Pancreatic index: A prognostic factor of upfront surgery for body or tail pancreatic ductal adenocarcinoma with vascular involvement—A retrospective study

Abstract Background The pancreatic index (PI) is a useful preoperative imaging predictor for pancreatic ductal adenocarcinoma (PDAC). In this retrospective study, we determined the predictive effect of PI to distinguish patients of pancreatic body/tail cancer (PBTC) with vascular involvement who can benefit from upfront surgery. Method All patients who received distal pancreatectomy for PDAC from 2016 to 2020 at the Pancreatic Disease Center, Ruijin Hospital, Shanghai Jiaotong University School of Medicine were considered for the study. A total of 429 patients with PBTC were assessed in relation to the value of PI. Fifty‐five patients were eventually included and divided into low PI group and 29 patients in the normal PI group. Results The median overall survival (mOS) was significantly shorter in the low PI group (13.1 vs. 30.0 months, p = 0.002) in this study, and PI ≥ 0.78 (OR = 0.552, 95% CI: 0.301–0.904, p = 0.020) was an independent influencing factor confirmed by multivariate analysis. Subgroup analysis showed that PI was an independent prognostic factor for LA‐PBTC (OR = 0.272, 95% CI: 0.077–0.969, p = 0.045). As for BR PBTC, PI (OR = 0.519, 95% CI: 0.285–0.947, p = 0.033) combined with carbohydrate antigen 125 (CA125) (OR = 2.806, 95% CI: 1.206–6.526, p = 0.017) and chemotherapy (OR = 0.327, 95% CI: 0.140–0.763, p = 0.010) were independent factors. Conclusion This study suggests that the PI can be used as a predictive factor to optimize the surgical indication for PBTC with vascular involvement. Preoperative patients with normal PI and CA125 can achieve a long‐term prognosis comparable to that of resectable PBTC patients.

Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignant tumor with an increasing worldwide incidence. 1adical surgery remains the primary curative method for pancreatic cancer.Pancreatic body/tail cancer (PBTC) developing from the dorsal primordium was recently proven to be different from pancreatic head cancer (PHC) by different molecular profiles such as different lymphatic metastatic pathway. 2,3PBTC usually involves more metastasis and less resectability than PHC at diagnosis. 4,5Despite the adoption of neoadjuvant, the probability of resection is uncontrolled, and prognosis varies significantly for PBTC with vascular involvement. 6The effect of neoadjuvant therapy and the rate of receiving chemotherapy are lower in PBTC as well. 48][9] Vein reconstruction has been proven to be irrelevant to the long-term prognosis of borderline resectable PBTC. 10 Recently, distal pancreatectomy with en bloc celiac axis resection (DP-CAR) has been performed by the worldwide high-volume pancreatic surgery center for patients with locally advanced PBTC with tumors contacting the celiac axis. 10,11owever, early recurrence after the surgery was observed in that study, which was not predicted preoperatively by current clinically measured factors. 12In general, individual and specific surgical indications and strategies for PBTC are needed.
Pancreatic index (PI) is a useful preoperative imaging predictor for pathological changes reflecting the progression of local fibrosis in the pancreas, and it has been proved to be a strong independent predictor for PDAC basically resolved the influence of subcutaneous and visceral fat, 13 and a decreased PI is relevant to precancerous lesions of PDAC. 14he aim of this study was to evaluate whether a decreased PI can be used to predict the prognosis of upfront surgery for PBTC with vascular involvement at a single institution.

| MATERIALS AND METHODS
The study has been approved by the Ruijin Hospital Ethic Committee of Shanghai Jiaotong University of Medicine (RJ2020-20), and the requirement for informed consent was waived due to the retrospective nature of the study.The study was in accordance with the Declaration of Helsinki.This retrospective cohort study is reported in line with the STROCSS 2021 guideline. 15he data collected were retrospectively controlled and classified.

| Patients
All patients who received distal pancreatectomy (DP) for PDAC from 2016 to 2020 at the Pancreatic Disease Center, Ruijin Hospital, Shanghai Jiaotong University School of Medicine were considered for the study.A total of 429 patients were assessed according to the criteria listed below, and 84 patients were eventually included in the study.
Inclusion criteria: 1. Patients with PDAC located in the body/tail of the pancreas who underwent curative radical DP. 2. PDAC was diagnosed as borderline resectable (BR) according to the consensus proposed by the National Comprehensive Cancer Network without neoadjuvant chemotherapy, as follows: (a) superior mesenteric vein (SMV)/portal vein (PV)-tumor contact less than 180° or greater or bilateral narrowing/occlusion, not exceeding the inferior border of the duodenum; (b) celiac axis (CA)-tumor contact of less than 180° (abutment) without showing deformity/stenosis.3. PDAC was diagnosed as locally advanced (LA) according to the following criteria: (a) CA-tumor contact greater than 180° (encasement) without contact with the proper hepatic artery (PHA), which makes distal pancreatectomy with en bloc celiac resection (DP-CAR) possible; and (b) CHA-encasement without showing contact with the PHA or the presence of variant arterial anatomy, such as the right hepatic artery (RHA) or common hepatic artery (CHA), which makes DP-CAR possible.4 There was no evidence of distant metastasis according to preoperative images.
Exclusion criteria: 1. Distant metastases were confirmed intraoperatively.2. Radical DP was not carried out, or total pancreatectomy was performed instead due to the unexpected progression of the tumor compared to preoperative image results.3. Patients received neoadjuvant therapy.4. Intraoperative freezing examination showed that the pancreatic transection margin was positive, and R0 resection was unrealized.Park W et al. 16 proposed the recent definition of vessel abutment (<180°) or encasement (>180°).Finally, 84 patients were included and divided into the low PI group and normal PI group according to cutoff value (Figure S1).The details of vascular involvement were shown in Table S1.

| Perioperative management and surgical procedure
Preoperative computed tomography (CT) was routinely performed for all patients.The diagnosis of the tumor, the identification of vascular involvement, and the surgical protocol were analyzed and determined by our multidisciplinary team, which consisted of the chief surgeon, first assistant, professor of anesthesiology, and at least two professors of radiology.The status of vascular involvement (encasement or abutment) was discussed, which was crucial for surgical indication.Vascular resection, reconstruction, and DP-CAR were performed to ensure a curative operation through cooperation with the professor of cardiovascular surgery if necessary.The final surgical procedure was intraoperatively chosen by the chief surgeon who was well-experienced (>100 cases) in undertaking a trial dissection.After discharge, all patients without contraindication were suggested to have a subsequent visit within 12 weeks after the operation, and their suitability for adjuvant chemotherapy was evaluated at the Department of Chemotherapy of our Pancreatic Disease Center.The primary chemotherapy regimens were categorized according to the Eastern Cooperative Oncology Group (ECOG) score as follows: ECOG 0-1 scores were treated with gemcitabine + capecitabine or FOLFIRINOX for really fit patients, while patients with ECOG 2 scores received gemcitabine mono. 17,18Due to variations in postoperative recovery, individual reasons, economic factors, and regional differences, patients who eventually did not receive any chemotherapy were categorized as the "no adjuvant chemotherapy" group during follow-up.

| Quantitative evaluation of the CT images
As previously shown, the PI was calculated by dividing the CT density of the pancreas by that of the spleen 13 on plain CT.PI on arterial phase (PI-a) was calculated by dividing the CT density of the pancreas by that of the spleen on contrast-enhanced CT.Preoperative CT scans were regularly performed for each patient.Two individual radiologists who were blinded to the subjects' clinical information measured the mean pancreatic CT number for the largest possible spherical region of interest (ROI) of the pancreatic parenchyma at the resection line determined by the contrast of postoperative imaging with the first assistant if necessary.The mean splenic CT number was measured by tracing the ROI of the entire spleen at the level of the splenic hilum (Figure S2).

| Follow-up visit
After discharge, follow-up was performed by telephone interviews every 2 months, recording the time and location of recurrence and patient survival.Disease-free survival (DFS) and overall survival (OS) were calculated from the date of the operation to the date of tumor recurrence and death, respectively.Tumor recurrence and death were considered event data; no tumor recurrence or death was classified as censored data.Patients lost to follow-up were classified based on the condition of the last followup.Two patients who had postoperative 90-day mortality were excluded.The follow-up visits included 82 patients, excluding the two with 90-day mortality.The rate of loss to follow-up was 1.2%.Follow-up visits ended in October 2022.The median follow-up time was 16.7 months.

| Statistical analysis
Statistical analyses were performed using SPSS statistical software (version 22).Continuous variables are expressed as means with standard deviation, as medians with range, or as rates (percentage).Continuous variables were compared using the Mann-Whitney U test.Categorical variables were compared using the chi-square test and Fisher's exact test in case of small frequencies expected.The cutoff value of PI was determined by a time-dependent receiver operating characteristic curve (ROC) using R language (version 4.2.1).For survival analysis, DFS and OS rates were analyzed by the Kaplan-Meier method and verified by the log-rank test.For all tests, p values less than 0.05 were considered significant.

| Patient characteristics and pathologic variables
There were 55 patients in the low PI group and 29 in the normal PI group included in this study.The ratio of BR (65.5% vs 89.7%) and LA (34.5% vs 10.3%) was significantly different between the low and normal PI groups (p = 0.016).The detailed classification of tumor involvement is shown in Table S1.The remaining demographic characteristics did not significantly differ between the two groups.
There were no significant differences in tumor diameter, surgical margin distance, number of positive lymph nodes, total retrieved lymph nodes, and proportion of chemotherapy between the two groups.The detailed data are shown in Table 1.

| Perioperative risk
Postoperative mortality, reoperation rate, complications, operation time, intraoperative bleeding, and intraoperative transfusion between the two groups were compared.In this study, two patients in the normal PI group experienced postoperative 90-day mortality.One patient experienced acute cerebral infarction on the third day after the operation.He died on the 29th day after the surgery due to circulatory failure and central hyperpyrexia caused by massive cerebral infarction.The other patient had severe postoperative pancreatic exocrine insufficiency and died on postoperative Day 70 at home due to internal milieu disorder.Three patients in the normal PI group and one patient in the low PI group received unplanned reoperation.One patient in the normal PI group suffering grade C POPF received unplanned reoperation on postoperative Day 20 owing to postoperative hemorrhage, and the rest two patients in the normal PI group and one patient in low PI group without POPF underwent reoperation for postoperative hemorrhage caused by the stump of the splenic artery on postoperative Days 3, 4, and 8. Transcatheter endovascular embolization (TME) of the splenic arterial stump by DSA was successfully conducted on postoperative Day 17 in one patient in the low PI group suspected to have abdominal hemorrhage caused by grade C POPF.There were no statistically significant differences in the incidence of postoperative complications such as pancreatic fistula, intra-abdominal abscess, intra-abdominal bleeding, operative time, intraoperative bleeding, or transfusion.Details are shown in Table S2.

| Tumor recurrence
The liver was the main site of tumor recurrence in both groups.The details of the other tumor recurrences are shown in Table S3, and there were no statistically significant differences.

| Survival analysis
In total, 82 patients participated in follow-up visits, excluding the two patients with 90-day mortality.The cutoff value of PI was determined by a time-dependent receiver operating characteristic (t-ROC) curve, which showed the T A B L E 1 Patient characteristics and pathologic variables between low and normal pancreatic index groups.
largest area under the curve (AUC) on postoperative Day 730 (Figure S3).ROC analysis was then carried out to find the cutoff at the greatest AUC of 2-year postoperative survival (Figure S4).The cutoff value of PI-a was calculated by the similar procedure; however, there was no statically diagnostic significance of PI-a (p = 0.506).Univariate and multivariate analyses were performed to verify the statistical significance of the PI and PI-a with well-known clinical risk factors.
The median disease-free time (mDFS) was significantly longer in patients with PI ≥ 0.78 (23.5 vs. 11.3 months, p = 0.015).Further multivariate analysis showed that PI ≥ 0.78 (OR = 0.462, 95% CI: 0.244-0.874,p = 0.018) was an independent factor of DFS.Similar to the univariate survival analysis for OS, we also compared the impact of DP-CAR and arterial abutment on the DFS time.The results indicated that DP-CAR (p = 0.206) and arterial abutment (0.164) were not independent factors affecting postoperative DFS time.Details are shown in Table 3.

| Subgroup analysis
In the subgroup analysis of patients with LA-PBTC, the OS time of patients in normal PI group was longer than that for those in low PI group (mOS: 26.5 months vs. 12.7 months, p = 0.033).Multivariate analysis confirmed that PI was an independent factor (OR = 0.272, 95% CI: 0.077-0.969,p = 0.045) (Figure 1).No other independent influencing factor of OS was found.There was no influencing factor proven independently relevant to the DFS time of LA-PBTC.
In the subgroup of patients with BR PBTC, the OS time and DFS time of patients in the normal PI group were longer than those in the low PI group (mOS: 31.3 months vs. 15.8 months, p = 0.016, mDFS: 19.0 months vs. 7.7 months, p = 0.010).The mOS of patients who received chemotherapy (mOS: 30.0 months vs. 10.7 months, p = 0.000) and patients with CA125 < 35 U/mL (mOS: 30.0 months vs. 10.8 months, p = 0.000) were significantly longer.Further multivariate analysis showed that normal PI (OR = 0.519, 95% CI: 0.285-0.947,p = 0.033), CA125 ≥ 35 U/mL (OR = 2.806, 95% CI: 1.206-6.526,p = 0.017), and chemotherapy (OR = 0.327, 95% CI: 0.140-0.763,p = 0.010) were independent factors of OS (Figure 2).The incidence of PDAC worldwide is projected to increase at an average annual growth of 1.1%. 19Cases of resectable PDAC remain low at 20% at diagnosis due to rapid tumor progression without obvious early clinical features. 20nfortunately, a consideration proportion of patients suffer from short disease-free and overall survival times due to lack of preoperative screening method.Approximately a quarter of patients with resectable tumors experience rapid relapse within 6 months after radical surgery. 21To our knowledge, there is currently no accessible preoperative evaluation method to screen patients with PBTC with early recurrence or poor prognosis, especially in cases with vascular involvement.It is thus necessary to evaluate the tumor condition and screen patients to optimize the treatment strategy before operating on DP or DP-CAR for these targeted patients.The diagnostic indicator, PI, which essentially eliminates the influence of individual differences in obesity was demonstrated to be a strong predictor for PDAC. 13he decrease in PI was closely related to local fatty degeneration and progression of fibrosis in the pancreas instead of visceral fat. 22Localized fat degeneration and intensified fibrosis are manifestations of local chronic inflammatory responses, accompanied by the infiltration of inflammatory cells which is crucial for the formation of an immunosuppressive tumor microenvironment. 23ncreatic epithelial cells are prone to undergoing metaplasia and then dysplasia in a persistent inflammatory environment, eventually leading to the formation of in situ carcinoma.Moreover, tumor cells themselves stimulate pancreatic stellate cells, which further activate the degree of local fibrosis. 24,25Especially for PBTC, tumor cells can spread through the main pancreatic duct to the rest of the pancreas.During this process, tumor cells might obstruct the pancreatic duct, leading to the activation of distant pancreatic stellate cells, thereby intensifying local inflammation and fibrosis.These mechanisms can all be reflected in the decrease of the PI value.Therefore, based on the results of this study and the aforementioned pathological foundation, to our acknowledge, this study is the first to propose preoperative PI should be utilized as a predictive factor to optimize the surgical indication for PBTC with vascular involvement who could benefit from upfront surgery.
Firstly, our results have unveiled that PBTCs with vascular involvement subjected to PI screening, irrespective of BR PBTC with arterial abutment, have attained a protracted survival akin to that observed in cases of resectable pancreatic cancer. 24,25Recent studies prove that simple vessel abutment on preoperative imaging might not be a surgical contraindication or a sign of poor prognosis with recent surgical techniques 7,26,27 which is consistent with our conclusion.These specific patients can benefit from upfront surgery, rather than risking tumor progression by receiving a uniform neoadjuvant treatment due to the lack of personalized regimens.An undeniable point to consider is that, imaging is currently the first-line method for the indication of upfront surgery and neoadjuvant therapy. 9However, it remains challenging to evaluate treatment response and resectability and optimize individual indications for neoadjuvant therapy. 28The unexpected progression of local tumors and a lower resection rate after neoadjuvant therapy were inevitably observed in a previous study. 29In addition, evaluation of artery invasion by preoperative imaging is sometimes ambiguous due to the elasticity of the arterial wall, which may lead to overestimation of local invasion and delay the possibility of primary resection. 30esides, PBTC differs from pancreatic head cancer in that, even when involving the CA, there is still an opportunity for DP-CAR, whereas the latter is more prone to SMA invasion, 31 making surgery unfeasible.2][23] Both previous literature 27 and the results of this study have confirmed the safety of DP-CAR.In this context, the findings of survival analysis of LA subgroup in this study further suggest that through PI screening, patients with high PI who undergo DP-CAR treatment achieve long-term survival similar to that of resectable pancreatic cancer.This emphasizes that even among PBTC cases involving CA, the tumor heterogeneity can lead to variations in the overall disease progression among patients. 32PI may serve as an effective indicator reflecting such differences.
In addition to the PI factor, in our subgroup analysis, we found that a high CA 125 level indicates a poorer prognosis for BR PBTC patients.CA 125 is an important marker that can help predict the outcomes for PDAC. 33hese patients may not benefit from upfront surgery, even if it may involve only venous abutment.The biomarker CA 199 has been identified as adverse factor for pancreatic cancer prognosis 34  node metastasis, thereby predicting the effectiveness of upfront surgery for resectable pancreatic cancer. 35owever, these two factors did not have a significant impact on patient prognosis in this study.We believe a possible explanation is that for BR PBTC patients, CA19-9 levels overall are relatively close and may not be as sensitive as in early diagnosis of pancreatic cancer.CA125, on the other hand, exhibits higher sensitivity in predicting post-surgery outcomes for this subgroup of patients.It is necessary to investigate and confirm the biomarkers, the status of lymph node metastasis, and other influencing factors, in conjunction with PI in future clinical trials to help predict the prognosis.
Alternatively, the predict value of PI-a was unsatisfactory in our study.The intraindividual variability in pancreatic tissue which was caused by abdominal perfusion, cardiac status, and age on contrast-enhanced CT could be an excuse as the previous study mentioned. 36wever, this retrospective study also had some limitations.The sample size of the study needs to be expanded which led to selection bias.Second, our study is a single-center study, and further multicenter clinical studies are needed to establish a more generalized baseline cutoff value.Besides, the predictive effect of PI for patients after neoadjuvant therapy should be verified by randomized controlled trials in future to optimize individual strategy based on the conclusion of this study.Our team aims to lay the theoretical foundation for future worldwide studies in various centers to investigate whether high PI patients benefit from neoadjuvant therapy by sharing the results of this study.

| CONCLUSION
This retrospective study provides evidence that PI should be utilized as a predictive factor to determine the feasibility for upfront surgery for PBTC with vascular involvement.For patients with normal PI and CA125 levels, direct surgery followed by chemotherapy is a safe and personalized high-quality treatment strategy.
and recent research further proposed that CA 19-9 levels can even predict lymph F I G U R E 1 Survival curve of LA subgroup according to the PI (</≥0.78).mOS, median overall survival.

F I G U R E 2
Survival curve of BR subgroup.A. Survival curve of OS according to the PI (</≥0.78),B. Survival curve of DFS according to the PI (</≥0.78),C. Survival curve of OS according to the adjuvant chemotherapy (±), D Survival curve of OS according to the CA125 (</≥35 U/mL).mDFS, median disease-free survival; mOS, median overall survival.QIAN et al.
Prognostic factors in univariate analysis (overall survival).