Diagnosis of thyroid neoplasm using support vector machine algorithms based on platelet RNA-seq

The use of platelets as a clinical tool in oncology: opportunities and challenges

Platelets are small circulating anucleated cells mainly involved in thrombosis and hemostasis processes. Moreover, platelets play an active role in tumorigenesis and cancer progression, stimulating angiogenesis and vascular remodelling, and protecting circulating cancer cells from shear forces and immune surveillance. Several reports indicate that platelet number in the blood circulation of cancer patients is associated with prognosis and response to treatment. However, the mechanisms of platelets "education" by cancer cells and the crosstalk between platelets and tumor are still unclear, and the role of "tumor educated platelets" (TEPs) is achieving growing interest in cancer research. TEPs are a biological source of cancer-derived biomarkers, especially RNAs that are protected by platelets membrane from circulating RNases, and could serve as a non-invasive tool for tumor detection, molecular profiling and evolution during therapy in clinical practice. Moreover, short platelet lifespan offers the possibility to get a snapshot assessment of cancer molecular profile, providing a real-time tool. We review and discuss the potential and the clinical utility, in terms of cancer diagnosis and monitoring, of platelet count together with other morphological parameters and of the more recent and innovative TEP profiling.

Multi-view progression diagnosis of thyroid cancer by integrating platelet transcriptomes and blood routine tests

Thyroid cancer is the most common type of endocrine system cancer. The pre-cancer and early stages are usually benign or slowly growing, and do not need invasive treatments. This study investigated the challenging classification task of four classes of samples, i.e., normal controls (N), thyroid adenomas (TA), papillary thyroid cancers (PTC) and metastasized papillary thyroid cancers (MPTC). We proposed a multi-view progression diagnosis framework ThyroidBloodTest to integrate the two views of RNAseq platelet transcriptomes (View-T) and blood routine (View-B) features. Platelet transcriptome represented the molecular-level information, while the blood routine features were easy to obtain in the clinical practice. Eleven feature selection algorithms and seven classifiers were evaluated for both views. The experimental data suggested the importance of choosing appropriate data analysis algorithms and feature engineering techniques like principal component analysis (PCA). The best ThyroidBloodTest model achieved Acc = 0.8750 for the four-class classification of the N/TA/PTC/MPTC samples based on the integrated feature space of View-T and View-B. The cellular localization cytosol and three post-translational modification types acetylation/phosphorylation/ubiquitination were observed to be enriched in the proteins encoded by the View-T biomarkers. The numbers of different immune cells also contributed positively to the progression diagnosis of thyroid cancer. The proposed multi-view prediction model demonstrated the necessity of integrating both platelet transcriptomes and blood routine tests for the progression diagnosis of thyroid cancer.

Bleeding and Thrombosis in Multiple Myeloma: Platelets as Key Players during Cell Interactions and Potential Use as Drug Delivery Systems

Multiple myeloma (MM) is a hematological malignancy originated in the bone marrow and characterized by unhindered plasma cell proliferation that results in several clinical manifestations. Although the main role of blood platelets lies in hemostasis and thrombosis, platelets also play a pivotal role in a number of other pathological conditions. Platelets are the less-explored components from the tumor microenvironment in MM. Although some studies have recently revealed that MM cells have the ability to activate platelets even in the premalignant stage, this phenomenon has not been widely investigated in MM. Moreover, thrombocytopenia, along with bleeding, is commonly observed in those patients. In this review, we discuss the hemostatic disturbances observed in MM patients and the dynamic interaction between platelets and myeloma cells, along with present and future potential avenues for the use of platelets for diagnostic and therapeutic purposes.

A systematic review on artificial intelligence techniques for detecting thyroid diseases

The use of artificial intelligence approaches in health-care systems has grown rapidly over the last few years. In this context, early detection of diseases is the most common area of application. In this scenario, thyroid diseases are an example of illnesses that can be effectively faced if discovered quite early. Detecting thyroid diseases is crucial in order to treat patients effectively and promptly, by saving lives and reducing healthcare costs. This work aims at systematically reviewing and analyzing the literature on various artificial intelligence-related techniques applied to the detection and identification of various diseases related to the thyroid gland. The contributions we reviewed are classified according to different viewpoints and taxonomies in order to highlight pros and cons of the most recent research in the field. After a careful selection process, we selected and reviewed 72 papers, analyzing them according to three main research questions, i.e., which diseases of the thyroid gland are detected by different artificial intelligence techniques, which datasets are used to perform the aforementioned detection, and what types of data are used to perform the detection. The review demonstrates that the majority of the considered papers deal with supervised methods to detect hypo- and hyperthyroidism. The average accuracy of detection is high (96.84%), but the usage of private and outdated datasets with a majority of clinical data is very common. Finally, we discuss the outcomes of the systematic review, pointing out advantages, disadvantages, and future developments in the application of artificial intelligence for thyroid diseases detection.

Tumor-educated platelet SNORA58, SNORA68 and SNORD93 as novel diagnostic biomarkers for esophageal cancer

Aim: The purpose of this study was to evaluate whether tumor-educated platelet (TEP) snoRNAs could be used as a diagnostic biomarker for esophageal cancer (ESCA). Methods: Platelet precipitates were obtained from platelet-rich plasma by low-speed centrifugation, and total RNA was extracted from platelets using Trizol™ reagent. RT-qPCR was used to detect snoRNA expression, and the receiver operating characteristic was used to assess its diagnostic potential. Results: SNORA58, SNORA68 and SNORD93 were significantly upregulated in TEPs from ESCA patients and early-stage patients compared with healthy controls. Importantly, the three snoRNAs were capable of serving as circulating biomarkers of diagnostics and early diagnosis of ESCA, possessing areas under the curve of 0.846 and 0.857, respectively. Conclusion: TEP SNORA58, SNORA68 and SNORD93 could potentially serve as noninvasive biomarkers for diagnosis and early diagnosis of ESCA.

A narrative review for platelets and their RNAs in cancers: New concepts and clinical perspectives

Recent years have witnessed a growing body of evidence suggesting that platelets are involved in several stages of the metastatic process via direct or indirect interactions with cancer cells, contributing to the progression of neoplastic malignancies. Cancer cells can dynamically exchange components with platelets in and out of blood vessels, and directly phagocytose platelets to hijack their proteome, transcriptome, and secretome, or be remotely regulated by metabolites or microparticles released by platelets, resulting in phenotypic, genetic, and functional modifications. Moreover, platelet interactions with stromal and immune cells in the tumor microenvironment lead to alterations in their components, including the ribonucleic acid (RNA) profile, and complicate the impact of platelets on cancers. A deeper understanding of the roles of platelets and their RNAs in cancer will contribute to the development of anticancer strategies and the optimization of clinical management. Encouragingly, advances in high-throughput sequencing, bioinformatics data analysis, and machine learning have allowed scientists to explore the potential of platelet RNAs for cancer diagnosis, prognosis, and guiding treatment. However, the clinical application of this technique remains controversial and requires larger, multicenter studies with standardized protocols. Here, we integrate the latest evidence to provide a broader insight into the role of platelets in cancer progression and management, and propose standardized recommendations for the clinical utility of platelet RNAs to facilitate translation and benefit patients.

Impact of Local Control on Clinical Course in Stage IVC Anaplastic Thyroid Carcinoma

BACKGROUND

The present study investigated the association between local resection and cause of death in anaplastic thyroid carcinoma (ATC) patients with stage IVC disease.

METHODS

A total of 54 ATC patients with stage IVC disease were included in the study. Information including patient characteristics, laboratory data including complete blood count, treatment, and death were collected for analysis.

RESULTS

The median overall survival (OS) for patients with or without resection was 8.4 [95% confidence interval (CI) 5.9-14.4)] and 4.2 (95% CI 2.5-6.2) months, respectively (p < 0.001). No patients survived without resection at 1 year. Univariate analysis revealed that resection (p < 0.001) and radiotherapy (p = 0.018) were significantly associated with OS. Multivariate analysis revealed that resection (hazard ratio 0.257; 95% CI 0.115-0.575; p < 0.001) was the only independent prognostic factor of OS. In ATC patients with known resection status, the median OS for the patients with a resection status of R0/1 (n = 28) and R2 (n = 7) were 13.0 (95% CI 7.5-18.7) and 1.7 (95% CI 0.1-6.2) months, respectively (p < 0.001). The most common specific cause of death was respiratory insufficiency (35%), followed by airway obstruction (25%) and cerebral cardiovascular-related death (5%). The frequency of airway obstruction was significantly lower in patients with resection (p = 0.018).

CONCLUSIONS

Resection probably impacts on clinical course in ATC patients despite the presence of distant metastasis. However, R2 resection is likely to be harmful and surgeons should carefully consider the resectability of thyroid tumors.

Detection and localization of early- and late-stage cancers using platelet RNA

Cancer patients benefit from early tumor detection since treatment outcomes are more favorable for less advanced cancers. Platelets are involved in cancer progression and are considered a promising biosource for cancer detection, as they alter their RNA content upon local and systemic cues. We show that tumor-educated platelet (TEP) RNA-based blood tests enable the detection of 18 cancer types. With 99% specificity in asymptomatic controls, thromboSeq correctly detected the presence of cancer in two-thirds of 1,096 blood samples from stage I-IV cancer patients and in half of 352 stage I-III tumors. Symptomatic controls, including inflammatory and cardiovascular diseases, and benign tumors had increased false-positive test results with an average specificity of 78%. Moreover, thromboSeq determined the tumor site of origin in five different tumor types correctly in over 80% of the cancer patients. These results highlight the potential properties of TEP-derived RNA panels to supplement current approaches for blood-based cancer screening.

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.

Tumor-Educated Platelets as a Promising Biomarker for Blood-Based Detection of Renal Cell Carcinoma

Purpose

Tumor-educated platelets (TEPs) are a promising liquid biopsy in many cancers. However, their role in renal cell carcinoma (RCC) is unknown. Thus, this study explored the diagnostic value of TEPs in RCC patients.

Methods

Platelets were prospectively collected from 24 RCC patients and 25 controls. RNA-seq was performed to identify the differentially expressed genes (DEGs) between RCC patients and controls. Besides, RNA-seq data of pan-cancer TEPs were downloaded and randomly divided into training and validation sets. A pan-cancer TEP model was developed in the training set using the support vector machine (SVM) and validated in the validation set and our RCC dataset. Finally, an RCC-based TEP model was developed and optimized through the SVM algorithms and recursive feature elimination (RFE) method.

Result

Two hundred three DEGs, 64 (31.5%) upregulated and 139 (68.5%) downregulated, were detected in the platelets of RCC patients compared with controls. The pan-cancer TEP model had a high accuracy in detecting cancer in the internal validation (training set, accuracy 98.8%, AUC: 0.999; validation set, accuracy 95.4%, AUC: 0.972; different tumor subtypes, accuracy 86.6%–96.1%, AUC: 0.952–1.000). However, the pan-cancer TEP model in the external validation had a scarce diagnostic value in RCC patients (accuracy 48.7%, AUC: 0.615). Therefore, to develop the RCC-based TEP model, the gene biomarkers mostly contributing to the model were selected using the RFE method. The RCC-based TEP model containing 68 gene biomarkers reached a diagnostic accuracy of 100% (AUC: 1.000) in the training set, 88.9% (AUC: 0.963) in the validation set, and 95.9% (AUC: 0.988) in the overall cohort.

Conclusion

TEPs could function as a minimally invasive blood biomarker in the detection of RCC.

A Novel and Effective Model to Predict Skip Metastasis in Papillary Thyroid Carcinoma Based on a Support Vector Machine

INTRODUCTION

Skip metastasis, referred to as lymph node metastases to the lateral neck compartment without involvement of the central compartment, is generally unpredictable in papillary thyroid carcinoma (PTC). This study aims to establish an effective predictive model for skip metastasis in PTC.

METERIALS AND METHODS

Retrospective analysis was performed of clinical samples from 18192 patients diagnosed with thyroid cancer between 2016 to 2020. The First Affiliated Hospital of Wenzhou Medical University. The lateral lymph node metastasis was occureed in the training set (630 PTC patients) and validation set (189 PTC patients). The univariate and multivariate analyses were performed to detect the predictors of skip metastasis and the support vector machine (SVM) was used to establish a model to predict skip metastasis.

RESULTS

The rate of skip metastasis was 13.3% (84/631). Tumor size (≤10 mm), upper location, Hashimoto's thyroiditis, extrathyroidal extension, absence of BRAFV600E mutation, and less number of central lymph node dissection were considered as independent predictors of skip metastasis in PTC. For the training set, these predictors performed with 91.7% accuracy, 86.4% sensitivity, 92.2% specificity, 45.2% positive predictive value (PPV), and 98.9% negative predictive value (NPV) in the model. Meanwhile, these predictors showed 91.5% accuracy,71.4% sensitivity, 93.1% specificity, 45.5% PPV, and 97.6% NPV in validation set.

CONCLUSION

This study screened the predictors of the skip lateral lymph node metastasis and to establish an effective and economic predictive model for skip metastasis in PTC. The model can accurately distinguish the skip metastasis in PTC using a simple and affordable method, which may have potential for daily clinical application in the future.

What makes a good prediction? Feature importance and beginning to open the black box of machine learning in genetics

Genetic data have become increasingly complex within the past decade, leading researchers to pursue increasingly complex questions, such as those involving epistatic interactions and protein prediction. Traditional methods are ill-suited to answer these questions, but machine learning (ML) techniques offer an alternative solution. ML algorithms are commonly used in genetics to predict or classify subjects, but some methods evaluate which features (variables) are responsible for creating a good prediction; this is called feature importance. This is critical in genetics, as researchers are often interested in which features (e.g., SNP genotype or environmental exposure) are responsible for a good prediction. This allows for the deeper analysis beyond simple prediction, including the determination of risk factors associated with a given phenotype. Feature importance further permits the researcher to peer inside the black box of many ML algorithms to see how they work and which features are critical in informing a good prediction. This review focuses on ML methods that provide feature importance metrics for the analysis of genetic data. Five major categories of ML algorithms: k nearest neighbors, artificial neural networks, deep learning, support vector machines, and random forests are described. The review ends with a discussion of how to choose the best machine for a data set. This review will be particularly useful for genetic researchers looking to use ML methods to answer questions beyond basic prediction and classification.

Identification of the circRNA-miRNA-mRNA regulatory network and its prognostic effect in colorectal cancer

BACKGROUND

The high morbidity and mortality of colorectal cancer (CRC) have posed great threats to human health. Circular RNA (CircRNA) and microRNA (miRNA), acting as competing endogenous RNAs (ceRNAs), have been found to play vital roles in carcinogenesis. However, the biological function of ceRNAs in CRC pathogenesis and prognosis remains largely unexplored.

AIM

To identify the CRC-specific circRNA-miRNA-mRNA regulatory network and uncover the subnetwork associated with its prognosis.

METHODS

CircRNAs, miRNAs and mRNAs differentially expressed (DE) in CRC tissues were selected by expression file analysis in the Gene Expression Omnibus (GEO) database, and the downstream target molecules of circRNAs and miRNAs were predicted. Then, the intersection of differentially expressed RNA molecules with the predicted targets was determined to obtain a ceRNA network. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were conducted to elucidate the possible mechanism of pathogenesis. A survival analysis using the gene profiles and clinical information in The Cancer Genome Atlas (TCGA) database was performed to identify the mRNAs associated with the clinical outcome of CRC patients and construct a prognostic subnetwork.

RESULTS

We downloaded three datasets (GSE126095, GSE41655 and GSE41657) of large-scale CRC samples from the GEO database. There were 55 DEcircRNAs, 114 DEmiRNAs and 267 DEmRNAs in CRC tissues compared with normal tissues. After intersecting these molecules with predicted targets, 19 circRNAs, 13 miRNAs and 28 mRNAs were chosen to develop a circRNA-miRNA-mRNA network. GO and KEGG functional enrichment analyses indicated that the retinol metabolic process, leukocyte chemotaxis, extracellular matrix remodeling, endoplasmic reticulum stress, alcohol dehydrogenase activity, gastric acid secretion, nitrogen metabolism and NOD-like receptor signaling pathway might participate in the tumorigenesis of CRC. After verifying the identified mRNA effect in the TCGA database, we finally recognized 3 mRNAs (CA2, ITLN1 and LRRC19) that were significantly associated with the overall survival of CRC patients and constructed a ceRNA subnetwork including 5 circRNAs (hsa_circ_0080210, hsa_circ_0007158, hsa_circ_0000375, hsa_circ_0018909 and hsa_circ_0011536) and 3 miRNAs (hsa-miR-601, hsa-miR-671-5p and hsa-miR-765), which could contain innovative and noninvasive indicators for the early screening and prognostic prediction of CRC.

CONCLUSION

We proposed a circRNA-miRNA-mRNA regulatory network closely associated with the progression and clinical outcome of CRC that might include promising biomarkers for carcinogenesis and therapeutic targets.