DICER1 alterations in thyroid lesions: a systematic review and meta-analysis with clinicopathologic implications

DICER1 is a key RNase III endoribonuclease involved in microRNA biogenesis and is implicated in both constitutional DICER1-related tumor predisposition and sporadic thyroid tumorigenesis. In thyroid pathology, DICER1 alterations have been reported across a broad spectrum of lesions, but their reported frequency and clinicopathologic significance vary substantially across published series. We performed a systematic review of thyroid-focused studies published between 2020 and 2025. Studies were included in the quantitative synthesis only if they provided an interpretable lesion-level numerator and denominator for thyroid lesions harbouring DICER1 alterations. Additional thyroid-focused studies that refined the cytomorphologic, histologic, and lesion-spectrum interpretation of DICER1-associated thyroid disease, but lacked a suitable denominator for prevalence meta-analysis, were retained for qualitative clinicopathologic synthesis. Random-effects meta-analysis of logit-transformed proportions was used for the quantitative component. Seven studies met criteria for quantitative synthesis, comprising 16,831 thyroid lesions, of which 317 were reported as DICER1-altered. Study-level proportions ranged from 1.4% in a large adult consecutive molecular-testing cohort to 22.0% in a pediatric follicular-patterned tumor cohort. The pooled proportion of thyroid lesions harbouring DICER1 alterations was 5.76% (95% CI, 2.49%-12.78%), with substantial heterogeneity (I² = 96.94%). Exploratory stratification showed lower pooled proportions in predominantly adult cohorts and higher pooled proportions in pediatric or young-enriched cohorts. Qualitative synthesis showed that DICER1 alterations occur across a wide thyroid lesion spectrum, including multinodular and follicular nodular disease, follicular adenoma, NIFTP, follicular thyroid carcinoma, follicular-patterned papillary thyroid carcinoma, and rare higher-grade or primitive malignant tumors such as thyroblastoma. Cytologically and histologically, these lesions frequently show follicular-patterned architecture, often with macrofollicular or mixed follicular growth and relatively bland nuclear features.Seven studies met criteria for quantitative synthesis, comprising 16,831 thyroid lesions, of which 317 were reported as DICER1-altered. Study-level proportions ranged from 1.4% in a large adult consecutive molecular-testing cohort to 22.0% in a pediatric follicular-patterned tumor cohort. The pooled proportion of thyroid lesions harbouring DICER1 alterations was 5.76% (95% CI, 2.49%-12.78%), with substantial heterogeneity (I² = 96.94%). Exploratory stratification showed lower pooled proportions in predominantly adult cohorts and higher pooled proportions in pediatric or young-enriched cohorts. Qualitative synthesis showed that DICER1 alterations occur across a wide thyroid lesion spectrum, including multinodular and follicular nodular disease, follicular adenoma, NIFTP, follicular thyroid carcinoma, follicular-patterned papillary thyroid carcinoma, and rare higher-grade or primitive malignant tumors such as thyroblastoma. Cytologically and histologically, these lesions frequently show follicular-patterned architecture, often with macrofollicular or mixed follicular growth and relatively bland nuclear features. DICER1 alterations are uncommon in large unselected adult thyroid cohorts but enriched in pediatric, young-adult, and follicular-patterned settings. They should not be interpreted as isolated lesion-specific markers. Rather, their diagnostic significance lies in the integration of molecular findings with lesion type, cytomorphology, histology, patient age, and clinical context, including the selective recognition of cases in which constitutional DICER1-related tumor predisposition should be considered.