In the absence of definite signs of benign lesion, a biopsy should always be performed. Consultation with the radiologist, the surgeon, and the pathologist allows them to define the biopsy tract using compartmental anatomy definitions [62, 63], the biopsy site (especially the most suspicious portion), and appropriate processing of biopsy specimens (tissue preservation for genetic studies). A surgical excisional biopsy or a percutaneous procedure (core needle biopsy) is decided according to the size and location of the mass. Although large lesions of the limbs can easily be biopsied without image guidance, deep-seated musculoskeletal lesions are difficult to target, and benefit from CT or US guidance. Imaging-guided percutaneous core needle biopsies are performed by trained radiologists, under local or general anesthesia, using CT or US guidance [64–68]. The procedure should ideally be performed by both the radiologist and the pathologist, the latter being the most qualified to evaluate the specimen quality and to separate the tissue for morphological and biological studies. If the specimen cannot be frozen immediately, it must not be fixed directly, but should be temporarily placed in culture medium such as Roswell Park Memorial Institute medium, RPMI.

Fine-needle aspiration (FNA) usually does not replace biopsy but, in our experience, constitutes an excellent first-line and reliable diagnostic procedure, provided that it is performed and examined by trained pathologists. It is an inexpensive technique, almost without morbidity [69], and it can be performed under local anaesthesia. Using fine needles (23 Gauge, 0.6 mm of outer diameter), with ultrasound guidance if necessary, provides highly contributive cell aspirates [70, 71]. In the diagnostic strategy, FNA may be used right after initial imaging when the decision to perform or not perform a biopsy is pending, e.g., in case of clinically and radiologically presumed benign or pseudotumoral lesion, in case of a highly vascularized lesion at risk for a biopsy, or for suspected relapses. FNA material is excellent for ancillary techniques (this material is tumor-cell rich and stroma-cell poor) and allows karyotyping and molecular analyses. Cells should be stored in ethylenediaminetetraacetic acid (EDTA) (Figs. 1.11 and 1.12).

Fluorescence in situ hybridization (FISH) is a molecular technique that allows identification of rearrangement, gain, deletion, or amplification of genetic material on chromosomes (Fig. 1.14). Pros and cons are summarized in Table 1.4. This method uses DNA probes labeled with different fluorescent dyes able to specifically hybridize to unique sequences on the DNA. In the pediatric diagnostic context, almost all the probes are of the break-apart type. These probes are designed to hybridize to centromeric and telomeric DNA sequences closely flanking a gene or locus of interest. In nuclei with normally paired chromosomes, each containing the gene or locus of interest, the signals are adjacent or overlying. In nuclei with balanced chromosomal translocation, instead, the signal of the telomeric probe is separated from that of the centromeric one, meaning that part of the genetic material was transferred to another segregated chromosome. This strategy is widely applied in the diagnostics of pediatric tumors to detect gene rearrangements. It has a great limitation, however: the fusion gene partner is unknown. Indeed, the probes highlight only that the gene or locus of interest is transferred, without providing any information about the new chromosomal attachment site. For instance, the demonstration of a chromosomal translocation using a break-apart set of probes specifically designed for the EWSR1 gene in a small round cell tumor is not diagnostic for ES/PNET. Indeed, EWSR1 may have combined with FLI1, ERG, E1AF, FEV, or ETV1 as it happens in ES/PNET or with WT1 as in desmoplastic small round cell tumor (DSRCT). In either case, this characteristic can also be an advantage since with a single set of probes all the spectrum of rearrangements of EWSR1 is covered. FISH is applicable to any type of cytological preparations, such as air-dried/alcohol-fixed smear, liquid-based preparation, and formalin-fixed paraffin-embedded (FFPE) cell-block sections. Of note, in smears and liquid-base preparations the nuclei are intact; the method is thus not affected by loss of nuclear material, which is one of the major technical pitfalls of histologic or cell-block FFPE section. Another application of in situ hybridization is the demonstration of viral infection, as for Epstein-Barr virus (EBV) in lymphomas.

PCR allows amplification of a target sequence of DNA included between the regions bound by two specifically designed primers. Thus, it is possible to determine the precise order of nucleotides of the amplicons through sequencing. This method is mainly used to detect gene mutations, insertions, duplications, and deletions. An example of application is the search for mutations in KIT and PDGFRA genes in the uncommon pediatric gastrointestinal stromal tumors (GIST). This is crucial not for the diagnosis but for the therapeutic implications. Indeed, patients with mutated GIST could benefit from targeted therapy.

A variant of PCR followed by sequencing provides an initial step of reverse transcription to obtain complementary DNA (cDNA) molecules from RNA templates. This enables the molecular investigation to move from a long DNA sequence (including both introns and exons of the gene) to its shorter messenger RNA (including only exons of the gene). The advantage is the possibility to overcome (at least partially) the problem of nucleic acids fragmentation and cross-linking due to the use of fixatives (mainly formalin). Indeed, a PCR analysis on fixed cells is reliable only if it is based on amplicons shorter than 200 base pairs, a length too limited to cover both introns and exons of a combined gene (other pros and cons of the technique are reported in Table 1.5). In messenger RNA, instead, amplicons of this length can extend into two juxtaposed exons. This is particularly useful in the analysis of known fusion genes resulting from chromosomal translocations. Indeed, by designing the primers so that each is positioned in one of the supposed combined genes, if the fusion occurred they will be able to drive the amplification (Fig. 1.15); if the fusion did not occur, they will not be able to drive the amplification. Thus, RT-PCR could be a valid alternative to FISH in this setting, since it demonstrates the occurrence of the rearrangement and with which gene. Regarding the example of small round cell tumor, using RT-PCR one can differentiate ES/PNET from DSRCT, being able to recognize the gene combined with EWSR1. However, all the possible fusion genes must be known a priori.

Another major limitation is related to the low sensitivity of classic Sanger sequencing that might bring about false negative results when applied to cytological samples. Indeed, the relative amount of tumor cells harboring the molecular alteration to be investigated could be reduced by an excess of inflammatory or non-tumoral cells in the sample. A neoplastic component higher than 25% of the cells is required for an adequate analysis. Thus, tumor cells enrichment through micro-dissection of the neoplastic cells must be performed in some cases. Obviously, the method suffers also when facing a low absolute amount of neoplastic cells. The result in these cases should be interpreted with caution. Where available, it is advisable to use a more sensitive technique. As for FISH, this method is feasible in any type of cytological sample. However, the best performances are achieved with fresh or frozen specimens. In any case, personnel must be aware of the risk of contamination and take all the necessary precautions. PCR and RT-PCR can be applied also for the detection of oncogenic viral DNA or RNA [78–80].

The current molecular biology workflow applied to FNA is destined to radically change in the coming years. Indeed, next-generation sequencing (NGS) will probably replace Sanger sequencing and FISH. NGS is a technique with an extraordinary sensitivity which, by parallelizing the sequencing process, is able to produce up to millions of sequences at the same time. This allows for performance of multiple gene analysis using a minimum of nucleic acids, a crucial aspect for aspirates. Moreover, NGS can cover all the range of genomic alterations, such as base substitutions, short insertions and deletions, amplifications, homozygous deletions, and gene rearrangements. Thus, in the same aspirate it will be possible to analyze concurrently several molecular abnormalities identifying a diagnostic molecular alteration, even if present in less than 1% of the cells.

ICC is a simple, valuable, and cost-effective method to reveal the lineage of differentiation of tumors and is available in almost all pathology laboratories worldwide. ICC is also used to analyze therapeutic targets on neoplastic cells and can be applied for the identification of molecular alteration [81, 82] since it can detect chimeric protein resulting from chromosomal translocation (pros and cons summarized in Table 1.6). In this latter context, ICC is effective in highlighting the presence of the protein product of a fusion transcript (Fig. 1.16). However, the antibodies usually bind to a small part of the protein (epitope) belonging to only one of the combined genes, and therefore does not provide direct information about the fusion gene partner. This is a limitation if the recognized epitope is in a protein that can combine with several others. For instance, a positive immunoreaction to ALK in inflammatory myofibroblastic tumors just means that a translocation involving the ALK gene happened. It is impossible to know if it concerned TPM3, TPM4, CLTC, RANBP2, ATIC, CARS, or SEC31A. However, indirect approximate clues about the translocation can be derived from the subcellular localization of the immunostaining, since the chimeric protein can have nuclear, cytoplasmic or membranous expression depending on the combined gene. ICC can be performed on air-dried/alcohol-fixed smears, liquid-based preparations, and FFPE cell block sections. However, ICC protocols must be optimized according to the material sampled, the fixative used, the times of each step, and the antibodies selected to guarantee reliability.

FC is a technique that enables concurrent testing of the presence of several antibodies labeled with different fluorochromes directed toward membranous or cytoplasmic antigens, which maximizes the amount of data obtained from a few cells, as in the case of cytological samples. It is used to help diagnose and classify neoplasms, mainly lymphomas (Fig. 1.17), to detect therapeutic molecular targets on the surface of malignant cells, and to monitor disease [83–85]. In order to determine the quantity and the immunophenotype of a malignant cell population, FC requires viable cells. Thus, fixation is forbidden. Aspirates must be immediately suspended in sterile saline solution or cell culture medium after sampling and processed as soon as possible to prevent cell loss. Indeed, FC analysis may be affected by nonspecific binding of the antibodies to dead or damaged cells. One should be aware that in such cases a negative search for neoplastic cells could be a false negative. Finally, even small tissue fragments may interfere with FC analysis, thus they must be broken down to allow the release of the cells in suspension into fluid before the analysis. Mechanical rather than enzymatic disaggregation is preferred, since enzymes could modify the epitopes recognized by the antibodies. Pros and cons of FC are summarized in Table 1.7

Rapid On-Site Evaluation (ROSE) [86, 87] is an invaluable procedure that should be mandatorily implemented in each service dealing with FNA, especially in the pediatric context. Indeed, ROSE allows the medical team to perform the following tasks:

Cases occur in which the cytopathologist performing ROSE needs a second opinion not attainable soon and is thus not able to decide how to handle the sample. In such circumstances, the cells may be collected and temporarily kept in a non-fixative solution (sterile saline or cell-culture medium) to delay the decision for a few hours. Then, the samples can be appropriately processed.