Analysis of the impact of oncocytic cells on the performance of the thyroid molecular classifier based on micrornas and DNA: results from the optimized version of the MIR-thype test

Marcos Santos; Oliveira Miriane de; de Oliveira Andrei Felix; Vilela Diego Nogueira; Rabelo Bruna Frizzo; Fredi Bruno Mari; Morales Augusto Isabela Fernanda; Deo Thamiris Gatti; Rodrigues Bruna Moretto

doi:10.1530/endoabs.101.PS2-15-07

PS2-15-07

Prev Next

Section Contents Cite

Endocrine Abstracts (2024) 101 PS2-15-07 | DOI: 10.1530/endoabs.101.PS2-15-07

ETA2024 Poster Presentations Translational thyroid cancer research-1 (10 abstracts)

Analysis of the impact of oncocytic cells on the performance of the thyroid molecular classifier based on micrornas and DNA: results from the optimized version of the MIR-thype test

Marcos Santos , Miriane de Oliveira , Andrei Félix de Oliveira , Diego Nogueira Vilela , Bruna Frizzo Rabelo , Bruno Mari Fredi , Isabela Fernanda Morales Augusto , Thamiris Gatti Deo & Bruna Moretto Rodrigues

Views

Author affiliations

Onkos Molecular Diagnostics, Department of Research & Development, Ribeirão Preto, Brazil

Background: In thyroid nodules, the presence of oncocytic cells poses a diagnostic challenge for both cytology and molecular analysis. This challenge arises from the distinctive morphological characteristics and molecular profile of these cells, making it difficult to differentiate between benign and cancerous conditions.

Objective: To evaluate the performance of the v2 algorithm version of a microRNA and DNA-based molecular classifier test (mir-THYpe full) in oncocytic subtypes of thyroid nodules.

Methods: The molecular data for nodule classification was obtained from preoperative FNA samples of 59 thyroid nodules (58 patients) (26 Bethesda-III; 32 Bethesda-IV; 1 Bethesda-V) with known post-surgery oncocytic lesions (45 oncocytic adenomas; 14 oncocytic carcinomas). The analysis of microRNA and DNA mutational data (BRAF V600E and pTERT C228T/C250T) were performed by qPCR.

Results: The molecular test correctly classified as negative for malignancy 38/45 (84.4%) oncocytic adenomas (two false-positives had TERT C228T mutation) and as positive for malignancy 10/14 (71.4%) oncocytic carcinomas (four true-positives had TERT C228T mutation). Forty-eighth out of the 59 oncocytic samples were correctly classified (81% accuracy). The performance of the algorithm was: 71% of sensibility, 84% of specificity, 59% of positive predictive value, 91% of negative predictive value at 23.7% disease prevalence.

Conclusion: Despite the diagnostic challenge imposed by oncocytic cells, the results showed that v2 version of the microRNA and DNA-based molecular test mir-THYpe full has an acceptable performance to early identification of benign and cancer oncocytic subtypes, when compared to Afirma and Thyroseq, that showed only 20% PPV on oncocytic lesions. Therefore, the molecular test has the potential to enhance prognostic insights by assessing pertinent mutations, thereby aiding in the prediction of patient outcomes.