Validation of a nomogram for predicting survival after resection for adenocarcinoma of the pancreas

Accuracy of models to prognosticate survival after surgery for pancreatic cancer in the era of neoadjuvant therapy

BACKGROUND

Outcomes for pancreatic adenocarcinoma (PDAC) remain difficult to prognosticate. Multiple models attempt to predict survival following the resection of PDAC, but their utility in the neoadjuvant population is unknown. We aimed to assess their accuracy among patients that received neoadjuvant chemotherapy (NAC).

METHODS

We performed a multi-institutional retrospective analysis of patients who received NAC and underwent resection of PDAC. Two prognostic systems were evaluated: the Memorial Sloan Kettering Cancer Center Pancreatic Adenocarcinoma Nomogram (MSKCCPAN) and the American Joint Committee on Cancer (AJCC) staging system. Discrimination between predicted and actual disease-specific survival was assessed using the Uno C-statistic and Kaplan-Meier method. Calibration of the MSKCCPAN was assessed using the Brier score.

RESULTS

A total of 448 patients were included. There were 232 (51.8%) females, and the mean age was 64.1 years (±9.5). Most had AJCC Stage I or II disease (77.7%). For the MSKCCPAN, the Uno C-statistic at 12-, 24-, and 36-month time points was 0.62, 0.63, and 0.62, respectively. The AJCC system demonstrated similarly mediocre discrimination. The Brier score for the MSKCCPAN was 0.15 at 12 months, 0.26 at 24 months, and 0.30 at 36 months, demonstrating modest calibration.

CONCLUSIONS

Current survival prediction models and staging systems for patients with PDAC undergoing resection after NAC have limited accuracy.

Prognostic models to predict survival in patients with pancreatic cancer: a systematic review

BACKGROUND

Pancreatic ductal adenocarcinoma (PDAC) has poor survival. Current treatments offer little likelihood of cure or long-term survival. This systematic review evaluates prognostic models predicting overall survival in patients diagnosed with PDAC.

METHODS

We conducted a comprehensive search of eight electronic databases from their date of inception through to December 2019. Studies that published models predicting survival in patients with PDAC were identified.

RESULTS

3297 studies were identified; 187 full-text articles were retrieved and 54 studies of 49 unique prognostic models were included. Of these, 28 (57.1%) were conducted in patients with advanced disease, 17 (34.7%) with resectable disease, and four (8.2%) in all patients. 34 (69.4%) models were validated, and 35 (71.4%) reported model discrimination, with only five models reporting values >0.70 in both derivation and validation cohorts. Many (n = 27) had a moderate to high risk of bias and most (n = 33) were developed using retrospective data. No variables were unanimously found to be predictive of survival when included in more than one study.

CONCLUSION

Most prognostic models were developed using retrospective data and performed poorly. Future research should validate instruments performing well locally in international cohorts and investigate other potential predictors of survival.

Perioperative risk of pancreatic head resection-nomogram-based prediction of severe postoperative complications as a decisional aid for clinical practice

PURPOSE

To develop nomograms for pre- and early-postoperative risk assessment of patients undergoing pancreatic head resection.

METHODS

Clinical data from 956 patients were collected in a prospectively maintained database. A test (n = 772) and a validation cohort (n = 184) were randomly generated. Uni- and multi-variate analysis and nomogram construction were performed to predict severe postoperative complications (Clavien-Dindo Grades III-V) in the test cohort. External validation was performed with the validation cohort.

RESULTS

We identified ASA score, indication for surgery, body mass index (BMI), preoperative white blood cell (WBC) count, and preoperative alkaline phosphatase as preoperative factors associated with an increased perioperative risk for complications. Additionally to ASA score, BMI, indication for surgery, and the preoperative alkaline phosphatase, the following postoperative parameters were identified as risk factors in the early postoperative setting: the need for intraoperative blood transfusion, operation time, maximum WBC on postoperative day (POD) 1-3, and maximum serum amylase on POD 1-3. Two nomograms were developed on the basis of these risk factors and showed accurate risk estimation for severe postoperative complications (ROC-AUC-values for Grades III-V-preoperative nomogram: 0.673 (95%, CI: 0.626-0.721); postoperative nomogram: 0.734 (95%, CI: 0.691-0.778); each p ≤ 0.001). Validation yielded ROC-AUC-values for Grades III-V-preoperative nomogram of 0.676 (95%, CI: 0.586-0.766) and postoperative nomogram of 0.677 (95%, CI: 0.591-0.762); each p = 0.001.

CONCLUSION

Easy-to-use nomograms for risk estimation in the pre- and early-postoperative setting were developed. Accurate risk estimation can support the decisional process, especially for IPMN-patients with an increased perioperative risk.

SOURCE-PANC: A Prediction Model for Patients With Metastatic Pancreatic Ductal Adenocarcinoma Based on Nationwide Population-Based Data

BACKGROUND

A prediction model for overall survival (OS) in metastatic pancreatic ductal adenocarcinoma (PDAC) including patient and treatment characteristics is currently not available, but it could be valuable for supporting clinicians in patient communication about expectations and prognosis. We aimed to develop a prediction model for OS in metastatic PDAC, called SOURCE-PANC, based on nationwide population-based data.

MATERIALS AND METHODS

Data on patients diagnosed with synchronous metastatic PDAC in 2015 through 2018 were retrieved from the Netherlands Cancer Registry. A multivariate Cox regression model was created to predict OS for various treatment strategies. Available patient, tumor, and treatment characteristics were used to compose the model. Treatment strategies were categorized as systemic treatment (subdivided into FOLFIRINOX, gemcitabine/nab-paclitaxel, and gemcitabine monotherapy), biliary drainage, and best supportive care only. Validation was performed according to a temporal internal-external cross-validation scheme. The predictive quality was assessed with the C-index and calibration.

RESULTS

Data for 4,739 patients were included in the model. Sixteen predictors were included: age, sex, performance status, laboratory values (albumin, bilirubin, CA19-9, lactate dehydrogenase), clinical tumor and nodal stage, tumor sublocation, presence of distant lymph node metastases, liver or peritoneal metastases, number of metastatic sites, and treatment strategy. The model demonstrated a C-index of 0.72 in the internal-external cross-validation and showed good calibration, with the intercept and slope 95% confidence intervals including the ideal values of 0 and 1, respectively.

CONCLUSIONS

A population-based prediction model for OS was developed for patients with metastatic PDAC and showed good performance. The predictors that were included in the model comprised both baseline patient and tumor characteristics and type of treatment. SOURCE-PANC will be incorporated in an electronic decision support tool to support shared decision-making in clinical practice.

Multi-institutional development and external validation of machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection

BACKGROUND

Surgical resection is the only potentially curative treatment for pancreatic ductal adenocarcinoma (PDAC) and the survival of patients after radical resection is closely related to relapse. We aimed to develop models to predict the risk of relapse using machine learning methods based on multiple clinical parameters.

METHODS

Data were collected and analysed of 262 PDAC patients who underwent radical resection at 3 institutions between 2013 and 2017, with 183 from one institution as a training set, 79 from the other 2 institution as a validation set. We developed and compared several predictive models to predict 1- and 2-year relapse risk using machine learning approaches.

RESULTS

Machine learning techniques were superior to conventional regression-based analyses in predicting risk of relapse of PDAC after radical resection. Among them, the random forest (RF) outperformed other methods in the training set. The highest accuracy and area under the receiver operating characteristic curve (AUROC) for predicting 1-year relapse risk with RF were 78.4% and 0.834, respectively, and for 2-year relapse risk were 95.1% and 0.998. However, the support vector machine (SVM) model showed better performance than the others for predicting 1-year relapse risk in the validation set. And the k neighbor algorithm (KNN) model achieved the highest accuracy and AUROC for predicting 2-year relapse risk.

CONCLUSIONS

By machine learning, this study has developed and validated comprehensive models integrating clinicopathological characteristics to predict the relapse risk of PDAC after radical resection which will guide the development of personalized surveillance programs after surgery.

Exploring the significance of novel immune-related gene signatures in the prognosis and immune features of pancreatic adenocarcinoma

BACKGROUND

Immune-related genes (IRGs) are associated with the prognosis of different cancers and are helpful for the diagnosis and management of systematic treatment for cancer patients. However, there have been a few corresponding studies in pancreatic adenocarcinoma (PAAD).

METHODS

The data of PAAD patients were obtained from the TCGA, GEO, and ICGC databases. Additionally, the expression profiles of the normal pancreas from the GTEx database were used to screen differentially expressed immune-related genes (DEIRGs). Cox regression analyses were used to explore overall survival (OS)- and progression-free survival (PFS)-related DEIRGs and to establish two nomograms for PAAD prognosis. Finally, transcription factor (TF), immune infiltration, and unsupervised consensus analyses were performed to understand the potential mechanisms.

RESULTS

An OS-prognostic signature based on seven DEIRGs and a PFS-prognostic signature based on seven DEIRGs were generated, and their robust prognostic ability was confirmed by ROC curves (OS: 0.736 ~ 0.774, PFS: 0.732 ~ 0.840). According to the risk score, the OS and PFS of the high-risk group were poorer than those of the low-risk group in the training set and four external validation sets. In addition, two nomograms based on the signatures and clinical variables also showed excellent discrimination. And two hub regulatory pathways were successfully validated in several independent datasets. Discernable patterns of DEIRGs in unsupervised consensus analysis showed that patients with low expression of immune checkpoints had a favorable prognosis.

CONCLUSION

Two DEIRG-based signatures can be used as independent tools for the prognostic prediction of PAAD and to provide potential novel immunotherapy targets.

Precision Oncology in Surgery: Patient Selection for Operable Pancreatic Cancer

Objective:

We aimed to define preoperative clinical and molecular characteristics that would allow better patient selection for operative resection.

Background:

Although we use molecular selection methods for systemic targeted therapies, these principles are not applied to surgical oncology. Improving patient selection is of vital importance for the operative treatment of pancreatic cancer (pancreatic ductal adenocarcinoma). Although surgery is the only chance of long-term survival, 80% still succumb to the disease and approximately 30% die within 1 year, often sooner than those that have unresected local disease.

Method:

In 3 independent pancreatic ductal adenocarcinoma cohorts (total participants = 1184) the relationship between aberrant expression of prometastatic proteins S100A2 and S100A4 and survival was assessed. A preoperative nomogram based on clinical variables available before surgery and expression of these proteins was constructed and compared to traditional measures, and a postoperative nomogram.

Results:

High expression of either S100A2 or S100A4 was independent poor prognostic factors in a training cohort of 518 participants. These results were validated in 2 independent patient cohorts (Glasgow, n = 198; Germany, n = 468). Aberrant biomarker expression stratified the cohorts into 3 distinct prognostic groups. A preoperative nomogram incorporating S100A2 and S100A4 expression predicted survival and nomograms derived using postoperative clinicopathological variables.

Conclusions:

Of those patients with a poor preoperative nomogram score, approximately 50% of patients died within a year of resection. Nomograms have the potential to improve selection for surgery and neoadjuvant therapy, avoiding surgery in aggressive disease, and justifying more extensive resections in biologically favorable disease.

Development and validation of a novel nomogram for predicting the prognosis of patients with resected pancreatic adenocarcinoma

The survival prediction for patients with resected pancreatic adenocarcinoma by using the Tumor-Node-Metastasis (TNM) staging system remains limited. A nomogram is a efficient tool that can be used to predict the outcome of patients with various types of malignancy. The present study aimed to develop and validate a nomogram for patients with resected pancreatic adenocarcinoma. A total of 368 patients (258 in the training set and 110 in the validation set) who underwent pancreatic adenocarcinoma resection at the China National Cancer Center between January 2008 and October 2018 were included in the present study. The nomogram was established according to the results from Cox multivariate analysis, which was validated by discrimination and calibration. The area under the receiver operating characteristic curve (AUC) was determined to assess the accuracy of survival predictions. The results from multivariate analysis in the training set demonstrated that blood transfusion, T-stage, N-stage, tumor grade, capsule invasion, carbohydrate antigen 199, neutrophil percentage and adjuvant therapy were independent prognostic factors for overall survival (OS; all P<0.05). Subsequently, a nomogram predicting the 1-year, 3-year and 5-year OS rates, with favorable calibration, was established based on the independent prognostic factors. The concordance indices of the nomogram were higher compared with the TNM staging system in both training and validation sets. Furthermore, a clear risk stratification system based on the nomogram was used to classify patients into the three following groups: Low-risk group (≤168), moderate-risk group (168–255) and high-risk group (>255). The risk stratification system demonstrated an improved ability in predicting the 1-year, 3-year and 5-year OS rates compared with the TNM system (AUC, 0.758, 0.709 and 0.672 vs. AUC, 0.614, 0.604 and 0.568; all P<0.05). The present study developed and validated a nomogram for patients with resected pancreatic adenocarcinoma by including additional independent prognostic factors, including tumor marker, immune index, surgical information, pathological data and adjuvant therapy. Taken together, the results from the present study indicated an improved performance of the nomogram in predicting the prognosis of patients with resected pancreatic adenocarcinoma compared with the TNM staging system.

Development and validation of a novel nomogram for pretreatment prediction of liver metastasis in pancreatic cancer

PURPOSE

The diagnostic value of nomogram in pancreatic cancer (PC) with liver metastasis (PCLM) is still largely unknown. We sought to develop and validate a novel nomogram for the prediction of liver metastasis in patients with PC.

METHOD

About 604 pathologically confirmed PC patients from the Sun Yat-sen University Cancer Center (SYSUCC) between July, 2001 and December, 2013 were retrospectively studied. The SYSUCC cohort was randomly assigned to as the training set and internal validation set. Using these two sets, we derived and validated a prognostic model by using concordance index and calibration curves. Another two independent cohorts between August, 2002 and December, 2013 from the Sun Yat-sen Memorial Hospital (SYSMH, n = 335) and Guangdong General Hospital (GDGH, n = 503) was used for external validation.

RESULT

Computed tomography (CT) reported liver metastasis status, carcinoembryonic antigen (CEA) level and differentiation type were identified as risk factors for PCLM in the training set. The final diagnostic model demonstrated good calibration and discrimination with a concordance index of 0.97 and had a robust internal validation. The score ability to diagnose PCLM was further externally validated in SYSMH and GDGH with a concordance index of 0.93. The model showed better calibration and discrimination than CT, CEA and differentiation in each cohort.

CONCLUSION

Based on a large multi-institution database and on the routinely observed CT-reported status, CEA level and tumor differentiation in clinical practice, we developed and validated a novel nomogram to predict PLCM.

A Nomogram for Individual Prediction of Poor Prognosis After Radical Surgery in Patients with Primary Pancreatic Duct Adenocarcinoma

Background

Pancreatic cancer is a highly malignant tumor characterized by poor prognosis. TNM stage cannot always provide accurate prediction of prognosis, which is vital for individualized treatment. Therefore, a novel way to identify patients with poor prognosis after radical surgery is urgently needed.

Material/Methods

The nomogram was established based on a discovery cohort that included 554 patients with PDAC who had received radical surgery from 2012 to 2016. The clinicopathological data were collected. Poor prognosis was evaluated using 25 features, in which appropriate features for a prediction model were identified. A prediction model incorporating the selected features was established. The discriminative capacity was assessed by C-index, calibration by calibration plot, and clinical usefulness by decision curve. The bootstrapping approach was used to perform internal validation.

Results

Characteristics included in the nomogram were coronary artery disease and stroke history, elevated CA125, AJCC stage >II, R0 resection, operating time >6 h, poor differentiation, nerve invasion, length of stay >30 days, and postoperative complications. A C-index of 0.713 indicated good discrimination of the prediction model, and the calibration curve showed acceptable calibration. Survival analysis showed that this model had better discriminative capacity than the AJCC staging system and could distinguish relatively good prognosis from poor prognosis in patients at stage II (especially IIa) and IV.

Conclusions

Our study presents a valid and practical model to predict prognosis of pancreatic cancer patients, which contributes to individualized therapy by assisting surgeons to predict poor prognosis in patients who received radical surgery.

Prediction of postoperative pancreatic fistula using a nomogram based on the updated definition

Purpose

The International Study Group on Pancreatic Fistula's definition of postoperative pancreatic fistula (POPF) has recently been updated. This study aimed to identify risk factors for POPF in patients having pancreaticoduodenectomy (PD) and to generate a nomogram to predict POPF.

Methods

Data on 298 patients who underwent PD from March 2012 to October 2017 was retrospectively reviewed and POPF statuses were redefined. A nomogram was constructed using data from 220 patients and validated using the remaining 78 patients. Independent risk factors for POPF were identified using univariate and multivariate analyses. A predictive nomogram was established based on the independent risk factors and was compared with existing models.

Results

Texture of the pancreas, size of the main pancreatic duct, portal vein invasion, and definitive pathology were the identified risk factors. The nomogram had a C-index of 0.793 and was internally validated. The nomogram performed better (C-index of 0.816) than the other most cited models (C-indexes of 0.728 and 0.735) in the validation cohort. In addition, the nomogram can assign patients into low- (less than 10%), intermediate- (10% to 30%), and high-risk (equal or higher than 30%) groups to facilitate personalized management.

Conclusion

The nomogram accurately predicted POPF in patients having PD.

International validation and update of the Amsterdam model for prediction of survival after pancreatoduodenectomy for pancreatic cancer

BACKGROUND

The objective of this study was to validate and update the Amsterdam prediction model including tumor grade, lymph node ratio, margin status and adjuvant therapy, for prediction of overall survival (OS) after pancreatoduodenectomy for pancreatic cancer.

METHODS

We included consecutive patients who underwent pancreatoduodenectomy for pancreatic cancer between 2000 and 2017 at 11 tertiary centers in 8 countries (USA, UK, Germany, Italy, Sweden, the Netherlands, Korea, Australia). Model performance for prediction of OS was evaluated by calibration statistics and Uno's C-statistic for discrimination. Validation followed the TRIPOD statement.

RESULTS

Overall, 3081 patients (53% male, median age 66 years) were included with a median OS of 24 months, of whom 38% had N2 disease and 77% received adjuvant chemotherapy. Predictions of 3-year OS were fairly similar to observed OS with a calibration slope of 0.72. Statistical updating of the model resulted in an increase of the C-statistic from 0.63 to 0.65 (95% CI 0.64-0.65), ranging from 0.62 to 0.67 across different countries. The area under the curve for the prediction of 3-year OS was 0.71 after updating. Median OS was 36, 25 and 15 months for the low, intermediate and high risk group, respectively (P < 0.001).

CONCLUSIONS

This large international study validated and updated the Amsterdam model for survival prediction after pancreatoduodenectomy for pancreatic cancer. The model incorporates readily available variables with a fairly accurate model performance and robustness across different countries, while novel markers may be added in the future. The risk groups and web-based calculator www.pancreascalculator.com may facilitate use in daily practice and future trials.

Preoperative Clinical and Computed Tomography (CT)-Based Nomogram to Predict Oncologic Outcomes in Patients with Pancreatic Head Cancer Resected with Curative Intent: A Retrospective Study

Background: Currently, proposed nomograms are mainly based on post-operative histopathology. The purpose of this study was to identify preoperative computed tomography (CT) and clinical information that allow prediction of disease-free (DFS) and overall survival (OS) of patients surgically treated for pancreatic head cancer. Methods: A total of 136 patients who underwent curative-intent surgery were retrospectively reviewed. Based on results from multivariate Cox regression analysis, a prediction model was constructed with preoperative CT features and clinical information. Overall performance of the nomogram was calculated by Harrell’s C-index. Results: Symptoms at diagnosis, preoperative serum CA 19-9 ≥ 34 U/mL, and four imaging features (necrosis (DFS, P = 0.066; OS, P = 0.002), possible venous invasion (DFS, P = 0.150, OS, P = 0.055), suspected metastatic regional lymph node (DFS, P = 0.001; OS, P = 0.099), and associated pancreatitis or pseudocyst (DFS, P = 0.013; OS, P = 0.041)) were included to build the nomogram. The c-statistics for the discrimination power of the proposed nomogram was 0.6496 for DFS and 0.6746 for OS. Conclusion: A nomogram derived from preoperative CT and clinical information could estimate the risk of recurrence and all-cause death after curative-intent surgery for radiologically resectable pancreatic head cancer.

Development and validation of a prognostic model to predict the prognosis of patients who underwent chemotherapy and resection of pancreatic adenocarcinoma: a large international population-based cohort study

BACKGROUND

Pancreatic cancer (PaC) remains extremely lethal worldwide even after resection. PaC resection rates are low, making prognostic studies in resected PaC difficult. This large international population-based study aimed at exploring factors associated with survival in patients with resected TNM stage I-II PaC receiving chemotherapy and at developing and internationally validating a survival-predicting model.

METHODS

Data of stage I-II PaC patients resected and receiving chemotherapy in 2003-2014 were obtained from the national cancer registries of Belgium, the Netherlands, Slovenia, and Norway, and the US Surveillance, Epidemiology, and End Results (SEER)-18 Program. Multivariable Cox proportional hazards models were constructed to investigate the associations of patient and tumor characteristics with overall survival, and analysis was performed in each country respectively without pooling. Prognostic factors remaining after backward selection in SEER-18 were used to build a nomogram, which was subjected to bootstrap internal validation and external validation using the European datasets.

RESULTS

A total of 11,837 resected PaC patients were analyzed, with median survival time of 18-23 months and 3-year survival rates of 21-31%. In the main analysis, patient age, tumor T stage, N stage, and differentiation were associated with survival across most countries, with country-specific association patterns and strengths. However, tumor location was mostly not significantly associated with survival. Resection margin, hospital type, tumor size, positive and harvested lymph node number, lymph node ratio, and comorbidity number were associated with survival in certain countries where the information was available. A median survival time- and 1-, 2-, 3-, and 5-year survival probability-predictive nomogram incorporating the backward-selected variables in the main analysis was established. It fits each European national cohort similarly well. Calibration curves showed very good agreement between nomogram-prediction and actual observation. The concordance index of the nomogram (0.60) was significantly higher than that of the T and N stage-based model (0.56) for predicting survival.

CONCLUSIONS

In these large international population-based cohorts, patients with resected PaC receiving chemotherapy have distinct characteristics independently associated with survival, with country-specific patterns and strengths. A robust benchmark population-based survival-predicting model is established and internationally validated. Like previous models predicting survival in resected PaC, our nomogram performs modestly.

Systematic review of clinical prediction models for survival after surgery for resectable pancreatic cancer

BACKGROUND

As more therapeutic options for pancreatic cancer are becoming available, there is a need to improve outcome prediction to support shared decision-making. A systematic evaluation of prediction models in resectable pancreatic cancer is lacking.

METHODS

This systematic review followed the CHARMS and PRISMA guidelines. PubMed, Embase and Cochrane Library databases were searched up to 11 October 2017. Studies reporting development or validation of models predicting survival in resectable pancreatic cancer were included. Models without performance measures, reviews, abstracts or more than 10 per cent of patients not undergoing resection in postoperative models were excluded. Studies were appraised critically.

RESULTS

After screening 4403 studies, 22 (44 319 patients) were included. There were 19 model development/update studies and three validation studies, altogether concerning 21 individual models. Two studies were deemed at low risk of bias. Eight models were developed for the preoperative setting and 13 for the postoperative setting. Most frequently included parameters were differentiation grade (11 of 21 models), nodal status (8 of 21) and serum albumin (7 of 21). Treatment-related variables were included in three models. The C-statistic/area under the curve values ranged from 0·57 to 0·90. Based on study design, validation methods and the availability of web-based calculators, two models were identified as the most promising.

CONCLUSION

Although a large number of prediction models for resectable pancreatic cancer have been reported, most are at high risk of bias and have not been validated externally. This overview of prognostic factors provided practical recommendations that could help in designing easily applicable prediction models to support shared decision-making.

Preoperative prediction of peripancreatic vein invasion by pancreatic head cancer

Background

Pancreatic adenocarcinoma is often diagnosed at an advanced stage when adjacent vascular invasion is present. Accurate evaluation of presence of vascular invasion can help guide therapy. The aim of this study was to construct a nomogram for preoperative prediction of peripancreatic vein invasion in patients with pancreatic head cancer.

Study design

Data of patients with carcinoma head of pancreas and suspected peripancreatic invasion (n = 247) who underwent pancreatic resection with venous reconstruction between January 2012 and January 2017 at four academic institutions were retrospectively analyzed. Univariate and multivariate analyses were used to identify independent risk factors for vein invasion from among demographic, biological, conditional host-related, and anatomical data. A predictive nomogram was constructed based on the identified independent risk factors.

Results

The nomogram was constructed using data from 181 patients while the validation cohort consisted of 66 patients. Length of tumor contact (P = 0.031), circumferential vein involvement (P = 0.048), and venous contour abnormalities (P = 0.001) were independent predictors of venous invasion. The C-index of the model in predicting venous invasion was 0.963 for the external validation cohort. Patients could be assigned into low- (< 50%), intermediate- (50–90%), and high-risk (> 90%) groups based on the nomogram to facilitate personalized management.

Conclusions

Vein invasion by pancreatic head cancer is mainly associated with anatomical factors. The nomogram for prediction of vein invasion was found to be practicable.

Electronic supplementary material

The online version of this article (10.1186/s40644-018-0179-z) contains supplementary material, which is available to authorized users.

Can physician gestalt predict survival in patients with resectable pancreatic adenocarcinoma?

PURPOSE

Clinician gestalt may hold unexplored information that can be capitalized upon to improve existing nomograms. The study objective was to evaluate physician ability to predict 2-year overall survival (OS) in resected pancreatic ductal adenocarcinoma (PDAC) patients based on pre-operative clinical characteristics and routine CT imaging.

METHODS

Ten surgeons and two radiologists were provided with a clinical vignette (including age, gender, presenting symptoms, and pre-operative CA19-9 when available) and pre-operative CT scan for 20 resected PDAC patients and asked to predict the probability of each patient reaching 2-year OS. Receiver operating characteristic curves were used to assess agreement and to compare performance with an established institutional nomogram.

RESULTS

Ten surgeons and 2 radiologists participated in this study. The area under the curve (AUC) for all physicians was 0.707 (95% CI 0.642-0.772). Attending physicians with > 5 years experience performed better than physicians with < 5 years of clinical experience since completion of post-graduate training (AUC = 0.710, 95% CI [0.536-0.884] compared to AUC = 0.662, 95% CI [0.398-0.927]). Radiologists performed better than surgeons (AUC = 0.875, 95% CI [0.765-0.985] compared to AUC = 0.656, 95% CI [0.580-0.732]). All but one physician outperformed the clinical nomogram (AUC = 0.604).

CONCLUSIONS

This pilot study demonstrated significant promise in the quantification of physician gestalt. While PDAC remains a difficult disease to prognosticate, physicians, particularly those with more clinical experience and radiologic expertise, are able to perform with higher accuracy than existing nomograms in predicting 2-year survival.

Proposal of a modified American Joint Committee on Cancer staging scheme for resectable pancreatic ductal adenocarcinoma with a lymph node ratio-based N classification: A retrospective cohort study

The recently launched 8th edition of the American Joint Committee on Cancer (AJCC) staging scheme for pancreatic ductal adenocarcinoma (PDAC) did not account for the impact of the total examined lymph node count on prognostic accuracy. In this population-based cohort study, we proposed a modified AJCC staging scheme by incorporating a lymph node ratio (LNR)-based N classification for patients with resectable PDAC.We analyzed 8615 patients with resectable PDAC from the Surveillance, Epidemiology, and End Results database between 2004 and 2013. The optimal cut-off points for LNR were identified by recursive partitioning, and an LNR-based N classification was designed accordingly.The LNR-based N classification could further stratify patients with the 8th AJCC N1 and N2 disease into subgroups with significantly different overall survival (P < .001 for both). By replacing the 8th AJCC N classification with the corresponding LNR-based N classification, we further proposed a modified AJCC staging scheme. The modified AJCC staging outperformed the 8th AJCC staging in terms of the discriminatory capacity measured by the concordance index and Akaike information criterion, and the prognostic homogeneity assessed by using the likelihood ratio chi-squared test and stratified survival analysis.Replacing the 8th AJCC N classification with the LNR-based N classification can improve the prognostic performance of the 8th AJCC staging scheme for PDAC.

Genome-scale analysis to identify prognostic markers in patients with early-stage pancreatic ductal adenocarcinoma after pancreaticoduodenectomy

BACKGROUND

Molecular analysis is a promising source of clinically useful prognostic biomarkers. The aim of this investigation was to identify prognostic biomarkers for patients with early-stage pancreatic ductal adenocarcinoma (PDAC) after pancreaticoduodenectomy.

METHODS

An RNA sequencing dataset of PDAC was obtained from The Cancer Genome Atlas. Survival analysis and weighted gene co-expression network analysis were used to investigate the prognostic markers of early-stage PDAC after pancreaticoduodenectomy.

RESULTS

Using whole genome expression level screening, we identified 1,238 markers that were related to the prognosis of PDAC after pancreaticoduodenectomy, and identified 9 hub genes (ARHGAP30, HCLS1, CD96, FAM78A, ARHGAP15, SLA2, CD247, GVINP1, and IL16) using the weighted gene co-expression network analysis approach. We also constructed a signature comprising the 9 hub genes and weighted by the regression coefficient derived from a multivariate Cox proportional hazards regression model to divide patients into a high-risk group, with increased risk of death, and a low-risk group, with significantly improved overall survival (adjusted P=0.026, adjusted HR =0.513, 95% CI =0.285-0.924). The prognostic signature of the 9 genes demonstrated good performance for predicting 1-year overall survival (area under the respective receiver operating characteristic curves =0.641).

CONCLUSION

Our results have provided a new prospect for prognostic biomarkers of PDAC after pancreaticoduodenectomy, and may have a value in clinical application.

Palliative radiation therapy in patients with metastasized pancreatic cancer - description of a rare patient group

Background

Pancreatic cancer (PAC) patients experience a high rate of locoregional recurrences and distant metastasis finally leading to their demise even after curatively-intended multidisciplinary treatment approaches including surgery, chemotherapy and radiotherapy. However, clinical reports on bone and brain metastases in PAC patients are extremely rare and thus timing and dose description are not well defined. Our work therefore summarizes a mono-institutional experience on the use of radiotherapy (RT) for PAC patients with metastatic disease with the aim of identifying overall survival and treatment response in this rarely reported patient group.

Method

Forty-four PAC patients with 66 metastatic lesions were treated with palliative radiotherapy (RT). Thirty-three patients (48 lesions), 7 patients (11 lesions) and 5 patients (7 lesions) with bone, liver and brain metastases analyzed respectively were analyzed; one patient had both bone and cerebral metastases and was treated for the lesions, thus including him in both subgroups. Indications for RT were pain, neurological impairment, risk of pathological fracture or imminent danger for development of any of these conditions in case of tumor progression. Median age was 64 years (range 38 to 78 years) and there were 27 male (61%) and 17 (39%) female patients. Analyses of overall survival (OS) and local control were performed. OS was calculated from the first day of RT.

Results

Median overall survival (mOS) of all patients after start of RT was 4.2 months. Survival rates after 1, 3 and 6 months were 79.3%, 55.3% and 30.3% respectively. Patients presenting with bone metastasis had a mOS of 3.1 months and after 1, 3 and 6 months, survival rates were 75.3%, 46.5% and 19.9% respectively. Symptomatic response to therapy was recorded in 85% of all evaluated patients with bone metastasis. Patients undergoing radiosurgery because of liver metastasis were locally controlled in all but one patient after a median follow-up of 8.3 months.

Conclusion

Overall survival of all patients with metastatic disease was considerably worse. A major goal for the future must be the selection of an appropriate RT treatment in terms of duration and technique for these PAC patients.

Can statistically determined prognostic factors predict the long-term survival of patients with pancreatic ductal adenocarcinoma following surgical resection?: Clinicopathological analysis of 82 long-term survivors

OBJECTIVE

The objective of this study were to analyze and describe the clinicopathological characteristics of long-term pancreatic ductal adenocarcinoma survivors and to determine if statistically identified prognostic factors can be used to predict the actual survival.

METHODS

Between January 2000 and December 2007, 537 patients with resectable pancreatic ductal adenocarcinoma underwent surgical resection at a single institute. Medical records were retrospectively reviewed, and 9 patients were excluded. Of the remaining 528, patients who survived for more than 5 years were classified as long-term survivors.

RESULTS

The actual 5-year survival rate of all 528 included patients was 15.5% (82 patients; median follow-up period, 82.7 months). Fifty-eight patients (70.7%) were diagnosed with cancers located in the head of pancreas, and the median size was 2.5 cm (range, 0.8-12.0 cm). Three patients had focal involvement noted on the resected surfaces. Poorly differentiated carcinoma (9.8%), lymph node metastasis (32.9%), lymphovascular invasion (25.6%), and perineural invasion (48.8%) were detected. Fifty-five of the 82 long-term survivors whose clinicopathological characteristics included several dismal predictors survived for more than 5 years without recurrence.

CONCLUSIONS

The long-term survivors identified in this study did not meet the appropriate prognostic criteria. Therefore, there are limitations in the use of statistically determined prognostic factors for estimating in the long term.

Pancreatic cancer and predictors of survival: comparing the CA 19-9/bilirubin ratio with the McGill Brisbane Symptom Score

INTRODUCTION

Few tools predict survival from pancreatic cancer (PAC). The McGill Brisbane Symptom Score (MBSS) based on symptoms at presentation (weight loss, pain, jaundice and smoking) was recently validated. The present study compares the ability of four strategies to predict 9-month survival: MBSS, carbohydrate antigen 19-9 (CA 19-9) alone, CA19-9-to-bilirubin ratio and a combination of MBSS and the CA19-9-to-bilirubin ratio.

METHODOLOGY

A retrospective review of 133 patients diagnosed with PAC between 2005 and 2011 was performed. Survival was determined from the Quebec civil registry. Blood CA 19-9 and bilirubin values were collected (n = 52) at the time of diagnosis. Receiver-operating characteristic (ROC) curves were used to determine a cutoff for optimal test characteristics of CA 19-9 and CA19-9-to-total bilirubin ratio in predicting survival at 9 months. Predictive characteristics were then calculated for the four strategies.

RESULTS

Of the four strategies, the one with the greatest negative predictive value was the MBSS: negative predictive value (NPV) was 90.2% (76.9-97.3%) and the positive likelihood ratio (LR) was the greatest. The ability of CA 19-9 levels alone, at baseline, to predict survival was low. For the CA19-9-to-bilirubin ratio, the test characteristics improved but remained non-significant. The best performing strategy according to likelihood ratios was the combined MBSS and CA19-9 to the bilirubin ratio.

CONCLUSION

CA19-9 levels and the CA19-9-to-bilirubin ratio are poor predictors of survival for PAC, whereas the MBSS is a far better predictor, confirming its clinical value. By adding the CA19-9-to-bilirubin ratio to the MBSS the predictive characteristics improved.

Does long-term survival in patients with pancreatic cancer really exist? Results from the CONKO-001 study

BACKGROUND

Long-term survival (LTS) in patients (pts) with pancreatic cancer is still uncommon, little data is available to identify long-term survivors. The CONKO-001 study, which established gemcitabine after resection as adjuvant therapy, may provide data to answer this question.

METHODS

CONKO-001 pts with an overall survival ≥5 years were compared to those who survived <5 years. Central re-evaluation of primary histology was performed. Univariate analysis with the χ(2) -test identified qualifying factors. Logistic regression was used to investigate the influence of these covariates on LTS.

RESULTS

Of the evaluable 354 CONKO-001 pts, 54 (15%) with an overall survival ≥5 years were identified. It was possible to obtain tumor specimens of 39 pts (72%). Histological re-evaluation confirmed adenocarcinoma in 38 pts, 1 showed a high-grade neuroendocrine tumor. Univariate analysis for all 53 LTS pts with adenocarcinoma compared to the remaining 300 non-LTS pts revealed as relevant active treatment, tumor grading, tumor size, lymph nodes. No significance could be demonstrated for resection margin, sex, age, Karnofsky performance status, CA 19-9 at study entry. In multivariate analysis, tumor grading, active treatment, tumor size, lymph node involvement were independent prognostic factors for LTS.

CONCLUSION

Long-term survival can be achieved in adenocarcinoma of the pancreas.

Poorly differentiated resectable pancreatic cancer: is upfront resection worthwhile?

BACKGROUND

Poorly differentiated, resectable pancreatic ductal adenocarcinoma is associated with early recurrence and may benefit from neoadjuvant treatment. The aim of this study was to evaluate clinicopathologic characteristics and survival of patients with resectable pancreatic ductal adenocarcinoma according to histologic grading.

METHODS

A total of 502 patients who underwent resection for pancreatic ductal adenocarcinoma between 1990 and 2008 were analyzed via the use of different histologic grading.

RESULTS

Well-differentiated (G1), moderately differentiated (G2), and poorly differentiated (G3) pancreatic ductal adenocarcinomas were found in 23 (4.5%), 310 (62%), and 169 (33.5%) patients. Adjuvant therapy, N status, grading, and R status were independent predictors of disease-specific survival for the entire cohort, with 1- and 5-year disease-specific survival rates of 81% and 21%, respectively. Only the presence of symptoms was a significant clinical predictor of G3 status (P = .035). G3 neoplasms were characterized by a greater rate of lymph node metastases, microvascular/perineural invasion, and R2 resections. Median disease-specific survival was 77, 26, and 20 months for G1, G2, and G3 neoplasms (P < .0001). Median disease-free survival was 63, 14, and 9 months for G1, G2, and G3 pancreatic ductal adenocarcinoma (P < .0001). Adjuvant therapy improved disease-specific survival in G2 (P < .04) and G3 (P < .0001) pancreatic ductal adenocarcinoma, with a greater survival benefit for G3 neoplasms (hazard ratio: 1.334 vs 2.116).

CONCLUSION

G3 pancreatic ductal adenocarcinoma is associated with a lesser rate of disease-free survival after resection and with the presence of other poor prognostic factors. The benefit of adjuvant therapy is greater in G3 than in G1 and G2 neoplasms. On the basis of these findings, patients with resectable G3 PDAC can be considered as possible targets for neoadjuvant treatment.

Downstaging in pancreatic cancer: a matched analysis of patients resected following systemic treatment of initially locally unresectable disease

BACKGROUND

Patients with locally unresectable pancreatic cancer (AJCC stage III) have a median survival of 10-14 months. The objective of this study was to evaluate outcome of initially unresectable patients who respond to multimodality therapy and undergo resection.

METHODS

Using a prospectively collected database, patients were identified who were initially unresectable because of vascular invasion and had sufficient response to nonoperative treatment to undergo resection. Overall survival (OS) was compared with a matched group of patients who were initially resectable. Case matching was performed using a previously validated pancreatic cancer nomogram.

RESULTS

A total of 36 patients with initial stage III disease were identified who underwent resection after treatment with either systemic therapy or chemoradiation. Initial unresectability was determined by operative exploration (n = 15, 42%) or by cross-sectional imaging (n = 21, 58%). Resection consisted of pancreaticoduodenectomy (n = 31, 86%), distal pancreatectomy (n = 4, 11%), and total pancreatectomy (n = 1, 3%). Pathology revealed T3 lesions in 26 patients (73%), node positivity in 6 patients (16%), and a negative margin in 30 patients (83%). The median OS in this series was 25 months from resection and 30 months since treatment initiation. There was no difference in OS from time of resection between the initial stage III patients and those who presented with resectable disease (P = .35).

CONCLUSIONS

In this study, patients who were able to undergo resection following treatment of initial stage III pancreatic cancer experienced survival similar to those who were initially resectable. Resection is indicated in this highly select group of patients.

External validation of nomogram for the prediction of recurrence after curative resection in early gastric cancer

BACKGROUND

Nomograms are statistics-based tools that provide the overall probability of a specific outcome. In our previous study, we developed a nomogram that predicts recurrence of early gastric cancer (EGC) after curative resection. We carried out this study to externally validate our EGC nomogram.

PATIENTS AND METHODS

The EGC nomogram was established from a retrospective EGC database that included 2923 consecutive patients. This nomogram was independently externally validated for a cohort of 1058 consecutive patients. For the EGC nomogram validation, we assessed both discrimination and calibration.

RESULTS

Within the follow-up period (median 37 months), a total of 11 patients (1.1%) experienced recurrence. The concordance index (c-index) was 0.7 (P = 0.02) and the result of the overall C index was 0.82 [P = 0.006, 95% confidence interval (CI) 0.59-1.00]. The goodness of fit test showed that the EGC nomogram had significantly good fit for 1- and 2-year survival intervals (P = 0.998 and 0.879, respectively). The actual and predicted survival outcomes showed good agreement, suggesting that the survival predictions from the nomogram are well calibrated externally.

CONCLUSIONS

A preexisting nomogram for predicting disease-free survival (DFS) of EGC after surgery was externally validated. The nomogram is useful for accurate and individual prediction of DFS, patient prognostication, counseling, and follow-up planning.

64-slice CT perfusion imaging of pancreatic adenocarcinoma and mass-forming chronic pancreatitis

RATIONALE AND OBJECTIVES

To investigate 64 computed tomography (CT) perfusion imaging features of patients with pancreatic cancer and mass-forming chronic pancreatitis.

MATERIALS AND METHODS

Between January 2003 and April 2010, 234 patients with pancreatic mass underwent 64-CT perfusion imaging. Among them, the histopathological results of 64 patients were proven to be pancreatic adenocarcinoma and 15 patients were proven to be mass-forming chronic pancreatitis. Additionally, CT perfusion imaging was performed in 33 healthy volunteers served as controls. The slice data were processed using CT perfusion software. Perfusion parameters including time density curve, blood flow, blood volume, permeability, peak enhancement, and time to peak were recorded.

RESULTS

Blood flow was 77% lower in patients with pancreatic adenocarcinoma than in controls, 48% lower in patients with mass-forming chronic pancreatitis than in controls, and 56% lower in patients with pancreatic adenocarcinoma than with mass-forming chronic pancreatitis (P < .016). Blood volume was 65% lower in pancreatic adenocarcinoma than in controls, 27% lower in mass-forming chronic pancreatitis than in controls, and 53% lower in cancer than mass-forming chronic pancreatitis (P < .016). Permeability was 559% higher in pancreatic adenocarcinoma than in controls, 821% higher in mass-forming chronic pancreatitis than in controls, and 28% lower in cancer than mass-forming chronic pancreatitis (P < .016). Peak enhancement was 27% lower and time to peak 23% longer in pancreatic adenocarcinoma than mass-forming chronic pancreatitis (P < .016). Time-density curve showed the peak of mass-forming chronic pancreatitis is earlier and higher than that of pancreatic adenocarcinoma, and the peak of mass-forming chronic pancreatitis is later and lower than that of controls.

CONCLUSION

CT perfusion imaging can provide additional quantitative hemodynamic information of pancreatic adenocarcinoma and mass-forming chronic pancreatitis.

Sociodemographics and comorbidities influence decisions to undergo pancreatic resection for neoplastic lesions

INTRODUCTION

Pancreatic resection is being performed with increasing frequency and safety. Technical outcomes and long-term survival for neoplastic lesions are well reported; however, reasons why patients do not undergo surgery for potentially resectable lesions are not well understood. The aim of this study was to determine the factors contributing to the decision not to operate for resectable pancreatic neoplasms.

METHODS

From 2004 to 2008, all patients with resectable pancreatic neoplasms at a single high-volume hepatopancreaticobiliary center were evaluated. The impact of patient factors, sociodemographics, medical comorbidities (Charlson combined comorbidity index (CCI) and ACCI), disease factors (tumor characteristics), and surgical factors (type of resection required) on the decision to undergo pancreatectomy were analyzed using univariate and multivariate binary logistic regression analysis.

RESULTS

Three hundred seventy-five patients with resectable pancreatic lesions were identified. The median age was 62 years (21-93); 203 out of 375 (54.1%) were males. Fifty-five (14.7%) did not undergo resection. On univariate analysis, age (odds ratio (OR) 1.116, p < 0.001), non-English speaking background (NESB; OR 4.276, p = 0.001), tumor type (p = 0.001 increased for cystic neoplasms including intraductal papillary mucinous neoplasm), CCI score (OR 1.239, p = 0.001), and ACCI score (OR 1.433, p < 0.001) were associated with an increased risk of not undergoing resection. Gender, age, marital status, and urban residence were not predictive. On multivariate analysis, NESB (p = 0.018) and the ACCI (p = 0.002) remained predictive of not undergoing resection. The majority of patients did not undergo surgery because the patient declined in 25 out of 55 (45.5%), and resection was not offered in 15 out of 55 (27.3%). In the remainder, medical contraindications precluded surgery. Advanced age, tumor type, comorbidities (27.3%), age (21.8%), surgical risk (29.1%), frailty (18.2%), and uncertain diagnosis (5.5%) were cited as reasons for not proceeding with surgery.

CONCLUSION

Patients with a higher ACCI and those from a NESB are less likely to undergo surgery for resectable neoplastic lesions of the pancreas. These factors must be taken into consideration in the decision-making process when considering surgery for patients with pancreatic neoplasms. Novel strategies should be employed to optimize access to surgery for patients with resectable pancreatic neoplasms.

A neoadjuvant strategy for pancreatic adenocarcinoma increases the likelihood of receiving all components of care: lessons from a single-institution database

BACKGROUND

Recent studies have shown adjuvant therapy improves outcomes from pancreatic cancer (PC). This study investigates receipt and timing of PC treatments, and association with outcomes.

METHODS

The analysis cohort consisted of patients with newly-diagnosed PC at a single institution over 5 years. Primary Endpoints were (i) receipt of recommended therapy, and (ii) overall survival (OS).

RESULTS

Among 102 patients, 52 underwent resection. Out of 36 localized resected and 16 locally advanced resected (LAR) patients, 26 and 13, respectively, received adjuvant therapy. Six of the latter group received neoadjuvant therapy. Median OS for resected patients was 15.7 months (range 0.6-51.4), compared with 7.7 for unresected patients (range 0.4-32.0) (P < 0.001), and 14.0 months for patients with resection alone (range 0.6-24.4) vs. 16.1 for patients who also received adjuvant therapy (range 3.2-51.4) (P= 0.027). Out of 46 patients undergoing up-front resection, 33 had R0 surgical margins. For the six LAR patients undergoing neoadjuvant therapy, all margins were R0.

CONCLUSION

After resection, a substantial proportion of patients do not receive adjuvant therapy, and have worse survival. In this study, neoadjuvant treatment increased both the proportion of patients receiving all components of recommended therapy and the R0 resection rate.