Thyroid Carcinoma: Molecular Signature by Histotype-Specific Mutations and Gene Expression Patterns Diagnostic, Prognostic and Therapeutic Value of Gene Signatures

Although the most important application of novel molecular markers is that on cytology, translation from the bench to the FNA is complex . To make DNA- and/or RNA-based testing cost-effective on cytology a close integration with morphology is needed . Thus, the informativeness of the sample for cytology needs to be preserved to keep the accuracy of microscopy high. To this end, each of the steps of traditional cytology, such as preparation of FNA material, search for morphological criteria, assignment to the correct diagnostic class, and suggestion of the appropriate post FNA options should not be altered by molecular analysis. Then, this latter can refine cytology. The final result is to effectively stratify into high-risk and low-risk categories, and the indeterminate cytology classes identified by the BSRTC . Thus, cytological specimens should be properly handled to provide both morphological and molecular information . Our method of preparation of FNA to harvest material sufficient for both tests was recently validated on a series of 128 routinely performed FNA . The rationale behind our sample collection method was to ensure first an adequate cytological diagnosis and, then, to exploit part of the diagnostic material for molecular testing . Thus, two passes from different areas of the lesion are performed. A representative air-dried Diff-Quik stained smear is prepared within few minutes and reviewed on site ] . In 44 cases, the cytological evidences were sufficient for morphological assessment and the third pass was directly collected in RNA or DNA buffer extraction. Conversely, in 84 cases the specimen was either deemed inadequate by the onsite evaluation or required an additional ethanol-fixed Papanicolaou-stained smear to better evaluate nuclear morphology. Thus, a third pass was dedicated to the preparation of an additional smears and only needle rinsing was collected for BRAF testing. Higher average of extracted DNA concentration was observed in the dedicated pass group (25.9 vs. 7.95 ng/ m l). However, the rate of successful exon 15 BRAF amplification was similar with (43/44; 97.7%) or without (79/84; 94%) the dedicated pass. Thus, our protocol is suitable for both tests. When necessary, BRAF testing may also be performed on the residual samples of thyroid nodules, without interfering with routine cytology. Similarly, as far as mRNA markers are concerned, we have shown that in most samples, qRT-PCR analysis does not interfere with cytology . In fact, in a recent study on UbcH10 expression, including 84 cases with a cytological diagnosis of either follicular neoplasm ( n = 57) or suspicious for malignancy ( n = 27), we found that most (73.8%) cases were adequate for both tests