Machine Learning Helps Identify New Drug Mechanisms in Triple-Negative Breast Cancer

Molecular Classification Models for Triple Negative Breast Cancer Subtype Using Machine Learning

Triple negative breast cancer (TNBC) lacks well-defined molecular targets and is highly heterogenous, making treatment challenging. Using gene expression analysis, TNBC has been classified into four different subtypes: basal-like immune-activated (BLIA), basal-like immune-suppressed (BLIS), mesenchymal (MES), and luminal androgen receptor (LAR). However, there is currently no standardized method for classifying TNBC subtypes. We attempted to define a gene signature for each subtype, and to develop a classification method based on machine learning (ML) for TNBC subtyping. In these experiments, gene expression microarray data for TNBC patients were downloaded from the Gene Expression Omnibus database. Differentially expressed genes unique to 198 known TNBC cases were identified and selected as a training gene set to train in seven different classification models. We produced a training set consisting of 719 DEGs selected from uniquely expressed genes of all four subtypes. The highest average accuracy of classification of the BLIA, BLIS, MES, and LAR subtypes was achieved by the SVM algorithm (accuracy 95–98.8%; AUC 0.99–1.00). For model validation, we used 334 samples of unknown TNBC subtypes, of which 97 (29.04%), 73 (21.86%), 39 (11.68%) and 59 (17.66%) were predicted to be BLIA, BLIS, MES, and LAR, respectively. However, 66 TNBC samples (19.76%) could not be assigned to any subtype. These samples contained only three upregulated genes (EN1, PROM1, and CCL2). Each TNBC subtype had a unique gene expression pattern, which was confirmed by identification of DEGs and pathway analysis. These results indicated that our training gene set was suitable for development of classification models, and that the SVM algorithm could classify TNBC into four unique subtypes. Accurate and consistent classification of the TNBC subtypes is essential for personalized treatment and prognosis of TNBC.

Supervised Machine Learning Predictive Analytics For Triple-Negative Breast Cancer Death Outcomes

OBJECTIVE

To use machine learning algorithms to predict the death outcomes of patients with triple-negative breast cancer, 5 years after discharge.

METHODS

1570 stage I-III breast cancer patients receiving treatment from Sun Yat-sen Memorial Hospital were analyzed. Machine learning was used to predict the death outcomes of patients with triple-negative breast cancer, 5 years after discharge.

RESULTS

The results showed that platelets, LMR (lymphocyte-to-monocyte ratio), age, PLR (the platelet-to-lymphocyte ratio) and white blood cell counts accounted for a significant weight in the 5-year prognosis of triple-negative breast cancer patients. The results of model prediction indicated that rankings for accuracy among the training group (from high to low) were forest, gbm, and DecisionTree (0.770335, 0.760766, 0.751994, 0.737640 and 0.734450, respectively). For AUC value (high to low), they were forest, Logistic and DecisionTree (0.896673, 0.895408, 0.776836, 0.722799 and 0.702804, respectively). The highest MSE value for DecisionTree was 0.2656, and the lowest MSE value for forest was 0.2297. In the test group, accuracy rankings (from high to low) were DecisionTree, and GradientBoosting (0.748408, 0.738854, 0.738854, 0.732484 and gbm, respectively). For AUC value (high to low), the rankings were GradientBoosting, gbm, and DecisionTree (0.731595, 0.715438, 0.712767, 0.708348 and 0.691960, respectively). The maximum MSE value for gbm was 0.2707, and the minimum MSE value for DecisionTree was 0.2516.

CONCLUSION

The machine learning algorithm can predict the death outcomes of patients with triple-negative breast cancer 5 years after discharge. This can be used to estimate individual outcomes for patients with triple-negative breast cancer.

Concepts Driving Pharmacogenomics Implementation Into Everyday Healthcare

Pharmacogenomics (PGx) is often promoted as the domain of precision medicine with the greatest potential to readily impact everyday healthcare. Rapid advances in PGx knowledge derived from extensive basic and clinical research along with decreasing costs of laboratory testing have led to an increased interest in PGx and expectations of imminent clinical translation with substantial clinical impact. However, the implementation of PGx into clinical workflows is neither simple nor straightforward, and comprehensive processes and multidisciplinary collaboration are required. Several national and international institutions have pioneered models for implementing clinical PGx, and these initial models have led to a better understanding of unresolved challenges. In this review, we have categorized and explored the most relevant of these challenges to highlight potential gaps and present possible solutions. We describe the ongoing need for basic and clinical research to drive further developments in evidence-based medicine. Integration into daily clinical workflows introduces new challenges requiring innovative solutions; specifically those related to the electronic health record and embedded clinical decision support. We describe advances in PGx testing and result reporting and describe the critical need for increased standardization in these areas across laboratories. We also explore the complexity of the PGx knowledge required for clinical practice and the need for educational strategies to ensure adequate understanding among members of current and future healthcare teams. Finally, we evaluate knowledge obtained from previous implementation efforts and discuss how to best apply these learnings to future projects. Despite these challenges, the future of precision medicine appears promising due to the rapidity of recent advances in the field and current multidisciplinary efforts to effectively translate PGx to everyday clinical practice.

Focus on Cdc42 in Breast Cancer: New Insights, Target Therapy Development and Non-Coding RNAs

Breast cancer is the most common malignant tumors in females. Although the conventional treatment has demonstrated a certain effect, some limitations still exist. The Rho guanosine triphosphatase (GTPase) Cdc42 (Cell division control protein 42 homolog) is often upregulated by some cell surface receptors and oncogenes in breast cancer. Cdc42 switches from inactive guanosine diphosphate (GDP)-bound to active GTP-bound though guanine-nucleotide-exchange factors (GEFs), results in activation of signaling cascades that regulate various cellular processes such as cytoskeletal changes, proliferation and polarity establishment. Targeting Cdc42 also provides a strategy for precise breast cancer therapy. In addition, Cdc42 is a potential target for several types of non-coding RNAs including microRNAs and lncRNAs. These non-coding RNAs is extensively involved in Cdc42-induced tumor processes, while many of them are aberrantly expressed. Here, we focus on the role of Cdc42 in cell morphogenesis, proliferation, motility, angiogenesis and survival, introduce the Cdc42-targeted non-coding RNAs, as well as present current development of effective Cdc42-targeted inhibitors in breast cancer.

Cdc42: A Novel Regulator of Insulin Secretion and Diabetes-Associated Diseases

Cdc42, a member of the Rho GTPases family, is involved in the regulation of several cellular functions including cell cycle progression, survival, transcription, actin cytoskeleton organization and membrane trafficking. Diabetes is a chronic and metabolic disease, characterized as glycometabolism disorder induced by insulin deficiency related to β cell dysfunction and peripheral insulin resistance (IR). Diabetes could cause many complications including diabetic nephropathy (DN), diabetic retinopathy and diabetic foot. Furthermore, hyperglycemia can promote tumor progression and increase the risk of malignant cancers. In this review, we summarized the regulation of Cdc42 in insulin secretion and diabetes-associated diseases. Organized researches indicate that Cdc42 is a crucial member during the progression of diabetes, and Cdc42 not only participates in the process of insulin synthesis but also regulates the insulin granule mobilization and cell membrane exocytosis via activating a series of downstream factors. Besides, several studies have demonstrated Cdc42 as participating in the pathogenesis of IR and DN and even contributing to promote cancer cell proliferation, survival, invasion, migration, and metastasis under hyperglycemia. Through the current review, we hope to cast light on the mechanism of Cdc42 in diabetes and associated diseases and provide new ideas for clinical diagnosis, treatment, and prevention.