Circulating inflammatory indices refine prognostic stratification and complication risk in pancreatic cancer following radical surgery

Introduction

An alarming death rate is related to pancreatic ductal adenocarcinoma (PDAC). A mere 10% of people with this condition live through 5 years (1). The sole curative treatment for patients is surgery. Nevertheless, most patients experience severe illness, and merely around 20% to 25% of cases are suitable for surgery (2,3). PDAC frequently invades critical blood vessels, further complicating surgical intervention, reducing the chances of successful resection, and increasing the risk of perioperative mortality (4-6). Even after successful tumor resection, the risk of recurrence remains high. It remains controversial whether direct surgery gives an extended lifespan benefit in individuals with vascular invasion (7-9). PDAC surgery also carries several risks, such as infections, slowed down gastric emptying, and pancreatic fistula. These consequences have the potential to drastically decrease patients’ quality of life. Not all eligible patients benefit from direct surgery, while neoadjuvant therapy (NAT) can advance patients’ quality of life and survival (10-12). Timing of surgery for borderline resectable (BR) patients remains controversial.

As a result, carefully selecting patients who might gain value from surgery is critical to improving their well-being and quality of life. Several clinical trials were undertaken to discover risk factors that can reliably estimate patient prognosis (13). Although numerous prognostic factors have been identified, including inflammatory factors, tumor size, and lymph node metastasis (14-16), there is currently no simple model that could be used to combine these factors and, hence improve the prediction of post-operative survival after surgery. As a result, there is a need to construct an uncomplicated clinical model that can reliably estimate survival following surgery and identify the best treatment option.

Inflammation plays a crucial role in tumor development and progression, including PDAC (17-21). An immune-suppressive milieu brought on by inflammation may enable malignancies to avoid detection. It can also disrupt the tumor microenvironment, promote angiogenesis, and facilitate metabolic reprogramming for supporting tumor growth (22). Obesity, pancreatitis, and tumor-related factors like obstruction-related pancreatitis can increase the inflammatory response in PDAC (23,24). The tumor can also obstruct the bile and pancreatic ducts, leading to inflammation that further triggers tumor progression (25). The goal of this research was to create a nomogram combining pre-operative clinical and inflammatory parameters to forecast surgical problems and post-operative survival in patients with PDAC. Depending on pre-operative risk factors, the results of this research may be utilized for enhancing PDAC treatment. We present this article in accordance with the TRIPOD reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0008/rc).

Methods

Eligibility criteria

The Ethics Committee of The First Affiliated Hospital of Soochow University (No. 2025252) gave its approval to this single-center retrospective study. The study was conducted in accordance with the Declaration of Helsinki and and its subsequent amendments. Patient consent was waived due to the retrospective nature of the study. This research recruited patients with pathologically diagnosed PDAC who had undergone intraoperative examination at our center between September 2012 and November 2020 to determine whether aggressive surgery for PDAC was feasible. Patients who had undergone pancreatic surgical procedures for trauma or to remove a benign or malignant tumor within the bile duct or duodenum were excluded. In addition, patients in whom the intraoperative exploration revealed unresectable malignant tumors and those who underwent surgical biopsy or diversion surgery in which the original lesion was not radically resected were also excluded. We excluded people with previous instances of malignant tumors, incomplete clinical data, and those who passed away during the perioperative phase. Furthermore, patients who received neoadjuvant chemotherapy or conversion therapy prior to surgery were not included in this study.

Data collection

We used medical records to collect pre-operative medical history, laboratory tests, and radiology results. Pre-operative enhanced magnetic resonance imaging (MRI) and computed tomography (CT) results were acquired within two weeks before surgery. All medical images were evaluated by two experienced radiologists to assess vascular invasion. Based on the vascular invasion levels, we divided the tumors as BR, unresectable, and resectable (R) (26). CT and MRI images were also used to assess the pancreatic duct’s diameter and the maximal diameter of the patient’s primary pancreatic lesion.

All laboratory tests including routine blood count, biochemistry, and inflammatory indices were performed within one week before surgery. We evaluated many serum inflammatory markers, such as the platelet-to-lymphocyte ratio (PLR), neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), and C-reactive protein (CRP). We calculated the inflammation-based index (IBI) and systemic immune-inflammation index (SII). We calculated the SII as follows: “SII = (neutrophil count × platelet count) / lymphocyte count”, where the neutrophil and platelet counts reflect inflammation, and the lymphocyte count indicates immune function. The IBI was calculated using the formula IBI = CRP (mg/L) × NLR, where CRP represents systemic inflammation and the equilibrium between neutrophil-mediated inflammation and lymphocyte-driven immunity is reflected in NLR.

We used the Glasgow Prognostic Score (GPS) as well as the prognostic nutritional index (PNI) to evaluate the patient’s overall nutritional and inflammatory status. We calculated the PNI as follows: “PNI = serum albumin level (g/L) + 5 × lymphocyte count (10⁹ cells/L)”. A higher PNI value indicates better nutritional and immune status and contributes to a better prognosis, while lower values suggest a higher risk of complications and poor outcomes (27). The serum albumin and CRP levels were used to determine the GPS. The GPS score ranges from 0 (normal) to 2 (poor). A GPS of 2 indicates elevated CRP levels (>10 mg/L) along with hypoalbuminemia (serum albumin <35 g/L), a GPS of 1 indicates either an elevated CRP or low albumin, and a GPS of 0 indicates normal CRP and albumin levels (28) (Table 1).

Surgical procedure

Distal pancreatectomy (DP), total pancreatectomy (TP), or pancreatoduodenectomy (PD) were performed on all patients. The PD resection area included the distal stomach, pancreas head, duodenum, upper jejunum, common bile duct, and gallbladder; we removed lymph nodes in the corresponding areas. After resection, the pancreas, bile duct, and stomach were anastomosed to the jejunum to reconstruct the digestive tract. The lymph nodes in the corresponding areas were also removed along with the left anterior renal fat capsule. TP involved the removal of the entire pancreas, gallbladder, duodenum, distal stomach, upper jejunum, common bile duct, and spleen. For both DP and TP resections, the lymph nodes in the corresponding areas were dissected along with the left anterior renal fat capsule. The surgical approach was recorded. The volume of operative bleeding and tumor location were also noted. The pathological tumor-node-metastasis (TNM) stage was determined according to the 8th American Joint Committee on Cancer (AJCC) staging system (29).

Post-operative assessment

We recorded the stay in hospital and post-operative complication incidence, such as non-clinically relevant postoperative pancreatic fistula (CR-POPF) and CR-POPF onset. The Clavien-Dindo complications score was also calculated (30). Complications following surgery are ranked by the Clavien-Dindo Classification according to whether they require intensive, medical, or surgical care. The Clavien-Dindo Classification categorizes surgical complications following pancreatic tumor resection as Grade I (minor, no intervention), Grade II (requires medication or blood transfusion), Grade III (requires radiological, endoscopic, or surgical intervention), Grade IV (lethal, demanding intensive care), and Grade V (death).

Follow-up in addition to review

The patients were followed up with every 2 months after discharge. Regular chemotherapy was given according to the recovery situation. During the follow-up appointments, we obtained serum blood samples to evaluate the blood count, liver function, kidney function, and tumor inflammatory indices. In addition, an abdominal-enhanced CT or MRI was also acquired to assess disease progression. If patients were unable to complete regular follow-up visits, they were contacted by telephone to determine their survival status.

Statistical analysis

The EXCEL program was used for collecting all clinical data. In light of the occurrence of severe (grade 3 or above) Clavien-Dindo complications, CR-POPF problems, and anatomical resection status, we divided the patients into two groups. Continuous variables were compared between these groups using the Mann-Whitney U test. Each patient was staged according to the TNM classification based on the post-operative pathological diagnosis. Continuous variables were compared between diverse staging groups using the Kruskal-Wallis test. For pairwise comparisons, we used the Dunn’s test as a post-hoc assessment. Version 25.0 of the Statistical Package for Social Sciences program was used to analyze the data.

The survival analysis was performed using R software, version 3.5.0. We used the “pROC” program to plot the receiver operating characteristic (ROC) curve for survival at 1, 3, and 5 years. We used the “timeROC” package to plot the time-dependent ROC curves, whereas we used the “pec::cindex” package to plot the time-dependent index curves. We performed the long-rank test and mapped Kaplan–Meier survival curves for a comparison of OS between diverse indices and staging groups. We used the Youden’s index to fix the ideal thresholds for risk scores as well as risk factors. We randomly separated data into validation and training sets in a 7:3 ratio. We used the training set data to develop the predictive nomogram and the validation set data to perform validation. To find isolated risk factors for overall survival (OS), we performed multivariate and univariate Cox analyses using the “survival” package. We performed the least absolute shrinkage and selection operator (LASSO) regression analysis using the “glmnet” package. In LASSO, we determined the optimal regularization parameter (λ) through cross-validation to minimize model error and prevent overfitting. Founded on the optimal λ, we used LASSO regression to classify variables significantly related to survival. The variables obtained from the LASSO regression were compared with multivariate analysis variables. Based on the findings from screening preoperative survival indicators in the training group using LASSO and multivariate analysis, we created a predictive nomogram. We plotted the nomogram using the R software’s “rms” package. ROCs were plotted for the training and validation groups. We used the area under the ROC curve (AUC) to evaluate the nomogram’s predictive capability. The time-dependent concordance index (C-index) for our nomogram was compared with other clinically known risk factors. A P value below 0.05 was deemed significant for all statistical tests.

Results

Post-operative survival

Patients in stage I (n=148) had a 27% 5-year survival rate and over 2 years of median survival time. Patients in stage II (n=163) had a 15% 5-year survival rate and a roughly 1.5 year median survival period. The median survival durations for patients in stage III (n=44) and IV (n=13) were 0.9 and 1.1 years, respectively. None of the stage III and IV patients survived 5 years (Figure 1).

Figure 1 Post-operative survival curves for stages I (red), II (green), III (blue), and IV (purple). The dotted line indicates the median survival for each stage.

Differences in inflammatory indicators between subgroups

A total of 81 patients developed CR-POPF, and 287 developed non-CR-POPF. Compared with patients who had no signs of CR-POPF, those who did exhibited higher white blood cell (WBC) levels (6.1×10⁹/L vs. 5.7×10⁹/L, P=0.01), monocyte count (0.51×10⁹/L vs. 0.40×10⁹/L, P=0.002), neutrophil (NE) levels (3.9×10⁹/L vs. 3.6×10⁹/L, P=0.03), IBI (18.8 vs. 9.3, P=0.002), and CRP (6.51 vs. 3.70 mg/L, P=0.001). Compared with patients who failed to develop CR-POPF, those who did tended to have higher PLR (207.0 vs. 144.4, P=0.79), NLR (2.8 vs. 2.6, P=0.25), MLR (0.33 vs. 0.30, P=0.07), and SII (659.5 vs. 502.9, P=0.08). Moreover, patients who developed CR-POPF were inclined to stay in the hospital for a longer time (25 vs. 18 days, P=0.054). Nevertheless, these differences are insignificant (Table S1).

Two groups of patients were created in accordance with the findings of the intraoperative exploration: anatomical BR (n=126) and anatomical R (n=242). Compared with the anatomical R group, the anatomical BR group showed higher length of hospital stay (21 vs. 19 days, P=0.004), hemoglobin (HGB) (121.0 vs. 126.0 g/L, P=0.01), CRP (6.32 vs. 3.11 mg/L, P<0.001), tumor size (4 vs. 3 cm, P<0.001), neutrophil (NE) (3.90×10⁹/L vs. 3.65×10⁹/L, P=0.048), NLR (2.855 vs. 2.61, P=0.03), IBI (17.75 vs. 7.95, P<0.001), MLR (0.35 vs. 0.30, P=0.005), cancer antigen 199 (CA19-9) (345.6 vs. 102.5 IU/mL, P=0.001). The anatomical BR group had higher levels of SII (589.35 vs. 502.93, P=0.053) and WBC (5.87×10⁹/L vs. 5.77×10⁹/L, P=0.09), though the differences were insignificant (Table 2).

In all, 345 patients had a low Clavien-Dindo score of (2 or less) and 23 had a high score (3 to 5). The patients with a high Clavien-Dindo score had significantly higher CRP (12.05 vs. 3.78 mg/L, P=0.001), IBI (32.12 vs. 9.87, P=0.001), total bilirubin (TBIL) (69.5 vs. 23.0 µmol/L, P=0.04) than those with a low score. Additionally, individuals with a high Clavien-Dindo score had a longer hospital stay (39 vs. 19 days, P<0.001). Furthermore, they reported a comparatively higher SII (568.84 vs. 529.25, P=0.792), tumor size (3.7 vs. 3 cm, P=0.17), NE (3.90×10⁹/L vs. 3.71×10⁹/L, P=0.37), NLR (3.25 vs. 2.64, P=0.11), MLR (0.34 vs. 0.31, P=0.06), CA19-9 (232.76 vs. 165.74 IU/mL, P=0.20), and WBC (6.07×10⁹/L vs. 5.80×10⁹/L, P=0.09), however the difference in these inflammatory indices was not statistically significant (Table S2).

Overall these results indicate that the pre-operative CRP and IBI could predict CR-POPF and high Clavien-Dindo score. Moreover, these markers could also be used to predict local and vasculature tumor invasion.

Differences in the continuous variables according to the PDAC clinical stage

The Kruskal-Wallis test showed that the median IBI [χ²(3)=12.448, P=0.006] and pre-operative CRP levels [χ²(3)=12.772, P=0.005] differed significantly between patients with different clinical stages. Furthermore, a CRP difference was found between stages II and III (P<0.001) and between stages I and III (P=0.002), according to Dunn’s post-hoc analysis. Nevertheless, stage II and the other subgroups did not differ significantly (P>0.05) (Figure S1A). The group’s median IBI also differed significantly between stages II and III (P<0.001) and between stages I and III (P<0.001), according to Dunn’s post-hoc analysis. The differences between the other groups were insignificant (P>0.05) (Figure S1B). These results suggest that the pre-operative CRP and IBI can effectively discriminate between stages I, II, and III and could also be used to predict prognosis and hence guide treatment interventions.

Differences in survival between subgroups

The optimal pre-operative cut-off values predictive of survival, based on Youden’s index, were a CRP of 8.13 mg/L, an IBI of 12.262, an SII of 263.438, an NLR of 2.530, an MLR of 0.164, a prealbumin (PAB) of 186.0 mg/L, a PNI of 45.3, and a maximum tumor diameter of 2.4 cm (Figure 2). According to the Kaplan-Meier curve analysis, all indices above the optimal threshold except for SII had a significant impact on survival. Higher PNI values were associated with improved OS. All other inflammatory indices and tumor size with values above their respective cut-offs were associated with a worse prognosis. Despite an insignificant difference, those with a higher SII tended to have a worse survival.

Figure 2 Survival curves stratified according to threshold clinical values set using the Youden’s index for CRP (A), IBI (B), SII (C), NLR (D), MLR (E), tumor diameter (F), PAB (G), and PNI (H). The dotted line indicates median survival, which differed significantly between the high and low optimal thresholds except for SII. CRP, C-reactive protein; IBI, inflammation-based index; MLR, monocyte-to-lymphocyte ratio; NLR, neutrophil-to-lymphocyte ratio; PAB, prealbumin; PNI, prognostic nutritional index; SII, systemic immune-inflammation index.

Construction of survival-related predictive models and validation

The univariate Cox regression of all available pre-operative clinical characteristics revealed thirteen factors predictive of survival including; CRP level, radiological R, CA19-9 levels, tumor size, GPS score, IBI, PAB, NLR, indirect bilirubin (IBIL), age, TBIL, SII, and a caudal tumor position (Table S3). Four final predictive variables, namely CRP (P=0.001), radiological R (P=0.005), CA19-9 level (P=0.009), and tumor size (P<0.001), were identified from the multivariate Cox analysis (Table 3). The LASSO regression model identified four variables significantly associated with survival (Figure S2A,S2B). All of the variables highlighted by the LASSO analysis remained statistically significant in the multivariate Cox regression, thus confirming their importance in predicting survival.

Nomogram’s predictive ability

Prediction nomograms for survival rates for 1, 3, and 5 years were constructed using the four risk factors found in the LASSO and multivariate analyses (Figure S2C). For the training set, AUC for survival rates for 1, 3, and 5 years were 0.69, 0.77, and 0.70, respectively (Figure S3A); for the validation set, they were 0.67, 0.67, and 0.72, respectively (Figure S3B). In both sets, the nomogram’s AUC was greater than the AUC of each of the other risk factors (sex, age, TNM stage, and CA19-9) in both training and validation datasets (Figure S3C,S3D). Similarly, the time-dependent c-index showed that compared with all four other risk factors, our nomogram achieved a relatively better performance in predicting post-operative survival in PDAC patients (Figure S3E,S3F).

Discussion

A multidisciplinary approach combining surgery and systemic chemotherapy offers the possibility of curing BR PDAC (31). The number of persons with borderline PDAC receiving surgery is rising as a result of the development of vein reconstruction approaches (9). However, it remains unclear which patients are more likely to have good long-term outcomes following early surgery. Resection of the PDAC can reduce the endocrine and exocrine pancreatic functions. In addition, some cases may require complex reconstructions of the digestive tract. As a result, surgical resection often leads to several complications, including severe diarrhea, weight loss, cholangitis, pancreaticojejunostomy stenosis, and blood glucose abnormalities post-operatively (32-35). The patient’s quality of life could be significantly impacted by these issues. Moreover, many patients are found to be inoperable during the surgical exploration due to the presence of non-reconstructable vascular invasion. Exploratory surgery may cause unnecessary trauma and delay the initiation of chemotherapy. In addition, some patients may not benefit from early surgery due to the recurrence and metastasis of the disease shortly after the procedure (13). Hence, researchers should recognize clinical risk factors that could be used to predict treatment outcomes after surgery and, hence, improve patient selection for surgical intervention and enhance OS rates by identifying those most likely to benefit from resection.

Tumor growth and metastasis are significantly influenced by the systemic inflammatory response (19,36). Fluctuations in inflammatory proteins, such as lymphocytes, CRP, and neutrophils, as well as peripheral blood cells, can be signs of systemic inflammation. Inflammatory indicators reflect variations in a variety of tumor microenvironments but also provide important predictive information for complications after surgery, recurrence of tumors, and life span (37-39). Elevated pre-operative IBI contributed to countless complications along with anatomical BR. Similarly, a CRP of 8.13 mg/L or higher yielded a 200-day median survival decrease. Thus, similar to traditional tumor markers, namely CA19-9, these inflammatory indices may indicate potential spread and more aggressive tumor behavior. Even if the PDAC is anatomically resectable, patients with elevated inflammatory markers are more susceptible to early relapse and metastasis after surgery. Therefore, NAT may enhance tumor resectability by reducing the need for partial revascularization, potentially shortening the perioperative stay in the hospital and improving lasting survival outcomes (40,41). The ability of the inflammatory indices to stratify BR patients suggests a shift in therapeutic decision-making for PDAC patients with high inflammatory factors.

Our nomogram based on pre-operative risk variables derived from the LASSO and Cox regression confirmed that inflammatory indices could be used to characterize the tumor and predict prognosis in PDAC patients. Similarly to previous studies, a large tumor diameter and elevated CA19-9 levels indicated poor prognosis and relapse (13,42). Furthermore, vascular invasion strongly predicted survival and PDAC relapse (43-47). Our findings indicate that the determination of BR based solely on pre-operative imaging may not reflect the actual surgical tumor resectability and survival. However, our nomogram based on radiological information inflammatory indices, tumor size, and biological resectability significantly improved the prediction of post-operative patient survival. In addition, the model achieved better time-dependent differentiation and even outperformed conventional pathological staging in our center’s internal validation data (AUC of 0.77 at 3 years vs. 0.58 for TNM staging).

We must be mindful of the various limitations of this research. The generalizability of our nomogram may be limited by the retrospective single-center approach, the absence of external dataset validation, and the small number of participants. Missing data may have also led to selection bias. Therefore, we recommend larger multicentre research to evaluate the model’s generalisability and relevance in different populations. Given the lack of prospective validation and decision curve analysis, these findings should be considered hypothesis-generating. Further prospective studies are warranted to validate its clinical utility and generalizability.

Conclusions

Inflammation indices were associated with postoperative complications. Compared to traditional indicators, our nomogram, based on elevated inflammatory indices combined with other clinical indications, demonstrated potential in predicting postoperative survival in patients with PDAC. Patients with PDAC who stand the best chance of benefiting after surgery can be recognized using our nomogram.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0008/rc

Data Sharing Statement: Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0008/dss

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0008/prf

Funding: This research was funded by The Standardized Residency Training Program at The First Affiliated Hospital of Soochow University (No. ZPKT20250111).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0008/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and approved by the Ethics Committee of The First Affiliated Hospital of Soochow University (No. 2025252). Patient consent was waived due to the retrospective nature of the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin 2020;70:7-30. [Crossref] [PubMed]
2. Tingstedt B, Andersson B, Jönsson C, et al. First results from the Swedish National Pancreatic and Periampullary Cancer Registry. HPB (Oxford) 2019;21:34-42. [Crossref] [PubMed]
3. Sugawara T, Ban D, Nishino J, et al. Prediction of early recurrence of pancreatic ductal adenocarcinoma after resection. PLoS One 2021;16:e0249885. [Crossref] [PubMed]
4. Neoptolemos JP, Kleeff J, Michl P, et al. Therapeutic developments in pancreatic cancer: current and future perspectives. Nat Rev Gastroenterol Hepatol 2018;15:333-48. [Crossref] [PubMed]
5. Raptis DA, Sánchez-Velázquez P, Machairas N, et al. Defining Benchmark Outcomes for Pancreatoduodenectomy With Portomesenteric Venous Resection. Ann Surg 2020;272:731-7. [Crossref] [PubMed]
6. Al Faraï A, Garnier J, Ewald J, et al. International Study Group of Pancreatic Surgery type 3 and 4 venous resections in patients with pancreatic adenocarcinoma:the Paoli-Calmettes Institute experience. Eur J Surg Oncol 2019;45:1912-8. [Crossref] [PubMed]
7. Gemenetzis G, Groot VP, Blair AB, et al. Survival in Locally Advanced Pancreatic Cancer After Neoadjuvant Therapy and Surgical Resection. Ann Surg 2019;270:340-7. [Crossref] [PubMed]
8. Navez J, Bouchart C, Lorenzo D, et al. What Should Guide the Performance of Venous Resection During Pancreaticoduodenectomy for Pancreatic Ductal Adenocarcinoma with Venous Contact? Ann Surg Oncol 2021;28:6211-22. [Crossref] [PubMed]
9. Groen JV, Stommel MWJ, Sarasqueta AF, et al. Surgical management and pathological assessment of pancreatoduodenectomy with venous resection: an international survey among surgeons and pathologists. HPB (Oxford) 2021;23:80-9. [Crossref] [PubMed]
10. Cloyd JM, Heh V, Pawlik TM, et al. Neoadjuvant Therapy for Resectable and Borderline Resectable Pancreatic Cancer: A Meta-Analysis of Randomized Controlled Trials. J Clin Med 2020;9:1129. [Crossref] [PubMed]
11. Rangelova E, Wefer A, Persson S, et al. Surgery Improves Survival After Neoadjuvant Therapy for Borderline and Locally Advanced Pancreatic Cancer: A Single Institution Experience. Ann Surg 2021;273:579-86. [Crossref] [PubMed]
12. Inoue Y, Saiura A, Oba A, et al. Neoadjuvant gemcitabine and nab-paclitaxel for borderline resectable pancreatic cancers: Intention-to-treat analysis compared with upfront surgery. J Hepatobiliary Pancreat Sci 2021;28:143-55. [Crossref] [PubMed]
13. Ushida Y, Inoue Y, Ito H, et al. High CA19-9 level in resectable pancreatic cancer is a potential indication of neoadjuvant treatment. Pancreatology 2021;21:130-7. [Crossref] [PubMed]
14. Jachnis A, Słodkowski MT. The Relationship between Nutritional Status and Body Composition with Clinical Parameters, Tumor Stage, CA19-9, CEA Levels in Patients with Pancreatic and Periampullary Tumors. Curr Oncol 2021;28:4805-20. [Crossref] [PubMed]
15. Taniai T, Haruki K, Furukawa K, et al. The novel index using preoperative C-reactive protein and neutrophil-to-lymphocyte ratio predicts poor prognosis in patients with pancreatic cancer. Int J Clin Oncol 2021;26:1922-8. [Crossref] [PubMed]
16. Sakamoto T, Sunaguchi T, Goto K, et al. Modified geriatric nutritional risk index in patients with pancreatic cancer: a propensity score-matched analysis. BMC Cancer 2022;22:974. [Crossref] [PubMed]
17. Lilly AC, Astsaturov I, Golemis EA. Intrapancreatic fat, pancreatitis, and pancreatic cancer. Cell Mol Life Sci 2023;80:206. [Crossref] [PubMed]
18. Padoan A, Plebani M, Basso D. Inflammation and Pancreatic Cancer: Focus on Metabolism, Cytokines, and Immunity. Int J Mol Sci 2019;20:676. [Crossref] [PubMed]
19. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860-7. [Crossref] [PubMed]
20. Proctor MJ, Morrison DS, Talwar D, et al. An inflammation-based prognostic score (mGPS) predicts cancer survival independent of tumour site: a Glasgow Inflammation Outcome Study. Br J Cancer 2011;104:726-34. [Crossref] [PubMed]
21. Pierce BL, Ballard-Barbash R, Bernstein L, et al. Elevated biomarkers of inflammation are associated with reduced survival among breast cancer patients. J Clin Oncol 2009;27:3437-44. [Crossref] [PubMed]
22. Gukovsky I, Li N, Todoric J, et al. Inflammation, autophagy, and obesity: common features in the pathogenesis of pancreatitis and pancreatic cancer. Gastroenterology 2013;144:1199-209.e4. [Crossref] [PubMed]
23. Das A, Bararia A, Mukherjee S, et al. Chronic pancreatitis as a driving factor for pancreatic cancer: An epidemiological understanding. World J Clin Oncol 2024;15:1459-62. [Crossref] [PubMed]
24. Alhobayb T, Peravali R, Ashkar M. The Relationship between Acute and Chronic Pancreatitis with Pancreatic Adenocarcinoma Diseases 2021;9:93. Review. [Crossref] [PubMed]
25. Michalak N, Małecka-Wojciesko E. Modifiable Pancreatic Ductal Adenocarcinoma (PDAC) Risk Factors. J Clin Med 2023;12:4318. [Crossref] [PubMed]
26. Bockhorn M, Uzunoglu FG, Adham M, et al. Borderline resectable pancreatic cancer: a consensus statement by the International Study Group of Pancreatic Surgery (ISGPS). Surgery 2014;155:977-88. [Crossref] [PubMed]
27. Yan L, Nakamura T, Casadei-Gardini A, et al. Long-term and short-term prognostic value of the prognostic nutritional index in cancer: a narrative review. Ann Transl Med 2021;9:1630. [Crossref] [PubMed]
28. Forrest LM, McMillan DC, McArdle CS, et al. Evaluation of cumulative prognostic scores based on the systemic inflammatory response in patients with inoperable non-small-cell lung cancer. Br J Cancer 2003;89:1028-30. [Crossref] [PubMed]
29. Nagaria TS, Wang H. Modification of the 8(th) AJCC staging system of pancreatic ductal adenocarcinoma. Hepatobiliary Surg Nutr 2020;9:95-7. [Crossref] [PubMed]
30. Clavien PA, Sanabria JR, Strasberg SM. Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery 1992;111:518-26.
31. Kwaśniewska D, Fudalej M, Nurzyński P, et al. How A Patient with Resectable or Borderline Resectable Pancreatic Cancer should Be Treated-A Comprehensive Review. Cancers (Basel) 2023;15:4275. [Crossref] [PubMed]
32. Uijterwijk BA, Moekotte A, Boggi U, et al. Oncological resection and perioperative outcomes of robotic, laparoscopic and open pancreatoduodenectomy for ampullary adenocarcinoma: a propensity score matched international multicenter cohort study. HPB (Oxford) 2025;27:318-29. [Crossref] [PubMed]
33. Jiménez-Romero C, de Juan Lerma A, Marcacuzco Quinto A, et al. Risk factors for delayed gastric emptying after pancreatoduodenectomy: a 10-year retrospective study. Ann Med 2025;57:2453076. [Crossref] [PubMed]
34. Montorsi RM, Strijbos BTM, Stommel MWJ, et al. Preventing and Treating Delayed Gastric Emptying After Pancreatic Surgery: A Systematic Review and Meta-analysis of Randomized Controlled Trials. Ann Surg 2025;282:954-62. [Crossref] [PubMed]
35. Wismans LV, Hendriks TE, Suurmeijer JA, et al. Preoperative stereotactic radiotherapy to prevent pancreatic fistula in high-risk patients undergoing pancreatoduodenectomy (FIBROPANC): prospective multicentre phase II single-arm trial. Br J Surg 2025;112:znae327. [Crossref] [PubMed]
36. Greten FR, Grivennikov SI. Inflammation and Cancer: Triggers, Mechanisms, and Consequences. Immunity 2019;51:27-41. [Crossref] [PubMed]
37. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646-74. [Crossref] [PubMed]
38. Valero C, Lee M, Hoen D, et al. Pretreatment neutrophil-to-lymphocyte ratio and mutational burden as biomarkers of tumor response to immune checkpoint inhibitors. Nat Commun 2021;12:729. [Crossref] [PubMed]
39. Wang D, Wang Y, Dong X, et al. The significance of preoperative neutrophil-to-lymphocyte ratio in predicting short-term complications and survival benefits of pancreaticoduodenectomy: A systematic review and meta-analysis. Am J Surg 2024;229:76-82. [Crossref] [PubMed]
40. Chae YS, Jung HS, Yun WG, et al. Clinical outcomes of preservation versus resection of portal/superior mesenteric vein during pancreaticoduodenectomy in pancreatic cancer patients who respond to neoadjuvant treatment: a retrospective cohort study. Int J Surg 2024;110:7150-8. [Crossref] [PubMed]
41. Li T, D'Cruz RT, Lim SY, et al. Somatostatin analogues and the risk of post-operative pancreatic fistulas after pancreatic resection - A systematic review & meta-analysis. Pancreatology 2020;20:158-68. [Crossref] [PubMed]
42. Secchettin E, Paiella S, Azzolina D, et al. Expert Judgment Supporting a Bayesian Network to Model the Survival of Pancreatic Cancer Patients. Cancers (Basel) 2025;17:301. [Crossref] [PubMed]
43. Stoop TF, Molnár A, Seelen LWF, et al. Tangential Versus Segmental Portomesenteric Venous Resection During Pancreatoduodenectomy for Pancreatic Cancer: An International Multicenter Cohort Study on Surgical and Oncological Outcome. Ann Surg 2025; Epub ahead of print. [Crossref]
44. Murakami Y, Satoi S, Motoi F, et al. Portal or superior mesenteric vein resection in pancreatoduodenectomy for pancreatic head carcinoma. Br J Surg 2015;102:837-46. [Crossref] [PubMed]
45. Han S, Choi DW, Choi SH, et al. Long-term outcomes following en bloc resection for pancreatic ductal adenocarcinoma of the head with portomesenteric venous invasion. Asian J Surg 2021;44:313-20. [Crossref] [PubMed]
46. Labib PLZ, Russell TB, Denson JL, et al. Patterns, timing and predictors of recurrence following pancreaticoduodenectomy for pancreatic ductal adenocarcinoma: an international multicentre retrospective cohort study. HPB (Oxford) 2025;27:445-60. [Crossref] [PubMed]
47. Zou Y, Liang Y, Ma X, et al. A Preoperative Model for Predicting Lymphovascular Invasion in Pancreatic Ductal Adenocarcinoma. J Surg Res 2025;313:488-99. [Crossref] [PubMed]