Profilo di espressione di microRNA su campioni citologici tiroidei processati con liquid-based cytology

Background Thyroid nodules are common and diagnosed with a prevalence of 50-70% with the use of highly sensitive imaging techniques. Fine needle aspiration cytology (FNAC) is the most reliable diagnostic tool for differential diagnosis of clinically suspicious thyroid nodules. The main limitation of FNAC remain indeterminate cythopathology. Indeterminate thyroid lesions on FNAC harbor malignancy in about 25% of cases and therefore up to 75% of patients undergo unnecessary thyroidectomy. The detection of molecular markers in addition to cytological analysis of fine needle aspiration samples is a promising way to improve and to refine preoperative diagnosis of indeterminate thyroid nodules. MicroRNA are small endogenous noncoding RNAs implicated with gene expression regulation involved in tumour development and are emerging as a promising biomarkers for diagnostic and prognostic purposes. The aim of the study was to evaluate miRNA expression profile on fine needle aspiration samples processed on liquid-based cytology, mostly diagnosed as indeterminate lesions, to establish whether it is possible to distinguish malignant and benign lesions by providing a predictive molecular diagnosis on FNAC. Materials and Methods Thirty-tree nodules subjected to fine needle aspiration processed by liquid-based cytology and with available histological diagnosis from September 2020 to March 2022 have been retrospectively selected. The samples set included 20 nodules cytologically indeterminate classified as 10 TIR3A e 10TIR3B, 5 nodules cytologically benign-TIR 2 and 8 nodules malignant-TIR5. The cases were selected based on histological diagnosis after thyroidectomy so that the cytological classes TIR2 and TIR3A were all benign lesions and cases TIR3B and TIR5 were all malignant lesions. The expression profile of 84 well characterized miRNAs was preliminarily evaluated using Human miFinder miRNA PCR array (MIHS-001Z) on a series of 12 selected nodules with different cytological classification (3 TIR2, 3 TIR3A, 3 TIR3B, 3 TIR5) with available histological diagnosis). The analysis of the results of this first evaluation allowed to identify 19 miRNAs differentially expressed (up or down regulated) in the different cytological subgroups compared to the control group (TIR2). Subsequently the differential expression of some of these miRNAs, that showed peculiar expression characteristics in terms of selectivity of cytological subtypes (miR-96-5p and miR-155-5p respectively overexpressed in TIR5 and in TIR3B; miR-7-5p, miR-150-5p respectively under-expressed in TIR5 and TIR3A) or quantitative expression (miR-126-3p, miR-143-3p, miR-9-5p) compared to control group and in accordance with current literature evidence regarding their expression on histological samples of thyroid lesions, was evaluated in the whole cohort of 33 cytological samples using Real-Time PCR. Results The results of the expression of 84 independent miRNAs using miRNA PCR Array form miFinder platform on the series of 12 nodules selected from the cohort of study showed that the expression of some miRNAs in the different subgroups was significantly downregulated or upregulated compared to controls. In particular, the expression of miR-96-5p was significantly upregulated in TIR5 (fold change = 6.7194) while miR-155-5p was significantly upregulated in TIR3B (fold change = 4.1989). From the analysis performed, it was also highlighted how some miRNAs, in particular miR-9-5p, miR-26b-5p, miR-125a-5p, miR-126-3p, miR-144-3p, miR-23b- 3p, miR-143-3p, miR-122-5p, were significantly under-expressed in both TIR5 and TIR3B.Furthermore, some miRNAs were significantly downregulated into specific cytological categories. In particular, in the TIR5 category there was a significant downregulation of miR-27b-3p (fold change = -4.352), let-7c-5p (fold change = -7.9539), miR-7-5p (fold change = - 4.3822), miR-196b-5p (fold change = -5.6634). Conversely miR-142-5p (fold change = -4.6913), miR-150-5p (fold change = -12.4666), miR-103a-3p (fold change = -6.0629), miR-223 -3p (fold change = -7.1602), miR-424-5p (fold change = -8.8766) and let-7f-5p (fold change = -4.2871) had significant downregulation exclusively in the TIR3A diagnostic subcategory. The comparison between the cytological samples also revealed an overlap in the downregulation of some miRNAs (miR-26a-5p, miR-185-5p, miR-29c-3p, miR-140-3p, miR-100-5p) in all subcategories analyzed, with no significant differences in their expression. Data relating to the evaluation of the expression profile of selected miRNAs on the entire study cohort using Real Time PCR proved that there was a significantly higher expression of miR-96-5p in the nodules of the cytological category TIR5 (p = 0.0003) and TIR3B (p = 0.0097) compared to the control group. In fact, this miRNA was upregulated by 2.6 times and approximately 2 times respectively in the TIR5 and TIR3B nodules compared to the control group, with a greater increase in the malignant cytological category. A significantly higher expression (4.526 vs 1.884; 2.4 times) of miR-155-5p was also observed in the TIR3B nodules compared to the control group. In contrast, the expression of three miRNAs (miR-9-5p, miR-126-3p and miR-143-3p) was significantly reduced in the diagnostic categories TIR3B and TIR5. In particular, the expression of miR-9-5p was significantly downregulated in both TIR3B and TIR5 nodules (both p <0.0001) (Figure 7), with a mean reduction in miR-9-5p expression of 4-fold (0.6244 vs 2.486) and fifteen-fold (0.160 vs 2.486) compared to the control group. The expression of miR-126-3p was significantly reduced in TIR3B (p = 0.0072) and TIR5 (p <0.0001) nodules compared to the control group and this miRNA was respectively under-expressed by approximately 2-fold and 6.6-fold in these categories compared to the control group. MiR-143-3p was significantly downregulated in both TIR3B and TIR 5 (both p <0.0001) with mean expression 2.6 and 4 times lower, respectively, than in the control group. Finally, the expression of miR-7-5p was confirmed to be significantly lower in TIR5 than in the control group (2.1233 vs 5.386667, p <0.0001), whereas that of miR-150-5p was reduced by approximately 2.6 times in the TIR3A category compared to controls (2.24 vs 5.37, p <0.0001). Conclusions The expression analysis/assessment of a miRNA panel consisting of miR-96-5p, miR-155-5p, miR-9-5p, miR-126-3p, miR-143-3p could be a promising diagnostic tool in differentiating between lesions with indeterminate cytology to be subjected to clinical-instrumental follow-up or surgery.