Constructing a risk prediction model for anastomotic leakage after esophageal cancer resection

Sex differences in survival following surgery for esophageal cancer: A systematic review and meta-analysis

The impact of sex on the prognosis of patients with esophageal cancer remains unclear. Evidence supports that sex- based disparities in esophageal cancer survival could be attributed to sex- specific risk exposures, such as age at diagnosis, race, socioeconomic status, smoking, drinking, and histological type. The aim of our study is to investigate the role of sex disparities in survival of patients who underwent surgery for esophageal cancer. A systematic review and meta-analysis of the existing literature in PubMed, EMBASE, and CENTRAL from December 1966 to February 2023, was held. Studies that reported sex-related differences in survival outcomes of patients who underwent esophagectomy for esophageal cancer were identified. A total of 314 studies were included in the quantitative analysis. Statistically significant results derived from 1-year and 2-year overall survival pooled analysis with Relative Risk (RR) 0.93 (95% Confidence Interval (CI): 0.90-0.97, I2 = 52.00) and 0.90 (95% CI: 0.85-0.95, I2 = 0.00), respectively (RR < 1 = favorable for men). In the postoperative complications analysis, statistically significant results concerned anastomotic leak and heart complications, RR: 1.08 (95% CI: 1.01-1.16) and 0.62 (95% CI: 0.52-0.75), respectively. Subgroup analysis was performed among studies with <200 and > 200 patients, histology types, study continent and publication year. Overall, sex tends to be an independent prognostic factor for esophageal carcinoma. However, unanimous results seem rather obscure when multivariable analysis and subgroup analysis occurred. More prospective studies and gender-specific protocols should be conducted to better understand the modifying role of sex in esophageal cancer prognosis.

Integration of IDPC Clustering Analysis and Interpretable Machine Learning for Survival Risk Prediction of Patients with ESCC

Precise forecasting of survival risk plays a pivotal role in comprehending and predicting the prognosis of patients afflicted with esophageal squamous cell carcinoma (ESCC). The existing methods have the problems of insufficient fitting ability and poor interpretability. To address this issue, this work proposes a novel interpretable survival risk prediction method for ESCC patients based on extreme gradient boosting improved by whale optimization algorithm (WOA-XGBoost) and shapley additive explanations (SHAP). Given the imbalanced nature of the data set, the adaptive synthetic sampling (ADASYN) is first used to generate the samples with high survival risk. Then, an improved clustering by fast search and find of density peaks (IDPC) algorithm based on cosine distance and K nearest neighbors is used to cluster the patients. Next, the prediction model for each cluster is obtained by WOA-XGBoost and the constructed model is visualized with SHAP to uncover the factors hidden in the structured model and improve the interpretability of the black-box model. Finally, the effectiveness of the proposed scheme is demonstrated by analyzing the data collected from the First Affiliated Hospital of Zhengzhou University. The results of the analysis reveal that the proposed methodology exhibits superior performance, as indicated by the area under the receiver operating characteristic curve (AUROC) of 0.918 and accuracy of 0.881.

Development of nomograms predictive of anastomotic leakage in patients before minimally invasive McKeown esophagectomy

Purpose

The present study aims to identify factors related to anastomotic leakage before esophagectomy and to construct a prediction model.

Methods

A retrospective analysis of 285 patients who underwent minimally invasive esophagectomy (MIE). An absolute shrinkage and selection operator was applied to screen the variables, and predictive models were developed using binary logistic regression.

Results

A total of 28 variables were collected in this study. LASSO regression analysis, combined with previous literature and clinical experience, finally screened out four variables, including aortic calcification, heart disease, BMI, and FEV1. A binary logistic regression was conducted on the four predictors, and a prediction model was established. The prediction model showed good discrimination and calibration, with a C-statistic of 0.67 (95% CI, 0.593-0.743), a calibration curve fitting a 45° slope, and a Brier score of 0.179. The DCA demonstrated that the prediction nomogram was clinically useful. In the internal validation, the C-statistic still reaches 0.66, and the calibration curve has a good effect.

Conclusions

When patients have aortic calcification, heart disease, obesity, and a low FEV1, the risk of anastomotic leakage is higher, and relevant surgical techniques can be used to prevent it. Therefore, the clinical prediction model is a practical tool to guide surgeons in the primary prevention of anastomotic leakage.

Does Intraoperative Blood Loss Affect the Short-Term Outcomes and Prognosis of Gastric Cancer Patients After Gastrectomy? A Meta-Analysis

Purpose

The purpose of the current meta-analysis was to analyze whether intraoperative blood loss (IBL) influenced the complications and prognosis of gastric cancer patients after gastrectomy.

Methods

We systematically searched the PubMed, Embase and Cochrane library databases on November 29, 2021. The Newcastle-Ottawa scale was used to evaluate the quality of included studies. This meta-analysis uses RevMan 5.3 for data analysis.

Results

A total of nine retrospective studies were included in this meta-analysis, involving 4653 patients. In terms of short-term outcomes, the Larger IBL group has a higher complication rate (OR = 1.94, 95% CI, 1.44 to 2.61, P < 0.0001) and a longer operation time (OR = 77.60, 95% CI, 41.95 to 113.25, P < 0.0001) compared with the smaller IBL group, but the Larger IBL group had higher total retrieved lymph nodes (OR = 3.68, 95% CI, 1.13 to 6.24, P = 0.005). After pooling up all the HRs, the Larger IBL group has worse overall survival (OS) (HR = 1.80, 95% CI, 1.27 to 2.56, P = 0.001) and disease-free survival (DFS) (HR = 1.48, 95% CI, 1.28 to 1.72, P < 0.00001).

Conclusion

Larger IBL increased operation time and postoperative complications, and decreased OS and DFS of gastric cancer patients. Therefore, surgeons should be cautious about IBL during operation.

Albumin and Derived Neutrophil-to-Lymphocyte Ratio is a Novel Prognostic Factor for Patients with Esophageal Squamous Cell Carcinoma

BACKGROUND

Multidisciplinary treatment combining neoadjuvant treatment (NAT) and surgery has slightly improved the prognosis of patients with esophageal squamous cell carcinoma (ESCC). Although various biomarkers targeting nutrition and inflammation are associated with cancer prognosis, most studies have focused on conditions prior to NAT. Developing real-time and sensitive biomarkers that monitor changes in systemic conditions during NAT is important. We established a novel nutritional and inflammatory index, represented as the albumin to derived neutrophil-to-lymphocyte ratio (Alb-dNLR), and calculated the change in Alb-dNLR (ΔAlb-dNLR) during neoadjuvant chemotherapy (nCT). In this study, we aimed to evaluate whether ΔAlb-dNLR is associated with prognosis in patients with ESCC.

METHODS

We investigated 172 patients who underwent nCT before esophagectomy between April 2010 and March 2018. The dNLR was calculated as the ratio of neutrophil count to (white blood cell count - neutrophil count), Alb-dNLR was calculated by dividing the serum albumin level by the dNLR, and ΔAlb-dNLR was evaluated by dividing the post-Alb-dNLR by the pre-Alb-dNLR. Patients were divided into 'high' and 'low' groups according to the ΔAlb-dNLR.

RESULTS

Thirty-nine patients (22.7%) had a low ΔAlb-dNLR (≤ 0.8), and the 5-year overall survival (OS) rates in patients with low and high ΔAlb-dNLR were 38.1% and 53.6%, respectively (p = 0.0072). Multivariate analyses demonstrated that estimated blood loss (p = 0.044), pathological T stage (p = 0.0005), pathological N stage (p = 0.017), and ΔAlb-dNLR (p = 0.005) were independent prognostic factors for OS.

CONCLUSIONS

ΔAlb-dNLR is a useful prognostic factor for OS in patients with ESCC receiving nCT.

A Novel Nomogram for Prediction of Early Postoperative Complications of Total Gastrectomy for Gastric Cancer

Background

Accurate prediction of postoperative complications is especially important for the formulation of treatment plans for patients with total gastrectomy (TG) for gastric cancer (GC). The purpose of this study was to establish a risk assessment model for early postoperative complications.

Methods

This retrospective study involved 363 patients with GC who underwent TG from January 2019 to December 2020. The influencing factors were explored by univariate and multivariable logistic regression; then, a nomogram was established and verified by internal verification.

Results

Linear stapler (OR=2.501, P=0.030), age (OR=1.052, P =0.024), blood transfusion (yes) (OR=2.450, P =0.021), one-time consumables for surgery (or=1.000, P =0.022), number of total lymph nodes (OR=1.060, P =0.011) and number of positive lymph nodes (OR=1.054, P =0.029) were independent risk factors for early postoperative complications in TG, and nomogram model was constructed. The C-index of primary cohort, modeling cohort and validation cohort was 0.787, 0.754 and 0.912. The calibration curves showed good accuracy.

Conclusion

This study used the indicators available before and during surgery to establish a nomogram model for early postoperative complications of total gastrectomy for gastric cancer, which found that linear stapler (LS), blood transfusion, one-time consumables for surgery, number of total lymph nodes and number of positive lymph nodes were factors.

Prediction Model of Anastomotic Leakage Among Esophageal Cancer Patients After Receiving an Esophagectomy: Machine Learning Approach

Background

Anastomotic leakage (AL) is one of the severe postoperative adverse events (5%-30%), and it is related to increased medical costs in cancer patients who undergo esophagectomies. Machine learning (ML) methods show good performance at predicting risk for AL. However, AL risk prediction based on ML models among the Chinese population is unavailable.

Objective

This study uses ML techniques to develop and validate a risk prediction model to screen patients with emerging AL risk factors.

Methods

Analyses were performed using medical records from 710 patients who underwent esophagectomies at the National Clinical Research Center for Cancer between January 2010 and May 2015. We randomly split (9:1) the data set into a training data set of 639 patients and a testing data set of 71 patients using a computer algorithm. We assessed multiple classification tools to create a multivariate risk prediction model. Our ML algorithms contained decision tree, random forest, naive Bayes, and logistic regression with least absolute shrinkage and selection operator. The optimal AL prediction model was selected based on model evaluation metrics.

Results

The final risk panel included 36 independent risk features. Of those, 10 features were significantly identified by the logistic model, including aortic calcification (OR 2.77, 95% CI 1.32-5.81), celiac trunk calcification (OR 2.79, 95% CI 1.20-6.48), forced expiratory volume 1% (OR 0.51, 95% CI 0.30-0.89); TLco (OR 0.56, 95% CI 0.27-1.18), peripheral vascular disease (OR 4.97, 95% CI 1.44-17.07), laparoscope (OR 3.92, 95% CI 1.23-12.51), postoperative length of hospital stay (OR 1.17, 95% CI 1.13-1.21), vascular permeability activity (OR 0.46, 95% CI 0.14-1.48), and fat liquefaction of incisions (OR 4.36, 95% CI 1.86-10.21). Logistic regression with least absolute shrinkage and selection operator offered the highest prediction quality with an area under the receiver operator characteristic of 72% in the training data set. The testing model also achieved similar high performance.

Conclusions

Our model offered a prediction of AL with high accuracy, assisting in AL prevention and treatment. A personalized ML prediction model with a purely data-driven selection of features is feasible and effective in predicting AL in patients who underwent esophagectomy.