Survival prediction for gallbladder carcinoma after curative resection: Comparison of nomogram and Bayesian network models

What eliminates the chance for cure: a multi-center evaluation on 10-year follow-up of gallbladder cancer after surgical resection

Curative resection stands as the sole potential cure for gallbladder cancer (GBC); nevertheless, a dearth of knowledge persists regarding long-term follow-up data and prognostic factors that hinder achieving a cure post-surgery. A retrospective cohort study was conducted by analyzing pathologically confirmed initial resections for GBC between 2000 and 2013 across three Chinese medical centers. The concept of observed cure refers to a 10-year survival period devoid of any disease recurrence. Employing a semiparametric proportional hazards mixture cure model enabled the identification of clinicopathological factors impeding a cure for GBC post-surgery. In our current study, a total of 331 patients were included, with a follow-up period exceeding a decade. The median overall survival (OS) was recorded at 31.6 months, with 39 patients (11.78%) achieving a 10-year OS, classified as 10-year survivors. Within this subset, 36 patients reached a 10-year relapse-free survival, denoting cure, and yielding an observed cure rate of 10.88%. Notably, factors such as combined surgical resection involving invaded organs, positive lymph node metastasis, and R1 resection (below 1%) were identified as virtually precluding a cure. Additionally, patients with T3-4 stage, hepatic invasion, advanced AJCC stage or poor tumor differentiation exhibited a low likelihood of achieving cure (below 5%). The discovery of these prognostic factors holds significant value in tailoring individualized treatment strategies and enhancing clinical decision-making processes.

Bayesian network-based survival prediction model for patients having undergone post-transjugular intrahepatic portosystemic shunt for portal hypertension

BACKGROUND

Portal hypertension (PHT), primarily induced by cirrhosis, manifests severe symptoms impacting patient survival. Although transjugular intrahepatic portosystemic shunt (TIPS) is a critical intervention for managing PHT, it carries risks like hepatic encephalopathy, thus affecting patient survival prognosis. To our knowledge, existing prognostic models for post-TIPS survival in patients with PHT fail to account for the interplay among and collective impact of various prognostic factors on outcomes. Consequently, the development of an innovative modeling approach is essential to address this limitation.

AIM

To develop and validate a Bayesian network (BN)-based survival prediction model for patients with cirrhosis-induced PHT having undergone TIPS.

METHODS

The clinical data of 393 patients with cirrhosis-induced PHT who underwent TIPS surgery at the Second Affiliated Hospital of Chongqing Medical University between January 2015 and May 2022 were retrospectively analyzed. Variables were selected using Cox and least absolute shrinkage and selection operator regression methods, and a BN-based model was established and evaluated to predict survival in patients having undergone TIPS surgery for PHT.

RESULTS

Variable selection revealed the following as key factors impacting survival: age, ascites, hypertension, indications for TIPS, postoperative portal vein pressure (post-PVP), aspartate aminotransferase, alkaline phosphatase, total bilirubin, prealbumin, the Child-Pugh grade, and the model for end-stage liver disease (MELD) score. Based on the above-mentioned variables, a BN-based 2-year survival prognostic prediction model was constructed, which identified the following factors to be directly linked to the survival time: age, ascites, indications for TIPS, concurrent hypertension, post-PVP, the Child-Pugh grade, and the MELD score. The Bayesian information criterion was 3589.04, and 10-fold cross-validation indicated an average log-likelihood loss of 5.55 with a standard deviation of 0.16. The model's accuracy, precision, recall, and F1 score were 0.90, 0.92, 0.97, and 0.95 respectively, with the area under the receiver operating characteristic curve being 0.72.

CONCLUSION

This study successfully developed a BN-based survival prediction model with good predictive capabilities. It offers valuable insights for treatment strategies and prognostic evaluations in patients having undergone TIPS surgery for PHT.

Textbook outcome in gallbladder carcinoma after curative-intent resection: a 10-year retrospective single-center study

BACKGROUND

Textbook outcome (TO) can guide decision-making among patients and clinicians during preoperative patient selection and postoperative quality improvement. We explored the factors associated with achieving a TO for gallbladder carcinoma (GBC) after curative-intent resection and analyzed the effect of adjuvant chemotherapy (ACT) on TO and non-TO patients.

METHODS

A total of 540 patients who underwent curative-intent resection for GBC at the Department of Hepatobiliary Surgery of the First Affiliated Hospital of Xi'an Jiaotong University from January 2011 to December 2020 were retrospectively analyzed. Multivariable logistic regression was used to investigate the factors associated with TO.

RESULTS

Among 540 patients with GBC who underwent curative-intent resection, 223 patients (41.3%) achieved a TO. The incidence of TO ranged from 19.0% to 51.0% across the study period, with a slightly increasing trend over the study period. The multivariate analysis showed that non-TO was an independent risk factor for prognosis among GBC patients after resection (P =0.003). Age ≤60 years (P =0.016), total bilirubin (TBIL) level ≤34.1 μmol/L (P <0.001), well-differentiated tumor (P =0.008), no liver involvement (P <0.001), and T1-2 stage disease (P =0.006) were independently associated with achieving a TO for GBC after resection. Before and after propensity score matching (PSM), the overall survival outcomes of non-TO GBC patients who received ACT and those who did not were statistically significant; ACT improved the prognosis of patients in the non-TO group (P <0.050).

CONCLUSION

Achieving a TO is associated with a better long-term prognosis among GBC patients after curative-intent resection, and ACT can improve the prognosis of those with non-TO.

A Bayesian Network Prediction Model for Microvascular Invasion in Patients with Intrahepatic Cholangiocarcinoma: A Multi-institutional Study

BACKGROUND

Microvascular invasion (MVI) has been reported to be an independent prognostic factor of recurrence and poor overall survival in patients with intrahepatic cholangiocarcinoma (ICC). This study aimed to explore the preoperative independent risk factors of MVI and establish a Bayesian network (BN) prediction model to provide a reference for surgical diagnosis and treatment.

METHODS

A total of 531 patients with ICC who underwent radical resection between 2010 and 2018 were used to establish and validate a BN model for MVI. The BN model was established based on the preoperative independent variables. The ROC curves and confusion matrix were used to assess the performance of the model.

RESULTS

MVI was an independent risk factor for relapse-free survival (RFS) (P < 0.05). MVI has a correlation with postoperative recurrence, early recurrence (< 6 months), median RFS and median overall survival (all P < 0.05). The preoperative independent risk variables of MVI included obstructive jaundice, prognostic nutritional index, CA19-9, tumor size, and major vascular invasion, which were used to establish the BN model. The AUC of the BN model was 78.92% and 83.01%, and the accuracy was 70.85% and 77.06% in the training set and testing set, respectively.

CONCLUSION

The BN model established based on five independent risk variables for MVI is an effective and practical model for predicting MVI in patients with ICC.

Development and validation of a nomogram to predict overall survival in patients with incidental gallbladder cancer: A retrospective cohort study

Objective

The aim of this study was to develop and validate a nomogram to predict the overall survival of incidental gallbladder cancer.

Methods

A total of 383 eligible patients with incidental gallbladder cancer diagnosed in Shanghai Eastern Hepatobiliary Surgery Hospital from 2011 to 2021 were retrospectively included. They were randomly divided into a training cohort (70%) and a validation cohort (30%). Univariate and multivariate analyses and the Akaike information criterion were used to identify variables independently associated with overall survival. A Cox proportional hazards model was used to construct the nomogram. The C-index, area under time-dependent receiver operating characteristic curves and calibration curves were used to evaluate the discrimination and calibration of the nomogram.

Results

T stage, N metastasis, peritoneal metastasis, reresection and histology were independent prognostic factors for overall survival. Based on these predictors, a nomogram was successfully established. The C-index of the nomogram in the training cohort and validation cohort was 0.76 and 0.814, respectively. The AUCs of the nomogram in the training cohort were 0.8, 0.819 and 0.815 for predicting OS at 1, 3 and 5 years, respectively, while the AUCs of the nomogram in the validation cohort were 0.846, 0.845 and 0.902 for predicting OS at 1, 3 and 5 years, respectively. Compared with the 8th AJCC staging system, the AUCs of the nomogram in the present study showed a better discriminative ability. Calibration curves for the training and validation cohorts showed excellent agreement between the predicted and observed outcomes at 1, 3 and 5 years.

Conclusions

The nomogram in this study showed excellent discrimination and calibration in predicting overall survival in patients with incidental gallbladder cancer. It is useful for physicians to obtain accurate long-term survival information and to help them make optimal treatment and follow-up decisions.

A Bayesian network prediction model for gallbladder polyps with malignant potential based on preoperative ultrasound

BACKGROUND

It is important to identify gallbladder polyps (GPs) with malignant potential and avoid unnecessary cholecystectomy by constructing prediction model. The aim of the study is to develop a Bayesian network (BN) prediction model for GPs with malignant potential in a long diameter of 8-15 mm based on preoperative ultrasound.

METHODS

The independent risk factors for GPs with malignant potential were screened by χ² test and Logistic regression model. Prediction model was established and validated using data from 1296 patients with GPs who underwent cholecystectomy from January 2015 to December 2019 at 11 tertiary hospitals in China. A BN model was established based on the independent risk variables.

RESULTS

Independent risk factors for GPs with malignant potential included age, number of polyps, polyp size (long diameter), polyp size (short diameter), and fundus. The BN prediction model identified relationships between polyp size (long diameter) and three other variables [polyp size (short diameter), fundus and number of polyps]. Each variable was assigned scores under different status and the probabilities of GPs with malignant potential were classified as [0-0.2), [0.2-0.5), [0.5-0.8) and [0.8-1] according to the total points of [- 337, - 234], [- 197, - 145], [- 123, - 108], and [- 62,500], respectively. The AUC was 77.38% and 75.13%, and the model accuracy was 75.58% and 80.47% for the BN model in the training set and testing set, respectively.

CONCLUSION

A BN prediction model was accurate and practical for predicting GPs with malignant potential patients in a long diameter of 8-15 mm undergoing cholecystectomy based on preoperative ultrasound.

Analysis of Prognostic Factors of Rectal Cancer and Construction of a Prognostic Prediction Model Based on Bayesian Network

Background

The existing prognostic models of rectal cancer after radical resection ignored the relationships among prognostic factors and their mutual effects on prognosis. Thus, a new modeling method is required to remedy this defect. The present study aimed to construct a new prognostic prediction model based on the Bayesian network (BN), a machine learning tool for data mining, clinical decision-making, and prognostic prediction.

Methods

From January 2015 to December 2017, the clinical data of 705 patients with rectal cancer who underwent radical resection were analyzed. The entire cohort was divided into training and testing datasets. A new prognostic prediction model based on BN was constructed and compared with a nomogram.

Results

A univariate analysis showed that age, Carcinoembryonic antigen (CEA), Carbohydrate antigen19-9 (CA19-9), Carbohydrate antigen 125 (CA125), preoperative chemotherapy, macropathology type, tumor size, differentiation status, T stage, N stage, vascular invasion, KRAS mutation, and postoperative chemotherapy were associated with overall survival (OS) of the training dataset. Based on the above-mentioned variables, a 3-year OS prognostic prediction BN model of the training dataset was constructed using the Tree Augmented Naïve Bayes method. In addition, age, CEA, CA19-9, CA125, differentiation status, T stage, N stage, KRAS mutation, and postoperative chemotherapy were identified as independent prognostic factors of the training dataset through multivariate Cox regression and were used to construct a nomogram. Then, based on the testing dataset, the two models were evaluated using the receiver operating characteristic (ROC) curve. The results showed that the area under the curve (AUC) of ROC of the BN model and nomogram was 80.11 and 74.23%, respectively.

Conclusion

The present study established a BN model for prognostic prediction of rectal cancer for the first time, which was demonstrated to be more accurate than a nomogram.

A Novel Clinically Prognostic Stratification Based on Prognostic Nutritional Index Status and Histological Grade in Patients With Gallbladder Cancer After Radical Surgery

Purpose

Gallbladder carcinoma (GBC) is the most common malignancy of the biliary tract, with a 5-year survival rate of 5%. The prognostic models to predict the prognosis of patients with GBC remain controversial. Therefore, to construct a prognosis prediction of GBC, a retrospective cohort study was carried out to investigate the prognostic nutritional index and histological grade in the long-term outcome of patients with GBC after radical surgery (RS).

Methods

A retrospective study of a total of 198 patients with GBC who underwent surgical treatment were enrolled. The hematological indicators, imageological data, and perioperative clinical data were acquired for statistical analysis and poor prognosis model construction.

Results

Prognostic nutrition index (PNI) < 45.88, maximum tumor diameter (MTD) > 2.24 cm, and jaundice (JD) were all associated with a poor prognosis in multivariate logistic regression analysis. The prognosis prediction model was based on the three risk factors, which indicated a superior predictive ability in the primary cohort [area under the curve (AUC) = 0.951] and validation cohort (AUC = 0.888). In multivariate Cox regression analysis, poorly differentiation (PD) was associated with poor 3-year survival. In addition, Kaplan-Meier (KM) survival analysis suggested that GBC patients with high-risk scores and PD had a better prognosis after RS (p < 0.05), but there was no significant difference in prognosis for patients with non-poorly differentiation (NPD) or low-risk scores after RS (p > 0.05).

Conclusion

Our prediction model for GBC patients with prognosis evaluation is accurate and effective. For patients with PD and high-risk scores, RS is highly recommended; a simple cholecystectomy can also be considered for acceptance for patients with NPD or low-risk score. The significant findings provide a new therapeutic strategy for the clinical treatment of GBC.

A Novel Prognostic Nomogram for Gallbladder Cancer after Surgical Resection: A Single-Center Experience

Background

Gallbladder cancer (GBC), which accounts for more than 80% of biliary tract malignancies, has a poor prognosis with an overall 5-year survival less than 10%. The study aimed to identify risk factors and develop a predictive model for GBC following surgical resection.

Methods

98 GBC patients who underwent surgical resection from Guangdong Provincial People's Hospital were enrolled in the study. Cox-regression analysis was performed to identify significant prognostic factors. A nomogram was constructed and Harrell's concordance index, calibration plot, and decision cure analysis were used to evaluate the discrimination and calibration of the nomogram.

Results

Liver resection, tumor size, perineural invasion, surgical margin, and liver invasion were identified as independent risk factors for overall survival (OS) in GBC patients who underwent surgical resection. Based on the selected risk factors, a novel nomogram was constructed. The C-index of the nomogram was 0.777, which was higher than the American Joint Committee on Cancer (AJCC) staging system (0.724) and Nevin staging system (0.659). Decision cure analysis revealed that the nomogram had a better net benefit and the calibration curves for the 1-, 3-, and 5-year survival probabilities were also well matched with the actual survival rates. Lastly, high-risk GBC were stratified based on the scores of the nomogram and we found high-risk GBC were associated with both worse OS and disease-free survival (DFS).

Conclusion

We developed a nomogram showing a better predictive capacity for patients' survival of resected GBC than the AJCC staging systems. The established model may help to stratify high-risk GBC and facilitate decision-making in the clinic.