Introduction/Purpose: Peroneal tendon disease (PerTD) is a major cause of lateral ankle pain, including tendinosis, tenosynovitis, tears, and instability, all involving inflammation or degeneration of the tendon and tenosynovium (TS). The TS maintains tendon health by providing lubrication and nutrients, and its role in disorders like rheumatoid arthritis and trigger finger is well established. In PerTD, however, its contribution is poorly understood. Research has largely focused on tendon structure, leaving cellular and subcellular TS features understudied. Despite its clinical importance, no current classification of PerTD incorporates both tendon and TS pathology. Such classifications are essential for predicting outcomes, guiding treatment, and studying disease mechanisms. We aimed to develop a clinically meaningful TS and tendon classification and link TS phenotypes to molecular signatures and patient-reported outcomes.
Methods: We developed a 5-stage PerTD classification for surgically treated patients (IRB-approved): Type 0 (non-inflamed TS, normal tendons, n=9), Type 1 (inflamed TS, normal tendons, n=16), Type 2 (<50% tendon tear/tendinosis, n=6), Type 3 (>50% tear, n=6), and Type 4 (near-complete rupture, n=3, Fig1B).
Transcriptomic differences across types were analyzed, and Gene Ontology (GO) enrichment was performed. TS specimens were formalin-fixed and paraffin-embedded (FFPE), H&E-stained, and qualitatively assessed. Picrosirius Red (PSR)–stained FFPE sections were imaged under fluorescence microscopy, tiled into 10,000µm2 regions, and collagen-positive fractions normalized per specimen. Multiplex IHC (CLIC5, CD68, CD31, DAPI) was imaged on the Akoya PhenoImager. PROMIS Pain Interference and Physical Function Scores were collected preoperatively and at 3, 6, 12, and 24 months postoperatively. Collagen fraction was modeled using mixed-effects beta regression with Tukey-adjusted estimated marginal means. PROMIS score trajectories were compared (Kruskal–Wallis test and Dunn’s correction). Analyses were two-sided (α=0.05) in R Studio.
Results: Transcriptomic analysis revealed stage-specific TS molecular signatures across PerTD (Fig1A) with distinct GO-term enrichments differentiating stages 1–4 (Fig1B). Early PerTD (types 1/2) showed enrichment of inflammatory and tissue-remodeling pathways, whereas advanced PerTD (types 3/4) demonstrated upregulation of mitochondrial genes, altered fatty acid metabolism, and neutrophil degranulation (Fig1B). Histopathology showed progressive dense sublining connective tissue and thickened TS lining (Fig1C-D). IHC for CLIC5 (TS fibroblasts) and CD68 (TS macrophages) confirmed these findings (Fig1E). Quantitative PSR analysis showed significantly increased collagen-positive area in tendinopathic PerTD (types 2–4) vs type 1 (p=0.03, Fig1F). Patients with dense connective tissue exhibited greater postoperative improvement in PROMIS Physical Function and reduction in Pain (Fig1G), trends not observed in patients without a dense connective tissue reaction (data not shown).
Conclusion: Our classification defines PerTD categories with distinct histopathological and molecular profiles, including macrophage enrichment and fibroblast expansion within thickened TS lining, consistent with a macrophage–fibroblast activation axis driving fibrotic remodeling. While fibrosis and lining thickening were more common in advanced stages, they did not fully align with clinical categories, underscoring the need for refinement through integration of histopathology and non-invasive imaging. The observed association between dense connective-tissue reaction and greater postoperative functional improvement highlights the classification’s potential as a prognostic tool and a platform for biomarker discovery, ultimately informing preoperative decision-making and targeted therapeutic strategies.
