FISH molecular testing in cytological preparations from solid tumors

Many of the exciting new developments in solid tumor molecular cytogenetics impact classical and molecular pathology. Fluorescence in situ hybridization to identify specific DNA target sequences in nuclei of non-dividing cells in solid neoplasms has contributed to the integration of molecular cytogenetics into cytology in spite of the remarkable promiscuity of cancer genes. Indeed, although it is a low-throughput assay, fluorescence in situ hybridization enables the direct disclosure and localization of genetic markers in single nuclei. Gene fusions are among the most prominent genetic alterations in cancer, providing markers that may be determinant in needle biopsies that are negative or suspicious for malignancy, and may contribute to the correct classification of the tumors. In view of the expanding use of fluorescence in situ hybridization in cytology, future challenges include automated sample evaluation and the specification of common criteria for interpreting and reporting results.


Background
Various types of genetic alterations, as well as epigenetic phenomena, have been identified and are now considered important in the classification, prognosis, and treatment of cancer. Correlations between genomic instability and carcinogenesis have been extensively investigated, leading to the recognition of an increasing number of genetic abnormalities as a tumor driving force. Currently, several molecular approaches are available to investigate tumor cell pathobiology at different levels (chromosome, gene, gene expression). The predominant approaches include immunohistochemistry, fluorescence in situ hybridization (FISH), polymerase chain reaction, arraybased and omics-based techniques [1][2][3][4][5]. The integration of results obtained using these platforms has been invaluable in clarifying genetic alterations associated with cancer and in interpreting the key role of the impaired signaling pathways. Gene gains and losses and gene disruptions by chromosome translocation, inversion, or deletion have been recognized as playing a pathogenetic role in many cancers. These exciting new developments in solid tumor molecular cytogenetics impact classical and molecular pathology, and an increasing number of chromosome markers have been integrated into World Health Organization tumor classifications [6]. Some of these markers are also relevant to selection of therapies targeting the protein products of gene fusions. In this scenario the impact of testing gene alterations by interphase FISH in material from needle biopsies and organic fluids has rapidly increased.
Promiscuity: a false dilemma?
The identification of a specific translocation in solid tumors dates back to 1983 when the t(11;22)(q24;q12) in Ewing's sarcoma was first described [7]. It took nine years before the underlying gene fusion, EWS/FLI1 [8] (today named EWSR1/FLI1), was discovered. The EWS/ FLI1 fusion was found to be closely associated with this type of sarcoma, and was thought to play a causal role in initiating the neoplastic process. Subsequent observations of variant translocations and the resulting EWSR1 fusions with different partner genes in the same tumor entity disclosed the tip of an iceberg, paving the way for discovering the phenomenon of gene promiscuity in cancer. Indeed, the molecular cytogenetics of Ewing's sarcoma family tumors (so called ESFT) and subsequently of other histologically unrelated soft tissue tumors, and finally of tumors arising in tissues of distinct embryological origin, demonstrated the ubiquitous involvement of the

POU5F1
Skin hydroadenoma CREB3L1 Sclerosing epitheliod fibrosarcoma Epithelial lineage; Hematological lineage; Mesenchymal lineage; Mesenchymal and Epithelial lineage; ca. = carcinoma; sa. = sarcoma; tu. = tumor Figure 1 Promiscuity of the EWSR1 gene in malignancy.  EWSR1 gene in a wide spectrum of cancers, from sarcoma to carcinoma and to hematological malignancy (Figure 1). It is now clear that most structural gene alterations mediated by chromosome rearrangements (examples in Figures 2 and 3) [9,10], drive malignancy in a variety of tumors of different hystogenetic types. Nevertheless, this evidence does not invalidate the role of gene fusions or deregulated genes as diagnostic tools. Indeed, the accumulated data prompted expansion of the understanding of gene promiscuity. For example, it is now clear that the EWSR1 gene fuses with several genes mainly encoding transcriptional regulator factor families, resulting in deregulation of specific molecular pathways. These include ETS, homeobox-genes, zinc finger, and leucine-zipper transcription factor families. Disruption of these pathways may influence the pathogenesis of specific tumor types through a variety of activation mechanisms [11]. In spite of promiscuity, searching for gene involved in chromosome alterations leading to illicit shuffling of coding or regulatory sequences in cancer is becoming an invaluable approach in cytological investigations as well.

Cytology and cytogenetics
Based on the above scenario, the relationship between cytology and cytogenetics has become increasingly close, and the use of cytogenetics as an ancillary supplementary tool in cytological diagnosis has been introduced in the pathology sector. In particular, the leitmotif of the union of cytology and cytogenetics [12] is the need for a close collaboration between the two parts, since on one hand asking for a FISH test implies being aware of the rearrangement to be investigated (Figure 4), and on the other the cytogenetic result needs to be interpreted in the context of the cytological (and possibly clinical) observations. In addition, a FISH test is often used as a confirmatory tool since a negative result is not informative, both because unknown alterations cannot be excluded and the availability of tumor cells in cytological preparations may be limited.
Considering the introduction of systematic genomic testing for some tumors (such as lung and breast cancer) [13,14], the consequent need for a correct evaluation of ratio value in the presence of genetic heterogeneity [15], and the growing demand for FISH tests in fine needle aspirations and organic fluids, two main challenges for the future can be foreseen: the implementation of automated FISH evaluation and the specification of common criteria for interpreting and reporting FISH results in as many tumor types as possible. A significant impediment to evaluating the ever increasing numbers of clinical FISH tests requested is imposed by the labor intensive nature of the assay, as each test requires scoring numerous interphase nuclei by double blind observation. Automated FISH, with strictly established parameters for standardization, could partly overcome these issues, although automation has yet to be perfected [16]. Specific recommendations and guidelines for FISH on tumors have been established within ACMG (American College of Medical Genetics and Genomics) [17] and E.C.A (European Cytogeneticists Association) [  the other hand, common objective interpretation criteria for FISH on cytological preparations, as well as quality control and quality assurance policies, remain limited [13,14], and require an extraordinary cooperative effort and interaction between cytogeneticists and cytologists. It would be desirable to convene expert advisory panels from scientific societies of clinical cytogeneticists and pathologists to establish evaluation criteria for the various tumors, based on expertise and a review of published literature, with a view to establishing common shared recommendations.

Conclusions
Many of the exciting new developments of molecular cytogenetics are having a profound impact on classical and molecular cytology. The growing demand for cytological FISH tests highlights the need for the specification of common criteria for interpreting and reporting FISH results, for quality control and quality assurance policies, and for possible implementation of automated FISH evaluation. Figure 4 Example of FISH in a cytological preparation. A cytological preparation from thyroid fine needle aspiration was simultaneously hybridized with RET (labeled with Spectrum Aqua/ Spectrum Red) and PPARg (labeled with Spectrum Green/Spectrum Gold respectively) split-apart probes. Broken RET is revealed by the split-apart centromeric aqua and telomeric red (arrowheads) probes. Contiguous dual-color signals indicate intact genes. Nuclei are counterstained with DAPI.