Differences in variant profile between follicular thyroid carcinoma and adenoma: targeted next-generation sequencing study

Introduction Preoperative ultrasound-guided fine-needle aspiration cytology cannot reliably distinguish follicular thyroid carcinoma (FTC) from follicular thyroid adenoma (FTA), making these neoplasms particularly challenging to diagnose. Although several studies have explored variant differences using targeted next-generation sequencing (NGS) panels, the molecular distinctions between FTC and FTA remain incompletely defined. In this study, targeted NGS was performed on DNA from 52 formalin-fixed, paraffin-embedded specimens. Methods Genomic DNA was extracted, quality assessed, and targeted NGS performed for 14 genes commonly mutated in thyroid neoplasms. Synonymous variants were excluded, and analyses then focused on pathogenic alterations. Variant frequencies were evaluated at both the locus and gene levels in FTC and FTA. Results After excluding synonymous variants, 1,074 variants across 14 genes were identified, with 92.8% clustered in five genes: TERT (70.5%), TSHR (8.6%), RET (5.9%), TP53 (4.2%), and BRAF (3.6%). FTC and FTA exhibited distinct variant landscapes, and TERT variants were significantly enriched in FTC. EIF1AX splice site variants (chrX:20148727; c.338–2A>C and c.338–2A>T) were significantly associated with FTC. In the pathogenic dataset, each additional EIF1AX variant increased the likelihood of FTC by 15.4-fold. In contrast, RET and PTEN variants were more frequent in FTA, and each additional RET variant reduced the probability of FTC by nearly half (odds ratio, 0.489). Conclusion These findings underscored the potential diagnostic utility of variant profiling. TERT and EIF1AX variants were predictors of malignant transformation, whereas RET and PTEN variants were more frequently detected in FTA. Their diagnostic and therapeutic utility warrants further validation in clinical settings.