Phase II Randomized, Double-Masked, Vehicle-Controlled Trial of Recombinant Human Nerve Growth Factor for Neurotrophic Keratitis.

PURPOSE To evaluate the safety and efficacy of topical recombinant human nerve growth factor (rhNGF) for treating moderate-to-severe neurotrophic keratitis (NK), a rare degenerative corneal disease resulting from impaired corneal innervation. DESIGN Phase II multicenter, randomized, double-masked, vehicle-controlled trial. PARTICIPANTS Patients with stage 2 (moderate) or stage 3 (severe) NK in 1 eye. METHODS The REPARO phase II study assessed safety and efficacy in 156 patients randomized 1:1:1 to rhNGF 10 μg/ml, 20 μg/ml, or vehicle. Treatment was administered 6 drops per day for 8 weeks. Patients then entered a 48- or 56-week follow-up period. Safety was assessed in all patients who received study treatment, whereas efficacy was by intention to treat. MAIN OUTCOME MEASURES Corneal healing (defined as <0.5-mm maximum diameter of fluorescein staining in the lesion area) was assessed by masked central readers at week 4 (primary efficacy end point) and week 8 (key secondary end point) of controlled treatment. Corneal healing was reassessed post hoc by masked central readers using a more conservative measure (0-mm staining in the lesion area and no other persistent staining). RESULTS At week 4 (primary end point), 19.6% of vehicle-treated patients achieved corneal healing (<0.5-mm lesion staining) versus 54.9% receiving rhNGF 10 μg/ml (+35.3%; 97.06% confidence interval [CI], 15.88-54.71; P < 0.001) and 58.0% receiving rhNGF 20 μg/ml (+38.4%; 97.06% CI, 18.96-57.83; P < 0.001). At week 8 (key secondary end point), 43.1% of vehicle-treated patients achieved less than 0.5-mm lesion staining versus 74.5% receiving rhNGF 10 μg/ml (+31.4%; 97.06% CI, 11.25-51.49; P = 0.001) and 74.0% receiving rhNGF 20 μg/ml (+30.9%; 97.06% CI, 10.60-51.13; P = 0.002). Post hoc analysis of corneal healing by the more conservative measure (0-mm lesion staining and no other persistent staining) maintained statistically significant differences between rhNGF and vehicle at weeks 4 and 8. More than 96% of patients who healed after controlled rhNGF treatment remained recurrence free during follow-up. Treatment with rhNGF was well tolerated; adverse effects were mostly local, mild, and transient. CONCLUSIONS Topical rhNGF is safe and more effective than vehicle in promoting healing of moderate-to-severe NK.