Long-term functional and structural outcomes in X-linked retinoschisis: implications for clinical trials.

Abstract

X-linked retinoschisis (XLRS) is an inherited retinal disease (IRD) caused by pathogenic mutations in the retinoschisin gene, RS1. Affected individuals develop retinal layer separation, leading to loss of visual acuity (VA). Several XLRS gene therapy trials have been attempted but none have met their primary endpoints. An improved understanding of XLRS natural history and clinical outcomes may better inform future trials. Here, we report the long-term functional and structural outcomes of XLRS and the relevance of RS1 genotypes to the visual prognosis of affected individuals. A retrospective chart review of patients with molecularly confirmed X-linked retinoschisis was performed. Functional and structural outcomes, and RS1 genotype data, were included for analysis. Fifty-two patients with XLRS from 33 families were included in the study. Median age at symptom onset was 5 years (range 0-49) and median follow-up was 5.7 years (range 0.1-56.8). Macular retinoschisis occurred in 103 of 104 eyes (99.0%), while peripheral retinoschisis occurred in 48 of 104 eyes (46.2%), most often in the inferotemporal quadrant (40.4%). Initial and final VA were similar (logMAR 0.498 vs. 0.521; p = 0.203). Fifty of 54 eyes (92.6%) developed detectable outer retinal loss by age 20, and 29 of 66 eyes (43.9%) had focal or diffuse outer retinal atrophy (ORA) by age 40. ORA but not central subfield thickness (CST) was associated with reduced VA. Inter-eye correlation was modest for VA (r-squared = 0.03; p = 0.08) and CST (r-squared = 0.15; p = 0.001). Carbonic anhydrase inhibitors (CAIs) improved CST (p = 0.026), but not VA (p = 0.380). Eight of 104 eyes (7.7%) had XLRS-related retinal detachment (RD), which was associated with poorer outcomes compared to eyes without RD (median final VA 0.875 vs. 0.487; p <0.0001). RS1 null genotypes had greater odds of at least moderate visual impairment at final follow-up (OR 7.81; 95% CI 2.17, 28.10; p = 0.002) which was independent of age at onset, initial CST, initial ORA, or previous RD. Overall, long-term follow-up of XLRS patients demonstrated relatively stable VA, with presenting CST, development of ORA, and null RS1 mutations associated with poorer long-term visual outcomes, indicating a clinically relevant genotype-phenotype correlation in XLRS.