Platform-independent approach for cancer detection from gene expression profiles of peripheral blood cells

In search of the ratio of miRNA expression as robust biomarkers for constructing stable diagnostic models among multi-center data

MicroRNAs (miRNAs) are promising biomarkers for the early detection of disease, and many miRNA-based diagnostic models have been constructed to distinguish patients and healthy individuals. To thoroughly utilize the miRNA-profiling data across different sequencing platforms or multiple centers, the models accounting the batch effects were demanded for the generalization of medical application. We conducted transcription factor (TF)-mediated miRNA-miRNA interaction network analysis and adopted the within-sample expression ratios of miRNA pairs as predictive markers. The ratio of the expression values between each miRNA pair turned out to be stable across multiple data sources. A genetic algorithm-based classifier was constructed to quantify risk scores of the probability of disease and discriminate disease states from normal states in discovery, with a validation dataset for COVID-19, renal cell carcinoma, and lung adenocarcinoma. The predictive models based on the expression ratio of interacting miRNA pairs demonstrated good performances in the discovery and validation datasets, and the classifier may be used accurately for the early detection of disease.

Integrated analysis of diverse cancer types reveals a breast cancer-specific serum miRNA biomarker through relative expression orderings analysis

PURPOSE

Serum microRNA (miRNA) holds great potential as a non-invasive biomarker for diagnosing breast cancer (BrC). However, most diagnostic models rely on the absolute expression levels of miRNAs, which are susceptible to batch effects and challenging for clinical transformation. Furthermore, current studies on liquid biopsy diagnostic biomarkers for BrC mainly focus on distinguishing BrC patients from healthy controls, needing more specificity assessment.

METHODS

We collected a large number of miRNA expression data involving 8465 samples from GEO, including 13 different cancer types and non-cancer controls. Based on the relative expression orderings (REOs) of miRNAs within each sample, we applied the greedy, LASSO multiple linear regression, and random forest algorithms to identify a qualitative biomarker specific to BrC by comparing BrC samples to samples of other cancers as controls.

RESULTS

We developed a BrC-specific biomarker called 7-miRPairs, consisting of seven miRNA pairs. It demonstrated comparable classification performance in our analyzed machine learning algorithms while requiring fewer miRNA pairs, accurately distinguishing BrC from 12 other cancer types. The diagnostic performance of 7-miRPairs was favorable in the training set (accuracy = 98.47%, specificity = 98.14%, sensitivity = 99.25%), and similar results were obtained in the test set (accuracy = 97.22%, specificity = 96.87%, sensitivity = 98.02%). KEGG pathway enrichment analysis of the 11 miRNAs within the 7-miRPairs revealed significant enrichment of target mRNAs in pathways associated with BrC.

CONCLUSION

Our study provides evidence that utilizing serum miRNA pairs can offer significant advantages for BrC-specific diagnosis in clinical practice by directly comparing serum samples with BrC to other cancer types.

Blood-Based mRNA Tests as Emerging Diagnostic Tools for Personalised Medicine in Breast Cancer

Molecular diagnostic tests help clinicians understand the underlying biological mechanisms of their patients' breast cancer (BC) and facilitate clinical management. Several tissue-based mRNA tests are used routinely in clinical practice, particularly for assessing the BC recurrence risk, which can guide treatment decisions. However, blood-based mRNA assays have only recently started to emerge. This review explores the commercially available blood mRNA diagnostic assays for BC. These tests enable differentiation of BC from non-BC subjects (Syantra DX, BCtect), detection of small tumours <10 mm (early BC detection) (Syantra DX), detection of different cancers (including BC) from a single blood sample (multi-cancer blood test Aristotle), detection of BC in premenopausal and postmenopausal women and those with high breast density (Syantra DX), and improvement of diagnostic outcomes of DNA testing (variant interpretation) (+RNAinsight). The review also evaluates ongoing transcriptomic research on exciting possibilities for future assays, including blood transcriptome analyses aimed at differentiating lymph node positive and negative BC, distinguishing BC and benign breast disease, detecting ductal carcinoma in situ, and improving early detection further (expression changes can be detected in blood up to eight years before diagnosing BC using conventional approaches, while future metastatic and non-metastatic BC can be distinguished two years before BC diagnosis).

Proteome expression profiling of red blood cells during the tumorigenesis of hepatocellular carcinoma

The early diagnosis of hepatocellular carcinoma (HCC) has not been clinically elucidated, leading to an increased mortality rate in patients with HCC. HCC is a systemic disease related to disorders of blood homeostasis, and the association between red blood cells (RBCs) and HCC tumorigenesis remains elusive. We performed data-independent acquisition proteomic analyses of 72 clinical RBC samples, including HCC (n = 30), liver cirrhosis (LC, n = 17), and healthy controls (n = 25), and characterized the clinical relevance of RBCs and tumorigenesis in HCC. We observed dynamic changes in RBCs during HCC tumorigenesis, and our findings indicate that, based on the protein expression profiles of RBCs, LC is a developmental stage closely approaching HCC. The expression of hemoglobin (HbA and HbF) in peripheral blood dynamically changed during HCC tumorigenesis, suggesting that immature erythroid cells exist in peripheral blood of HCC patients and that erythropoiesis is influenced by the onset of LC. We also identified the disrupted autophagy pathway in RBCs at the onset of LC, which persisted during HCC tumorigenesis. The oxytocin and GnRH pathways were disrupted and first identified during the development of LC into HCC. Significantly differentially expressed SMIM1, ANXA7, HBA1, and HBE1 during tumorigenesis were verified as promising biomarkers for the early diagnosis of HCC using parallel reaction monitoring technology. This study may enhance the understanding of HCC tumorigenesis from a different point of view and aid the early diagnosis of HCC.

Ensemble classification based signature discovery for cancer diagnosis in RNA expression profiles across different platforms

Molecular signatures have been excessively reported for diagnosis of many cancers during the last 20 years. However, false-positive signatures are always found using statistical methods or machine learning approaches, and that makes subsequent biological experiments fail. Therefore, signature discovery has gradually become a non-mainstream work in bioinformatics. Actually, there are three critical weaknesses that make the identified signature unreliable. First of all, a signature is wrongly thought to be a gene set, each component of which keeps differential expressions between or among sample groups. Second, there may be many false-positive genes expressed differentially found, even if samples derived from cancer or normal group can be separated in one-dimensional space. Third, cross-platform validation results of a discovered signature are always poor. In order to solve these problems, we propose a new feature selection framework based on ensemble classification to discover signatures for cancer diagnosis. Meanwhile, a procedure for data transform among different expression profiles across different platforms is also designed. Signatures are found on simulation and real data representing different carcinomas across different platforms. Besides, false positives are suppressed. The experimental results demonstrate the effectiveness of our method.

Gene expression analysis of combined RNA-seq experiments using a receiver operating characteristic calibrated procedure

Because of rapid advancements in sequencing technology, the experimental platforms of RNA-seq are updated frequently. It is quite common to combine data sets from several experimental platforms for analysis in order to increase the sample size and achieve more powerful tests for detecting the presence of differential gene expression. The data sets combined from different experimental platforms will have a complex data distribution, which causes a major problem in statistical modeling as well as in multiple testing. Although plenty of research have studied this problem by modeling the batch effects, there are no general and robust data-driven procedures for RNA-seq analysis. In this paper we propose a new robust procedure which combines the use of popular methods (packages) with a data-driven simulation (DDS). We construct the average receiver operating characteristic curve through the DDS to provide the calibrated levels of significance for multiple testing. Instead of further modifying the adjusted p-values, we calibrated the levels of significance for each specific method and mean effect model. The procedure was demonstrated with several popular RNA-seq analysis methods (edgeR, DEseq2, limma+voom). The proposed procedure relaxes the stringent assumptions of data distributions for RNA-seq analysis methods and is illustrated using colorectal cancer studies from The Cancer Genome Atlas database.

Immune-related gene expression signatures in colorectal cancer

The immune system is crucial in regulating colorectal cancer (CRC) tumorigenesis. Identification of immune-related transcriptomic signatures derived from the peripheral blood of patients with CRC would provide insights into CRC pathogenesis, and suggest novel clues to potential immunotherapy strategies for the disease. The present study collected blood samples from 59 patients with CRC and 62 healthy control patients and performed whole blood gene expression profiling using microarray hybridization. Immune-related gene expression signatures for CRC were identified from immune gene datasets, and an algorithmic predictive model was constructed for distinguishing CRC from controls. Model performance was characterized using an area under the receiver operating characteristic curve (ROC AUC). Functional categories for CRC-specific gene expression signatures were determined using gene set enrichment analyses. A Kaplan-Meier plotter survival analysis was also performed for CRC-specific immune genes in order to characterize the association between gene expression and CRC prognosis. The present study identified five CRC-specific immune genes [protein phosphatase 3 regulatory subunit Bα (PPP3R1), amyloid β precursor protein, cathepsin H, proteasome activator subunit 4 and DEAD-Box Helicase 3 X-Linked]. A predictive model based on this five-gene panel showed good discriminatory power (independent test set sensitivity, 83.3%; specificity, 94.7%, accuracy, 89.2%; ROC AUC, 0.96). The candidate genes were involved in pathways associated with ‘adaptive immune responses’, ‘innate immune responses’ and ‘cytokine signaling’. The survival analysis found that a high level of PPP3R1 expression was associated with a poor CRC prognosis. The present study identified five CRC-specific immune genes that were potential diagnostic biomarkers for CRC. The biological function analysis indicated a close association between CRC pathogenesis and the immune system, and may reveal more information about the immunogenic and pathogenic mechanisms driving CRC in the future. Overall, the association between PPP3R1 expression and survival of patients with CRC revealed potential new targets for CRC immunotherapy.

Machine learning-based classification and diagnosis of clinical cardiomyopathies

Dilated cardiomyopathy (DCM) and ischemic cardiomyopathy (ICM) are two common types of cardiomyopathies leading to heart failure. Accurate diagnostic classification of different types of cardiomyopathies is critical for precision medicine in clinical practice. In this study, we hypothesized that machine learning (ML) can be used as a novel diagnostic approach to analyze cardiac transcriptomic data for classifying clinical cardiomyopathies. RNA-Seq data of human left ventricle tissues were collected from 41 DCM patients, 47 ICM patients, and 49 nonfailure controls (NF) and tested using five ML algorithms: support vector machine with radial kernel (svmRadial), neural networks with principal component analysis (pcaNNet), decision tree (DT), elastic net (ENet), and random forest (RF). Initial ML classifications achieved ~93% accuracy (svmRadial) for NF vs. DCM, ~82% accuracy (RF) for NF vs. ICM, and ~80% accuracy (ENet and svmRadial) for DCM vs. ICM. Next, 50 highly contributing genes (HCGs) for classifying NF and DCM, 68 HCGs for classifying NF and ICM, and 59 HCGs for classifying DCM and ICM were selected for retraining ML models. Impressively, the retrained models achieved ~90% accuracy (RF) for NF vs. DCM, ~90% accuracy (pcaNNet) for NF vs. ICM, and ~85% accuracy (pcaNNet and RF) for DCM vs. ICM. Pathway analyses further confirmed the involvement of those selected HCGs in cardiac dysfunctions such as cardiomyopathies, cardiac hypertrophies, and fibrosis. Overall, our study demonstrates the promising potential of using artificial intelligence via ML modeling as a novel approach to achieve a greater level of precision in diagnosing different types of cardiomyopathies.

A qualitative transcriptional signature for predicting microsatellite instability status of right-sided Colon Cancer

Background

Microsatellite instability (MSI) accounts for about 15% of colorectal cancer and is associated with prognosis. Today, MSI is usually detected by polymerase chain reaction amplification of specific microsatellite markers. However, the instability is identified by comparing the length of microsatellite repeats in tumor and normal samples. In this work, we developed a qualitative transcriptional signature to individually predict MSI status for right-sided colon cancer (RCC) based on tumor samples.

Results

Using RCC samples, based on the relative expression orderings (REOs) of gene pairs, we extracted a signature consisting of 10 gene pairs (10-GPS) to predict MSI status for RCC through a feature selection process. A sample is predicted as MSI when the gene expression orderings of at least 7 gene pairs vote for MSI; otherwise the microsatellite stability (MSS). The classification performance reached the largest F-score in the training dataset. This signature was verified in four independent datasets of RCCs with the F-scores of 1, 0.9630, 0.9412 and 0.8798, respectively. Additionally, the hierarchical clustering analyses and molecular features also supported the correctness of the reclassifications of the MSI status by 10-GPS.

Conclusions

The qualitative transcriptional signature can be used to classify MSI status of RCC samples at the individualized level.