Meta- and reanalysis of gene expression profiles of hot and cold thyroid nodules and papillary thyroid carcinoma for gene groups

Molecular Tests for Risk-Stratifying Cytologically Indeterminate Thyroid Nodules: An Overview of Commercially Available Testing Platforms in the United States

The past decade has witnessed significant advances in the application of molecular diagnostics for the pre-operative risk-stratification of cytologically indeterminate thyroid nodules. The tests that are currently marketed in the United States for this purpose combine aspects of tumor genotyping with gene and/or microRNA expression profiling. This review compares the general methodology and clinical validation studies for the three tests currently offered in the United States: ThyroSeq v3, Afirma GSC and Xpression Atlas, and ThyGeNEXT/ThyraMIR.

Updates in Thyroid Cytology

Genomic, clinical, and pathologic studies have prompted a more risk-stratified approach to the management of patients with thyroid nodules. The recent nomenclature change concerning noninvasive follicular thyroid neoplasm with papillary-like nuclear features reflects the clinical trend toward conservative treatment choices for carefully selected low-risk thyroid neoplasms. These developments have occurred in parallel with a growing array of molecular tests intended to improve clinical triage for patients with indeterminate fine needle aspiration diagnoses. This review discusses the implications of the nomenclature revision on the interpretation of thyroid fine needle aspiration and updates available ancillary molecular tests for thyroid fine needle aspirations.

Radiation signatures in childhood thyroid cancers after the Chernobyl accident: possible roles of radiation in carcinogenesis

After the Tokyo Electric Power Company Fukushima Daiichi nuclear power plant accident, cancer risk from low-dose radiation exposure has been deeply concerning. The linear no-threshold model is applied for the purpose of radiation protection, but it is a model based on the concept that ionizing radiation induces stochastic oncogenic alterations in the target cells. As the elucidation of the mechanism of radiation-induced carcinogenesis is indispensable to justify the concept, studies aimed at the determination of molecular changes associated with thyroid cancers among children who suffered effects from the Chernobyl nuclear accident will be overviewed. We intend to discuss whether any radiation signatures are associated with radiation-induced childhood thyroid cancers.

Classification of follicular cell-derived thyroid cancer by global RNA profiling

The incidence of thyroid cancer is increasing worldwide and thyroid nodules are a frequent clinical finding. Diagnosing follicular cell-derived cancers is, however, challenging both histopathologically and especially cytopathologically. The advent of high-throughput molecular technologies has prompted many researchers to explore the transcriptome and, in recent years, also the miRNome in order to generate new molecular classifiers capable of classifying thyroid tumours more accurately than by conventional cytopathological and histopathological methods. This has led to a number of molecular classifiers that may differentiate malignant from benign thyroid nodules. Molecular classification models based on global RNA profiles from fine-needle aspirations are currently being evaluated; results are preliminary and lack validation in prospective clinical trials. There is no doubt that molecular classification will not only contribute to our biological insight but also improve clinical and pathological examinations, thus advancing thyroid tumour diagnosis and ultimately preventing superfluous surgery. This review evaluates the status of classification and biological insights gained from molecular profiling of follicular cell-derived thyroid cancers.

Identification of SERPINA1 as single marker for papillary thyroid carcinoma through microarray meta analysis and quantification of its discriminatory power in independent validation

Background

Several DNA microarray based expression signatures for the different clinically relevant thyroid tumor entities have been described over the past few years. However, reproducibility of these signatures is generally low, mainly due to study biases, small sample sizes and the highly multivariate nature of microarrays. While there are new technologies available for a more accurate high throughput expression analysis, we show that there is still a lot of information to be gained from data deposited in public microarray databases. In this study we were aiming (1) to identify potential markers for papillary thyroid carcinomas through meta analysis of public microarray data and (2) to confirm these markers in an independent dataset using an independent technology.

Methods

We adopted a meta analysis approach for four publicly available microarray datasets on papillary thyroid carcinoma (PTC) nodules versus nodular goitre (NG) from N2-frozen tissue. The methodology included merging of datasets, bias removal using distance weighted discrimination (DWD), feature selection/inference statistics, classification/crossvalidation and gene set enrichment analysis (GSEA). External Validation was performed on an independent dataset using an independent technology, quantitative RT-PCR (RT-qPCR) in our laboratory.

Results

From meta analysis we identified one gene (SERPINA1) which identifies papillary thyroid carcinoma against benign nodules with 99% accuracy (n = 99, sensitivity = 0.98, specificity = 1, PPV = 1, NPV = 0.98). In the independent validation data, which included not only PTC and NG, but all major histological thyroid entities plus a few variants, SERPINA1 was again markedly up regulated (36-fold, p = 1:3*10^-10) in PTC and identification of papillary carcinoma was possible with 93% accuracy (n = 82, sensitivity = 1, specificity = 0.90, PPV = 0.76, NPV = 1). We also show that the extracellular matrix pathway is strongly activated in the meta analysis data, suggesting an important role of tumor-stroma interaction in the carcinogenesis of papillary thyroid carcinoma.

Conclusions

We show that valuable new information can be gained from meta analysis of existing microarray data deposited in public repositories. While single microarray studies rarely exhibit a sample number which allows robust feature selection, this can be achieved by combining published data using DWD. This approach is not only efficient, but also very cost-effective. Independent validation shows the validity of the results from this meta analysis and confirms SERPINA1 as a potent mRNA marker for PTC in a total (meta analysis plus validation) of 181 samples.

Copy number and gene expression alterations in radiation-induced papillary thyroid carcinoma from chernobyl pediatric patients

BACKGROUND

Following exposure to radiation during the Chernobyl fallout tragedy, papillary thyroid carcinoma (PTC) increased significantly in individuals who were children at the time of the accident. We have used two high-throughput, whole genome platforms to analyze radiation-induced PTCs from pediatric patients from the Chernobyl region.

METHODS

We performed comparative genomic hybridization using Affymetrix 50K Mapping arrays and gene expression profiling on 10 pediatric post-Chernobyl PTCs obtained from patients living in the region. We performed an overlay analysis of these two data sets.

RESULTS

Many regions of copy number alterations (CNAs) were detected including novel regions that had never been associated with PTCs. Increases in copy numbers were consistently found on chromosomes 1p, 5p, 9q, 12q, 13q, 16p, 21q, and 22q. Deletions were observed less frequently and were mapped to 1q, 6q, 9q, 10q, 13q, 14q, 21q, and 22q. Gene expression analysis revealed that most of the altered genes were also perturbed in sporadic adult PTC; however, 141 gene expression changes were found to be unique to the post-Chernobyl tumors. The genes with the highest increases in expression that were novel to the pediatric post-Chernobyl tumors were TESC, PDZRN4, TRAa/TRDa, GABBR2, and CA12. The genes showing the largest expression decreases included PAPSS2, PDLIM3, BEXI, ANK2, SORBS2, and PPARGCIA. An overlay analysis of the gene expression and CNA profiles was then performed. This analysis identified genes showing both CNAs and concurrent gene expression alterations. Many of these are commonly seen in sporadic PTC such as SERPINA, COL8A, and PDX, while others were unique to the radiation-induced profiles including CAMK2N1, AK1, DHRS3, and PDE9A.

CONCLUSIONS

This type of analysis allows an assessment of gene expression changes that are associated with a physical mechanism. These genes and chromosomal regions are potential markers for radiation-induced PTC.

Increasing the number of thyroid lesions classes in microarray analysis improves the relevance of diagnostic markers

Background

Genetic markers for thyroid cancers identified by microarray analysis have offered limited predictive accuracy so far because of the few classes of thyroid lesions usually taken into account. To improve diagnostic relevance, we have simultaneously analyzed microarray data from six public datasets covering a total of 347 thyroid tissue samples representing 12 histological classes of follicular lesions and normal thyroid tissue. Our own dataset, containing about half the thyroid tissue samples, included all categories of thyroid lesions.

Methodology/Principal Findings

Classifier predictions were strongly affected by similarities between classes and by the number of classes in the training sets. In each dataset, sample prediction was improved by separating the samples into three groups according to class similarities. The cross-validation of differential genes revealed four clusters with functional enrichments. The analysis of six of these genes (APOD, APOE, CLGN, CRABP1, SDHA and TIMP1) in 49 new samples showed consistent gene and protein profiles with the class similarities observed. Focusing on four subclasses of follicular tumor, we explored the diagnostic potential of 12 selected markers (CASP10, CDH16, CLGN, CRABP1, HMGB2, ALPL2, ADAMTS2, CABIN1, ALDH1A3, USP13, NR2F2, KRTHB5) by real-time quantitative RT-PCR on 32 other new samples. The gene expression profiles of follicular tumors were examined with reference to the mutational status of the Pax8-PPARγ, TSHR, GNAS and NRAS genes.

Conclusion/Significance

We show that diagnostic tools defined on the basis of microarray data are more relevant when a large number of samples and tissue classes are used. Taking into account the relationships between the thyroid tumor pathologies, together with the main biological functions and pathways involved, improved the diagnostic accuracy of the samples. Our approach was particularly relevant for the classification of microfollicular adenomas.

Retinoic acid receptor and retinoid X receptor subtype expression for the differential diagnosis of thyroid neoplasms

BACKGROUND

Although differential expression of retinoic acid receptor (RAR) subtypes between benign and malignant thyroid tissues has been described, their diagnostic value has not been reported.

AIM

To investigate the diagnostic accuracy of RAR and retinoid X receptor (RXR) subtype protein expression for the differential diagnosis of thyroid neoplasms.

METHODS

We used a tissue array containing 93 benign thyroid tissues (normal thyroid, multinodular goiter, and follicular adenoma (FA)) and 77 thyroid carcinomas (papillary thyroid carcinoma (PTC), follicular thyroid carcinoma, and follicular variant of PTC (FVPTC)). Immunostaining was done for RAR and RXR subtypes. Staining was analyzed semiquantitatively based on receiver operating curve analyses and using hierarchical cluster analysis.

RESULTS

We found increased expression of cytoplasmic (c) RARA, cRARG, cRXRB and decreased expression of nuclear (n) RARB, nRARG, and nRXRA in thyroid carcinomas compared with benign tissues. We found three proteins differently expressed between FA and FTC and five proteins differentially expressed between FA and FVPTC, with high diagnostic accuracies. Using cluster analysis, the combination of negative staining of membranous RXRB and positive staining for cRXRB had a high positive predictive value (98%) for malignant thyroid disease, whereas the combination of positive nRXRA and negative cRXRB staining had a high predictive value (91%) for benign thyroid lesions.

CONCLUSION

We conclude that differences in RAR and RXR subtype protein expression may be valuable for the differential diagnosis of thyroid neoplasms. The results of this study and especially the value of cluster analysis have to be confirmed in subsequent studies.

Perspectives for improved and more accurate classification of thyroid epithelial tumors

CONTEXT

Histologic examination of thyroid nodules is the current standard to distinguish benign from malignant thyroid epithelial tumors and to classify histologic subtypes. This review analyzes the problems in histological differential diagnosis as well as contradictions between histology and molecular data and describes possibilities to combine histology with molecular data in an effort to more accurately classify thyroid epithelial tumors.

EVIDENCE ACQUISITION

Published literature, addressing the current recommendations for thyroid tumor classification, as well as literature on the application of histology and molecular studies on the etiology of thyroid tumors is analyzed.

EVIDENCE SYNTHESIS

The current histologic criteria to classify thyroid tumors, especially follicular-patterned tumors, are hampered by considerable interobserver variability. The detection of somatic mutations via genotyping and the definition of potentially informative gene expression signatures by microarray analyses, which can distinguish cancer subtypes as well as low- and high-risk cohorts, have recently demonstrated significant diagnostic potential. Moreover, in a routine diagnostic setting, micro-RNA profiling appears most promising due to their relative stability and the high accuracy of their expression profiles.

CONCLUSIONS

It is very likely that molecular definitions of thyroid tumors mentioned in the current World Health Organization classification will be further developed, leading to future progress in defining thyroid tumor types by an integrated histologic and molecular approach. These integrated classifications need to be evaluated for their specific impact on thyroid tumor diagnosis and prognosis.

"Hook"-calibration of GeneChip-microarrays: chip characteristics and expression measures

Background

Microarray experiments rely on several critical steps that may introduce biases and uncertainty in downstream analyses. These steps include mRNA sample extraction, amplification and labelling, hybridization, and scanning causing chip-specific systematic variations on the raw intensity level. Also the chosen array-type and the up-to-dateness of the genomic information probed on the chip affect the quality of the expression measures. In the accompanying publication we presented theory and algorithm of the so-called hook method which aims at correcting expression data for systematic biases using a series of new chip characteristics.

Results

In this publication we summarize the essential chip characteristics provided by this method, analyze special benchmark experiments to estimate transcript related expression measures and illustrate the potency of the method to detect and to quantify the quality of a particular hybridization. It is shown that our single-chip approach provides expression measures responding linearly on changes of the transcript concentration over three orders of magnitude. In addition, the method calculates a detection call judging the relation between the signal and the detection limit of the particular measurement. The performance of the method in the context of different chip generations and probe set assignments is illustrated. The hook method characterizes the RNA-quality in terms of the 3'/5'-amplification bias and the sample-specific calling rate. We show that the proper judgement of these effects requires the disentanglement of non-specific and specific hybridization which, otherwise, can lead to misinterpretations of expression changes. The consequences of modifying probe/target interactions by either changing the labelling protocol or by substituting RNA by DNA targets are demonstrated.

Conclusion

The single-chip based hook-method provides accurate expression estimates and chip-summary characteristics using the natural metrics given by the hybridization reaction with the potency to develop new standards for microarray quality control and calibration.

Transcriptome analysis of endocrine tumors: clinical perspectives

There is considerable interest in the application of DNA microarrays to the pathologic evaluation of endocrine neoplasms. Improvements in tumor classification and prognostication, prediction of response to therapy, and comprehensive assessment of tumoral hormone production represent the major anticipated benefits. Here, some of the microarray studies that support the clinical use of transcriptome profiling for endocrine tumors are reviewed. In addition, some of the barriers to clinical implementation are discussed.

Genome-wide studies in thyroid neoplasia

There is much interest in the application of genome biology to the field of thyroid neoplasia, despite the relatively low mortality rate associated with thyroid cancer in general. The principal reason for this interest is that the field of thyroid neoplasia stands to benefit from the application of genomic information to address a variety of pathologic and clinical issues. In addition to practical patient care issues, there is an excellent opportunity of expand the basic understanding of thyroid carcinogenesis. In this article, the most relevant genomic work on thyroid tumors performed to date is reviewed along with some general comments about the potential impact of genomic biology on thyroid pathology and the management of patients with thyroid nodules and cancer.

Utility of malignancy markers in fine-needle aspiration cytology of thyroid nodules: comparison of Hector Battifora mesothelial antigen-1, thyroid peroxidase and dipeptidyl aminopeptidase IV

The purpose of this study was to compare the diagnostic interest of Hector Battifora mesothelial antigen-1 (HBME-1), thyroid peroxidase (TPO), and dipeptidyl aminopeptidase IV (DPP4) in thyroid fine-needle aspirates obtained from 200 resected thyroid lesions (55 colloid nodules, 54 follicular adenomas, 59 papillary cancers, and 32 follicular carcinomas). Hector Battifora mesothelial antigen-1 or TPO expression (% positive cells) and DPP4 staining score (12-point scale) were evaluated. Receiver operating characteristic (ROC) curves were plotted and optimal cutoff values for diagnosing malignancy were determined. The TPO ROC curve was consistently higher than the HBME-1 ROC curve. The TPO curve was also higher than the DPP4 curve with regard to sensitivity, but dipped below the DPP4 curve with regard to specificity. Using a cutoff value of <80% positive cells for TPO, >10% positive cells for HBME-1, and staining score > or =1 for DPP4, sensitivity to specificity ratios were 98-83% for TPO, 90-60% for HBME-1, and 88-80% for DPP4. Two particularly interesting findings of this study were the low negative likelihood ratio of TPO (0.02) allowing highly reliable exclusion of malignancy and the 100% specificity of DPP4 staining scores=12. Due to poor performance on follicular lesions, HBME-1 showed no advantage over TPO or DPP4.

Microarray analysis refines classification of non-medullary thyroid tumours of uncertain malignancy

Conventional histology failed to classify part of non-medullary thyroid lesions as either benign or malignant. The group of tumours of uncertain malignancy (T-UM) concerns either atypical follicular adenomas or the recently called 'tumours of uncertain malignant potential'. To refine this classification we analysed microarray data from 93 follicular thyroid tumours: 10 T-UM, 3 follicular carcinomas, 13 papillary thyroid carcinomas and 67 follicular adenomas, compared to 73 control thyroid tissue samples. The diagnosis potential of 16 selected genes was validated by real-time quantitative RT-PCR on 6 additional T-UM. The gene expression profiles in several groups were examined with reference to the mutational status of the RET/PTC, BRAF and RAS genes. A pathological score (histological and immunohistochemical) was estimate for each of the T-UM involved in the study. The correlation between the T-UM gene profiles and the pathological score allowed a separation of the samples in two groups of benign or malignant tumours. Our analysis confirms the heterogeneity of T-UM and highlighted the molecular similarities between some cases and true carcinomas. We demonstrated the ability of few marker genes to serve as diagnosis tools and the need of a T-UM pathological scoring.

Insights from molecular pathways: potential pharmacologic targets of benign thyroid nodules

PURPOSE OF REVIEW

To describe molecular pathways that might be of relevance for a potential pharmacologic therapy of benign thyroid nodules.

RECENT FINDINGS

Constitutively activating thyrotropin receptor mutations have been found in about 60% of hot nodules. Its predominant role for signaling in hot nodules has been confirmed by in-vitro mutagenesis studies, thyrotropin receptor modeling and microarray studies. In contrast, the basic molecular cause of cold thyroid nodules is so far largely unknown. Defective sodium/iodide symporter trafficking, accumulation of T4-deficient, insufficiently iodinated thyroglobulin, increased oxidative stress and differential expression of several Gqalpha-protein kinase C pathway-associated genes have, however, recently been identified in cold thyroid nodules.

SUMMARY

As disturbed thyrotropin receptor signaling plays a central role in hot thyroid nodules, the identification of effective low-molecular-weight thyrotropin receptor ligands, such as thyrotropin receptor agonists, inverse agonists and antagonists has a pharmacologic potential in the diagnosis and treatment of thyroid cancer, Graves' disease and hot thyroid nodules, respectively. Further studies have to clarify the pharmacologic potential of the enhancement of antioxidative mechanisms and the inhibition of Gqalpha-protein kinase C signaling in cold thyroid nodules.

A multi-gene approach to differentiate papillary thyroid carcinoma from benign lesions: gene selection using support vector machines with bootstrapping

Selection of novel molecular markers is an important goal of cancer genomics studies. The aim of our analysis was to apply the multivariate bioinformatical tools to rank the genes - potential markers of papillary thyroid cancer (PTC) according to their diagnostic usefulness. We also assessed the accuracy of benign/malignant classification, based on gene expression profiling, for PTC. We analyzed a 180-array dataset (90 HG-U95A and 90 HG-U133A oligonucleotide arrays), which included a collection of 57 PTCs, 61 benign thyroid tumors, and 62 apparently normal tissues. Gene selection was carried out by the support vector machines method with bootstrapping, which allowed us 1) ranking the genes that were most important for classification quality and appeared most frequently in the classifiers (bootstrap-based feature ranking, BBFR); 2) ranking the samples, and thus detecting cases that were most difficult to classify (bootstrap-based outlier detection). The accuracy of PTC diagnosis was 98.5% for a 20-gene classifier, its 95% confidence interval (CI) was 95.9-100%, with the lower limit of CI exceeding 95% already for five genes. Only 5 of 180 samples (2.8%) were misclassified in more than 10% of bootstrap iterations. We specified 43 genes which are most suitable as molecular markers of PTC, among them some well-known PTC markers (MET, fibronectin 1, dipeptidylpeptidase 4, or adenosine A1 receptor) and potential new ones (UDP-galactose-4-epimerase, cadherin 16, gap junction protein 3, sushi, nidogen, and EGF-like domains 1, inhibitor of DNA binding 3, RUNX1, leiomodin 1, F-box protein 9, and tripartite motif-containing 58). The highest ranking gene, metallophosphoesterase domain-containing protein 2, achieved 96.7% of the maximum BBFR score.

Perspectives and limitations of microarray-based gene expression profiling of thyroid tumors

Microarray technology has become a powerful tool to analyze the gene expression of tens of thousands of genes simultaneously. Microarray-based gene expression profiles are available for malignant thyroid tumors (i.e., follicular thyroid carcinoma, and papillary thyroid carcinoma), and for benign thyroid tumors (such as autonomously functioning thyroid nodules and cold thyroid nodules). In general, the two main foci of microarray investigations are improved understanding of the pathophysiology/molecular etiology of thyroid neoplasia and the detection of genetic markers that could improve the differential diagnosis of thyroid tumors. Their results revealed new features, not known from one-gene studies. Simultaneously, the increasing number of microarray analyses of different thyroid pathologies raises the demand to efficiently compare the data. However, the use of different microarray platforms complicates cross-analysis. In addition, there are other important differences between these studies: 1) some studies use intraindividual comparisons, whereas other studies perform interindividual comparisons; 2) the reference tissue is defined as strictly nonnodular healthy tissue or also contains benign lesions such as goiter, follicular adenoma, and hyperplastic nodules in some studies; and 3) the widely used Affymetrix GeneChip platform comprises several GeneChip generations that are only partially compatible. Moreover, the different studies are characterized by strong differences in data analysis methods, which vary from simple empiric filters to sophisticated statistic algorithms. Therefore, this review summarizes and compares the different published reports in the context of their study design. It also illustrates perspectives and solutions for data set integration and meta-analysis, as well as the possibilities to combine array analysis with other genetic approaches.

Molecular determination of benign and malignant thyroid tumors

Recent molecular studies have revolutionized our understanding of the pathogenesis of thyroid tumors and particular advances have been made in three areas. First, toxic thyroid nodules, which originate from constitutive activation of thyroid-stimulating hormone receptor/Gs α signaling and represent the dominant cause of thyrotoxicosis in regions with iodine deficiency. Second, papillary thyroid cancer, the most frequent thyroid malignancy, which is characterized by a common fingerprint of constitutive mitogen-activated protein kinase activation. Importantly, this is caused by distinct genetic alterations in radiation-induced (RET/PTC, NTRK and AKAP9/BRAF rearrangements) and sporadic tumors (BRAF and RAS point mutation) and, recently, there exciting in vitro have emerged explaining the structural basis for this. These findings suggest a scenario in which the fate of a thyroid tumor is determined by the specific genetic defect at the beginning. Third, application of microarray analysis in nodular pathologies in which the oncogenic pathway is less clear, notably follicular neoplasia, has led to the identification of a number of promising genetic markers (TFF-3, Gal-3, PLAB, CCND2 and PCKD2) for the diagnostic distinction of follicular adenoma and carcinoma. In addition to the diagnostic perspective, the identification of molecular fingerprints of thyroid tumors opens novel avenues for an improved therapeutic approach; for example, selective antagonism of cell signaling in treatment-refractory thyroid cancer.

Cellular transcriptomics -- the next phase of endocrine expression profiling

Transcriptome analysis, or global gene expression profiling, has become a commonly used and valuable tool in both basic and clinical endocrine research. Novel endocrine regulators have 'surfaced' and greater consideration is now given to understanding function at the level of gene networks. Recent developments have shown that the transcriptome is considerably larger and more divergently expressed than was previously thought. Endocrine cells express a great variety of coding and noncoding RNAs in a highly cell-specific manner. If further value is to be taken from this research area, then steps towards defined cellular transcriptomics must be taken. New sampling techniques that utilize novel genetic models are a key first step.