Endocrine Abstracts

Introduction: Accurately discern malignant from benign nodules sometimes represents a real challenge. Fine needle aspiration (FNA) biopsy, ultrasound and The Bethesda System for Reporting Thyroid Cytopathology (BSRTC) allow for accurate diagnosis in 60–80% of all nodules. However, 10–20% of thyroid FNAs yield an indeterminate cytology (BSRTC III and IV categories). The real malignancy rate of BSRTC III category is difficult to ascertain because histopathological analysis of the lesion is not always available. Molecular testing may assist clinical practice in the management of controversial FNAs since 67% of thyroid cancers bear at least one genetic alteration. Next-generation sequencing (NGS) provides parallel high-throughput alternative to assess multiple targets of the genome and, in recent years, several mutational panels that include genetic alterations identified in thyroid cancer have been developed and tested for performance.

Aim: In view of the above, the aim of this study was to investigate the presence of not yet described genetic alterations occurring in FNA samples with indeterminate cytology, by using a custom multi-gene NGS panel, in order to better characterize indeterminate cytology nodules.

Materials and methods: The study includes 56 random FNA samples yielding Bethesda III or IV cytological analysis that underwent surgery, therefore histopathological diagnosis is available: 32 are benign nodules, 22 are papillary thyroid carcinomas (PTCs) and 11 are follicular thyroid carcinomas (FTCs). NGS was performed by using a custom library on Ion PGM™ sequencer. Reporter variant caller pipeline was used for data analysis and R and PLINK software were used to compute logistic regression and principal component (PCA) analyses.

Results: PCA analysis did not show any significant variation between PTC, FTC and benign nodules, indicating the heterogeneous nature of thyroid nodules. However, we did find genetic alterations in intronic regions of PI3KCA and HRAS genes that were significantly associated with our population of FTCs, showing an odds ratio>15 and P>0.001.

Conclusions: Our results indicate that FNA samples with Bethesda III or IV cytological analysis show genetic alterations in intronic regions of PI3KCA and HRAS genes in 16% of the samples and associate with a final histology consistent with FTCs. Several groups have developed Genetic Classifiers for thyroid cancer in order to increase sensitivity and specificity of molecular testing. However, unknown and/or intronic genetic alterations are usually discarded. We found an intronic HRAS point mutation that was strongly and positively associated with FTC phenotype that should be investigated for its’ potential predictive value.