Automated density-based counting of FISH amplification signals for HER2 status assessment.

BACKGROUND: Automated image analysis can make quantification of FISH signals in histological sections more efficient and reproducible. Current detection-based methods, however, often fail to accurately quantify densely clustered FISH signals. METHODS: We propose a novel density-based approach to quantifying FISH signals. Instead of detecting individual signals, this approach quantifies FISH signals in terms of the integral over a density map predicted by Deep Learning. We apply the density-based approach to the task of counting and determining ratios of ERBB2 and CEN17 signals and compare it to common detection-based and area-based approaches. RESULTS: The ratios determined by our approach were strongly correlated with results obtained by manual annotation of individual FISH signals (Pearson's r = 0.907). In addition, they were highly consistent with cutoff-scores determined by a pathologist (balanced concordance = 0.971). The density-based approach generally outperformed the other approaches. Its superiority was particularly evident in the presence of dense signal clusters. CONCLUSIONS: The presented approach enables accurate and efficient automated quantification of FISH signals. Since signals in clusters can hardly be detected individually even by human observers, the density-based quantification performs better than detection-based approaches.

Impact of molecular screening for point mutations and rearrangements in routine air-dried fine-needle aspiration samples of thyroid nodules.

BACKGROUND: The diagnostic limitations of thyroid fine-needle aspiration (FNA), such as the indeterminate category, can be partially overcome by molecular analyses. However, until now, rearrangements have only been detected in fresh FNA material and the number of follicular thyroid carcinomas (FTCs) was rather low in previous studies. We aimed at investigating the impact of point mutations and rearrangement detection in a set of routine air-dried FNA smears with a higher percentage of FTCs. METHODS: RNA and DNA was extracted from 310 FNAs (164 indeterminate, 57 malignant, 89 benign) and corresponding formalin-fixed paraffin-embedded tissue (156 follicular adenomas [FAs], 32 FTCs, 44 papillary thyroid carcinomas [PTCs], 9 follicular variant PTCs, and 69 goiters). PAX8/PPARG and RET/PTC rearrangements were detected by qPCR, BRAF and RAS mutations by high-resolution melting PCR and by pyrosequencing. RESULTS: Forty-seven mutations were detected in the FNAs: 22 BRAF, 13 NRAS, and 3 HRAS mutations, 8 PAX8/PPARG, and one RET/PTC-rearrangement. While the presence of a BRAF and RET/PTC mutation was associated with cancer in 100% of samples each, the presence of a RAS and PAX8/PPARG mutation was associated with cancer in only 12% and 50% of samples, respectively. In the indeterminate group 4 of 25 carcinomas were identified by molecular FNA screening, which increased the sensitivity from 67% (cytology alone) to 75% (cytology plus molecular screening). CONCLUSION: Molecular screening for point mutations and rearrangements is feasible in air-dried FNAs. Although the impact of detecting point mutations and rearrangements in FNAs has most likely been overestimated in previous studies, molecular FNA analyses improve presurgical diagnostics. The detection of BRAF mutations in FNA may improve the choice of surgery and postsurgical treatment. Further data are necessary to elucidate the true impact of detecting RAS and PAX8/PPARG mutations in FNAs. The inclusion of additional rare somatic mutations and miRNA markers might further improve the impact of molecular FNA diagnostics.

HMGA2 expression in the PC-3 prostate cancer cell line is autonomous of growth factor stimulation.

BACKGROUND: High-mobility group AT-hook 2 (HMGA2) protein acts as an oncofoetal transcriptional regulator. In mesenchymal tissues, its expression can be induced by a variety of growth factors such as fibroblast growth factor-1 (FGF1) and platelet-derived growth factor-BB (PDGF-BB) as well as by foetal bovine serum (FBS), thus enhancing proliferation. MATERIALS AND METHODS: To examine these effects in epithelial malignancies, we used the PC-3 prostate cancer cell line for assaying proliferation and HMGA2 expression in response to incubation with growth factors and FBS. The HMGA2 locus was investigated by fluorescence in situ hybridisation (FISH) for loss, amplification or re-arrangement. RESULTS: PC-3 is a cell line that moderately overexpresses HMGA2. None of the growth factors nor FBS caused significantly increased expression of HMGA2. In contrast, a significantly augmented proliferation rate was observed when applying FGF1 or PDGF-BB for 12 h. CONCLUSION: HMGA2 is expressed independently of external stimuli, whereas proliferation stimulated by growth factors is independent of further elevated HMGA2 expression.

Activation of the two microRNA clusters C19MC and miR-371-3 does not play prominent role in thyroid cancer.

Chromosomal rearrangements of band 19q13.4 are frequent cytogenetic alterations in benign thyroid adenomas. Apparently, these alterations lead to the upregulation of genes encoding microRNAs of two clusters mapping to the breakpoint region, i.e. miR-371-3 and C19MC. Since members of both clusters have been associated with neoplastic growth in other tumor entities the question arises whether or not their upregulation predisposes to malignant transformation of follicular cells of the thyroid. To address this question we have quantified the expression of miR-372 and miR-520c-3p in samples of 114 thyroid cancers including eight anaplastic thyroid carcinomas, 25 follicular thyroid carcinomas, 78 papillary thyroid carcinomas (including 13 follicular variants thereof), two medullary thyroid carcinomas and one oncocytic thyroid carcinoma. Additionally, we quantified miR-371a-3p and miR-519a-3p in selected samples. While in neither of the cases miR-520c-3p and miR-519a-3p were found to be upregulated, one papillary and one anaplastic thyroid carcinoma, respectively, showed upregulation of miR-372 and miR-371a-3p. However, in these cases fluorescence in situ hybridization did not reveal rearrangements of the common breakpoint region as affected in adenomas. Thus, these rearrangements do apparently not play a major role as first steps in malignant transformation of the thyroid epithelium. Moreover, there is no evidence that 19q13.4 rearrangements characterize a subgroup of thyroid adenomas associated with a higher risk to undergo malignant transformation. Vice versa, the mechanisms by which 19q13.4 rearrangements contribute to benign tumorigenesis in the thyroid remain to be elucidated.

Monosomy and ring chromosome 13 in a thyroid nodular goiter-do we underestimate its relevance in benign thyroid lesions?

Classical cytogenetic examination of a thyroid nodular goiter revealed the existence of two different cytogenetically aberrant cell clones. They were characterized by monosomy 13 as the sole abnormality in one clone, and loss of one chromosome 13 and a ring chromosome that was found to consist of chromosome 13 material by fluorescence in situ hybridization in the other clone. We have concluded that during the course of karyotypic evolution, the instability of the ring chromosome has led to its loss and subsequent monosomy 13. In the literature, two cases of partial monosomy 13 have been reported in adenomatous goiters, suggesting that this abnormality characterizes a rare but distinct subgroup of benign thyroid lesions histologically presenting as adenomatous goiters. Possible target genes of these deletions are the retinoblastoma (RB1) gene locus and the MIR16-1/15A cluster. Based on similar changes in other tumors, it seems reasonable to also analyze a large number of adenomatous goiters for submicroscopic deletions of the long arm of chromosome 13.

MED12 mutations in uterine fibroids--their relationship to cytogenetic subgroups.

Recurrent chromosomal alterations are found in roughly 20% of all uterine fibroids but in the majority cytogenetic changes are lacking. Recently, mutations of the gene mediator subcomplex 12 (MED12) have been detected in a majority of fibroids but no information is available whether or not they co-occur with cytogenetic subtypes as, e.g., rearrangements of the genes encoding high mobility group AT-hook (HMGA) proteins. In a total of 80 cytogenetically characterized fibroids from 50 patients, we were not only able to confirm the frequent occurrence of MED12 mutations but also to stratify two mutually exclusive pathways of leiomyomagenesis with either rearrangements of HMGA2 reflected by clonal chromosome abnormalities affecting 12q14~15 or by mutations affecting exon 2 of MED12. On average the latter mutations were associated with a significantly smaller tumor size. However, G>A transitions of nucleotides c.130 or c.131 correlate with a significantly larger size of the fibroids compared to other MED12 mutations thus explaining the high prevalence of the former mutations among clinically detectable fibroids. Interestingly, fibroids with MED12 mutations expressed significantly higher levels of the gene encoding wingless-type MMTV integration site family, member 4 (WNT4). Based on these findings and data from the literature, we hypothesize that estrogen and the mutated MED12 cooperate in activating the Wnt pathway which in turn activates ß-catenin known to cause leiomyoma-like lesions in a mouse model. The occurrence of a "fibroid-type mutation" in a rare histologic subtype of endometrial polyps suggests that this mechanism is not confined to uterine leiomyomas.

Detection of PAX8-PPARG fusion transcripts in archival thyroid carcinoma samples by conventional RT-PCR.

The t(2;3)(q13;p25) occurs in a subgroup of follicular-patterned thyroid tumors and leads to a fusion of the genes encoding for the thyroid-specific transcription factor paired box 8 (PAX8) and the peroxisome proliferator-activated receptor gamma (PPARγ). Although initially discovered in follicular carcinomas (FTC), the fusion transcripts were also detected in a small fraction of follicular adenomas and rarely in follicular variants of papillary carcinomas (FV-PTC). In most RT-PCR based studies, fresh or snap-frozen tissue samples were used. The aim of the present study was to develop a method for the detection of chimeric PAX8-PPARG transcripts in formalin-fixed paraffin-embedded (FFPE) thyroid tumor samples by conventional RT-PCR. For this purpose, RNA from FFPE samples of 21 FTC, seven FV-PTC, and one bone metastasis derived from an FTC was subjected to RT-PCR with subsequent gel electrophoretic separation of the products. Fusion transcripts were detected in 2/21 primary FTC (9.5%) and in the bone metastasis, but they were undetectable in all seven FV-PTC under investigation. The RT-PCR approach described herein allows to detect all known variants of PAX8-PPARG fusion transcripts and is applicable to FFPE tissues. Thus, it can be used to screen archival thyroid tumor samples for the gene fusion.

Decrease in thyroid adenoma associated (THADA) expression is a marker of dedifferentiation of thyroid tissue.

BACKGROUND: Thyroid adenoma associated (THADA) has been identified as the target gene affected by chromosome 2p21 translocations in thyroid adenomas, but the role of THADA in the thyroid is still elusive. The aim of this study was to quantify THADA gene expression in normal tissues and in thyroid hyper- and neoplasias, using real-time PCR. METHODS: For the analysis THADA and 18S rRNA gene expression assays were performed on 34 normal tissue samples, including thyroid, salivary gland, heart, endometrium, myometrium, lung, blood, and adipose tissue as well as on 85 thyroid hyper- and neoplasias, including three adenomas with a 2p21 translocation. In addition, NIS (sodium-iodide symporter) gene expression was measured on 34 of the pathological thyroid samples. RESULTS: Results illustrated that THADA expression in normal thyroid tissue was significantly higher (p < 0.0001, exact Wilcoxon test) than in the other tissues. Significant differences were also found between non-malignant pathological thyroid samples (goiters and adenomas) and malignant tumors (p < 0.001, Wilcoxon test, t approximation), anaplastic carcinomas (ATCs) and all other samples and also between ATCs and all other malignant tumors (p < 0.05, Wilcoxon test, t approximation). Furthermore, in thyroid tumors THADA mRNA expression was found to be inversely correlated with HMGA2 mRNA. HMGA2 expression was recently identified as a marker revealing malignant transformation of thyroid follicular tumors. A correlation between THADA and NIS has also been found in thyroid normal tissue and malignant tumors. CONCLUSIONS: The results suggest THADA being a marker of dedifferentiation of thyroid tissue.

Interphase fluorescence in situ hybridization analysis detects a much higher rate of thyroid tumors with clonal cytogenetic deviations of the main cytogenetic subgroups than conventional cytogenetics.

In benign thyroid lesions, three main cytogenetic subgroups, characterized by trisomy 7 or structural aberrations involving either chromosomal region 19q13.4 or 2p21, can be distinguished by conventional cytogenetics (CC). As a rule, these aberrations seem to be mutually exclusive. Interphase fluorescence in situ hybridization (I-FISH) analysis on benign as well as malignant thyroid neoplasias has been performed in the past, but rarely in combination with CC. In the present paper, we have analyzed 161 benign thyroid lesions both with CC and I-FISH on touch preparations by using a multi-target, triple-color FISH assay as well as dual-color break-apart probes for detection of the main cytogenetic subgroups. Within the samples, I-FISH detected tumors belonging to either of the subgroups more frequently than CC (23 vs. 11.4%), either due to small subpopulations of aberrant cells or to cryptic chromosomal rearrangements (three cases). Thus, I-FISH seems to be more sensitive than CC, particularly in the detection of subpopulations of cells harboring cytogenetic aberrations that may be overlooked by CC. In summary, I-FISH on touch preparations of benign thyroid lesions seems to be a favorable method for cytogenetic subtyping of thyroid lesions.

On the prevalence of the PAX8-PPARG fusion resulting from the chromosomal translocation t(2;3)(q13;p25) in adenomas of the thyroid.

The chromosomal translocation t(2;3)(q13;p25) characterizes a subgroup of tumors originating from the thyroid follicular epithelium and was initially discovered in a few cases of adenomas. Later, a fusion of the genes PAX8 and PPARG resulting from this translocation was frequently observed in follicular carcinomas and considered as a marker of follicular thyroid cancer. According to subsequent studies, however, this rearrangement is not confined to carcinomas but also occurs in adenomas, with considerably varying frequencies. Only five cases of thyroid adenomas with this translocation detected by conventional cytogenetics have been documented. In contrast, studies using reverse-transcription polymerase chain reaction (RT-PCR) detected fusion transcripts resulting from that translocation in an average of 8.2% of adenomas. The aim of this study was to determine the frequency of the PAX8-PPARG fusion in follicular adenomas and to use the HMGA2 mRNA level of such tumors as an indicator of malignancy. In cytogenetic studies of 192 follicular adenomas, the t(2;3)(q13;p25) has been identified in only two cases described herein. Histopathology revealed no evidence of malignancy in either case, and, concordantly, HMGA2 mRNA levels were not elevated. In summary, the fusion is a rare event in follicular adenomas and its prevalence may be overestimated in many RT-PCR-based studies.

6p21 rearrangements in uterine leiomyomas targeting HMGA1.

To quantify the expression of HMGA1 mRNA in uterine leiomyomas, the expression of HMGA1 was analyzed in a series including tumors with aberrations of chromosome 6 (n = 7) and cytogenetically normal tumors (n = 8) as a control group by quantitative reverse transcriptase-polymerase chain reaction. The average expression level in the 6p21 group was found to be 5.6 times higher than that in the control group, and with one exception, all cases with 6p21 alteration revealed a high expression of HMGA1 mRNA than cytogenetically normal tumors. Nevertheless, compared to fibroids with a normal karyotype, the upregulation of the HMGA1 mRNA in these cases was much less strong than that of HMGA2 mRNA in case of 12q14∼15 aberrations identified in previous studies.

Cell culture and senescence in uterine fibroids.

The in vitro growth of cells from uterine fibroids is characterized by an early onset of senescence. Often, an even lower growth potential than that of matching myometrial cells is noted. Also, the tremendous differences in the expression of the high mobility group protein HMGA2 seen when comparing fibroids of different genetic subtypes are surprisingly not reflected by significant differences in their growth potential in vitro. We aimed to evaluate possible changes of the HMGA2 expression level between the native tissue and cell cultures, so we performed quantitative real-time polymerase chain reaction studies that revealed a marked decrease of the HMGA2 mRNA in culture in those cases with overexpression of HMGA2. In the two cases initially showing the highest expression, it decreased by approximately 97%. Associated with the decrease of HMGA2 was a clearly increased expression of the senescence-associated p19(Arf). Together, these findings explain the similar behavior of cell cultures from fibroids of different genetic subgroups and may also offer an explanation for the early onset of in vitro senescence in these cell cultures.

The two stem cell microRNA gene clusters C19MC and miR-371-3 are activated by specific chromosomal rearrangements in a subgroup of thyroid adenomas.

Thyroid adenomas are common benign human tumors with a high prevalence of about 5% of the adult population even in iodine sufficient areas. Rearrangements of chromosomal band 19q13.4 represent a frequent clonal cytogenetic deviation in these tumors making them the most frequent non-random chromosomal translocations in human epithelial tumors at all. Two microRNA (miRNA) gene clusters i.e. C19MC and miR-371-3 are located in close proximity to the breakpoint region of these chromosomal rearrangements and have been checked for a possible up-regulation due to the genomic alteration. In 4/5 cell lines established from thyroid adenomas with 19q13.4 rearrangements and 5/5 primary adenomas with that type of rearrangement both the C19MC and miR-371-3 cluster were found to be significantly overexpressed compared to controls lacking that particular chromosome abnormality. In the remaining cell line qRT-PCR revealed overexpression of members of the miR-371-3 cluster only which might be due to a deletion accompanying the chromosomal rearrangement in that case. In depth molecular characterization of the breakpoint in a cell line from one adenoma of this type reveals the existence of large Pol-II mRNA fragments as the most likely source of up-regulation of the C19MC cluster. The up-regulation of the clusters is likely to be causally associated with the pathogenesis of the corresponding tumors. Of note, the expression of miRNAs miR-520c and miR-373 is known to characterize stem cells and in terms of molecular oncology has been implicated in invasive growth of epithelial cells in vitro and in vivo thus allowing to delineate a distinct molecular subtype of thyroid adenomas. Besides thyroid adenomas rearrangements of 19q13.4 are frequently found in other human neoplasias as well, suggesting that activation of both clusters might be a more general phenomenon in human neoplasias.

Overexpression of HMGA2 in uterine leiomyomas points to its general role for the pathogenesis of the disease.

An overexpression of HMGA2 is supposed to be a key event in the genesis of leiomyoma with chromosomal rearrangements affecting the region 12q14-15 targeting the HMGA2 gene, but gene expression data regarding differences between uterine leiomyomas with and those without 12q14-15 aberrations are insufficient. To address the question whether HMGA2 is only upregulated in the 12q14-15 subgroup, the expression of HMGA2 was analyzed in a comprehensive set of leiomyomas (n = 180) including tumors with 12q14-15 chromosomal aberrations (n = 13) and matching myometrial tissues (n = 51) by quantitative RT-PCR. The highest expression levels for HMGA2 were observed in tumors with rearrangements affecting the region 12q14-15, but although HMGA2 is expressed at lower levels in leiomyomas without such aberrations, the comparison between the expression in myomas and matching myometrial tissues indicates a general upregulation of HMGA2 regardless of the presence or absence of such chromosomal abnormalities. The significant (P < 0.05) overexpression of HMGA2 also in the group of fibroids without chromosomal aberrations of the 12q14-15 region suggests a general role of HMGA2 in the development of the disease.

Chromosomal assignment of canine THADA gene to CFA 10q25.

BACKGROUND: Chromosomal translocations affecting the chromosome 2p21 cluster in a 450 kb breakpoint region are frequently observed in human benign thyroid adenomas. THADA (thyroid adenoma associated) was identified as the affected gene within this breakpoint region. In contrast to man tumours of the thyroid gland of dogs (Canis lupus familiaris) constitute mainly as follicular cell carcinomas, with malignant thyroid tumours being more frequent than benign thyroid adenomas. In order to elucidate if the THADA gene is also a target of chromosomal rearrangements in thyroid adenomas of the dog we have physically mapped the canine THADA gene to canine chromosome 10.A PCR was established to screen a canine genome library for a BAC clone containing the gene sequence of canine THADA. Further PCR reactions were done using the identified BAC clone as a template in order to verify the corresponding PCR product by sequencing.Canine whole blood was incubated with colcemid in order to arrest the cultured cells in metaphases. The verified BAC DNA was digoxigenin labeled and used as a probe in fluorescence in situ hybridization (FISH). Ten well spread metaphases were examined indicating a signal on canine chromosome 10 on both chromatids. A detailed fine mapping was performed indicating the canine THADA gene locus on the q-arm of chromosome 10. RESULTS: The canine THADA gene locus was mapped on chromosome 10q25. Our mapping results obtained in this study following the previously described nomenclature for the canine karyotype. CONCLUSION: We analysed whether the THADA gene locus is a hotspot of canine chromosomal rearrangements in canine neoplastic lesions of the thyroid and in addition might play a role as a candidate gene for a possible malignant transformation of canine thyroid adenomas. Although the available cytogenetic data of canine thyroid adenomas are still insufficient the chromosomal region to which the canine THADA has been mapped seems to be no hotspot of chromosomal aberrations seen in canine thyroid adenomas.

A domain of the thyroid adenoma associated gene (THADA) conserved in vertebrates becomes destroyed by chromosomal rearrangements observed in thyroid adenomas.

THADA, mapping to chromosomal band 2p21 is target gene of specific chromosomal rearrangements observed in thyroid benign tumors. Thus, it is one of the most common gene targets in chromosomal rearrangements in benign epithelial tumors. Nevertheless, nothing is known about the function of its protein. Therefore, we have analyzed the genetic structure of THADA homologous genes in selected vertebrates (Canis familiaris, Chlorocebus aethiops, Gallus gallus, and Mus musculus), which are not characterized up to now. The coding sequences of the mRNA of these species have been sequenced and analyzed revealing similarities to ARM repeat structures which indicates an involvement in protein-protein interactions. Using multiple alignments we identified the most conserved part of the protein (aa 1033-1415 Homo sapiens) with an identity of 70.5% between the most different organisms implying a putative important functional domain. The truncations observed in human thyroid adenomas disrupt this conserved domain of the protein indicating a loss of function of THADA contributing to the development of the follicular neoplasias of the thyroid.

Identification of a gene rearranged by 2p21 aberrations in thyroid adenomas.

Thyroid adenomas belong to the cytogenetically best investigated human epithelial tumors. Cytogenetic studies of about 450 benign lesions allow one to distinguish between different cytogenetic subgroups. Two chromosomal regions, that is, 19q13 and 2p21, are frequently rearranged in these tumors. Although 2p21 aberrations only account for about 10% of the benign thyroid tumors with clonal cytogenetic deviations, 2p21 rearrangements belong to the most common cytogenetic rearrangements in epithelial tumors due to the high frequency of these benign lesions. The 2p21 breakpoint region recently has been delineated to a region of 450 kbp, but the gene affected by the cytogenetic rearrangements still has escaped detection. Positional cloning and 3' RACE-PCR allowed us to clone that gene which we will refer to as thyroid adenoma associated (THADA) gene. In cells from two thyroid adenomas characterized by translocations t(2;20;3) (p21;q11.2;p25) and t(2;7)(p21;p15), respectively, we performed 3'-RACE-PCRs and found two fusions of THADA with a sequence derived from chromosome band 3p25 or with a sequence derived from chromosome band 7p15. The THADA gene spans roughly 365 kbp and, based on preliminary results, encodes a death receptor-interacting protein.

Molecular-cytogenetic analysis of fragmentation of chromosome 17 in the breast cancer cell line EFM-19.

Because a previous study by conventional cytogenetics had revealed a nullisomy 17 in the breast cancer cell line EFM-19, we analysed that cell line by SKY-FISH and by FISH using different probes derived from chromosome 17. A bicolor FISH using a HER2-specific probe and a chromosome 17 centromeric probe showed five HER2 and six centromeric signals all appearing on different chromosomes A further bicolor FISH using a chromosome 17-specific painting probe and a HER2-specific probe revealed that the HER2 signals were always localized within chromosome 17 segments constituting part of structurally altered chromosomes as deduced from their G-banding. Further FISH analyses using single-locus probes of chromosome 17, i.e., for MDS, p53, SMS and RARA, showed that all five chromosome 17 painting segments contained material from the long arm but only two painting segments had additional material from the short arm. A SKY-FISH confirmed the results of the chromosome 17 painting by FISH, except for one structurally altered chromosome showing additional chromosome 17 material detected by the SKY experiment. These results allow us to conclude that, in this cell line, polysomy 17 has preceeded the fragmentation of chromosome 17 leading to amplification of small parts of that chromosome as well as to extended losses. As to a general mechanism, polysomy 17 and a fragility of this, chromosome in breast cancer cells may not only account for part of the cases with HER2 amplification but, at the same time, may further support malignant progression due to the loss of tumor suppressor genes as e.g. p53.