Abstract
Recent studies have reported that inflammatory markers, such as neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and advanced lung cancer inflammation index, are associated with invasiveness of intraductal papillary mucinous neoplasm (IPMN). This study aimed to develop and validate a new nomogram that includes inflammatory markers for predicting the invasiveness of IPMN.
The data of 365 patients who underwent surgical resection for IPMN at 4 centers between 1995 and 2016 were retrospectively reviewed to develop a new nomogram. For external validation, a separate patient cohort was used. The predictive ability of the nomogram was evaluated using the area under the receiver operating characteristic curve.
The new nomogram was developed using the following variables which were identified as risk factors for invasive IPMN: body mass index, preoperative serum bilirubin level, carbohydrate antigen 19-9, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, advanced lung cancer inflammation index, main duct type, presence of solid portion, and tumor size. After external validation, the area under the curve value was 0.649 (95% CI: 0.578-0.720, P < .001).
To the best of our knowledge, this study is the first to predict and externally validate the invasiveness in IPMN using inflammatory markers. Further research is necessary to improve predictability of the model for selecting patients for surgical resection.
Keywords: inflammatory markers, intraductal papillary mucinous neoplasm, malignancy risk, nomogram, pancreatic cancer
1. Introduction
Intraductal papillary mucinous neoplasms (IPMNs) of the pancreas are cystic neoplasms that have a broad histological spectrum ranging from benign to malignant disease. The incidence has been increasing over time as advanced imaging tests facilitate the recognition of pancreatic cystic lesions.[1] From histopathological point of view, IPMN is often mentioned together with mucinous cystic neoplasm, since both have malignant potential. IPMNs can be distinguished from mucinous cystic neoplasms by the presence or absence of communication with the pancreatic duct.[2,3]
Prediction of malignant IPMN based on radiologic features has become feasible using magnetic resonance cholangiopancreatography scan or endoscopic ultrasound guided biopsy. According to the recently revised consensus guidelines of the International Association of Pancreatology (IAP),[4] “high-risk stigmata,” which suggests malignancy of branch duct intraductal papillary mucinous neoplasm (BD-IPMN), includes measurement of the enhanced mural nodule and the size of the main pancreatic duct in the imaging tests. Several studies have attempted to invent nomograms for risk analysis including the imaging features, and reported variable predictive values.[5-8]
However, in addition to image findings that directly reflect the characteristics of the lesions, using other factors that are associated with malignant IPMN would be useful to develop a more accurate quantitative scoring model. Previous studies have shown that inflammatory markers, such as neutrophil-tolymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) are associated with poor survival rates in patients with pancreatic cancer and other malignancies.[9-11] These markers were identified as useful predictors of malignant potential in IPMN.[12-15]
The purpose of this study is to develop a newly expanded nomogram by including the inflammatory markers, that quantifies the risk of malignancy in IPMN patients. We also validated the predictability of the new model using external data.
2. Methods
2.1. Data collection
The cohort for developing a nomogram included patients who were diagnosed with IPMN based on pathologic examination after surgical resection, between April 1995 and December 2016 at 4 different centers: Samsung Medical Center (SMC), Seoul National University Boramae Medical Center (BRM), Ilsan Paik Hospital (IPH), and Dongguk University Ilsan Hospital (DUH). The pathologic diagnosis and review of IPMN was done by each hospital’s pathologists according to the 4th edition of WHO classification.[16]
Among 468 patients, 103 patients were excluded for the following reasons: history of pancreatitis, history of cholangitis or other malignancy, or absence of data regarding carbohydrate antigen 19-9 (CA19-9). Finally, a total of 365 patients were included in the analyses. This study was approved by the Institutional Review Board of Samsung Medical Center (Seoul, Korea, the approval number: 2017-07-016).
2.2. Clinical variables for analysis
The patients’ demographics, preoperative laboratory results, and image findings were retrospectively reviewed. Blood laboratory tests were performed at least once within a month before surgery, and the results of the most recent examinations were used in the analysis. NLR was calculated as the total count of neutrophils divided by the total count of lymphocytes and PLR as the total count of platelets divided by the total count of lymphocytes. Advanced lung cancer inflammation index (ALI) was calculated as follows: body mass index (BMI) × serum Albumin/NLR. The values that were required for the calculations were measured before the operations. The radiological assessments including computed tomography, magnetic resonance imaging, endoscopic ultrasound were conducted preoperatively with a focus on the type and size of IPMN, as well as the presence of solid portion and the size of the main pancreatic duct. The sizes of tumor and pancreatic duct were defined as the largest diameter of the main mass and the main pancreatic duct, respectively. In case of patients with >2 preoperative images by different modalities, the largest figures were adopted. The type of IPMN was classified as branch duct type, main duct type, or mixed type in accordance with the 2017 revised Fukuoka guidelines.[4] After surgical resection, pathological analysis and histological grading were performed according to the criteria of the WHO classification system for digestive neoplasms. This classification system includes low or intermediate dysplasia, high-grade dysplasia, and IPMN with invasive cancer.[16]
2.3. Development of nomogram
In our previous study,[15] a comparative analysis was performed between the patients with non-invasive and invasive IPMN in the development cohort. After the risk factor analysis for invasive IPMN, we selected variables for a new nomogram.
A nomogram was built after fitting in the regression model. First, we used the distribution summaries of the included variables with plotting ranges and values to be adjusted. Using the linear predictor method, the factor with the greatest impact was determined and the other factors were assigned in consecutive order. All the predictors with assigned values were mapped in the “Point” axis. To estimate the predictability of the new nomogram, internal validation was conducted using bootstrap resamples.
2.4. External validation
For external validation, we extracted another patient cohort from Seoul National University Hospital (SNUH), which included 384 pathologically-proven IPMN patients after surgical resection between May 1995 and December 2016. The pathology reports were reviewed according to the 4th edition of WHO classification.[14] The same exclusion criteria used for the development set were applied to the external validation set. The new receiver operating characteristic curve was drawn to verify the predictability of the model.
The development of the nomogram and its internal and external validation were conducted using R 3.4.3 (Vienna, Austria).
3. Results
The demographic and clinical data from the development set (n = 365) are summarized in Table 1. The median age of the patients was 63 years and 66.3% were men. The median values of NLR, PLR, and ALI were 1.8, 109.5, and 56.3, respectively. The most common type of IPMN was BD-IPMN (50.7%). The solid portion in the preoperative imaging was found in 89 (24.4%) patients. The median size of tumor and duct was 3.1cm and 4.7mm, respectively. Among these patients, 98 (26.8%) patients demonstrated the presence of invasive IPMN in the surgical pathology.
Table 1 .
Demographic and clinical data of the development set (n = 365).
|
Variables |
N (%) mean or median (IQR) |
|
|---|---|---|
|
Age, median, yrs |
63 (57-69) |
|
|
Sex, male |
242 (66.3) |
|
|
BMI, median, kg/m2 |
23.8 (22.1-25.5) |
|
|
Maximum pre-op serum bilirubin, mean, mg/dL |
0.7 (0.5-1.2) |
|
|
CA 19-9, median, U/mL |
12.7 (6.4-24.7) |
|
|
NLR, median |
1.8 (1.3-2.5) |
|
|
PLR, median |
109.5 (85.7-152.4) |
|
|
ALI, median |
56.3 (36.8-80.6) |
|
|
Type of IPMN, n (%) | ||
|
Main duct type |
76 (20.8) |
|
|
Branch duct type |
185 (50.7) |
|
|
Mixed type |
104 (28.5) |
|
|
Tumor size, mean, cm |
3.1 (2.4-4.2) |
|
|
Pancreatic duct size, mean, mm |
4.7 (3.0-7.0) |
|
|
Solid portion†, n (%) |
89 (24.4) |
|
|
Operation, n (%) | ||
|
Pancreaticoduodenectomy |
184 (50.4) |
|
|
Distal pancreatectomy |
120 (32.9) |
|
|
Central pancreatectomy |
13 (3.6) |
|
|
Others |
48 (13.1) |
|
|
Pathology, n (%) | ||
|
Non-invasive* |
267 (73.2) |
|
|
Invasive |
98 (26.8) |
|
|
ALI = advanced lung cancer inflammation index, BMI = body mass index, CA 19-9 = carbohydrate antigen 19-9, IPMN = intraductal papillary mucinous neoplasm, IQR = interquartile range, NLR = neutrophil-to-lymphocyte ratio, PLR = platelet-to-lymphocyte ratio, Pre-op = preoperative. | ||
* Low, intermediate or high-grade dysplasia.
† Solid portion, mural nodule, or enhanced cyst.
Based on our previous study, the following variables which were identified as significant were included in developing the nomogram[13]: NLR, PLR, ALI, maximum preoperative serum bilirubin level, the solid portion of the tumor, BMI, size of the tumor, carbohydrate antigen 19-9, and type of IPMN. A draft of the new nomogram is shown in Fig. 1. To obtain a total point, points from the point axis for each predictor were added together. The individual risk for invasive IPMN can be obtained using the axis of “predicted value.”
Figure 1. The new nomogram predicting risk the of malignancy in IPMN. IPMN = intraductal papillary mucinous neoplasm.
To internally validate the predictive performance of the model, resampling with a thousand bootstrap samples was done (Fig. 2). In the figure, the x-axis is the predicted probability and the y-axis is the actual probability of invasive IPMN. The dashed line indicates the relationship between the actual and predicted probabilities when prediction is perfect. The dotted line was drawn using the new nomogram and very close to the ideal line, with the mean absolute error of 0.025, which means good calibration. Based on the results, we created a web calculator that can be easily used to predict the malignancy risk of IPMN by entering clinical data (http://ipmn.smchbp.org/) (Fig. 3).
Figure 2. Calibration plot of internal validation through 1000 bootstrap resamples.
Figure 3. The website for calculating malignancy risk of IPMN. IPMN = intraductal papillary mucinous neoplasm.
Table 2 summarizes the demographic and clinical features of the external validation set (n = 384). The median values of NLR, PLR, and ALI were 1.8, 114.5, and 54.3, respectively. The mean tumor size was 3.5cm and the mean diameter of the pancreatic duct was 6.0 mm. Among the patients in the validation cohort, BD-IPMN was identified as the most common type (n = 185, 55.7%), and the solid portion was found in 99 (25.8%) patients. Based on the pathology reports, 102 (25.3%) patients were identified to have invasive IPMN. After excluding the patients with missing variables, a total of 308 patients were included in external validation. The receiver operating characteristic curve is shown in Fig. 4, and the area under the curve (AUC) of the nomogram was identified as 0.649 (95% CI: 0.578-0.720, P < .001).
Table 2 .
Demographic and clinical data of the external validation set (n = 384).
|
Variables |
N (%) mean or median (IQR) |
|
|---|---|---|
|
Age, median, yrs |
65 (59-70) |
|
|
Sex, male 219 |
(57.0) |
|
|
BMI, median, kg/m2 |
23.2 (20.9-25.1) |
|
|
Maximum pre-op serum bilirubin, mean, mg/dL |
1.0 (0.6-1.1) |
|
|
CA 19-9, median, U/mL |
10.1 (5.0-24.2) |
|
|
NLR, median |
1.8 (1.3-2.3) |
|
|
PLR, median |
114.5 (89.4-149.1) |
|
|
ALI, median |
54.3 (36.0-74.9.6) |
|
|
Type of IPMN, n (%) | ||
|
Main duct type |
38 (9.9) |
|
|
Branch duct type |
214 (55.7) |
|
|
Mixed type 132 (34.4) |
||
|
Tumor size, mean, cm |
3.5 (2.1-4.0) |
|
|
Pancreatic duct size, mean, mm |
6.0 (2.3-8.0) |
|
|
Solid portion†, n (%) |
99 (25.8) |
|
|
Operation, n (%) | ||
|
Pancreaticoduodenectomy |
192 (50.0) |
|
|
Distal pancreatectomy |
122 (31.8) |
|
|
Central pancreatectomy |
17 (4.4) |
|
|
Others |
53 (13.8) |
|
|
Pathology, n (%) | ||
|
Non-invasive* |
282 (74.7) |
|
|
Invasive |
102 (25.3) |
|
|
ALI = advanced lung cancer inflammation index, BMI = body mass index, CA 19-9 = carbohydrate antigen 19-9, IPMN = intraductal papillary mucinous neoplasm, IQR = interquartile range, NLR = neutrophil-to-lymphocyte ratio, PLR = platelet-to-lymphocyte ratio, Pre-op = preoperative. | ||
* Low, intermediate or high-grade dysplasia.
† Solid portion, mural nodule or enhanced cyst.
Figure 4. The receiver operating curve (ROC) of the new nomogram.
4. Discussion
To the best of our knowledge, this is the first study to propose a nomogram with inflammatory markers for predicting invasiveness in patients with IPMN. The predictive power of the nomogram demonstrated in the external validation was acceptable (AUC: 0.649, 95% CI: 0.578-0.720, P < .001). Nomograms are considered useful as they provide a graphical interface for interpreting the effect sizes of predictors in multivariable models. In the present study, we have also established a user-friendly web site for calculation.
Since IPMNs are known to have malignant potential, it is clinically important to determine surgical candidates by predicting the invasiveness of IPMN preoperatively. In addition, the fact that IPMN mostly occurs in older patients raises the need for more sophisticated surgical indications to minimize unnecessary surgery.[17] The International consensus guidelines, which are regarded as the most reliable resources, have been updated based on cytology and image findings to predict malignancy and present a flow chart for surveillance of IPMN.[4] However, several studies have argued that the positive predictive value of the guideline remains limited (27-62%).[18-20] Among these, one study evaluated the utility of the current guidelines and identified 40 (10.5%) patients with malignancy among 382 low-risk patients.[18] This finding suggests that the risk of malignancy may be underestimated based on the guidelines alone and further research for identifying potential risk factors is needed.
Several studies have shown the clinical significance of NLR and PRL in predicting the prognosis of pancreas ductal adenocarcinoma,[9,21] as well as other solid tumors.[10,11,22] In a metaanalysis that assessed the prognostic ability of NLR on overall survival in patients with pancreatic cancer,[23] a high preoperative NLR was identified to be associated with a poorer prognosis than that of patients with lower NLR values, with the cut-off values ranging from 2 to 5. In pancreas ductal adenocarcinoma initiation, neutrophil activity and the release of inflammatory mediators may play central roles in mediating peritumoral inflammation and failed protective immunity, which results in decreased lymphocyte counts.[24] Following this, the association of tumor-associated neutrophils (TANs) with malignant progression in IPMNs was identified.[25] Another study showed that NLR was significantly higher in malignant IPMN patients compared with non-invasive IPMN.[26] The authors also found that NLR in the non-invasive IPMN group was higher than in healthy patients. In addition, NLR decreased after surgical resection in both invasive and non-invasive IPMN groups. Likewise, in our previous study, we identified NLR as an independent prognostic factor of malignant IPMN. The higher recurrence rate of invasive IPMN in patients with NLR ≥3.5 supports a close association between NLR and invasiveness of IPMN.[15]
In terms of ALI, some studies reported that low ALI could serve as a predictor for worse outcomes in other malignancies.[27,28] Also, BMI and serum albumin level, which are the components of ALI, have been studied as prognostic factors for pancreatic cancer.[29,30] We hypothesized that ALI might be useful in identifying high-risk IPMNs, considering that BMI and NLR were also related to the invasiveness of IPMN. There exist only limited data on the association between ALI and invasive IPMN, and there have been studies investigating the new predictive factors such as cystic fluid inflammatory markers.[31,32] Apparently, there remains significant room for more research to validate the roles of these factors, and to develop another platform with improved predictability including the valid factors.
The present study has several limitations, including a limited number of patients and the retrospective design. Above all, the AUC was 0.649, which was below the reported values from previous nomogram studies (0.747-0.787).[6,7] Shimizu et al[6] suggested a simple nomogram with only 4 factors: sex, type of lesion, size of mural nodules, and cytology of pancreatic juice. Following that, another nomogram with 6 variables including carbohydrate antigen 19-9 was proposed in a cross-national multicenter study.[7] In spite of adding inflammatory markers, our new developed nomogram showed a relatively lower AUC. This might be mainly due to heterogeneity of data, caused by individual hospital’s different diagnostic facilities and faculty members who reviewed the clinical data. The study may also have been subject to selection bias, as we only included patients who underwent surgical resection and did not consider the patients who were on active surveillance. Nonetheless, this research was the first to attempt to include inflammatory markers in the predictive platform for invasive IPMN, and it could be the groundwork for future studies. A nationwide prospective study that includes molecular biology with the combination of artificial intelligence techniques is under consideration at our institution to improve the accuracy and predictive value of our risk assessment models.
In conclusion, we have developed a nomogram with inflammatory markers which are recently investigated predictors of invasive IPMN. The results of the external validation leave considerable room for improvement. Further extensive research using different methodology is needed to more accurately predict the risk of malignancy in IPMN.
Acknowledgment
The authors would like to thank Hyemin Kim (Data Manager, Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine) for help with data collection.
Author contributions
(I) Conception and design: So Jeong Yoon, Hongbeom Kim, Chang-Sup Lim, Yong Chan Shin, Wooil Kwon, Jin-Young Jang, Sang Hyun Shin, Jin Seok Heo, In Woong Han; (II) Administrative support: Jin-Young Jang, In Woong Han; (III) Provision of study materials or patients: So Jeong Yoon, Hongbeom Kim, Chang-Sup Lim, Yong Chan Shin, Wooil Kwon, Jin-Young Jang, Sang Hyun Shin, Jin Seok Heo, In Woong Han; (IV) Collection and assembly of data: So Jeong Yoon, Hongbeom Kim, Chang-Sup Lim, Yong Chan Shin, Wooil Kwon, Jin-Young Jang, Sang Hyun Shin, Jin Seok Heo, In Woong Han; (V) Data analysis and interpretation: all authors; (VI) Manuscript writing: all authors; (VII) Final approval of manuscript: all authors
Footnotes
SJY and HK authors contributed equally to this work.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The authors have no conflict of interest to disclose.
Reporting checklist: The authors have completed the STROBE checklist.
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.
How to cite this article: Yoon SJ, Kim H, Lee OJ, Jung JH, Lim CS, Shin YC, Kwon W, Jang JY, Shin SH, Heo JS, Han IW. Development and external validation of a nomogram with inflammatory markers for predicting invasiveness of intraductal papillary mucinous neoplasm of pancreas. Medicine 2022;101:00(e29036).
Abbreviations: ALI = advanced lung cancer inflammation index, AUC = area under the curve, BD-IPMN = branch duct intraductal papillary mucinous neoplasm, BMI = body mass index, CA 19-9 = carbohydrate antigen 19-9, IPMN = intraductal papillary mucinous neoplasm, NLR = neutrophil-to-lymphocyte ratio, PLR = platelet-to-lymphocyte ratio, ROC curve = receiver operating characteristic curve.
References
- [1].Klibansky DA, Reid-Lombardo KM, Gordon SR, Gardner TB. The clinical relevance of the increasing incidence of intraductal papillary mucinous neoplasm. Clin Gastroenterol Hepatol 2012;10:555- 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Testini M, Gurrado A, Lissidini G, Venezia P, Greco L, Piccinni G. Management of mucinous cystic neoplasms of the pancreas. World J Gastroenterol 2010;16:5682- 92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Gurzu S, Bara T, Molnar C, et al. The epithelial-mesenchymal transition induces aggressivity of mucinous cystic neoplasm of the pancreas with neuroendocrine component: an immunohistochemistry study. Pathol Res Pract 2019;215:82- 9. [DOI] [PubMed] [Google Scholar]
- [4].Tanaka M, Fernández-Del Castillo C, Kamisawa T, et al. Revisions of international consensus Fukuoka guidelines for the management of IPMN of the pancreas. Pancreatology 2017;17:738- 53. [DOI] [PubMed] [Google Scholar]
- [5].Correa-Gallego C, Do R, Lafemina J, et al. Predicting dysplasia and invasive carcinoma in intraductal papillary mucinous neoplasms of the pancreas: development of a preoperative nomogram. Ann Surg Oncol 2013;20:4348- 55. [DOI] [PubMed] [Google Scholar]
- [6].Shimizu Y, Yamaue H, Maguchi H, et al. Validation of a nomogram for predicting the probability of carcinoma in patients with intraductal papillary mucinous neoplasm in 180 pancreatic resection patients at 3 high-volume centers. Pancreas 2015;44:459- 64. [DOI] [PubMed] [Google Scholar]
- [7].Jang JY, Park T, Lee S, et al. Proposed nomogram predicting the individual risk of malignancy in the patients with branch duct type intraductal papillary mucinous neoplasms of the pancreas. Ann Surg 2017;266:1062- 8. [DOI] [PubMed] [Google Scholar]
- [8].Attiyeh MA, Fernandez-Del Castillo C, Al Efishat M, et al. Development and validation of a multi-institutional preoperative nomogram for predicting grade of dysplasia in intraductal papillary mucinous neoplasms (IPMNs) of the pancreas: a report from the Pancreatic Surgery Consortium. Ann Surg 2018;267:157- 63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Yang JJ, Hu ZG, Shi WX, Deng T, He SQ, Yuan SG. Prognostic significance of neutrophil to lymphocyte ratio in pancreatic cancer: a meta-analysis. World J Gastroenterol 2015;21:2807- 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Graziosi L, Marino E, De Angelis V, Rebonato A, Cavazzoni E, Donini A. Prognostic value of preoperative neutrophils to lymphocytes ratio in patients resected for gastric cancer. Am J Surg 2015;209:333- 7. [DOI] [PubMed] [Google Scholar]
- [11].Yang HJ, Guo Z, Yang YT, et al. Blood neutrophil-lymphocyte ratio predicts survival after hepatectomy for hepatocellular carcinoma: a propensity score-based analysis. World J Gastroenterol 2016;22:5088- 95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Goh BK, Teo JY, Allen JC, Jr, et al. Preoperative platelet-to-lymphocyte ratio improves the performance of the international consensus guidelines in predicting malignant pancreatic cystic neoplasms. Pancreatology 2016;16:888- 92. [DOI] [PubMed] [Google Scholar]
- [13].Gemenetzis G, Bagante F, Griffin JF, et al. Neutrophil-to-lymphocyte ratio is a predictive marker for invasive malignancy in intraductal papillary mucinous neoplasms of the pancreas. Ann Surg 2017;266:339- 45. [DOI] [PubMed] [Google Scholar]
- [14].Hata T, Mizuma M, Motoi F, et al. Diagnostic and prognostic impact of neutrophil-to-lymphocyte ratio for intraductal papillary mucinous neoplasms of the pancreas with high-grade dysplasia and associated invasive carcinoma. Pancreas 2019;48:99- 106. [DOI] [PubMed] [Google Scholar]
- [15].Kim H, Jung W, Chan Shin Y, et al. The diagnostic and prognostic values of inflammatory markers in intraductal papillary mucinous neoplasm. HPB (Oxford) 2021;23:1623- 8. [DOI] [PubMed] [Google Scholar]
- [16].Swerdlow SH. WHO Classification of Tumors of the Digestive System. 4th ed. 2010;IARC, 304-313.
- [17].Khalid A, Brugge W. ACG practice guidelines for the diagnosis and management of neoplastic pancreatic cysts. Am J Gastroenterol 2007; 102:2339- 49. [DOI] [PubMed] [Google Scholar]
- [18].Goh BK, Lin Z, Tan DM, et al. Evaluation of the Fukuoka Consensus Guidelines for intraductal papillary mucinous neoplasms of the pancreas: results from a systematic review of 1,382 surgically resected patients. Surgery 2015;158:1192- 202. [DOI] [PubMed] [Google Scholar]
- [19].Srinivasan N, Teo JY, Chin YK, et al. Systematic review of the clinical utility and validity of the Sendai and Fukuoka Consensus Guidelines for the management of intraductal papillary mucinous neoplasms of the pancreas. HPB (Oxford) 2018;20:497- 504. [DOI] [PubMed] [Google Scholar]
- [20].Nguyen AH, Toste PA, Farrell JJ, et al. Current recommendations for surveillance and surgery of intraductal papillary mucinous neoplasms may overlook some patients with cancer. J Gastrointest Surg 2015; 19:258- 65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Luo G, Guo M, Liu Z, et al. Blood neutrophil-lymphocyte ratio predicts survival in patients with advanced pancreatic cancer treated with chemotherapy. Ann Surg Oncol 2015;22:670- 6. [DOI] [PubMed] [Google Scholar]
- [22].Ahn HK, Hwang IC, Lee JS, Sym SJ, Cho EK, Shin DB. Neutrophillymphocyte ratio predicts survival in terminal cancer patients. J Palliat Med 2016;19:437- 41. [DOI] [PubMed] [Google Scholar]
- [23].Mowbray NG, Griffith D, Hammoda M, Shingler G, Kambal A, Al-Sarireh B. A meta-analysis of the utility of the neutrophil-to-lymphocyte ratio in predicting survival after pancreatic cancer resection. HPB (Oxford) 2018;20:379- 84. [DOI] [PubMed] [Google Scholar]
- [24].Steele CW, Jamieson NB, Evans TR, et al. Exploiting inflammation for therapeutic gain in pancreatic cancer. Br J Cancer 2013;108:997- 1003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Sadot E, Basturk O, Klimstra DS, et al. Tumor-associated neutrophils and malignant progression in intraductal papillary mucinous neoplasms: an opportunity for identification of high-risk disease. Ann Surg 2015;262:1102- 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Arima K, Okabe H, Hashimoto D, et al. The neutrophil-to-lymphocyte ratio predicts malignant potential in intraductal papillary mucinous neoplasms. J Gastrointest Surg 2015;19:2171- 7. [DOI] [PubMed] [Google Scholar]
- [27].He X, Zhou T, Yang Y, et al. Advanced lung cancer inflammation index, a new prognostic score, predicts outcome in patients with small-cell lung cancer. Clin Lung Cancer 2015;16:e165- 71. [DOI] [PubMed] [Google Scholar]
- [28].Jank BJ, Kadletz L, Schnoll J, Selzer E, Perisanidis C, Heiduschka G. Prognostic value of advanced lung cancer inflammation index in head and neck squamous cell carcinoma. Eur Arch Otorhinolaryngol 2019;276:1487- 92. [DOI] [PubMed] [Google Scholar]
- [29].Shi YQ, Yang J, Du P, et al. Effect of body mass index on overall survival of pancreatic cancer: a meta-analysis. Medicine (Baltimore) 2016;95:e3305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Ruiz-Tovar J, Martín-Pérez E, Fernández-Contreras ME, Reguero-Callejas ME, Gamallo-Amat C. Impact of preoperative levels of hemoglobin and albumin on the survival of pancreatic carcinoma. Rev Esp Enferm Dig 2010;102:631- 6. [DOI] [PubMed] [Google Scholar]
- [31].Harrington KA, Williams TL, Lawrence SA, et al. Multimodal radiomics and cyst fluid inflammatory markers model to predict preoperative risk in intraductal papillary mucinous neoplasms. J Med Imaging (Bellingham) 2020;7:31507doi: 10.1117/1.JMI.7.3.031507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Morgell A, Reisz J.A, Ateeb Z, et al. Metabolic Characterization of Plasma and Cyst Fluid from Cystic Precursors to Pancreatic Cancer Patients Reveal Metabolic Signatures of Bacterial Infection. J Proteome Res 2021;20:2725- 38. [DOI] [PubMed] [Google Scholar]