection of the nerve plexus.
Therefore, we believe that surgeons should strictly refrain from the prophylactic dissection of the nerve plexus because this technique can result in refractory diarrhea, malnutrition, and poor quality of life. [6][7][8]20 In recent years, preoperative chemotherapy or chemoradiotherapy plus adjuvant therapy in the treatment of resectable or borderline resectable pancreatic cancer have been widely spread as the multidisciplinary therapy. This trend basically indicates the difficulty to completely cure the pancreatic cancer by surgery alone. In this situation, one of the clinical benefits for this neoadjuvant therapy could be the prevention of microinvasion or micrometastasis around the primary tumor. Although our study could not demonstrate that the prophylactic dissection of the SMA right-half nerve plexus could yield a prognostic benefit in the patients with resectable pancreatic cancer, the preoperative chemotherapy or chemoradiotherapy should be strongly recommended to reduce the incidence of local recurrence in future studies.
In conclusion, we conducted a multicenter randomized phase nerve plexus in pancreatoduodenectomy for pancreatic head cancer.
We did not demonstrate a clinical impact on the locoregional recurrence rate or survival outcomes. Therefore, the prophylactic dissection of the SMA nerve plexus is unnecessary given that refractory diarrhea could be induced by this technique.