Surgical Method for Implantation of a Biosynthetic Retinal Pigment Epithelium Monolayer for Geographic Atrophy: Experience from a Phase 1/2a Study

To report the intraoperative methods and anatomic results for subretinal implantation of an investigational human embryonic stem cell–derived retinal pigment epithelium (RPE) monolayer seeded on a synthetic substrate (California Project to Cure Blindness Retinal Pigment Epithelium 1 [CPCB-RPE1]) in geographic atrophy (GA). Single-arm, open label, prospective, nonrandomized, Phase 1/2a study. Advanced non-neovascular age-related macular degeneration (NNAMD). The worse-seeing eye (≤20/200) of each subject underwent subretinal implantation of a single 3.5×6.25 mm CPCB-RPE1 implant with a preplanned primary end point of safety and efficacy at 365 days. Commercially available 23-gauge vitrectomy equipment, custom surgical forceps, and operating microscope with or without intraoperative OCT (iOCT) were used. Exact Wilcoxon rank-sum tests and Spearman rank correlation coefficients were used to assess the association of the percentage of the GA area covered by the implant with patient and surgery characteristics. The partial Spearman correlation coefficient was calculated for the correlation between duration of surgery and baseline GA size after adjustment for surgeon experience. Intraoperative exploratory measures are reported, including area of GA covered by implant, subretinal position of implant, duration of surgery, and incidence of adverse events. Operative recordings and reports were used to determine exploratory outcome measures. Sixteen subjects were enrolled with a median age of 78 years (range, 69–85 years). Median duration of the surgery for all subjects was 160 minutes (range, 121–466 minutes). Intraoperative OCT was used to guide subretinal placement in 9 cases. Intraoperative OCT was potentially useful in identifying pathology not evident with standard intraoperative visualization. Median GA area at baseline was 13.8 mm 2 (range, 6.0–46.4 mm 2 ), and median GA area left uncovered by the implant was 1.7 mm 2 (range, 0–20.4 mm 2 ). On average, 86.9% of the baseline GA area was covered by the implant. In 5 subjects, >90% of the GA area was covered. Baseline GA size was inversely correlated with percentage of GA area covered by the implant (r s =−0.72; P = 0.002). No unanticipated serious adverse events related to the implant or surgery were reported. Surgical implantation of CPCB-RPE1 targeted to the area of GA in subjects with advanced NNAMD is feasible in an outpatient setting. Intraoperative OCT is not necessary but potentially useful in identifying subretinal pathology and confirming implant location.