Back

Juvenile and adult open-angle glaucoma form a single polygenic continuum

Schmidt, J.; Guevarra, M. C. B.; Ni, G.; Hassall, T. A. D.; Ji, S.; Nguyen, T. T.; Hollitt, G. L.; Kolovos, A.; Souzeau, E.; Le, N. Q.; Gharahkhani, P.; Turner-Maier, J.; Collantes, E. R. A.; Aboobakar, I. F.; Segre, A. V.; Hewitt, A. W.; Haan, N.; MacGregor, S.; Wiggs, J. L.; Craig, J. E.; Siggs, O. M.

2026-09-14 ophthalmology
10.64898/2026.09.13.26361381 medRxiv
Show abstract

Purpose: To test whether juvenile open-angle glaucoma (JOAG) and primary congenital glaucoma (PCG) are polygenically continuous with primary open-angle glaucoma (POAG), using a multi-trait POAG polygenic risk score (PRS). Design: Retrospective cohort study. Participants: 3,102 open-angle glaucoma index cases, of whom 2,836 had a recorded age at diagnosis (52 PCG, 226 JOAG, and 2,558 POAG) and 999 European-ancestry controls from a glaucoma disease registry and a population-based cohort. Two replication cohorts comprised 70 JOAG and 31 PCG probands with 230 controls (Mass Eye and Ear, MEE), and 9 PCG and 120 POAG cases (Genetics of Glaucoma Study, GOGS). Methods: One index case per family was analysed. PRS were calculated from weighted SNPs and adjusted for ancestry. Cases were stratified by age at diagnosis (PCG <=3 years, JOAG >=4 and <40 years, POAG >=40 years) and presence of pathogenic/likely pathogenic variants ("Mendelian"). Groups were compared by Kruskal-Wallis test, with logistic regression and area under the curve for per-SD effects. Main Outcome Measures: PRS centile values compared across glaucoma subtypes and Mendelian and non-Mendelian subgroups. Results: Mean PRS was highest in cases diagnosed in the fourth decade of life. Among subgroups, it was highest in non-Mendelian JOAG (84.1%), exceeding controls (53.3%), Mendelian JOAG (68.9%), and non-Mendelian POAG (79.3%) (all P <= 0.003). Non-Mendelian JOAG was also the best discriminated from controls (OR 3.66 per SD increase, 95% CI 3.02-4.49; AUC 0.814, 95% CI 0.782-0.845). Neither PCG subgroup differed from controls. Among Mendelian cases, only MYOC p.Gln368Ter showed significantly elevated PRS (OR 2.49 per SD, 95% CI 1.89-3.34). In the MEE replication cohort, JOAG was associated with higher PRS (OR 2.12 per SD, 95% CI 1.50-3.07) while PCG was not (OR 0.90, P = 0.73). In GOGS, mean PRS was higher in POAG than PCG (84.8 vs 59.7 centile; P = 0.026). Conclusions: A POAG PRS predicts JOAG risk with discrimination comparable to or greater than POAG itself, supporting a shared polygenic architecture across juvenile and adult open-angle glaucoma. PCG showed no PRS elevation regardless of Mendelian status, indicating a distinct architecture for which rare-variant sequencing remains appropriate.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.