Correlation of androgen receptor with ultrasound, clinicopathological features and clinical outcomes in breast cancer.

BACKGROUND: This study aimed to explore whether there is an association between androgen receptor (AR) expression and ultrasound, clinicopathological features and prognosis of breast cancer. METHODS: A total of 141 breast cancer patients were included in this retrospective study. AR expression was analyzed by immunohistochemistry. The images of B-mode, color Doppler and strain elastography from 104 patients were collected continuously, and the corresponding ultrasound characteristics were obtained. The differences in ultrasound and clinicopathological features in different AR status were analyzed. Progression-free survival (PFS) of patients was obtained through up to 90 months of follow-up; then, the effect of AR on PFS was analyzed. Subsequently, a nomogram was constructed to predict the AR status. The predictive accuracy was calculated using C-index. RESULTS: The positive expression of AR (AR +) was associated with lower histological grade (p = 0.034) and lower Ki-67 level (p = 0.029). Triple-negative breast cancer (TNBC) had the lowest probability of AR + (p < 0.001). The AR + group mostly showed unsmooth margin (p < 0.001), posterior acoustic shadowing (p = 0.002) and higher elasticity score (p = 0.022) on ultrasound. The echo pattern of most tumors with AR + was heterogeneous (p = 0.024) in Luminal A subtype. AR + could be a sign of a better prognosis in overall breast cancer (p < 0.001), as well as in human epidermal growth factor receptor 2 (HER2) overexpression and Luminal B subtypes (p = 0.001 and 0.025). The nomogram showed relatively reliable performance with a C-index of 0.799. CONCLUSION: Our research demonstrated that AR expression was closely related to ultrasound, clinicopathological features and prognosis of breast cancer.

Radiogenomic analysis of prediction HER2 status in breast cancer by linking ultrasound radiomic feature module with biological functions.

BACKGROUND: Human epidermal growth factor receptor 2 (HER2) overexpressed associated with poor prognosis in breast cancer and HER2 has been defined as a therapeutic target for breast cancer treatment. We aimed to explore the molecular biological information in ultrasound radiomic features (URFs) of HER2-positive breast cancer using radiogenomic analysis. Moreover, a radiomics model was developed to predict the status of HER2 in breast cancer. METHODS: This retrospective study included 489 patients who were diagnosed with breast cancer. URFs were extracted from a radiomics analysis set using PyRadiomics. The correlations between differential URFs and HER2-related genes were calculated using Pearson correlation analysis. Functional enrichment of the identified URFs-correlated HER2 positive-specific genes was performed. Lastly, the radiomics model was developed based on the URF-module mined from auxiliary differential URFs to assess the HER2 status of breast cancer. RESULTS: Eight differential URFs (p < 0.05) were identified among the 86 URFs extracted by Pyradiomics. 25 genes that were found to be the most closely associated with URFs. Then, the relevant biological functions of each differential URF were obtained through functional enrichment analysis. Among them, Zone Entropy is related to immune cell activity, which regulate the generation of calcification in breast cancer. The radiomics model based on the Logistic classifier and URF-module showed good discriminative ability (AUC = 0.80, 95% CI). CONCLUSION: We searched for the URFs of HER2-positive breast cancer, and explored the underlying genes and biological functions of these URFs. Furthermore, the radiomics model based on the Logistic classifier and URF-module relatively accurately predicted the HER2 status in breast cancer.

Deep learning for the diagnosis of suspicious thyroid nodules based on multimodal ultrasound images.

Objectives: This study aimed to differentially diagnose thyroid nodules (TNs) of Thyroid Imaging Reporting and Data System (TI-RADS) 3-5 categories using a deep learning (DL) model based on multimodal ultrasound (US) images and explore its auxiliary role for radiologists with varying degrees of experience. Methods: Preoperative multimodal US images of 1,138 TNs of TI-RADS 3-5 categories were randomly divided into a training set (n = 728), a validation set (n = 182), and a test set (n = 228) in a 4:1:1.25 ratio. Grayscale US (GSU), color Doppler flow imaging (CDFI), strain elastography (SE), and region of interest mask (Mask) images were acquired in both transverse and longitudinal sections, all of which were confirmed by pathology. In this study, fivefold cross-validation was used to evaluate the performance of the proposed DL model. The diagnostic performance of the mature DL model and radiologists in the test set was compared, and whether DL could assist radiologists in improving diagnostic performance was verified. Specificity, sensitivity, accuracy, positive predictive value, negative predictive value, and area under the receiver operating characteristics curves (AUC) were obtained. Results: The AUCs of DL in the differentiation of TNs were 0.858 based on (GSU + SE), 0.909 based on (GSU + CDFI), 0.906 based on (GSU + CDFI + SE), and 0.881 based (GSU + Mask), which were superior to that of 0.825-based single GSU (p = 0.014, p< 0.001, p< 0.001, and p = 0.002, respectively). The highest AUC of 0.928 was achieved by DL based on (G + C + E + M)US, the highest specificity of 89.5% was achieved by (G + C + E)US, and the highest accuracy of 86.2% and sensitivity of 86.9% were achieved by DL based on (G + C + M)US. With DL assistance, the AUC of junior radiologists increased from 0.720 to 0.796 (p< 0.001), which was slightly higher than that of senior radiologists without DL assistance (0.796 vs. 0.794, p > 0.05). Senior radiologists with DL assistance exhibited higher accuracy and comparable AUC than that of DL based on GSU (83.4% vs. 78.9%, p = 0.041; 0.822 vs. 0.825, p = 0.512). However, the AUC of DL based on multimodal US images was significantly higher than that based on visual diagnosis by radiologists (p< 0.05). Conclusion: The DL models based on multimodal US images showed exceptional performance in the differential diagnosis of suspicious TNs, effectively increased the diagnostic efficacy of TN evaluations by junior radiologists, and provided an objective assessment for the clinical and surgical management phases that follow.

Nomograms predicting recurrence in patients with triple negative breast cancer based on ultrasound and clinicopathological features.

OBJECTIVES: The clinicopathological and ultrasound features associated with recurrence in patients with triple negative breast cancer (TNBC) were used to develop a nomogram to predict the prognosis of TNBC. METHODS: Clinicopathological data of 300 patients with TNBC treated between July 2012 and September 2014 were retrospectively reviewed. The endpoint was progression-free survival (PFS). Prognostic factors were screened by multivariate COX regression to develop nomograms. The C-index and calibration curves were used to evaluate the predictive accuracy and discriminatory ability of nomograms. RESULTS: Of 300 patients with TNBC followed-up for 5 years, 80 (26.7%) had PFS events. Five informative prognostic factors (large size, vertical orientation, posterior acoustic enhancement, lymph node involvement, and high pathological stage) were screened and used to construct a nomogram for PFS. The C-index of the PFS nomogram was 0.88 (p < 0.01, 95% confidence interval, 0.85-0.90), indicating good predictive accuracy. CONCLUSIONS: We developed and validated a nomogram for predicting PFS in TNBC. Vertical orientation and posterior acoustic enhancement in ultrasound images of TNBC were associated with worse outcomes. ADVANCES IN KNOWLEDGE: Patients with TNBC have a very poor prognosis and patients have a high risk of recurrence, and our study developed a nomogram based on ultrasound and clinicopathological features for TNBC patients to improve the accuracy of individualized prediction of recurrence and provide help for clinical treatment.

Predicting Pathological Complete Response After Neoadjuvant Chemotherapy in Advanced Breast Cancer by Ultrasound and Clinicopathological Features Using a Nomogram.

BACKGROUND AND AIMS: Prediction of pathological complete response (pCR) after neoadjuvant chemotherapy (NAC) for breast cancer is critical for surgical planning and evaluation of NAC efficacy. The purpose of this project was to assess the efficiency of a novel nomogram based on ultrasound and clinicopathological features for predicting pCR after NAC. METHODS: This retrospective study included 282 patients with advanced breast cancer treated with NAC from two centers. Patients received breast ultrasound before NAC and after two cycles of NAC; and the ultrasound, clinicopathological features and feature changes after two cycles of NAC were recorded. A multivariate logistic regression model was combined with bootstrapping screened for informative features associated with pCR. Then, we constructed two nomograms: an initial-baseline nomogram and a two-cycle response nomogram. Sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) were analyzed. The C-index was used to evaluate predictive accuracy. RESULTS: Sixty (60/282, 21.28%) patients achieved pCR. Triple-negative breast cancer (TNBC) and HER2-amplified types were more likely to obtain pCR. Size shrinkage, posterior acoustic pattern, and elasticity score were identified as independent factors by multivariate logistic regression. In the validation cohort, the two-cycle response nomogram showed better discrimination than the initial-baseline nomogram, with the C-index reaching 0.79. The sensitivity, specificity, and NPV of the two-cycle response nomogram were 0.77, 0.77, and 0.92, respectively. CONCLUSION: The two-cycle response nomogram exhibited satisfactory efficiency, which means that the nomogram was a reliable method to predict pCR after NAC. Size shrinkage after two cycles of NAC was an important in dependent factor in predicting pCR.

Identifying ultrasound features of positive expression of Ki67 and P53 in breast cancer using radiomics.

PURPOSE: To examine the relationship between ultrasonic findings and positive expression of Ki67 and P53 in breast cancer. MATERIAL AND METHODS: Surgical resection specimens of 263 breast cancer lesions were examined. Ultrasound examination and pathological examination were performed on each lesion for retrospective analysis. We applied regression analysis to the ultrasonic features related to the positive expression of Ki67 and P53 and obtained the corresponding models. To analyze diagnostic efficiency, we calculated the area under the curve (AUC). Additionally, we created a heat map to show the results of the cluster analysis. RESULTS: Lesions with higher Ki67 expression were associated with posterior acoustic enhancement, absence of an echo halo and a higher Breast Imaging Reporting and Data System (BI-RADS) category. P53-positive cancer were associated with an absence of an echo halo and a higher BI-RADS category. The AUC of the regression models of Ki67 and P53 was 0.78 and 0.71, respectively. CONCLUSIONS: Our study revealed that breast cancer ultrasonic findings were closely related to expression of molecular indicators, suggesting that ultrasound can be used to provide useful information to clinicians.

Inferences of Individual Drug Response-Related Long Non-coding RNAs Based on Integrating Multi-omics Data in Breast Cancer.

Differences in individual drug responses are obstacles in breast cancer (BRCA) treatment, so predicting responses would help to plan treatment strategies. The accumulation of cancer molecular profiling and drug response data provide opportunities and challenges to identify novel molecular signatures and mechanisms of tumor responsiveness to drugs in BRCA. This study evaluated drug responses with a multi-omics integrated system that depended on long non-coding RNAs (lncRNAs). We identified drug response-related lncRNAs (DRlncs) by combining expression data of lncRNA, microRNA, messenger RNA, methylation levels, somatic mutations, and the survival data of cancer patients treated with drugs. We constructed an integrated and computational multi-omics approach to identify DRlncs for diverse chemotherapeutic drugs in BRCA. Some DRlncs were identified with Adriamycin, Cytoxan, Tamoxifen, and all samples for BRCA patients. These DRlncs showed specific features regarding both expression and computational accuracies. The DRlnc-gene co-expression networks were constructed and analyzed. Key DRlncs, such as HOXA-AS2 (Ensembl: ENSG00000253552), in the drug Adriamycin were characterized. The experimental analysis also suggested that HOXA-AS2 (Ensembl: ENSG00000253552) was a key DRlnc in Adriamycin drug resistance in BRCA patients. Some DRlncs were associated with survival and some specific functions. A possible mechanism of DRlnc HOXA-AS2 (Ensembl: ENSG00000253552) in the Adriamycin drug response for BRCA resistance was inferred. In summary, this study provides a framework for lncRNA-based evaluation of clinical drug responses in BRCA. Understanding the underlying molecular mechanisms of drug responses will facilitate improved responses to chemotherapy and outcomes of BRCA treatment.