A nomogram based on preoperative inflammatory markers predicting the overall survival of pancreatic ductal adenocarcinoma

Influence of sarcopenia on postoperative complications and long-term survival in pancreatic cancer patients undergone pancreaticoduodenectomy

Background

Sarcopenia has the potential to impact the postoperative results and extended prognosis of various types of tumors. Nevertheless, the specific impact of sarcopenia on the postoperative results and long-term survival of pancreatic cancer (PC) following pancreaticoduodenectomy (PD) remains inadequately elucidated. This study investigates the significance of sarcopenia according to various Asian standards on postoperative complications and long-term prognosis in PC patients who have undergone PD.

Methods

This retrospective study systematically analyzed patients with PC who underwent PD from January 2015 to December 2022. Sarcopenia was diagnosed by the skeletal muscle index (SMI) obtained by the skeletal muscle area normalized for height squared on the third lumbar vertebra on computed tomography (CT) images. Univariate and multivariate logistic regression analysis were performed to analyze the correlation between sarcopenia and postoperative complications, while Cox regression analysis was utilized to explore the influence of sarcopenia on overall survival (OS) and recurrence-free survival (RFS) in PC patients after PD.

Results

We enrolled 162 patients with PC after PD (92 males and 70 females, mean age: 63.78 ± 10.27 years), including 83 and 79 patients with sarcopenia and non-sarcopenia, respectively. Compared with non-sarcopenia patients, sarcopenia exhibited higher rates of recurrence rate (75% versus 59%, p = 0.039). Univariate and multivariate logistic regression analysis showed that sarcopenia did not affect the incidence of complications in patients with PC after PD in three Asian sarcopenia criteria. Multivariate Cox regression analysis indicated that sarcopenia was an independent risk factor for OS (hazard ratio [HR]: 2.49, 95% confidence interval [CI]: 1.73-3.60, p < 0.001) and RFS(hazard ratio [HR]: 1.70, 95%confidence interval [CI]: 1.12-2.50, p = 0.012) of PC patients with PD in Japanese Society of Hepatology criteria. Meanwhile, according to the Asian pancreatic cancer population standard, sarcopenia is an independent risk factor affecting the long-term OS (hazard ratio [HR]: 2.59, 95% confidence interval [CI]: 1.80-3.70, p < 0.001) and RFS (hazard ratio [HR]: 2.00, 95% confidence interval [CI]: 1.36-3.00, p < 0.001) of PC after PD. While sarcopenia is recognized as a risk factor for OS (hazard ratio [HR]: 1.81, 95% confidence interval [CI]: 1.08-3.10, p = 0.025) in PC patients based on the Fujiwara criteria, it is not found to be associated with RFS (hazard ratio [HR]: 1.60, 95% confidence interval [CI]: 0.90-3.00, p = 0.10). The model based on sarcopenia and clinical characteristics has high predictive ability for OS and RFS.

Conclusion

Various Asian diagnostic criteria do not link sarcopenia with postoperative complications in PC patients after PD. Nevertheless, sarcopenia remains a significant independent risk factor for long-term survival, and its combination with clinical characteristics can aid clinicians in predicting long-term survival outcomes.

The ratio of skeletal muscle mass to body mass index combined with inflammatory immune markers to stratify survival of pancreatic cancer after pancreatoduodenectomy

BACKGROUND

We sought to combine skeletal muscle index and inflammatory immune markers to stratify long-term survival in patients with pancreatic cancer after pancreatoduodenectomy (PD).

METHODS

A total of 581 patients with pancreatic cancer underwent PD were included, and divided into the training and validation cohort. Image analysis of computed tomography scans was used to calculate the ratio of skeletal muscle (SM) area to body mass index (BMI). Naples prognostic score (NPS) was calculated from blood-test inflammatory immune markers. Propensity score matching (PSM) analysis was performed to minimize biases of clinicopathological characteristics. To estimate the overall survival (OS), a nomogram was developed using the training cohort. The predictive accuracy of nomogram was estimated by concordance index (C-index), calibration curve, and receiver operating characteristics (ROC) curve.

RESULTS

After PSM analysis, SM/BMI ratio, NPS, lymph node metastasis, TNM stage, surgical margin, tumor grade and adjuvant therapy were independent predictors of OS, which were all assembled into nomogram. The SM/BMI ratio was the best single-predictor for 3- and 5-year OS, with an AUC of 0.805 (95% CI: 0.755-0.855) and 0.812 (95% CI: 0.736-0.888), respectively. Harrell's c-index of the nomogram in the training cohort was 0.786 (95% CI: 0.770-0.802), and the area under ROC curve of 1-year, 3- and 5-year OS prediction were 0.869 (95%CI: 0.837-0.901), 0.846 (95%CI: 0.810-0.882) and 0.849 (95%CI: 0.801-0.896).

CONCLUSIONS

The nomogram based on SM/BMI ratio and NPS had excellent predictive performance, which should be incorporated to conventional risk scores to stratify survival of patients with PDAC after PD.

Glucose-to-Lymphocyte Ratio (GLR) as an Independent Prognostic Factor in Patients with Resected Pancreatic Ductal Adenocarcinoma-Cohort Study

Background: We retrospectively evaluated the usefulness of an elevated glucose-to-lymphocyte ratio (GLR) as a sensitive prognostic biomarker of disease-specific survival in 338 patients who underwent surgical resection of pancreatic ductal adenocarcinoma (PDAC). Methods: The optimal GLR cutoff value was determined using the method of Contal and O'Quigley. Patient demographics, clinical information, and imaging data were analyzed to identify preoperative predictors of long-term survival outcomes. Results: Elevated GLR correlated significantly with aggressive tumor biologic behaviors, such as a high carbohydrate antigen (CA) 19-9 level (p = 0.003) and large tumor size (p = 0.011). Multivariate analysis identified (1) GLR > 92.72 [hazard ratio (HR) = 2.475, p < 0.001], (2) CA 19-9 level > 145.35 (HR = 1.577, p = 0.068), and (3) symptoms (p = 0.064) as independent predictors of long-term, cancer-specific survival. These three risk factors were used to group patients into groups 1 (0 factors), 2 (1-2 factors), and 3 (3 factors), which corresponded to significantly different 5-year overall survival rates (50.2%, 34.6%, and 11.7%, respectively; p < 0.001). Conclusions: An elevated preoperative GLR is associated with aggressive tumor characteristics and is an independent predictor of poor postoperative prognosis in patients with PDAC. Further prospective studies are required to verify these findings.

The Development of a Prediction Model Based on Random Survival Forest for the Postoperative Prognosis of Pancreatic Cancer: A SEER-Based Study

Simple Summary

Surgery is the main treatment to cure pancreatic cancer (PC). However, the 5-year survival rate of surgical resection is only 10–20%. The aim of our study was to develop a prediction model with the novel machine learning algorithm random survival forest (RSF) and to offer easy-to-use prediction tools, including risk stratification and individual prognosis. The study would benefit patients and physicians in postoperative management and facilitate personalized medicine.

Abstract

Accurate prediction for the prognosis of patients with pancreatic cancer (PC) is a emerge task nowadays. We aimed to develop survival models for postoperative PC patients, based on a novel algorithm, random survival forest (RSF), traditional Cox regression and neural networks (Deepsurv), using the Surveillance, Epidemiology, and End Results Program (SEER) database. A total of 3988 patients were included in this study. Eight clinicopathological features were selected using least absolute shrinkage and selection operator (LASSO) regression analysis and were utilized to develop the RSF model. The model was evaluated based on three dimensions: discrimination, calibration, and clinical benefit. It found that the RSF model predicted the cancer-specific survival (CSS) of the postoperative PC patients with a c-index of 0.723, which was higher than the models built by Cox regression (0.670) and Deepsurv (0.700). The Brier scores at 1, 3, and 5 years (0.188, 0.177, and 0.131) of the RSF model demonstrated the model’s favorable calibration and the decision curve analysis illustrated the model’s value of clinical implement. Moreover, the roles of the key variables were visualized in the Shapley Additive Explanations plotting. Lastly, the prediction model demonstrates value in risk stratification and individual prognosis. In this study, a high-performance prediction model for PC postoperative prognosis was developed, based on RSF The model presented significant strengths in the risk stratification and individual prognosis prediction.

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.

Development and validation of a nomogram for predicting post-operative abdominal infection in patients undergoing pancreaticoduodenectomy

AIM

The aim of this retrospective study is to develop and validate a predictive nomogram for predicting the risk of post-operative abdominal infection (PAI) in patients undergoing pancreaticoduodenectomy (PD).

METHODS

A total of 360 patients who underwent PD were enrolled into this research and randomly divided into the development and validation group. The clinical data of patients were statistically compared and the nomogram was constructed based on the results of multivariate logistic regression analysis and stepwise (stepAIC) selection. The nomogram was internally and crossly validated by the development and validation cohort. The discriminatory ability of the nomogram was estimated by AUC (Area Under the receiver operating characteristic Curve), calibration curve and decision curve analysis.

RESULTS

After PD, post-operative abdominal infection occurred in 33.89% (n = 122) of patients. The nomogram showed that preoperative biliary drainage and C-reactive protein (CRP), direct bilirubin (DB), alkaline phosphatase (AKP) levels on the 3rd postoperative day (POD3) were independent prognostic factors for abdominal infection after PD. The internal and cross validation of Receiver Operating Characteristic (ROC) curve was statistically significant (AUC = 0.723 and 0.786, respectively). The calibration curves showed good agreement between nomogram predictions and actual observations. The decision curves showed that the nomogram was of great clinical value.

CONCLUSION

A nomogram based on perioperative risk factors such as preoperative biliary drainage, CRP, DB and AKP could simply and accurately predict the risk degree of PAI in patients undergoing PD.

Prognostic Value of Pretreatment Controlling Nutritional Status Score for Patients With Pancreatic Cancer: A Meta-Analysis

Background

Previous studies have explored the prognostic value of the pretreatment Controlling Nutritional Status (CONUT) score of patients with pancreatic cancer. However, the results of those studies were inconsistent. We used meta-analysis to investigate the impact of the CONUT score on the prognosis for patients with pancreatic cancer.

Methods

We thoroughly searched the PubMed, Web of Science, Embase, and Cochrane Library databases for relevant articles from inception to November 19, 2021. Combined hazard ratios (HRs) and 95% confidence intervals (95% CIs) were used to estimate the prognostic value of the CONUT score with respect to survival duration. The pooled odds ratios (ORs) and 95% CIs were used to estimate the correlation between the CONUT score and clinical characteristics.

Results

The database search found seven studies with 2,294 patients for inclusion in this meta-analysis. A high CONUT score was significantly associated with poor overall survival (OS) (HR = 1.56, 95% CI = 1.13–2.16, p = 0.007), but not with recurrence-free survival (RFS) (HR = 1.47, 95% CI = 0.97–2.23, p = 0.072) of patients with pancreatic cancer. Moreover, there was a significant association between an elevated CONUT score and male patients (OR = 1.34, 95% CI = 1.03–1.75, p = 0.029). However, there was no significant association between the CONUT score and the clinical stage (OR = 1.11, 95% CI = 0.46–2.71, p = 0.576), lymph node metastasis (OR = 0.73, 95% CI = 0.39–1.36, p = 0.323), lymphatic vessel invasion (OR = 0.84, 95% CI = 0.55–1.28, p = 0.411), invasion of the portal vein system (OR = 1.04, 95% CI = 0.51–2.13, p = 0.915), and nerve plexus invasion (OR = 1.22, 95% CI = 0.83–1.80, p = 0.318) in patients with pancreatic cancer.

Conclusions

The results of our meta-analysis indicate that a high CONUT score predicts a poor OS in patients with pancreatic cancer. The CONUT score may be an effective prognostic factor in pancreatic cancer in clinical practice.

Predictive factors of survival in patients with borderline resectable pancreatic cancer who received neoadjuvant therapy

BACKGROUND/OBJECTIVES

This study aimed to develop the prognostic score (PS) based on clinical factors to stratify the prognosis in borderline resectable pancreatic cancer (BRPC) patients treated with neoadjuvant therapy (NAT).

METHODS

This retrospective study included 57 BRPC patients who received NAT between April 2012 and December 2017. A score was assigned to each prognostic factor available before and after NAT, according to their β coefficients.

RESULTS

Multivariate analysis identified the following six prognostic factors, and scores were assigned as follows: being a familial PC patient (HR 4.98, p = 0.029), post-NAT CA19-9 ≥37 U/ml (HR 3.08, p = 0.020), reduction rate of CA19-9 <70% (HR 3.71, p = 0.008), pre-NAT neutrophil-to-lymphocyte ratio ≥2.8 (HR 4.32, p = 0.003), and non-resection (HR 3.98, p = 0.009) were scored as 1; and post-NAT albumin-to-globulin ratio <1.33 (HR 8.31, p < 0.001) was scored as 2. The PS was calculated by summing the scores assigned to each prognostic factor. Patients were then classified into three risk groups (low- [0-1 points], moderate- [2-3 points], and high-risk [4-6 points] groups). Median overall survival in the low-, moderate-, and high-risk groups were not reached, 37.5 months, and 11.8 months, respectively, and there were significant differences in survival among the three groups (p < 0.01 in each group).

CONCLUSIONS

This study showed that the PS may be useful for predicting the prognosis of BRPC patients treated with NAT.

The decreased platelet-to-lymphocyte ratio could predict a good prognosis in patients with oligometastatic colorectal cancer: a single-center cohort retrospective study

BACKGROUND

Inflammation markers have an important effect on tumor proliferation, invasion, and metastasis. Oligometastatic disease (OMD) is an intermediate state between widespread metastases and locally confined disease, where curative strategies may be effective for some patients. We aimed to explore the predictive value of inflammatory markers in patients with oligometastatic colorectal cancer (OMCC) and build a nomogram to predict the prognosis of these patients.

METHODS

Two hundred nine patients with OMCC were retrospectively collected in this study. The Kaplan-Meier survival curves and Cox regression analysis were used to estimate overall survival (OS) and progression-free survival (PFS). A multivariate Cox analysis model was utilized to establish the nomogram. The concordance index (C-index), calibration curve, and receiver operating characteristics (ROC) were established to verify the validity and accuracy of the prediction model.

RESULTS

According to the multivariate analysis, decreased platelet-to-lymphocyte ratio (PLR) might independently improve OS in patients with OMCC (HR = 2.396, 95% CI 1.391-4.126, P = 0.002). Metastases of extra-regional lymph nodes indicated poor OS (HR = 2.472, 95% CI 1.247-4.903, P = 0.010). While the patients with early N stage had better OS (HR = 4.602, 95% CI 2.055-10.305, P = 0.001) and PFS (HR = 2.100, 95% CI 1.364-3.231, P = 0.007). Primary tumor resection (HR = 0.367, 95% CI 0.148-0.908, P = 0.030) and lower fibrinogen (HR = 2.254, 95% CI 1.246-4.078, P = 0.007) could significantly prolong the OS in patients with OMCC. PLR, metastases of extra-regional lymph nodes, N stage, primary tumor resection, and fibrinogen were used to make up the nomogram. The C-index and area under the curve (AUC) of the ROC in nomogram were 0.721 and 0.772 respectively for OS, showed good consistency between predictive probability of OS and actual survival.

CONCLUSIONS

Decreased PLR could predict a good prognosis in patients with OMCC. The nomogram including inflammatory factors and clinicopathological markers was credible and accurate to predict survivals in patients with OMCC.

Clinical importance of preoperative red-cell volume distribution width as a prognostic marker in patients undergoing radical surgery for pancreatic cancer

Background and purpose

A new noninvasive biomarker is being sought to predict the prognosis of patients with pancreatic cancer. Red-cell volume distribution width (RDW), a descriptive parameter for erythrocyte variation, has been shown to have prognostic value for some tumor types. Our purpose was to assess the RDW value to predict the prognosis of patients with pancreatic cancer.

Methods

The subjects of this retrospective study were 792 patients who underwent radical surgery for pancreatic cancer, divided into high-RDW and low-RDW groups based on receiver operating characteristic (ROC) curve analysis (15.6%). The controlling nutritional status (CONUT) score was used to assess preoperative nutritional status. Statistical analysis was conducted to investigate the differences between the high and low RDW groups, and to explore the possibility of the RDW being used as prognostic predictor for patients with pancreatic cancer.

Results

The immune-nutritional status was worse in the high-RDW group than in the low-RDW group. The high-RDW group patients also had a poorer prognosis. Risk factor analysis showed that the RDW could be an independent risk factor for pancreatic cancer.

Conclusions

The RDW is associated with immune-nutritional status in pancreatic cancer patients and can be used as an independent prognostic factor for their postoperative survival.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00595-021-02374-7.

CT Radiomics-Based Preoperative Survival Prediction in Patients With Pancreatic Ductal Adenocarcinoma

OBJECTIVE. Pancreatic ductal adenocarcinoma (PDAC) is often a lethal malignancy with limited preoperative predictors of long-term survival. The purpose of this study was to evaluate the prognostic utility of preoperative CT radiomics features in predicting postoperative survival of patients with PDAC. MATERIALS AND METHODS. A total of 153 patients with surgically resected PDAC who underwent preoperative CT between 2011 and 2017 were retrospectively identified. Demographic, clinical, and survival information was collected from the medical records. Survival time after the surgical resection was used to stratify patients into a low-risk group (survival time > 3 years) and a high-risk group (survival time < 1 year). The 3D volume of the whole pancreatic tumor and background pancreas were manually segmented. A total of 478 radiomics features were extracted from tumors and 11 extra features were computed from pancreas boundaries. The 10 most relevant features were selected by feature reduction. Survival analysis was performed on the basis of clinical parameters both with and without the addition of the selected features. Survival status and time were estimated by a random survival forest algorithm. Concordance index (C-index) was used to evaluate performance of the survival prediction model. RESULTS. The mean age of patients with PDAC was 67 ± 11 (SD) years. The mean tumor size was 3.31 ± 2.55 cm. The 10 most relevant radiomics features showed 82.2% accuracy in the classification of high-risk versus low-risk groups. The C-index of survival prediction with clinical parameters alone was 0.6785. The addition of CT radiomics features improved the C-index to 0.7414. CONCLUSION. Addition of CT radiomics features to standard clinical factors improves survival prediction in patients with PDAC.

Nomograms to predict the prognosis in locally advanced oral squamous cell carcinoma after curative resection

BACKGROUND

Oral squamous cell carcinoma (OSCC) is the dominant histologic type of oral cancer. Locally advanced OSCC remains a major therapeutic challenge. Our study aimed to develop and validate nomograms predicting survival prognosis in patients with locally advanced oral squamous cell carcinoma (OSCC) after curative resection.

METHODS

A total of 269 consecutive patients with primary OSCC who received curative resection between September 2007 and March 2020 were retrospectively enrolled in our study. Patients were randomly assigned to the training cohort (n = 201) or the validation cohort (n = 68). Multivariate Cox regression analyses were conducted to determine independent prognostic factors for overall survival (OS) and cancer specific survival (CSS) in the training set, which were used to develop nomogram models estimating 3-, and 5-year OS and CSS. We also evaluated the nomograms using concordance indices (c-index), calibration curves, and decision curve analyses (DCA), and compared those with the AJCC 8th staging system. The results were externally validated in the validation cohort.

RESULTS

Age, Kaplan-Feinstein (KFI) index, pT, the number of positive nodes and systemic inflammatory index (SII) were significant prognostic predictors for OS and CSS. The OS nomogram had c-index values of 0.712 in the training set and 0.697 in the validation set, while the CSS nomogram exhibited c-index values of 0.709 in the training set and 0.675 in the validation set. These data were superior to those of AJCC 8th staging system, suggesting high discriminative ability of the nomograms. Calibration curves exhibited good agreement between observed and predicted survival. DCA curves indicated the nomograms were with potential clinical usefulness. These results were validated in the validation set.

CONCLUSIONS

The novel nomograms incorporating clinically available characteristics for OS and CSS prediction were developed in the locally advanced OSCC patients after curative surgery. Validation revealed good discrimination and calibration, indicating the clinical utility of the nomograms in the individualized prognosis prediction of locally advanced OSCC after curative surgery.

Prognostic Nomogram for Resected Pancreatic Adenocarcinoma: A TRIPOD-Compliant Retrospective Long-Term Survival Analysis

BACKGROUND

Prognostic prediction had been widely used in various cancer entities, from early screening to end-stage patient caring. Currently, there is hardly any well-validated nomogram which exists for long-term survival prediction in pancreatic adenocarcinoma (PC) patients in a post-surgery setting. Our objectives are to identify possible prognostic factors in PC patients following radical resection and to develop a prognostic nomogram based on independent survival predictors.

METHODS

From 2009 to 2014, a total of 432 PC patients who underwent curative intended surgeries with complete follow-up data were included in this current retrospective long-term survival analysis. Clinicopathological data were extracted from medical records, and all missing values (percentage 0.9-8.3%) were imputed five times with the "PMM" method. Cox proportional hazards models were utilized. A nomogram was formulated based on results from the multivariate regression model so as to predict OS at 1-, 2- and 3-year as well as median OS. Validations, including discrimination and calibration, were carried out with 1000 bootstrap resamples. External validation was conducted in order to verify the accuracy of our nomogram at 1 and 2 years by utilizing the clinicopathological data of 122 PC patients who underwent curative intended surgeries in 2015 in our centre.

RESULTS

Age, abdominal pain, back pain, tumour location, preoperative neutrophil-lymphocyte ratio, preoperative CA19-9, tumour differentiation, microscopic nerve invasion, microscopic vascular invasion, T stage, lymph node ratio, M stage and adjuvant chemotherapy were all assembled into nomogram. The concordance index (C-index) of internal and external validation was 0.702 and 0.688, respectively. The C-index of the TNM staging system was 0.572 (P < 0.001 vs. nomogram).

CONCLUSION

Our prognostic nomogram based on clinicopathological parameters shows good performance in long-term survival prediction in PC patients following radical surgery and could play a role in further clinical utilization.

Preoperative Platelet to Lymphocyte Ratio as a Prognostic Factor for Resectable Pancreatic Cancer: A Systematic Review and Meta-Analysis

INTRODUCTION

Various inflammatory markers have been investigated for a prognostic role in patients with resectable pancreatic cancer. However, the value of preoperative platelet to lymphocyte ratio (PLR) remains controversial. We performed a systematic review and meta-analysis of PLR as a preoperative prognostic factor for resectable pancreatic cancer.

MATERIAL AND METHODS

Systematic literature search was conducted for studies assessing PLR influence as a preoperative prognostic factor in resectable pancreatic cancer patients. Random-effects model was applied for pooling hazard ratios and 95% confidence intervals related to overall survival (OS) and disease-free survival (DFS).

RESULTS

Fourteen articles with 2,743 patients were included in the study. According to the analysis, high PLR had no correlation with decreased OS. Due to high heterogeneity among studies, subgroup analysis was performed. Better OS was associated with low PLR in Asian patients, patients with mixed type of operation performed, and patients with preoperative PLR ≤150. Low PLR was associated with significantly better DFS.

CONCLUSIONS

PLR is a predictive factor of better DFS in patients with resectable pancreatic cancer. However, available evidence does not support PLR as a reliable prognostic factor for OS.

Score for the Overall Survival Probability of Patients With Pancreatic Adenocarcinoma of the Body and Tail After Surgery: A Novel Nomogram-Based Risk Assessment

Pancreatic adenocarcinoma of the body and tail often has a dismal prognosis and lacks a specific prognostic stage. The aim of this study was to construct a nomogram for predicting survival of patients with pancreatic adenocarcinoma of the body and tail after surgery. Data of patients were selected from the Surveillance, Epidemiology, and End Results (SEER) database and from medical records of Sun Yat-sen University Cancer Center (SYSUCC). In a multivariate analysis for overall survival (OS), the following six variables were identified as independent predictors and incorporated into the nomogram: age, tumor differentiation, tumor size, lymph node ratio (LNR), and chemotherapy. A nomogram was built based on independent risk predictors. The concordance index (C-index) for nomogram, Tumor-Node-Metastasis (TNM) 7th and 8th stage system were 0.775 [95% confidence interval (CI), 0.731–0.819], 0.617 (95%CI, 0.575–0.659), and 0.632 (95%CI, 0.588–0.676), respectively. The calibrated nomogram predicted survival rates which closely corresponded to the actual survival rates. Furthermore, the values of the area under receiver operating characteristic (ROC) curves (AUC) of the nomograms were higher than those of the TNM 7th or 8th stage system in predicting 1-, 2-, and 3-year survival of patients in training and external validation cohorts. The well-calibrated nomogram could be used to predict prognosis for patients with pancreatic adenocarcinoma of the body and tail after surgery.

Development, Validation and Comparison of Artificial Neural Network Models and Logistic Regression Models Predicting Survival of Unresectable Pancreatic Cancer

Background: Prediction models for the overall survival of pancreatic cancer remain unsatisfactory. We aimed to explore artificial neural networks (ANNs) modeling to predict the survival of unresectable pancreatic cancer patients.

Methods: Thirty-two clinical parameters were collected from 221 unresectable pancreatic cancer patients, and their prognostic ability was evaluated using univariate and multivariate logistic regression. ANN and logistic regression (LR) models were developed on a training group (168 patients), and the area under the ROC curve (AUC) was used for comparison of the ANN and LR models. The models were further tested on the testing group (53 patients), and k-statistics were used for accuracy comparison.

Results: We built three ANN models, based on 3, 7, and 32 basic features, to predict 8 month survival. All 3 ANN models showed better performance, with AUCs significantly higher than those from the respective LR models (0.811 vs. 0.680, 0.844 vs. 0.722, 0.921 vs. 0.849, all p < 0.05). The ability of the ANN models to discriminate 8 month survival with higher accuracy than the respective LR models was further confirmed in 53 consecutive patients.

Conclusion: We developed ANN models predicting the 8 month survival of unresectable pancreatic cancer patients. These models may help to optimize personalized patient management.

Prognostic value of nomogram based on pre-treatment inflammatory markers in patients with pulmonary-only synchronous metastases from colorectal cancer

BACKGROUND

The inflammatory markers were important factors affecting proliferation, invasiveness and metastasis of tumors. However, the prognostic value of inflammatory markers and related nomogram in patients with pulmonary-only synchronous metastases (POSM) from colorectal cancer has not been reported.

MATERIALS AND METHODS

A total of 98 POSM patients undergoing resection of colorectal cancer were included. The Kaplan-Meier survival analysis and Cox regression analysis were used to estimate the overall survival (OS) and progression-free survival (PFS). The nomogram was built based on multivariate Cox analysis model and evaluated by calibration curve, concordance index (C-index) and receiver operating characteristics (ROC).

RESULTS

The multivariate analysis showed that neutrophil-to-lymphocyte ratio (NLR) was prognostic factor of OS (HR = 2.73, 95%CI: 1.54-4.84) and PFS (HR = 1.78, 95%CI: 1.13-2.82). Elevated alkaline phosphatase (ALP) also was predictor of poor OS (HR = 1.99, 95%CI: 1.12-3.51) and PFS (HR = 2.23, 95%CI: 1.40-3.55). The early N stage and solitary pulmonary metastases had significant survival benefit for OS, while advanced T stage was independently related with worse PFS. Nomograms were consist of above significant risk factors, with C-index of 0.742 for OS and 0.656 for PFS. The calibration curves showed non-significant deviations between predicted and actual probability of OS and PFS.

CONCLUSIONS

Pre-treatment NLR and ALP were independently associated with OS and PFS in POSM patients. The nomograms involving inflammatory markers and clinicopathological factors were practical in predicting survival, which may help to guide use of therapeutic strategy and cancer surveillance.

HisCoM-PAGE: Hierarchical Structural Component Models for Pathway Analysis of Gene Expression Data

Although there have been several analyses for identifying cancer-associated pathways, based on gene expression data, most of these are based on single pathway analyses, and thus do not consider correlations between pathways. In this paper, we propose a hierarchical structural component model for pathway analysis of gene expression data (HisCoM-PAGE), which accounts for the hierarchical structure of genes and pathways, as well as the correlations among pathways. Specifically, HisCoM-PAGE focuses on the survival phenotype and identifies its associated pathways. Moreover, its application to real biological data analysis of pancreatic cancer data demonstrated that HisCoM-PAGE could successfully identify pathways associated with pancreatic cancer prognosis. Simulation studies comparing the performance of HisCoM-PAGE with other competing methods such as Gene Set Enrichment Analysis (GSEA), Global Test, and Wald-type Test showed HisCoM-PAGE to have the highest power to detect causal pathways in most simulation scenarios.

A systematic review of methodological quality of model development studies predicting prognostic outcome for resectable pancreatic cancer

OBJECTIVES

To assess the methodological quality of prognostic model development studies pertaining to post resection prognosis of pancreatic ductal adenocarcinoma (PDAC).

DESIGN/SETTING

A narrative systematic review of international peer reviewed journals DATA SOURCE: Searches were conducted of: MEDLINE, Embase, PubMed, Cochrane database and Google Scholar for predictive modelling studies applied to the outcome of prognosis for patients with PDAC post resection. Predictive modelling studies in this context included prediction model development studies with and without external validation and external validation studies with model updating. Data was extracted following the Checklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies (CHARMS) checklist.

PRIMARY AND SECONDARY OUTCOME MEASURES

Primary outcomes were all components of the CHARMS checklist. Secondary outcomes included frequency of variables included across predictive models.

RESULTS

263 studies underwent full text review. 15 studies met the inclusion criteria. 3 studies underwent external validation. Multivariable Cox proportional hazard regression was the most commonly employed modelling method (n=13). 10 studies were based on single centre databases. Five used prospective databases, seven used retrospective databases and three used cancer data registry. The mean number of candidate predictors was 19.47 (range 7 to 50). The most commonly included variables were tumour grade (n=9), age (n=8), tumour stage (n=7) and tumour size (n=5). Mean sample size was 1367 (range 50 to 6400). 5 studies reached statistical power. None of the studies reported blinding of outcome measurement for predictor values. The most common form of presentation was nomograms (n=5) and prognostic scores (n=5) followed by prognostic calculators (n=3) and prognostic index (n=2).

CONCLUSIONS

Areas for improvement in future predictive model development have been highlighted relating to: general aspects of model development and reporting, applicability of models and sources of bias.

TRIAL REGISTRATION NUMBER

CRD42018105942.

Clinical significance of traditional clinical parameters and inflammatory biomarkers for the prognosis of patients with spinal chondrosarcoma: a retrospective study of 150 patients in a single center

BACKGROUND

To investigate the clinical significance of five inflammatory biomarkers and conventional clinical parameters in prognostic prediction of spinal chondrosarcoma.

METHODS

Univariate and multivariate analyses were performed to investigate independent prognostic factors for recurrence and death of patients with spinal chondrosarcoma. Disease-free survival (DFS) and overall survival (OS) were estimated by Kaplan-Meier curve, and differences were analyzed by log-rank test. The optimal cutoff values for NLR, PLR, LMR, and CAR were determined by X-tile program.

RESULTS

The optimal cutoff value for NLR, PLR, LMR, AGR, and CAR was 2.7, 200, 3.0, 1.5, and 0.2, respectively. Of the 150 patients included, recurrence was detected in 105 patients, and death occurred in 78 patients. Multivariate analysis indicated that Tomita I-III, total resection, and CAR < 0.2 were significantly associated with longer DFS. Meanwhile, preoperative Frankel score D-E, total resection, and CAR < 0.2 were favorable prognostic factors for OS. Subtype analysis showed that only total resection was an independent prognostic factor for DFS of recurrent spinal chondrosarcoma.

CONCLUSION

Total resection could significantly reduce the recurrence rate of spinal chondrosarcoma and improve OS of chondrosarcoma patients. Tomita classification I-III was a favorable factor for DFS, and preoperative Frankel score A-C was an adverse prognostic factor for OS. CAR was the most robust prognostic indicator with a discriminatory ability as compared with other inflammatory indicators. These slides can be retrieved under Electronic Supplementary Material.

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.

A sample model established by S-index predicting overall survival after curative resection of primary hepatocellular carcinoma

Purpose

Prognostic prediction after curative resection of primary hepatocellular carcinoma (PHCC) remains an arduous task. The S-index calculated from γ-glutamyl transpeptidase, albumin, and platelets is reported to predict the severity of liver fibrosis. We constructed a nomogram for predicting the survival probability of PHCC based on a new indicator, the S-index, combined with other routine clinical parameters.

Patients and methods

We selected 490 patients with PHCC postradical surgery at the First Affiliated Hospital of Wenzhou Medical University between January 2007 and January 2014. The subjects were randomly allocated into the training cohort and the validation cohort in the ratio 7:3 by the digital method. Important variables screened by univariate analysis were included in multivariate analysis to obtain independent risk factors for predicting the prognosis of PHCC. The construction of the nomogram was based on Cox proportional hazard regression models. The concordance index (C-index) was used in the nomogram for evaluating the model performance for prognosis. We drew time-dependent receiver operating characteristic curves to compare our model with other staging systems.

Results

The nomogram based on six independent risk factors after multivariate analyses had good predictive power after radical surgery of PHCC. In the training and validation groups, the C-index of the nomogram was highly consistent for evaluating survival from PHCC. Compared with the traditional scoring system, the areas under time-dependent receiver operating characteristic curves were 0.7382, 0.7293, and 0.7520 for 1-, 3-, and 5-year overall survival, respectively. In summary, the nomogram showed excellent results in terms of prognosis of PHCC.

Conclusion

Based on the S-index and the other clinical indicators, we developed a precise nomogram that predicts the survival probability of patients with PHCC after radical surgery. This tool can provide effective information for surgeons and patients.

A Prognostic Nomogram for Disease-Specific Survival in Patients with Pancreatic Ductal Adenocarcinoma of the Head of the Pancreas Following Pancreaticoduodenectomy

BACKGROUND This study developed and validated a nomogram to predict patient prognosis for pancreatic ductal adenocarcinoma (PDAC) of the head of the pancreas following pancreaticoduodenectomy. MATERIAL AND METHODS Retrospective data were obtained from 4,383 patients with PDAC of the head of the pancreas who underwent pancreaticoduodenectomy between 2004-2013 from 11 Registries Research Data of the Surveillance, Epidemiology,and End Results (SEER) database. Cox proportional hazards model was used to identify independent risk factors. The predictive accuracy of the nomogram was determined by the concordance index (C-index) and calibration curve. The results were externally validated by comparison with data from 1,743 patients from 7 other Registries Research Data. RESULTS Of the 4,383 patients in the training dataset, median disease-specific survival (DSS) was 17.0 months (range, 1.0-131 months), and postoperative 1-year, 3-year, and 5-year DSS rates were 70.3%, 26.1%, and 16.8%, respectively. Multivariate analysis showed that patient sex, age, tumor grade, regional lymph nodes examined, positive regional lymph nodes, tumor size, extent of local invasion, and tumor metastases were independent risk factors for DSS. The C-index of the internal validation dataset for prediction of DSS was 0.64 (95% CI, 0.63-0.65), which was superior to the American Joint Committee on Cancer (AJCC) staging, 0.57 (95% CI, 0.56-0.58) (P<0.001). The 5-year DSS rates and median DSS time for patients in the low-risk group were significantly greater compared with high-risk group (P<0.001). CONCLUSIONS A validated prognostic disease-specific nomogram for patient survival in PDAC of the head of the pancreas following pancreaticoduodenectomy was developed.

A nomogram for predicting survival of patients with locally advanced pancreatic cancer treated with chemoradiotherapy

BACKGROUND

We developed a nomogram for predicting survival of patients with locally advanced pancreatic cancer (LAPC) after concurrent chemoradiotherapy (CRT) using ¹⁸F-flurodeoxyglucose-positron emission tomography (FDG-PET) parameters and CA 19-9 levels.

METHODS

Based on 426 patients with LAPC who received concurrent CRT between 2004 and 2015, we investigated significant prognostic factors for survival to build a nomogram, including the maximum standardized uptake value (SUV_max) and CA 19-9 levels. Predictive accuracy and discriminative ability were then measured.

RESULTS

Median progression-free survival and overall survival (OS) were 9.4 and 15.4 months, respectively, at a median 15-month follow-up. High-dose radiation (EQD2, ≥61 Gy), initial SUV_max <3.5 and CA 19-9 ≤400 U/mL, and surgical resection after CRT were significantly related to prolonged OS by multivariate analysis (p < 0.05). A nomogram model for OS was established and showed good calibration and acceptable discrimination (c-index 0.656). Using the nomogram, 3 different prognosis groups could be identified with a median OS of 25, 15, and 11 months (p < 0.001).

CONCLUSION

A nomogram was developed with high-dose radiation (EQD2, ≥61 Gy), initial SUV_max <3.5, CA 19-9 ≤400 U/mL, and surgical resection after CRT for patients with LAPC. This will help in clinical decision-making and in selecting patients for CRT.

Prognostic Roles of Inflammatory Markers in Pancreatic Cancer: Comparison between the Neutrophil-to-Lymphocyte Ratio and Platelet-to-Lymphocyte Ratio

Background/Objectives

This meta-analysis is aimed at investigating the prognostic roles of the inflammatory markers neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) in patients with pancreatic cancer.

Methods

The correlations between high inflammatory marker expression levels and prognosis in 7105 patients with pancreatic cancer from 34 eligible studies were investigated. Additionally, subgroup analyses based on study location, tumor stage, treatment, and value cutoffs were performed.

Results

High NLR and PLR values were considered to be 2.0–5.0 and 150–200, respectively. Using a random-effects model, the estimated rates of high NLR and PLR were 0.379 (95% confidence interval [CI] 0.310–0.454) and 0.490 (95% CI 0.438–0.543), respectively. High NLRs were frequently found in patients with lower tumor stages and in those who underwent surgery. There were significant correlations between high NLR and PLR and poor survival rates (hazard ratio [HR] 1.737, 95% CI 1.502–2.009 and HR 1.143, 95% CI 1.037–1.259, resp.). Interestingly, the NLR and PLR had no prognostic value in patients who underwent chemoradiotherapy.

Conclusion

Taken together, our results showed that inflammatory markers are useful for predicting prognosis in patients with pancreatic cancer. The NLR is a more suitable parameter for predicting prognosis regardless of the patient's condition.

Survival Prediction in Pancreatic Ductal Adenocarcinoma by Quantitative Computed Tomography Image Analysis

BACKGROUND

Pancreatic cancer is a highly lethal cancer with no established a priori markers of survival. Existing nomograms rely mainly on post-resection data and are of limited utility in directing surgical management. This study investigated the use of quantitative computed tomography (CT) features to preoperatively assess survival for pancreatic ductal adenocarcinoma (PDAC) patients.

METHODS

A prospectively maintained database identified consecutive chemotherapy-naive patients with CT angiography and resected PDAC between 2009 and 2012. Variation in CT enhancement patterns was extracted from the tumor region using texture analysis, a quantitative image analysis tool previously described in the literature. Two continuous survival models were constructed, with 70% of the data (training set) using Cox regression, first based only on preoperative serum cancer antigen (CA) 19-9 levels and image features (model A), and then on CA19-9, image features, and the Brennan score (composite pathology score; model B). The remaining 30% of the data (test set) were reserved for independent validation.

RESULTS

A total of 161 patients were included in the analysis. Training and test sets contained 113 and 48 patients, respectively. Quantitative image features combined with CA19-9 achieved a c-index of 0.69 [integrated Brier score (IBS) 0.224] on the test data, while combining CA19-9, imaging, and the Brennan score achieved a c-index of 0.74 (IBS 0.200) on the test data.

CONCLUSION

We present two continuous survival prediction models for resected PDAC patients. Quantitative analysis of CT texture features is associated with overall survival. Further work includes applying the model to an external dataset to increase the sample size for training and to determine its applicability.

Prognostic value of pretreatment albumin/globulin ratio in digestive system cancers: A meta-analysis

The albumin/globulin ratio (AGR) has been widely reported to be a potential predictor of prognosis in digestive system cancers (DSCs), but convincing conclusions have not been made. Therefore, herein, we performed a meta-analysis of relevant studies regarding this topic to evaluate the prognostic value of AGR in patients with DSCs. Three databases, including PubMed, EMBase, and Web of science, were searched comprehensively for eligible studies through September 8, 2017. The outcomes of interest included overall survival (OS), disease-free survival (DFS), and cancer-specific survival (CSS). In our meta-analysis, pooled analysis of 13 studies with 9269 patients showed that a low AGR was significantly correlated with poor OS (HR = 1.94; 95% CI: 1.57-2.38; P <0.001). Five studies with 6538 participants involved DFS, and our pooled analysis of these studies also demonstrated that there was a significant association of a low AGR with worse DFS (HR = 1.49; 95% CI: 1.10 to 2.00; P < 0.001). In addition, only 2 studies referred to CSS, and we also detected a significant relationship between a low AGR and worse CSS from the results of our meta-analysis. In summary, a low pretreatment AGR was related to unfavorable survival in human digestive system cancers. A low pretreatment AGR may be a useful predictive prognostic biomarker in human digestive system cancers.

Pretreatment albumin to globulin ratio can serve as a prognostic marker in human cancers: a meta-analysis

PURPOSE

Our meta-analysis aims to investigate the prognostic role of pretreatment albumin to globulin ratio (AGR) in human cancers.

METHODS

Available databases were searched up to Sept 25th, 2017. Pooled hazard ratios (HRs) and risk ratio (RRs) with their corresponding 95% confidence intervals (CIs) were used to assess the prognostic impact of AGR on overall survival (OS)/disease-free survival (DFS)/progression-free survival (PFS) and 5-year mortality respectively.

RESULTS

Totally, 28 studies with 15 356 cancer patients were included. Our results demonstrated that low pretreatment AGR is associated with poor OS (HR=2.08, 95%CI:1.78-2.44, univariate results; HR=1.75, 95%CI:1.56-1.97, multivariate results), poor DFS (HR=1.96, 95%CI:1.48-2.59, univariate results; HR=1.64, 95%CI:1.26-2.14, multivariate results) and poor PFS (HR=1.89, 95%CI:1.61-2.22, univariate results; HR=1.66, 95%CI:1.32-2.0, multivariate results). Meanwhile, low pretreatment AGR is also associated with increased 5-year mortality (RR=2.12, 95%CI:1.48-3.03). Moreover, this significant correlation was not altered by stratified analysis according to publication times, sample sizes, patient origins, AGR cutoff values, cancer systems, treatment methods or HR sources.

CONCLUSION

Low pretreatment AGR is associated with poor prognosis in human cancers, and AGR should be used as a prognostic marker during cancer therapy.