An alternative approach in endocrine pathology research: MALDI-IMS in papillary thyroid carcinoma

To the Editor, Many different molecular techniques (polymerase chain reaction (PCR), DNA sequencing, fluorescence in situ hybridization (FISH)) have been introduced in thyroid pathology [1]. Fewer studies evaluated the role of proteomic analysis in the research of new useful targets [2, 3]. Matrixassisted laser desorption/ionization (MALDI) imaging mass spectrometry (IMS) is a unique technology that explores the spatial distribution of biomolecules directly in situ, thus integrating molecular and morphological information. Therefore, we are investigating the potential role of MALDIIMS in detecting new diagnostic targets in papillary thyroid carcinoma (PTC). We have addressed the issue of molecular stratification of PTC in a small cohort of samples, evaluating both the genomic profile of the main genes of interest (BRAF, N-H-K RAS point mutations, and PAX8/PPARγ or RET/PTC rearrangements) and the proteomic profile shown byMALDIIMS analysis. We have collected ex vivo cytological specimens (see “Technical Note”), in order to analyse a sample perfectly mimicking the in vivo fine-needle aspiration (FNA) biopsy, while respecting the ethical requirements. Unsupervised proteomic analysis of the samples was followed by comparison with histopathology and genetic classification of the patients (Table 1). Data generated by MALDI-IMS were submitted to hierarchical cluster analysis (HCA), in order to evaluate the different proteomic expressions in the cases under study. HCA allowed to cluster proteomic spectra based on their similarity on a dendrogram: spectra showing comparable features were grouped under the same node of the dendrogram; then, it was possible to select nodes and assign them a specific colour (Fig. 1) [4]. Node A groups together three cytological specimens (two collected from patient 1 and one from patient 2), all histologically diagnosed as hyperplastic benign nodules. Node B groups together specimens collected from patients 3, 4 and 5, who were affected by PTC comparable for stage and histotype (Table 1). Node C, instead, groups a microcarcinoma (follicular variant (fv) of PTC), from patient 6, and a nodule histologically classified as uncertain malignant potential (UMP) tumour (patient 7). The last one, originated in a multinodular goitre environment and showed ambiguous morphological features, defined as borderline between a hyperplastic nodule and an “incipient” fv,PTC. The three distinct nodes (A, B, C) generated by HCA analysis confirmed that MALDI-IMS can potentially discriminate between benign and malignant thyroid lesions. Moreover, the homologies between cases 6 and 7 highlight that MALDI-IMS is able to identify a PTC even when the classic diagnostic morphological aspects are still unclear and ambiguous (mild nuclear clearing, rare grooves, no pseudoinclusions). This finding is in agreement Veronica Mainini and Fabio Pagni equally contributed to this paper.