Four-Dimensional Machine Learning Radiomics for the Pretreatment Assessment of Breast Cancer Pathologic Complete Response to Neoadjuvant Chemotherapy in Dynamic Contrast-Enhanced MRI

A clinicopathological-imaging nomogram for the prediction of pathological complete response in breast cancer cases administered neoadjuvant therapy

OBJECTIVE

To construct a user-friendly nomogram with MRI and clinicopathological parameters for the prediction of pathological complete response (pCR) after neoadjuvant therapy (NAT) in patients with breast cancer (BC).

METHODS

We retrospectively enrolled consecutive female patients pathologically confirmed with breast cancer who received NAT followed by surgery between January 2018 and December 2022 as the development cohort. Additionally, we prospectively collected eligible candidates between January 2023 and December 2023 as an external validation group at our institution. Pretreatment MRI features and clinicopathological variables were collected, and the pre- and post-treatment background parenchymal enhancement (BPE) and the changes in BPE on two MRIs were compared between patients who achieved pCR and those who did not. Multivariable logistic regression analysis was used to identify independent variables associated with pCR in the development cohort. These independent variables were combined into a predictive nomogram for which performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration plot, decision curve analysis, and external validation.

RESULTS

In the development cohort, there were a total of 276 female patients with a mean age of 48.3 ± 8.7 years, while in the validation cohort, there were 87 female patients with a mean age of 49.0 ± 9.5 years. Independent prognostic factors of pCR included small tumor size, HER2(+), high Ki-67 index,high signal enhancement ratio (SER), low minimum value of apparent diffusion coefficient (ADCmin), and significantly decreased BPE after NAT(change of BPE). The nomogram, which incorporates the above parameters, demonstrated excellent predictive performance in both the development and external validation cohorts, with AUC values of 0.900 and 0.850, respectively. Additionally, the nomogram showed excellent calibration capacities, as indicated by Hosmer-Lemeshow test p values of 0.508 and 0.423 in the two cohorts. Furthermore, the nomogram provided greater net benefits compared to the default simple schemes in both cohorts.

CONCLUSION

A nomogram constructed using tumor size, HER2 status, Ki-67 index, SER, ADCmin, and changes in pre- and post-NAT BPE demonstrated strong predictive performance, calibration ability, and greater net benefits for predicting pCR in patients with BC after NAT. This suggests that the user-friendly nomogram could be a valuable imaging biomarker for identifying suitable candidates for NAT.

Use of Pretreatment Perfusion MRI-based Intratumoral Heterogeneity to Predict Pathologic Response of Triple-Negative Breast Cancer to Neoadjuvant Chemoimmunotherapy

Background Neoadjuvant chemoimmunotherapy (NACI) has significantly increased the rate of pathologic complete response (pCR) in patients with early-stage triple-negative breast cancer (TNBC), although predictors of response to this regimen have not been identified. Purpose To investigate pretreatment perfusion MRI-based radiomics as a predictive marker for pCR in patients with TNBC undergoing NACI. Materials and Methods This prospective study enrolled women with early-stage TNBC who underwent NACI at two different centers from August 2021 to July 2023. Pretreatment dynamic contrast-enhanced MRI scans obtained using scanners from multiple vendors were analyzed using the Tofts model to segment tumors and analyze pharmacokinetic parameters. Radiomics features were extracted from the rate constant for contrast agent plasma-to-interstitial transfer (or Ktrans), volume fraction of extravascular and extracellular space (Ve), and maximum contrast agent uptake rate (Slopemax) maps and analyzed using unsupervised correlation and least absolute shrinkage and selector operator, or LASSO, to develop a radiomics score. Score effectiveness was assessed using the area under the receiver operating characteristic curve (AUC), and multivariable logistic regression was used to develop a multimodal nomogram for enhanced prediction. The discrimination, calibration, and clinical utility of the nomogram were evaluated in an external test set. Results The training set included 112 female participants from center 1 (mean age, 52 years ± 11 [SD]), and the external test set included 83 female participants from center 2 (mean age, 47 years ± 11). The radiomics score demonstrated an AUC of 0.80 (95% CI: 0.70, 0.89) for predicting pCR. A nomogram incorporating the radiomics score, grade, and Ki-67 yielded an AUC of 0.86 (95% CI: 0.78, 0.94) in the test set. Associations were found between higher radiomics score (>0.25) and tumor size (P < .001), washout enhancement (P = .01), androgen receptor expression (P = .009), and programmed death ligand 1 expression (P = .01), demonstrating a correlation with tumor immune environment in participants with TNBC. Conclusion A radiomics score derived from pharmacokinetic parameters at pretreatment dynamic contrast-enhanced MRI exhibited good performance for predicting pCR in participants with TNBC undergoing NACI, and could potentially be used to enhance clinical decision making. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Rauch in this issue.

Development and Validation of an Interpretable Machine Learning Prediction Model for Total Pathological Complete Response after Neoadjuvant Chemotherapy in Locally Advanced Breast Cancer: Multicenter Retrospective Analysis

Objective: This study aims to develop an interpretable machine learning (ML) model to accurately predict the probability of achieving total pathological complete response (tpCR) in patients with locally advanced breast cancer (LABC) following neoadjuvant chemotherapy (NAC). Methods: This multi-center retrospective study included pre-NAC clinical pathology data from 698 LABC patients. Post-operative pathological outcomes divided patients into tpCR and non-tpCR groups. Data from 586 patients at Shanghai Ruijin Hospital were randomly assigned to a training set (80%) and a test set (20%). In comparison, data from our hospital's remaining 112 patients were used for external validation. Variable selection was performed using the Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis. Predictive models were constructed using six ML algorithms: decision trees, K-nearest neighbors (KNN), support vector machine, light gradient boosting machine, and extreme gradient boosting. Model efficacy was assessed through various metrics, including receiver operating characteristic (ROC) curves, precision-recall (PR) curves, confusion matrices, calibration plots, and decision curve analysis (DCA). The best-performing model was selected by comparing the performance of different algorithms. Moreover, variable relevance was ranked using the SHapley Additive exPlanations (SHAP) technique to improve the interpretability of the model and solve the "black box" problem. Results: A total of 191 patients (32.59%) achieved tpCR following NAC. Through LASSO regression analysis, five variables were identified as predictive factors for model construction, including tumor size, Ki-67, molecular subtype, targeted therapy, and chemotherapy regimen. The KNN model outperformed the other five classifier algorithms, achieving area under the curve (AUC) values of 0.847 (95% CI: 0.809-0.883) in the training set, 0.763 (95% CI: 0.670-0.856) in the test set, and 0.665 (95% CI: 0.555-0.776) in the external validation set. DCA demonstrated that the KNN model yielded the highest net advantage through a wide range of threshold probabilities in both the training and test sets. Furthermore, the analysis of the KNN model utilizing SHAP technology demonstrated that targeted therapy is the most crucial factor in predicting tpCR. Conclusion: An ML prediction model using clinical and pathological data collected before NAC was developed and verified. This model accurately predicted the probability of achieving a tpCR in patients with LABC after receiving NAC. SHAP technology enhanced the interpretability of the model and assisted in clinical decision-making and therapy optimization.

A preoperative radiogenomic model based on quantitative heterogeneity for predicting outcomes in triple-negative breast cancer patients who underwent neoadjuvant chemotherapy

BACKGROUND

Triple-negative breast cancer (TNBC) is highly heterogeneous, resulting in different responses to neoadjuvant chemotherapy (NAC) and prognoses among patients. This study sought to characterize the heterogeneity of TNBC on MRI and develop a radiogenomic model for predicting both pathological complete response (pCR) and prognosis.

MATERIALS AND METHODS

In this retrospective study, TNBC patients who underwent neoadjuvant chemotherapy at Fudan University Shanghai Cancer Center were enrolled as the radiomic development cohort (n = 315); among these patients, those whose genetic data were available were enrolled as the radiogenomic development cohort (n = 98). The study population of the two cohorts was randomly divided into a training set and a validation set at a ratio of 7:3. The external validation cohort (n = 77) included patients from the DUKE and I-SPY 1 databases. Spatial heterogeneity was characterized using features from the intratumoral subregions and peritumoral region. Hemodynamic heterogeneity was characterized by kinetic features from the tumor body. Three radiomics models were developed by logistic regression after selecting features. Model 1 included subregional and peritumoral features, Model 2 included kinetic features, and Model 3 integrated the features of Model 1 and Model 2. Two fusion models were developed by further integrating pathological and genomic features (PRM: pathology-radiomics model; GPRM: genomics-pathology-radiomics model). Model performance was assessed with the AUC and decision curve analysis. Prognostic implications were assessed with Kaplan‒Meier curves and multivariate Cox regression.

RESULTS

Among the radiomic models, the multiregional model representing multiscale heterogeneity (Model 3) exhibited better pCR prediction, with AUCs of 0.87, 0.79, and 0.78 in the training, internal validation, and external validation sets, respectively. The GPRM showed the best performance for predicting pCR in the training (AUC = 0.97, P = 0.015) and validation sets (AUC = 0.93, P = 0.019). Model 3, PRM and GPRM could stratify patients by disease-free survival, and a predicted nonpCR was associated with poor prognosis (P = 0.034, 0.001 and 0.019, respectively).

CONCLUSION

Multiscale heterogeneity characterized by DCE-MRI could effectively predict the pCR and prognosis of TNBC patients. The radiogenomic model could serve as a valuable biomarker to improve the prediction performance.

Comparison of radiomics-based machine-learning classifiers for the pretreatment prediction of pathologic complete response to neoadjuvant therapy in breast cancer

Background

Machine learning classifiers are increasingly used to create predictive models for pathological complete response (pCR) in breast cancer after neoadjuvant therapy (NAT). Few studies have compared the effectiveness of different ML classifiers. This study evaluated radiomics models based on pre- and post-contrast first-phase T1 weighted images (T1WI) in predicting breast cancer pCR after NAT and compared the performance of ML classifiers.

Methods

This retrospective study enrolled 281 patients undergoing NAT from the Duke-Breast-Cancer-MRI dataset. Radiomic features were extracted from pre- and post-contrast first-phase T1WI images. The Synthetic Minority Oversampling Technique (SMOTE) was applied, then the dataset was randomly divided into training and validation groups (7:3). The radiomics model was built using selected optimal features. Support vector machine (SVM), random forest (RF), decision tree (DT), k-nearest neighbor (KNN), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM) were classifiers. Receiver operating characteristic curves were used to assess predictive performance.

Results

LightGBM performed best in predicting pCR [area under the curve (AUC): 0.823, 95% confidence interval (CI) [0.743-0.902], accuracy 74.0%, sensitivity 85.0%, specificity 67.2%]. During subgroup analysis, RF was most effective in pCR prediction in luminal breast cancers (AUC: 0.914, 95% CI [0.847-0.981], accuracy 87.0%, sensitivity 85.2%, specificity 88.1%). In triple-negative breast cancers, LightGBM performed best (AUC: 0.836, 95% CI [0.708-0.965], accuracy 78.6%, sensitivity 68.2%, specificity 90.0%).

Conclusion

The LightGBM-based radiomics model performed best in predicting pCR in patients with breast cancer. RF and LightGBM showed promising results for luminal and triple-negative breast cancers, respectively.

A novel approach correlating pathologic complete response with digital pathology and radiomics in triple-negative breast cancer

This rapid communication highlights the correlations between digital pathology-whole slide imaging (WSI) and radiomics-magnetic resonance imaging (MRI) features in triple-negative breast cancer (TNBC) patients. The research collected 12 patients who had both core needle biopsy and MRI performed to evaluate pathologic complete response (pCR). The results showed that higher collagenous values in pathology data were correlated with more homogeneity, whereas higher tumor expression values in pathology data correlated with less homogeneity in the appearance of tumors on MRI by size zone non-uniformity normalized (SZNN). Higher myxoid values in pathology data are correlated with less similarity of gray-level non-uniformity (GLN) in tumor regions on MRIs, while higher immune values in WSIs correlated with the more joint distribution of smaller-size zones by small area low gray-level emphasis (SALGE) in the tumor regions on MRIs. Pathologic complete response (pCR) was associated with collagen, tumor, and myxoid expression in WSI and GLN and SZNN in radiomic features. The correlations of WSI and radiomic features may further our understanding of the TNBC tumoral microenvironment (TME) and could be used in the future to better tailor the use of neoadjuvant chemotherapy (NAC). This communication will focus on the post-NAC MRI features correlated with pCR and their association with WSI features from core needle biopsies.

Artificial intelligence in breast imaging: potentials and challenges

Breast cancer, which is the most common type of malignant tumor among humans, is a leading cause of death in females. Standard treatment strategies, including neoadjuvant chemotherapy, surgery, postoperative chemotherapy, targeted therapy, endocrine therapy, and radiotherapy, are tailored for individual patients. Such personalized therapies have tremendously reduced the threat of breast cancer in females. Furthermore, early imaging screening plays an important role in reducing the treatment cycle and improving breast cancer prognosis. The recent innovative revolution in artificial intelligence (AI) has aided radiologists in the early and accurate diagnosis of breast cancer. In this review, we introduce the necessity of incorporating AI into breast imaging and the applications of AI in mammography, ultrasonography, magnetic resonance imaging, and positron emission tomography/computed tomography based on published articles since 1994. Moreover, the challenges of AI in breast imaging are discussed.

Exploring Neoadjuvant Chemotherapy, Predictive Models, Radiomic, and Pathological Markers in Breast Cancer: A Comprehensive Review

Breast cancer retains its position as the most prevalent form of malignancy among females on a global scale. The careful selection of appropriate treatment for each patient holds paramount importance in effectively managing breast cancer. Neoadjuvant chemotherapy (NACT) plays a pivotal role in the comprehensive treatment of this disease. Administering chemotherapy before surgery, NACT becomes a powerful tool in reducing tumor size, potentially enabling fewer invasive surgical procedures and even rendering initially inoperable tumors amenable to surgery. However, a significant challenge lies in the varying responses exhibited by different patients towards NACT. To address this challenge, researchers have focused on developing prediction models that can identify those who would benefit from NACT and those who would not. Such models have the potential to reduce treatment costs and contribute to a more efficient and accurate management of breast cancer. Therefore, this review has two objectives: first, to identify the most effective radiomic markers correlated with NACT response, and second, to explore whether integrating radiomic markers extracted from radiological images with pathological markers can enhance the predictive accuracy of NACT response. This review will delve into addressing these research questions and also shed light on the emerging research direction of leveraging artificial intelligence techniques for predicting NACT response, thereby shaping the future landscape of breast cancer treatment.

Longitudinal dynamic contrast-enhanced MRI radiomic models for early prediction of response to neoadjuvant systemic therapy in triple-negative breast cancer

Early prediction of neoadjuvant systemic therapy (NAST) response for triple-negative breast cancer (TNBC) patients could help oncologists select individualized treatment and avoid toxic effects associated with ineffective therapy in patients unlikely to achieve pathologic complete response (pCR). The objective of this study is to evaluate the performance of radiomic features of the peritumoral and tumoral regions from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) acquired at different time points of NAST for early treatment response prediction in TNBC. This study included 163 Stage I-III patients with TNBC undergoing NAST as part of a prospective clinical trial (NCT02276443). Peritumoral and tumoral regions of interest were segmented on DCE images at baseline (BL) and after two (C2) and four (C4) cycles of NAST. Ten first-order (FO) radiomic features and 300 gray-level-co-occurrence matrix (GLCM) features were calculated. Area under the receiver operating characteristic curve (AUC) and Wilcoxon rank sum test were used to determine the most predictive features. Multivariate logistic regression models were used for performance assessment. Pearson correlation was used to assess intrareader and interreader variability. Seventy-eight patients (48%) had pCR (52 training, 26 testing), and 85 (52%) had non-pCR (57 training, 28 testing). Forty-six radiomic features had AUC at least 0.70, and 13 multivariate models had AUC at least 0.75 for training and testing sets. The Pearson correlation showed significant correlation between readers. In conclusion, Radiomic features from DCE-MRI are useful for differentiating pCR and non-pCR. Similarly, predictive radiomic models based on these features can improve early noninvasive treatment response prediction in TNBC patients undergoing NAST.

An MRI-based machine learning radiomics can predict short-term response to neoadjuvant chemotherapy in patients with cervical squamous cell carcinoma: A multicenter study

BACKGROUND AND PURPOSE

Neoadjuvant chemotherapy (NACT) has become an essential component of the comprehensive treatment of cervical squamous cell carcinoma (CSCC). However, not all patients respond to chemotherapy due to individual differences in sensitivity and tolerance to chemotherapy drugs. Therefore, accurately predicting the sensitivity of CSCC patients to NACT was vital for individual chemotherapy. This study aims to construct a machine learning radiomics model based on magnetic resonance imaging (MRI) to assess its efficacy in predicting NACT susceptibility among CSCC patients.

METHODS

This study included 234 patients with CSCC from two hospitals, who were divided into a training set (n = 180), a testing set (n = 20), and an external validation set (n = 34). Manual radiomic features were extracted from transverse section MRI images, and feature selection was performed using the recursive feature elimination (RFE) method. A prediction model was then generated using three machine learning algorithms, namely logistic regression, random forest, and support vector machines (SVM), for predicting NACT susceptibility. The model's performance was assessed based on the area under the receiver operating characteristic curve (AUC), accuracy, and sensitivity.

RESULTS

The SVM approach achieves the highest scores on both the testing set and the external validation set. In the testing set and external validation set, the AUC of the model was 0.88 and 0.764, and the accuracy was 0.90 and 0.853, the sensitivity was 0.93 and 0.962, respectively.

CONCLUSIONS

Machine learning radiomics models based on MRI images have achieved satisfactory performance in predicting the sensitivity of NACT in CSCC patients with high accuracy and robustness, which has great significance for the treatment and personalized medicine of CSCC patients.

Characterization of Breast Tumors from MR Images Using Radiomics and Machine Learning Approaches

Determining histological subtypes, such as invasive ductal and invasive lobular carcinomas (IDCs and ILCs) and immunohistochemical markers, such as estrogen response (ER), progesterone response (PR), and the HER2 protein status is important in planning breast cancer treatment. MRI-based radiomic analysis is emerging as a non-invasive substitute for biopsy to determine these signatures. We explore the effectiveness of radiomics-based and CNN (convolutional neural network)-based classification models to this end. T1-weighted dynamic contrast-enhanced, contrast-subtracted T1, and T2-weighted MR images of 429 breast cancer tumors from 323 patients are used. Various combinations of input data and classification schemes are applied for ER⁺ vs. ER^-, PR⁺ vs. PR^-, HER2⁺ vs. HER2^-, and IDC vs. ILC classification tasks. The best results were obtained for the ER⁺ vs. ER^- and IDC vs. ILC classification tasks, with their respective AUCs reaching 0.78 and 0.73 on test data. The results with multi-contrast input data were generally better than the mono-contrast alone. The radiomics and CNN-based approaches generally exhibited comparable results. ER and IDC/ILC classification results were promising. PR and HER2 classifications need further investigation through a larger dataset. Better results by using multi-contrast data might indicate that multi-parametric quantitative MRI could be used to achieve more reliable classifiers.

Next generation immuno-oncology tumor profiling using a rapid, non-invasive, computational biophysics biomarker in early-stage breast cancer

Background

Immuno-oncology (IO) therapies targeting the PD-1/PD-L1 axis, such as immune checkpoint inhibitor (ICI) antibodies, have emerged as promising treatments for early-stage breast cancer (ESBC). Despite immunotherapy's clinical significance, the number of benefiting patients remains small, and the therapy can prompt severe immune-related events. Current pathologic and transcriptomic predictions of IO response are limited in terms of accuracy and rely on single-site biopsies, which cannot fully account for tumor heterogeneity. In addition, transcriptomic analyses are costly and time-consuming. We therefore constructed a computational biomarker coupling biophysical simulations and artificial intelligence-based tissue segmentation of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRIs), enabling IO response prediction across the entire tumor.

Methods

By analyzing both single-cell and whole-tissue RNA-seq data from non-IO-treated ESBC patients, we associated gene expression levels of the PD-1/PD-L1 axis with local tumor biology. PD-L1 expression was then linked to biophysical features derived from DCE-MRIs to generate spatially- and temporally-resolved atlases (virtual tumors) of tumor biology, as well as the TumorIO biomarker of IO response. We quantified TumorIO within patient virtual tumors (n = 63) using integrative modeling to train and develop a corresponding TumorIO Score.

Results

We validated the TumorIO biomarker and TumorIO Score in a small, independent cohort of IO-treated patients (n = 17) and correctly predicted pathologic complete response (pCR) in 15/17 individuals (88.2% accuracy), comprising 10/12 in triple negative breast cancer (TNBC) and 5/5 in HR+/HER2- tumors. We applied the TumorIO Score in a virtual clinical trial (n = 292) simulating ICI administration in an IO-naïve cohort that underwent standard chemotherapy. Using this approach, we predicted pCR rates of 67.1% for TNBC and 17.9% for HR+/HER2- tumors with addition of IO therapy; comparing favorably to empiric pCR rates derived from published trials utilizing ICI in both cancer subtypes.

Conclusion

The TumorIO biomarker and TumorIO Score represent a next generation approach using integrative biophysical analysis to assess cancer responsiveness to immunotherapy. This computational biomarker performs as well as PD-L1 transcript levels in identifying a patient's likelihood of pCR following anti-PD-1 IO therapy. The TumorIO biomarker allows for rapid IO profiling of tumors and may confer high clinical decision impact to further enable personalized oncologic care.

Brachytherapy outcome modeling in cervical cancer patients: A predictive machine learning study on patient-specific clinical, physical and dosimetric parameters

PURPOSE

To predict clinical response in locally advanced cervical cancer (LACC) patients by a combination of measures, including clinical and brachytherapy parameters and several machine learning (ML) approaches.

METHODS

Brachytherapy features such as insertion approaches, source metrics, dosimetric, and clinical measures were used for modeling. Four different ML approaches, including LASSO, Ridge, support vector machine (SVM), and Random Forest (RF), were applied to extracted measures for model development alone or in combination. Model performance was evaluated using the area under the curve (AUC) of receiver operating characteristics curve, sensitivity, specificity, and accuracy. Our results were compared with a reference model developed by simple logistic regression applied to three distinct clinical features identified by previous papers.

RESULTS

One hundred eleven LACC patients were included. Nine data sets were obtained based on the features, and 36 predictive models were built. In terms of AUC, the model developed using RF applied to dosimetric, physical, and total BT sessions features were found as the most predictive [AUC; 0.82 (0.95 confidence interval (CI); 0.79 -0.93), sensitivity; 0.79, specificity; 0.76, and accuracy; 0.77]. The AUC (0.95 CI), sensitivity, specificity, and accuracy for the reference model were found as 0.56 (0.52 ...0.68), 0.51, 0.51, and 0.48, respectively. Most RF models had significantly better performance than the reference model (Bonferroni corrected p-value < 0.0014).

CONCLUSION

Brachytherapy response can be predicted using dosimetric and physical parameters extracted from treatment parameters. Machine learning algorithms, including Random Forest, could play a critical role in such predictive modeling.