A phase I clinical trial of human embryonic stem cell‐derived retinal pigment epithelial cells for early‐stage Stargardt macular degeneration: 5‐years' follow‐up

Abstract Objectives To evaluate the long‐term biosafety and efficacy of transplantation of human embryonic stem cells‐derived retinal pigment epithelial (hESC‐RPE) cells in early‐stage of Stargardt macular degeneration (STGD1). Materials and methods Seven patients participated in this prospective clinical study, where they underwent a single subretinal transplantation of 1 × 105 hESC‐RPE cells in one eye, whereas the fellow eye served as control. These patients were reassessed for a 60‐month follow‐up through systemic and ophthalmic examinations. Results None of the patients experienced adverse reactions systemically or locally, except for two who had transiently high intraocular pressure post‐operation. Functional assessments demonstrated that all of the seven operated eyes had transiently increased or stable visual function 1‐4 months after transplantation. At the last follow‐up visit, two of the seven eyes showed visual function loss than the baseline; however, one of them showed a stable visual acuity when compared with the change of fellow eye. Obvious small high reflective foci in the RPE layer were displayed after the transplantation, and maintained until the last visit. Interestingly, three categories of patients who were classified based on autofluorescence, exhibited distinctive patterns of morphological and functional change. Conclusions Subretinal transplantation of hESC‐RPE in early‐stage STGD1 is safe and tolerated in the long term. Further investigation is needed for choosing proper subjects according to the multi‐model image and function assessments.


| INTRODUC TI ON
Stargardt macular degeneration (STGD1), also known as juvenile macular degeneration, is the most common monogenic hereditary macular/retinal dystrophy caused by autosomal recessive mutations in the ATP binding cassette subfamily A member 4 (Abca4) gene in chromosome 1. 1,2 This gene encodes the Rim protein (RmP or ABCA4), which is specifically expressed in photoreceptor cells and retinal pigment epithelium (RPE) cells. Mutations in the photoreceptor-specific flippase transporter ABCA4 leads to an accumulation of the toxic Nretinylidene-N-retinylethanolamine (A2E), resulting in atrophy of the RPE and in the death of photoreceptor cells. [2][3][4] Despite the high incidence of this disease, no curative treatments for it have been developed. While gene replacement therapies involving adeno-associated virus (AAV) have been successfully used for some ocular genes, the 6.8 kb cDNA of ABCA4 is too large to be packaged into AAV vectors.
Alternatively, lentiviral vectors, non-viral compacted DNA nanoparticles, CRISPR/Cas9-mediated genome editing of patient skin cellderived-induced pluripotent stem cells (iPSCs) or human embryonic stem cells (hESCs) could be used in gene-or cell-based therapies. [5][6][7] Recent studies have demonstrated that both genome-edited iPSCs and hESCs could be differentiated into retinal pigment epithelial (RPE) cells in vitro for transplantation. Moreover, the RPE cells implanted into human subjects are well-tolerated and could potentially restore some vision loss in STGD1 patients. [8][9][10][11] Thus, transplantation of hESC-RPE cells may become a promising new treatment option for macular degeneration. However, the long-term safety of stem cell implantation into human eyes had remained unclear, and the field lacked an objective visual function evaluation system for macular degeneration patients.
In our previous studies, we successfully generated RPE cells from clinical grade (CTS)-hESCs, which met the standard requirements for clinical applications. 12,13 We demonstrated that the CTS-

hESC-RPE cells, transplanted into the subretinal space of the Royal
College of Surgeons (RCS) rats with inherited retinal degeneration and light-damaged pigs, protected the animals against retinal degeneration. 14,15 In a recent clinical trial, we implanted CTS-hESC-RPE cells into the subretinal space of patients for the treatment of wet age-related macular degeneration (wet-AMD) due to neovascular disruption. 8 With optimized perioperative management and surgical protocols, wet-AMD patients who underwent the transplantation showed a new RPE-like cell layer in a previously damaged retinal area, without adverse health events. Moreover, visual function test results indicated partial vision improvement. 8 Our studies indicated that RPE cell replacement therapy is safe and feasible for wet-AMD, and possibly for other macular degeneration diseases.
In this study, we extended our hESC-RPE cell-based therapy to patients with early-stage STGD1 and conducted a phase I clinical trial. We employed a new fundus autofluorescence classification approach to group the subjects after stem cell transplantation, and we monitored their clinical outcomes. The patients were followed up for 5 years to evaluate the long-term safety and efficacy of the hESC-RPE therapy.

| hESC-RPE preparation
The hESCs, specifically the Q-CTS-hESC-2 cell line, were provided by the Institute of Zoology, Chinese Academy of Sciences and were authenticated by the Institute of Pharmaceutical and Biological Products Certification in China. The CTS-hESCs were differentiated into RPE, and the CTS-RPE cells were expanded, purified and certified by the GMP Laboratory of Cell Biotherapy Center, Southwest Hospital of Army Medical University, as described previously. 8 Each vial contains less than 0.5 EU/mL endotoxin and was pathogen-free (based on the results of the tests for fungi, bacteria, mycoplasma, syphilis, human immunodeficiency virus, and hepatitis B and C). The CTS-RPE cells had a >95% viability.

| Patient recruitment and Study design
Seven patients (two males and five females) who met the inclusion criteria were enrolled in this prospective study, which was an open-label, self-control, and single-centre study conducted from May 2015 to May 2020. Each participant underwent a single subretinal transplantation of 1 × 10 5 hESC-RPE cells into the operated eye, and the fellow eye served as control. We chose the eye with worse best-corrected visual acuity (BCVA). If the BCVA was same in both eyes, the eye with worse retinal electroretinography (ERG) and autofluorescence (AF) was chosen. The patients were followed up for 9-60 months after implantation. This study was approved by the Medical Ethics Committee of Southwest Hospital, Army Medical University (2015-18). The clinical trial was registered at the clinicaltrials.gov with a reference number

NCT02749734.
Conclusions: Subretinal transplantation of hESC-RPE in early-stage STGD1 is safe and tolerated in the long term. Further investigation is needed for choosing proper subjects according to the multi-model image and function assessments.

| Surgical procedure
Surgical vitrectomy and subretinal injection were carried out as reported. 8,16 In brief, a three-port pars plana vitrectomy was performed; this procedure involved both the creation of a complete posterior vitreous detachment and total vitreous excision. A small amount of saline was infused using a 41-gauge injection cannula (Bausch & Lomb Storz) into the sub-retinal space to detach the temporal retina, which is located next to the macula. A total of 10 5 Q-CTS-hESC-2-RPE cells suspended in 100 μL volume were slowly injected into the sub-retinal space, creating a localized domeshaped retinal detachment in the macular area (Video S1). Then the silicon oil was tamponaded after air-fluid exchange. After the surgery, the patient remained in a supine position overnight until the subretinal fluid was absorbed; thereafter, the patient changed to a prone position, which was maintained for 1 week. The silicon oil was removed after 3 months of transplantation. To inhibit potential immune response and inflammation, we administrated immunosuppression drugs to the patients before and after surgery as described previously. 8,17 Specifically, the patients received oral mycophenolate mofetil (500 mg, bid), tacrolimus (0.1 mg/kg/d, bid) and prednisone (0.5 mg/kg/d, qd) 1 week before surgery. After surgery, mycophenolate mofetil was given for 4 weeks. Tacrolimus was titrated to keep its serum levels within 3-7 ng/mL and was maintained until the 12th week. The prednisone dosage was decreased (0.25 mg/kg/d, qd) at the 4th week and then stopped at the 12th week.

| Clinical evaluation
The clinical examination protocol is summarized in the Figure 1

| Statistical analysis
Data were analysed using SPSS17.0 software and were expressed as mean ± standard deviation. Visual acuity, visual field and amplitudes as well as peak time of electrophysiology were measured during the follow-up visits. The visual parameters for the same eye and which were obtained at different time points (follow-up visits) were compared with the baseline measurement to evaluate any vision improvement. Moreover, the ETDRS letter scores of the fellow eye during BCVA assessments, which were conducted at the same time points, were compared with the baseline. Hybrid model test was employed in the statistical analyses. P < .01 indicated a significant difference between two groups. Supplementary information is available at Cell Proliferation's website.

| RE SULTS
Seven patients including two males and five females aged 19-27 years (median age: 23 years) were enrolled in this study. These STGD1 patients underwent surgery, wherein one eye was operated (operated eye) and the other eye served as a control (fellow eye). These patients were followed up for 9-60 months. As shown in Table 1, all patients were characterized as ffERG type 1 before transplantation. 1 They had stable vital signs, and they did not show any adverse conditions, such as allergy, immune rejection, fever, headache, or other systemic adverse reactions after subretinal hESC-RPE injection (Table S3). Moreover, they did not experience any severe local complications, such as endophthalmitis or retinal detachment.
Systemic examination revealed that neither tumour-like appearance nor elevated levels of tumour markers in the bloodstream were detected throughout the treatment period. However, 1-2 months after operation, two of the seven patients (P4 and P7) had a transiently high intraocular pressure ranging from 26 to 32 mm Hg, which was relieved by eye drops and cured after silicone oil removal 3 months after transplantation. Other adverse events, including conjunctival haemorrhage, hyperaemia and eye sting pain, were transient and did not require any intervention (Table S3). The VFQ-25 questionnaire did not reveal any significant changes before and after stem cell transplantation.
To evaluate the retinal function, we tested the patients' visual acuity using the ETDRS, and we found that all of the seven operated eyes remained stable in terms of visual acuity, compared with the fellow eyes at the same time point (P = .52, Figure 2). Of the seven patients, two (29%, P3 and P5) regained vision at the 1st month and one patient (14%, P5) regained vision at the 4th month of follow-up.
Three of those seven patients (42%, P2, P3 and P6) had ETDRS letter scores that were lower by more than five letters compared with the baseline during the last visit, indicating visual acuity loss. However, only one of them (14%, P2) was considered to have a significant visual acuity decrease after stem cell transplantation, considering the difference in EDTRS scores between the operated eye and the fellow eye from the baseline to the visiting time point (Table S1). Moreover, there was no significant difference in the overall retinal sensitivity of the visual field of the treated eyes before and after operation, nor between the treated eyes and the fellow eyes (P = .09, Figure 2). Subsequently, we evaluated the patients' visual acuity by using the PVEP functional test. We found that the patients' amplitude of  Figure S3, Table S2). The three other patients with category I and III classifications (42%; P1, P2 and P7) displayed a stable mfERG and a stable (P1 and P2) or a slightly decreased (P7) visual field in the SHR area ( Figure 6 and Figure S3, Table S2). For the category III patients (P2, P5 and P7), the postoperative AF showed a massive change in hypo-autofluorescence ( Figure 6B,C), whereas the fundus photography showed a pigmentation change ( Figure 6A) and OCT displayed only a partial SHR (Figure 6d,e,f) in the AF-altered area.
The long-term follow-up revealed that the partial SHR remained stable over time. Interestingly, our microperimetry test revealed that the fixation points in all of the seven patients shifted towards the transplanted area ( Figure 7). The fixation stability in microperimetry in all patients were stable or increased slightly (>5%), except in P3 (category II) and P7 (category III) whose fixation stability decreased. Although the category III patients showed a variable degree of light threshold reduction in their previous visual field, they showed a partial improvement in their fixation stability, as seen in P2 and P5 ( Figure 7C,D and Table S2).  Our ffERG and mfERG could more reliably identify patients with early-stage STGD1 because it is more accurate and sensitive than the AF and OCT tests, albeit AF imaging is still widely used as a monitoring tool for disease progression. 1,25,26 It has been reported that an abnormal fundus appearance is not always consistent with mild functional damage in rod and cone cells in early-stage STGD1. 1,24 (2) Our combined visual function assessments enabled monitoring of non-immunosuppressed rhesus monkeys were no longer detectable 3 weeks after transplantation due to rejection by the immune system. 27 Although the iPS-derived RPE cells originated from the patient's adult cells and were distinct from the hESC-RPE cells we applied, it did evoke serious concerns about the intraocular immune response after stem cell transplantation. 19 Idelson et al 28

DATA AVA I L A B I L I T Y S TAT E M E N T
All supporting data are included in the article and its additional files.