A topical VEGF inhibitor optimized for ocular indications

Abstract

Purpose : VEGF inhibitors approved for oncology, reformulated as eye drops, have to date failed in wet AMD trials due to lack of efficacy. Could a VEGF inhibitor be optimized for ocular indications? We report in vitro, in vivo efficacy and ocular PK of such a molecule, LHA510.

Methods : LHA510 was assessed in a FRET based assay measuring VEGFR2 kinase activity (n = 6 experiments) and in a cell line dependent on VEGFR2 signaling for survival (n = 5 experiments). Topical LHA510 was assessed in a rat laser CNV assay. CNV area was measured by analyzing images of fixed eye tissues 10 -14 days after laser injury, N = 10 rats/20 eyes per group. Brown Norway (BN) Rats were dosed with placebo or 0.1 – 3% LHA510 b.i.d. or t.i.d. (n = 11 experiments). Retinal leakage in male Dutch belted rabbits dosed with topical 3% LHA510 or placebo was measured 48 hrs after an IVT challenge of 400 ng of human VEGF₁₆₅ (n = 6 rabbits/group, n = 1 experiment). All samples were analyzed and exclusions applied on masked randomized images. LHA510 concentrations were determined in rat and rabbit ocular tissues using liquid chromatography and mass spectrometry. Rats were dosed with 4 uL of 1% LHA510 bilaterally. Plasma, retina or RPE/choroid/sclera samples were collected after a single dose or after 11 days of t.i.d. dosing ( n = 2 rats/time point, 7 time points, 5 min – 24 hrs). New Zealand white and red crossed rabbits were dosed topically with 30 uL of 2% LHA510 t.i.d. uniocularly for either a single dose or for 7 days t.i.d. and retina, choroid or plasma collected ( n = 4 rabbits/time point, 7 time points ranging from 0.5 hr – 96 hrs). Statistics were assessed with a one-way ANOVA with a post hoc analysis. ED50/90 values were calculated with Excel fit.

Results : LHA510 IC₅₀ in enzymatic and cellular KDR assays were 19 ±14 and 17 ±14 (nM ±SD). The ED50/90 for CNV area inhibition of topical LHA510 was 1.0 and 2.6% b.i.d. and 0.5 and 1.2% t.i.d. Topical LHA510 fully inhibited retinal leakage in rabbits challenged with IVT VEGF. LHA510 had high retinal and choroidal exposure and minimal plasma exposure following a single drop and accumulated in the ocular tissues with multiple doses (see tables). LHA510 was higher in the dosed eye than the non-dosed eye. This direct penetration accounted for 35.8% and 56.7% of the total AUClast in RPE/choroid and retina, respectively.

Conclusions : Efficacy of topical LHA510 in wet AMD patients will readout in 2017 (clin.trial.gov NCT02355028).

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.