Effect of Trastuzumab among HER2-Positive Breast Cancer Patients that Achieved Pathologic Complete Response after Neoadjuvant Chemotherapy

Application of machine learning with multiparametric dual-energy computed tomography of the breast to differentiate between benign and malignant lesions

BACKGROUND

Multiparametric dual-energy computed tomography (mpDECT) is widely used to differentiate various kinds of tumors; however, the data regarding its diagnostic performance with machine learning to diagnose breast tumors is limited. We evaluated univariate analysis and machine learning performance with mpDECT to distinguish between benign and malignant breast lesions.

METHODS

In total, 172 patients with 214 breast lesions (55 benign and 159 malignant) who underwent preoperative dual-phase contrast-enhanced DECT were included in this retrospective study. Twelve quantitative features were extracted for each lesion, including CT attenuation (precontrast, arterial, and venous phases), the arterial-venous phase difference in normalized effective atomic number (nZ_eff), normalized iodine concentration (NIC), and slope of the spectral Hounsfield unit (HU) curve (λ_Hu). Predictive models were developed using univariate analysis and eight machine learning methods [logistic regression, extreme gradient boosting (XGBoost), stochastic gradient descent (SGD), linear discriminant analysis (LDA), adaptive boosting (AdaBoost), random forest (RF), decision tree, and linear support vector machine (SVM)]. Classification performances were assessed based on the area under the receiver operating characteristic curve (AUROC). The best performances of the conventional univariate analysis and machine learning methods were compared using the Delong test.

RESULTS

The univariate analysis showed that the venous phase λ_Hu had the highest AUROC (0.88). Machine learning with mpDECT achieved an excellent and stable diagnostic performance, as shown by the mean classification performances in the training dataset (AUROC, 0.88-0.99) and testing (AUROC, 0.83-0.96) datasets. The performance of the AdaBoost model based on mpDECT was more stable than the other machine learning models and superior to the univariate analysis (AUROC, 0.96 vs. 0.88; P<0.001).

CONCLUSIONS

The performance of the AdaBoost classifier based on mpDECT data achieved the highest mean accuracy compared to the other machine learning models and univariate analysis in differentiating between benign and malignant breast lesions.

Comparison of breast cancer surrogate subtyping using a closed-system RT-qPCR breast cancer assay and immunohistochemistry on 100 core needle biopsies with matching surgical specimens

Background

Routine clinical management of breast cancer (BC) currently depends on surrogate subtypes according to estrogen- (ER) and progesterone (PR) receptor, Ki-67, and HER2-status. However, there has been growing demand for reduced immunohistochemistry (IHC) turnaround times. The Xpert® Breast Cancer STRAT4* Assay (STRAT4)*, a standardized test for ESR1/PGR/MKi67/ERBB2 mRNA biomarker assessment, takes less than 2 hours. Here, we compared the concordance between the STRAT4 and IHC/SISH, thereby evaluating the effect of method choice on surrogate subtype assessment and adjuvant treatment decisions.

Methods

In total, 100 formalin-fixed paraffin-embedded core needle biopsy (CNB) samples and matching surgical specimens for 98 patients with primary invasive BC were evaluated using the STRAT4 assay. The concordance between STRAT4 and IHC was calculated for individual markers for the CNB and surgical specimens. In addition, we investigated whether changes in surrogate BC subtyping based on the STRAT4 results would change adjuvant treatment recommendations.

Results

The overall percent agreement (OPA) between STRAT4 and IHC/SISH ranged between 76 and 99% for the different biomarkers. Concordance for all four biomarkers in the surgical specimens and CNBs was only 66 and 57%, respectively. In total, 74% of surgical specimens were concordant for subtype, regardless of the method used. IHC- and STRAT4-based subtyping for the surgical specimen were shown to be discordant for 25/98 patients and 18/25 patients would theoretically have been recommended a different adjuvant treatment, primarily receiving more chemotherapy and trastuzumab.

Conclusions

A comparison of data from IHC/in situ hybridization and STRAT4 demonstrated that subsequent changes in surrogate subtyping for the surgical specimen may theoretically result in more adjuvant treatment given, primarily with chemotherapy and trastuzumab.