Preoperative T2-weighted MR imaging texture analysis of gastric cancer: prediction of TNM stages

Machine learning applications in upper gastrointestinal cancer surgery: a systematic review

BACKGROUND

Machine learning (ML) has seen an increase in application, and is an important element of a digital evolution. The role of ML within upper gastrointestinal surgery for malignancies has not been evaluated properly in the literature. Therefore, this systematic review aims to provide a comprehensive overview of ML applications within upper gastrointestinal surgery for malignancies.

METHODS

A systematic search was performed in PubMed, EMBASE, Cochrane, and Web of Science. Studies were only included when they described machine learning in upper gastrointestinal surgery for malignancies. The Cochrane risk-of-bias tool was used to determine the methodological quality of studies. The accuracy and area under the curve were evaluated, representing the predictive performances of ML models.

RESULTS

From a total of 1821 articles, 27 studies met the inclusion criteria. Most studies received a moderate risk-of-bias score. The majority of these studies focused on neural networks (n = 9), multiple machine learning (n = 8), and random forests (n = 3). Remaining studies involved radiomics (n = 3), support vector machines (n = 3), and decision trees (n = 1). Purposes of ML included predominantly prediction of metastasis, detection of risk factors, prediction of survival, and prediction of postoperative complications. Other purposes were predictions of TNM staging, chemotherapy response, tumor resectability, and optimal therapy.

CONCLUSIONS

Machine Learning algorithms seem to contribute to the prediction of postoperative complications and the course of disease after upper gastrointestinal surgery for malignancies. However, due to the retrospective character of ML studies, these results require trials or prospective studies to validate this application of ML.

Accuracy of preoperative clinical staging for locally advanced gastric cancer in KLASS-02 randomized clinical trial

Purpose

The discrepancy between preoperative and final pathological staging has been a long-standing challenge for the application of clinical trials or appropriate treatment options. This study aimed to demonstrate the accuracy of preoperative staging of locally advanced gastric cancer using data from a large-scale randomized clinical trial.

Materials and methods

Of the 1050 patients enrolled in the clinical trial, 26 were excluded due to withdrawal of consent (n = 20) or non-surgery (n = 6). The clinical and pathological staging was compared. Risk factor analysis for underestimation was performed using univariate and multivariate analyses.

Results

Regarding T staging by computed tomography, accuracy rates were 74.48, 61.62, 58.56, and 85.16% for T1, T2, T3 and T4a, respectively. Multivariate analysis for underestimation of T staging revealed that younger age, ulcerative gross type, circular location, larger tumor size, and undifferentiated histology were independent risk factors. Regarding nodal status estimation, 54.9% of patients with clinical N0 disease were pathologic N0, and 36.4% of patients were revealed to have pathologic N0 among clinical node-positive patients. The percentage of metastasis involvement at the D1, D1+, and D2 lymph node stations significantly increased with the advanced clinical N stage. Among all patients, 29 (2.8%), including 26 with peritoneal seeding, exhibited distant metastases.

Conclusions

Estimating the exact pathologic staging remains challenging. A thorough evaluation is mandatory before treatment selection or trial enrollment. Moreover, we need to set a sufficient case number when we design the clinical trial considering the stage migration.

Differentiating T1a-T1b from T2 in gastric cancer lesions with three different measurement approaches based on contrast-enhanced T1W imaging at 3.0 T

Background

To explore the diagnostic value of three different measurement approaches in differentiating T1a–T1b from T2 gastric cancer (GC) lesions.

Methods

A total of 95 consecutive patients with T1a–T2 stage of GC who performed preoperative MRI were retrospectively enrolled between January 2017 and November 2020. The parameters MRI T stage (subjective evaluation), thickness, maximum area and volume of the lesions were evaluated by two radiologists. Specific indicators including AUC, optimal cutoff, sensitivity, specificity, accuracy, positive likelihood ratio (PLR), negative likelihood ratio (NLR), positive predictive value (PPV) and negative predictive value (NPV) of MRI T stage, thickness, maximum area and volume for differentiating T1a–T1b from T2 stage lesions were calculated. The ROC curves were compared by the Delong test. Decision curve analysis (DCA) was used to evaluate the clinical benefit.

Results

The ROC curves for thickness (AUC = 0.926), maximum area (AUC = 0.902) and volume (AUC = 0.897) were all significantly better than those of the MRI T stage (AUC = 0.807) in differentiating T1a–T1b from T2 lesions, with p values of 0.004, 0.034 and 0.041, respectively. The values corresponding to the thickness (including AUC, sensitivity, specificity, accuracy, PPV, NPV, PLR and NLR) were all higher than those corresponding to the MRI T stage, maximum area and volume. The DCA curves indicated that the parameter thickness could provide the highest clinical benefit if the threshold probability was above 35%.

Conclusions

Thickness may provide an efficient approach to rapidly distinguish T1a–T1b from T2 stage GC lesions.

The Discordance of Clinical and Pathologic Staging in Locally Advanced Gastric Adenocarcinoma

BACKGROUND

Clinical staging guides decisions about optimal treatment sequence in patients with gastric cancer, although the preoperative accuracy is not strongly established. This study investigates concordance of clinical and pathologic stage as well as its impact on the survival of patients with gastric adenocarcinoma.

METHODS

Patients with clinical stage T2-4, N0, M0 gastric adenocarcinoma who underwent surgery without neoadjuvant therapy were identified from the National Cancer Database (2010-2015). The primary outcome was up-staging, defined as cT < pT, pN1-3, and/or pM1 (AJCC 7^th edition). Multivariable logistic regression analysis was performed to predict up-staging. Survival analysis was performed using the Kaplan-Meier method.

RESULTS

In total, 2254 patients were identified. cTNM staging was discordant with pTNM staging in 65.6% of cases, with 50.4% up-staged and 15.2% down-staged. On multivariable logistic regression, younger age (OR 0.991, 95% CI 0.984-0.999, p=0.0188), male sex (versus female; OR 1.392, 95% CI 1.158-1.673, p=0.0004), poor or undifferentiated tumor grade (versus well differentiated or moderately differentiated; OR 2.399, 95% CI 1.987-2.896; p<0.0001), positive margin status (versus negative; OR 4.575, 95% CI 3.360-6.230; p<0.0001), and days from diagnosis to surgery (15-32 days versus ≤ 14 days; OR 1.411, 95% CI 1.098-1.814, p=0.0072) were predictive of up-staging. Patients who were up-staged had a decreased survival compared to patients who were accurately staged (median survival 27.9 months versus 67.6 months; log-rank p<0.0001).

CONCLUSION

This study found a substantial discordance between clinical and pathologic staging of resectable locally advanced gastric adenocarcinoma. These data support that patients may have more advanced disease at presentation than reflected in clinical staging and may benefit from improved diagnostic modalities and neoadjuvant chemotherapy.