Thyroid Capsular Elastic Fiber Fragmentation as a Non-Invasive Surrogate for Capsular Invasion

Abstract Background: In encapsulated follicular-patterned thyroid tumors, proving capsular invasion requires exhaustive capsule submission, which is labor intensive and reproducible to varying degrees. We tested whether a Capsular Elastic Fiber Fragmentation Score (CEFFS), derived from Verhoeff–Van Gieson–stained sections, could serve as a histology-only surrogate for capsular invasion. Methods: In a single-center, blinded accuracy study, we identified 50 cases (25 minimally invasive follicular thyroid carcinoma MI-FTC, 25 follicular adenoma FA) defined by WHO 2022 criteria. Two pathologists, blinded to diagnosis, selected three capsule hotspots per case and scored elastic fiber continuity loss, fragmentation proportion, and tangential re-orientation. Hotspot sums were averaged to yield CEFFS (range 0–9). Demographics were compared after Shapiro–Wilk normality checks using Welch’s t-test or Mann–Whitney U for continuous variables and Fisher’s exact test for sex. Group differences in CEFFS and capsule thickness used Welch’s t-tests; category distributions were compared using chi-square tests. Logistic regression related MI-FTC probability to CEFFS; performance was summarized by area under ROC curve (AUC, DeLong 95% CI), calibration intercept and slope, and Brier score with 3,000 bootstrap resamples. Results: Groups were demographically balanced. CEFFS was higher in MI-FTC than in FA, and each 1-point increase in CEFFS approximately doubled the odds of MI-FTC (odds ratio 2.34). The CEFFS-based model showed good discrimination between MI-FTC and FA (AUC 0.87). Capsule thickness and pericapsular fibroinflammatory reaction were also greater in MI-FTC than in FA. Conclusions: Quantified elastic fiber disruption robustly distinguishes MI-FTC from FA and provides high-discrimination risk estimates in this single-center accuracy study. Within existing WHO/CAP diagnostic frameworks, CEFFS could be implemented as a histology-only triage tool to flag cases that warrant full circumferential capsule submission or intensified level sampling. This approach has the potential to accelerate reporting when exhaustive sampling is impractical, but validation and reproducibility studies, including external multicenter cohorts, are warranted.