Corneal tissue engineering: From research to industry, quality of life impact, and Latin American ophthalmologists' perspectives

2.1 Bibliometric analysis

2.1.1 Search strategy

We searched the Web of Science-Core Collection (WOS-SCC)¹² database using topic as a field of search with the following keywords: tissue engineering, regenerative, reconstruction, corneal endothelium, and descement in the following query: TS=(“tissue* engineer*” OR regener* OR reconstruct*) AND TS=((cornea* AND endothel*) OR descement*). This search yielded a total of 1030 articles. The documents were then filtered by document type, including only original articles published in any language between 2003 and 2023. The data were searched and extracted on the same day (June 29, 2023).

2.2 Data extraction and harmonization

The full records, including the cited references of all 614 selected articles in plain text format from the WOS-SCC, were exported. The raw file was uploaded to VOS-Viewer Software¹³ https://www.vosviewer.com/ to screen the author’s names and to create a thesaurus file. The ‘found and replace’ function was used in open-line software to update all the information in the database and to make the metadata consistent between the VOS-Viewer and Biblioshiny formats.

2.3 Bibliometric analysis

The bibliometric analysis was conducted using VOS-Viewer software ³¹³ and the bibliometrix and biblioshiny R packages¹⁴ https://www.bibliometrix.org/home/index.php/layout/biblioshiny. VOS-Viewer Software was used to filter the top 25 most relevant authors and journals by creating a list according to their selection thresholds of the minimum number of documents and citations. Biblioshiny was used to perform a descriptive analysis of the bibliometric indicators, evaluate the citations of the top 25 authors and journals, and determine the impact factor and collaboration networks between the top authors and countries.

2.2 Patent analysis

We searched the Patent Public Search (USPTO)¹⁵ database. We used the following query: “corneal endothelium” AND (“tissue engineering” OR “regenerative” OR “reconstruction”) AND (biomaterials OR cell therapy OR scaffold OR membrane) NOT (“organoids” OR “epithelium”). This search yielded a total of 126 patents. A search of the European Patent Office database¹⁶ using the following query: “corneal endothelium in the title AND tissue engineering OR scaffold OR biomaterial in the title or abstract” retrieved 6 results. Given the difference in the number of results retrieved from these two databases, we conducted a third search using the Lens database¹⁷ with the following query: (title:((cornea AND endothelium) AND (cell therapy OR scaffold OR biomaterial OR tissue engineering OR membrane OR reconstruction NOT epithelium)) OR abstract:((cornea AND endothelium) AND (cell therapy OR scaffold OR biomaterial OR tissue engineering OR membrane OR reconstruction NOT epithelium)) OR claim:((cornea AND endothelium) AND (cell therapy OR scaffold OR tissue engineering OR membrane OR reconstruction NOT epithelium))). This search retrieved 181 results. The data search was conducted on August 28, 2023. The “analysis” function of the Lens database was used to generate graphs about the number of patents over time, type of documents, legal status, top inventors, jurisdiction, and most cited documents from 1980 to present.

2.3 Company analysis

We used the results on Biblioshiny based on the publication affiliation from the bibliometric analysis search. We chose the affiliations that were companies, and then we selected those that had an active website. From those companies, we chose those that had products related to engineered corneal tissue. Data relating to the company’s country, type of product, year of foundation, clinical use and phase of production were analyzed.

2.4 QALY analysis

For the QALY analysis, we created a Markov model¹⁸ that considered both the quality and quantity of life in the Mexican population using parameters such as life expectancy, cost of corneal transplant, and mean age of transplant. We considered three categories of vision health: blindness, sight, and mortality. Each category has several transition possibilities. Patients can transition from “sight” to “blind” and vice versa and then to death, referring to the end as the absorbing state, given that once an individual enters the state, they remain there.¹⁹ Only the transition from blindness to sight incurs expenses, including the cost of obtaining cells from one corneal donor and generating the necessary cells for up to ten engineered corneal endothelium constructs.²⁰ We used a hypothetical cohort of 1000 blind patients aged 35 years and ran the program for 40 years to calculate life expectancy. A discount rate of 3.5% over a 10-year period was used in the computations. We projected that the intervention would reduce the waiting list from three to one year since more transplants would be accessible. We calculated the cost per QALY gained for each intervention and compared it to Mexico’s GDP per capita (8,346.7 USD) to determine whether the interventions were cost-effective, assuming the same cost per patient. The reference cost for a corneal transplant was $6000 USD,²¹ and the calculated cost of an engineered corneal endothelium tissue was $2400 USD (taken from a preliminary cost analysis from data not shown considering donor cornea acquisition, laboratory consumables, reagents, biocompatible scaffold production cost, QA testing, and a GMP-compliant laboratory, among others). The transplant survival rate/duration was 87% after one year, 72% after three years, 54% after six years, and 42% after ten years.²²

2.5 LATAM survey

We created an online 10-question formulary using Survey Monkey, an online software platform with free license options, to assess Latin American ophthalmology surgeons’ attitudes toward the use of engineered tissue for corneal endothelium restoration. The formulary was distributed during October 2022 by email throughout the community of Latin American ophthalmology surgeons, and information about the respondent’s location, willingness to use a synthetic biocompatible membrane to transplant corneal endothelial cells, preferred characteristics (thickness, insertion port size, presentation, rigidity, and transparency level), and price range of the bioengineered tissue was recorded to assess the potential of transplanting engineered corneal endothelium instead of donated corneal endothelium. The replies were analyzed using descriptive statistics.

3.1 Bibliometric analysis

A total of 416 of the 1030 screened articles were excluded (those published before 2003, review articles, meeting abstracts, proceedings papers, book chapters, editorial material, letters, early access, corrections, or retracted publications), and 614 were included in the bibliometric analysis. A flow diagram describing the flow information used for article screening and selection is shown in Figure 1.

Figure 1. Engineered corneal endothelium—Study selection for bibliometric analysis.

The articles were published by 2,572 authors from 49 different countries in six different languages—English, French, German, Korean, Portuguese, and Russian—in 206 different journals and used a total of 17,194 references. In total, these articles had 12,042 citations, excluding self-citations. To analyze the trend in the number of publications and correlate it to the year, we performed a polynomial regression. Using a log transformation of the year as an independent variable, we observed a continuous annual growth rate that stabilized in 2021. We fitted a squared polynomial equation to predict the number of articles as a function of year, and we obtained a statistically significant R-squared value of 0.89 and a p value of 1.53×10^-9, demonstrating a trend toward an increase in the number of articles related to corneal endothelium regeneration, as shown in Figure 2a.

Figure 2. Bibliometric analysis of corneal endothelium engineering techniques.

a) Annual growth rate of the number of publications on corneal endothelium regeneration. b) Top 25 authors. c) Top 25 Journals. d) Author collaboration.

The analysis included a total of 2,572 authors contributing to the 614 selected original articles, resulting in an average of 0.23 documents per author. Eight authors had single-authored documents. The average number of coauthors per document was 6.34, and the average number of international coauthorships was 25.57%. Figure 2b shows the top 25 authors according to their publication rate, and Figure 2c shows the relationships between the top authors in the field who had more than 8 publications and 159 citations. The most productive authors on this specific topic are Metha JS (28 articles), Peh GSL (24 articles), Kinoshita S (23 articles), Koizumi N (23 articles), and Okumura N (23 articles). There is a strong coauthor relationship between Peh GS and Metha JS, as well as between Kinoshita S, Koizumi N, and Okumura, all of whom currently form the most productive research groups in the field.

The study included 206 journals with 614 published papers, yielding a total of 0.33 documents per journal. The top 25 journals with at least 5 articles and 35 citations are shown in Figure 2d. According to JCR 2022, these 25 journals have an average impact factor of 4.238 and are in quartile 2; 11 (52%) are in the category of ophthalmology, 5 (20%) are in the category of cell & tissue engineering, 3 (12%) are in the category of multidisciplinary sciences, 2 (8%) are in the category of material science, and 2 (8%) are in other categories. These journal articles have been referenced more than 558 times and have a local H-index greater than 13. Exp. Eye Res. (47) is the most relevant journal.

3.2 Patent analysis

A search of the Lens database returned 181 results. We applied the categorization filter to the following: A61f2/14 (eye parts, e.g. lenses, corneal implants; artificial eyes making thereof from organic plastic material), A61f2/142 (artificial or natural cornea replacement implant for repair of defective corneal tissue), A61p27/02 (Drugs for disorders of the senses, ophthalmic agents) A61l27/24 (materials for {grafts or} prostheses or for coating {grafts or} prostheses, collagen) A61l27/3604 (materials for {grafts or} prostheses or for coating {grafts or} prostheses, {characterized by the human or animal origin of the biological material, e.g. hair, fascia, fish scales, silk, shellac, pericardium, pleura, renal tissue, amniotic membrane, parenchymal tissue, fetal tissue, muscle tissue, fat tissue, enamel}), A61l27/38 (materials for {grafts or} prostheses or for coating {grafts or} prostheses, {containing added animal cells}), A61l27/3834 (materials for {grafts or} prostheses or for coating {grafts or} prostheses {Cells able to produce different cell types, e.g. hematopoietic stem cells, mesenchymal stem cells, marrow stromal cells, embryonic stem cells}, and C12n5/071 (Vertebrate cells or tissues, e.g. human cells or tissues). There were 26 patents who continued to be examined. The top inventors were Cong R., Fan T., Kinoshita S., Ueno M., Zhao J., Koizumi, Gutermuth A., Yang X., Duang H., and Hori J (Figure 3a). Among those, Kinoshita, Ueno, Koizumi, and Duan were also among the most cited authors in the bibliometric analysis. Four patents were granted, one was an additional patent, and twenty-one were patent applications between 1992 and 2024 (Figure 3b). The legal status of 10 of the patents was active, with only three being discontinued (Figure 3c). The jurisdiction with the largest number of patents is the United States, followed by the World Intellectual Property Organization (WIPO) and European patents (Figure 3d). The most cited patent, with 62 citations, was “Methods employed in replacement of the corneal endothelium”, of White T., granted in 1992; it was followed by “Human Corneal Endothelial Cell-Derived Precursor Cells, Cellular Aggregates, Methods for Manufacturing the Same, and Methods for Transplanting Precursor Cells and Cellular Aggregates”, of Amano, published in 2009, with 19 citations, currently discontinued; and “In Vitro Cornea Equivalent Model”, of Parenteau, published in 1994, with 6 citations.

Figure 3. Patent analysis of corneal endothelium engineering techniques.

a) Number of documents per inventor. b) Document type versus publication date. c) Number of documents by legal status. d) Number of documents per jurisdiction (data and images retrieved from Lens.org).

3.3 Company analysis

The affiliations of the articles included in the bibliometric analysis yielded 53 companies. Twenty of those had an active webpage. Of those, 10 had products related to the original research paper from which the affiliation was taken (Table 1). These companies were founded from 2001 to 2022. These countries are in four different regions: five in the USA, two in Japan, two in China and one in Spain. Five of these companies have products on the market (Linkocare, CorneaGen, Takara Bio, BioTissue, and Cell Science & Technology Institute); two are developing phase II clinical trials (Trefoil Therapeutics and Emmecell); and one is in a phase I clinical trial (Body Organ Biomedical Corp); two of them are in the preclinical phase (Vissum Group and Cellusion). Three of these companies have products whose main use is corneal transplantation (Linkocare, CorneaGen, and Cellusion). One of the companies (Corneagen) appears as the affiliation of three documents with five of the top authors from the bibliometric analysis (Kinoshita, Hamuro, Ueno, Okumura, and Koizumi); two of them are also among the top patent inventors (Kinoshita and Koizumi).

3.4 QALY analysis

The Markov model based QALY study showed that engineered corneal endothelium intervention is associated with greater QALY gains than full keratoplasty is. The analysis revealed that endothelial corneal transplant had a QALY increase of 0.07 compared to that of engineered corneal endothelium intervention (0.74 QALY gain on a scale of 0 to 1). Table 2 shows the comparison between the QALYs gained and the cost of the transplantation of engineered tissue and full keratoplasty.

3.5 Latin American survey

A total of 30 ophthalmology surgeons with subspecialties in the cornea answered the questionnaire. The participants were from 8 Latin American countries: Colombia, Ecuador, Venezuela, Mexico, Spain, Costa Rica, Paraguay, and the Dominican Republic (Figure 4a). In a question with open answers, the key features for selecting an engineered corneal endothelium as the first choice for transplantation were biocompatibility, cost, manipulability, and transparency. More respondents mentioned efficacy, ease of insertion, orientation, resistance during manipulation, accessibility, thickness, adherence, and viability time (Figure 4b).

Figure 4. Survey answer analysis.

a) Nationality distribution of the survey respondents. b) Features favoring engineered corneal endothelium for transplantation over full keratoplasty.

The majority of participants (97%) expressed the belief that synthetic membrane-based engineered tissue would be beneficial for corneal endothelium transplantation. The participants’ preferences for key characteristics of an engineered corneal endothelium are outlined in Table 3.