Concordance of the molecular subtype classification between core needle biopsy and surgical specimen in primary breast cancer

Multiparametric MRI-based Interpretable Machine Learning Radiomics Model for Distinguishing Between Luminal and Non-luminal Tumors in Breast Cancer: A Multicenter Study

RATIONALE AND OBJECTIVES

To construct and validate an interpretable machine learning (ML) radiomics model derived from multiparametric magnetic resonance imaging (MRI) images to differentiate between luminal and non-luminal breast cancer (BC) subtypes.

METHODS

This study enrolled 1098 BC participants from four medical centers, categorized into a training cohort (n = 580) and validation cohorts 1-3 (n = 252, 89, and 177, respectively). Multiparametric MRI-based radiomics features, including T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC), and dynamic contrast-enhanced (DCE) imaging, were extracted. Five ML algorithms were applied to develop various radiomics models, from which the best performing model was identified. A ML-based combined model including optimal radiomics features and clinical predictors was constructed, with performance assessed through receiver operating characteristic (ROC) analysis. The Shapley additive explanation (SHAP) method was utilized to assess model interpretability.

RESULTS

Tumor size and MR-reported lymph node status were chosen as significant clinical variables. Thirteen radiomics features were identified from multiparametric MRI images. The extreme gradient boosting (XGBoost) radiomics model performed the best, achieving area under the curves (AUCs) of 0.941, 0.903, 0.862, and 0.894 across training and validation cohorts 1-3, respectively. The XGBoost combined model showed favorable discriminative power, with AUCs of 0.956, 0.912, 0.894, and 0.906 in training and validation cohorts 1-3, respectively. The SHAP visualization facilitated global interpretation, identifying "ADC_wavelet-HLH_glszm_ZoneEntropy" and "DCE_wavelet-HLL_gldm_DependenceVariance" as the most significant features for the model's predictions.

CONCLUSION

The XGBoost combined model derived from multiparametric MRI may proficiently differentiate between luminal and non-luminal BC and aid in treatment decision-making.

CRITICAL RELEVANCE STATEMENT

An interpretable machine learning radiomics model can preoperatively predict luminal and non-luminal subtypes in breast cancer, thereby aiding therapeutic decision-making.

Machine learning radiomics based on intra and peri tumor PA/US images distinguish between luminal and non-luminal tumors in breast cancers

PURPOSE

This study aimed to evaluate a radiomics model using Photoacoustic/ultrasound (PA/US) imaging at intra and peri-tumoral area to differentiate Luminal and non-Luminal breast cancer (BC) and to determine the optimal peritumoral area for accurate classification.

MATERIALS AND METHODS

From February 2022 to April 2024, this study continuously collected 322 patients at Shenzhen People's Hospital, using standardized conditions for PA/US imaging of BC. Regions of interest were delineated using ITK-SNAP, with peritumoral regions of 2 mm, 4 mm, and 6 mm automatically expanded using code from the Pyradiomic package. Feature extraction was subsequently performed using Pyradiomics. The study employed Z-score normalization, Spearman correlation for feature correlation, and LASSO regression for feature selection, validated through 10-fold cross-validation. The radiomics model integrated intra and peri-tumoral area, evaluated by receiver operating characteristic curve(ROC), Calibration and Decision Curve Analysis(DCA).

RESULTS

We extracted and selected features from intratumoral and peritumoral PA/US images regions at 2 mm, 4 mm, and 6 mm. The comprehensive radiomics model, integrating these regions, demonstrated enhanced diagnostic performance, especially the 4 mm model which showed the highest area under the curve(AUC):0.898(0.78-1.00) and comparably high accuracy (0.900) and sensitivity (0.937). This model outperformed the standalone clinical model and combined clinical-radiomics model in distinguishing between Luminal and non-Luminal BC, as evidenced in the test set results.

CONCLUSION

This study developed a radiomics model integrating intratumoral and peritumoral at 4 mm region PA/US model, enhancing the differentiation of Luminal from non-Luminal BC. It demonstrated the diagnostic utility of peritumoral characteristics, reducing the need for invasive biopsies and aiding chemotherapy planning, while emphasizing the importance of optimizing tumor surrounding size for improved model accuracy.

Ultrasound-based radiomics nomogram for predicting HER2-low expression breast cancer

Purpose

Accurate preoperative identification of Human epidermal growth factor receptor 2 (HER2) low expression breast cancer (BC) is critical for clinical decision-making. Our aim was to use machine learning methods to develop and validate an ultrasound-based radiomics nomogram for predicting HER2-low expression in BC.

Methods

In this retrospective study, 222 patients (108 HER2-0 expression and 114 HER2-low expression) with BC were included. The enrolled patients were randomly divided into a training cohort and a test cohort with a ratio of 8:2. The tumor region of interest was manually delineated from ultrasound image, and radiomics features were subsequently extracted. The features underwent dimension reduction using the least absolute shrinkage and selection operator (LASSO) algorithm, and rad-score were calculated. Five machine learning algorithms were applied for training, and the algorithm demonstrating the best performance was selected to construct a radiomics (USR) model. Clinical risk factors were integrated with rad-score to construct the prediction model, and a nomogram was plotted. The performance of the nomogram was assessed using receiver operating characteristic curve and decision curve analysis.

Results

A total of 480 radiomics features were extracted, out of which 11 were screened out. The majority of the extracted features were wavelet features. Subsequently, the USR model was established, and rad-scores were computed. The nomogram, incorporating rad-score, tumor shape, border, and microcalcification, achieved the best performance in both the training cohort (AUC 0.89; 95%CI 0.836-0.936) and the test cohort (AUC 0.84; 95%CI 0.722-0.958), outperforming both the USR model and clinical model. The calibration curves showed satisfactory consistency, and DCA confirmed the clinical utility of the nomogram.

Conclusion

The nomogram model based on ultrasound radiomics exhibited high prediction value for HER2-low BC.

MRI features of breast cancer immunophenotypes with a focus on luminal estrogen receptor low positive invasive carcinomas

To compare the magnetic resonance imaging (MRI) features of different immunophenotypes of breast carcinoma of no special type (NST), with special attention to estrogen receptor (ER)-low-positive breast cancer. This retrospective, single-centre, Institutional Review Board (IRB)-approved study included 398 patients with invasive breast carcinoma. Breast carcinomas were classified as ER-low-positive when there was ER staining in 1-10% of tumour cells. Pretreatment MRI was reviewed to assess the tumour imaging features according to the 5th edition of the Breast Imaging Reporting and Data System (BI-RADS) lexicon. Of the 398 cases, 50 (12.6%) were luminal A, 191 (48.0%) were luminal B, 26 (6.5%) were luminal ER-low positive, 64 (16.1%) were HER2-overexpressing, and 67 (16.8%) were triple negative. Correlation analysis between MRI features and tumour immunophenotype showed statistically significant differences in mass shape, margins, internal enhancement and the delayed phase of the kinetic curve. An oval or round shape and rim enhancement were most frequently observed in triple-negative and luminal ER-low-positive tumours. Spiculated margins were most common in luminal A and luminal B tumours. A persistent kinetic curve was more frequent in luminal A tumours, while a washout curve was more common in the triple-negative, HER2-overexpressing and luminal ER-low-positive immunophenotypes. Multinomial regression analysis showed that luminal ER-low-positive tumours had similar results to triple-negative tumours for almost all variables. Luminal ER-low-positive tumours present with similar MRI findings to triple-negative tumours, which suggests that MRI can play a fundamental role in adequate radiopathological correlation and therapeutic planning in these patients.

Predicting hormone receptor status in invasive breast cancer through radiomics analysis of long-axis and short-axis ultrasound planes

The hormone receptor (HR) status plays a significant role in breast cancer, serving as the primary guide for treatment decisions and closely correlating with prognosis. This study aims to investigate the predictive value of radiomics analysis in long-axis and short-axis ultrasound planes for distinguishing between HR-positive and HR-negative breast cancers. A cohort of 505 patients from two hospitals was stratified into discovery (Institute 1, 416 patients) and validation (Institute 2, 89 patients) cohorts. A comprehensive set of 788 ultrasound radiomics features was extracted from both long-axis and short-axis ultrasound planes, respectively. Utilizing least absolute shrinkage and selection operator (LASSO) regression analysis, distinct models were constructed for the long-axis and short-axis data. Subsequently, radiomics scores (Rad-scores) were computed for each patient. Additionally, a combined model was formulated by integrating data from long-axis and short-axis Rad-scores along with clinical factors. The diagnostic efficacy of all models was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC). The long-axis and short-axis models, consisting of 11 features and 15 features, respectively, were established, yielding AUCs of 0.743 and 0.751 in the discovery cohort, and 0.795 and 0.744 in the validation cohort. The calculated long-axis and short-axis Rad-scores exhibited significant differences between HR-positive and HR-negative groups across all cohorts (all p < 0.001). Univariate analysis identified ultrasound-reported tumor size as an independent predictor. The combined model, incorporating long-axis and short-axis Rad-scores along with tumor size, achieved superior AUCs of 0.788 and 0.822 in the discovery and validation cohorts, respectively. The combined model effectively distinguishes between HR-positive and HR-negative breast cancers based on ultrasound radiomics features and tumor size, which may offer a valuable tool to facilitate treatment decision making and prognostic assessment.

The Clinical and Biological Significance of Estrogen Receptor-Low Positive Breast Cancer

Estrogen receptor (ER) status in breast cancer (BC) is determined using immunohistochemistry (IHC) with nuclear expression in ≥1% of cells defined as ER-positive. BC with 1%-9% expression (ER-low-positive), is a clinically and biologically unique subgroup. In this study, we hypothesized that ER-low-positive BC represents a heterogeneous group with a mixture of ER-positive and ER-negative tumor, which may explain their divergent clinical behavior. A large BC cohort (n = 8171) was investigated and categorized into 3 groups: ER-low-positive (1%-9%), ER-positive (≥10%), and ER-negative (<1%) where clinicopathological and outcome characteristics were compared. A subset of ER-low-positive cases was further evaluated using IHC, RNAscope, and RT-qPCR. PAM50 subtyping and ESR1 mRNA expression levels were assessed in ER-low-positive cases within The Cancer Genome Atlas data set. The reliability of image analysis software in assessment of ER expression in the ER-low-positive category was also assessed. ER-low-positive tumors constituted <2% of BC cases examined and showed significant clinicopathological similarity to ER-negative tumors. Most of these tumors were nonluminal types showing low ESR1 mRNA expression. Further validation of ER status revealed that 45% of these tumors were ER-negative with repeated IHC staining and confirmed by RNAscope and RT-qPCR. ER-low-positive tumors diagnosed on needle core biopsy were enriched with false-positive ER staining. BCs with 10% ER behaved similar to ER-positive, rather than ER-negative or low-positive BCs. Moderate concordance was found in assessment of ER-low-positive tumors, and this was not improved by image analysis. Routinely diagnosed ER-low-positive BC includes a proportion of ER-negative cases. We recommend repeat testing of BC showing 1%-9% ER expression and using a cutoff ≥10% expression to define ER positivity to help better inform treatment decisions.

Deep learning radiopathomics based on preoperative US images and biopsy whole slide images can distinguish between luminal and non-luminal tumors in early-stage breast cancers

BACKGROUND

For patients with early-stage breast cancers, neoadjuvant treatment is recommended for non-luminal tumors instead of luminal tumors. Preoperative distinguish between luminal and non-luminal cancers at early stages will facilitate treatment decisions making. However, the molecular immunohistochemical subtypes based on biopsy specimens are not always consistent with final results based on surgical specimens due to the high intra-tumoral heterogeneity. Given that, we aimed to develop and validate a deep learning radiopathomics (DLRP) model to preoperatively distinguish between luminal and non-luminal breast cancers at early stages based on preoperative ultrasound (US) images, and hematoxylin and eosin (H&E)-stained biopsy slides.

METHODS

This multicentre study included three cohorts from a prospective study conducted by our team and registered on the Chinese Clinical Trial Registry (ChiCTR1900027497). Between January 2019 and August 2021, 1809 US images and 603 H&E-stained whole slide images (WSIs) from 603 patients with early-stage breast cancers were obtained. A Resnet18 model pre-trained on ImageNet and a multi-instance learning based attention model were used to extract the features of US and WSIs, respectively. An US-guided Co-Attention module (UCA) was designed for feature fusion of US and WSIs. The DLRP model was constructed based on these three feature sets including deep learning US feature, deep learning WSIs feature and UCA-fused feature from a training cohort (1467 US images and 489 WSIs from 489 patients). The DLRP model's diagnostic performance was validated in an internal validation cohort (342 US images and 114 WSIs from 114 patients) and an external test cohort (270 US images and 90 WSIs from 90 patients). We also compared diagnostic efficacy of the DLRP model with that of deep learning radiomics model and deep learning pathomics model in the external test cohort.

FINDINGS

The DLRP yielded high performance with area under the curve (AUC) values of 0.929 (95% CI 0.865-0.968) in the internal validation cohort, and 0.900 (95% CI 0.819-0.953) in the external test cohort. The DLRP also outperformed deep learning radiomics model based on US images only (AUC 0.815 [0.719-0.889], p = 0.027) and deep learning pathomics model based on WSIs only (AUC 0.802 [0.704-0.878], p = 0.013) in the external test cohort.

INTERPRETATION

The DLRP can effectively distinguish between luminal and non-luminal breast cancers at early stages before surgery based on pretherapeutic US images and biopsy H&E-stained WSIs, providing a tool to facilitate treatment decision making in early-stage breast cancers.

FUNDING

Natural Science Foundation of Guangdong Province (No. 2023A1515011564), and National Natural Science Foundation of China (No. 91959127; No. 81971631).

Breast Magnetic Resonance Imaging Can Predict Ki67 Discordance Between Core Needle Biopsy and Surgical Samples

BACKGROUND

Disagreement in assessments of Ki67 expression based on core-needle biopsy and matched surgical samples complicates decisions in the treatment of breast cancer.

PURPOSE

To examine whether preoperative breast MRI could be useful in predicting Ki67 discordance between core-needle biopsy and surgical samples.

STUDY TYPE

Retrospective.

POPULATION

Three hundred and sixty-five breast cancer patients with MRI scans and having both core-needle biopsy and surgical samples from 2017 to 2019.

FIELD STRENGTH/SEQUENCE

3.0 T, T2-weighted iterative decomposition of water and fat with echo asymmetry and least squares estimation sequence, diffusion-weighted sequence using b-values 0/1000, dynamic contrast enhanced image by volume imaged breast assessme NT.

ASSESSMENT

We collected clinicopathologic variables and preoperative MRI features (tumor size, lesion type, shape of mass, spiculated margin, internal enhancement, peri-tumoral edema, intra-tumoral necrosis, multifocal/multicentric, apparent diffusion coefficient [ADC] minimum, ADC mean, ADC maximum, ADC difference).

STATISTICAL TESTS

K-means clustering, multivariable logistic regression, receiver operating characteristic curve.

RESULTS

Sixty-one patients showed Ki67 discordance and 304 patients show Ki67 concordance according to our definition using K-means clustering. Multivariable regression analysis showed that the following parameters were independently associated with Ki67 discordance: peri-tumoral edema, odds ratio (OR) 2.662, 95% confidence interval (CI) 1.432-4.948; ADCmin ≤ 0.829 × 10^-3  mm² /sec, OR 2.180, 95% CI 1.075-4.418; and ADCdiff > 0.317 × 10^-3  mm² /sec, OR 3.365, 95% CI 1.698-6.669. This multivariable model resulted in an AUC of 0.758 (95% CI 0.711-0.802) with sensitivity and specificity being 0.803 and 0.621, respectively.

CONCLUSION

Presence of peri-tumoral edema, smaller ADCmin and greater ADCdiff in preoperative breast MRI may indicate high risk of Ki67 discordance between core-needle biopsy and surgical samples. For patients with these MRI-based risk factors, clinicians should not rely on Ki67 assessment only from core-needle biopsy.

Systematically higher Ki67 scores on core biopsy samples compared to corresponding resection specimen in breast cancer: a multi-operator and multi-institutional study

Ki67 has potential clinical importance in breast cancer but has yet to see broad acceptance due to inter-laboratory variability. Here we tested an open source and calibrated automated digital image analysis (DIA) platform to: (i) investigate the comparability of Ki67 measurement across corresponding core biopsy and resection specimen cases, and (ii) assess section to section differences in Ki67 scoring. Two sets of 60 previously stained slides containing 30 core-cut biopsy and 30 corresponding resection specimens from 30 estrogen receptor-positive breast cancer patients were sent to 17 participating labs for automated assessment of average Ki67 expression. The blocks were centrally cut and immunohistochemically (IHC) stained for Ki67 (MIB-1 antibody). The QuPath platform was used to evaluate tumoral Ki67 expression. Calibration of the DIA method was performed as in published studies. A guideline for building an automated Ki67 scoring algorithm was sent to participating labs. Very high correlation and no systematic error (p = 0.08) was found between consecutive Ki67 IHC sections. Ki67 scores were higher for core biopsy slides compared to paired whole sections from resections (p ≤ 0.001; median difference: 5.31%). The systematic discrepancy between core biopsy and corresponding whole sections was likely due to pre-analytical factors (tissue handling, fixation). Therefore, Ki67 IHC should be tested on core biopsy samples to best reflect the biological status of the tumor.