Association analysis of genetic risk variants with response to treatment with intravitreal anti-vascular endothelial growth factor in Korean neovascular age-related macular degeneration patients

Abstract

Introduction: Although intravitreal anti-vascular endothelial growth factor (VEGF) injection has revolutionized the outcome of neovascular age-related macular degeneration (AMD) treatment, there is wide spectrum of clinical response, and genetic factors seems to be one of determinants. The purpose of this study is to investigate the association between genetic risk variants for AMD and response to intravitreal anti-VEGF in Korean neovascular AMD patients.

Methods: This prospective study included 366 treatment-naïve patients (366 eyes) with subfoveal neovascular AMD. Patients were genotyped for 17 single-nucleotide polymorphisms within 13 AMD-relevant genes. Initially, patients underwent three monthly injections of intravitreal ranibizumab (0.5 mg / 0.05ml) and were followed up monthly. Additional treatments with intravitreal ranibizumab or bevacizumab (1.25 mg / 0.05ml) were administered when the retreatment criteria were met. Main outcome measures were visual response based on visual acuity change criteria and amount of change in best-corrected visual acuity (BCVA). Tomographic response and angiographic response, which were based on optical coherence tomography and fluorescein angiography findings respectively, and changes in central retinal thickness (CRT) and lesion size from baseline were also evaluated. Genotypic association of all categorical and continuous outcome variables at month 6, 12, and 24 was evaluated with regression analysis for each candidate polymorphisms. To evaluate the potential gene-gene interaction, multifactor dimensionality reduction (MDR) analysis was performed. In addition, to reappraise the inconsistent previous pharmacogenetic results on anti-VEGF for neovascular AMD, meta-analysis was performed including the data from this cohort.

Results: At month 24, BCVA improved by 4.5 ± 22.5 letters and CRT decreased by 69.4 ± 112.6 µm from baseline. Regression analyses for genotypic association revealed that minor allele homozygotes of VEGFA gene rs3025039 had significantly higher chance of good visual response than other genotypes at month 24 (Odds ratio [OR], 5.46

95% Confidence Interval [CI], 1.79 – 16.70

P = 0.0029). Minor allele homozygotes of CFH rs800292 showed significantly lower chance of good angiographic response than other genotypes (OR, 0.26

95% CI, 0.11 - 0.62

P = 0.0021). Minor allele homozygotes for ARMS2 rs10490924 and HTRA1 rs11200638 showed larger amount of CRT reduction at month 12 with borderline significance when corrected for multiple testing. MDR analysis revealed the significant interaction between CFH rs800292 and PEDF rs1136287 for tomographic response at month 12. In meta-analysis, combined data from this study and 5 previous pharmacogenetic studies showed that minor allele homozygosity for ARMS2 or HTRA1 were associated with larger amount of visual improvement after anti-VEGF treatment for neovascular AMD.

Conclusion: In this Korean neovascular AMD cohort, VEGFA rs3025039, CFH rs800292, ARMS2 rs10490924, HTRA1 rs11200638, and PEDF rs1136287 showed possible association with response to anti-VEGF treatment. With more evidence of pharmacogenetic association with anti-VEGF agent, individualized therapeutic approaches based on genetic background may lead to optimal treatment outcome in neovascular AMD.