A 19‑miRNA Support Vector Machine classifier and a 6‑miRNA risk score system designed for ovarian cancer patients

Improved lung cancer classification by employing diverse molecular features of microRNAs

MiRNAs are edited or modified in multiple ways during their biogenesis pathways. It was reported that miRNA editing was deregulated in tumors, suggesting the potential value of miRNA editing in cancer classification. Here we extracted three types of miRNA features from 395 LUAD and control samples, including the abundances of original miRNAs, the abundances of edited miRNAs, and the editing levels of miRNA editing sites. Our results show that eight classification algorithms selected generally had better performances on combined features than on the abundances of miRNAs or editing features of miRNAs alone. One feature selection algorithm, i.e., the DFL algorithm, selected only three features, i.e., the frequencies of hsa-miR-135b-5p, hsa-miR-210-3p and hsa-mir-182_48u (an edited miRNA), from 316 training samples. Seven classification algorithms achieved 100% accuracies on these three features for 79 independent testing samples. These results indicate that the additional information of miRNA editing is useful in improving the classification of LUAD samples.

HER2 classification in breast cancer cells: A new explainable machine learning application for immunohistochemistry

The immunohistochemical (IHC) evaluation of epidermal growth factor 2 (HER2) for the diagnosis of breast cancer is still qualitative with a high degree of inter-observer variability, and thus requires the incorporation of complementary techniques such as fluorescent in situ hybridization (FISH) to resolve the diagnosis. Implementing automatic algorithms to classify IHC biomarkers is crucial for typifying the tumor and deciding on therapy for each patient with better performance. The present study aims to demonstrate that, using an explainable Machine Learning (ML) model for the classification of HER2 photomicrographs, it is possible to determine criteria to improve the value of IHC analysis. We trained a logistic regression-based supervised ML model with 393 IHC microscopy images from 131 patients, to discriminate between upregulated and normal expression of the HER2 protein. Pathologists' diagnoses (IHC only) vs. the final diagnosis complemented with FISH (IHC + FISH) were used as training outputs. Basic performance metrics and receiver operating characteristic curve analysis were used together with an explainability algorithm based on Shapley Additive exPlanations (SHAP) values to understand training differences. The model could discriminate amplified IHC from normal expression with better performance when the training output was the IHC + FISH final diagnosis (IHC vs. IHC + FISH: area under the curve, 0.94 vs. 0.81). This may be explained by the increased analytical impact of the membrane distribution criteria over the global intensity of the signal, according to SHAP value interpretation. The classification model improved its performance when the training input was the final diagnosis, downplaying the weighting of the intensity of the IHC signal, suggesting that to improve pathological diagnosis before FISH consultation, it is necessary to emphasize subcellular patterns of staining.

KAZN as a diagnostic marker in ovarian cancer: a comprehensive analysis based on microarray, mRNA-sequencing, and methylation data

Background

Ovarian cancer (OC) is among the deadliest malignancies in women and the lack of appropriate markers for early diagnosis leads to poor prognosis in most cases. Previous studies have shown that KAZN is involved in multiple biological processes during development, such as cell proliferation, differentiation, and apoptosis, so defects or aberrant expression of KAZN might cause queer cell behaviors such as malignancy. Here we evaluated the KAZN expression and methylation levels for possible use as an early diagnosis marker for OC.

Methods

We used data from Gene Expression Omnibus (GEO) microarrays, The Cancer Genome Atlas (TCGA), and Clinical Proteomic Tumor Analysis Consortium (CPTAC) to investigate the correlations between KAZN expression and clinical characteristics of OC by comparing methylation levels of normal and OC samples. The relationships among differentially methylated sites in the KAZN gene, corresponding KAZN mRNA expression levels and prognosis were analyzed.

Results

KAZN was up-regulated in ovarian epithelial tumors and the expression of KAZN was correlated with the patients’ survival time. KAZN CpG site cg17657618 was positively correlated with the expression of mRNA and the methylation levels were significantly differential between the group of stage “I and II” and the group of stage “III and IV”. This study also presents a new method to classify tumor and normal tissue in OC using DNA methylation pattern in the KAZN gene body region.

Conclusions

KAZN was involved in ovarian cancer pathogenesis. Our results demonstrate a new direction for ovarian cancer research and provide a potential diagnostic biomarker as well as a novel therapeutic target for clinical application.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-09747-2.

Discovering Common miRNA Signatures Underlying Female-Specific Cancers via a Machine Learning Approach Driven by the Cancer Hallmark ERBB

Big data processing, using omics data integration and machine learning (ML) methods, drive efforts to discover diagnostic and prognostic biomarkers for clinical decision making. Previously, we used the TCGA database for gene expression profiling of breast, ovary, and endometrial cancers, and identified a top-scoring network centered on the ERBB2 gene, which plays a crucial role in carcinogenesis in the three estrogen-dependent tumors. Here, we focused on microRNA expression signature similarity, asking whether they could target the ERBB family. We applied an ML approach on integrated TCGA miRNA profiling of breast, endometrium, and ovarian cancer to identify common miRNA signatures differentiating tumor and normal conditions. Using the ML-based algorithm and the miRTarBase database, we found 205 features and 158 miRNAs targeting ERBB isoforms, respectively. By merging the results of both databases and ranking each feature according to the weighted Support Vector Machine model, we prioritized 42 features, with accuracy (0.98), AUC (0.93–95% CI 0.917–0.94), sensitivity (0.85), and specificity (0.99), indicating their diagnostic capability to discriminate between the two conditions. In vitro validations by qRT-PCR experiments, using model and parental cell lines for each tumor type showed that five miRNAs (hsa-mir-323a-3p, hsa-mir-323b-3p, hsa-mir-331-3p, hsa-mir-381-3p, and hsa-mir-1301-3p) had expressed trend concordance between breast, ovarian, and endometrium cancer cell lines compared with normal lines, confirming our in silico predictions. This shows that an integrated computational approach combined with biological knowledge, could identify expression signatures as potential diagnostic biomarkers common to multiple tumors.

A novel cross-species differential tumor classification method based on exosome-derived microRNA biomarkers established by human-dog lymphoid and mammary tumor cell lines' transcription profiles

Background and Aim:

Exosome-derived microRNA (miRNA) has been widely studied as a non-invasive candidate biomarker for tumor diagnosis in humans and dogs. Its application, however, was primarily focused on intraspecies usage for individual tumor type diagnosis. This study aimed to gain insight into its application as a cross-species differential tumor diagnostic tool; we demonstrated the process of identifying and using exosome-derived miRNA as biomarkers for the classification of lymphoid and mammary tumor cell lines in humans and dogs.

Materials and Methods:

Exosome-derived miRNA sequencing data from B-cell lymphoid tumor cell lines (n=13), mammary tumor cell lines (n=8), and normal mammary epithelium cultures (n=4) were pre-processed in humans and dogs. F-test and rank product (RP) analyses were used to select candidate miRNA orthologs for tumor cell line classification. The classification was carried out using an optimized support vector machine (SVM) with various kernel classifiers, including linear SVM, polynomial SVM, and radial basis function SVM. The receiver operating characteristic and precision-recall curves were used to assess the performance of all models.

Results:

MIR10B, MIR21, and MIR30E were chosen as the candidate orthologs from a total of 236 human-dog miRNA orthologs (p≤0.01, F-test score ≥10, and RP score ≤10). Their use of polynomial SVM provided the best performance in classifying samples from various tumor cell lines and normal epithelial culture.

Conclusion:

The study successfully demonstrated a method for identifying and utilizing candidate human-dog exosome-derived miRNA orthologs for differential tumor cell line classification. Such findings shed light on a novel non-invasive tumor diagnostic tool that could be used in both human and veterinary medicine in the future.

Circulating Exosomal miRNAs as Biomarkers in Epithelial Ovarian Cancer

Failure to detect early-stage epithelial ovarian cancer (EOC) is a major contributing factor to its low survival rate. Increasing evidence suggests that different subtypes of EOC may behave as distinct diseases due to their different cells of origins, histology and treatment responses. Therefore, the identification of EOC subtype-specific biomarkers that can early detect the disease should be clinically beneficial. Exosomes are extracellular vesicles secreted by different types of cells and carry biological molecules, which play important roles in cell-cell communication and regulation of various biological processes. Multiple studies have proposed that exosomal miRNAs present in the circulation are good biomarkers for non-invasive early detection of cancer. In this review, the potential use of exosomal miRNAs as early detection biomarkers for EOCs and their accuracy are discussed. We also review the differential expression of circulating exosomal miRNAs and cell-free miRNAs between different biofluid sources, i.e., plasma and serum, and touch on the issue of endogenous reference miRNA selection. Additionally, the current clinical trials using miRNAs for detecting EOCs are summarized. In conclusion, circulating exosomal miRNAs as the non-invasive biomarkers have a high potential for early detection of EOC and its subtypes, and are likely to be clinically important in the future.

Identification of Seven-Gene Hypoxia Signature for Predicting Overall Survival of Hepatocellular Carcinoma

Background

Hepatocellular carcinoma (HCC) is ranked fifth among the most common cancer worldwide. Hypoxia can induce tumor growth, but the relationship with HCC prognosis remains unclear. Our study aims to construct a hypoxia-related multigene model to predict the prognosis of HCC.

Methods

RNA-seq expression data and related clinical information were download from TCGA database and ICGC database, respectively. Univariate/multivariate Cox regression analysis was used to construct prognostic models. KM curve analysis, and ROC curve were used to evaluate the prognostic models, which were further verified in the clinical traits and ICGC database. GSEA analyzed pathway enrichment in high-risk groups. Nomogram was constructed to predict the personalized treatment of patients. Finally, real-time fluorescence quantitative PCR (RT-qPCR) was used to detect the expressions of KDELR3 and SCARB1 in normal hepatocytes and 4 HCC cells. The expressions of SCARB1 in hepatocellular carcinoma tissue in 46 patients were detected by immunohistochemistry, and the correlation between its expressions and disease free survival of patient was calculated.

Results

Through a series of analyses, seven prognostic markers related to HCC survival were constructed. HCC patients were divided into the high and low risk group, and the results of KM curve showed that there was a significant difference between the two groups. Stratified analysis, found that there were significant differences in risk values of different ages, genders, stages and grades, which could be used as independent predictors. In addition, we assessed the risk value in the clinical traits analysis and found that it could accelerate the progression of cancer, while the results of GSEA enrichment analysis showed that the high-risk group patients were mainly distributed in the cell cycle and other pathways. Then, Nomogram was constructed to predict the overall survival of patients. Finally, RT-qPCR showed that KDELR3 and SCARB1 were highly expressed in HepG2 and L02, respectively. Results of IHC staining showed that SCARB1 was highly expressed in cancer tissues compared to adjacent normal liver tissues and its expression was related to hepatocellular carcinoma differentiation status. The Kaplan-Meier survival showed a poor percent survival in the SCARB1 high group compared to that in the SCARB1 low group.

Conclusion

This study provides a potential diagnostic indicator for HCC patients, and help clinicians to deepen the comprehension in HCC pathogenesis so as to make personalized medical decisions.

A Support Vector Machine Model Predicting the Risk of Duodenal Cancer in Patients with Familial Adenomatous Polyposis at the Transcript Levels

Objective

Familial adenomatous polyposis (FAP) is one major type of inherited duodenal cancer. The estimate of duodenal cancer risk in patients with FAP is critical for selecting the optimal treatment strategy.

Methods

Microarray datasets related with FAP were retrieved from the Gene Expression Omnibus (GEO) database. Differentially expressed genes were identified by FAP vs. normal samples and FAP and duodenal cancer vs. normal samples. Furthermore, functional enrichment analyses of these differentially expressed genes were performed. A support vector machine (SVM) was performed to train and validate cancer risk prediction model.

Results

A total of 196 differentially expressed genes were identified between FAP compared with normal samples. 177 similarly expressed genes were identified both in FAP and duodenal cancer, which were mainly enriched in pathways in cancer and metabolic-related pathway, indicating that these genes in patients with FAP could contribute to duodenal cancer. Among them, Cyclin D1, SDF-1, AXIN, and TCF were significantly upregulated in FAP tissues using qRT-PCR. Based on the 177 genes, an SVM model was constructed for prediction of the risk of cancer in patients with FAP. After validation, the model can accurately distinguish FAP patients with high risk from those with low risk for duodenal cancer.

Conclusion

This study proposed a cancer risk prediction model based on an SVM at the transcript levels.

Identification and validation of seven prognostic long non-coding RNAs in oral squamous cell carcinoma

Oral squamous cell carcinoma (OSCC) is one of the most common malignancies worldwide, due to poor diagnosis and treatment. There is increasing evidence that demonstrates the involvement of long non-coding RNAs (lncRNAs) in carcinogenesis and cancer progression. Therefore, the aim of the present study was to explore potential lncRNA-associated features of patients with OSCC as a valuable and independent prognostic biomarker. A total of 268 lncRNA expression profiles and clinical patient information on OSCC were downloaded from The Cancer Genome Atlas database. The clinical information was exploited for prescreening, using Cox regression analysis, and differentially expressed lncRNAs (DElncRNAs) were identified using edgeR software. Using the 'caret' package, the datasets were categorized into test datasets and training datasets, respectively. Through bioinformatics, seven prognostic DElncRNAs were selected. Using the regression coefficients, a risk score based on the seven-DElncRNA signature was developed to assess the prognostic function of key DElncRNAs. According to the median risk score, patients were classified into high-risk and low-risk groups in the training and test datasets. Additionally, receiver operating characteristic (ROC) curve analysis was conducted to evaluate the sensitivity and specificity of the prognostic DElncRNAs, and the optimal cut-off point was obtained from ROC analysis. Based on the optimal cut-off point, the patients were also categorized into high-risk and low-risk groups. Notably, the optimal cut-off point was more sensitive than the median risk score, particularly in the test dataset. The Kaplan-Meier survival and log rank test analysis results indicated that the P-value, based on the optimal cut-off, was less than the median risk cut-off. Additionally, stratified analysis results revealed that the seven-DElncRNAs signature was also independent of OSCC age. Furthermore, the findings of the present study suggested that the seven-DElncRNA signature can be used as a potential prognostic indicator and may have important clinical significance in OSCC.

Recurrence-Associated Multi-RNA Signature to Predict Disease-Free Survival for Ovarian Cancer Patients

Ovarian cancer (OvCa) is an intractable gynecological malignancy due to the high recurrence rate. Several molecular biomarkers have been previously screened for early identifying patients with a high recurrence risk and poor prognosis. However, all the known studies focused on a single type of RNAs, not integrating various types. This study was to construct a new multi-RNA-based model to predict the recurrence and prognosis for OvCa patients by using the messenger RNA (mRNA, including long noncoding RNA (lncRNA)) and microRNA (miRNA) sequencing data of The Cancer Genome Atlas database. After univariate Cox regression and least absolute shrinkage and selection operator analyses, a multi-RNA-based signature (2 miRNAs: hsa-miR-508, hsa-miR-506; 1 lncRNA: TM4SF1-AS1; 11 mRNAs: MAGI3, SLAMF7, GLI2, PDK1, ARID3A, PLEKHG4B, TNFAIP8L3, C1QTNF3, NDUFAF1, CH25H, TMEM129) was generated and used to establish a risk score model. The high- and low-risk patients classified by the median risk score exhibited significantly different recurrence risks (89% versus 61%, p < 0.001) and survival time (the area under the receiver operating characteristic curve (AUC) = 0.901 for 5-year disease-free survival (DFS)). This risk model was independent of other clinical features and superior to pathologic staging for DFS prediction (AUC, 0.906 versus 0.524; C-index, 0.633 versus 0.510). Furthermore, some new interaction axes were revealed to explain the possible functions of these RNAs (competing endogenous RNA: TM4SF1-AS1-miR-186-STEAP2, LINC00536-miR-508-STEAP2, LINC00475-miR-506-TMEM129; coexpression: LINC00598-PLEKHG4B). In conclusion, this multi-RNA-based risk model may be clinically useful to stratify OvCa patients with different recurrence risks and survival outcomes and included RNAs may be potential therapeutic targets.

Role of microRNAs as Clinical Cancer Biomarkers for Ovarian Cancer: A Short Overview

Ovarian cancer has the highest mortality rate among gynecological cancers. Early clinical signs are missing and there is an urgent need to establish early diagnosis biomarkers. MicroRNAs are promising biomarkers in this respect. In this paper, we review the most recent advances regarding the alterations of microRNAs in ovarian cancer. We have briefly described the contribution of miRNAs in the mechanisms of ovarian cancer invasion, metastasis, and chemotherapy sensitivity. We have also summarized the alterations underwent by microRNAs in solid ovarian tumors, in animal models for ovarian cancer, and in various ovarian cancer cell lines as compared to previous reviews that were only focused the circulating microRNAs as biomarkers. In this context, we consider that the biomarker screening should not be limited to circulating microRNAs per se, but rather to the simultaneous detection of the same microRNA alteration in solid tumors, in order to understand the differences between the detection of nucleic acids in early vs. late stages of cancer. Moreover, in vitro and in vivo models should also validate these microRNAs, which could be very helpful as preclinical testing platforms for pharmacological and/or molecular genetic approaches targeting microRNAs. The enormous quantity of data produced by preclinical and clinical studies regarding the role of microRNAs that act synergistically in tumorigenesis mechanisms that are associated with ovarian cancer subtypes, should be gathered, integrated, and compared by adequate methods, including molecular clustering. In this respect, molecular clustering analysis should contribute to the discovery of best biomarkers-based microRNAs assays that will enable rapid, efficient, and cost-effective detection of ovarian cancer in early stages. In conclusion, identifying the appropriate microRNAs as clinical biomarkers in ovarian cancer might improve the life quality of patients.