Animal Models for the Study of Keratoconus

**Background:** Keratoconus (KC) is a bilateral, asymmetric corneal disorder characterized by localized central thinning and cone-like protrusion of the cornea. Its precise etiology remains undetermined, though both genetic and environmental factors contribute to disease susceptibility. Due to KC's complex nature, there is currently no ideal animal model that represents both the corneal phenotype and underlying pathophysiology. This review discusses the current landscape of non-human animal models in KC research, including mice, rats, rabbits, and proposed novel animals such as avian chicks and tree shrews.

**Methods:** The authors conducted a narrative review of the available literature on keratoconus animal models. They categorized models as genetic, spontaneous, treatment-induced, or proposed. For each model, they evaluated the methods used to induce or identify KC-like phenotypes, including ultraviolet radiation (UVR) exposure, collagenase treatment, endo-β-galactosidase, atopic dermatitis, vitamin A deficiency, glucocorticoid treatment, and genetic modifications. Assessment techniques included optical coherence tomography (OCT), slit lamp biomicroscopy, corneal topography, pachymetry, histology (H&E staining, immunofluorescence, transmission electron microscopy), and biomechanical testing.

**Key Results:** - **Mouse models:** Genetic models include Itgb1 conditional knockout (stromal thinning, loss of epithelial cell layers, edema, scarring), Zfp469 mutation (stromal thinning but not progressive or localized), Ppip5k2 gene-trap (abnormal corneal surfaces, changes in anterior chamber depth, thickened epithelia, abnormal curvature, thinning of CCT), Tgfbr2 conditional knockout (stromal-specific thinning, thickened epithelium, reduced TGFB2 expression, diminished collagen fibril density), and FGFR2 knockout (progressive, localized stromal thinning at 3 months, corneal angle steepening, keratocyte apoptosis, corneal hydrops). Spontaneous models include SKC mice (cone-shaped corneas, autosomal recessive inheritance with male bias, androgen-dependent phenotype). Treatment-induced models include UVR exposure (corneal protrusion, stromal thinning, collagenolysis), atopic dermatitis/eye rubbing (cone-shaped corneas, epithelial thinning, stromal disorganization), collagenase injection (damage to collagen fibrils, thinning of cornea and epithelium, corneal rupture), and endo-β-galactosidase (cone-like shape, epithelial and stromal thinning). - **Rat models:** UVR exposure induced apoptosis in epithelium, stroma, and endothelium, with neutrophil infiltration in the stroma. SAL006 (ISR agonist) reduced stromal keratocyte density and collagenous ECM production. Vitamin A deficiency led to keratoconus, xerophthalmia, and perforation, but methods were inconsistent and controls were absent. - **Rabbit models:** Collagenase type II treatment (via corneal trephine, topical solution, or intrastromal injection) consistently produced increased keratometry (Km), decreased central corneal thickness (CCT), looser collagen fibril arrangement, and reduced biomechanical stiffness. Glucocorticoid (fluorometholone) treatment decreased material stiffness after 8 weeks. - **Proposed novel animals:** Avian chicks (pop-eye strain with keratoglobus, corneal thinning, altered curvature) and tree shrews (corneal diameter ~8.5 mm, CCT 202–301 µm, similar collagen structure to humans) have been suggested but not yet employed for KC modeling.

**Clinical Implications:** The lack of a validated in vivo animal model is a critical impediment to studying KC pathogenesis and developing new treatments. An ideal model would demonstrate cone-like morphology, localized central thinning, and degradation of corneal layers, confirmed by slit-lamp microscopy, OCT, pachymetry, topography, histology, immunofluorescence, and biomechanical measurements. Current models have drawbacks: genetic models may not capture environmental influences; treatment-induced models may not reflect genetic contributions; and many studies lack comprehensive assessment of all relevant variables. Future directions include combining genetic susceptibility with environmental factors (e.g., UVR, eye rubbing) in mice, confirming spontaneous animal lines at multiple levels, and exploring novel animals like tree shrews. A robust model would accelerate understanding of KC mechanisms and enable testing of therapies such as collagen crosslinking and pharmacological interventions.