Defining a Nomogram for Predicting Early Recurrence in Gastric Cancer Patients After Neoadjuvant Chemotherapy and Radical Gastrectomy

Machine Learning Prediction of Early Recurrence in Gastric Cancer: A Nationwide Real-World Study

BACKGROUND

Patients with gastric cancer (GC) who experience early recurrence (ER) within 2 years postoperatively have poor prognoses. This study aimed to analyze and predict ER after curative surgery for patients with GC using machine learning (ML) methods.

PATIENTS AND METHODS

This multicenter population-based cohort study included data from ten large tertiary regional medical centers in China. The clinical, pathological, and laboratory parameters were retrospectively collected from the records of 11,615 patients. The patients were randomly divided into training (70%) and test (30%) cohorts. A total of ten ML models were developed and validated to predict the ER. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), calibration plots, and Brier score (BS). SHapley Additive exPlanations (SHAP) was used to rank the input features and interpret predictions.

RESULTS

ER was reported in 1794 patients (15%) during follow-up. The stacking ensemble model achieved AUCs of 1.0 and 0.8 in the training and testing cohorts, respectively, with a BS of 0.113. SHAP dependency plots revealed that tumor staging, elevated tumor marker levels, lymphovascular invasion, perineural invasion, and tumor size > 5 cm were associated with higher ER risk. The impact of age and the number of lymph nodes harvested on ER risk exhibited a "U-shaped distribution." Additionally, an online prediction tool based on the best model was developed to facilitate clinical applications.

CONCLUSIONS

We developed a robust clinical model for predicting the risk of ER after surgery for GC, which may aid in individualized clinical decision-making.

Peritoneal recurrence in gastric cancer after curative gastrectomy: risk factors and predictive score model

BACKGROUND

Peritoneal recurrence (PR) remains the most common pattern of relapse in gastric cancer (GC), even after curative resection. Given its dismal prognosis, the identification of risk factors for PR is essential for developing new treatment modalities and selecting a more appropriate subgroup of patients. This study aimed to evaluate the risk factors and survival outcomes of patients with GC who had PR and to develop a risk score to predict PR.

METHODS

All patients with GC who underwent curative gastrectomy were included. For analysis, patients were divided into no recurrence (NR), recurrence in other sites (ROS), and PR. Risk factors for PR were analyzed to build a risk score.

RESULTS

Among 622 patients with GC, 460 (74.0%) had NR, 98 (15.7%) had ROS, and 64 (10.3%) had PR. Female patients, linitis on computed tomography, depth of tumor invasion, and diffuse/mixed type were associated with PR. Patients with PR had worse overall survival than those with ROS (22.0 vs 29.8 months, respectively; P = .008). The median survival estimates after recurrence were 5.0 months in the PR group and 9.9 months in the ROS group (P < .001). The scoring system developed with 8 variables had an accuracy of 81% in predicting PR. Accordingly, 385 patients (61.9%) were classified as low risk, and 237 patients (38.1%) were classified as high risk. Among the 64 patients with PR, 53 (82.8%) were correctly classified as high risk (P < .001).

CONCLUSION

Patients with PR had distinct clinicopathologic characteristics and extremely restricted survival compared with patients with recurrence in other sites. The risk-scoring model was able to identify patients at higher risk of PR.

Comparison of efficacy and safety of nab-paclitaxel and oxaliplatin + S-1 and standard S-1 and oxaliplatin chemotherapy regimens for treatment of gastric cancer

BACKGROUND

Gastric cancer (GC) is a relatively frequent clinical phenomenon, referring to malignant tumors emerging in the gastric mucosal epithelial cells. It has a high morbidity and mortality rate, posing a significant threat to the health of patients. Hence, how to diagnose and treat GC has become a heated topic in this research field.

AIM

To discuss the effectiveness and safety of nab-paclitaxel in combination with oxaliplatin and S-1 (P-SOX) for the treatment of GC, and to analyze the factors that may influence its outcomes.

METHODS

A total of 219 eligible patients with advanced GC, who were treated at Qinghai University Affiliated Hospital Gastrointestinal Oncology between January 2018 and March 2020, were included in the study. Among them, 149 patients received SOX regimen and 70 patients received S-1 regimen. All patients underwent both preoperative and postoperative chemotherapy consisting of 2-4 cycles each, totaling 6-8 cycles, along with parallel D2 radical surgical treatment. The patients were followed up for a period of three years or until reaching the event endpoint.

RESULTS

The short-term and long-term efficacy of the P-SOX group was significantly higher than that of the SOX group, and the safety was manageable. Cox multivariate analysis revealed that progression-free survival was associated with perioperative chemotherapy efficacy, tumor diameter ≤ 2cm, high differentiation, and early cTNM (T stands for invasion depth; N stands for node metastasis; M stands for distant invasion) stage.

CONCLUSION

In comparison to the SOX regimen, the P-SOX regimen demonstrates improved short-term and long-term efficacy with tolerable adverse reactions. It is anticipated that the P-SOX regimen will emerge as a first-line chemotherapy option for GC. Patients with GC who receive effective perioperative chemotherapy (Response Evaluation Criteria in Solid Tumors 1.1, Tumor Regression Grade), have a tumor diameter ≤ 2cm, exhibit high degree of differentiation, and are at an early cTNM stage show better prognosis.

Development and validation of nomograms for predicting the prognosis of early and late recurrence of advanced gastric cancer after radical surgery based on post-recurrence survival

In this study, we aimed to explore the risk factors influencing post-recurrence survival (PRS) of early recurrence (ER) and late recurrence (LR) in stage advanced gastric cancer (AGC) patients after radical surgery, respectively, and to develop predictive models in turn. Medical records of 192 AGC patients who recurred after radical gastrectomy were retrospectively reviewed. They were randomly divided into the training and validation set at a ratio of 2:1. Nomograms were built based on risk factors influencing PRS of ER and LR explored by Cox regression analyses, respectively. Concordance index (C-index) values and calibration curves were used to evaluate predictive power of nomograms. Body mass index < 18.5 kg/m2, prealbumin level < 70.1 mg/L, positive lymph nodes ratio ≥ 0.486 and palliative treatment after recurrence were independent risk factors for the prognosis of ER. In contrast, prealbumin level < 170.1 mg/L, CEA ≥ 18.32 μg/L, tumor diameter ≥ 5.5 cm and palliative treatment after recurrence were independent risk factors for the prognosis of LR. The C-index values were 0.801 and 0.772 for ER and LR in the training set, respectively. The calibration curves of validation set showed a C-index value of 0.744 and 0.676 for ER and LR, respectively. Nomograms which were constructed to predict the prognosis of ER and LR of AGC after surgery showed great predictive power and could provide reference for clinicians' treatment strategies to some extent.

Development and validation of a preoperative CT-based risk scoring system for predicting recurrence-free survival in patients undergoing curative surgery for gastric cancer

PURPOSE

The objective of this study was to establish and validate a preoperative risk scoring system that incorporated both clinical and computed tomography(CT) variables to predict recurrence-free survival (RFS) in gastric cancer(GC) patients who underwent curative resection.

METHOD

We retrospectively included consecutive patients with surgically confirmed GC who underwent preoperative CT scans between October 2017 and January 2022. Multivariate Cox regression analysis was employed in the derivation set to identify clinical and CT variables associated with RFS and to construct a risk score. This risk score was subsequently validated in an independent test set.

RESULTS

A total of 346 patients were included in the study, with 213 in the derivation set and 133 in the test set. Five variables, namely ctEMVI, ctBorrmann, visceral obesity, sarcopenia, and NLR, were independently associated with RFS. In the test set, the preoperative risk score exhibited a c-index of 0.741, which outperformed the predictive accuracy of pathological tumor staging (c-index of 0.673, p = 0.021) at various time points. The preoperative risk score effectively stratified patients into low and high-risk groups.

CONCLUSION

The developed preoperative risk scoring system demonstrated the ability to predict RFS following curative resection in GC patients.