Predictor Of Prognosis In Breast Cancer Research Articles

Integrating multi-omics data through deep learning for accurate cancer prognosis prediction

BackgroundGenomic information is nowadays widely used for precise cancer treatments. Since the individual type of omics data only represents a single view that suffers from data noise and bias, multiple types of omics data are required for accurate cancer prognosis prediction. However, it is challenging to effectively integrate multi-omics data due to the large number of redundant variables but relatively small sample size. With the recent progress in deep learning techniques, Autoencoder was used to integrate multi-omics data for extracting representative features. Nevertheless, the generated model is fragile from data noises. Additionally, previous studies usually focused on individual cancer types without making comprehensive tests on pan-cancer. Here, we employed the denoising Autoencoder to get a robust representation of the multi-omics data, and then used the learned representative features to estimate patients’ risks. ResultsBy applying to 15 cancers from The Cancer Genome Atlas (TCGA), our method was shown to improve the C-index values over previous methods by 6.5% on average. Considering the difficulty to obtain multi-omics data in practice, we further used only mRNA data to fit the estimated risks by training XGboost models, and found the models could achieve an average C-index value of 0.627. As a case study, the breast cancer prognosis prediction model was independently tested on three datasets from the Gene Expression Omnibus (GEO), and shown able to significantly separate high-risk patients from low-risk ones (C-index>0.6, p-values<0.05). Based on the risk subgroups divided by our method, we identified nine prognostic markers highly associated with breast cancer, among which seven genes have been proved by literature review. ConclusionOur comprehensive tests indicated that we have constructed an accurate and robust framework to integrate multi-omics data for cancer prognosis prediction. Moreover, it is an effective way to discover cancer prognosis-related genes.

GPDBN: deep bilinear network integrating both genomic data and pathological images for breast cancer prognosis prediction.

MotivationBreast cancer is a very heterogeneous disease and there is an urgent need to design computational methods that can accurately predict the prognosis of breast cancer for appropriate therapeutic regime. Recently, deep learning-based methods have achieved great success in prognosis prediction, but many of them directly combine features from different modalities that may ignore the complex inter-modality relations. In addition, existing deep learning-based methods do not take intra-modality relations into consideration that are also beneficial to prognosis prediction. Therefore, it is of great importance to develop a deep learning-based method that can take advantage of the complementary information between intra-modality and inter-modality by integrating data from different modalities for more accurate prognosis prediction of breast cancer.ResultsWe present a novel unified framework named genomic and pathological deep bilinear network (GPDBN) for prognosis prediction of breast cancer by effectively integrating both genomic data and pathological images. In GPDBN, an inter-modality bilinear feature encoding module is proposed to model complex inter-modality relations for fully exploiting intrinsic relationship of the features across different modalities. Meanwhile, intra-modality relations that are also beneficial to prognosis prediction, are captured by two intra-modality bilinear feature encoding modules. Moreover, to take advantage of the complementary information between inter-modality and intra-modality relations, GPDBN further combines the inter- and intra-modality bilinear features by using a multi-layer deep neural network for final prognosis prediction. Comprehensive experiment results demonstrate that the proposed GPDBN significantly improves the performance of breast cancer prognosis prediction and compares favorably with existing methods.Availabilityand implementationGPDBN is freely available at https://github.com/isfj/GPDBN.Supplementary informationSupplementary data are available at Bioinformatics online.

Multi-modal advanced deep learning architectures for breast cancer survival prediction

Breast cancer is the most frequently occurring cancer and has compelling contributions to increasing mortality rates among women. The manual prognosis and diagnosis of this disease take long hours, even for a medical professional. A model with better predictive power can benefit cancer patients from going through the toxic side effects and extra medical expenses related to unnecessary treatment. Medical professionals can be benefited from early-stage detection and selection of the appropriate cancer treatment plan. The availability of multi-modal cancer data, i.e., genomic details, histopathology images, and clinical details, supports the researchers in proceeding with the development of multi-modal based advanced deep-learning models. This research proposes gated attentive deep learning models stacked with random forest classifiers, which use multi-modal data and produce informative features to enhance the breast cancer prognosis prediction. It is designed as a bi-phase model; phase one uses a sigmoid gated attention convolutional neural network to generate the stacked features, while phase two passes the stacked features to the random forest classifier. The comparative study of the proposed and other existing methods over METABRIC (1980 patients) and TCGA-BRCA (1080 patients) datasets illustrate significant enhancements, 5.1% in sensitivity values, in the survival estimation of breast cancer patients.

Identification of novel cell glycolysis related gene signature predicting survival in patients with breast cancer

One of the most frequently identified tumors and a contributing cause of death in women is breast cancer (BC). Many biomarkers associated with survival and prognosis were identified in previous studies through database mining. Nevertheless, the predictive capabilities of single-gene biomarkers are not accurate enough. Genetic signatures can be an enhanced prediction method. This research analyzed data from The Cancer Genome Atlas (TCGA) for the detection of a new genetic signature to predict BC prognosis. Profiling of mRNA expression was carried out in samples of patients with TCGA BC (n = 1222). Gene set enrichment research has been undertaken to classify gene sets that vary greatly between BC tissues and normal tissues. Cox models for additive hazards regression were used to classify genes that were strongly linked to overall survival. A subsequent Cox regression multivariate analysis was used to construct a predictive risk parameter model. Kaplan–Meier survival predictions and log-rank validation have been used to verify the value of risk prediction parameters. Seven genes (PGK1, CACNA1H, IL13RA1, SDC1, AK3, NUP43, SDC3) correlated with glycolysis were shown to be strongly linked to overall survival. Depending on the 7-gene-signature, 1222 BC patients were classified into subgroups of high/low-risk. Certain variables have not impaired the prognostic potential of the seven-gene signature. A seven-gene signature correlated with cellular glycolysis was developed to predict the survival of BC patients. The results include insight into cellular glycolysis mechanisms and the detection of patients with poor BC prognosis.

DNA damage response as a prognostic indicator in metastatic breast cancer via mutational analysis

BackgroundHigh tumor heterogeneity contributes to breast cancer recurrence and metastasis. However, the lack of indicators to serve as precise and reliable means of predicting breast cancer prognosis has yet to be addressed. This study aims to reveal the prognostic relevance of mutations in metastatic breast cancer (MBC) by large-scale circulating tumor DNA (ctDNA) analysis in China.MethodsWe performed ctDNA panel-captured sequencing of 958 blood samples from MBC patients including 494 hormone receptor (HR)-positive cases, 130 human epidermal growth factor receptor 2-positive cases, and 177 triple-negative breast cancer (TNBC) cases. The somatic mutations and potential targets were assessed. Progression-free survival (PFS) was analyzed using the Kaplan-Meier method.ResultsIn 801 of the 958 MBC blood samples, 663 mutated genes and 5,829 nonsynonymous alterations were identified. Mutated genes of the highest frequency were tumor protein p53 (TP53, 54%), phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha (PIK3CA, 41%), estrogen receptor 1 (ESR1, 12%), myeloid/lymphoid or mixed-lineage leukemia protein 3 (MLL3, 11%), DNA (cytosine-5)-methyltransferase 3A (DNMT3A, 10%), erb-b2 receptor tyrosine kinase 2 (ERBB2, 10%), GATA binding protein 3 (GATA3, 8%), FAT atypical cadherin 1 (FAT1, 7%), phosphatase and tensin homolog (PTEN, 6%), and mitogen-activated protein kinase kinase kinase 1 (MAP3K1, 6%). Enriched mutations and driver genes in MBC varied across stages and in multiple subtypes. Moreover, TP53, ERBB2, or coexisting TP53/PIK3CA mutations in MBC were remarkably related with shorter PFS. Mutated DNA damage response (DDR) genes were significantly associated with tumor mutation burden and mutant-allele tumor heterogeneity score, as well as with worse clinical outcome.ConclusionsOur findings indicate that the mutations of TP53, PIK3CA, ERBB2, and in particular, DDR genes, in MBC might be relevant indicators of unfavorable prognosis in MBC.

A Novel Framework to Predict Breast Cancer Prognosis Using Immune-Associated LncRNAs.

Background: Breast cancer (BC) is one of the most frequently diagnosed malignancies among females. As a huge heterogeneity of malignant tumor, it is important to seek reliable molecular biomarkers to carry out the stratification for patients with BC. We surveyed immune- associated lncRNAs that may be used as potential therapeutic targets in BC.Methods: LncRNA expression data and clinical information of BC patients were downloaded from the TCGA database for a comprehensive analysis of candidate genes. A model consisting of immune-related lncRNAs enriched in BC cancerous tissues was established using the univariate Cox regression analysis and the iterative Lasso Cox regression analysis. The prognostic performance of this model was validated in two independent cohorts (GSE21653 and BC-KR), and compared with known prognostic biomarkers. A nomogram that integrated the immune-related lncRNA signature and clinicopathological factors was constructed to accurately assess the prognostic value of this signature. The correlation between the signature and immune cell infiltration in BC was also analyzed.Results: The Kaplan-Meier analysis showed that the OS of Patients in the low-risk group had significantly better survival than those in the high-risk group, Clinical subgroup analysis showed that the predictive ability was independent of clinicopathological factors. Univariate/multivariate Cox regression analysis showed immune lncRNA signature is an important prognostic factor and an independent prognostic marker. In addition, GSEA and GSVA analysis as well as comprehensive analysis of immune cells showed that the signature was significantly correlated with the infiltration of immune cells.Conclusion: We successfully constructed an immune-associated lncRNA signature that can accurately predict BC prognosis.

Association of variably methylated tumour DNA regions with overall survival for invasive lobular breast cancer

BackgroundTumour DNA methylation profiling has shown potential to refine disease subtyping and improve the diagnosis and prognosis prediction of breast cancer. However, limited data exist regarding invasive lobular breast cancer (ILBC). Here, we investigated the genome-wide variability of DNA methylation levels across ILBC tumours and assessed the association between methylation levels at the variably methylated regions and overall survival in women with ILBC.MethodsTumour-enriched DNA was prepared by macrodissecting formalin-fixed paraffin embedded (FFPE) tumour tissue from 130 ILBCs diagnosed in the participants of the Melbourne Collaborative Cohort Study (MCCS). Genome-wide tumour DNA methylation was measured using the HumanMethylation 450K (HM450K) BeadChip array. Variably methylated regions (VMRs) were identified using the DMRcate package in R. Cox proportional hazards regression models were used to assess the association between methylation levels at the ten most significant VMRs and overall survival. Gene set enrichment analyses were undertaken using the web-based tool Metaspace. Replication of the VMR and survival analysis findings was examined using data retrieved from The Cancer Genome Atlas (TCGA) for 168 ILBC cases. We also examined the correlation between methylation and gene expression for the ten VMRs of interest using TCGA data.ResultsWe identified 2771 VMRs (P < 10−8) in ILBC tumours. The ten most variably methylated clusters were predominantly located in the promoter region of the genes: ISM1, APC, TMEM101, ASCL2, NKX6, HIST3H2A/HIST3H2BB, HCG4P3, HES5, CELF2 and EFCAB4B. Higher methylation level at several of these VMRs showed an association with reduced overall survival in the MCCS. In TCGA, all associations were in the same direction, however stronger than in the MCCS. The pooled analysis of the MCCS and TCGA data showed that methylation at four of the ten genes was associated with reduced overall survival, independently of age and tumour stage; APC: Hazard Ratio (95% Confidence interval) per one-unit M-value increase: 1.18 (1.02–1.36), TMEM101: 1.23 (1.02–1.48), HCG4P3: 1.37 (1.05–1.79) and CELF2: 1.21 (1.02–1.43). A negative correlation was observed between methylation and gene expression for CELF2 (R = − 0.25, P = 0.001), but not for TMEM101 and APC.ConclusionsOur study identified regions showing greatest variability across the ILBC tumour genome and found methylation at several genes to potentially serve as a biomarker of survival for women with ILBC.

The utility of serum tumor markers CEA and CA 15–3 for breast cancer prognosis and their association with clinicopathological parameters

BackgroundWe aimed to evaluate the association of serum carcinoembryonic antigen (CEA) and cancer antigen 15–3 (CA 15–3) levels with clinicopathological parameters in patients with breast cancer (BC) and their efficiency for the prediction of recurrence.Methods: The records of 482 female patients with breast cancer diagnosis followed in Medical Oncology and Radiation Oncology clinics of Kartal Dr. Lutfu Kirdar Education and Research Hospital were evaluated retrospectively. ResultsThe median age of the patients was 49. CEA levels were significantly higher in postmenopausal patients (p = 0.022). There was no association between CEA and CA 15–3 levels and nodal involvement (p = 0.689, 0.379; respectively). CEA levels were significantly higher in hormone receptor-positive patients (p = 0.007). HER2 negative patients had significantly higher levels of CEA and CA 15–3 (p = 0.017 and 0.011, respectively). The evaluation of metastatic patients showed that CEA and CA 15–3 levels before metastasis were significantly elevated (p = 0.016 ve 0.008, respectively). There was no relation between the metastasis site and CEA, CA 15–3 levels (p = 0.936, 0.201, respectively). Receiver operating characteristic (ROC) analysis was performed to determine the role of CEA and CA 15–3 levels in the prediction of metastasis, and cut-off values were 1.39 ng/ml and 14.54 U/ml, respectively. Sensitivities of CA 15–3 and CEA levels were 82.1% and 88.3%; specificities were 47.3% and 46.2%, respectively. ConclusionsCEA and CA 15–3 are useful as tumor markers for early diagnosis of metastases, and their elevations were associated with unfavorable clinicopathological parameters of breast cancer patients. Since these markers are considered a cheap and accessible way of predicting breast cancer prognosis, there is an increasing interest in the prognostic value of serum levels of tumor markers in recent years. More sensitive cut-off values for each marker are needed to be validated with further studies.

Dynamic changes of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio predicts breast cancer prognosis

BackgroundWe aimed to identify whether neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) are more useful predictors after initial intention to treat than at the time of diagnosis.MethodsWe collected the medical data of 533 patients. The results of the peripheral blood sampling before the primary treatments were labeled as initial cohort, and those obtained between 24 and 36 months after initial treatment were defined as the 2nd cohort. Delayed metastasis has been defined as distant metastasis 2 years after treatment, and survival outcome was estimated and compared across groups.ResultsMedian follow-up duration was 74 months (24–162 months), and 53 patients experienced delayed metastasis. In univariate analysis, metastasis-free survival, patient age at diagnosis, tumor size, axillary lymph node metastasis, HER-2 status, initial NLR and PLR, and 2nd NLR and PLR were found to be significantly associated with delayed metastasis. However, in multivariate analysis, only the 2nd NLR and PLR were found to be significantly associated with delayed metastasis, excluding initial NLR and PLR. Metastasis-free survival was analyzed through the pattern changes of NLR or PLR. The results revealed that patients with continued low NLR and PLR values at pre- and post-treatment (low initial values and 2nd values) showed a significantly better prognosis than those with a change in value or continued high NLR and PLR.ConclusionsWe identified that patients with persistent high NLR and PLR after initial treatment have significant worse prognosis in terms of late metastasis. Therefore, these results suggest that NLR and PLR are more useful in predicting prognosis post-treatment.

In-silico protein-ligand docking studies against the estrogen protein of breast cancer using pharmacophore based virtual screening approaches.

Breast cancer in woman is the most common cancer and in 2018 there were around 2 million new cases recorded. The maximum rate of breast cancer is reported in Belgium followed by Luxembourg. It is the second most general cancer, Lung cancer being the first. If the cancer tumor is located only in the breast, the survival rate would be 99%. If the tumor has wide to lymph nodes around the survival rate would be 85% and if the tumor had extend to distant parts, the survival rate would come down to 27%. Mammary gland is an important organ in mammals which has potential function to secrete, synthesize and deliver milk to the infants for nourishment, improvement and protection. Generally, cancer is named after the body part in which it originated; thus, breast cancer refers to the erratic development and proliferation of cells that originate in the breast tissue (7). There are some kinds of tumors that may grow within various areas of the breast. Most tumors are the outcome of benign (non-cancerous) alters within the breast. The estrogen receptors (ER) in ordinary and diseased states are significant for the improvement of relevant therapeutic strategies. Two main forms of ER exist, ERα and ERβ, which are encoded by separate genes. Estrogens play a central role in breast cancer improvement with ERα status being the mainly significant predictor of breast cancer prognosis. The potent lead molecule binding mode, residue-interaction patterns and docking energy were examined by molecular docking and binding free energy studies. The lead compounds and 3ERT complex structural stability and dynamic behavior were monitored by molecular dynamics analysis. The drug-likeness properties of lead compounds were predicted ADME analysis.

Clinical value of CTLA4 combined with clinicopathological factors in evaluating the prognosis of breast cancer.

Clinical prediction of breast cancer prognosis relies on both clinical-pathological features and biological markers. Many studies have revealed that tumor cytotoxic T lymphocyte antigen 4 (CTLA4) expression may present prognostic predicting value in cancers. We intended to explore the prognostic value of significant clinicopathological parameters and CTLA4 for predicting survival of patients with breast cancer. A total of 229 breast cancer patients who had radical surgery treatment between Sep 2009 and April 2011 were enrolled in this study. Immunohistochemical staining was performed to evaluate CTLA4 grade and Ki-67 index in breast cancer tissue. Univariate and multivariate logistic analysis, Kaplan-Meier survival analysis and ROC curve were used to explore the association between CTLA4 or clinicopathological parameters and disease-free survival (DFS). A nomogram was constructed based on the regression model to predict DFS of patients with breast cancer. CTLA4 grade (OR 1.730, 95% CI: 1.213-2.468, P=0.002), Ki-67 (OR 1.449, 95% CI: 1.069-1.964, P=0.017) and N stage (lymph node metastasis) (OR 2.268, 95% CI: 1.588-3.303, P=0.000) showed significantly association with DFS of breast cancer patients. All these factors were independent predictors for poor survival, as patients with stage N2-3 tumors, high CTLA4 grade and Ki-67 index showed low survival probability (P<0.01). The conjunction of these factors exhibited good discrimination value (AUC 0.815, 95% CI: 0.749-0.882, P=0.000). Nomogram performed based on CTLA4 grade, Ki-67 index and N stage provided an efficient method to predict DFS of patients with breast cancer. The high expression of CTLA4 and Ki-67 together with lymph node metastasis in breast cancer are independent risk factors that affect the prognosis of breast cancer patients. They have the potentiality to be utilized conjunctively as predictor in clinical practice.

Breast Lesion Elastography Region of Interest Selection and Quantitative Heterogeneity: A Systematic Review and Meta-Analysis

Abstract Background: Breast cancer is one of the most common causes of death among women worldwide. Early detection and diagnosis will be helpful to reduce mortality and improve prognosis. It is urgent to develop efficient detection technology for breast cancer. Mammographic screening is a valuable tool for early detection of breast cancer. However, the increased density of breast tissue significantly reduces the diagnostic accuracy. Aim of Study: To provide an overview of the different reported elasticities of specific breast pathologies based on ultrasound elastography. Material and Methods: A total of 35 articles including 8316 patients and 9057 breast lesions were included in the pooled analysis of which 3060 malignant lesions were included from 40 studies. The median incidence of malignant breast lesion is 37.1% calculated from the incidence of malignant lesions of all included studies. Seven of the included studies assessed VTIQ. Mean age varied along all included studies. Results: The sensitivity and specificity of Emax, Emean and Eratio for the diagnosis of breast cancer varied according to the interpretative criteria used to define a test as positive. The summary estimates of sensitivity and specificity were 82.58% (95% CI 78.32% to 86.16%) and 84.12% (95% CI 79.07% to 87.07%) for Emean, 86.19% (95% CI 81.60% to 89.77%) and 88.56% (95% CI 88.56% to 91.54%) for Emax, and 87.50% (95% CI 77.47% to 93.44%) and 79.30% (95% CI 68.21% to 87.24%) for Eratio respectively. Regarding DOR, Emax achieved the highest value 48.32 (95% CI 28.7 to 67.8) which means there are 48 times the odds of obtaining an Emax positive result in a diseased rather than a non-diseased person. Meta-regression analysis was conducted to assess the impact of two covaries; Emean and Emax using Likelihood ratio test and revealed significant difference existed with higher summary sensitivity (x2=35.04, p < 001) and specificity (x2=18.65, p < 001) in Emax than Emean. SROC curves were used to show the distribution of sensitivity and specificity of Emax, Emean and Eratio in the Roc space as well as the prediction region. Conclusion: Our meta-analysis demonstrates that SWE is an accurate and reliable diagnostic tool in discriminating malignant and benign breast lesions. With wide application, SWE may significantly improve the early diagnostic of breast cancer. SWE can provide additional information on predicting breast cancer prognosis. However, the possibility of false-positive and false-negative results should be considered during interpretation.

Multi-Modal Classification for Human Breast Cancer Prognosis Prediction: Proposal of Deep-Learning Based Stacked Ensemble Model.

Breast Cancer is a highly aggressive type of cancer generally formed in the cells of the breast. Despite significant advances in the treatment of primary breast cancer in the last decade, there is a dire need to attempt of an accurate predictive model for breast cancer prognosis prediction. Researchers from various disciplines are working together to develop methods to save people from this fatal disease. A good predictive model can help in correct prognosis prediction of breast cancer. This accurate prediction can have several benefits like detection of cancer in the early stage, spare patients from getting unnecessary treatment and medical expenses related to it. Previous works rely mostly on uni-modal data (selected gene expression)for predictive model design. In recent years, however, multi-modal cancer data sets have become available (gene expression, copy number alteration and clinical). Motivated by the enhancement of deep-learning based models, in the current study, we propose to use some deep-learning based predictive models in a stacked ensemble framework to improve the prognosis prediction of breast cancer from available multi-modal data sets. One of the unique advantages of the proposed approach lies in the architecture of the model. It is a two-stage model. Stage one uses a convolutional neural network for feature extraction, while stage two uses the extracted features as input to the stack-based ensemble model. The predictive performance evaluated using different performance measures shows that this model produces better result than already existing approaches. This model results in AUC value of 0.93 and accuracy of 90.2 percent at medium stringency level (Specificity = 95 percent and threshold = 0.45). Keras 2.2.1, along with Tensorflow 1.12, is used for implementing the source code of the model. The source code can be downloaded from Github: https://github.com/nikhilaryan92/BreastCancer.

Generative Adversarial Networks for Robust Breast Cancer Prognosis Prediction with Limited Data Size.

Accurate cancer patient prognosis stratification is essential for oncologists to recommend proper treatment plans. Deep learning models are capable of providing good prediction power for such stratification. The main challenge is that only a limited number of labeled patients are available for cancer prognosis. To overcome this, we proposed Wasserstein Generative Adversarial Network-based Deep Adversarial Data Augmentation (wDADA) that leverages generative adversarial networks to perform data augmentation and assist in model training. We used the proposed framework to train our model for predicting disease-specific survival (DSS) of breast cancer patients from the METABRIC dataset. We found that wDADA achieved 0.6726± 0.0278, 0.7538±0.0328, and 0.6507 ±0.0248 in terms of accuracy, AUC, and concordance index in predicting 5-year DSS, respectively, which is comparable to our previously proposed Bimodal model (accuracy: 0.6889±0.0159; AUC: 0.7546± 0.0183; concordance index: 0.6542±0.0120), which needs careful calibration and extensive search on pre-trained network architectures. The flexibility of the proposed wDADA allows us to incorporate it with ensemble learning and semi-supervised learning to further improve performance. Our results indicate that it is possible to utilize generative adversarial networks to train deep models in medical applications, wherein only limited data are available.

Prognostic and predictive value of monocarboxylate transporter 4 in patients with breast cancer.

The Warburg effect explains the large amount of lactic acid that tumour cells produce to establish and maintain the acidic characteristics of the tumour microenvironment, which contributes to the migration, invasion and angiogenesis of tumour cells. Monocarboxylate transporter 4 (MCT-4) is a key marker of tumour glycolysis and lactic acid production; however, the role of MCT-4 in breast cancer remains unclear. In the present study, immunohistochemistry (IHC) was used to detect the expression levels of MCT-4 in tissue microarrays of 145 patients diagnosed with invasive ductal breast cancer. The IHC score was used to assess the intensity of staining and the proportion of positive cells. Western blotting and reverse transcription-quantitative PCR were also performed to detect the expression levels of MCT-4 in 30 pairs of breast cancer tissues and adjacent normal tissues. In vitro experiments (EdU incoporation and Cell Counting Kit-8) were performed to examine the role of MCT-4 in the breast cancer MCF-7 cell line. The results of the present study indicated that high MCT-4 expression was associated with pT status (P=0.018), oestrogen receptor (ER) status (P=0.001), progesterone receptor (PR) status (P=0.024), Ki67 index (P=0.043) and androgen receptor (AR) status (P=0.033). In addition, an association between MCT-4 expression and pathological grade was observed (P=0.030). Furthermore, univariate (P=0.027) and multivariate (P=0.001) survival analysis revealed that MCT-4 expression and lymph node involvement were significant independent predictors of breast cancer prognosis. In addition, silencing MCT-4 expression attenuated breast cancer cell viability. Therefore, MCT-4 may be used as a potential predictor of invasive breast cancer.

Elevated levels of tumour apolipoprotein D independently predict poor outcome in breast cancer patients.

Apolipoprotein D (ApoD) is a protein that is regulated by androgen and oestrogen, and is a major constituent of breast cysts. Although ApoD has been reported to be a marker of breast cancer, its prognostic importance in invasive breast cancer is unclear. The aim of this study was to investigate the relationship between ApoD protein expression, oestrogen receptor-α (ERα) expression and androgen receptor (AR) expression in predicting breast cancer outcome. ApoD levels were measured by the use of immunohistochemistry and video image analysis on tissue sections from a breast cancer cohort (n=214). We assessed the associations of ApoD expression with disease-free survival (DFS), metastasis-free survival (MFS), and overall survival (OS). We also assessed the relationship between ApoD expression, AR expression and ERα expression in predicting OS. ApoD expression (>1% ApoD positivity) was found in 72% (154/214) of tissues. High ApoD positivity (≥20.7%, fourth quartile) was an independent predictor of MFS and OS, and conferred a 2.2-fold increased risk of developing metastatic disease and a 2.1-fold increased risk of breast cancer-related death. ApoD positivity was not associated with AR or ERα nuclear positivity. However, patients with (≥1%) ERα-positive cancers with low (<20.7%) ApoD positivity, or those showing high (≥78%) AR positivity and low (<20.7%) ApoD positivity had better OS than other patient groups. ApoD expression could be used to predict breast cancer prognosis independently of ERα and AR expression.

CD31 and D2-40 Contribute to Peritoneal Metastasis of Colorectal Cancer by Promoting Epithelial-Mesenchymal Transition.

Background/AimsColorectal cancer (CRC) patients often exhibit peritoneal metastasis, which negatively impacts their prognosis. CD31 and D2-40 have recently been suggested to be predictors of breast cancer prognosis, but their role in colorectal peritoneal metastasis (CRPM) remains unknown.MethodsThe expression profiles of CD31 and D2-40 were analyzed in CRC patients with or without CRPM and in CRC cell lines with increasing metastatic potential. Overexpression and short hairpin RNA knockdown assays were performed in CRC cells, and the effects of these alterations on epithelial-mesenchymal transition (EMT) in vitro, growth of xenograft tumors in vivo, and peritoneal metastasis potential in a mouse model of CRPM were examined.ResultsThe expressions of CD31 and D2-40 were upregulated in CRC tumor tissues and was elevated further in tumor tissues from patients with CRPM. CD31 and D2-40 expression levels exhibited increasing trends parallel to the EMT potential of CRC cells. CD31 and D2-40 are essential for CRC cell EMT in vitro as well as for xenograft tumor growth and peritoneal metastasis in vivo.ConclusionsCD31 and D2-40 contribute to CRPM by promoting EMT and may serve as prognostic markers and therapeutic targets for CRC, particularly in patients with peritoneal metastasis.

Highly robust model of transcription regulator activity predicts breast cancer overall survival

BackgroundWhile several multigene signatures are available for predicting breast cancer prognosis, particularly in early stage disease, effective molecular indicators are needed, especially for triple-negative carcinomas, to improve treatments and predict diagnostic outcomes. The objective of this study was to identify transcriptional regulatory networks to better understand mechanisms giving rise to breast cancer development and to incorporate this information into a model for predicting clinical outcomes.MethodsGene expression profiles from 1097 breast cancer patients were retrieved from The Cancer Genome Atlas (TCGA). Breast cancer-specific transcription regulatory information was identified by considering the binding site information from ENCODE and the top co-expressed targets in TCGA using a nonlinear approach. We then used this information to predict breast cancer patient survival outcome.ResultWe built a multiple regulator-based prediction model for breast cancer. This model was validated in more than 5000 breast cancer patients from the Gene Expression Omnibus (GEO) databases. We demonstrated our regulator model was significantly associated with clinical stage and that cell cycle and DNA replication related pathways were significantly enriched in high regulator risk patients.ConclusionOur findings demonstrate that transcriptional regulator activities can predict patient survival. This finding provides additional biological insights into the mechanisms of breast cancer progression.

Machine Learning techniques in breast cancer prognosis prediction: A primary evaluation.

More than 750 000 women in Italy are surviving a diagnosis of breast cancer. A large body of literature tells us which characteristics impact the most on their prognosis. However, the prediction of each disease course and then the establishment of a therapeutic plan and follow‐up tailored to the patient is still very complicated. In order to address this issue, a multidisciplinary approach has become widely accepted, while the Multigene Signature Panels and the Nottingham Prognostic Index are still discussed options. The current technological resources permit to gather many data for each patient. Machine Learning (ML) allows us to draw on these data, to discover their mutual relations and to esteem the prognosis for the new instances. This study provides a primary evaluation of the application of ML to predict breast cancer prognosis. We analyzed 1021 patients who underwent surgery for breast cancer in our Institute and we included 610 of them. Three outcomes were chosen: cancer recurrence (both loco‐regional and systemic) and death from the disease within 32 months. We developed two types of ML models for every outcome (Artificial Neural Network and Support Vector Machine). Each ML algorithm was tested in accuracy (=95.29%‐96.86%), sensitivity (=0.35‐0.64), specificity (=0.97‐0.99), and AUC (=0.804‐0.916). These models might become an additional resource to evaluate the prognosis of breast cancer patients in our daily clinical practice. Before that, we should increase their sensitivity, according to literature, by considering a wider population sample with a longer period of follow‐up. However, specificity, accuracy, minimal additional costs, and reproducibility are already encouraging.

Diagnostic and prognostic value of the cancer-testis antigen lactate dehydrogenase C4 in breast cancer

BackgroundLactate dehydrogenase C4 (LDH-C4) as a cancer/testis antigen (CTA) is abnormally expressed in some malignant tumors. However, the expression and clinical significance of LDH-C4 in breast cancer (BC) has not been characterized. MethodsWe determined LDHC mRNA expression in serum and serum-derived exosomes of BC patients by quantitative RT-PCR. We also evaluated the protein expression of LDH-C4 in BC tissues using high-throughput tissue microarray analysis and immunohistochemistry. ResultsOur results showed high mRNA expression level of LDHC in serum and serum-derived exosomes of BC patients. The LDHC level in serum and exosomes could distinguish BC cases from healthy individuals based on their AUCs of 0.9587 and 0.9464, respectively. Besides, the LDHC level in exosomes of BC patients associated with tumor size, and positively correlated with HER2 and Ki-67 expressions (all with P < 0.05). Serum and exosomal level of LDHC negatively correlated with medical treatment and positively with the recurrence of BC. Survival analysis showed that LDH-C4 expression negatively correlated with BC prognosis. ConclusionSerum and exosomal LDHC may be an effective indicator for the diagnosis, efficacy evaluation, and monitoring the recurrence of BC. LDH-C4 may act as a biomarker that predicts BC prognosis.