Assessment of Standardization Circumferential Resection Margin Status Inuence Survival following Pancreaticoduodenectomy for Pancreatic Ductal Adenocarcinoma

Background: Not only to assess the prognostic inuence on standardization circumferential resection margin R0 and R1 Status but also to research the prognostic inuence on adjuvant chemotherapy to PV/SMV, SMA resection margins (cid:0) 1mm. The SMV and SMA resection margins had an important prognostic inuence to PDAC patients, and the survival prognosis of R1 status(resection margin (cid:0) 1mm) was poor. Methods: A total of 228 patients performed PD between 2015 and 2019 were included, which was assessment of standardization circumferential resection margin status and survival prognosis. There were cancer cells within 1mm clearance of PV/SMV and SMA resection margins named R1 PV/SMV, SMA , and no cancer cells named R0 PV/SMV, SMA. Results: The resection margin 1mm clearance of PV/SMV, SMA (P=0.010) and postoperative adjuvant chemotherapy (P=0.001) were prognostic independent predictors. The median survival time was 22 months of 166 R0 PV/SMV, SMA patients (73%) compared to 15 months of 62 R1 PV/SMV, SMA patients (27%) (P=0.005). There was the statistical signicance of survival time between the adjuvant chemotherapy group and the none-adjuvant chemotherapy group (P=0.000). In the R1 PV/SMV, SMA group, there was no statistical signicance of survival time between the adjuvant chemotherapy patients and the none-adjuvant chemotherapy patients (P=0.208). Conclusions: Patients undergoing PD for PDAC, postoperative adjuvant chemotherapy could not improve the poor survival prognosis of R1 PV/SMV, SMA resection patients. The resection margins of PV/SMV, SMA had a greater prognostic inuence on survival than postoperative adjuvant chemotherapy. Trial This study was a retrospective cohort study which was approved by West China Hospital All patients underwent surgical treatment in the Department of Pancreatic Surgery, Sichuan University West China Hospital, during a 4-year period (January 1, 2015 to March 31, 2019). The decision to perform surgical treatment was made by a multidisciplinary team (MDT) including radiologists, surgeons, pathologists and oncologists. All patients were performed classic PD by 2 surgeons (XB.L and NW.K). This study was limited to patients undergoing PD for resection of pancreatic head adenocarcinoma which all the pathological diagnosis was PDAC. And other lesions such as ampullary, intraductal papillary mucinous neoplasms (IPMN), mucinous cystadenocarcinomas, duodenal or distal bile duct adenocarcinomas and were excluded. And we also had excluded: (a) the patients had received neoadjuvant therapy, (b) the pathological report had not clearly dened the distance between the circumferential margin (CRM) and the tumor.


Background
Pancreatic ductal adenocarcinoma (PDAC) is the malignant tumor of the digestive system with a poor prognosis. The latest research report showed that the annual incidence of PDAC was 12.9/100000, and the death rate was 11/100000 [1] . In the last 20 years, because of progression the comprehensive treatment, the 5-year survival rate of PDAC was higher than before [2,3] . Compared to the other malignant tumor, however, the 5-year survival rate of PDAC was only 8% and it was still lower [4] . The 5-year survival rate of PDAC patients who had received surgical treatment would reach 12.4%, and the 5-year survival rate of PDAC patient who was metastatic disease could only be 2.9% [3,4] . And the latest studies predict that the pancreatic adenocarcinoma will be the second most fatal malignancy in 2030 [5] . Nowadays, surgical treatment was still the only radical cure method for PDAC [6][7][8] .
The 90% of PDAC located in pancreatic head. Previous studies had shown that resection margin was an independent prognostic factor affecting the prognosis of pancreatic head adenocarcinoma [9][10][11][12][13][14][15] . However, some studies had shown that the resection margin was not the independent prognostic factor [16][17][18][19] . The reasons caused those results might be none standardized de nition of resection margins or none standardized pancreaticoduodenectomy(PD). Then, the Royal College of Pathologists (RCP) had recommended the de nition of R1 status was that there were the cancer cells within the 1 mm clearance of the circumferential margin of pancreatic surgical specimen [20,21] . After that, the International Study Group of Pancreatic Surgery (ISGPS) and College of American Pathologists (CAP) also recommended whether there were cancer cells within the resection margin 1 mm to de ne the R0 status or R1 status [22,23] .
With the standard of pathological examination for the pancreatic resection margins, many studies had begun to focus on prognostic in uence of pancreatic resection margins [10,11,14,24] . Those studies had revealed that in resectable pancreatic head carcinoma patients (exclude vascular invasion patients) who the cancer cells were not microscopic examined within 1 mm clearance of resection margins, the 5-year survival rate could reach 48%. However, in the patients who the cancer cells were microscopic examined within 1 mm clearance of resection margins, the 5-year survival rate was only 14% [10,11,14,24] . Those studies also revealed that the resection margins were an independent risk factor for survival prognosis, and the R1resection status had had a poor survival prognosis. And there were multiple resection margins of PD. Whether each resection margin had the same effect on survival prognosis? And, whether it was prognostic signi cance to study each resection margin? Those questions were controversial. Previous studies had suggested that each resection margin of PDAC had the same effect on survival prognosis [25][26][27][28][29] .In 2020, a study in < Pancreatology > had revealed that not all resection margin positivity had the same prognostic signi cance, and superior mesenteric artery(SMA) and superior mesenteric vein(SMV) margin positivity had a worse survival impact [30] .
In the previous studies, the SMV and SMA resection margins had an important prognostic in uence to PDAC patients, and the survival prognosis of R1 status(resection margin 1 mm) was poor. However, the previous results had not analyzed the prognostic in uence of adjuvant chemotherapy on the R1 status patients. Whether e cacy of adjuvant therapy could improve the survival prognosis of R1 status patients? And whether the R1 status patients who had received adjuvant chemotherapy could be the same survival rate to the R0 status patients? If so, then the patients of resectable pancreatic head carcinoma would perform PD rst and then received adjuvant chemotherapy.
In this study, we had examined all the resection margins of PD and assessed the Port Vein/SMV, SMA resection margins. We de ned the microscopic examined cancer cells within 1 mm clearance of PV/SMV, SMA resection margins as R1 resection(resection margins 1 mm), and none microscopic examined cancer cells within 1 mm clearance of PV/SMV, SMA resection margins as R0 resection(resection margins 1 mm). Consequently, we not only assessed the prognostic in uence on R0 and R1 status but also researched the prognostic in uence on adjuvant chemotherapy to PV/SMV, SMA resection margins 1 mm.  March 31, 2019). The decision to perform surgical treatment was made by a multidisciplinary team (MDT) including radiologists, surgeons, pathologists and oncologists. All patients were performed classic PD by 2 surgeons (XB.L and NW.K). This study was limited to patients undergoing PD for resection of pancreatic head adenocarcinoma which all the pathological diagnosis was PDAC. And other lesions such as ampullary, intraductal papillary mucinous neoplasms (IPMN), mucinous cystadenocarcinomas, duodenal or distal bile duct adenocarcinomas and were excluded. And we also had excluded: (a) the patients had received neoadjuvant therapy, (b) the pathological report had not clearly de ned the distance between the circumferential margin (CRM) and the tumor.
The inclusion criteria were: (a) age 18-75 years, (b) the computed tomography (CT) showed the tumor had localized in the head of the pancreas, (c) there was no greater than 180° circumferential involvement of SMV and no overt arterial involvement [14] , (d) follow-up data were completed.
No patients got neoadjuvant therapy. Postoperatively, some patients did not choose to receive chemotherapy. Follow-up which was up to December 4, 2019, comprised the out-patient and telephone reviews. When the Carbohydrate antigen19-9 (CA19-9) was high, CT scans were given to the postoperative patients.

Pathology Assessment
Measurement, stained and xation of the surgical specimen and de nition of the resection margins, pathological examination, and pathological report were all based on the guidelines of the RCPath. (https://www.rcpath.org/profession/guidelines/cancer-data sets-and-tissue-pathways.html) We had paid particular attention to the CRM de nition of the postoperative specimens, which included surface of anterior, posterior, SMV, and SMA (FIGURE 1A). And also, we had examined the transection margins which consisted of the corpus/neck margin, proximal and distal gastric/ jejunum margins and the bile duct margin. Then we had stained the PV/SMV and SMA resection margins(FIGURE 1B), with glacial acetic acid strengthened. The standard of choosing a transverse plane was not only to show the relationship between tumor and adjacent tissues but to show the distance between tumor and resection margin. Thereafter 3 cuts were made from the tumor in head of pancreas. Then the specimens were xed with 10% neutral formalin solution for 24 to 48 hours. After that, the specimen was further microscopic examined as follows: the tumor, anterior margin, posterior margin, PV/SMV and SMA margin, proximal gastric margin, distal jejunum margin, common bile duct margin and gallbladder, corpus/neck margin [26] .
In microscopic assessment, each resection margin was measured at 1/10 mm intervals up to 5 mm which the blocks of resection margins were cut. And the standard pathological report included as follows: maximum tumor diameter and extent; tumor grade and location of local spread; lymphatic, venous and perineural invasion; total number of lymph nodes examined; number positive lymph nodes. We had performed Tumor Node Metastasis (TNM) which corresponded to the staging system of Union for International Cancer Control (UICC) or American Joint Commission on Cancer (AJCC).
In this study, we had microscopically assessed the R0 or R1 status according to the RCPath criteria. And we had de ned the microscopically positive margin as<1 mm when there were cancer cells within 1 mm clearance of PV/SMV and SMA resection margins, and the other margins were not. Those formed the PV/SMV and SMA resection margins>1 mm group(R1 PV/SMV, SMA ). Also, we had de ned the microscopically negative margin as>1 mm when there were no cancer cells within 1 mm clearance of PV/SMV and SMA resection margins, and the other margins were not as well. And those formed the PV/SMV and SMA resection margins>1 mm group(R0 PV/SMV, SMA ).

Statistical Analysis
All statistical analyses were processed by SPSS version 20.0 (SPSS Inc. Chicago, IL). The χ 2 test was used for counting data, and Fisher's exact test was used for frequency which was less than 5. Shapiro-Wilk test analyzed the normal distribution of data, and the Levebe test analyzed the homogeneity of variance. The T-test was used for Gaussian distribution, which was expressed as mean ± standard deviation, and the statistical value was t. And the Mann-Whitney U test was used for the abnormal distribution, which was expressed as M (P25, P75), and the statistical value was Z. Kaplan-Meier survival analysis analyzed the overall survival withdrawing a survival curve. The Log-Rank test was used for univariate analysis. The proportional hazards regression (Cox regression) model was used for multifactor analysis. Statistical signi cance was set at a P<0.05.

Different Resection Margin Status of the Patient Cohort from Pathological Report
Patients whose resection margins were not clearly indicated in the pathological report were excluded.
There were a total of 228 patients performed PD between 2015 and 2019 in this study. (FIGURE 2)According to the Pathological report: there were no cancer cells in anterior margin, posterior margin, proximal gastric margin, distal jejunum margin, common bile duct margin and gallbladder, corpus/neck margin of the all the 228 patients. 62 (27%) patients had histologically positive which there were cancer cells within 1 mm (<1 mm) in PV/SMV and SMA margins and were R1 PV/SMV, SMA resections group. The others 166(73%) patients which there were no cancer cells within 1 mm (>1 mm) were R0 PV/SMV, SMA resections group.

Clinico-pathologic Characteristics Of Different Resection Margin Status Groups
In this study, tumor grade is categorized into poorly differentiated tumors, poorly-moderately differentiated tumors, moderately differentiated tumors, moderately-well differentiated tumors and well differentiated tumors. All patients were tumor stage T1, T2 or T3. And the AJCC criteria of TNM staging were applied. The clinico-pathologic characteristics of the different resection margin status groups are summarized in Table 1.  The results of the Log-rank test showed that resection margins status (P = 0.005) and adjuvant chemotherapy (P = 0.000) were associated with survival prognosis (Table 2A). The results of Cox regression showed that resection margins status (P = 0.010) and adjuvant chemotherapy (P = 0.001) were independent risk factors for survival prognosis (Table 2B).

Relationship Between Survival And Resection Margin Status
In a total of 228 patients, 215(94.3%) patients were followed up after surgical. 152(91.6%) patients were followed up in the R0 PV/SMV, SMA group. And 57(91.9%) patients were followed up in the R1 PV/SMV, SMA group. The follow-up time ranged 6-60 months, and the median follow-up time was 20 months (Table 3) There was the statistical signi cance of survival time between the adjuvant chemotherapy group and the none-adjuvant chemotherapy group (P = 0.000), (Table 3 In the R1 PV/SMV, SMA group, there was no statistical signi cance of survival time between the adjuvant chemotherapy patients and the none-adjuvant chemotherapy patients (P = 0.208) ( Table 3) (FIGURE 3F).

Discussion
The previous study had suggested that the patients of R1 Previous studies had revealed that smoking and obesity were modi able risk factors for PDAC, and that was to say, smoking cessation and diet could reduce the risk of PDAC [31] . Male had nearly 30% more than women for the incidence of PDAC [32] . In our study, 137 (60%) patients were male, and 91 (40%) patients were female, which was consistent with previous studies. The cause of difference in incidence might be male's preference for alcohol, smoking and obesity. However, the gender, age and BMI were not the independent prognostic factors in this study, which might relate to the small sample and PDAC had caused patients to lose weight quickly.
A series of previous studies on resection margins in which the standardized histopathological examination had reported that the resection margins had prognostic in uence on survival. Popescu I et al had reported that R0 resection of PD in PDAC could improve overall postoperative survival rate [33] .
Tummers W.S et al had researched the relationship between the R0 and (or) R1 resection and postoperative local recurrence, overall survival, which had reported that the R status of resection margin was associated with the postoperative local recurrence, overall survival of PDAC patients [34] . Especially in N1staging patients, there were having similar recurrence characteristics between R0 and R1 resection [34] . Demir I.E et al had reported that the PD for advanced PDAC was more likely to be R1 resection, and the resection margins could be independent risk factor for survival prognosis [35] . In a prospective randomized controlled study, Delpero J.R et al had also reported that the R status of resection margin was associated with the survival prognosis, and they further reported that resection margins < 1 mm was independent risk factor for survival prognosis [36] . After standardized histopathological examination on the resection margins of specimen, Hank T et al had reported that the resection margins 1 mm was independent risk factor for survival prognosis [37] . In this study, there was statistical signi cance between R1 PV/SMV, SMA group and R0 PV/SMV, SMA group (P = 0.005). And the 1, 2, 3-year survival rate of R0 PV/SMV, SMA group patients were signi cantly better than that of R1 PV/SMV, SMA group(73%VS 61% 40%VS 13% 19%VS 10%). Therefore, the resection margin 1 mm clearance of PV/SMV, SMA was the independent risk factor for survival prognosis in this study.
With the standardized histopathological examination on the specimen, most R0 resection of PD in the past was actually R1 resection, which the reason might be that the PV/SMV, SMA resection margins were nonstandardized of radical resection. For this cohort of patients, neoadjuvant therapy and adjuvant chemotherapy might have been critical to improving survival prognosis. NCCN had recommended adjuvant chemotherapy following resection of PDAC as standard treatment since 2000. Postoperatively, however, less than 50% of patients received adjuvant chemotherapy [38] . In a retrospective study of PDAC, Shaib W.L et al had reported that the adjuvant chemotherapy could prolong survival time [39] . In the same study, Nagrial A.M et al had reported that the elder PDAC patients who had received postoperative adjuvant chemotherapy could live longer [40] . This study had provided adjuvant chemotherapy evidence for elderly PDAC patients after PD. In a study of adjuvant chemotherapy with Gemcitabine, Oettle H et al had reported that the PDAC patients who had received standardized treatment of Gemcitabine for 6 months would live longer and get higher disease-free survival [41] . Then in the study of system evaluation in PDAC patients, Parmar A et al had not only con rmed that postoperative adjuvant chemotherapy could signi cantly improve the survival prognosis but also had recommended the mFOLFIRINOX as the preferred adjuvant chemotherapy regimen [42] . And in our study, there was statistical signi cance of survival time between the adjuvant chemotherapy group and the none-adjuvant chemotherapy group (P = 0.000). The adjuvant chemotherapy group patients had higher 1-year, 2-year, and 3-year survival rates than the none-adjuvant chemotherapy group (83%vs65%, 56%vs 30%,31%vs14%), which was consistent with previous studies.     Flow chart of screening patients.

Figure 2
Flow chart of screening patients.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. STROBEchecklistcohort.docx