Bridging Discoveries and Treatments: The New Landscape of Breast Cancer Research.

Welcome to our Special Issue, "Advances in Breast Cancer Research and Treatment" of Life, where we have embarked on a comprehensive exploration of groundbreaking studies that advance our understanding and management of breast cancer [...].

MSGM: An Advanced Deep Multi-Size Guiding Matching Network for Whole Slide Histopathology Images Addressing Staining Variation and Low Visibility Challenges.

Matching whole slide histopathology images to provide comprehensive information on homologous tissues is beneficial for cancer diagnosis. However, the challenge arises with the Giga-pixel whole slide images (WSIs) when aiming for high-accuracy matching. Learning-based methods are difficult to generalize well with large-size WSIs, necessitating the integration of traditional matching methods to enhance accuracy as the size increases. In this paper, we propose a multi-size guiding matching method applicable high-accuracy requirements. Specifically, we design learning multiscale texture to train deep descriptors, called TDescNet, that trains 64 ×64×256 and 256 ×256×128 size convolution layer as C64 and C256 descriptors to overcome staining variation and low visibility challenges. Furthermore, we develop the 3D-ring descriptor using sparse keypoints to support the description of large-size WSIs. Finally, we employ C64, C256, and 3D-ring descriptors to progressively guide refined local matching, utilizing geometric consistency to identify correct matching results. Experiments show that when matching WSIs of size 4096×4096 pixels, our average matching error is 123.48 [Formula: see text] and the success rate is 93.02 % in 43 cases. Notably, our method achieves an average improvement of 65.52 [Formula: see text] in matching accuracy compared to recent state-of-the-art methods, with enhancements ranging from 36.27 [Formula: see text] to 131.66 [Formula: see text]. Therefore, we achieve high-fidelity whole-slice image matching, and overcome staining variation and low visibility challenges, enabling assistance in comprehensive cancer diagnosis through matched WSIs.

Genetic abnormalities assist in pathological diagnosis and EBV-positive cell density impact survival in Chinese angioimmunoblastic T-cell lymphoma patients.

Objective: To explore the application of genetic abnormalities in the diagnosis of angioimmunoblastic T-cell lymphoma (AITL) and the reliable pathological prognostic factors. Methods: This study included 53 AITL cases, which were reviewed for morphological patterns, immunophenotypes, presence of Hodgkin and Reed-Sternberg (HRS)-like cells, and co-occurrence of B cell proliferation. The Epstein-Barr virus (EBV)-positive cells in tissues were counted, and cases were classified into "EBV encoded RNA (EBER) high-density" group if >50/HPF. Targeted exome sequencing was performed. Results: Mutation data can assist AITL diagnosis: 1) with considerable HRS-like cells (20 cases): RHOA mutated in 14 cases (IDH2 co-mutated in 3 cases, 4 cases with rare RHOA mutation), TET2 was mutated in 5 cases (1 case co-mutated with DNMT3A), and DNMT3A mutated in 1 case; 2) accompanied with B cell lymphoma (7 cases): RHOA mutated in 4 cases (1 case had IDH2 mutation), TET2 mutated in 2 cases and DNMT3A mutated in 1 case; 3) mimic peripheral T cell lymphoma, not otherwise specified (5 cases): RHOA mutated in 2 cases (IDH2 co-mutated in 1 case), TET2 mutated in 3 cases, and DNMT3A mutated in 1 case; 4) pattern 1 (1 case), RHOA and TET2 co-mutated. Besides RHOAG17V (30/35), rare variant included RHOAK18N, RHOAR68H, RHOAC83Y, RHOAD120G and RHOAG17del, IDH2R172 co-mutated with IDH2M397V in one case. There were recurrent mutations of FAT3, PCLO and PIEZO1 and genes of epigenetic remodeling, T-cell activation, APC and PI3K/AKT pathway. EBER high-density independently indicated adverse overall survival and progression-free survival (P=0.046 and P=0.008, Kaplan-Meier/log-rank). Conclusions: Over half AITL cases might be confused in diagnosis for certain conditions without mutation data. Targeted exome sequencing with a comprehensive panel is crucial to detect both hot-spot and rare mutation variants for RHOA and IDH2 and other recurrent mutated genes in addition to TET2 and DNMT3A. EBER high-density independently indicated adverse survival.

An orientation-free ring feature descriptor with stain-variability normalization for pathology image matching.

Comprehensively analyzing the corresponding regions in the images of serial slices stained using different methods is a common but important operation in pathological diagnosis. To help increase the efficiency of the analysis, various image registration methods are proposed to match the corresponding regions in different images, but their performance is highly influenced by the rotations, deformations, and variations of staining between the serial pathology images. In this work, we propose an orientation-free ring feature descriptor with stain-variability normalization for pathology image matching. Specifically, we normalize image staining to similar levels to minimize the impact of staining differences on pathology image matching. To overcome the rotation and deformation issues, we propose a rotation-invariance orientation-free ring feature descriptor that generates novel adaptive bins from ring features to build feature vectors. We measure the Euclidean distance of the feature vectors to evaluate keypoint similarity to achieve pathology image matching. A total of 46 pairs of clinical pathology images in hematoxylin-eosin and immunohistochemistry straining to verify the performance of our method. Experimental results indicate that our method meets the pathology image matching accuracy requirements (error ¡ 300µm), especially competent for large-angle rotation cases common in clinical practice.

Loss of Human Epidermal Receptor 2 Expression in Formalin-Fixed Paraffin-Embedded Breast Cancer Samples and the Rescuing Effect of Enhanced Antigen Retrieval and Signal Amplification.

As an important therapeutic target in breast cancer, HER2 expression assessed by immunohistochemistry plays a critical role in breast cancer treatment. Recent advances in HER2 antibody-drug conjugate therapy have enabled patients with HER2-low expression breast cancer to benefit from the drugs. However, it is not known whether the HER2-low expression in breast cancer FFPE blocks would be lost as storage time increased. In this study, we aimed to assess the loss of HER2 antigenicity in stored FFPE blocks of breast cancer and the rescue effect of modifying the protocol of antigen staining. We selected archived HER2-low breast cancer FFPE blocks with stored time ranging from 1 year to over 15 years and re-detected the expression of HER2. Our study showed that HER2 antigenicity loss increased with storage time and could cause false negativity in HER2-low detection. Moreover, we showed that by either increasing the antigen retrieval time or applying the tyramide signal amplification (TSA) kit, the HER2 signal can be rescued and detected in about half of the cases with HER2-low loss without causing false positivity.

Using Whole Slide Gray Value Map to Predict HER2 Expression and FISH Status in Breast Cancer.

Accurate detection of HER2 expression through immunohistochemistry (IHC) is of great clinical significance in the treatment of breast cancer. However, manual interpretation of HER2 is challenging, due to the interobserver variability among pathologists. We sought to explore a deep learning method to predict HER2 expression level and gene status based on a Whole Slide Image (WSI) of the HER2 IHC section. When applied to 228 invasive breast carcinoma of no special type (IBC-NST) DAB-stained slides, our GrayMap+ convolutional neural network (CNN) model accurately classified HER2 IHC level with mean accuracy 0.952 ± 0.029 and predicted HER2 FISH status with mean accuracy 0.921 ± 0.029. Our result also demonstrated strong consistency in HER2 expression score between our system and experienced pathologists (intraclass correlation coefficient (ICC) = 0.903, Cohen's κ = 0.875). The discordant cases were found to be largely caused by high intra-tumor staining heterogeneity in the HER2 IHC group and low copy number in the HER2 FISH group.

Adverse Events of PD-1 or PD-L1 Inhibitors in Triple-Negative Breast Cancer: A Systematic Review and Meta-Analysis.

(1) Background: This study aimed to develop a comprehensive understanding of the treatment-related adverse events when using PD-1 or PD-L1 inhibitors in triple-negative breast cancer (TNBC). (2) Methods: We conducted a meta-analysis of Phase II/III randomized clinical trials. Studies were searched for using PubMed, Embase, and Cochrane Library from 1 March 1980 till 30 June 2022. Data on adverse events were mainly extracted from ClinicalTrials.gov and published articles. A generalized linear mixed model with the logit transformation was employed to obtain the overall incidence of adverse events across all studies. For serious adverse events with low incidences, the Peto method was used to calculate the odds ratio (OR) and 95% confidence interval (95%CI) in the PD-1 or PD-L1 inhibitors groups compared to the control groups. (3) Results: Nine studies were included in the meta-analysis, including a total of 2941 TNBC patients treated with PD-1 or PD-L1 inhibitors (including atezolizumab, pembrolizumab and durvalumab) and 2339 patients in the control groups. Chemotherapy alone was the control group in all studies. The average incidences of all serious immune-related adverse events of interest (hypothyroidism, hyperthyroidism, pneumonitis, pruritus, rash) were less than 1%, except for adrenal insufficiency (1.70%, 95%CI: 0.50-5.61%) in the PD-1 or PD-L1 groups. PD-1 or PD-L1 inhibitors significantly increased the risk of serious pneumonitis (OR = 2.52, 95%CI: 1.02-6.26), hypothyroidism (OR = 5.92, 95%CI: 1.22-28.86), alanine aminotransferase (ALT) elevation (OR = 1.66, 95%CI: 1.12-2.45), and adrenal insufficiency (OR = 18.81, 95%CI: 3.42-103.40). For non-serious adverse events, the patients treated with PD-1 or PD-L1 inhibitors had higher risk of aspartate aminotransferase (AST) elevation (OR =1.26, 95%CI: 1.02-1.57), hypothyroidism (OR = 3.63, 95%CI: 2.92-4.51), pruritus (OR = 1.84, 95%CI: 1.30-2.59), rash (OR = 1.29, 95%CI: 1.08-1.55), and fever (OR = 1.77, 95%CI: 1.13-2.77), compared with chemotherapy alone. (4) Conclusions: The incidence of serious immune-related adverse events in PD-1 or PD-L1 inhibitors groups is low but significantly higher than in chemotherapy groups. When using PD-1 or PD-L1 inhibitors for the treatment of TNBC, serious pneumonitis, hypothyroidism, ALT elevation, and adrenal insufficiency should be considered. Non-serious adverse events, such as AST elevation, rash, and fever, should also be taken into consideration.

AR Expression Correlates with Distinctive Clinicopathological and Genomic Features in Breast Cancer Regardless of ESR1 Expression Status.

Androgen receptor (AR) expression is frequently observed in breast cancer, but its association with estrogen receptor (ER) expression in breast cancer remains unclear. This study analyzed the clinicopathological and molecular features associated with AR negativity in both ER-positive and ER-negative breast cancer, trying to elucidate the molecular correlation between AR and ER. Our results showed that AR negativity was associated with different clinicopathological characteristics and molecular features in ER-positive and ER-negative breast cancer. Moreover, AR-positive breast cancer has better clinicopathological features than AR-negative breast cancer, especially in the ER-negative subtype. These results suggest that the role of AR in ER-negative breast cancer is distinctive from that in ER-positive breast cancer.

Two-Dose Vaccination Significantly Prolongs the Duration from Symptom Onset to Death: A Retrospective Study Based on 173,894 SARS-CoV-2 Cases in Khyber Pakhtunkhwa, Pakistan.

This research was carried out to quantify the duration from symptom onset to recovery/death (SOR/SOD) during the first four waves and the Alpha/Delta period of the epidemic in Khyber Pakhtunkhwa, Pakistan, and identify the associated factors. A total of 173,894 COVID-19 cases were admitted between 16 March 2020 and 30 November 2021, including 458 intensive care unit (ICU) cases. The results showed that the case fatality rate (CFR) increased with age, and females had a higher CFR. The median SOR of ICU cases was longer than that of non-ICU cases (27.6 vs. 17.0 days), while the median SOD was much shorter (6.9 vs. 8.4 days). The SOR and SOD in the Delta period were slightly shortened than the Alpha period. Age, cardiovascular diseases, chronic lung disease, diabetes, fever, breathing issues, and ICU admission were risk factors that were significantly associated with SOD (p < 0.001). A control measure, in-home quarantine, was found to be significantly associated with longer SOD (odds ratio = 9.49, p < 0.001). Infected vaccinated individuals had longer SOD than unvaccinated individuals, especially for cases that had received two vaccine doses (p < 0.001). Finally, an advice on getting full-dose vaccination is given specifically to individuals aged 20-59 years.

Comparative proteomic and clinicopathological analysis of breast adenoid cystic carcinoma and basal-like triple-negative breast cancer.

Background: Adenoid cystic carcinoma (ACC) is a rare type of triple-negative breast cancer that has an indolent clinical behavior. Given the substantial overlapping morphological, immunohistochemical, and molecular features with other basal-like triple-negative breast cancer (BL-TNBC), accurate diagnosis of ACC is crucial for effective clinical treatment. The integrative analysis of the proteome and clinicopathological characteristics may help to distinguish these two neoplasms and provide a deep understanding on biological behaviors and potential target therapy of ACC. Methods: We applied mass spectrometry-based quantitative proteomics to analyze the protein expression in paired tumor and adjacent normal breast tissue of five ACC and five BL-TNBC. Bioinformatic analyses and the clinicopathological characteristics, including histological features, immunohistochemistry, and FISH results, were also collected to get comprehensive information. Results: A total of 307 differentially expressed proteins (DEPs) were identified between ACC and BL-TNBC. Clustering analysis of DEPs clearly separated ACC from BL-TNBC. GSEA found downregulation of the immune response of ACC compared with BL-TNBC, which is consistent with the negative PD-L1 expression of ACC. Vesicle-mediated transport was also inhibited, while ECM organization was enriched in ACC. The top upregulated proteins in DEPs were ITGB4, VCAN, and DPT. Moreover, in comparison with normal breast tissue, ACC showed elevated ribosome biogenesis and RNA splicing activity. Conclusion: This study provides evidence that ACC presents a substantially different proteomic profile compared with BL-TNBC and promotes our understanding on the molecular mechanisms and biological processes of ACC, which might be useful for differential diagnosis and anticancer strategy.

Mechanism and effect of stress granule formation in cancer and its potential roles in breast cancer therapy.

Stress granules are non-membranous cytoplasmic foci induced by various stress conditions. It is a protective strategy used by cells to suppress overall translation during stress. In cancer cells, it was thought that the formation of stress granules could protect them from apoptosis and induces resistance towards anti-cancer drugs or radiation treatment which makes the stress granules a potential target for cancer treatment. However, most of our understanding of stress granules are still in the stage of molecular and cell biology, and a transitional gap for its actual effect on clinical settings remains. In this review, we summarize the mechanism and effect of stress granules formation in cancer and try to illuminate its potential applications in cancer therapy, using breast cancer as an example.

Expression of DNA Helicase Genes Was Correlated with Homologous Recombination Deficiency in Breast Cancer.

Homologous recombination deficiency which is currently measured by the homologous recombination deficiency (HRD) score including score of telomeric allelic imbalance (TAI), large-scale transition (LST), and loss of heterozygosity (LOH) is highly related with sensitivity to platinum-containing drug and PARP inhibitors. DNA helicases are essential components for the homologous recombination repair process in which DNA helicases unwind double-strand DNA utilizing ATP hydrolysis. In our study, the correlation between the expression of DNA helicase genes and HRD score in breast cancer was analyzed. The overexpression in half of the DNA helicase genes was found to be highly correlated with a high HRD score both in BRCA-mutated and BRCA wild-type breast cancer. Moreover, HRD score can be predicted by a linear function contributed by five DNA helicase genes. In conclusion, our study revealed a close relation between the overexpression of certain DNA helicase genes and the deficiency of homologous recombination repair in breast cancer.

BM-Net: CNN-Based MobileNet-V3 and Bilinear Structure for Breast Cancer Detection in Whole Slide Images.

Breast cancer is one of the most common types of cancer and is the leading cause of cancer-related death. Diagnosis of breast cancer is based on the evaluation of pathology slides. In the era of digital pathology, these slides can be converted into digital whole slide images (WSIs) for further analysis. However, due to their sheer size, digital WSIs diagnoses are time consuming and challenging. In this study, we present a lightweight architecture that consists of a bilinear structure and MobileNet-V3 network, bilinear MobileNet-V3 (BM-Net), to analyze breast cancer WSIs. We utilized the WSI dataset from the ICIAR2018 Grand Challenge on Breast Cancer Histology Images (BACH) competition, which contains four classes: normal, benign, in situ carcinoma, and invasive carcinoma. We adopted data augmentation techniques to increase diversity and utilized focal loss to remove class imbalance. We achieved high performance, with 0.88 accuracy in patch classification and an average 0.71 score, which surpassed state-of-the-art models. Our BM-Net shows great potential in detecting cancer in WSIs and is a promising clinical tool.

Detection of Structural Variations and Fusion Genes in Breast Cancer Samples Using Third-Generation Sequencing.

Background: Structural variations (SVs) are common genetic alterations in the human genome that could cause different phenotypes and diseases, including cancer. However, the detection of structural variations using the second-generation sequencing was limited by its short read length, which restrained our understanding of structural variations. Methods: In this study, we developed a 28-gene panel for long-read sequencing and employed it to Oxford Nanopore Technologies and Pacific Biosciences platforms. We analyzed structural variations in the 28 breast cancer-related genes through long-read genomic and transcriptomic sequencing of tumor, para-tumor, and blood samples in 19 breast cancer patients. Results: Our results showed that some somatic SVs were recurring among the selected genes, though the majority of them occurred in the non-exonic region. We found evidence supporting the existence of hotspot regions for SVs, which extended our previous understanding that they exist only for single nucleotide variations. Conclusion: In conclusion, we employed long-read genomic and transcriptomic sequencing to identify SVs from breast cancer patients and proved that this approach holds great potential in clinical application.

The natural history of breast cancer: a chronological analysis of breast cancer progression using data from the SEER database.

Background: Numerous mathematical models and retrospective analyses have been conducted to explore the natural progression of breast cancer, but there is no precise timeline as to how long it takes breast cancer to progress from stages 0 to IV. Studying the natural history of breast cancer requires the follow-up of a large cohort of patients who received no treatment following diagnosis for a long period, but this has become unpractical in the modern era. In this study, we quantified the natural progression of breast cancer using the Surveillance, Epidemiology, and End Results (SEER) database which was collected the clinical-pathological characteristics and survival data of untreated breast cancer patients. Methods: We included 12,687 female patients who did not receive surgical treatment, radiotherapy, or chemotherapy, had complete information on tumor staging, and had only 1 primary tumor in their lifetime in our study as the untreated group. Overall survival (OS) was calculated from the date of the latest diagnosis of breast cancer to the date of death. Medium survival time (MST) was defined as the time at which 50% of the patients reached the endpoint. The medium survival time (MST) was calculated for each stage. The progression time was calculated by subtracting the MST of a higher stage from that of a lower stage, which also indicated "life loss". Estrogen receptor (ER) negative and positive patients were analyzed as two separate groups. Results: For the ER positive patients, the median progression times from stages 0 to I, I to II, II to III, and III to IV were 5.2±1.2, 4.2±0.5, 2±0.3, and 0.9±0.2 years, respectively, which were similar to those of the ER negative patients of 5.4±2.9, 4.7±1.3, 2.5±0.7, and 1.1±0.4 years (P=0.49). The survival of stage IV patients was longer in ER positive patients than ER negative patients (2.7±0.1 vs. 0.9±0.2 year, P<0.001). Conclusions: ER negative patients appear to have the same progression time as that of ER positive patients in the primary site, but had a much worse process in the metastasis site.

Intraparenchymal breast leiomyoma and atypical leiomyoma.

BACKGROUND: Breast leiomyoma is a rare benign mesenchymal tumor, accounting for less than 1% of all breast neoplasms. Cases of breast atypical leiomyoma is even more rarely reported and its diagnostic criteria together with its clinical courses is not cleared defined. CASE PRESENTATION: We described two patients with breast leiomyomas. One has unilateral benign breast leiomyoma, the other one has bilateral breast leiomyomas. For the bilateral case, the left-side tumor was diagnosed as benign leiomyoma while the right-side tumor was diagnosed as atypical leiomyoma. The morphological features that lead to the diagnosis of atypical leiomyoma are its invasive growth pattern, mild nuclear atypia, and mitotic figures up to 3mitoses/10HPF. CONCLUSIONS: Atypical breast leiomyoma appears to behave like benign leiomyoma without recurrence in our study with nine-year follow-up. Due to the limited experience, cases presented as atypical intraparenchymal breast leiomyoma should be closely followed.

Modeling effective tumor burden of primary lesion and metastatic lymph node in breast cancer patients from the SEER database.

BACKGROUND: Characterizing the role of the primary tumor and axillary lymph nodes (ALNs) in metastasis progression can provide a rational basis for effective local control and systemic treatments. METHODS: We employed data from the Surveillance, Epidemiology, and End Results (SEER) database to perform survival analysis and modeling to estimate breast cancer (BC) progression in both untreated and treated patients. Effective tumor burden was defined as the number of cancer cells that can lead to distant metastasis. Quantified analysis was conducted for primary lesion and metastatic nodes, respectively. A tumor growth model was built based on mathematical modeling: N ( t ) = m N T ( t ) + n N N ( t ) . In this model, the parameters "m" and "n" represent the contributions of the primary lesion and metastatic lymph nodes (LNs) to distant metastasis, respectively. RESULTS: We found that both the primary lesion and the metastatic LNs contribute to the effective tumor burden. Through fitting the built tumor growth model, "m" in treated groups was determined to be 1, and "n" was determined to be 1.5; while in the untreated group, "m" was determined to be 1, and "n" was 0.15, which was 10 times smaller than in treated groups. CONCLUSIONS: Our study revealed that the prognostic value of the anatomic stage in BC progression varied in different historical periods, due to the development of oncological treatments. While tumor size was a significant prognostic factor in both untreated and treated patients, the prognostic value of nodal status was only significant in patients who received locoregional treatment and systemic treatment.

Triple-Negative Apocrine Breast Carcinoma Has Better Prognosis despite Poor Response to Neoadjuvant Chemotherapy.

Apocrine carcinoma is a rare subtype of invasive ductal breast cancer that shows apocrine differentiation and largely triple-negative immunohistology. Triple-negative breast cancers are known to have more aggressive clinical courses. However, unlike most other subtypes, it is reported that triple-negative apocrine carcinoma (TNAC) has a better prognosis. Due to the scarcity of reported studies, our knowledge regarding its clinical behavior, prognosis and response to therapy is very limited. In this study, we retrospectively retrieved 41 triple-negative apocrine carcinoma cases from our breast cancer database, with an average follow-up of 32.8 months. It was found that TNAC had a poorer response to neoadjuvant therapy but a better prognosis than other nonapocrine types of triple-negative breast cancer. Meanwhile, TNAC has a low proliferative nature, as indicated by its low Ki-67 index. An updated analysis of the Surveillance, Epidemiology, and End Results database showed that chemotherapy did not improve breast-cancer-specific survival in TNAC patients. Our results suggest that TNAC is a special subtype of triple-negative breast cancer with a better short-term prognosis despite poor response to neoadjuvant chemotherapy.

Biomarker Alteration after Neoadjuvant Endocrine Therapy or Chemotherapy in Estrogen Receptor-Positive Breast Cancer.

In estrogen receptor (ER)-positive breast cancer, changes in biomarker expression after neoadjuvant therapy indicate the therapeutic response and are prognostic. However, there is limited information about the biomarker alteration caused by neoadjuvant endocrine therapy in ER-positive and human epidermal growth factor receptor 2 (HER2)-negative breast cancer. We recruited ER-positive/HER2-negative breast cancer patients who received neoadjuvant chemotherapy (NCT), neoadjuvant endocrine therapy (NET), or sequential neoadjuvant endocrine-chemotherapy (NECT) at Peking University Cancer Hospital from 2015 to 2021. A total of 579 patients had paired immunohistochemistry information in both diagnostic biopsy samples and post-neoadjuvant therapy surgical samples. Through a paired comparison of the immunohistochemical information in pre-treatment and post-treatment samples, we found that progesterone receptor (PR) expression reductions were more frequent than ER expression reductions (70.8% vs. 35.2%) after neoadjuvant therapy. The percentage of patients who had a decreased Ki-67 index in the post-operative samples was similar in the three groups (79.8% vs. 79.7% vs. 78.4%). Moreover, PR losses caused by NET were related to low baseline PR expression (p = 0.001), while we did not find a significant association between PR losses and Ki-67 reductions (p = 0.428) or ER losses (p = 0.274). All three types of neoadjuvant therapies caused a reduction in ER, PR, and Ki-67 expression. In conclusion, we found that PR loss after NET was only significantly related to low baseline PR expression, and there is no significant difference in the extent of prognostic factor change including Ki-67 and ER between the PR loss and non-loss groups.

A Machine Learning Approach to Differentiate Two Specific Breast Cancer Subtypes Using Androgen Receptor Pathway Genes.

Triple-negative breast cancer is a heterogeneous disease with different molecular and histological subtypes. The Androgen receptor is expressed in a portion of triple-negative breast cancer cases and the activation of the androgen receptor pathway is thought to be a molecular subtyping signature as well as a therapeutic target for triple-negative breast cancer. Thus, identification of the androgen receptor pathway status is important for both molecular characterization andclinical management. In this study, we investigate the expression of the androgen receptor pathway in metaplastic breast cancer and luminal androgen receptor subtypes of triple-negative breast cancer and found that the androgen receptor pathway was downregulated in metaplastic breast cancer compared to luminal androgen receptor subtype. Using random forest, we found that the two subtypes of breast cancer can be molecularly classified with the gene expression of the androgen receptor pathway.