Using Breast Cancer Gene Expression Signatures in Clinical Practice: Unsolved Issues, Ongoing Trials and Future Perspectives

Enhancing cancer stage prediction through hybrid deep neural networks: a comparative study

Efficiently detecting and treating cancer at an early stage is crucial to improve the overall treatment process and mitigate the risk of disease progression. In the realm of research, the utilization of artificial intelligence technologies holds significant promise for enhancing advanced cancer diagnosis. Nonetheless, a notable hurdle arises when striving for precise cancer-stage diagnoses through the analysis of gene sets. Issues such as limited sample volumes, data dispersion, overfitting, and the use of linear classifiers with simple parameters hinder prediction performance. This study introduces an innovative approach for predicting early and late-stage cancers by integrating hybrid deep neural networks. A deep neural network classifier, developed using the open-source TensorFlow library and Keras network, incorporates a novel method that combines genetic algorithms, Extreme Learning Machines (ELM), and Deep Belief Networks (DBN). Specifically, two evolutionary techniques, DBN-ELM-BP and DBN-ELM-ELM, are proposed and evaluated using data from The Cancer Genome Atlas (TCGA), encompassing mRNA expression, miRNA levels, DNA methylation, and clinical information. The models demonstrate outstanding prediction accuracy (89.35%-98.75%) in distinguishing between early- and late-stage cancers. Comparative analysis against existing methods in the literature using the same cancer dataset reveals the superiority of the proposed hybrid method, highlighting its enhanced accuracy in cancer stage prediction.

Analytical validation and algorithm improvement of HepatoPredict kit to assess hepatocellular carcinoma prognosis before a liver transplantation

Objectives

To verify the analytical performance of the HepatoPredict kit, a novel tool developed to stratify Hepatocellular Carcinoma (HCC) patients according to their risk of relapse after a Liver Transplantation (LT).

Methods

The HepatoPredict tool combines clinical variables and a gene expression signature in an ensemble of machine-learning algorithms to forecast the benefit of a LT in HCC patients. To ensure the accuracy and reliability of this method, extensive analytical validation was conducted to verify its specificity and robustness. The experiments were designed following the guidelines for multi-target genomic assays such as ISO201395-2019, MIQE, CLSI-MM16, CLSI-MM17, and CLSI-EP17-A. The validation process included reproducibility between operators and between RNA extractions and RT-qPCR runs, and interference of input RNA levels or varying reagent levels. A recently retrained version of the HepatoPredict algorithms was also tested.

Results

The validation process demonstrated that the HepatoPredict kit met the required standards for robustness (p > 0.05), analytical specificity (inclusivity of 95 %), and sensitivity (LoB, LoD, linear range, and amplification efficiency between 90 and 110 %). The operator, equipment, input RNA, and reagents used had no significant effect on the HepatoPredict results. Additionally, the testing of a recently retrained version of the HepatoPredict algorithm, showed that this new version further improved the accuracy of the kit and performed better than existing clinical criteria in accurately identifying HCC patients who are more likely to benefit LT.

Conclusions

Even with the introduced variations in molecular and clinical variables, the HepatoPredict kit's prognostic information remains consistent. It can accurately identify HCC patients who are more likely to benefit from a LT. Its robust performance also confirms that it can be easily integrated into standard diagnostic laboratories.

Biomarkers in Breast Cancer: An Old Story with a New End

Breast cancer is the second most frequent cancer in the world. It is a heterogeneous disease and the leading cause of cancer mortality in women. Advances in molecular technologies allowed for the identification of new and more specifics biomarkers for breast cancer diagnosis, prognosis, and risk prediction, enabling personalized treatments, improving therapy, and preventing overtreatment, undertreatment, and incorrect treatment. Several breast cancer biomarkers have been identified and, along with traditional biomarkers, they can assist physicians throughout treatment plan and increase therapy success. Despite the need of more data to improve specificity and determine the real clinical utility of some biomarkers, others are already established and can be used as a guide to make treatment decisions. In this review, we summarize the available traditional, novel, and potential biomarkers while also including gene expression profiles, breast cancer single-cell and polyploid giant cancer cells. We hope to help physicians understand tumor specific characteristics and support decision-making in patient-personalized clinical management, consequently improving treatment outcome.

Identification of Novel Diagnostic and Prognostic Gene Signature Biomarkers for Breast Cancer Using Artificial Intelligence and Machine Learning Assisted Transcriptomics Analysis

BACKGROUND

Breast cancer (BC) is one of the most common female cancers. Clinical and histopathological information is collectively used for diagnosis, but is often not precise. We applied machine learning (ML) methods to identify the valuable gene signature model based on differentially expressed genes (DEGs) for BC diagnosis and prognosis.

METHODS

A cohort of 701 samples from 11 GEO BC microarray datasets was used for the identification of significant DEGs. Seven ML methods, including RFECV-LR, RFECV-SVM, LR-L1, SVC-L1, RF, and Extra-Trees were applied for gene reduction and the construction of a diagnostic model for cancer classification. Kaplan-Meier survival analysis was performed for prognostic signature construction. The potential biomarkers were confirmed via qRT-PCR and validated by another set of ML methods including GBDT, XGBoost, AdaBoost, KNN, and MLP.

RESULTS

We identified 355 DEGs and predicted BC-associated pathways, including kinetochore metaphase signaling, PTEN, senescence, and phagosome-formation pathways. A hub of 28 DEGs and a novel diagnostic nine-gene signature (COL10A, S100P, ADAMTS5, WISP1, COMP, CXCL10, LYVE1, COL11A1, and INHBA) were identified using stringent filter conditions. Similarly, a novel prognostic model consisting of eight-gene signatures (CCNE2, NUSAP1, TPX2, S100P, ITM2A, LIFR, TNXA, and ZBTB16) was also identified using disease-free survival and overall survival analysis. Gene signatures were validated by another set of ML methods. Finally, qRT-PCR results confirmed the expression of the identified gene signatures in BC.

CONCLUSION

The ML approach helped construct novel diagnostic and prognostic models based on the expression profiling of BC. The identified nine-gene signature and eight-gene signatures showed excellent potential in BC diagnosis and prognosis, respectively.

Applying a GAN-based classifier to improve transcriptome-based prognostication in breast cancer

Established prognostic tests based on limited numbers of transcripts can identify high-risk breast cancer patients yet are approved only for individuals presenting with specific clinical features or disease characteristics. Deep learning algorithms could hold potential for stratifying patient cohorts based on full transcriptome data, yet the development of robust classifiers is hampered by the number of variables in omics datasets typically far exceeding the number of patients. To overcome this hurdle, we propose a classifier based on a data augmentation pipeline consisting of a Wasserstein generative adversarial network (GAN) with gradient penalty and an embedded auxiliary classifier to obtain a trained GAN discriminator (T-GAN-D). Applied to 1244 patients of the METABRIC breast cancer cohort, this classifier outperformed established breast cancer biomarkers in separating low- from high-risk patients (disease specific death, progression or relapse within 10 years from initial diagnosis). Importantly, the T-GAN-D also performed across independent, merged transcriptome datasets (METABRIC and TCGA-BRCA cohorts), and merging data improved overall patient stratification. In conclusion, the reiterative GAN-based training process allowed generating a robust classifier capable of stratifying low- vs high-risk patients based on full transcriptome data and across independent and heterogeneous breast cancer cohorts.

PROCURE European consensus on breast cancer multigene signatures in early breast cancer management

Breast cancer multigene signatures (BCMS) have changed how patients with early-stage breast cancer (eBC) are managed, as they provide prognostic information and can be used to select patients who may avoid adjuvant chemotherapy. Clinical guidelines make recommendations on the use of BCMS; however, little is known on the current use of BCMS in clinical practice. We conduct a two-round Delphi survey to enquire about current use and perceived utility for specific patient profiles, and unmet needs of BCMS. Overall, 133 panellists experienced in breast cancer across 11 European countries have participated, most using BCMS either routinely (66.2%) or in selected cases (27.1%). Our results show that BCMS are mainly used to assess the risk of recurrence and to select patients for adjuvant chemotherapy; notably, no consensus has been reached on the lack of utility of BCMS for selecting the type of chemotherapy to administer. Also, there are discrepancies between the recommended and current use of BCMS in clinical practice, with use in certain patient profiles for which there is no supporting evidence. Our study suggests that physician education initiatives are needed to ensure the correct use and interpretation of BCMS to, ultimately, improve management of patients with eBC.

Neoadjuvant endocrine therapy in postmenopausal women with HR+/HER2- breast cancer

INTRODUCTION

While endocrine therapy is the standard-of-care adjuvant treatment for hormone receptor-positive (HR+) breast cancers, there is also extensive evidence for the role of pre-operative (or neoadjuvant) endocrine therapy (NET) in HR+ postmenopausal women.

AREAS COVERED

We conducted a thorough review of the published literature, to summarize the evidence to date, including studies of how NET compares to neoadjuvant chemotherapy, which NET agents are preferable, and the optimal duration of NET. We describe the importance of on-treatment assessment of response, the different predictors available (including Ki67, PEPI score, and molecular signatures) and the research opportunities the pre-operative setting offers. We also summarize recent combination trials and discuss how the COVID-19 pandemic led to increases in NET use for safe management of cases with deferred surgery and adjuvant treatments.

EXPERT OPINION

NET represents a safe and effective tool for the management of postmenopausal women with HR+/HER2- breast cancer, enabling disease downstaging and a wider range of surgical options. Aromatase inhibitors are the preferred NET, with evidence suggesting that longer regimens might yield optimal results. However, NET remains currently underutilised in many territories and institutions. Further validation of predictors for treatment response and benefit is needed to help standardise and fully exploit the potential of NET in the clinic.

Optimal Choice of Neoadjuvant Chemotherapy for HER2-Negative Breast Cancer: Clinical Insights

The neoadjuvant setting provides immense opportunities for translational research and drug development. The acceptance of pathological complete response (pCR) as a surrogate endpoint for clinical benefit has led to the widespread use of neoadjuvant treatment. Optimal neoadjuvant therapies are determined based on their ability to achieve the highest rates of pCR. Predicted rates of pCR for triple negative breast cancer (TNBC) treated with sequential taxane/anthracycline regimens range from 35% to 48%. With the addition of a platinum agent pCR rates of 55% are predicted. Further increases have been observed with the addition of immune checkpoint inhibitors to this standard chemotherapy backbone. In the pivotal KEYNOTE-522 clinical trial pCR rates of 65% and 69% were reported for chemotherapy plus pembrolizumab in the overall and PD-L1-positive subgroup respectively. The role of the neoadjuvant chemotherapy is less clear in hormone receptor (HR)-positive, human epidermal growth factor receptor-2 (HER2)-negative breast cancer. In general, HR–positive cancers have the least chance of achieving a pCR after neoadjuvant chemotherapy, especially if they are low-grade. If neoadjuvant chemotherapy is given for high-risk HR-positive, HER2-negative breast cancer, standard adjuvant anthracycline/taxane regimens are appropriate. Optimum endocrine therapy is the standard-of-care in the adjuvant setting regardless of pCR. There are several genomic signatures available to guide decisions regarding adjuvant chemotherapy use however these assays are not routinely used in the neoadjuvant setting. For high-risk patients meeting the criteria for the monarchE trial adjuvant abemaciclib in addition to endocrine therapy is associated with an improvement in disease free survival (DFS) at 3 years. Based on the OlympiA trial patients with germline BRCA mutations should be considered for adjuvant olaparib therapy. In this article we review neoadjuvant clinical trials that guide optimum treatment options for TNBC and HR-positive, HER2-negative breast cancer.

The characteristics and clinical significance of atypical mitosis in breast cancer

Atypical mitosis is considered a feature of malignancy, however, its significance in breast cancer (BC) remains elusive. Here, we aimed to assess the clinical value of atypical mitoses in BC and to explore their underlying molecular features. Atypical and typical mitotic figures were quantified and correlated with clinicopathological variables in a large cohort of primary BC tissue sections (n = 846) using digitalized hematoxylin and eosin whole-slide images (WSIs). In addition, atypical mitoses were assessed in The Cancer Genome Atlas (TCGA) BC dataset (n = 1032) and were linked to the genetic alterations and pathways. In this study, the median of typical mitoses was 17 per 3 mm² (range 0-120 mitoses), while the median of atypical mitoses was 4 (range 0-103 mitoses). High atypical mitoses were significantly associated with parameters characteristic of aggressive tumor behavior. The total number of mitoses, and a high atypical-to-typical mitoses ratio (>0.27) were associated with poor BC specific survival (BCSS), (p = 0.04 and 0.01, respectively). The atypical-to-typical mitoses ratio dichotomized triple negative-BC (TNBC) patients into two distinct groups in terms of the association with the outcome, while the overall number of mitoses was not. Moreover, TNBC patients with high atypical-to-typical mitoses ratio treated with adjuvant chemotherapy were associated with shorter survival (p = 0.003). Transcriptomic analysis of the TCGA-BRCA cohort dichotomized based on atypical mitoses identified 2494 differentially expressed genes. These included genes linked to pathways involved in chromosomal localization and segregation, centrosome assembly, spindle and microtubule formation, regulation of cell cycle and DNA repair. To conclude, the atypical-to-typical mitoses ratio has prognostic value independent of the overall mitotic count in BC patients and could predict the response to chemotherapy in TNBC.