Henry N. Marshall, Xikun Han, Sean Mullany, Ayub Qassim, Ella C. Berry, Mark M. Hassall, Thi Nguyen, Georgina L. Hollitt, Lachlan S. W. Knight, Bronwyn Ridge, Emmanuelle Souzeau, Angela Schulz, Richard A. Mills, Ashish Agar, Anna Galanopoulos, John Landers, Paul Healey, Stuart Graham, Alex W. Hewitt, Robert J. Casson, Stuart MacGregor, Owen M. Siggs, Jamie E. Craig

Purpose: To investigate the relationship between cardio- vascular diseasegenetic risk score and glaucomatous progression.

Methods: A cardiovascular disease polygenic risk score was calculated for 828 participants from the PROG- RESSA study. Participants were characterised as showing either predominantly macular ganglion cell inner plexiform layer (mGCIPL) thinning, predominantly pRNFL thinning, or equivalent mGCIPL and pRNFL thinning at enrolment. The cardiovascular disease polygenic risk score for these study groups was compared to a reference group of stable glaucoma suspects and to an external normative population. Replication was undertaken by phenotyping 634 participants from the Australian and New Zealand Registry of Advanced Glaucoma (ANZRAG).

Results: After accounting for age, sex and intraocular pressure, participants with predominantly mGCIPL thinning had a higher cardiovascular disease genetic risk score than reference participants (OR 1.69/SD, 95% CI [1.14, 2.52] P = 0.007), and then external normative population (OR 1.27/SD 95% CI [1.09, 1.49] p = 0.003). This was reproduced in a replication cohort of perimetric glaucoma participants from ANZRAG (OR 1.39/SD; 95% CI [1.05, 1.83] p = 0.022). Review of longitudinal spectral domain optical coherence tomography data showed that a higher cardiovascular disease genetic risk score was associated with a faster rate of mGCIPL progression (β: 0.095 μm/ year/quintile 95% CI [0.003, 0.18] p = 0.043). Review of visual field data showed that a higher cardiovascular dis- ease score was associated with a greater likelihood of para- central visual field change (OR 1.86 95% CI: [1.14, 3.05] p = 0.013). Participants with predominantly mGCIPL thinning also exhibited a higher VCDR genetic risk score (OR 1.40/SD 95% CI [1.18,1.66] p < 0.001), but a comparable intraocular pressure genetic risk score (OR 1.12/SD 95% CI [0.95, 1.32] p = 0.180) to the normative population. Conclusion: A polygenic risk score highlights the susceptibility of macular GCIPL to cardiovascular disease in glaucoma.