Potential roles of lymphovascular space invasion based on tumor characteristics provide important prognostic information in T1 tumors with ER and HER2 positive breast cancer

MRI Radiomics-Based Machine Learning to Predict Lymphovascular Invasion of HER2-Positive Breast Cancer

This study aims to develop and prospectively validate radiomic models based on MRI to predict lymphovascular invasion (LVI) status in patients with HER2-positive breast cancer. A total of 225 patients with HER2-positive breast cancer who preoperatively underwent breast MRI were selected, forming the training set (n = 99 LVI-positive, n = 126 LVI-negative). A prospective validation cohort included 130 patients with breast cancer from the Affiliated Zhongshan Hospital of Dalian University (n = 57 LVI-positive, n = 73 LVI-negative). A total of 390 radiomic features and eight conventional radiological characteristics were extracted. For the optimum feature selection phase, the LASSO regression model with tenfold cross-validation (CV) was employed to identify features with non-zero coefficients. The conventional radiological (CR) model was determined based on visual morphological (VM) features and the optimal radiomic features correlated with LVI, identified through multivariate logistic analyses. Subsequently, various machine learning (ML) models were developed using algorithms such as support vector machine (SVM), k-nearest neighbor (KNN), gradient boosting machine (GBM), and random forest (RF). The performance of ML and CR models. The results show that the AUC of the CR model in the training and validation sets were 0.81 (95% confidence interval [CI], 0.74-0.86) and 0.82 (95% CI, 0.69-0.89), respectively. The ML model achieved the best performance, with AUCs of 0.96 (95% CI, 0.99-1.00) in the training set and 0.95 (95% CI, 0.89-0.96) in the validation set. There were significant differences between the CR and ML models in predicting LVI status. Our study demonstrated that the machine learning models exhibited superior performance in predicting LVI status based on pretreatment MRI compared to the CR model, which does not necessarily rely on a priori knowledge of visual morphology.

Ultrasound-Based Deep Learning Radiomics Nomogram for the Assessment of Lymphovascular Invasion in Invasive Breast Cancer: A Multicenter Study

RATIONALE AND OBJECTIVES

The aim of this study was to develop a deep learning radiomics nomogram (DLRN) based on B-mode ultrasound (BMUS) and color doppler flow imaging (CDFI) images for preoperative assessment of lymphovascular invasion (LVI) status in invasive breast cancer (IBC).

MATERIALS AND METHODS

In this multicenter, retrospective study, 832 pathologically confirmed IBC patients were recruited from eight hospitals. The samples were divided into training, internal test, and external test sets. Deep learning and handcrafted radiomics features reflecting tumor phenotypes on BMUS and CDFI images were extracted. The BMUS score and CDFI score were calculated after radiomics feature selection. Subsequently, a DLRN was developed based on the scores and independent clinic-ultrasonic risk variables. The performance of the DLRN was evaluated for calibration, discrimination, and clinical usefulness.

RESULTS

The DLRN predicted the LVI with accuracy, achieving an area under the receiver operating characteristic curve of 0.93 (95% CI 0.90-0.95), 0.91 (95% CI 0.87-0.95), and 0.91 (95% CI 0.86-0.94) in the training, internal test, and external test sets, respectively, with good calibration. The DLRN demonstrated superior performance compared to the clinical model and single scores across all three sets (p < 0.05). Decision curve analysis and clinical impact curve confirmed the clinical utility of the model. Furthermore, significant enhancements in net reclassification improvement (NRI) and integrated discrimination improvement (IDI) indicated that the two scores could serve as highly valuable biomarkers for assessing LVI.

CONCLUSION

The DLRN exhibited strong predictive value for LVI in IBC, providing valuable information for individualized treatment decisions.

A nomogram of preoperative indicators predicting lymph vascular space invasion in cervical cancer

PURPOSE

To develop predictive nomograms of lymph vascular space invasion (LVSI) in patients with early-stage cervical cancer.

METHODS

We identified 403 patients with cervical cancer from the Affiliated Hospital of Jiangnan University from January 2015 to December 2019. Patients were divided into the training set (n = 242) and the validation set (n = 161), with patients in the training set subdivided into LVSI (+) and LVSI (-) groups according to postoperative pathology. Preoperative hematologic indexes were compared between the two subgroups. Univariate and multivariate logistic regression analyses were used to analyze the independent risk factors for LVSI, from which a nomogram was constructed using the R package.

RESULTS

LVSI (+) was present in 94 out of 242 patients in the training set, accompanied by a significant increase in the preoperative squamous cell carcinoma antigen (SCC), white blood cells (WBC), neutrophil (NE), platelet (PLT), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), systemic inflammation index (SII), and tumor size (P < 0.05). Univariate analysis showed that SCC, WBC, NE, NLR, PLR, SII, and tumor size were correlated with LVSI (P < 0.05), and multivariate analysis showed that tumor size, SCC, WBC, and NLR were independent risk factors for LVSI (P < 0.05). A nomogram was correspondingly established with good performance in predicting LVSI [training: ROC-AUC = 0.845 (95% CI: 0.731-0.843) and external validation: ROC-AUC = 0.704 (95% CI: 0.683-0.835)] and high accuracy (training: C-index = 0.787; external validation: C-index = 0.759).

CONCLUSION

The nomogram based on preoperative tumor size, SCC, WBC, and NLR had excellent accuracy and discriminative capability to assess the risk of LVSI in early-stage cervical cancer patients.

Peptidylarginine Deiminase Type 2 Predicts Tumor Progression and Poor Prognosis in Patients with Curatively Resected Biliary Tract Cancer

(1) Background: PADI2 is a post-translational modification (PTM) enzyme that catalyzes citrullination, which then triggers autoimmune disease and cancer. This study aimed to evaluate the prognostic value of peptidylarginine deiminase 2 (PADI2) protein expression in biliary tract cancer (BTC) patients. (2) Methods: Using immunohistochemistry, the PADI2 protein expression in BTC tissues was analyzed. The correlations between PADI2 protein expression and clinicopathologic characteristics were analyzed using Chi-square tests. The Kaplan-Meier procedure was used for comparing survival distributions. We used Cox proportional hazards regression for univariate and multivariate analyses. From 2014 to 2020, 30 resected BTC patients were enrolled in this study. (3) Results: Patients with high PADI2 protein expression were associated with shorter progress-free survival (PFS; p = 0.041), disease-specific survival (DSS; p = 0.025), and overall survival (OS; p = 0.017) than patients with low PADI2 protein expression. (4) Conclusions: The results indicated that PADI2 protein expression was an independent poor prognostic factor for BTC patients regarding PFS, DSS, and OS.

Further predictive value of lymphovascular invasion explored via supervised deep learning for lymph node metastases in breast cancer

Lymphovascular invasion, specifically lymph-blood vessel invasion (LBVI), is a risk factor for metastases in breast invasive ductal carcinoma (IDC) and is routinely screened using hematoxylin-eosin histopathological images. However, routine reports only describe whether LBVI is present and does not provide other potential prognostic information of LBVI. This study aims to evaluate the clinical significance of LBVI in 685 IDC cases and explore the added predictive value of LBVI on lymph node metastases (LNM) via supervised deep learning (DL), an expert-experience embedded knowledge transfer learning (EEKT) model in 40 LBVI-positive cases signed by the routine report. Multivariate logistic regression and propensity score matching analysis demonstrated that LBVI (OR 4.203, 95% CI 2.809-6.290, P < 0.001) was a significant risk factor for LNM. Then, the EEKT model trained on 5780 image patches automatically segmented LBVI with a patch-wise Dice similarity coefficient of 0.930 in the test set and output counts, location, and morphometric features of the LBVIs. Some morphometric features were beneficial for further stratification within the 40 LBVI-positive cases. The results showed that LBVI in cases with LNM had a higher short-to-long side ratio of the minimum rectangle (MR) (0.686 vs. 0.480, P = 0.001), LBVI-to-MR area ratio (0.774 vs. 0.702, P = 0.002), and solidity (0.983 vs. 0.934, P = 0.029) compared to LBVI in cases without LNM. The results highlight the potential of DL to assist pathologists in quantifying LBVI and, more importantly, in exploring added prognostic information from LBVI.

Preoperative prediction of lymphovascular invasion in patients with T1 breast invasive ductal carcinoma based on radiomics nomogram using grayscale ultrasound

Purpose

The aim of this study was to develop a radiomics nomogram based on grayscale ultrasound (US) for preoperatively predicting Lymphovascular invasion (LVI) in patients with pathologically confirmed T1 (pT1) breast invasive ductal carcinoma (IDC).

Methods

One hundred and ninety-two patients with pT1 IDC between September 2020 and August 2022 were analyzed retrospectively. Study population was randomly divided in a 7: 3 ratio into a training dataset of 134 patients (37 patients with LVI-positive) and a validation dataset of 58 patients (19 patients with LVI-positive). Clinical information and conventional US (CUS) features (called clinic_CUS features) were recorded and evaluated to predict LVI. In the training dataset, independent predictors of clinic_CUS features were obtained by univariate and multivariate logistic regression analyses and incorporated into a clinic_CUS prediction model. In addition, radiomics features were extracted from the grayscale US images, and the radiomics score (Radscore) was constructed after radiomics feature selection. Subsequent multivariate logistic regression analysis was also performed for Radscore and the independent predictors of clinic_CUS features, and a radiomics nomogram was developed. The performance of the nomogram model was evaluated via its discrimination, calibration, and clinical usefulness.

Results

The US reported axillary lymph node metastasis (LNM) (US_LNM) status and tumor margin were determined as independent risk factors, which were combined for the construction of clinic_CUS prediction model for LVI in pT1 IDC. Moreover, tumor margin, US_LNM status and Radscore were independent predictors, incorporated as the radiomics nomogram model, which achieved a superior discrimination to the clinic_CUS model in the training dataset (AUC: 0.849 vs. 0.747; P < 0.001) and validation dataset (AUC: 0.854 vs. 0.713; P = 0.001). Calibration curve for the radiomic nomogram showed good concordance between predicted and actual probability. Furthermore, decision curve analysis (DCA) confirmed that the radiomics nomogram had higher clinical net benefit than the clinic_CUS model.

Conclusion

The US-based radiomics nomogram, incorporating tumor margin, US_LNM status and Radscore, showed a satisfactory preoperative prediction of LVI in pT1 IDC patients.