Cytological biomarkers and risk of malignancy in indeterminate thyroid nodules: a systematic review and meta-analysis

Abstract

Purpose Thyroid nodules with indeterminate cytology are often referred for diagnostic surgery. This systematic review evaluated diagnostic markers in fine-needle aspiration cytology samples of indeterminate nodules to improve differentiation between benign and malignant nodules. Methods We searched MEDLINE, Embase, and the Cochrane Library from 1995 to 2025. The literature was screened, and data were extracted from the included studies. The quality of studies was assessed with QUADAS-2. We performed split component meta-analysis of diagnostic accuracy using R-studios and the SCS meta-function (split component synthesis). Results Forty-one relevant studies were identified, of which 23 eligible studies were included in the meta-analysis. Data on Galectin-3, HBME-1, and BRAF mutation from 1,564 patients were evaluated. Galectin-3 and HBME-1 demonstrated promising diagnostic accuracy, with diagnostic odds ratios (DORs) of 20.3 (95% CI: 9.7–42.5) and 15.0 (95% CI: 7.3–31.0), sensitivities of 0.74 (95% CI: 0.64–0.82) and 0.77 (95% CI: 0.67–0.84), and specificities of 0.88 (95% CI: 0.80–0.93) and 0.82 (95% CI: 0.73–0.89), respectively. Combined positivity yielded the highest DOR of 45.8 (95% CI: 17.7–118.6), sensitivity of 0.90 (95% CI: 0.81–0.95), and specificity of 0.84 (95% CI: 0.73–0.91). BRAF mutation had the lowest DOR of 6.6 (95% CI: 2.1–20.4) and sensitivity of 0.17 (95% CI: 0.09–0.29) but the highest specificity of 0.97 (95% CI: 0.93–0.99). Conclusion Our meta-analysis indicates that Galectin-3 and HBME-1, particularly when combined, may improve differentiation between benign and malignant nodules. BRAF mutation is highly specific but lacks sensitivity. These biomarkers show potential, but none warrant stand-alone clinical use, emphasizing the need for large-scale validation and potentially a multi-marker approach to improve diagnostic accuracy.