Molecular Vision 2025; 31:55-67
<http://www.molvis.org/molvis/v31/55>
Received 25 August 2024 | Accepted 21 March 2025 | Published 23 March 2025
Molecular Vision 2025; 31:55-67
<http://www.molvis.org/molvis/v31/55>
Received 25 August 2024 |
Accepted 21 March 2025 |
Published 23 March 2025
Anshuman Verma,1 Arif O. Khan,2,3 Venkatesh Pochaboina,1 Sirisha Senthil4
The first and last authors contributed equally to this study.
1Institute of Rare Eye Diseases and Ocular Genetics, LV Prasad Eye Institute, Hyderabad, India; 2Eye Institute, Cleveland Clinic Abu Dhabi, Abu Dhabi, United Arab Emirates; 3Department of Ophthalmology, Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH; 4VST Center for Glaucoma Care, L V Prasad Eye Institute, Hyderabad, India
Correspondence to: Sirisha Senthil, L.V Prasad Eye Institute, Kallam Anji Reddy campus, L V Prasad Marg, Road No: 2, Banjara Hills, Hyderabad, India-500034; Phone: +91 40 6810 2619; email: sirishasenthil@lvpei.org
Purpose: This study describes a distinct spectrum of latent transforming growth factor-β-binding protein 2 (LTBP2)–related ocular phenotypes in pediatric glaucoma with supporting genetic evidence and highlights our clinical experiences in its management.
Methods: A total of 189 children with childhood glaucoma underwent whole-exome sequencing–based genetic testing. Of these, 24 children displayed LTBP2-related phenotypes, among whom 18 cases who tested positive for LTBP2 variants were included in the study. The identified variants were confirmed through Sanger sequencing whenever possible and analyzed using in silico tools. The clinical presentation, genetic variants, and management of these 18 cases were thoroughly reviewed and presented.
Results: All 36 eyes of the 18 children with biallelic LTBP2 variants exhibited megalocornea without Descemet break, iridodonesis, and ectopia lentis. Pupillary changes were noted in all eyes, with persistent pupillary membrane in 78% (28/36) and ectropion uveae in 19% (7/36) eyes. Secondary glaucoma was observed in 72% (26/36) eyes, requiring surgery in 13 of these. Retinal pathology was noted in 47% (17/36) eyes. Lensectomy was performed in 94% (34/36) eyes with a mean age of 4.09 ± 3.5 years. Logistic regression analysis suggested that older age at lensectomy increased the risk of secondary glaucoma (hazard ratio, 1.69; [95% Confidence Interval: 1.00, 2.86], p < 0.05). The identified LTBP2 variants included five stop-gain variations, six frameshift variations, and one substitution variation, with five being novel and seven classified as rare variants.
Conclusions: The study expands the classic LTBP2-related phenotype spectrum in an Indian pediatric glaucoma cohort, highlighting additional features such as persistent pupillary membrane, ectropion uveae, and associated retinal pathology. These ocular manifestations were predominantly linked to nonsense LTBP2 variants. From a management standpoint, early lensectomy can help prevent secondary glaucoma, while timely identification and treatment of peripheral retinal pathology can reduce the risk of sight-threatening complications.
Latent transforming growth factor-β-binding protein 2 (LTBP2) plays a key role in the structural integrity of the extracellular matrix and its interactions with microfibril-containing elastic tissues [1]. LTBP2-related ocular dysgenesis is a rare autosomal recessive condition characterized by megalocornea, zonular weakness, and a predisposition to secondary lens-related glaucoma. The phenotypic spectrum associated with LTBP2 variants is wide, and its ocular manifestations can be confused with those seen in primary congenital glaucoma (PCG) [2]. The diagnosis of PCG is clinical, with typical features including corneal enlargement, corneal edema or haze, Descemet membrane breaks (Haab striae), optic nerve cupping, ocular enlargement (buphthalmos), myopia, and astigmatism. The most common identifiable cause for PCG is biallelic mutations in cytochrome P4501B1 (CYP1B1) [3].
In 2009, LTBP2 was suggested as a second gene that can be associated with PCG [4]. However, later, it became evident that biallelic LTBP2 variants more commonly result in a PCG-mimicking phenotype: primary megalocornea with zonular weakness and secondary lens-related glaucoma [2]. Recognizing the classic LTBP2-related phenotype and differentiating it from PCG is crucial, as management strategies differ; LTBP2-related cases require lensectomy, while PCG typically necessitates angle surgery. In this study, we report our experience and observations for the largest genetically confirmed cohort of the classical LTBP2-related phenotype reported to date. We also describe additional phenotypic features, novel LTBP2 variants, and management recommendations based on our clinical experiences.
The patients were recruited from April 2020 to December 2023 at the L V Prasad Eye Institute, Hyderabad, India. The study had received approval from the institutional ethics committee, and explicit informed consent was obtained from the legal guardians. The study strictly adhered to the ethical guidelines outlined in the Code of Ethics of the World Medical Association, as per the Declaration of Helsinki.
A total of 189 children diagnosed with childhood glaucoma were subjected to whole-exome sequencing (WES)–based genetic testing. This group included children with typical or atypical features of PCG, secondary developmental glaucoma, or syndromic glaucoma. Among these, children with PCG (identified with the CYP1B1 variant), syndromic ectopia lentis (Marfan syndrome, homocystinuria, etc.), secondary glaucoma due to posterior segment pathology, and isolated congenital megalocornea and keratoglobus, all negative for LTBP2 variants, were excluded from the study. Phenotypically, 24 children exhibited classic LTBP2-associated features, including primary megalocornea without Haab striae and zonular weakness. Among these, two children had characteristic features, but no pathogenic variants were detected. Another child with similar features tested positive for a growth hormone receptor (GH-R) gene variant, indicative of Laron dwarfism. Additionally, three children with microspherophakia and ectopia lentis, along with suspected megalocornea, tested positive for C3 and PZP-like alpha-2-macroglobulin domain containing 8 (CPAMD8), cystathionine β-synthase (CBS), and fibrillin 1 (FBN1) genes, respectively. Following the exclusion of these six cases, 18 children with LTBP2-related ocular phenotypes and positive for biallelic LTBP2 variants were included in the study. Consequently, the relative frequency of apparent biallelic LTBP2 variants among the pediatric glaucoma cohort that underwent genetic testing was 9.5% (18 of 189), and among those exhibiting characteristic LTBP2-related features, the prevalence was 75% (18 of 24).
All the included 18 patients underwent a detailed ophthalmic examination in the clinic physically or under anesthesia. Anterior segment evaluation included horizontal corneal diameter (HCD), corneal clarity, iris or lens details, refraction, and measurement of intraocular pressure (IOP) with a Perkins and Goldmann applanation tonometer. Posterior segment evaluation involved examining the optic disc and assessing the retina with a 78D lens or 20D lens and indirect ophthalmoscopy. Patients also underwent ocular biometry, and anterior and posterior segment imaging was performed whenever possible. They were evaluated for their clinical presentation, treatment, and outcomes until the last follow-up in December 2023.
Patients and their families received pretest and posttest counseling from a certified genetic counselor at the institute, providing education regarding the genetic nature of the condition, its inheritance pattern, and limitations of genetic testing. A multigenerational pedigree was drawn for all 14 families except one (family 11), who was unavailable for genetic counselling.
Next-generation sequencing–based genetic testing was performed in all probands. Briefly, WES with an approximately 30-Mb region was conducted using the Illumina (San Diego, CA) NovaSeq platform. The targeted regions covered approximately 99% of the reference consensus coding sequence (CCDS) with an average sequencing depth of 80× to 100× and over 90% of bases covered at a depth of 20× within the target region. Variant calling against the reference genome was performed using the Genomic Analysis Tool Kit. The identified variants underwent annotation and filtering using the Golden Helix VarSeq and Varsome analysis workflow, adhering to the American College of Medical Genetics and Genomics (ACMG) 2015 update guidelines for the classification of sequence variants [5]. All potential modes of inheritance were considered and identified variants were evaluated based on family history and clinical information to assess pathogenicity and causality. The variant nomenclature followed the Human Genome Variation Society (HGVS) standard, with the LTBP2 reference sequence (NM_000428.3).
An in silico analysis of identified LTBP2 variants was performed using Mutation Taster (Berlin, Germany), Franklin by Genoox (Palo Alto, CA), and Ensembl Variant Effect Predictor (VEP; Cambridge, UK). Different parameters under these tools were examined to determine the pathogenicity of the identified variants.
The clinical parameters of the 18 patients were analyzed for their association with the development of glaucoma using logistic regression analysis. The data were analyzed for normality using the Shapiro-Wilk test. Descriptive statistics included mean and standard deviation for normally distributed variables and median and interquartile range for nonnormally distributed variables. Categorical data were described in proportions. A p value of <0.05 was considered statistically significant.
A total of 18 children (11 females, 7 males), with a median presenting age of 4 years (range: 4 months to 14 years), from 14 families (13 consanguineous and 1 non-consanguineous), were included in the study based on the established inclusion and exclusion criteria. Complete clinical details for the cases in this cohort, including the number of patients and eyes with associated ocular features, are presented in Appendix 1. The selected cohort exhibited no major systemic or syndromic abnormalities beyond ocular features through the last follow-up. Figure 1 shows the representation of different ocular features (in percentages) in 36 eyes of the 18 patients in the study cohort. In all 36 eyes (100%) of 18 children, there was megalocornea (HCD >12 mm) without Descemet breaks, evident gross iridodonesis, and the presence of ectopia lentis (Figure 2, Figure 3, and Figure 4). The mean HCD was 13.37 ± 0.8 mm (range, 12.5-15 mm). A persistent pupillary membrane (PPM) (Figure 2A, B) was noted in 28 of 36 eyes (78%) and ectropion uveae (Figure 2C) in seven of 36 eyes (19%), and in one eye, the pupil could not be visualized due to a hazy cornea. Several of them had nondilating pupils. There was a posteriorly dislocated crystalline lens (Figure 2D) in 12 of 36 eyes (33.3%), an anteriorly dislocated lens (Figure 2E) in two of 36 eyes (5.5%), a subluxated lens (Figure 2F) in 17 of 36 eyes (47.2%), and a subluxated cataractous lens in one of 36 eyes (3%), and 4 of 36 eyes (11%) were aphakic and had lensectomy performed for subluxated lens elsewhere. The central corneal thickness was available in 14 children, and the mean central corneal thickness was 547 ± 103 microns (range, 402–709). The median axial length of 19 eyes was 24.33 (22.74, 25.17) mm. Ten eyes had a shallow anterior chamber (AC), 23 eyes had a deep AC, two eyes had an irregular AC depth, and one had emulsified silicone oil bubbles in AC after vitreoretinal surgery for retinal detachment (RD; Appendix 1). Secondary glaucoma was detected in 72% (26 of 36) eyes showing disc damage (Figure 3A), which was intervened through lensectomy or glaucoma drainage devices (Figure 3B, C); there was no glaucoma in nine eyes; and two eyes were phthisical and hence glaucoma could not be assessed. Retinal pathology was noted in 47% eyes (17 of 36), including RD in seven eyes, peripheral retinal degeneration in eight eyes, and suprachoroidal hemorrhage in two eyes (Figure 4A-C and Figure 2G). The median follow-up of these children was 0.94 years (0.76, 6.51).
Different surgical management strategies, including lensectomy, glaucoma surgeries, and retinal surgeries, were performed according to clinician judgment (Appendix 1).
Thirty-four of 36 eyes underwent pars plana lensectomy (four of these eyes had lensectomy done for ectopia lentis elsewhere), and two eyes were phthisical (one spontaneous RD and one eye phthisical after a primary glaucoma surgery performed elsewhere). The mean age at lensectomy was 4.09 ± 3.5 years (youngest, four months; oldest, 14 years). The mean refractive error after lensectomy was 9.2 ± 3.1 D.
Secondary glaucoma, noted in 26 of 36 eyes, had a mean IOP at presentation of 16.35 ± 10.2 mm Hg, and four eyes had an IOP less than 5 mm Hg (RD or phthisis). The mean cup-to-disc ratio was 0.51 ± 0.28, and nine eyes had >0.8 cup-to-disc ratio. Thirteen eyes with glaucoma required surgery during the follow-up period for IOP control, with a few needing more than one surgery with a mean of 1.46 ± 0.6 surgeries. Types of surgeries included six glaucoma drainage devices, three goniotomies, one combined trabeculectomy with trabeculotomy, one trabeculectomy with mitomycin C, one external trabeculotomy alone, and seven cyclophotocoagulations. At the final follow-up, 19 eyes were on topical antiglaucoma medications with a mean of 1.2 antiglaucoma medications, and the mean IOP was 15.8 ± 9.2 mm Hg.
Seventeen eyes (47%) had coexisting retinal pathologies, with seven of 17 eyes developing RD following lensectomy or a glaucoma procedure. Two of the 17 eyes developed suprachoroidal hemorrhage after undergoing combined trabeculectomy with trabeculotomy, which was mistakenly performed elsewhere due to a misdiagnosis of PCG. One of the misdiagnosed eyes was resolved with conservative management, and vision improved to 1.5 logarithm of the minimum angle of resolution, while the other eye (that had RD as well) became phthisical. Five of 17 eyes with lattice, with or without retinal holes, underwent peripheral retinal barrage laser intraoperatively during pars plana lensectomy with complete or core vitrectomy (pars plana vitrectomy) and did not develop RD until the last follow-up. Three of 17 eyes had peripheral pigmentary chorioretinal atrophic changes and were not treated but were kept under observation. None of the eyes had triamcinolone-guided posterior vitreous detachment induction during vitrectomy. RD after lensectomy occurred one year later in two eyes and two years later in the other two eyes and one year later after glaucoma surgery. All were rhegmatogenous RD, and one had a giant retinal tear.
Pedigree analysis-- The pedigree analysis showed 93% (13 of 14) parental consanguinity in the cohort (Appendix 2). Additionally, none of the children or their parents gave a history of any significant other systemic illness.
Variant analysis-- Using WES, genetic analysis successfully detected disease-related variants in the LTBP2 gene as suspected across all 18 children from 14 families (Table 1). The identified LTBP2 variants were subsequently confirmed through Sanger sequencing in the family (Appendix 3). These genetic alterations were observed in either a homozygous state in 16 cases or a compound heterozygous form in 2 cases.
In 14 families, 12 distinct LTBP2 genetic variants were identified, including five stop-gain variations, six frameshift variations, and 1 substitution variation (Table 1 and Appendix 4). Variations p.Gln329Ter in exon 4 and p.Val586Glyfs*17 in exon 8 were recurrent in two unrelated families and therefore could be potentially hotspot LTBP2 variants. The relatively higher frequency of these variants in a small cohort with prevalent consanguinity suggests they are likely founder variants. However, future reports of these frequent variants would provide clear insights.
No additional gene variants, including CYP1B1 associated with PCG or any other genes involved in anterior segment dysgenesis, were identified except in two cases (P5 and P18) that reported additional heterozygote variants in genes myocilin (MYOC) and nuclear receptor subfamily 0 group B member 2 (NR0B2) genes, respectively (refer to Table 1 for details). The MYOC (NM_000261) gene is associated with glaucoma type 1A and primary open-angle glaucoma [6]. The identified MYOC heterozygous frameshift variant in P5 c.358delG (p.Glu120Serfs*4) was not observed in the 1000 Genomes Database but was seen in the ExAC database at a 0.00002 minor allele frequency. Phenotypically, P5 exhibited central corneal opacity (scar secondary to anteriorly dislocated lens) in the right eye (RE) with extensive staphyloma and secondary glaucoma and a typical LTBP2 phenotype in the left eye. The RE had a posteriorly dislocated lens along with secondary glaucoma, which was managed medically (Appendix 5). The NR0B2 (NM_021969.3) variant c.227delT (p.Phe76Serfs*30) in P18 has been reported to the ClinVar database (submission accession SCV004113399) and is associated with early-onset obesity. Phenotypically, while there was no sign of obesity, P18 had a severe anterior segment LTBP2 phenotype along with bilateral RD, with one eye having a giant retinal tear. P18 underwent surgery for RD with vitrectomy and silicone oil implantation, with its removal after a few months. P18, who developed uncontrolled intraocular pressure with emulsified silicone oil bubbles and advanced glaucoma, received sequential transscleral cyclophotocoagulation, which controlled the IOP, but ultimately lost vision in one eye due to recurrent vitreous hemorrhage and hyphema (Appendix 6).
Among the different protein structural domains of LTBP2 [7], most of the identified variants (6 of 12) were in the calcium-binding epidermal growth factor (EGF)–like domain of LTBP2 (Figure 5). Variants not reported in population genome databases, like the 1000 Genomes Database or the ExAC database, were given novel status, while other variants (with rs ID) had a minor allele frequency less than 0.01, signifying their rare status (Table 1). Among the 12 variants, 11 were classified as likely pathogenic, and one was a variant of unknown significance according to the ACMG criteria (Table 1). Extensive bioinformatic analyses using various in silico tools consistently categorized these variants as pathogenic, underscoring their significant detrimental effects on gene function (Table 2). Interestingly, most of these variants caused the generation of a premature stop codon due to nonsense or frameshift variations (details in Table 2).
For the risk of developing glaucoma, clinical parameters like type of ectopia lentis, HCD, age at lensectomy, presence of retinal pathology, and IOP at presentation were evaluated. In a logistic regression analysis, older age at lensectomy was associated with a greater risk of developing glaucoma (hazard ratio, 1.69; [95% CI: 1, 2.86]; p < 0.05). Other clinical parameters did not show any significant association.
This study of 18 Indian children with LTPB2-related pediatric glaucoma is the largest genetically confirmed LTBP2 group documented to date. Our findings expand the understanding of the LTBP2 classic phenotype, characterized by congenital megalocornea with zonular weakness and lens-related secondary glaucoma, to encompass additional features such as PPM, ectropion uveae, and peripheral RD. Making an initial diagnosis can be tricky in such a spectrum due to the presence of megalocornea and lens subluxation, which can mimic PCG and might lead to a consideration of glaucoma surgeries. However, certain signs, such as the absence of Descemet breaks (Haab striae) despite megalocornea, along with ectopia lentis, iridodonesis, and a whitish persistent pupillary margin, suggest a different clinical picture that needs to be recognized. Our cohort represents a different LTBP2 clinical spectrum other than PCG that is reinforced by the absence of pathogenic variants in the CYP1B1 gene, which is a common cause of PCG. We support our clinical findings with genetic evidence identifying novel LTBP2 pathogenic variants, including stop-gain mutations (50%), frameshift mutations (45%), and substitutions (5%) in this cohort. The annotation and characterization of identified LTBP2 variants using in silico tools further supported their pathogenicity. Furthermore, we provide practical recommendations for managing these patients based on our clinical insights.
LTBP2 is a member of the LTBP protein family, which is crucial for transforming growth factor β activation and regulation. Unlike other LTBP family members, LTBP2 operates independently of transforming growth factor B and is directly involved in stabilizing microfibril bundles and regulating elastic fiber assembly in the extracellular matrix (ECM) [8]. LTBP2 is widely expressed in tissues such as the liver, bone, trabecular meshwork, and ciliary processes [9]. Small interfering RNA–mediated knockdown of LTBP2 in human trabecular meshwork cell cultures altered ECM gene regulation and induced apoptosis [10]. The role of LTBP2 in ocular phenotypes is evident from animal studies as well. LTBP2 knockout studies in animal models have shown weakened ciliary zonules and ectopia lentis [11]. Other phenotypes include altered skin features, fewer tail vertebrae, and longer bleeding times [12]. In a report, a homozygous duplication (insertion) c.5446dupC in LTBP2 was found in affected family members with isolated microspherophakia [13]. Similarly, in three consanguineous families with secondary glaucoma linked to spherophakia or ectopia lentis, affected children in each family carried unique LTBP2 variants: p.Ser338ProfsTer4 (c.1012delT), p.Gln1619Ter (c.4855C>T), and p.Cys1438Tyr (c.4313G>A), respectively [2]. In a previous study, genetic variants in LTBP2 were identified in affected children from two families exhibiting distinct ocular phenotypes, including megalocornea, impaired vision, microspherophakia, ectopia lentis, and myopia, alongside Marfanoid features such as a high-arched palate and tall stature with an unusually large arm span over body height ratio [14]. Notably, there were no Marfan-like features or any other systemic features in our study cohort as of the last follow-up. Eight novel LTBP2 variants were identified in the Chinese congenital ectopia lentis cohort, characterized by features of megalocornea, spherophakia, high myopia, and glaucoma [15]. The clinical spectrum of LTBP2 involvement has also expanded to other diseases, including Weill-Marchesani syndrome and alveolar capillary dysplasia without misalignment of pulmonary veins [16,17]. LTBP2 is also associated with low bone mineral density and may serve as a circulatory biomarker for pulmonary arterial hypertension and acute kidney injury [18,19]. Its overexpression in various cancers, such as nasopharyngeal cancer, esophageal cancer, melanoma, and cervical cancer, has shown prognostic significance [20]. Thus, the range of LTBP2 clinical phenotypes is expanding in both ocular and nonocular forms.
The LTBP2 variants identified in our cohort were novel or rare with low population frequency. Variants c.985C>T and c.1756dupG could be observed in probands from two unrelated families in a consanguineous population, suggesting they could be frequent causative founder variants. Most LTBP2 pathogenic variants in our study were interpreted as likely pathogenic status according to ACMG guidelines, probably due to a lack of previous reports. One case (Cys1413Gly) was primarily classified as a variant of uncertain significance. However, the patient with this variant had a distinct LTBP2 phenotype, and no other gene variants were found. This variant was absent from the 1000 Genomes Database and presented at a low frequency (0.000008) in the ExAC database. Multiple bioinformatics tools, including SIFT, PolyPhen, MutationTaster, and PROVEAN, predicted it to be deleterious.
LTBP2 has EGF-like repeats and four 8-cysteine motifs, similar to fibrillin 1 and fibrillin 2. Most genetic variants in our study were seen in the EGF domain (Figure 5). Interestingly, many of these identified variants were causing a premature stop codon with or without the frameshift of the LTBP2 protein. All of these protein truncation variants (premature termination codons [PTCs]) could activate the nonsense-mediated decay (NMD) pathway (Table 2).
NMD is a crucial quality control mechanism that eliminates fully or partially processed messenger RNA transcripts containing PTCs. The activation of NMD largely depends on the location of the PTC. According to the canonical rule, PTCs situated in the final exon or within 50 nucleotides upstream of the second to last exon generally evade NMD. In our cohort, none of the identified PTCs were located in terminal or preterminal exons, making them direct targets for NMD without escape. If a PTC escapes NMD, the resulting frameshifted, altered, or truncated proteins may retain partial function, exhibit dominant-negative effects, display gain of function, or have no significant impact. In contrast, robust NMD activation leads to the complete degradation of the messenger RNA, resulting in a total loss of the encoded protein and its associated function. While NMD can sometimes mitigate the effects of pathogenic mutations, most human PTC variants show that NMD typically exacerbates the detrimental consequences of truncating variants. Furthermore, studies show that activation of NMD may influence the expression of other normal genes as well [21]. In our cohort, where most PTC variants are predicted to be NMD targets, it suggests that the distinct and severe ocular manifestations associated with the LTBP2 gene phenotype may be driven by NMD activation, leading to a complete loss of protein function. The phenotypic spectrum seen in our cohort in the context of LTBP2 genotype needs to be understood for identifying the potential disease process and mechanism and genotype-phenotype correlations.
The primary pathology in these eyes is likely to be zonulopathy, which predisposes to luxation or dislocation of the lens. This weakening of the zonules has been closely linked to findings in LTBP2 knockout mice, in which the structural organization of the zonules remains intact, but their tensile strength is compromised, leading to their degradation with movement and, consequently, reduced longevity [22]. The clinical manifestation of the displaced lens in our cohort could be seen as early as four months, signifying zonular breakage at an early stage. Therefore, we hypothesize the following: (1) absence of LTBP2 protein results in low-tensile strength zonules; (2) as the human eye undergoes saccades, it causes strain on the nasal and temporal zonules, which start breaking as early as two to three months of age; and (3) this results in ectopia lentis and subluxation.
Microfibrils are an essential component of elastic fibers in body connective tissue, encoded by the FBN1 (fibrillin) gene. Structurally, they are 0.5 to 1.0 microns and have a central core and peripheral surrounding ring cover. FBN1 and FBN2 are present in the inner and outer rings. The interactions between LTBP2 and the FBN1/2 protein are required for the structural integrity and functionality of microfibrils [23]. In the absence of LTBP2 protein, these interactions may be compromised, which may further lead to the weakening of connective tissue in the ciliary zone. This, in turn, can further contribute to lens dislocation, secondary to which iridodonesis may be noted. Also, there could be a possibility of dilator muscle hypoplasia that results in iridodonesis and a small nondilating pupil.
PPM is a common congenital ocular anomaly in neonates, arising from residual tissue of the anterior tunica vasculosa lentis, which nourishes the lens during the first six months of fetal embryonic development. During embryogenesis, the embryologic lens is supplied by the anterior tunica vasculosa lentis network of capillaries. These blood vessels follow the course of the zonules in the channels. They regress within one to two weeks in the postnatal period [24]. Due to the earlier occurrence of zonular loss or breakage attributed to LTBP2 genetic alteration, the zonules break before the vessels regress and may retain the PPM. In the presence of PPM, with time, it can lead to ectropion uveae, as reflected in our LTBP2 cohort. This may also contribute to the nondilating pupil.
The HCD at birth is between 9 and 10.5 mm. HCD >11 mm at birth and diameter >13 mm at any age are considered abnormal and referred to as megalocornea [25]. The enlarged corneal diameter was seen in 100% of patients in our cohort without the presence of Haab striae, indicating the absence of a corneal stretch or Descemet break. Moreover, it is unusual to have secondary megalocornea from increased IOP without Descemet breaks when HCD is >13 mm. The fact that the enlarged HCD in our LTBP2 cohort could not be due to elevated IOP is further supported by the evidence that LTBP2 null mice did not have elevated IOP [22]. Genetic variation in the chordin-like 1 (CHRDL1) gene has been reported to lead to X-linked megalocornea, characterized by enlarged anterior eye segments, mosaic corneal degeneration, presenile cataract, and secondary glaucoma [26]. Megalocornea seen in our LTBP2 cohort may be associated with deep AC, an enlarged ciliary ring, weak zonules, and a small pupil or iris atrophic patches. The weakened connective tissue and ECM component due to LTBP2 alterations can affect the structural integrity of the anterior segment of the eye, including the cornea. This could lead to progressive enlargement of the cornea leading to megalocornea.
The secondary glaucoma seen in our cohort can be caused by pupillary block, angle closure due to intermittent pupillary block, or a dislocated lens. However, the possibility of an inherent angle abnormality cannot be ruled out. The glaucoma did not develop in eyes that underwent lensectomy earlier than 2 years of age in the cohort. This suggests early lensectomy in this cohort prevents secondary glaucoma. Angle surgeries might be helpful when glaucoma occurs in the context of dysgenetic angles. As evident in case P13, the child underwent goniotomy for uncontrolled IOP one year after lensectomy, with open angles and high iris insertion that resulted in well-controlled IOP and stable visual acuity (Appendix 7). However, in established glaucoma, glaucoma drainage devices or transscleral cyclophotocoagulation may be required. Preventing sudden hypotony in these large aphakic eyes is of utmost important when performing any glaucoma intervention.
An important clinical feature noted in our cohort in a significant proportion was peripheral retinal degeneration and RD. We hypothesize that abnormal and stretched peripheral retina associated with retinal thinning and lattice degeneration may predispose to RD. Thorough peripheral retinal examination and treating these lesions would help prevent RD in these eyes. As was also reported earlier, performing primary glaucoma surgery in these eyes results in sight-threatening retinal complications [4,12].
In conclusion, this study highlights our experience with the unique ophthalmic phenotypic spectrum related to LTBP2. Early and appropriate management of secondary glaucoma in these cases involves the removal of the crystalline lens rather than glaucoma surgery. Regular and thorough peripheral retinal examinations, along with prompt treatment, are essential to prevent sight-threatening retinal complications. Additionally, early genetic testing and counseling are beneficial for accurate differential diagnosis, facilitating appropriate management. Further research and similar case studies will enhance understanding and optimize management strategies for these groups of cases.
Appendix 1. Supplementary Table 1.
Appendix 2. Supplementary Figure 1.
Appendix 3. Supplementary Figure 2.
Appendix 4. Supplementary Figure 3.
Appendix 5. Supplementary Figure 4.
We would like to thank Mrinal Singh for assistance in figures and Deepika C Parameshwarappa for her help in managing the retinal problems in these children and for her contribution. Funding: The study is financially supported by Hyderabad Eye Research Foundation (HERF25061) and Hyderabad Eye Institute (HEI), LV Prasad eye institute, Hyderabad, India Competing Interests: “The authors have no competing interests or relevant financial or non-financial interests to disclose.”