Dataset Information

LASSO-Based Machine Learning Algorithm for Prediction of Lymph Node Metastasis in T1 Colorectal Cancer.

ABSTRACT:

Purpose

The role of tumor-infiltrating lymphocytes (TILs) in predicting lymph node metastasis (LNM) in patients with T1 colorectal cancer (CRC) remains unclear. Furthermore, clinical utility of a machine learning-based approach has not been widely studied.

Materials and methods

Immunohistochemistry for TILs against CD3, CD8, and forkhead box P3 in both center and invasive margin of the tumor were performed using surgically resected T1 CRC slides. Three hundred and sixteen patients were enrolled and categorized into training (n=221) and validation (n=95) sets via random sampling. Using clinicopathologic variables including TILs, the least absolute shrinkage and selection operator (LASSO) regression model was applied for variable selection and predictive signature building in the training set. The predictive accuracy of our model and the Japanese criteria were compared using area under the receiver operating characteristic (AUROC), net reclassification improvement (NRI)/integrated discrimination improvement (IDI), and decision curve analysis (DCA) in the validation set.

Results

LNM was detected in 29 (13.1%) and 12 (12.6%) patients in training and validation sets, respectively. Nine variables were selected and used to generate the LASSO model. Its performance was similar in training and validation sets (AUROC, 0.795 vs. 0.765; p=0.747). In the validation set, the LASSO model showed better outcomes in predicting LNM than Japanese criteria, as measured by AUROC (0.765 vs. 0.518, p=0.003) and NRI (0.447, p=0.039)/IDI (0.121, p=0.034). DCA showed positive net benefits in using our model.

Conclusion

Our LASSO model incorporating histopathologic parameters and TILs showed superior performance compared to conventional Japanese criteria in predicting LNM in patients with T1 CRC.

SUBMITTER: Kang J

PROVIDER: S-EPMC8291173 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Distant metastasis (DM) is relatively uncommon in T1 stage gastric cancer (GC). The aim of this study was to develop and validate a predictive model for DM in stage T1 GC using machine learning (ML) algorithms. Patients with stage T1 GC from 2010 to 2017 were screened from the public Surveillance, Epidemiology and End Results (SEER) database. Meanwhile, we collected patients with stage T1 GC admitted to the Department of Gastrointestinal Surgery of the Second Affiliated Hospital of Nanchang University from 2015 to 2017. We applied seven ML algorithms: logistic regression, random forest (RF), LASSO, support vector machine, k-Nearest Neighbor, Naive Bayesian Model, Artificial Neural Network. Finally, a RF model for DM of T1 GC was developed. The AUC, sensitivity, specificity, F1-score and accuracy were used to evaluate and compare the predictive performance of the RF model with other models. Finally, we performed a prognostic analysis of patients who developed distant metastases. Independent risk factors for prognosis were analysed by univariate and multifactorial regression. K-M curves were used to express differences in survival prognosis for each variable and subvariable. A total of 2698 cases were included in the SEER dataset, 314 with DM, and 107 hospital patients were included, 14 with DM. Age, T-stage, N-stage, tumour size, grade and tumour location were independent risk factors for the development of DM in stage T1 GC. A combined analysis of seven ML algorithms in the training and test sets found that the RF prediction model had the best prediction performance (AUC: 0.941, Accuracy: 0.917, Recall: 0.841, Specificity: 0.927, F1-score: 0.877). The external validation set ROCAUC was 0.750. Meanwhile, survival prognostic analysis showed that surgery (HR = 3.620, 95% CI 2.164-6.065) and adjuvant chemotherapy (HR = 2.637, 95% CI 2.067-3.365) were independent risk factors for survival prognosis in patients with DM from stage T1 GC. Age, T-stage, N-stage, tumour size, grade and tumour location were independent risk factors for the development of DM in stage T1 GC. ML algorithms had shown that RF prediction models had the best predictive efficacy to accurately screen at-risk populations for further clinical screening for metastases. At the same time, aggressive surgery and adjuvant chemotherapy can improve the survival rate of patients with DM.

Project description:Lymph node involvement increases the risk of breast cancer recurrence. An accurate non-invasive assessment of nodal involvement is valuable in cancer staging, surgical risk, and cost savings. Radiomics has been proposed to pre-operatively predict sentinel lymph node (SLN) status; however, radiomic models are known to be sensitive to acquisition parameters. The purpose of this study was to develop a prediction model for preoperative prediction of SLN metastasis using deep learning-based (DLB) features and compare its predictive performance to state-of-the-art radiomics. Specifically, this study aimed to compare the generalizability of radiomics vs DLB features in an independent test set with dissimilar resolution. Dynamic contrast-enhancement images from 198 patients (67 positive SLNs) were used in this study. Of these subjects, 163 had an in-plane resolution of 0.7 × 0.7 mm2, which were randomly divided into a training set (approximately 67%) and a validation set (approximately 33%). The remaining 35 subjects with a different in-plane resolution (0.78 × 0.78 mm2) were treated as independent testing set for generalizability. Two methods were employed: (1) conventional radiomics (CR), and (2) DLB features which replaced hand-curated features with pre-trained VGG-16 features. The threshold determined using the training set was applied to the independent validation and testing dataset. Same feature reduction, feature selection, model creation procedures were used for both approaches. In the validation set (same resolution as training), the DLB model outperformed the CR model (accuracy 83% vs 80%). Furthermore, in the independent testing set of the dissimilar resolution, the DLB model performed markedly better than the CR model (accuracy 77% vs 71%). The predictive performance of the DLB model outperformed the CR model for this task. More interestingly, these improvements were seen particularly in the independent testing set of dissimilar resolution. This could indicate that DLB features can ultimately result in a more generalizable model.