Introduction

Corneal endothelial failure (CEF) results from diminished pump and barrier functions due to abnormalities in corneal endothelial cells (CECs), which are often indicated by a significant reduction in CEC density, possibly accompanied by extensive guttae formation. This pathogenesis is frequently seen in cases of Fuchs endothelial corneal dystrophy (FECD), exfoliation syndrome, and corneal endothelial trauma from cataract, glaucoma, vitreoretina, and laser-iridotomy surgeries [1]. It also occurs in instances of chronic corneal graft failure. In these situations, known as bullous keratopathy (BK), corneal endothelial transplant procedures such as Descemet stripping automated endothelial keratoplasty (DSAEK) [2,3,4] and Descemet membrane endothelial keratoplasty (DMEK) [5,6,7], which utilize donor corneas from the central cornea, have been effective in restoring corneal clarity for extended periods [8].

Our novel allogeneic cultured human CEC (cHCEC)-transplant therapy, leveraging regenerative medicine technology, offers the potential for restoring healthy corneal endothelium with better prognosis than the above-mentioned conventional keratoplasty [9]. This therapy involves the surgical transfer of cHCECs with a Rho-associated kinase (ROCK) inhibitor, Y-27632, into the anterior chamber [10, 11]. Our previous clinical trial showed that for optimal surgical results, it is crucial to use mature differentiated cHCECs that resemble healthy in vivo human CECs [9].

After developing procedures to reproducibly produce cHCECs at the Kyoto Prefectural University of Medicine Cell Processing Center (KPUM CPC), adhering to Standard Operating Procedures (SOP) aligned with Good Manufacturing Practices (GMP), we conducted the physician-initiated phase II and phase III clinical trials of the cHCEC product made at the KPUM CPC under the support of the Japan Agency for Medical Research and Development (AMED) after receiving the clinical-trial protocol approval of the Pharmaceuticals and Medical Devices Agency (PMDA).

We conducted a multicenter, double-blind physician-initiated phase II clinical trial, titled "Exploratory physician-initiated trial for the establishment of innovative regenerative medicine using cultured corneal endothelial cells for BK (UMIN000028324)." This exploratory study used cHCECs with a fixed dose of 100 µM Y-27632 as an investigational product, administered in dosages of 2 x 105 cells (low dose), 5 x 105 cells (medium dose), and 1 x 106 cells (high dose), each in a volume of 300 µL and injected into the anterior chamber. The cell dose served as a control in a stratified block randomization, with the site as a stratification factor, divided into 3 groups: low, medium, and high dose. Fifteen eyes, with 5 eyes per group, were enrolled.

Subsequently, a multicenter, open-label physician-initiated phase III clinical trial, named "Confirmatory physician-initiated trial for the establishment of innovative regenerative medicine using cultured corneal endothelial cells for BK (UMIN000034334)," was conducted as a verification trial. This trial used the high dose (1 x 106 cells) with 100 µM Y-27632 from the phase II study, albeit from a different product lot. The investigational product, consisting of 1 x 106 cHCECs, was administered in the same manner as in phase II. This study included 12 eyes.

Here, we detail the combined analysis of the phase II and phase III clinical trials, henceforth referred to as the combined trial. The findings from this combined analysis provide essential data for the Common Technical Document (CTD) for regulatory submission, responses to PMDA inquiries, and other relevant purposes.

Participants and methods

Study participants

The characteristics of the enrolled eyes are presented in Figure 1. A total of 29 consecutive eyes that met the following inclusion and exclusion criteria were enrolled in the phase II and phase III clinical trials: (1) best-corrected visual acuity (BCVA) of worse than 0.5, (2) CECs unobservable via specular microscopy or endothelial cell density below 500 cells/mm2, (3) corneal thickness higher than 630 μm with the presence of corneal epithelial edema, (4) patient age between 20 and 90 years, and (5) written consent obtained. One participant from each trial (1 eye each) withdrew consent before cHCEC-transplant therapy (case number 01-02 for phase II, 31-12 for phase III). The remaining 27 eyes received the protocol treatment, and no major protocol deviations such as violations of discontinuation criteria or implementation of prohibited concomitant treatments occurred. Excluding those who withdrew consent, the population of patients was defined as the Safety Analysis Set (SAS). The Full Analysis Set (FAS), the most extensive analysis set, excluded patients assigned to the low- and medium-dose groups in the phase II study.

Fig. 1
figure 1

Study design and overall case disposition. Sixteen eyes were enrolled in the phase II clinical trial, and 13 eyes, in the phase III clinical trial. One eye from each trial withdrew consent before cHCEC-transplant therapy. The remaining 27 eyes received the protocol treatment, and in none was the treatment discontinued. The population of eyes that underwent the protocol treatment was defined as the Safety Analysis Set (SAS). The Full Analysis Set (FAS), the most extensive analysis set, excluded eyes assigned to the low-dose group of 2 x 105 cells and the medium -dose group of 5 x 105 cells in the phase II study

Full size image

The distribution of the 27 SAS eyes across each center was as follows: 23 eyes at the University Hospital Kyoto Prefectural University of Medicine, 2 eyes at the Kyoto University Hospital, and 2 eyes at the National Center for Geriatrics and Gerontology.

These trials were conducted in accordance with the tenets set forth in the Declaration of Helsinki and in compliance with the “Good Clinical Practice” stipulated by the Japanese Ministry of Health, Labour and Welfare. The protocols were approved by the PMDA, the Japanese regulatory authority, and approved by the institutional review boards of the University Hospital Kyoto Prefectural University of Medicine, Kyoto University Hospital, and National Center for Geriatrics and Gerontology. Written informed consent was obtained from all the participants before their involvement in the trials.

Case background, cHCEC product, and cHCEC-transplant therapy

The background details of the cases in the SAS are summarized in Table 1. The median age at enrollment for both studies was 75.0 years (range, 41–88 years), and 15 of the eyes were from female patients. The most common cause of BK in this study was FECD, in 10 eyes, followed by pseudophakic BK (PBK) and pseudoexfoliation syndrome-related BK (PEX-BK) in 6 eyes. Notably, 51.9% (14 eyes) had ocular complications, 11.1% (3 eyes) had previous corneal transplant in their fellow eyes, and all 27 eyes had undergone previous ocular surgeries excluding corneal transplant.

Table 1 Case demographics and baseline characteristics (Safety Analysis Set)
Full size table

The method used for the production of the cHCEC product was slightly modified from that previously described [9,10,11,12]. The cell products for these clinical trials were examined to verify that they met the criteria for mature-differentiated cHCECs resembling healthy in vivo human CECs (Supplementary Table S1). All surgical procedures were performed with the patient under local anesthesia, as described previously [9,10,11]. Briefly, after polishing removal of the abnormal extracellular matrix and the degenerated cells on the Descemet membrane in an 8-mm diameter area of the central cornea, a 26-gauge needle was used to inject a suspension of cHCECs into the anterior chamber. The patient was then immediately placed in a prone position for 3 hours to enhance the adhesion of the injection-transplanted cells. Table 2 details the findings at the cHCEC-transplant therapy in the SAS. All 27 eyes where the protocol treatment was initiated were treated as prescribed [9,10,11]. After cell transplant, all the patients were given systemic and topical glucocorticoids and antimicrobial agents in accordance with the drug regimen we use in our regular corneal transplant procedures (Supplementary Table S2).

Table 2 Summary of operative findings (Safety Analysis Set)
Full size table

No patients underwent prohibited concurrent therapies such as corneal transplant, and no concomitant surgeries were conducted during the protocol treatment period.

Efficacy analysis

Efficacy analysis was conducted on the FAS consisting of 17 eyes. The primary endpoint was achieving a CEC density of 1000 cells/mm2 or more at 24 weeks post-cHCEC-transplant. The secondary endpoints included (1) corneal thickness below 630 µm without corneal epithelial edema at 24 weeks post-cHCEC-transplant, (2) change in corneal thickness from pretransplant to 24 weeks post-cHCEC-transplant, and (3) improvement in BCVA from before to 24 weeks post-cHCEC-transplant.

In vivo CEC images were obtained via slit-scanning contact specular microscopy and its computer algorithm by the center method (CellChek; Konan Medical). The central corneal thickness was measured by use of Scheimpflug imaging with a focus on the thickness at the pupillary center (Pentacam HR; Oculus Optikgerate).

Safety assessment

Safety assessment was conducted on the SAS comprising 27 eyes. In the phase II study, the assessment period was divided into 2 phases: up to 24 weeks post-transplant and from 25 to 52 weeks post-transplant. For the SAS as a whole, the occurrence of adverse events up to 24 weeks post-transplant was tabulated.

Statistical and analytical considerations

All the clinical data and findings were re-examined and verified by a third-party organization, and statistical analysis was performed by the Department of Biostatics, Kyoto Prefectural University of Medicine, Kyoto, Japan. Statistical analyses were conducted using SAS version 9.4 (SAS Institute).

Results

Primary endpoint

The success rate for achieving a CEC density of 1000 cells/mm2 or more at 24 weeks post-transplant is detailed in Table 3. In the phase II trial, this was achieved in 80% of the eyes (4 of 5; 95% CI, 28.4%, 100.0%); in the phase III trial, in 100% of the eyes (12 of 12; 95% CI, 73.5%, 100.0%); and in the combined trial, in 94.1% of the eyes (16 of 17; 95% CI, 71.3%, 99.9%). The CEC densities at each evaluation point are shown in Fig. 2a and Supplementary Table S3. Owing to limited pretransplant data, changes from pretransplant levels could not be examined.

Table 3 Success rate for the primary and secondary endpoints (Full Analysis Set)
Full size table
Fig. 2
figure 2

Postoperative follow-up clinical data of corneal endothelial cell (CEC) density, central corneal thickness (CCT), and best-corrected visual acuity (BCVA). Measures of a CEC density, b CCT, and c BCVA are provided in the box plots. BCVA is expressed as logMAR visual acuity, which was converted from decimal visual acuity. The horizontal line in the boxes represents the median, and the bottom and top of the boxes represent the lower and upper quartiles, respectively. I bars indicate 1.5 times the lower and upper quartiles, and the dots represent outliers. Wk week

Full size image

Secondary endpoints

Table 3 presents the percentage of eyes with corneal thickness under 630 µm and no corneal epithelial edema at 24 weeks post-transplant: 100.0% (5 of 5; 95% CI, 48%, 100%) in the phase II trial, 75.0% (9 of 12; 95% CI, 42.8%, 94.5%) in the phase III trial, and 82.4% (14 of 17; 95% CI, 56.6%, 96.2%) in the combined trial. The measured corneal thicknesses at each evaluation point are shown in Fig. 2b and Supplementary Table S3.

A summary of the changes in corneal thickness from pretransplant to 24 weeks post-transplant is shown in Table 4. The average reduction was −182.8 µm (95% CI, −267.2 µm, −98.4 µm) in the phase II trial, −189.3 µm (95% CI, −263.6 µm, −114.9 µm) in the phase III trial, and −187.4 µm (95% CI, −240.2 µm, −134.5 µm) in the combined trial. The details of changes from pretransplant levels at each evaluation point are shown in Supplementary Table S3.

Table 4 Change in corneal thickness from pre-cHCEC transplant to 24 weeks post-cHCEC transplant (Full Analysis Set)
Full size table

Table 3 also indicates the percentage of eyes with improvement in BCVA (logMAR) from pretransplant to 24 weeks post-transplant: 100.0% (5 of 5 eyes; 95% CI, 47.8%, 100.0%) in the phase II trial, 100.0% (12 of 12; 95% CI, 73.5%, 100.0%) in the phase III trial, and 100.0% (95% CI, 80.5%, 100.0%) in the combined trial, with improvement defined as a decrease of 0.2 or more in logMAR visual acuity. These measurements are shown in Fig. 2c and Supplementary Table S3.

The efficacy response data are illustrated in Fig. 2; they show the trends in CEC density, corneal thickness, and BCVA (logMAR). Figure 3 shows a representative patient with pseudophakic BK. Corneal endothelium was reconstituted with high cellular density of over 4000 cells/mm2, and corneal transparency and thickness were recovered within 4 weeks of surgery and maintained for up to 24 weeks.

Fig. 3
figure 3

Clinical results of a representative eye with pseudophakic bullous keratopathy. Slit-lamp microscopy images (upper), slit-scanning wide-field contact specular microscopy images of the central cornea (middle), and Scheimpflug images (lower) of pre-cHCEC transplant (left column), 4 weeks post-transplant (middle), and 12 weeks post-transplant (right column). Corneal endothelium was reconstituted with high endothelial cellular density (ECD), and corneal transparency was recovered within 4 weeks of surgery. The rapid decrease in central corneal thickness (CCT) is apparent in the 2-dimensional corneal pachymetry images. Best-corrected decimal visual acuity (BCVA) was improved. Scale bar: 100 μm

Full size image

Adverse events

By 24 weeks post-transplant, 88.9% of the cases (24 of 27) had experienced adverse events in the SAS. Adverse events as listed in the Case Report Forms (CRFs) were compiled and categorized using the Japanese version of the ICH International Glossary of Pharmaceutical Terms (MedDRA/J version 23.1). Categorized adverse events according to System Organ Class (SOC) and preferred term (PT) are detailed in Supplementary Table S4. The most common adverse events occurring in 4 or more cases were eye pain in 9 cases (33.3%), nasopharyngitis in 6 cases (22.2%), constipation in 5 cases (18.5%), and intraocular pressure (IOP) increased in 4 cases (14.8%).

Specifically, in terms of the "eye disorders" of SOC, eye pain was reported in 9 cases (33.3%), and eyelid edema and lacrimation increased in 2 cases each (7.4%). Adverse events occurring in 4 or more cases and classified outside the category of "eye disorders" of SOC included nasopharyngitis in 6 cases (22.2%), constipation in 5 cases (18.5%), and IOP increased in 4 cases (14.8%), from which all the cases recovered. Regarding severity, 5 cases (18.5%) experienced moderate levels of severity: 2 cases (7.4%) with ocular pain and 1 case (3.7%) with eyelid edema in "eye disorders," and 1 (3.7%) with musculoskeletal pain, stomach cancer, papillary thyroid cancer, insomnia, and pruritus, all in the nonocular categories. There were no cases with severe or fatal adverse events.

Adverse reactions occurred in 12 of 27 cases (44.4%) by 24 weeks post-transplant. Table 5 details the adverse reactions categorized by SOC and PT. Within the "eye disorders" of SOC, eye pain was reported in 9 cases (33.3%), and eyelid edema and lacrimation increased were noted in 2 cases (7.4%) each. IOP increased was also reported in 2 cases (7.4%) within the "investigations" of SOC; in both these cases, the elevated IOP resolved.

Table 5 Summary of adverse reactions (Safety Analysis Set)
Full size table

Serious adverse events were observed in 2 cases of the phase II study: 1 case (case no. 01-04) developed gastric cancer and papillary thyroid cancer before 24 weeks postinjection, and the other case (case no. 01-10) suffered femoral neck fracture between 24 and 52 weeks postinjection. No serious adverse events were observed in the phase III study.

Discussion

We conducted a physician-initiated phase II clinical trial to determine the optimal dosage of cHCECs for the treatment of BK. The high-dose (1 × 106 cells) group showed higher efficacy than those of the medium-dose (5 × 105 cells) and low-dose (2×105 cells) groups while maintaining safety (Supplementary Tables S5 and S6). From these results, we determined that the optimal dosage of cHCECs to be injected into the anterior chamber is 1 × 106 cells. In this combined analysis, we included 5 eyes using the highest dose from the phase II trial and 12 eyes from the phase III trial that underwent transplant using the same dose for the efficacy analysis. The combined results from the FAS of 17 eyes were obtained for the reason described above. The FAS consisted of 6 eyes with FECD, 4 eyes with PBK and laser iridotomy-induced BK (LI-BK), 2 eyes with PBK and PEX-BK, 1 eye with FECD and LI-BK, 1 eye with PEX-BK, 1 eye with congenital hereditary endothelial dystrophy (CHED), 1 eye with intraocular surgery-related BK (IOS-BK), and 1 eye with nonguttata FECD. The improvements in this diverse group of BK etiologies show the robustness of cHCECs in improving vision and decreasing corneal thickness [13]. Previous reports have presented findings suggesting that the overall health of CECs following corneal transplant may be influenced by an impaired microenvironment. This includes the cytokine profiles, miRNA, metabolite and iris damage in the anterior chamber [14,15,16,17] as well as the size and height of the corneal guttae [18]. Despite the variety of microenvironments in the transplanted site, the efficacy of cHCEC-transplant therapy was confirmed in the FAS.

In only 1 eye was the primary endpoint not achieved, that of a 69-year-old woman with PBK and LI-BK. The CEC density in this eye was already as low as 1285 cells/mm2 at 4 weeks post-transplant, suggesting that the number of cHCECs adhering to the posterior surface of the cornea was low, possibly because of instability of the prone position. The secondary endpoint of achieving corneal thickness below 630 μm without corneal epithelial edema at 24 weeks post-cHCEC transplant was not met in 3 exceptionally severely affected eyes: those of a 46-year-old man with CHED, a 74-year-old woman with nonguttata FECD, and a 70-year-old man with IOS-BK following cataract surgery, vitrectomy for retinal detachment, and extraction and scleral suturing of an intraocular lens. However, the corneal thickness consistently decreased, from 794 to 685 μm, 1212–728 μm, and 854–668 μm, respectively, without corneal epithelial edema within 24 weeks of surgery. The BCVA of these 3 eyes improved from 0.4 to 0.7, from 0.01 to 0.7, and from 0.1 to 0.2, respectively. Efficacy was also confirmed in these exceptionally severely affected eyes.

From the combined results from the SAS of the 27 cases, no obvious safety concerns were identified; none of the serious adverse events were deemed related to the study treatment, and most adverse reactions related to cHCEC-transplant therapy were of an ocular nature and transient. With respect to potential adverse events specific to cHCEC-transplant therapy, we focused on 3 pathologic events including elevated IOP, allogeneic immune reaction, and anterior uveitis. In the SAS, no instances of IOP elevation within 1 week of cHCEC-transplant therapy occurred. This suggests the absence of acute IOP elevation triggered by cHCEC transplant into the anterior chamber. Two of 15 eyes (13.3%) in the phase II trial and 2 of 12 eyes (16.6%) experienced elevation of IOP, which was successfully managed with antiglaucoma drugs. When compared with a reported incidence of IOP elevation greater than 21 mmHg after DSAEK at 54% [19], the incidence of IOP elevation after cHCEC transplant therapy was comparatively low. The reported incidences of immune rejection after DSAEK and DMEK were 10.0% and 1.9%, respectively [4, 7]. Notably, neither allogeneic immune reaction nor anterior uveitis occurred in the SAS. Moreover, in our reported animal-model experiment, antigen recognition after CEC injection into the anterior chamber did not occur [20]. Given these combined results, the risk of allogeneic immune reaction in cHCEC-transplant therapy is presumed to be extremely low.

In these phase II and phase III trials, polished removal of abnormal materials from the Descemet membrane was successful in all the eyes without descemetorhexis (Table 2). Descemet membrane stripping only (DSO) is a procedure that involves the removal of a small area of the failed central endothelium and of the Descemet membrane without the need for transplant of a donor corneal tissue graft [21]. From the aspect of management of guttae in FECD eyes, cHCEC-transplant therapy combined with DSO may yield a slightly better visual outcome than that of the current cHCEC-transplant therapy. The preliminary data of cHCEC-transplant therapy with descemetorhexis have shown positive results [10]. The clinical outcomes of cHCEC-transplant therapy and of cHCEC-transplant therapy with DSO with/without ROCK-inhibitor topical application should be compared in the future as cHCEC-transplant therapy for FECD becomes optimized.

This study was subject to certain limitations. First, it did not include cHCEC transplant for phakic eyes. Second, data on cHCEC transplant combined with cataract surgery were lacking. Third, the study excluded patients with severe systemic disorders, such as chronic renal failure necessitating hemodialysis. It is important to note that anterior-chamber microenvironments immediately after cataract surgery or during hemodialysis may vary considerably when compared with those observed in the eyes included in this study.

In conclusion, the overall success, measured against the primary and secondary endpoints, exceeded the required threshold for regulatory approval in Japan. Postsurgery, some cases experienced mild or moderate adverse events such as transient ocular pain and lid edema; however, no severe adverse event was observed. Owing to the credibility established by these data, the cHCEC product was approved in Japan and is now marketed under the name Vyznova.