PAM50- and immunohistochemistry-based subtypes of breast cancer and their relationship with breast cancer mortality in a population-based study

Identification and validation of a novel necroptosis-related molecular signature to evaluate prognosis and immune features in breast cancer

Necroptosis has been shown to play an important role in the development of tumors. However, the characteristics of the necroptosis-related subtypes and the associated immune cell infiltration in the tumor microenvironment (TME) of breast cancer (BRCA) remain unclear. In this study, we identified three clusters related to necroptosis using the expression patterns of necroptosis-relevant genes (NRGs), and found that these three clusters had different clinicopathological features, prognosis and immune cell infiltration in the TME. Cluster 2 was characterized by less infiltration of immune cells in the TME and was associated with a worse prognosis. Then, a necroptosis risk score (NRS) composed of 14 NRGs was constructed using the least absolute shrinkage and selection operator regression (LASSO) Cox regression method. Based on NRS, all BRCA patients in the TCGA datasets were classified into a low-risk group and a high-risk group. Patients in the low-risk group were characterized by longer overall survival (OS), lower mutation burden, and higher infiltration level of immune cells in the TME. Moreover, the NRS was significantly associated with chemotherapeutic drug sensitivity. Finally, the knockdown of VDAC1 reduced the proliferation and migration of BRCA cells, and promoted cell death induced by necroptosis inducer. This study identified a novel necroptosis-related subtype of BRCA, and a comprehensive analysis of NRGs in BRCA revealed its potential roles in prognosis, clinicopathological features, TME, chemotherapy, tumor proliferation, and tumor necroptosis. These results may improve our understanding of NRGs in BRCA and provide a reference for developing individualized therapeutic strategies.

High FLT3 expression indicates favorable prognosis and correlates with clinicopathological parameters and immune infiltration in breast cancer

Purpose: Breast cancer is a highly heterogeneous malignancy, seriously threatening female health worldwide and inducing higher mortalities. Few have the studies evaluated Fms-like TyrosineKinase-3 (FLT3) in prognostic risk, immunotherapy or any other treatment of breast cancer. Our study focused on investigating the function of FLT3 in breast cancer.

Patients and methods: Based on transcriptome and methylation data mined from The Cancer Gene Atlas (TCGA), we explored the clinical features of FLT3 expression in 1079 breast cancer samples. RT-qPCR in cell lines and tissue samples was used to verify the expression difference of FLT3. Kaplan–Meier survival analysis and cox regression models were employed for screening of FLT3 with potential prognostic capacity. Subsequently, functional analysis of the co-expressed genes was conducted using Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene-set enrichment analysis (GSEA). The correlation between FLT3 expression and tumor immune infiltration was jointly analyzed with estimate, ssGSEA, TIMER, and TISIDB. Then we employed checkpoint-related molecules, immunophenoscore (IPS), and tumor mutation burden (TMB) to assess the efficacy of immuno-checkpoint inhibitors (ICIs). Pearson correlation coefficient was employed to exam the association between DNA methylation and FLT3 expression.

Results: FLT3 displays an elevated expression in breast cancer than normal pairs and is significantly associated with multiple clinical characteristics like age, menopause status, histological type, pathological stage, and molecular subtype as well as increased overall survival (OS). Additionally, FLT3 is a favorable independent prognostic factor. GO, KEGG, and GSEA suggested that FLT3 was associated with diversified immune-related features. FLT3 expression is correlated with the abundance of various immune cells namely CD4+T cell, CD8+ T cell, myeloid dendritic cell, and neutrophil as well as immune inhibitors especially CTLA4, which is positively correlated with FLT3 expression. Moreover, TMB displayed a negative correlation with FLT3 expression while IPS showed adverse tendency. Ultimately, the methylation of FLT3 downregulates the gene expression and closely binds to a few clinical parameters.

Conclusion: FLT3 can be used for prognostic prediction and is relevant to immune infiltration in breast cancer. FLT3 may pave the way for future novel immunotherapies.

Combining Organoid Models with Next-Generation Sequencing to Reveal Tumor Heterogeneity and Predict Therapeutic Response in Breast Cancer

Estrogen receptor-positive (ER+) breast cancer (BC) is a common subtype of BC with a relatively good prognosis. However, recurrence and death from ER+ BC occur because of tumor heterogeneity. This study aimed to explore tumor heterogeneity using next-generation sequencing (NGS) and tumor-organoid models to promote BC precise therapy. We collected needle biopsy, surgical excision, and cerebrospinal fluid (CSF) samples to establish tumor organoids. We found that the histological characteristics of organoids were consistent with original lesions and recapitulated their heterogenicity. In addition, the NGS results showed that PIK3CA and TP53 genes had detrimental mutations. BAP1, RET, AXIN2, and PPP2R2A genes had mutations with unknown function. The score for homologous recombination deficiency (HRD) of genome was 56, indicating that the tumor was likely sensitive to PARPi. The mutant-allele tumor heterogeneity (MATH) value of the tumor genome was 68.03, indicating high tumor heterogeneity. At last, we performed a drug screening on organoids. The toxicity of different drugs toward BC organoids originated from needle biopsy and surgical excision was tested, respectively. The IC₅₀ values in the needle biopsy groups were paclitaxel 2.83 μM, carboplatin 61.47 μM, neratinib 0.8 μM, lapatinib >100 μM; in the surgical excision groups: trastuzumab >100 μM, docetaxel 0.036 μM, tamoxifen 20.54 μM, olaparib 5.478 μM, BYL719 < 0.1 μM. The toxicity data showed that the BC organoids could show dynamic characteristics of tumor progression and reflect the heterogeneity of BC. Our study demonstrates that the combined use of tumor organoids and NGS is a potential way to test tumor heterogeneity and predict drug response in ER + BC, which contributes to the development of personalized therapy.

The Clinical Research and Latest Application of Genomic Assays in Early-Stage Breast Cancer

Breast cancer is a kind of malignant tumor that seriously endangers women's life and health. Once diagnosed, most patients will receive a combination of treatments to achieve a cure. However, breast cancer is a heterogeneous disease. Even with the same clinical stage and pathological features, its response to treatment and postoperative recurrence risk may still be completely different. With the advent of genomic assay, some patients with early-stage breast cancer who originally needed treatment can still achieve long-term disease-free survival without adjuvant chemotherapy, so as to achieve personalized and accurate treatment mode to a certain extent. In this paper, we reviewed the 5 most widely used and studied genomic panel technologies in breast cancer, namely Oncotype DX, MammaPrint, RecurIndex, PAM50, and EndoPredict, according to accessibility and availability. Based on the results of the completed or ongoing clinical studies, we summarized the origin, applicable population, and clinical efficacy of each detection method, and discussed the potential development prospect of detection technology in the future.

Transfer Learning Strategy Based on Unsupervised Learning and Ensemble Learning for Breast Cancer Molecular Subtype Prediction Using Dynamic Contrast-Enhanced MRI

BACKGROUND

Imaging-driven deep learning strategies focus on training from scratch and transfer learning. However, the performance of training from scratch is often impeded by the lack of large-scale labeled training data. Additionally, owing to the differences between source and target domains, analyzing medical image tasks satisfactorily via transfer learning based on ImageNet is difficult.

PURPOSE

To investigate two transfer learning algorithms for breast cancer molecular subtype prediction (luminal and non-luminal) based on unsupervised pre-training and ensemble learning: M_EL and B_EL, using malignant and benign datasets as the source domain, respectively.

STUDY TYPE

Retrospective.

POPULATION

Eight hundred and thirty-three female patients with histologically confirmed breast lesions (567 benign and 266 malignant cases) were selected. In the 5-fold cross-validation, the malignant cohort was randomly divided into 5 subsets to form a training set (80%) and a validation set (20%).

FIELD STRENGTH/SEQUENCE

3.0 T, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) using T1-weighted high-resolution isotropic volume examination.

ASSESSMENT

First, three datasets acquired at different times post-contrast were preprocessed as unlabeled source domains. Second, three baseline networks corresponding to the different MRI post-contrast phases were built, optimized by a combination of mutual information maximization between high- and low-level representations and prior distribution constraints. Next, the pre-trained networks were fine-tuned on the labeled target domain. Finally, prediction results were integrated using weighted voting-based ensemble learning.

STATISTICAL TESTS

Mean accuracy, precision, specificity, and area under receiver operating characteristic curve (AUC) were obtained with 5-fold cross-validation. P < 0.05 was considered to be statistically significant.

RESULTS

Compared with a convolutional long short-term memory network, pre-trained VGG-16, VGG-19, and DenseNet-121 from ImageNet, M_EL and B_EL exhibited significantly more optimized prediction performance (specificity: 90.5% and 89.9%; accuracy: 82.6% and 81.1%; precision: 91.2% and 90.9%; AUC: 0.836 and 0.823, respectively).

DATA CONCLUSION

Transfer learning based on unsupervised pre-training may facilitate automatic prediction of breast cancer molecular subtypes.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.