Primary Congenital and Juvenile Glaucoma

Primary congenital glaucoma (CYP1B1, LTBP2, TEK, ANGPT1) and juvenile open-angle glaucoma (MYOC, EFEMP1, FOXC1 duplications): dysgeneses of the aqueous humour outflow system that present from birth to the fourth decade of life.

Literature and regulatory information update: 10 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Definitions and scope

Glaucoma is defined as an optic neuropathy with characteristic visual field loss, with or without raised intraocular pressure (IOP). Classification by mechanism distinguishes primary open-angle glaucoma (POAG), resulting from an intrinsic disease of the trabecular meshwork, angle-closure glaucoma, acute or chronic, and secondary glaucomas. In adults the disease is mostly multifactorial and family history is a recognised risk factor; in childhood glaucoma may form part of a broader condition, but it is more often isolated [19]. Childhood and juvenile glaucomas form the group in which the monogenic contribution is highest and in which genetic testing has direct consequences for diagnosis, counselling and surveillance of relatives.

Primary congenital glaucoma (PCG), also called buphthalmos, results from an isolated dysgenesis of the iridocorneal angle, with incomplete development of the trabecular meshwork and outflow system before and/or after birth; resistance to aqueous humour outflow raises IOP and, in an eye whose walls are still distensible, causes globe enlargement and optic neuropathy [1]. The Childhood Glaucoma Research Network (CGRN) classification, adopted by the 6th edition of the European Glaucoma Society (EGS) guidelines, requires for a diagnosis of PCG the absence of other ocular or systemic anomalies and of acquired conditions, the presence of glaucoma according to formal criteria and the presence of buphthalmos; age at onset separates the neonatal (0–1 month), infantile (>1 to 24 months) and late-onset or late-recognised (>2 years) forms [1, 2]. The same classification allows for a "self-limited" form of PCG (self-healed), with typical signs of PCG but normal IOP and a spontaneously non-progressive course [1]. The CGRN classification was approved by consensus at the 9th World Glaucoma Association (WGA) consensus in 2013 and adopted by the American Board of Ophthalmology; it replaced terms such as "developmental", "congenital" or "infantile" glaucoma, considered poorly defined, defines childhood according to national criteria (under 18 years in the USA; 16 years in the United Kingdom, in Europe and for UNICEF) and is applied to each eye separately [20, 21]. Thau and colleagues accept that young children with mild, characteristic angle anomalies and glaucoma, but without buphthalmos, may be classified as PCG, although the flowchart requires buphthalmos for this category [20].

Juvenile open-angle glaucoma (JOAG) is, in the CGRN classification, primary glaucoma of the child without buphthalmos, without congenital ocular anomalies or syndromes, with an open angle and raised IOP; the EGS places typical onset between 4 and 35 years [1, 2]. Thau and colleagues place the earliest onset generally at 4 years, extending to 30 or 40 years, the age from which glaucoma is classified as adult POAG, and describe JOAG as typically autosomal dominant with variable penetrance and expressivity [20]. In the glaucoma genetics literature, JOAG is also defined as the subset of POAG diagnosed before age 40, familial or sporadic, in which MYOC is the most frequent genetic association [12]. The EGS further describes a late-onset juvenile glaucoma, after puberty or in young adulthood, frequently dominant when familial, associated with MYOC, with a widely open and often poorly differentiated angle [1]. The EGS itself stresses that classification by age does not take genetic background into account and reflects prognosis more than pathophysiology [1]: the same CYP1B1 variants can cause neonatal PCG, JOAG or early-onset POAG, and MYOC variants can cause anything from juvenile glaucoma to typical late-onset POAG [4, 22].

This article covers PCG (genes CYP1B1, LTBP2, TEK and ANGPT1, and loci without an identified gene) and JOAG (MYOC, EFEMP1, FOXC1 duplications and CYP1B1). Anterior segment dysgeneses with secondary glaucoma, such as Axenfeld-Rieger anomaly and syndrome (PITX2, FOXC1) and Peters anomaly, and aniridia (PAX6) appear here only as differential diagnoses and where they share genes with PCG and JOAG.

1.2. Historical overview and milestones of molecular discovery

The first mapping of a glaucoma gene was achieved in 1993 by Sheffield and colleagues in a family with autosomal dominant JOAG, with 37 members examined and 22 affected, linked to the interval 1q21–q31 (GLC1A locus) with a maximum lod score of 6.5 [23]. In 1997, Stone and colleagues identified the gene at this locus, which encodes the trabecular meshwork protein then called TIGR (trabecular meshwork-induced glucocorticoid response protein) and later myocilin (MYOC): 13 patients with glaucoma carried one of three variants (3.9% of the population studied), also observed in one control (0.2%) [24]. The following year, analysis of 716 patients with POAG from Iowa found likely pathogenic MYOC variants in 4.6%, with ages at diagnosis from 8 to 77 years and maximum IOP from 12 to 77 mmHg, showing that the same gene spans juvenile glaucoma and typical late-onset POAG [22].

In PCG, Sarfarazi and colleagues mapped the first locus (GLC3A) to 2p21 in 1995, in 17 families, with a combined lod score of 11.50; six families were unlinked, the first evidence of genetic heterogeneity [25]. In 1996 the second locus, GLC3B, was mapped to 1p36.2–p36.1, and four families linked to neither locus indicated at least one further recessive locus [26]. In 1997, Stoilov, Akarsu and Sarfarazi identified CYP1B1 as the main cause of GLC3A-linked PCG, with three homozygous truncating variants in five families and expression of the gene in the trabecular meshwork [27]. In 2009, two independent groups established LTBP2 as a recessive cause of PCG, in consanguineous families from Pakistan and in patients of Gypsy (Roma) ethnicity [28] and in Iranian families [29].

The angiopoietin–TEK axis entered PCG genetics in 2016, when Souma and colleagues described rare loss-of-function TEK variants in 10 of 189 families with PCG and showed, in the mouse, that Tek hemizygosity produces a severely hypomorphic Schlemm's canal and raised IOP [9]. In 2017, the same group showed that angiopoietin-1 is required for development of Schlemm's canal and found rare ANGPT1 variants in 3 of 284 patients with PCG [30]; in 2020, SVEP1 was described as a modifier of TEK-associated PCG [10]. In JOAG, rare EFEMP1 variants were associated in 2022 with familial glaucoma in three Filipino families [7]. In 2019, Siggs and colleagues showed that 6.1% of patients with a clinical suspicion of PCG had FOXC1 variants, all with some Axenfeld-Rieger feature on re-evaluation [13], and in 2026 duplications encompassing FOXC1 were described as the second genetic cause of JOAG, after MYOC, in an Australian cohort [8].

1.3. Nomenclature, classification and status of reference sources

PCG loci follow the GLC3A to GLC3E nomenclature: GLC3A corresponds to CYP1B1 (2p22.2), GLC3B to a 1p36.2–p36.1 locus without an identified gene, GLC3C to a 14q24.3 locus without an identified gene, GLC3D to LTBP2 (14q24.3) and GLC3E to TEK (9p21.2) [4]. GeneReviews gives the cytoband 14q24.2 for GLC3C in one table and 14q24.3 in another; OMIM places it at 14q24.3, in a 2.9 cM interval defined by Stoilov and Sarfarazi (2002) in a five-generation consanguineous Turkish family linked to neither GLC3A nor GLC3B [4, 31]. The GLC3D locus, mapped by Firasat and colleagues (2008) to 14q24.2–q24.3 in two consanguineous families from Pakistani Punjab, is adjacent to GLC3C without overlapping it, and LTBP2 lies about 1.3 Mb proximal to the GLC3C locus [28, 32]. The MYOC JOAG locus is GLC1A (OMIM 137750) [6].

ClinGen classifies as definitive the validity of the associations between CYP1B1 and CYP1B1-related glaucoma with or without anterior segment dysgenesis (autosomal recessive, 2023) and between TEK and TEK-related primary glaucoma (autosomal dominant, 2022); the association of ANGPT1 with PCG is classified as limited (February 2024) [5, 33, 34]. For LTBP2 the association with PCG 3D is definitive [35]. The condition name currently used by the ClinGen glaucoma expert panel is "CYP1B1-related glaucoma with or without anterior segment dysgenesis" for CYP1B1 and "Open-angle glaucoma" for MYOC [36].

The GeneReviews chapter devoted to PCG, published in 2004 and last revised on 17 August 2017, was withdrawn on 2 April 2026 because it was out of date and no authors were available to revise it; it is retained as a historical reference, and its statements have been updated in this article with later sources where these existed (for example, inherited transmission of TEK variants, which GeneReviews described only as de novo) [4, 10]. There is no GeneReviews chapter for JOAG. The European reference guideline is the 6th edition of the EGS guidelines, published in September 2025 [1], complemented by the 2024 CGRN classification system and algorithm [2] and, for genetic testing in JOAG, by the 2025 American Academy of Ophthalmology (AAO) POAG Preferred Practice Pattern (PPP) [14]. A systematic review of guidelines published between 2010 and 2020 found no high-quality guideline developed specifically for childhood glaucoma [37].

2. Molecular Genetics and Pathophysiology

2.1. The iridocorneal angle and the common mechanism

Aqueous humour produced by the ciliary body leaves the eye mainly through the conventional pathway: it crosses the trabecular meshwork, enters Schlemm's canal, a lymphatic-like vessel encircling the limbus, and drains into the episcleral veins. The cornea, iris and trabecular meshwork form the anterior segment, whose development is interdependent, which explains the phenotypic, allelic and genetic heterogeneity of the dysgeneses of this region [38]. In PCG the primary defect is dysgenesis of the trabecular meshwork and angle (trabeculodysgenesis); gonioscopy shows anterior iris insertion forming a scalloped line, persistent uveal tissue and poorly differentiated structures [1]. Reduced outflow raises IOP; in the infant, the still-distensible sclera and cornea enlarge (buphthalmos), Descemet's membrane ruptures (Haab striae) and optic neuropathy sets in [1, 4]. In JOAG the angle is open, but the EGS frequently describes a variable dysgenesis similar to that of PCG, with high iris insertion or iris processes, and the review by Selvan and colleagues reports immaturity of the conventional outflow pathways not always visible on gonioscopy or anterior segment optical coherence tomography (OCT) [1, 12].

The known genes fall into three mechanisms: a developmental defect of the angle with a metabolic basis (CYP1B1), a defect of the microfibrillar network of the extracellular matrix (LTBP2) and a vascular defect of Schlemm's canal through the angiopoietin–TEK axis (TEK, ANGPT1, with SVEP1 as a modifier). In JOAG a mechanism of protein toxicity in trabecular meshwork cells predominates (MYOC and probably EFEMP1), together with the dosage effect of a transcription factor of anterior segment development (FOXC1 duplications).

2.2. CYP1B1

CYP1B1 (2p22.2; OMIM 601771; HGNC:2597) spans about 12 kb, has three exons, of which exons 2 and 3 are coding, and an mRNA of 1,631 bases [4, 33]. It encodes an endoplasmic reticulum cytochrome P450 mono-oxygenase that oxidises steroid hormones, fatty acids and retinoids: it hydroxylates 17β-oestradiol and oestrone at position 4, converts all-trans retinol to retinal and all-trans retinoic acid in extrahepatic tissues, epoxidises arachidonic acid and activates procarcinogens such as polycyclic aromatic hydrocarbons [33]. The most widely accepted mechanistic hypothesis is that the enzyme produces a signalling metabolite, steroid or retinoid, required for angle development [33]. Functional studies by Jansson and colleagues (2001), cited by GeneReviews, showed that NM_000104.4:c.182G>A p.(Gly61Glu) loses about 60% of its stability, that NM_000104.4:c.1405C>T p.(Arg469Trp) retains about 80% and that metabolic activity is reduced by 50–70% for all substrates [4].

Biallelic loss of function of CYP1B1 is the most frequent known genetic cause of PCG [1, 33]. The spectrum includes missense and nonsense variants, small insertions, deletions and duplications, exonic deletions and whole-gene deletions, sometimes extending to adjacent genes; GeneReviews counted almost 200 pathogenic variants in 2017 [4]. A meta-analysis of studies published between 2011 and 2020 pooled 161 variants in 1,641 patients, 78 of them novel, with 64.3% of studies conducted in Asia, 17.9% in Europe, 10.7% in the Americas and 7.1% in Africa; the most frequent variant was c.182G>A p.(Gly61Glu) [39]. The truncating variants described in the founding paper (a 13 bp deletion in exon 3, a cytosine insertion in exon 2 and a larger deletion of the 5′ end of exon 3) correspond to null alleles [27].

The ClinGen glaucoma expert panel reviewed CYP1B1 variants in November 2025. The classic founder variants p.(Gly61Glu), p.(Glu387Lys) and p.(Arg469Trp) and the truncating variants c.1064_1076del and c.868dup were classified as pathogenic; the variant p.(Arg368His), considered prevalent in India, remained a variant of uncertain significance (VUS), which matters for interpreting genotypes of Indian patients and older studies that counted it as causal [36, 40]. The Brazilian variant described as "4340delG" uses a historical genomic numbering. Simões and colleagues identify the most frequent Brazilian allele (20.2%) as A179RfsX18, which corresponds to c.535del p.(Ala179ArgfsTer18) in the MANE transcript [41, 42]; the ClinVar record for this variant (VCV 523943) gives the historical name "4339delG", used in Moroccan patients, and not "4340delG" [36, 43]. As the equivalence with "4340delG" is not confirmed in ClinVar, the historical Brazilian name is kept in quotation marks without conversion to Human Genome Variation Society (HGVS) nomenclature [36, 44].

2.3. LTBP2

LTBP2 (14q24.3; OMIM 602091; HGNC:6715) has 36 exons and encodes a protein of 1,821 amino acids (NM_000428.2; NP_000419.1; the current MANE Select transcript is NM_000428.3, which encodes the same protein), the largest member of the latent TGF-β binding protein (LTBP) family and the one most similar to the fibrillins [4, 35]. It takes part in the latent complex of transforming growth factor β (TGF-β), is a structural component of microfibrils, with which it associates together with fibrillin-1, and is involved in cell adhesion [4, 35]. It is expressed in the trabecular meshwork, ciliary body and ciliary processes, and Ali and colleagues proposed a structural role in ciliary muscle tone [28, 29]. The variants described are null or loss-of-function: four missense variants and small deletions in four consanguineous families from Pakistan and in Roma patients [28], and p.Ser472fsX3 and p.Tyr1793fsX55 in Iranian families [29]; GeneReviews notes that they interfere with binding to fibrillin-1 and fibulin-5 [4]. The mechanism common to LTBP2 phenotypes is disruption of the microfibrillar network of the anterior segment and zonule [35].

2.4. TEK, ANGPT1 and Schlemm's canal

TEK (TIE2; 9p21.2; OMIM 600221; HGNC:11724) encodes a receptor tyrosine kinase expressed almost exclusively in endothelium, activated by the angiopoietins ANGPT1, ANGPT2 and ANGPT4, which signals through the PI3K-AKT and ERK pathways [34]. The longest transcript (NM_000459.4; the current MANE Select transcript is NM_000459.5) is about 4.7 kb long and has 23 exons [4]. The functional direction of the variant determines the phenotype: gain-of-function variants cause venous malformations (the germline variant p.Arg849Trp in familial cutaneomucosal venous malformations and the somatic p.Leu914Phe in unifocal venous malformations), whereas heterozygous germline loss-of-function variants cause PCG type E [34, 45]. The PCG-associated variants described by Souma and colleagues cause haploinsufficiency through absent protein production, aggregate formation, proteasomal degradation, altered subcellular localisation or reduced ligand response [9]. Young and colleagues estimated that TEK haploinsufficiency accounts for about 5% of PCG in diverse populations, with low penetrance attributed to variable dysgenesis of Schlemm's canal [10].

ANGPT1 (8q23.1; OMIM 601667) encodes the TEK ligand. In the mouse, Angpt1 knockout produces a hypomorphic Schlemm's canal and raised IOP; ANGPT2 is dispensable but partially compensates, and the double knockout abolishes the canal [30]. Tissue-specific deletion of Angpt1 or Svep1 in trabecular meshwork cells reproduces PCG in the mouse, showing that the trabecular meshwork signals to the Schlemm's canal endothelium [15]. In patients, ANGPT1 nonsense variants reduce TEK activation; the 2024 ClinGen curation considered the evidence limited, with four or more probands in two publications and only two nonsense variants scored [5]. In OMIM, the only phenotype entry for ANGPT1 is hereditary angioedema with normal C1 inhibitor (type 5, OMIM 619361) [5].

2.5. MYOC

MYOC (1q24.3; OMIM 601652; HGNC:7610) encodes myocilin, a glycoprotein secreted by the trabecular meshwork, induced by glucocorticoids, which regulates cell–matrix adhesion, the cytoskeleton and cell migration through the Rho and Wnt pathways [6]. Pathogenic and likely pathogenic variants cluster in the olfactomedin domain encoded by exon 3 [46]. The mechanism is a toxic gain of function: the misfolded mutant protein is not secreted, accumulates in the endoplasmic reticulum of trabecular meshwork cells, causes reticulum stress and cell death and raises IOP by obstructing outflow [6, 16]. A patient with a complex maternal deletion of chromosome 1 that included the whole MYOC gene did not have glaucoma, nor did her relatives, showing that myocilin haploinsufficiency does not cause glaucoma (Wiggs and Vollrath 2001) [47]. This mechanism explains why reducing expression of the mutant allele, rather than replacing it, is the strategy of the gene therapies under study [48, 49]. More than 250 MYOC variants have been described; at least 93% of the pathogenic ones lie in exon 3 and 84% are missense, although the most frequent, p.(Gln368Ter), is nonsense [12].

The variant p.(Gln368Ter) is the most frequent: among 1,703 patients from five populations it was present in 27 (1.6%), in all groups except the Japanese, and flanking markers suggest a common founder [50]. Unlike the missense variants associated with severe JOAG, p.(Gln368Ter) behaves in the population as a risk allele of low to moderate penetrance (section 3.6) [51].

In 2022 the ClinGen glaucoma expert panel published specifications of the ACMG/AMP (American College of Medical Genetics and Genomics/Association for Molecular Pathology) rules for MYOC: 15 of the 28 criteria were adapted and 13 considered not applicable, with thresholds for allele frequency, proband and segregation counts and assessment of functional assays; in a pilot of 81 variants, classification changed in 40% of those already classified in ClinVar, and functional evidence influenced 18 variants [52]. In a July 2026 preprint, not yet peer reviewed, the same panel curated 271 MYOC variants with updated rules (11 benign, 45 likely benign, 166 VUS, 35 likely pathogenic and 14 pathogenic); clinically definitive classifications rose from 28% to 39%, and functional evidence was lacking for 93% of VUS [46].

2.6. EFEMP1, FOXC1 duplications and other genes

EFEMP1 encodes fibulin-3, an extracellular matrix protein. Collantes and colleagues described three Filipino families with autosomal dominant JOAG in which the variants NM_001039348.3:c.238A>T p.(Asn80Tyr), NM_001039348.3:c.1480T>C p.(Ter494GlnextTer29) and NM_001039348.3:c.1429C>T p.(Arg477Cys) cosegregated; the proposed mechanism is aggregation and intracellular retention of the protein, as with myocilin [7]. Orphanet classifies EFEMP1 as a causal gene of juvenile glaucoma [3], and the EGS includes it among the genes associated with PCG [1].

FOXC1 (6p25) encodes a transcription factor of anterior segment development. Sequence variants and deletions mainly cause the Axenfeld-Rieger spectrum, in which FOXC1 duplications are associated with severe glaucoma and the mean age at presentation of glaucoma is 6 years in patients with FOXC1 and 18 years in patients with PITX2 [53]. In 2026, Maxwell and colleagues described 20 individuals from 10 families with duplications encompassing FOXC1, all of those tested having glaucoma, 17 of whom had been referred as JOAG, one as POAG, one as PCG and one as anterior segment dysgenesis; only two (10.5%) had systemic features, subtle dental features or mild facial dysmorphism [8]. These data place FOXC1 duplications as a cause of apparently isolated JOAG.

The EGS also lists MYOC and EFEMP1 among the genes associated with PCG [1]. GeneReviews describes one individual with PCG heterozygous for variants in MYOC and CYP1B1 (Kaur 2005, suggesting digenism) and a Chinese family with POAG and PCG in which homozygous MYOC was proposed as a cause of PCG (Zhuo 2006) [4]. The GLC3B (1p36) and GLC3C (14q24) loci still have no identified gene [4].

2.7. Normalised variants

The variants cited in this article were normalised in VariantValidator (GRCh38, MANE Select transcripts, Matched Annotation from NCBI and EMBL-EBI) and checked against ClinVar in October 2026 [36, 42].

The 13 bp deletion that Della Paolera and colleagues designated g.7901_7913del13 in historical genomic numbering has no alias in ClinVar confirming correspondence with c.1064_1076del, and the two designations are therefore not treated here as equivalent [36, 54].

2.8. Genotype–phenotype correlations, penetrance and modifiers

In CYP1B1 PCG there is no consistent genotype–phenotype correlation: identical variants produce different severities between and within families, and the two eyes of the same patient may be asymmetric [4]. Studies comparing patients with and without variants reached opposite conclusions. In Brazil, among 30 unrelated patients, the nine with CYP1B1 variants had more surgeries to control IOP (p=0.003 for both eyes; p=0.011 for the worse eye) and bilateral disease in 9/9, against 11/21 in the others (p=0.013) [54]. In China, among 238 Han patients, variant carriers (17.2%) had earlier onset (median 2 months against 6 months) and, among the 192 operated on, the mutated genotype and poorer corneal clarity were associated with better IOP control after surgery [55]. In Saudi Arabia, patients with variants had more postoperative corneal clouding and greater need for medication [4]. Comparison of cohorts from India (301 patients with PCG analysed) and Brazil (150) showed more homozygotes in India (p=0.024), more compound heterozygotes in Brazil (p=0.012), only six shared variants and no correlation with sex, consanguinity, IOP or corneal diameter at presentation [40]. In Brazil, the "4340delG" variant was associated with early bilateral glaucoma requiring multiple surgeries [44]. In Portugal, three series found bilateral disease in all patients with biallelic CYP1B1 variants: in Lisbon (21 children), 100% against 46.7% without variants (p=0.017), with diagnosis at birth (0 against 4.5 months; p<0.01) and no differences in postoperative IOP, number of reoperations or final visual acuity [56]; in Coimbra (30 patients), 100% against 62.5% (p=0.014), with higher IOP at diagnosis (34.6 against 29.7 mmHg; p=0.034) [41]; and in Porto (71 patients), 100% against 68% (p<0.001), with earlier onset (median 0 against 5.5 months), more surgeries (median 2 against 1), more medications, higher final IOP and worse final visual acuity [57]. Among the Roma, homozygosity for the LTBP2 founder variant NM_000428.3:c.895C>T p.(Arg299Ter) (R299X in OMIM nomenclature) was associated with a more severe phenotype and worse outcome despite more surgeries, with a spectrum ranging from PCG with primary trabecular dysgenesis to a Marfan-like zonular disease with ectopia lentis and later secondary glaucoma (Azmanov 2011) [32].

Penetrance of PCG due to biallelic CYP1B1 is incomplete. In 61 Saudi families, Bejjani and colleagues (2000) found, in 22 families, 40 apparently unaffected individuals with the same variants and haplotypes as their affected siblings; two were later diagnosed with glaucoma and another two had findings compatible with mild forms, and the analysis suggested a dominant modifier locus not linked to CYP1B1 [43]. Among the Roma of Slovakia, by contrast, inheritance was described as recessive with complete penetrance, whereas in non-Roma families studied in the same region 26.6% of patients had unilateral disease, later onset and a milder course (Gencik 1980) [43]. The review by Leysen and colleagues brings together studies in which biallelic carriers did not develop PCG and others in which heterozygous carriers did, and describes identical twins homozygous for the same CYP1B1 variant, one with PCG and the other with JOAG; the authors interpret CYP1B1-associated PCG, JOAG and POAG as a continuum dependent on residual enzyme activity [58]. The EGS describes recessive inheritance with variable penetrance in most cases [1]. For LTBP2 and TEK no genotype–phenotype correlation has been described; patients with LTBP2 may have ectopia lentis, a high-arched palate and mild to moderate osteopenia, findings not observed with CYP1B1 or TEK, and TEK PCG has variable severity and age at onset [4]. In one family with TEK PCG of greater penetrance and severity, the SVEP1 variant NM_153366.4:c.2989C>T p.(Arg997Cys) (p.R997C in the publication) was present in four of five affected individuals, and histology of one eye showed absence of Schlemm's canal; SVEP1 was proposed as a modifier [10].

In CYP1B1 JOAG, Vincent and colleagues (2002) described a Canadian family with adult POAG and JOAG in which all affected individuals carried the MYOC variant; those who also carried the CYP1B1 variant had JOAG, with a mean age at onset of 27 years, against 51 years in carriers of the MYOC variant only, and carriers of the CYP1B1 variant only were unaffected, suggesting CYP1B1 as a modifier of MYOC [4, 47]. Rarely, a homozygous CYP1B1 genotype alone causes JOAG (Acharya 2006), but in Saudi Arabia 12 of 14 unrelated patients with JOAG had at least one CYP1B1 variant (eight homozygous for p.(Gly61Glu), two compound heterozygous p.(Gly61Glu)/NM_000104.4:c.1294C>G p.(Leu432Val) and two heterozygous for p.(Gly61Glu) without a second variant), and none had MYOC or LTBP2 variants (Abu-Amero 2015) [4]. In French patients with POAG, CYP1B1 was described as a risk factor for early onset and as a modifier in patients without a MYOC variant (Melki 2004) [4]; in the same work, of four sisters compound heterozygous for CYP1B1 variants, two had PCG and two had POAG with onset at 35 and 40 years [43]. In a Spanish patient with PCG homozygous for a CYP1B1 variant inherited only from the father, segregation of chromosome 2 markers showed paternal uniparental isodisomy (López-Garrido 2009) [43].

The MYOC-associated phenotype varies within the same family, from individuals without abnormalities to ocular hypertension and severe glaucoma (Cobb 2002) [19]; in a pair of monozygotic twins carrying the risk haplotype, one had severe glaucoma diagnosed at 19 years and the other only mildly raised IOP with normal fields (Wiggs 1995) [47]. Penetrance depends on the variant: in pedigrees with p.(Pro370Leu) it reached 100% by 30 years, against about 50% for NM_000261.2:c.1348A>T p.(Asn450Tyr) and p.(Gly367Arg) and 25% for p.(Gln368Ter), according to unpublished data cited in the review by Selvan and colleagues [12]. Variants at Asp380 give a picture intermediate between the severe forms associated with p.(Pro370Leu) or NM_000261.2:c.1267A>G p.(Lys423Glu) and the milder form of p.(Gln368Ter) [47]. In the UK Biobank, the prevalence of disc-defined glaucoma among p.(Gln368Ter) carriers increased with each decile of a POAG polygenic risk score, and diagnosed cases had higher scores (adjusted odds ratio [OR] of 27.5 for the top decile) [59]. In the angiopoietin–TEK axis, a POAG-protective polymorphism in the 3′ untranslated region (3′UTR) of ANGPT2 (rs76020419, T allele) increased ANGPT2 expression and Schlemm's canal area in the mouse [60], and common variants of ANGPT1 and SVEP1 are POAG risk alleles in genome-wide association studies [15].

2.9. Allelic disorders

3. Clinical Spectrum and Manifestations by System

3.1. Primary congenital glaucoma: ocular manifestations

PCG is characterised by raised IOP, globe enlargement (buphthalmos), corneal oedema and opacification with breaks in Descemet's membrane (Haab striae), thinning of the anterior sclera, iris atrophy, an abnormally deep anterior chamber and a structurally normal posterior segment, apart from progressive glaucomatous optic atrophy [4]. Manifestations are confined to the eye; PCG has no systemic involvement, and the presence of extraocular anomalies or other structural ocular anomalies points to secondary glaucoma (section 6) [2]. The classic data of deLuise and Anderson (1983) and of Ho and Walton (2004), compiled by GeneReviews and taken up by the EGS, indicate a male predominance (65%) and bilateral disease in 70% of patients [1, 4].

The classic symptoms are epiphora, photophobia, blepharospasm, eye rubbing and irritability; parents may notice cloudy or unusually large corneas, and corneal enlargement generally occurs before age 3 [1, 4]. The disease is not always symptomatic [1]. Epithelial oedema results from the rise in IOP, and Haab striae, curvilinear lines horizontal or concentric to the limbus, record breaks in Descemet's membrane caused by corneal stretching. Optic disc cupping typically appears only after some months and may regress after effective treatment [1]. Increased axial length causes myopia, and myopia and anisometropia frequently cause amblyopia [1, 33].

3.2. Forms by age at onset

The neonatal form (from birth to 1 month) is the most severe: the newborn may present with corneal opacity, increased corneal diameter, raised IOP and an enlarged globe [4]. In a series of 35 newborns with PCG described by Walton (1998), corneal oedema was present in every eye, diffuse in 90% and localised in 10% [4]. In some populations the neonatal form is more associated with CYP1B1 than the other forms, but the subtypes do not correspond to a specific genetic cause [4]; in a Chinese cohort, the median age at onset was 2 months in patients with CYP1B1 variants and 6 months in the others [55]. The infantile form is the one recognised clinically between 1 month and 2 years (Walton and Katsavounidou, 2005) [2, 4]. In the late-onset or late-recognised form, after 2 years, IOP rises gradually, corneal oedema and opacity are less evident and progressive globe enlargement generally no longer occurs after 3 to 4 years, when the sclera loses distensibility [4].

3.3. Untreated natural history and course with treatment

Without treatment, PCG progresses to blindness; GeneReviews states that blindness invariably occurs in untreated patients [4]. PCG is an important cause of childhood visual impairment, with a greater burden in low- and middle-income countries [33]; in the Indian state of Andhra Pradesh it accounts for about 4.2% of childhood blindness, and in Saudi Arabia and among the Roma of Slovakia it is the most frequent cause of blindness in childhood [4].

With treatment, the visual outcome depends on severity at diagnosis, other ocular anomalies, response to surgery and IOP control during follow-up; the earlier the clinical onset, the worse the prognosis [4]. Infants with raised IOP and cloudy corneas at birth have the worst prognosis, and forms that are severe at birth may reach legal blindness despite multiple surgeries, owing to corneal opacity, glaucomatous damage and amblyopia; mild forms with late presentation often respond to a single operation, with an excellent visual prognosis [4]. A postoperative IOP below 19 mmHg was associated with better visual function [4]. Unlike in adults, pressure-induced disc cupping in the infant is reversible, especially in the early stages, which is attributed to the elasticity of optic nerve tissues [4]. In an Indian cohort of 1,223 eyes of 775 children operated on between 1993 and 2023, with a minimum follow-up of one year, cupping reversal (reduction in cup-to-disc ratio of at least 0.2) occurred in 43.2% of eyes; each additional millimetre of baseline corneal diameter reduced the probability of reversal by 50% (adjusted OR 0.50; 95% confidence interval [CI] 0.42–0.59), and neonatal (adjusted OR 2.03; 95% CI 1.04–3.93) and infantile onset (2.45; 1.30–4.60) were associated with more reversal than late onset [63].

3.4. Differences between genes in PCG

CYP1B1 PCG is recessive, mostly bilateral, and its surgical course varies inconsistently between populations (section 2.8). In LTBP2 PCG, besides glaucoma, there may be ectopia lentis, microspherophakia, megalocornea, a high-arched palate and mild to moderate osteopenia [4, 35]. In TEK PCG severity and age at onset are variable, there are unaffected carriers, and one eye may be severely affected while the fellow eye is normal; the structural defect is hypoplasia or absence of Schlemm's canal [4, 9, 10]. In the original series of 10 families, the disease was unilateral in five and bilateral in the other five, and 8 of 22 carriers (36%) did not have the typical early-onset PCG phenotype [64]. The patients described with ANGPT1 variants had onset before age 3, and clinical data are scarce [5, 30]. Patients with FOXC1 variants who had been classified as PCG showed, on re-evaluation, at least one ocular or systemic Axenfeld-Rieger feature [13].

3.5. Juvenile open-angle glaucoma

JOAG is frequently asymptomatic until there is advanced visual field loss, and findings are often asymmetric [1]. There is no enlargement of the globe or cornea and there are no congenital ocular anomalies; the angle is open, IOP is raised, and gonioscopy may show high iris insertion or iris processes [1]. The review by Selvan and colleagues describes high IOP, frequent need for surgery, mean age at diagnosis in the third decade, male predominance, frequent myopia and large discs with deep, steep cupping; progression rates are comparable to those of adult primary glaucoma, but cumulative lifetime visual disability is greater because of the early age [12]. The EGS estimates that 40–70% of patients with JOAG need surgery [1]. The review by Selvan and colleagues adds that JOAG is usually bilateral, but up to 25% of patients present with unilateral disease; that the proportion of familial cases ranges from 28% in Korea, 30% in India and 43% in the USA to as much as 88% in African populations; that 42–72% of patients are myopic; and that patients generally present at a moderate to advanced stage, with an initial mean deviation below −8.5 dB [12]. Structural and functional progression over five years was 4.7% and 7.1% in an Indian series, and functional progression was 28% over seven years in a Korean series; raised IOP, its fluctuation and higher myopia were associated with progression [12].

MYOC-associated JOAG is autosomal dominant with frequently very high IOP, and large JOAG pedigrees were central to linkage to MYOC [65]. In the original 1q-linked family, mean age at diagnosis was 18 years, IOP at first examination was often above 50 mmHg, the angle was open without abnormal pigmentation or embryonic tissue, and eye drops initially controlled IOP but surgery was eventually needed (Johnson 1993); in five pedigrees linked to 1q21–q31, the 23 affected individuals were diagnosed at a mean age of 18.5 years (5–30), with a mean initial IOP of 38.5 mmHg (30–53), myopia in 87% and need for surgery in 83% (Wiggs 1995) [47]. MYOC variants were much more frequent in familial (27%) than in sporadic JOAG (2%) in an Indian series, and account for proportions of JOAG ranging from 5.8% in Chinese cohorts to 34% in Brazil [12]. For the six recurrent MYOC variants in the Iowa series, age at diagnosis ranged from 8 to 77 years and maximum IOP from 12 to 77 mmHg [22]. Families with JOAG have more affected relatives than those with adult POAG, but many patients have no known family history [19]. In EFEMP1-associated JOAG, the 34 affected carriers in the three Filipino families had a mean age at onset of 16 years, and 76% became blind [7]. In FOXC1 duplications, penetrance was high (all carriers tested had glaucoma) and systemic Axenfeld-Rieger features were rare and subtle [8].

3.6. The MYOC p.(Gln368Ter) variant in the population

The p.(Gln368Ter) variant illustrates the difference between risk estimated from patient registries and population risk. In the UK Biobank (411,337 white British participants recruited between 2006 and 2010), the variant was present in 1 in 786 participants (0.13%) and was associated with POAG with an OR of 6.76 (95% CI 4.05–11.29); penetrance was 7.6% for glaucoma, 24.3% for ocular hypertension and 30.8% for the two combined, and about half of carriers aged 65 years or older had glaucoma or ocular hypertension [51]. In Australian glaucoma registries (3,071 cases and 6,750 historical controls), the OR was 12.16 (6.34–24.97) for advanced glaucoma and 3.97 (1.55–9.75) for non-advanced glaucoma, and estimated penetrance rose to 56.1% (69.5% including suspects); the authors attribute the difference to ascertainment bias in the registries [51]. In an analysis of 77,959 UK Biobank fundus photographs, 45 of the 177 carriers with gradable photographs (25.4%) had probable or definite disc-defined glaucoma in at least one eye, 69% of them without a previous diagnosis, and most carriers had IOP of 21 mmHg or lower [59]. These data indicate that, in a p.(Gln368Ter) carrier, IOP alone cannot be used to exclude glaucoma and that disc imaging and perimetry are needed.

4. Clinical Recognition and Warning Signs

4.1. In the infant and young child

PCG should be suspected in an infant with photophobia, epiphora, blepharospasm or eye rubbing, especially if the child is tearful and irritable in the first weeks or first year of life, and in any child with large eyes or corneas, a cloudy cornea or Haab striae [1, 4]. GeneReviews adds as a suggestive criterion the absence of structural anterior chamber changes compatible with anterior segment dysgenesis or systemic disease [4]. The epiphora and red eye of PCG are confused with congenital nasolacrimal duct obstruction, conjunctivitis and corneal abrasion; the presence of photophobia, a cloudy cornea or increased corneal diameter makes it mandatory to exclude glaucoma [1, 4]. When faced with corneal clouding in a newborn it is essential to exclude congenital glaucoma, along with other malformations such as Peters anomaly and congenital aphakia due to FOXE3 [66]. A clear megalocornea with otherwise normal parameters is distinguished from buphthalmos, in which the cornea is cloudy [38].

Situations of increased risk include siblings of a child with PCG, children of consanguineous parents and children from populations with CYP1B1 founder variants (section 9). GeneReviews recommends evaluating at-risk siblings in the neonatal period [4].

4.2. In the older child, adolescent and young adult

JOAG has no early warning signs: it is frequently asymptomatic until there is advanced field loss and may be asymmetric [1]. It should be suspected in a young person with markedly raised IOP, especially if there is myopia and a dominant family history of early-onset glaucoma, and in a young person with increased or asymmetric disc cupping [1, 12]. The EGS recommends offering MYOC testing to families with several members affected by glaucoma at a relatively young age [1]. When PCG or JOAG is suspected, subtle Axenfeld-Rieger features should be sought: posterior embryotoxon, iris hypoplasia, corectopia, dental anomalies, midface hypoplasia and redundant periumbilical skin [13, 53].

4.3. Urgent signs

An infant with a cloudy cornea, a corneal diameter increased for age (at least 11 mm in the newborn, more than 12 mm in the first year, more than 13 mm at any age) and Haab striae are grounds for urgent referral to a paediatric ophthalmologist experienced in glaucoma [1, 2]. Surgery should not be delayed in an attempt at medical control of IOP [4]. In a patient with JOAG, very high IOP with progressive field loss, as in the case in section 10, may require urgent surgery [67].

4.4. Assessment by the geneticist

Genetic assessment includes a family history of at least three generations and enquiry about consanguinity; growth (short stature in chromosomal anomalies, in Peters plus syndrome and in Rieger syndrome, the latter with possible pituitary anomalies); psychomotor development; and examination directed at orofacial clefts (Peters plus, Kivlin), dental anomalies, hypospadias and redundant periumbilical skin (Rieger) and at changes in the sense of smell (PAX6) [19]. On ophthalmological examination it is important to exclude iris hypoplasia, aniridia, sclerocornea, megalocornea, microphthalmia and lens anomalies, and to record whether involvement is unilateral or bilateral; bilateral involvement may only be detected on examining the apparently normal eye [19, 38]. The history should include teratogens and infections in pregnancy [38]. The parents should be examined for minimal signs of anterior segment dysgenesis [19, 38].

4.5. Newborn screening

The sources consulted describe no population newborn screening programme specific to PCG. Detection relies on clinical suspicion, examination of newborns with corneal signs and targeted assessment of siblings and other at-risk children [1, 4].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic criteria for childhood glaucoma

The diagnosis of PCG and JOAG is clinical. The CGRN criteria, reproduced by the EGS, require at least two of the following items to establish a diagnosis of glaucoma in a child; a single item defines a glaucoma suspect [1, 2].

GeneReviews used as a clinical criterion an IOP above 21 mmHg in one or both eyes, measured by applanation, rebound tonometry or pneumotonometry on at least two occasions, and gave the normal IOP in a child as 12.02 ± 3.74 mmHg (Sihota 2006) [4]. The EGS insists that the IOP value alone is insufficient to diagnose PCG: a child's IOP fluctuates with context and the normal threshold is not comparable to that of the adult [1]. The most reliable values are obtained in a calm, awake and cooperative child, with a handheld tonometer; crying gives falsely high values and most anaesthetics give falsely low values; rebound tonometry is valid in children and can be used for home tonometry; epithelial oedema underestimates IOP and conditions that increase corneal rigidity overestimate it; digital palpation can serve as an estimate [1]. Central corneal thickness should be measured, because it may be increased even without oedema [1]. The EGS recommendation can be summarised as not deciding on IOP alone, looking for structural corneal changes and measuring axial length, which is needed to detect and monitor the disease [1]. The 9th World Glaucoma Association consensus (WGA, 2013) already defined childhood glaucoma as IOP-related ocular damage and considered globe enlargement, Haab striae and the cup-to-disc ratio more important than the IOP value in the infant [21]. The consensus points add that all inhaled anaesthetics lower IOP, sometimes rapidly and markedly, whereas chloral hydrate, ketamine and midazolam appear not to lower it, and that the same anaesthetic should be used for serial examinations; that central corneal thickness should not be used to correct IOP in the child; that increased corneal diameter is the cardinal sign of glaucoma in the infant and is useful for diagnosis and follow-up up to age 3; that gonioscopy should be performed at least once; that an axial length outside normal limits strongly suggests glaucoma and its continued growth suggests insufficient treatment; and that no child should be labelled as having glaucoma or operated on without reasonable diagnostic certainty and exclusion of mimicking conditions [21]. The EGS recommendation to measure central corneal thickness and the WGA consensus recommendation not to use it to correct IOP are not contradictory: the first serves to interpret the measurement, the second advises against correction formulae [1, 21].

5.2. CGRN classification

Source: [2, 20]; the EGS reproduces the same classification and adds the category of self-limited PCG [1]. According to Thau and colleagues, glaucoma following cataract surgery takes precedence over other assignments, even if the eye has buphthalmos and Haab striae; glaucoma due to an acquired mechanism resulting from a congenital anomaly, such as angle closure due to microspherophakia, retains the classification of glaucoma associated with a non-acquired condition; and each eye is classified separately [20]. Retrospective application of the classification to the records of children with glaucoma or suspected glaucoma reduced 26 different diagnoses to the seven CGRN categories, with no overlap between them (Hoguet 2016) [20]. In the definition published by Thau and colleagues, a glaucoma suspect requires IOP >21 mmHg on two occasions, which confirms the error in the EGS table [1, 20].

5.3. Examination under anaesthesia and ophthalmological investigation

In an infant with suspected PCG, examination under anaesthesia or sedation of both eyes includes IOP measurement in the first minutes of anaesthesia, corneal diameter, anterior segment examination, direct gonioscopy to exclude secondary glaucoma, dilated fundoscopy with optic nerve assessment, ultrasound biomicroscopy (UBM) or anterior segment OCT when the cornea is opaque, and axial length [4]. Cycloplegic refraction documents myopia [33]. Prior consent may be obtained to perform the appropriate surgery under the same anaesthetic, and the genetics clinic is part of the initial assessment [4]. In the adolescent and young adult with JOAG, assessment includes gonioscopy, tonometry, pachymetry, OCT imaging of the disc and retinal nerve fibre layer, and perimetry [1, 59].

5.4. Indications for genetic testing

The EGS recommends genetic testing in PCG to exclude other congenital anomalies, to facilitate counselling on family planning for patients and parents and to help in disease management, and strongly recommends it in secondary childhood glaucoma because of the large overlap of phenotypes; these are consensus recommendations without formal grading [1]. In open-angle glaucoma, the EGS recommends offering MYOC testing, in liaison with clinical genetics, to individuals from families with several members affected at a relatively young age, and advises against routine genotyping of all adult glaucoma patients, direct-to-consumer tests as a basis for clinical decisions and the use of polygenic scores, none of which is validated [1]. The AAO, in the 2025 PPP, does not recommend routine genetic testing for risk alleles in POAG, with the exception of testing the children of patients with JOAG, and also considers that no polygenic score is validated for clinical use [14]. The two positions agree in essence; the EGS centres the indication on the family with several young affected members and the AAO on the offspring of the patient with JOAG. The 2013 WGA consensus considers genetic assessment especially important when there is a known genotype–phenotype correlation, accepts carrier testing in at-risk relatives when both causal alleles of the patient have been identified, proposes considering genetic screening of first- and second-degree relatives in families with dominant MYOC JOAG, and stresses that rigorous ophthalmological diagnosis, including examination of parents and siblings, is the basis of counselling [21].

CYP1B1 testing is not needed to diagnose PCG, which is clinical, but it establishes the aetiology and enables carrier testing, reproductive counselling and targeted screening in populations with founder variants; a negative result does not exclude PCG [33, 58]. GeneReviews accepted molecular testing as confirmation when the clinical picture is inconclusive: biallelic pathogenic or likely pathogenic variants of CYP1B1 or LTBP2, or a heterozygous loss-of-function TEK variant; a VUS neither confirms nor excludes the diagnosis [4].

5.5. Testing strategy

The rationale for including FOXC1 and CNV analysis in PCG panels comes from the study by Siggs and colleagues: of 131 individuals from Australia (84) and Italy (47) with PCG suspected by the specialist, 8 (6.1%) had rare deleterious FOXC1 variants, or 8 of 166 (4.8%) counting those already explained by CYP1B1, and all had at least one ocular or systemic Axenfeld-Rieger feature on re-evaluation [13]. In JOAG, probands with a genetic diagnosis in the Australian registry had FOXC1 duplications in 13.5% (95% CI 6.7–25.3%), behind MYOC (53.8%; 40.5–66.7%), and in the Massachusetts Eye and Ear cohort in 9.5% (2.7–28.9%); the authors recommend routine CNV analysis in JOAG [8]. In the EGS, the prevalence of MYOC variants is 2–4% in patients with POAG and rises to 16–40% when patients with young onset, high IOP and a strong family history are selected [1].

5.6. Diagnostic yield by gene

Sources: [4, 10, 13, 30]. The GeneReviews percentages in the last two columns refer to the proportion of pathogenic variants of each gene detected by each method, and not to the clinical sensitivity of the test for PCG, which depends mainly on the proportion of patients with an identifiable genetic cause in the population studied [4]. The probability of finding CYP1B1 variants increases with family history, consanguinity, bilateral disease and severity, and sampling differences between studies make the estimates imprecise [4].

5.7. Limitations of testing

A negative test does not exclude PCG or JOAG: some patients have no identifiable cause among the known genes, the GLC3B and GLC3C loci still have no gene, and panel or exome sequencing may miss exonic or whole-gene deletions of CYP1B1 and FOXC1 duplications if CNV analysis is not validated [4, 8, 58]. Sequencing does not detect deep intronic variants outside the regions analysed or balanced rearrangements. Parental mosaicism for a de novo TEK variant may not be detected in blood [4]. In CYP1B1 PCG, identification of a single heterozygous variant is not sufficient for the diagnosis of a recessive disease and requires a search for the second variant by CNV analysis and, if necessary, by methods covering non-coding regions; the case in section 10.2 illustrates this situation [4, 68]. In MYOC JOAG, a whole-gene deletion does not cause glaucoma (the mechanism is toxic, not haploinsufficiency), so CNV analysis is not useful for MYOC, unlike CYP1B1 and FOXC1 [47]. For TEK, the laboratory must indicate the functional direction of the variant, because gain of function causes venous malformations and loss of function causes PCG [34]. Genotypes that include CYP1B1 p.(Arg368His), now classified as a VUS by the expert panel, should not be considered diagnostic [36]. Interpretation of MYOC variants should follow the specifications of the ClinGen glaucoma expert panel; most curated variants remain VUS for lack of functional evidence [46, 52].

5.8. Report content and interpretation

The report should state the gene, the versioned reference transcript (for example NM_000104.4 for CYP1B1 and NM_000261.2 for MYOC), HGVS nomenclature at DNA and protein level, zygosity, the ACMG/AMP class with the criteria applied, the method and regions covered, whether or not CNV analysis was included and, for TEK, the expected functional direction [34, 42]. In recessive diseases, the phase of the two variants should be established by parental testing; in dominant diseases with incomplete penetrance (TEK, MYOC), segregation analysis in affected and unaffected relatives contributes to classification, but a carrier without glaucoma does not exclude pathogenicity [10, 51]. VUS should be re-evaluated periodically, because expert panel revisions change classifications [46, 52].

5.9. Prenatal and preimplantation diagnosis

Prenatal diagnosis and preimplantation genetic testing are possible once the causal variants in the family have been identified [4, 19]. Second-trimester ultrasound has low sensitivity for anterior segment malformations [38]. Counselling aspects are set out in section 8.

6. Differential Diagnosis

6.1. Comparative table

6.2. Differential diagnosis of PCG in the infant

The first step is to distinguish glaucoma from non-glaucomatous causes of epiphora, photophobia and corneal enlargement or clouding. Congenital nasolacrimal duct obstruction and conjunctivitis cause epiphora and red eye, but do not increase corneal diameter or produce Haab striae [1, 4]. X-linked megalocornea gives large, clear corneas, with normal IOP, axial length and disc [38]. In 18 patients with CHRDL1 variants, the cornea was large and thin, without oedema or breaks in Descemet's membrane, and the anterior chamber was deeper than in PCG; a ratio of anterior chamber depth to total axial length above 0.19 on ultrasound makes megalocornea very likely, except in extreme axial myopia (Davidson 2014) [43]. CHED gives bilateral corneal opacification with generally normal diameter and IOP, but may coexist with congenital glaucoma, which requires careful tonometry and biometry [4]. The EGS further includes in the differential posterior polymorphous dystrophy, birth trauma, ocular hypertension and large optic discs [1].

The second step is to distinguish PCG from glaucomas secondary to non-acquired ocular anomalies, especially anterior segment dysgeneses. About half of patients with anterior segment dysgenesis develop glaucoma [19]. In Axenfeld-Rieger anomaly and syndrome the risk of glaucoma is 50–60%, generally in the second or third decade and occasionally congenital, with a mechanism similar to that of PCG (late arrest of angle development); forms with few extraocular manifestations are sometimes classified and treated as PCG [53]. GeneReviews notes that severe or advanced PCG may be difficult to distinguish from some dysgeneses, such as Peters anomaly [4]. Peters anomaly is characterised by central corneal opacity with iridocorneal or lenticulocorneal adhesions and absence of central endothelium and Descemet's membrane; it is bilateral in 80% of cases, asymmetric, and associated glaucoma is very frequent [38]. In Peters plus syndrome, autosomal recessive and due to B3GLCT, about 70% of patients are homozygous for the splicing variant NM_194318.4:c.660+1G>A, and rhizomelic short stature, cleft lip (45%), cleft palate (33%) and heart disease (<30%) are associated [69]. Aniridia combines iris and foveal hypoplasia with nystagmus, and glaucoma appears later [4].

The third step is the search for systemic disease or syndrome. The EGS lists trisomy 21, connective tissue disorders (Marfan, Weill-Marchesani, Stickler), metabolic disorders (homocystinuria, Lowe syndrome, mucopolysaccharidoses), phakomatoses (neurofibromatoses 1 and 2, Sturge-Weber, Klippel-Trenaunay-Weber) and congenital rubella [1]. In Sturge-Weber syndrome, glaucoma presents before age 2 if the tissues destined to form the angle are affected, and frequently only after age 5 if only the conjunctival and episcleral vessels are involved; heterochromia, with the darker iris on the side of the glaucoma, marks the cases that will develop it [70]. The EGS considers alpha-2 agonists contraindicated in children with a compromised blood–brain barrier, as in the leptomeningeal angiomatosis of Sturge-Weber [1]. Trisomies 13 and 18, Walker-Warburg syndrome and Zellweger syndrome are also associated with childhood glaucoma [4]. The congenital glaucomas described occasionally in SHORT, MIDAS, otopalatodigital type II, Melnick-Needles and Frank-ter Haar syndromes are accompanied by extraocular anomalies that distinguish them [71–75]. The 6p25 deletion, which includes FOXC1, combines Rieger anomaly, congenital glaucoma and posterior embryotoxon with developmental delay, heart disease, deafness and dysmorphism [62]. In congenital Zika syndrome, congenital glaucoma is among the ocular abnormalities, together with chorioretinal scars, optic nerve hypoplasia and pallor, congenital cataract, lens subluxation and iris coloboma [76].

6.3. Differential diagnosis of JOAG

JOAG is distinguished from late-recognised PCG by the absence of globe or corneal enlargement, and from secondary glaucomas by the absence of congenital ocular anomalies, syndromes and acquired causes [1]. Glaucomas associated with acquired conditions (uveitis, trauma, corticosteroids, tumours) are excluded by history and examination, and an important genetic differential is Axenfeld-Rieger with few signs, especially in families with a FOXC1 duplication, in which systemic features are present in only about 10% [1, 8]. In Weill-Marchesani syndrome, glaucoma results from pupillary block due to microspherophakia, and increased central corneal thickness may lead applanation to overestimate IOP; these patients have a paradoxical response to miotics and mydriatics [61]. In nail-patella syndrome, open-angle glaucoma is associated with absent or abnormal nails and requires surveillance for nephropathy [19].

7. Treatments and Risk Management Strategies

7.1. Principles

The aim of treatment is to lower IOP to prevent corneal opacification, globe enlargement and optic atrophy; early treatment reverses some of the complications, including disc cupping [1, 4]. PCG is almost always treated surgically: initial surgery is indicated in almost all cases, and medical treatment is usually neither effective nor practicable in the long term [1, 4]. Surgery should not be delayed in an attempt at medical control [4]. In JOAG, medical treatment is more effective than in PCG, but 40–70% of patients eventually need surgery [1]. IOP follow-up is lifelong, and refractive correction and orthoptic follow-up are an integral part of treatment, because anisometropic amblyopia remains an important cause of visual morbidity even with controlled IOP [1]. The EGS recommends that these children be referred to tertiary centres with experience, volume and a multidisciplinary team [1].

7.2. Surgery in primary congenital glaucoma

Surgery seeks to eliminate the outflow resistance caused by the structural anomalies of the angle [4]. The EGS and the CGRN algorithm recommend angle surgery as first choice, goniotomy or trabeculotomy, ab interno or ab externo, with trabeculotomy preferred when the cornea is cloudy [1, 2].

The 2013 WGA consensus considers angle surgery, conventional or circumferential, the procedure of choice in PCG, with the choice dictated by corneal clarity and the surgeon's experience; it points out that the first operation is often the best opportunity; it notes that initial combined trabeculotomy-trabeculectomy is proposed by some as more effective in certain populations, without prospective comparisons to confirm this; it places trabeculectomy or a drainage device as the next procedure after failure of angle surgery; and it considers that diode laser cyclophotocoagulation has limited long-term success and often requires repetition and continued medication [21]. The EGS advises against other operations in childhood glaucoma, and trabecular minimally invasive glaucoma surgery (MIGS) devices should not be used in congenital glaucoma [1]. The CGRN management algorithm proposes: treat medically; if the eye is suitable for angle surgery, perform it (goniotomy or trabeculotomy) while continuing medication; if IOP is not controlled or the eye is not suitable, trabeculectomy with or without antifibrotic or a drainage device, depending on findings and surgeon preference; diode laser cyclophotocoagulation if there is failure or limited visual potential; refraction and aggressive amblyopia treatment in all cases; and, when there is visually significant corneal opacity, assessment of the clear corneal periphery, retinoscopy after dilatation, optical iridectomy or corneal surgery [2].

GeneReviews gave success rates of about 80% for each initial procedure, a better outcome when surgery is performed between the second and eighth month of life and reduced effectiveness in preserving vision with increasing age [4]. More than one intervention is often needed, and surgical success may decline over time, which justifies lifelong surveillance [4, 33]. The indicators of postoperative course are media clarity, control of ocular dimensions (corneal diameter and axial length) and the state of the optic nerve [4]. In LTBP2 eyes, circumferential trabeculotomy has the advantage of opening the entire angle in one procedure, and in case of failure trabeculectomy or a drainage device is used [35].

7.3. Comparative evidence for surgery

The evidence base is limited. None of the trials in the Cochrane review reported quality of life or costs, most were conducted in a single region, follow-up ranged from 6 to 80 months and the authors concluded that an international multicentre trial is needed [77]. The 2026 meta-analyses combine randomised trials with retrospective studies, with moderate to very high heterogeneity (I² of 55% and 97.2%), and their success ORs should not be read as absolute differences without the baseline rate of each study [79, 81]. Results of series with different populations, definitions of success and durations should not be compared indirectly; definitions of complete success (without medication) and qualified success (with medication) vary between studies. The Turkish cohort shows that the Ahmed valve after failed trabeculotomy has low five-year survival [80], and the GATT series in MYOC JOAG has no control group [78]. A systematic review of guidelines found no high-quality guideline specific to childhood glaucoma; the available recommendations rest on low- to moderate-quality evidence or expert opinion [37]. In France, a prospective multicentre study (24 centres) on the diagnosis and treatment of congenital glaucoma in children under 4 years, with a primary outcome of visual acuity at age 4 and an estimated n of 250, completed primary data collection in November 2021 without results posted in the registry [82].

7.4. Medical treatment in childhood

In PCG, ocular hypotensive drugs are used before surgery, to protect the optic nerve, avoid sudden decompression of the globe and clear the cornea for examination and surgery; after surgery, until its success is confirmed; while the surgical approach is being decided or other options are being prepared after failure; and when surgery carries a risk to life or control is incomplete [1, 4]. Carbonic anhydrase inhibitors (CAIs) and beta-blockers are the first-line choices in infants with congenital glaucoma when medical treatment is considered [1]. Systemic absorption of eye drops is proportionally greater in children: the UK summary of product characteristics (SmPC) for latanoprost describes systemic exposure to latanoprost acid about twice as high between 3 and 12 years and about six times as high below 3 years as in adults [83].

GeneReviews and the EGS single out alpha-2 agonists as the class with serious adverse effects in infants and young children, and GeneReviews recommends avoiding them in children [1, 4]. The paediatric dose of acetazolamide comes from the EGS and not from the label, which in the USA does not establish safety or efficacy in childhood [1, 93]. The latanoprost and travoprost SmPCs point out that, in children under 3 years who mainly suffer from PCG, surgery (trabeculotomy or goniotomy) remains the first-line treatment [83, 85]. The UK SmPCs for timolol and dorzolamide and the US labels for pilocarpine and echothiophate iodide were consulted [88, 90, 96, 97]. The pilocarpine label gives paediatric safety and efficacy as established, but recommends caution in PCG because of the paradoxical rises in IOP described; the echothiophate label confirms the interaction with suxamethonium that justifies stopping it before general anaesthesia [4, 96, 97].

7.5. Paediatric drug trials

The two prostaglandin analogue trials showed no significant difference from timolol, and efficacy below 3 years rests on very small samples: in the latanoprost trial, only 13 patients aged 0 to <3 years received latanoprost, with no relevant efficacy in the 4 patients under 1 year; in the travoprost trial, 4 of 9 children aged 2 months to <3 years had no relevant IOP reduction [83, 85]. The brinzolamide trial was negative for efficacy [92]. These results reinforce the guideline position that, in infant PCG, drugs are a bridge to surgery.

7.6. Comparative regulatory status

Sources: [1, 83–95, 98, 99]. Access to and reimbursement of these eye drops in Portugal (INFARMED) and Brazil (ANVISA) are not covered in the sources consulted.

7.7. Treatment of juvenile open-angle glaucoma

In JOAG, medical treatment with any effective topical regimen may control IOP for some time, but surgery is needed in 40–70% of patients and often early [1]. Angle surgery (trabeculotomy, goniotomy) or combined angle and filtering surgery give good results; filtering surgery carries a higher risk of hypotony in younger patients, and drainage implants are an option in refractory cases; the use of antimetabolites is considered [1]. Selective laser trabeculoplasty (SLT) is not recommended in JOAG, which the EGS lists, together with angle dysgenesis, among the contraindications to this laser [1]. This position distinguishes JOAG from adult POAG, in which NICE guideline NG81 places 360° SLT as first line [6]. The review by Selvan and colleagues presents a different position: older studies of argon laser trabeculoplasty in JOAG showed poor response and IOP spikes, but in one SLT study more than 70% of young patients with JOAG or juvenile ocular hypertension achieved success at 12 months, with a mean IOP reduction of 30%, and in another, as secondary therapy, 43% of eyes had a reduction greater than 20%, with a better prognosis in the absence of angle dysgenesis; the authors conclude that SLT may not be a contraindication in JOAG [12]. The same review notes that carriers of MYOC variants generally respond less well to drugs and laser, which justifies early surgery [12]. The EGS recommends referral of these patients to tertiary centres [1].

Surgical results in JOAG come from uncontrolled series compiled by the same review: goniotomy with 53% complete success and a further 23.5% qualified success after a mean of 9 years; microcatheter circumferential trabeculotomy with 90–100% success at 15–43 months in two studies and 53% at 36 months in another, with transient hyphaema in up to 20%; trabeculectomy with 50–90% success at 3 years, 80% complete success at 5 years and 52% at 10 years, with early bleb failure in 20% and hypotony in 20–25% with MMC; and Ahmed and Baerveldt drainage devices with 90.7% success at 1 year [12]. Up to 50% of patients in an Indian series maintained IOP control with drugs for 5 years, but visual field deterioration was more evident in those treated medically than in those operated on; alpha agonists are generally avoided below 10 years [12].

In MYOC-associated JOAG, a series of eight eyes of four children with a strong family history treated with 360° gonioscopy-assisted transluminal trabeculotomy (GATT) achieved a mean IOP reduction of 26 mmHg (68%), discontinuation of all medication and no loss of efficacy or progression over 2.2 to 4.1 years, with visual acuity of 20/20 or better and one case of cupping reversal; the authors describe it as the first genotype-directed glaucoma surgery [78]. The rationale is that outflow resistance in MYOC glaucoma is located in the trabecular meshwork, which GATT opens around the whole circumference. It is a series without a control group, and does not allow GATT to be recommended as routine in genetic JOAG. The case in section 10.1 illustrates the course of MYOC JOAG treated with trabeculectomy with MMC [67].

7.8. Refraction, amblyopia and visual rehabilitation

Axial myopia and anisometropia are frequent in PCG and JOAG [12, 33]. Refractive errors and amblyopia must be treated to obtain the best visual function, and correction of myopia, amblyopia treatment and low-vision rehabilitation form part of the therapeutic plan [4, 33]. The EGS recommends regular orthoptic follow-up and refractive correction, because anisometropic amblyopia remains a cause of visual morbidity with controlled IOP, and access to vision therapy, mobility training, school support and support groups for carers [1].

7.9. Perioperative care and agents to avoid

Drugs such as echothiophate iodide (Phospholine Iodide®), a cholinesterase inhibitor, should be stopped before surgery, especially if suxamethonium is to be used, because of the risk of prolonged apnoea [4]; the US label states that suxamethonium should be administered only with great caution, if at all, to patients treated with anticholinesterases, because of the risk of respiratory or cardiovascular collapse [97]. Alpha-2 agonists should be avoided in children because of the risk of apnoea and bradycardia [4, 94]. IOP measured under anaesthesia is as a rule falsely low with most anaesthetics, and should be measured in the first minutes of anaesthesia [1, 4]. In Weill-Marchesani syndrome and LTBP2 microspherophakia, miotics and mydriatics should be avoided, because they may induce pupillary block; management includes early removal of the ectopic lens, peripheral iridectomy and trabeculectomy in advanced chronic angle-closure glaucoma, and ocular trauma should be avoided [35, 61].

7.10. Pregnancy and breastfeeding

A woman with JOAG or with PCG treated in childhood who is planning a pregnancy should discuss the treatment strategy before and during pregnancy, including alternatives such as laser or surgery [1]. No ocular hypotensive agent is specifically approved in pregnancy or breastfeeding, and the level of evidence on fetal risk is low [1]. According to the EGS, CAIs are contraindicated, especially in the first trimester, because of their teratogenic potential; prostaglandin analogues increase the risk of miscarriage; brimonidine may cause apnoea in the infant and should not be used in late pregnancy or during breastfeeding; beta-blockers and pilocarpine have considerable experience and are generally considered safe [1]. The US labels for pilocarpine and echothiophate state that there are no animal reproduction studies and that they should be used in pregnancy only if clearly needed; in breastfeeding, the echothiophate label asks for a choice between stopping breastfeeding or the drug, because of the risk of serious reactions in the infant [96, 97]. As IOP may fall in pregnancy, temporary discontinuation of treatment may be considered with close surveillance; if treatment is needed, the lowest effective dose is used and systemic absorption is reduced with punctal occlusion and eyelid closure [1]. In breastfeeding, prostaglandin analogues may be acceptable, netarsudil has negligible systemic exposure but no clinical data, and CAIs and beta-blockers may be used according to the American Academy of Pediatrics [1]. In advanced glaucoma, the Valsalva manoeuvre in labour transiently raises IOP, and caesarean section may be considered in severe or end-stage cases after multidisciplinary discussion, although the evidence is limited [1].

7.11. Transition to adulthood

PCG and JOAG require lifelong IOP follow-up [1, 4]. Transition from paediatric to adult care should ensure continuity of surveillance, genetic counselling of the affected or carrier young adult on the risks to offspring and reproductive options, and information on drugs to avoid in pregnancy [1, 4]. GeneReviews recommends storing DNA from probands without a molecular diagnosis, for future reanalysis [4].

7.12. New developments and research

The search carried out for this update in Europe PMC (September 2025 to October 2026) and ClinicalTrials.gov identified no drug trials in PCG and no regulatory applications or decisions specific to PCG or JOAG; this negative result does not demonstrate the absence of ongoing studies. The only human study of MYOC-directed gene therapy identified is suspended, and the other gene-directed approaches remain at the preclinical stage [16, 17, 49].

7.13. Guideline recommendations and divergences

Sources: [1, 2, 14, 21]. For JOAG, the WGA consensus considers medical treatment first line depending on age, although surgery is often needed, and the evidence on the best initial operation weak; for drugs, it notes that they rarely have sustained efficacy as primary treatment in PCG, that systemic absorption should be reduced with eyelid closure and nasolacrimal occlusion, that the lowest frequency and concentration should be used, that adverse effects may be atypical (nocturnal cough with beta-blockers) and that brimonidine should be avoided in young children [21]. The systematic review by Lingham and colleagues shows that guidelines prior to 2020 were not developed specifically for childhood glaucoma and rest on low- to moderate-quality evidence or expert opinion [37]. The ERN-EYE network had no guideline of its own on congenital glaucoma or JOAG on its guidelines page in October 2026 [18].

8. Reproductive Options and Family Counselling

8.1. Autosomal recessive PCG (CYP1B1, LTBP2)

The 2013 WGA consensus describes PCG as usually autosomal recessive, with a family history in 10–40% of cases and more frequent in consanguineous populations, and considers clinical screening of current and future siblings essential when there is parental consanguinity [21]. The parents of a child with PCG due to biallelic CYP1B1 or LTBP2 variants are obligate heterozygotes, asymptomatic and not at risk of developing PCG [4]. In each pregnancy, the risk for a sibling is 25% of being affected (subject to the caveat of incomplete penetrance), 50% of being an asymptomatic carrier and 25% of inheriting neither variant [4]. The children of a patient are obligate carriers, and the parents' siblings have a 50% probability of being carriers; detection of carriers in the family requires knowledge of the familial variants [4]. Consanguinity increases the risk of PCG: incidence may be 5 to 10 times higher with consanguineous parents [1], and in 22 Arab countries consanguinity was present in 45–100% of patients [102]. In populations with a founder variant, targeted testing for that variant is the first step in carrier testing [4, 39]. When only one parent carries a variant found in the homozygous state in the child, uniparental isodisomy is a possible explanation, already described in CYP1B1 PCG, and it changes the recurrence risk to values close to that of the population [43]. The review by Leysen and colleagues points out that the 25% risk applies only when pedigree and variant analysis demonstrate a typical recessive pattern; in other cases an exact percentage cannot be given, the risk is higher in familial than in sporadic cases, and in families with a proven pattern the risk of JOAG and POAG in older relatives should also be mentioned [58].

There are two caveats. A single CYP1B1 variant in a patient with PCG does not establish the diagnosis, and the recurrence risk cannot be calculated reliably without the second variant (section 5.7). And CYP1B1 heterozygotes may have an increased risk of early-onset POAG, according to data from French patients, and the variant may modify the severity of MYOC glaucoma [4]; the sources consulted do not quantify this risk for carriers.

8.2. Autosomal dominant PCG (TEK, ANGPT1)

GeneReviews, revised in 2017, described all probands with TEK PCG known up to then as carriers of de novo variants, and indicated that, if the variant is not detected in the parents' leucocytes, the risk to siblings is about 1% because of germline mosaicism [4]. Later data show autosomal dominant transmission with incomplete penetrance and variable expressivity: the original study already described dominant transmission with variable expressivity, and Young and colleagues described eight families with TEK PCG, two with father-to-son transmission [9, 10]. Counselling should therefore test both parents; each child of a carrier has a 50% probability of inheriting the variant, and the probability that the carrier will develop glaucoma is below 100% and is not quantified in the sources consulted [10]. In an asymptomatic carrier of a TEK variant, ophthalmological surveillance of the children is the practical consequence. For ANGPT1, transmission data are insufficient [5].

8.3. Autosomal dominant JOAG (MYOC, EFEMP1, FOXC1 duplications)

MYOC JOAG is transmitted in an autosomal dominant manner: each child of a patient has a 50% probability of inheriting the variant, with variable penetrance and expressivity, greater for missense variants associated with JOAG than for p.(Gln368Ter) [6, 51]. Predictive testing is possible when the familial variant is known, and serves to direct surveillance [19]. The EGS gives the example of a young unaffected member of a family with a MYOC variant: if the test is negative for the familial variant, he or she does not have the increased risk attributable to that variant (retaining the general population risk of glaucoma); if positive, he or she should have close surveillance and early treatment; and counselling should explore motivation, process and the impact of the result [1]. The AAO recommends genetic testing of the children of patients with JOAG [14].

Cascade testing brings diagnosis forward. In Australia and New Zealand, among 73 carriers of MYOC variants, those identified by cascade testing were, at first examination, 83% without glaucoma and 17% suspects, whereas those who came through the usual clinical route were 44% suspects, 28% with glaucoma and 28% with advanced glaucoma; those identified by cascade testing presented earlier (40.6 ± 12.5 vs 47.5 ± 16.7 years), with lower maximum IOP (17.6 vs 32.2 mmHg), lower cup-to-disc ratio (0.48 vs 0.65) and lower field mean deviation (−1.2 vs −10.0 dB), differences that persisted after adjustment for p.(Gln368Ter) [103]. Among 40 unaffected individuals who underwent predictive MYOC testing, the motivations were to obtain appropriate interventions and to reduce uncertainty, all perceived medical or emotional benefit, and informing their children was a strong motivation [104]. Familial risk is measurable even outside Mendelian forms: in the Rotterdam study, cited by the AAO, the odds of POAG were 9.2 times higher with a first-degree relative with confirmed POAG [14], and in the Baltimore Eye Survey the age-adjusted ORs were 3.69 with an affected sibling, 2.17 with a parent and 1.12 with a child [19].

Families with JOAG due to EFEMP1 and FOXC1 duplication follow the same dominant model, with high penetrance in FOXC1 duplications [7, 8].

8.4. Testing of minors

In PCG, GeneReviews recommends testing at-risk siblings soon after birth when the familial variants are known, to establish the diagnosis early and avoid repeated examinations under anaesthesia in children who have not inherited the genotype; in siblings with the at-risk genotype, and in families without an identified variant, clinical surveillance is maintained (section 11) [4]. In MYOC JOAG with onset before 18 years in the family, testing of minors is justified by the direct medical benefit: in three Australian families with six at-risk children, the parents chose to test half, none of those tested was a carrier, and the decision factors were the age at onset and severity in the family and the age of the child; testing may allow early detection or discharge from routine clinics [105]. The age at onset in the family is thus one of the central elements in the decision to test a child [105].

The review by Leysen and colleagues proposes a counselling and follow-up protocol for relatives of a proband with PCG: a pedigree with all types of open-angle glaucoma (PCG, JOAG, POAG); a panel with all known genes, also in sporadic cases, explaining that a negative result does not exclude a genetic cause; genetic testing of newborn siblings; in children with a positive test, four glaucoma examinations with IOP measurement and optic nerve assessment in the first year, then every 6 months for 2 years and then annually if stable, with examination under anaesthesia (IOP, axial length, pachymetry, optic nerve) if suspicion arises; annual examination of the parents (fundoscopy, IOP, automated perimetry and optic nerve biometry) because of the risk of late-onset glaucoma; and examination of second- and third-degree relatives every 2 years or at a longer interval depending on findings, preferably with an examination in the first year of life, and more frequently in consanguineous families [58]. This is an authors' proposal based on a literature review, not a society guideline.

8.5. Families without a molecular diagnosis

Without an identified genotype and without consanguinity, the empirical recurrence risk of PCG for siblings is about 5% [19]. In these families, exclusion of PCG in a sibling requires clinical screening, including IOP measurement under anaesthesia or sedation when necessary [4]. The proband's DNA should be stored for future reanalysis [4]. The parents and siblings of a patient with childhood glaucoma should be examined for minimal signs of anterior segment dysgenesis, which would change the inheritance model [19, 38]. In first-degree relatives of patients with juvenile glaucoma, ophthalmological surveillance is recommended, and the starting age depends on the family history [19].

8.6. Prenatal and preimplantation diagnosis

Once the causal variants in the family have been identified, prenatal diagnosis and preimplantation genetic testing are possible [4]. Most glaucomas are treatable with surveillance and early diagnosis, but some parents wish to have prenatal diagnosis in PCG, because the visual outcome may be poor despite treatment [19]. Views on prenatal diagnosis of a treatable disease differ between individuals and families, and counselling should present them non-directively [4]. Planning should take place before pregnancy, and affected young adults, carriers or those at risk of being carriers should receive genetic counselling on the risks to offspring and reproductive options [4]. In dominant diseases with incomplete penetrance (TEK, MYOC), identification of the variant in an embryo or fetus does not predict whether glaucoma will occur or its severity [10, 51].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency of PCG: what each figure measures

Estimates of PCG frequency vary more than tenfold depending on the source, because they measure different things. The EGS gives about 1 in 12,000–18,000 births in white Europeans [1], and GeneReviews a birth prevalence of 1 in 5,000–22,000 in Western countries [4]; GeneGuidance gives 1 in 10,000–20,000 live births in Ireland, the United Kingdom and the USA [35]. In Spain, the Spanish Collaborative Study of Congenital Malformations (ECEMC), with 1,124,654 consecutive births, found a birth incidence of about 3/100,000 (Bermejo and Martínez-Frías, 1998, cited by the Oxford Desk Reference) [19]. OMIM gives an incidence between 1 in 30,000 and 1 in 1,250 [43]. Orphanet, based on the regional population-based congenital anomaly registries of the EUROCAT network between 2000 and 2016, gives a birth prevalence of 2.2/100,000 births in Europe (about 1 in 45,000) [11]. EUROCAT data come from regional population-based congenital anomaly registries and do not constitute an exhaustive national prevalence; the difference from clinical estimates may reflect the way the registries ascertain cases, and the values should not be read as the prevalence of each country [11].

Source: Orphanet, based on EUROCAT data (congenital glaucoma of all causes; validated values) [11]. Orphanet has no validated point prevalence value for Europe (class 1–9/100,000, not validated) [11]. These values do not transfer to other regions.

Outside Western Europe, birth prevalence is 1 in 2,500 in the Middle East (in Saudi Arabia, according to GeneGuidance), 1 in 1,250 among the Roma of Slovakia and 1 in 3,300 in the Indian state of Andhra Pradesh [4, 35]. These high frequencies reflect consanguinity and CYP1B1 founder variants.

9.2. Proportion of PCG attributable to CYP1B1 and founder variants

The proportion of PCG due to CYP1B1 ranges from about 20% to 100% depending on the population, with the highest frequency in consanguineous populations [4, 29]. The 2021 meta-analysis proposes testing the founder or common variants of each population first [39]. In the systematic review of the 22 Arab countries, among 361 patients and 88 families, 77 causal variants were identified, 33 exclusive to Arab populations: 69 in CYP1B1, 5 in MYOC and one in each of the genes NTF4, FOXC1 and WDR36 [102]. In Brazil, CYP1B1 variants were twice as frequent in affected individuals of European ancestry as in those of African ancestry, and most were associated with the 5′-C-C-G-G-T-A-3′ haplotype, which contains the high-activity V432 allele [44].

9.3. JOAG and MYOC by population

The EGS estimates the prevalence of JOAG at about 1 in 50,000 [1], close to the point prevalence of 2.0/100,000 in the USA given by Orphanet, whose source is a review that does not present the figure in its abstract [3, 65]. In the analysis of 1,703 glaucoma patients from five populations, likely pathogenic MYOC variants were present in 4.3% in Iowa (31/727), 2.6% in African Americans from New York (8/312), 2.8% in Japan (3/107), 3.0% in Canada (5/167) and 2.8% in Australia (11/390), a total of 3.4%; 76% of the variants were present in only one population [50]. Among probands with JOAG and a genetic diagnosis in the Australian registry, MYOC accounted for 53.8% [8]. In Nigeria, in a case–control study with 45 patients with JOAG, the pathogenic variant p.(Pro370Leu) was present in 3 (6.7%) and the prevalence of pathogenic MYOC variants was estimated at 8.9%; the authors conclude that MYOC is the most common known cause of JOAG in populations of European ancestry and in sub-Saharan Africa, but accounts for a minority of cases [106]. In POAG in general, MYOC accounts for 3–5% of cases [6, 59]. In the Philippines, EFEMP1 was identified in three families with JOAG [7]; in Saudi Arabia, 12 of 14 patients with JOAG had CYP1B1 variants [4].

9.4. Portugal and Brazil

In Portugal, birth prevalence in the EUROCAT registries from 2000 to 2016 was 0.9/100,000 [11]. Three hospital series studied CYP1B1. In Lisbon (Hospital de D. Estefânia), 6 of 21 children (28.6%) had biallelic variants, all compound heterozygous, with seven variants, two novel [56]. In Coimbra (Centro Hospitalar e Universitário de Coimbra), 22 of 30 unrelated patients (73.3%) had biallelic variants, 10 homozygous and 12 compound heterozygous; the most frequent allele was A179RfsX18, that is, c.535del p.(Ala179ArgfsTer18), associated with the ancestral CGGTA haplotype, followed by p.(Glu387Lys), and segregation confirmed recessive inheritance in the families studied; the authors attribute the difference from Lisbon to the greater stability of the population of the central region [41]. In Porto, 43 of 71 patients (60.6%; 66 probands and 5 relatives) had two plausibly causal variants, among 19 variants, four novel; the most frequent were c.535del (43.5%) and c.1200_1209dup (28.2%), and the panel in patients without biallelic variants found three heterozygous TEK VUS [57]. The variant p.(Gly61Glu), described by the 2021 meta-analysis as recurrent in Portugal, appeared in a single patient in the Lisbon series, does not appear among the nine variants of the Coimbra series and is not among the two most frequent in Porto [39, 41, 56, 57]. Simões and colleagues note that A179RfsX18 is also the most frequent allele in Brazil and Morocco, and relate this distribution to Portuguese expansion [41]. The Porto data come from the abstract and public excerpts of the paper. The Centro Hospitalar e Universitário de Coimbra is a member of ERN-EYE [18]. In Brazil, a series of 52 patients with PCG found CYP1B1 variants in 50%, with "4340delG" in 20.2% of individuals screened [44]; another, of 30 patients, found variants in 30% and an association with more surgeries and bilaterality [54]; the comparison with India showed more compound heterozygotes in Brazil [40]. These data come from tertiary centres and do not estimate national prevalence.

9.5. Special populations

Consanguinity increases the incidence of PCG 5 to 10 times and is a factor of severity and poor prognosis [1]. In India and the Middle East, patients present with more corneal oedema and megalocornea, which contributes to the preference for primary combined trabeculotomy-trabeculectomy in those centres [33]. PCG is more frequent in males (65%) [1], and JOAG also shows a male predominance [12]. In low- and middle-income countries, late diagnosis contributes to the burden of PCG in childhood blindness [4, 33]. In studies of MYOC p.(Gln368Ter), the variant was not found in the Japanese group [50].

9.6. European reference network and patient associations

ERN-EYE, the European reference network for rare eye diseases coordinated by the University Hospitals of Strasbourg, has a working group on rare paediatric ophthalmology (WG3) and another on rare anterior segment diseases (WG4), cross-cutting groups on genetic diagnosis, registries and low vision, virtual case consultation (CPMS), online training and patient representatives (ePAG) [18]. The Childhood Glaucoma Research Network provides the classification, algorithms and resources for families [107]. The CGRN international paediatric glaucoma registry (Robison D. Harley, MD CGRN International Pediatric Glaucoma Registry), hosted at Wills Eye Hospital, uses the CGRN classification as a common language, had participants from six continents in 2018 and is linked to a DNA bank for genetic eye diseases, where participants can deposit samples from patients with glaucoma [20]. Patient associations and support organisations include Glaucoma UK [108], the Glaucoma Research Foundation in the USA [109], the Children's Glaucoma Foundation and the Grajewski Lyra (GL) Foundation for Children with Glaucoma [4]. The sources consulted did not identify Portuguese or Brazilian associations specific to childhood glaucoma.

10. Illustrative Clinical Case

10.1. MYOC-associated juvenile open-angle glaucoma with reversal and return of cupping

El Helwe and colleagues described a 25-year-old Puerto Rican woman with severe-stage MYOC-associated JOAG, diagnosed at age 20, seen at Massachusetts Eye and Ear in February 2018 for blurred vision and intermittent pain in the left eye [67]. She had a strong family history, with several affected first-degree relatives; the MYOC variant NM_000261.2:c.1153G>A p.(Glu385Lys), described as novel, was present in her affected father, sister and brother and absent in unaffected relatives, and was identified in a research protocol [67]. The review by Selvan and colleagues, which predates publication of the case, already included p.Glu385Lys among the MYOC variants described in one or a few families, which contradicts the designation as a novel variant [12]. Normalisation in VariantValidator confirms the correspondence between c.1153G>A and p.(Glu385Lys) in the MANE transcript (NC_000001.11:g.171636287C>T); the paper gives the protein incompletely in the text of the case [42, 67].

The right eye had undergone trabeculectomy with MMC in 2014, successfully, and the left eye SLT in early 2018, without a significant fall in IOP; the maximum recorded IOPs were 40 mmHg in the right eye and 46 mmHg in the left [67]. In February 2018, corrected visual acuity was 20/25 in the right eye and 20/30 in the left; IOP was 10 mmHg in the right eye without medication and 46 mmHg in the left on dorzolamide twice daily, timolol twice daily, brimonidine twice daily and bimatoprost at night. There was diffuse thinning of the neuroretinal rim, with a cup-to-disc ratio of about 0.90 in the right eye and 0.80 in the left; the visual field showed constriction with a central island in the right eye and a superior arcuate defect larger than the inferior one in the left; OCT showed diffuse thinning of the retinal nerve fibre layer (RNFL), with a thickness of 71 µm in the left eye and a mean deviation of −9.83 dB [67].

Urgent trabeculectomy with MMC 0.4 mg/ml was performed in the left eye in February 2018. On the first postoperative day IOP was 5 mmHg; there was a limbal leak, treated with a bandage contact lens and two bleb revisions, and a single episode of IOP of 2 mmHg without hypotony maculopathy (normal macular OCT in April 2018) [67]. At 8 months (October 2018), IOP was 5 mmHg without medication and the cup-to-disc ratio had decreased to 0.45, with an increase in mean neuroretinal rim thickness (cupping reversal; Figure 3), while the RNFL fell from 71 to 56 µm and then stabilised [67]. In May 2019, an IOP spike of 23 mmHg was accompanied by the return of cupping, with a cup-to-disc ratio of 0.70 (Figure 4); she started timolol in the left eye, and the ratio remained at 0.70 until January 2022, with RNFL between 52 and 58 µm and a stable visual field (maximum change of −1.32 dB from baseline, within the test–retest variability of ±2.44 dB) [67]. At the last examination, in January 2022, visual acuity was 20/30 and IOP 7 mmHg on timolol twice daily in the left eye [67].

The authors present the case as the first description of reversal and return of cupping in an adult with MYOC JOAG, with structural changes without functional correspondence in the visual field or RNFL [67]. The case illustrates several points of this article: the dominant family history with segregation of the variant, the very high IOP with advanced field loss in a young patient, the lack of response to SLT, which the EGS advises against in JOAG, and the need for filtering surgery [1]. The patient was receiving brimonidine; in adulthood this drug does not have the paediatric restrictions described in section 7.4. Cupping reversal, recognised as a sign of effective treatment in paediatric glaucoma, was here partial and transient, and subsequent surveillance relied on IOP, OCT and visual field together [63, 67]. The management described is that of a single case and does not constitute a protocol.

10.2. Primary congenital glaucoma with one CYP1B1 variant: a case with an incomplete molecular diagnosis

Abu-Amero and Sesma described in 2026 a boy referred at 2 months to a tertiary centre in Riyadh for photophobia, epiphora and bilateral corneal opacification, without family history and with normal development [68]. On examination under anaesthesia there were bilateral Haab striae and diffuse corneal oedema, IOP of 35 mmHg in the right eye and 30 mmHg in the left before intubation (27 mmHg in both after induction), a horizontal corneal diameter of 13 mm in both eyes, central corneal thickness of 737 and 731 µm and, on ultrasound, mild to moderate cupping and bilateral posterior staphyloma (Figure 1, bilateral buphthalmos with corneal oedema, more marked on the right) [68]. The difference between IOP before and after induction illustrates the lowering effect of anaesthesia described by the EGS [1].

Four operations were needed over 24 months: bilateral deep sclerectomy with MMC at 2 months; at 7 months, in the right eye (IOP 35 mmHg, corneal diameter 15 mm, cup-to-disc ratio 0.6), the planned repeat procedure was converted to microcatheter-assisted 360° ab externo trabeculotomy with MMC; at 18 months, in the left eye (IOP 24 mmHg, diameter 14 mm, ratio 0.5), repeat deep sclerectomy with MMC [68]. Axial length went from 21.31/21.63 mm at 2 months to 26.27/23.40 mm at 8 months, with axial myopia of −10.00 D in the right eye, up to a maximum of 27.84/27.56 mm [68]. At 5 years, IOP was 14 and 4 mmHg without medication, corrected visual acuity 20/400 in the right eye and 20/60 in the left, and the cup-to-disc ratio 0.4 and 0.3 (partial regression); there were peripheral anterior synechiae on the right, band keratopathy and a filtering bleb on the left, and amblyopia treatment was continuing [68].

The exome, confirmed by Sanger sequencing, identified the Saudi founder variant CYP1B1 NM_000104.4:c.182G>A p.(Gly61Glu) in the heterozygous state, with an allele fraction of 61%, and a heterozygous CPAMD8 variant, NM_015692.5:c.3061G>A, classified as a VUS; the panel was negative for PAX6, PITX3, FOXE3, FOXC1, LTBP2, TEK, MYOC and OPTN, and the authors propose a digenic model [68]. The case requires three caveats. First, the paper gives the CPAMD8 protein variant as p.Val102Met, but VariantValidator converts c.3061G>A to p.(Val1021Met) [42, 68]. Second, the paper assigns a parental origin to each variant and, in the discussion, mentions a lack of parental segregation data, an internal inconsistency [68]. Third, a single CYP1B1 variant is not sufficient to explain an autosomal recessive disease, the CPAMD8 variant is a VUS and the digenic mechanism has not been demonstrated functionally, as the authors acknowledge; the allele fraction of 61% raises, according to the authors themselves, the hypothesis of artefact or mosaicism [68]. In practice, this genotype does not allow carrier testing or prenatal diagnosis to be offered reliably, and justifies the search for a second CYP1B1 variant by CNV analysis and analysis of non-coding regions (section 5.7). The clinical course, with several operations, marked axial myopia and amblyopia, is representative of severe early-onset PCG with a cloudy cornea at presentation, which GeneReviews associates with the worst prognosis [4].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

Without treatment, PCG progresses to blindness [4]. With treatment, the visual prognosis depends on severity at diagnosis, the presence of other ocular anomalies, the response to surgery and IOP control during follow-up; earlier onset, raised IOP and a cloudy cornea at birth are associated with the worst prognosis, and forms that are severe at birth may reach legal blindness despite several operations [4]. The best outcome is obtained when surgery is performed between the second and eighth month of life, and the success rate of each initial procedure is about 80% [4]. A postoperative IOP below 19 mmHg is associated with better vision [4]. Consanguinity is a factor of severity and poor prognosis [1]. The WGA consensus considers cupping reversal frequent in successfully treated childhood glaucoma [21]. Reversal occurred in 43.2% of operated eyes in an Indian cohort of 1,223 eyes, with a lower probability the larger the initial corneal diameter [63]. Mild forms with late presentation often respond to a single operation, with an excellent visual prognosis [4]. Anisometropic amblyopia, axial myopia and corneal scarring continue to limit vision even with controlled IOP, as shown by the case in section 10.2 [1, 68]. Severe visual impairment is frequent [1], and the impact includes carers' anxiety and grief, school integration and the need for visual rehabilitation [1].

In JOAG, progression rates are comparable to those of adult primary glaucoma, but cumulative lifetime disability is greater because of the early age [12]. In an Indian series, the projected lifetime risk of bilateral blindness was 10% even with treatment; in an Indian tertiary centre, 15% of patients were blind in both eyes and 21% in one eye at presentation, and in Nigeria about two-thirds had acuity of 6/18 or worse in the better eye; with surgical control of IOP, cupping reversed in up to 17% of cases [12]. In EFEMP1 JOAG in the Filipino families, 76% of affected carriers became blind [7]. Early diagnosis through cascade testing identifies MYOC carriers before damage occurs: 83% of those identified by cascade testing had no glaucoma at first examination, against 28% with advanced glaucoma among those who came through the clinical route [103]. Without treatment, glaucoma leads to irreversible constriction of the visual field [19].

11.2. Surveillance

GeneReviews notes that the literature is imprecise regarding the age at onset of glaucoma, especially in families with identified variants, and therefore accepts annual screening until early adulthood in this high-risk group [4]; the proposal by Leysen and colleagues is more intensive in the first year of life and extends surveillance to parents and second- and third-degree relatives [58]. Serial measurement of axial length helps to monitor IOP control after surgery: in a series of 12 eyes, eyes with controlled IOP returned to axial growth parallel to the normal curve for age, and the two with poorly controlled IOP maintained above-normal growth (Law 2001) [43]. For carriers of MYOC variants, the sources consulted do not set a numerical interval; the EGS recommends close surveillance and early treatment of carriers [1], and UK Biobank data show that most p.(Gln368Ter) carriers with glaucoma had normal IOP and were undiagnosed, so surveillance should include disc imaging and perimetry [59]. The EGS does not set numerical surveillance intervals specific to PCG or JOAG and insists on lifelong IOP follow-up [1].

11.3. Organisation of care

Follow-up should take place in tertiary centres with volume and a multidisciplinary team, with age-appropriate functional assessment, carer support, visual rehabilitation, school support and contact with patient associations [1]. ERN-EYE offers virtual consultation of complex cases between European centres [18]. Patients without a molecular diagnosis should have their DNA stored and the case re-evaluated as new genes and new variant classifications are published [4, 46].

12. Summary and Final Considerations

Primary congenital glaucoma and juvenile open-angle glaucoma are primary glaucomas of children and young adults in which the monogenic contribution is high. PCG results from dysgenesis of the angle and outflow system; it usually presents in the first year of life with photophobia, epiphora, blepharospasm, corneal enlargement and clouding with Haab striae, and progresses to blindness without treatment. Its birth frequency in Europe is 2.2/100,000 in the EUROCAT registries and about 1 in 12,000–18,000 according to the EGS, and it is much higher in consanguineous populations with CYP1B1 founder variants, such as Saudis and the Roma of Slovakia. JOAG appears between 4 and 35 years, with an open angle, markedly raised IOP and a silent course, and has an estimated prevalence of 1 in 50,000.

The most frequent gene in PCG is CYP1B1, with autosomal recessive inheritance and incomplete penetrance, whose proportion ranges from about 20% to 100% depending on the population; LTBP2 is relevant in consanguineous populations, and TEK and ANGPT1 revealed the role of Schlemm's canal and dominant inheritance with incomplete penetrance. In JOAG, MYOC is the most frequent cause, with a mechanism of misfolded protein toxicity, and FOXC1 duplications emerge as the second cause in Australian cohorts. FOXC1 variants are also found in patients with suspected PCG who, on re-evaluation, have Axenfeld-Rieger features, which justifies panels that include this gene and CNV analysis. Variant interpretation is evolving: ClinGen curation reclassified the Indian variant CYP1B1 p.(Arg368His) as a VUS and continues to review MYOC variants.

Diagnosis is clinical, according to the CGRN criteria adopted by the EGS, and should not be based on IOP alone; corneal diameter, Haab striae, axial length and the optic disc carry equal or greater weight. Treatment of PCG is surgical from the outset, with angle surgery as first choice and 360° circumferential trabeculotomy having a higher success rate than conventional trabeculotomy in moderate-certainty trials; drugs serve as a bridge and have age restrictions, in particular the contraindication of brimonidine below 2 years, and there are divergences between US and EU labels for prostaglandin analogues. In JOAG, surgery is needed in 40–70% of patients and SLT is not recommended. There is no approved gene-directed therapy; approaches to silencing or editing MYOC and to activating the angiopoietin–TEK pathway are at the preclinical stage, and the only human MYOC editing study identified is suspended.

Genetic testing changes family management: it allows siblings to be tested at birth in PCG and repeated examinations under anaesthesia to be avoided, carrier testing and prenatal or preimplantation diagnosis to be offered, and MYOC carriers to be identified by cascade testing before damage occurs, which is the best-documented benefit. Follow-up is lifelong, centred on tertiary centres, and integrates refractive correction, amblyopia treatment, visual rehabilitation and family support.

13. References

1. Pazos M, Traverso CE, Viswanathan A; European Glaucoma Society. European Glaucoma Society - Terminology and guidelines for glaucoma, 6th Edition. Br J Ophthalmol. 2025;109(Suppl 1):1-212. doi:10.1136/bjophthalmol-2025-egsguidelines. PMID: 41026937.
2. Childhood Glaucoma Research Network. CGRN Classification System; CGRN Anomaly Algorithm [Internet]. Grajewski Lyra Foundation for Children with Glaucoma; 2024. Available at: http://www.childhoodglaucoma.com/_files/ugd/fdfc49_145b1c5500c5411c9854c68d084f39f9.pdf
3. Orphanet. Juvenile glaucoma (ORPHA:98977). Available at: https://www.orpha.net/en/disease/detail/98977
4. Abu-Amero KK, Edward DP. Primary Congenital Glaucoma – RETIRED CHAPTER, FOR HISTORICAL REFERENCE ONLY. 2004 Sep 30 [updated 2017 Aug 17; retired 2026 Apr 2]. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1135/. PMID: 20301314.
5. GeneGuidance. ANGPT1 — gene record (clinical curation; version of 2026-09-26).
6. GeneGuidance. MYOC — gene record (clinical curation; version of 2026-09-26).
7. Collantes ERA, Delfin MS, Fan B, Torregosa JMR, Siguan-Bell C, Florcruz NVG, et al. EFEMP1 rare variants cause familial juvenile-onset open-angle glaucoma. Hum Mutat. 2022;43(2):240-252. doi:10.1002/humu.24320. PMID: 34923728. PMC8972201.
8. Maxwell GE, Schmidt JM, Kolovos A, Nguyen TT, Zamora-Alejo K, Ruddle JB, et al. Association of FOXC1 Duplications With Juvenile Open-Angle Glaucoma. JAMA Ophthalmol. 2026;144(6):517-526. doi:10.1001/jamaophthalmol.2026.1183. PMID: 42096227. PMC13154036.
9. Souma T, Tompson SW, Thomson BR, Siggs OM, Kizhatil K, Yamaguchi S, et al. Angiopoietin receptor TEK mutations underlie primary congenital glaucoma with variable expressivity. J Clin Invest. 2016;126(7):2575-87. doi:10.1172/JCI85830. PMID: 27270174. PMC4922711.
10. Young TL, Whisenhunt KN, Jin J, LaMartina SM, Martin SM, Souma T, et al. SVEP1 as a Genetic Modifier of TEK-Related Primary Congenital Glaucoma. Invest Ophthalmol Vis Sci. 2020;61(12):6. doi:10.1167/iovs.61.12.6. PMID: 33027505. PMC7545080.
11. Orphanet. Congenital glaucoma (ORPHA:98976). Available at: https://www.orpha.net/en/disease/detail/98976
12. Selvan H, Gupta S, Wiggs JL, Gupta V. Juvenile-onset open-angle glaucoma - A clinical and genetic update. Surv Ophthalmol. 2022;67(4):1099-1117. doi:10.1016/j.survophthal.2021.09.001. PMID: 34536459. PMC9192165.
13. Siggs OM, Souzeau E, Pasutto F, Dubowsky A, Smith JEH, Taranath D, et al. Prevalence of FOXC1 Variants in Individuals With a Suspected Diagnosis of Primary Congenital Glaucoma. JAMA Ophthalmol. 2019;137(4):348-355. doi:10.1001/jamaophthalmol.2018.5646. PMID: 30653210. PMC6459212.
14. Gedde SJ, Bowden EC, Challa P, Vinod K, Kolomeyer NN, Chopra V, et al. Primary Open-Angle Glaucoma Preferred Practice Pattern®. Ophthalmology. 2026;133(4):P1-P103. doi:10.1016/j.ophtha.2025.12.029. PMID: 41665583.
15. Thomson BR, Liu P, Onay T, Du J, Tompson SW, Misener S, et al. Cellular crosstalk regulates the aqueous humor outflow pathway and provides new targets for glaucoma therapies. Nat Commun. 2021;12(1):6072. doi:10.1038/s41467-021-26346-0. PMID: 34663817. PMC8523664.
16. Yacoub S, Kaipa BR, Li L, Rios S, Kasetti R, Maddineni P, et al. Therapeutic mRNA delivery of CRISPR-Cas9 to the trabecular meshwork reverses ocular hypertension in myocilin glaucoma. Mol Ther Methods Clin Dev. 2025;33(4):101614. doi:10.1016/j.omtm.2025.101614. PMID: 41210166. PMC12594927.
17. Shanghai BDgene Co., Ltd. A Clinical Study on CRISPR/Cas9 Instantaneous Gene Editing Therapy to Primary Open-angle Glaucoma With Elevated Intraocular Pressure and MYOC Gene Mutation. ClinicalTrials.gov identifier NCT06465537. Available at: https://clinicaltrials.gov/study/NCT06465537
18. ERN-EYE — European Reference Network for rare eye diseases [Internet]. Strasbourg: Hôpitaux Universitaires de Strasbourg. Available at: https://www.ern-eye.eu/
19. Firth HV, Hurst JA. Glaucoma. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 428–9.
20. Thau A, Lloyd M, Freedman S, Beck A, Grajewski A, Levin AV. New classification system for pediatric glaucoma: implications for clinical care and a research registry. Curr Opin Ophthalmol. 2018;29(5):385-94. doi:10.1097/ICU.0000000000000516. PMID: 30096087.
21. Weinreb RN, Grajewski A, Papadopoulos M, Grigg J, Freedman S, editors. Childhood Glaucoma: The 9th Consensus Report of the World Glaucoma Association. Amsterdam: Kugler Publications; 2013. Summary Consensus Points available at: https://wga.one/consensus/consensus-9/
22. Alward WL, Fingert JH, Coote MA, Johnson AT, Lerner SF, Junqua D, et al. Clinical features associated with mutations in the chromosome 1 open-angle glaucoma gene (GLC1A). N Engl J Med. 1998;338(15):1022-7. doi:10.1056/NEJM199804093381503. PMID: 9535666.
23. Sheffield VC, Stone EM, Alward WL, Drack AV, Johnson AT, Streb LM, et al. Genetic linkage of familial open angle glaucoma to chromosome 1q21-q31. Nat Genet. 1993;4(1):47-50. doi:10.1038/ng0593-47. PMID: 8513321.
24. Stone EM, Fingert JH, Alward WL, Nguyen TD, Polansky JR, Sunden SL, et al. Identification of a gene that causes primary open angle glaucoma. Science. 1997;275(5300):668-70. doi:10.1126/science.275.5300.668. PMID: 9005853.
25. Sarfarazi M, Akarsu AN, Hossain A, Turacli ME, Aktan SG, Barsoum-Homsy M, et al. Assignment of a locus (GLC3A) for primary congenital glaucoma (Buphthalmos) to 2p21 and evidence for genetic heterogeneity. Genomics. 1995;30(2):171-7. doi:10.1006/geno.1995.9888. PMID: 8586416.
26. Akarsu AN, Turacli ME, Aktan SG, Barsoum-Homsy M, Chevrette L, Sayli BS, et al. A second locus (GLC3B) for primary congenital glaucoma (Buphthalmos) maps to the 1p36 region. Hum Mol Genet. 1996;5(8):1199-203. doi:10.1093/hmg/5.8.1199. PMID: 8842741.
27. Stoilov I, Akarsu AN, Sarfarazi M. Identification of three different truncating mutations in cytochrome P4501B1 (CYP1B1) as the principal cause of primary congenital glaucoma (Buphthalmos) in families linked to the GLC3A locus on chromosome 2p21. Hum Mol Genet. 1997;6(4):641-7. doi:10.1093/hmg/6.4.641. PMID: 9097971.
28. Ali M, McKibbin M, Booth A, Parry DA, Jain P, Riazuddin SA, et al. Null mutations in LTBP2 cause primary congenital glaucoma. Am J Hum Genet. 2009;84(5):664-71. doi:10.1016/j.ajhg.2009.03.017. PMID: 19361779. PMC2680998.
29. Narooie-Nejad M, Paylakhi SH, Shojaee S, Fazlali Z, Rezaei Kanavi M, Nilforushan N, et al. Loss of function mutations in the gene encoding latent transforming growth factor beta binding protein 2, LTBP2, cause primary congenital glaucoma. Hum Mol Genet. 2009;18(20):3969-77. doi:10.1093/hmg/ddp338. PMID: 19656777.
30. Thomson BR, Souma T, Tompson SW, Onay T, Kizhatil K, Siggs OM, et al. Angiopoietin-1 is required for Schlemm's canal development in mice and humans. J Clin Invest. 2017;127(12):4421-4436. doi:10.1172/JCI95545. PMID: 29106382. PMC5707160.
31. Online Mendelian Inheritance in Man, OMIM®. Glaucoma 3, primary congenital, C; GLC3C. MIM Number: 613085. Baltimore (MD): Johns Hopkins University. Available at: https://omim.org/entry/613085
32. Online Mendelian Inheritance in Man, OMIM®. Glaucoma 3, primary congenital, D; GLC3D. MIM Number: 613086. Baltimore (MD): Johns Hopkins University. Available at: https://omim.org/entry/613086
33. GeneGuidance. CYP1B1 — gene record (clinical curation; version of 2026-09-26).
34. GeneGuidance. TEK — gene record (clinical curation; version of 2026-09-26).
35. GeneGuidance. LTBP2 — gene record (clinical curation; version of 2026-09-26).
36. National Center for Biotechnology Information. ClinVar [Internet]. Bethesda (MD): NLM. VCV 7730, 7739, 7735, 7733, 282564, 68468 (CYP1B1); 7949, 7948, 7952, 7946 (MYOC). Available at: https://www.ncbi.nlm.nih.gov/clinvar/
37. Lingham G, Thakur S, Safi S, Gordon I, Evans JR, Keel S. A systematic review of clinical practice guidelines for childhood glaucoma. BMJ Open Ophthalmol. 2022;7(1):e000933. doi:10.1136/bmjophth-2021-000933. PMID: 35136841. PMC8804684.
38. Firth HV, Hurst JA. Anterior segment eye malformations. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 62–4.
39. Haddad A, Ait Boujmia OK, El Maaloum L, Dehbi H. Meta-analysis of CYP1B1 gene mutations in primary congenital glaucoma patients. Eur J Ophthalmol. 2021;31(6):2796-2807. doi:10.1177/11206721211016308. PMID: 34020567.
40. de Melo MB, Mandal AK, Tavares IM, Ali MH, Kabra M, de Vasconcellos JP, et al. Genotype-Phenotype Correlations in CYP1B1-Associated Primary Congenital Glaucoma Patients Representing Two Large Cohorts from India and Brazil. PLoS One. 2015;10(5):e0127147. doi:10.1371/journal.pone.0127147. PMID: 25978063. PMC4433271.
41. Simões MJ, Carmona S, Roberts R, Wainwright G, Faro C, Silva E, et al. CYP1B1 mutational screening in a Portuguese cohort of primary congenital glaucoma patients. Ophthalmic Genet. 2017;38(2):197-9. doi:10.1080/13816810.2016.1188121. PMID: 27268095.
42. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. PMID: 28967166. PMC5765404. Tool: https://variantvalidator.org/
43. Online Mendelian Inheritance in Man, OMIM®. Glaucoma 3, primary congenital, A; GLC3A. MIM Number: 231300. Baltimore (MD): Johns Hopkins University. Available at: https://omim.org/entry/231300
44. Stoilov IR, Costa VP, Vasconcellos JP, Melo MB, Betinjane AJ, Carani JC, et al. Molecular genetics of primary congenital glaucoma in Brazil. Invest Ophthalmol Vis Sci. 2002;43(6):1820-7. PMID: 12036985.
45. Seront E, Boon LM, Vikkula M. TEK-Related Venous Malformations. 2008 Sep 18 [updated 2023 Mar 2]. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1967/ PMID: 20301733.
46. Tompson SWJ, Graham P, Hadler J, Pasutto F, Whisenhunt KN, Chakrabarti S, et al. ClinGen Glaucoma Variant Curation Expert Panel recommendations enhance classification of myocilin variants [preprint]. medRxiv. 2026 Jul 31. doi:10.64898/2026.07.29.26359282
47. Online Mendelian Inheritance in Man, OMIM®. Glaucoma 1, open angle, A; GLC1A. MIM Number: 137750. Baltimore (MD): Johns Hopkins University. Available at: https://omim.org/entry/137750
48. Jain A, Zode G, Kasetti RB, Ran FA, Yan W, Sharma TP, et al. CRISPR-Cas9-based treatment of myocilin-associated glaucoma. Proc Natl Acad Sci U S A. 2017;114(42):11199-11204. doi:10.1073/pnas.1706193114. PMID: 28973933. PMC5651749.
49. Wan Y, Liu X, Yan X, Wu S, Teng Y, Han L, et al. AAV-DJ-Mediated MYOC Silencing as a Gene Therapy Approach for Myocilin-Associated Glaucoma. Invest Ophthalmol Vis Sci. 2026;67(1):48. doi:10.1167/iovs.67.1.48. PMID: 41569028. PMC12849824.
50. Fingert JH, Héon E, Liebmann JM, Yamamoto T, Craig JE, Rait J, et al. Analysis of myocilin mutations in 1703 glaucoma patients from five different populations. Hum Mol Genet. 1999;8(5):899-905. doi:10.1093/hmg/8.5.899. PMID: 10196380.
51. Han X, Souzeau E, Ong JS, An J, Siggs OM, Burdon KP, et al. Myocilin Gene Gln368Ter Variant Penetrance and Association With Glaucoma in Population-Based and Registry-Based Studies. JAMA Ophthalmol. 2019;137(1):28-35. doi:10.1001/jamaophthalmol.2018.4477. PMID: 30267046. PMC6439786.
52. Burdon KP, Graham P, Hadler J, Hulleman JD, Pasutto F, Boese EA, et al. Specifications of the ACMG/AMP variant curation guidelines for myocilin: Recommendations from the clingen glaucoma expert panel. Hum Mutat. 2022;43(12):2170-2186. doi:10.1002/humu.24482. PMID: 36217948. PMC9771967.
53. Jones KL, Jones MC, del Campo M. Axenfeld-Rieger syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 834–7.
54. Della Paolera M, de Vasconcellos JP, Umbelino CC, Kasahara N, Rocha MN, Richeti F, et al. CYP1B1 gene analysis in primary congenital glaucoma Brazilian patients: novel mutations and association with poor prognosis. J Glaucoma. 2010;19(3):176-82. doi:10.1097/IJG.0b013e3181a98bae. PMID: 19528825.
55. Chen X, Chen Y, Wang L, Jiang D, Wang W, Xia M, et al. CYP1B1 genotype influences the phenotype in primary congenital glaucoma and surgical treatment. Br J Ophthalmol. 2014;98(2):246-51. doi:10.1136/bjophthalmol-2013-303821. PMID: 24227805.
56. Cardoso MS, Anjos R, Vieira L, Ferreira C, Xavier A, Brito C. CYP1B1 gene analysis and phenotypic correlation in Portuguese children with primary congenital glaucoma. Eur J Ophthalmol. 2015;25(6):474-7. doi:10.5301/ejo.5000618. PMID: 25952714.
57. Rodrigues R, Alves D, Esteves-Leandro J, Silva M, Barbosa-Breda J, Tavares-Ferreira J, et al. Influence of CYP1B1 Variants on Phenotypic Characteristics and Therapeutic Outcomes in Primary Congenital Glaucoma. Ophthalmol Glaucoma. 2025;8(5):457-65. doi:10.1016/j.ogla.2025.04.012. PMID: 40349919.
58. Leysen L, Cassiman C, Vermeer S, Casteels I, Balikova I. Genetics in primary congenital glaucoma: Implications in disease management and counseling. Eur J Med Genet. 2022;65(1):104378. doi:10.1016/j.ejmg.2021.104378. PMID: 34748994.
59. Zebardast N, Sekimitsu S, Wang J, Elze T, Gharahkhani P, Cole BS, et al. Characteristics of p.Gln368Ter Myocilin Variant and Influence of Polygenic Risk on Glaucoma Penetrance in the UK Biobank. Ophthalmology. 2021;128(9):1300-1311. doi:10.1016/j.ophtha.2021.03.007. PMID: 33713785. PMC9134646.
60. Kiyota N, Onay T, Leeaw P, Liu P, Deb DK, Thomson BR, et al. Glaucoma-Protective Human Single-Nucleotide Polymorphism in the Angpt2 Locus Increased ANGPT2 Expression and Schlemm Canal Area in Mice-Brief Report. Arterioscler Thromb Vasc Biol. 2024;44(10):2207-2212. doi:10.1161/ATVBAHA.124.321555. PMID: 39206543. PMC11952684.
61. Marzin P, Cormier-Daire V, Tsilou E. Weill-Marchesani Syndrome. 2007 Nov 1 [updated 2020 Dec 10]. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1114/ PMID: 20301293.
62. Firth HV, Hurst JA. Deletions and duplications (microdeletions). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 660–5.
63. Panigrahi A, Yadav SK, Pathak AK, Kumar A, Das SK, Gupta S, et al. Reversal of optic disc cupping in primary congenital glaucoma: analysis of 1223 eyes. Br J Ophthalmol. 2026;. doi:10.1136/bjo-2026-330658. PMID: 42744592.
64. Online Mendelian Inheritance in Man, OMIM®. Glaucoma 3, primary congenital, E; GLC3E. MIM Number: 617272. Baltimore (MD): Johns Hopkins University. Available at: https://omim.org/entry/617272
65. Turalba AV, Chen TC. Clinical and genetic characteristics of primary juvenile-onset open-angle glaucoma (JOAG). Semin Ophthalmol. 2008;23(1):19-25. doi:10.1080/08820530701745199. PMID: 18214788.
66. Firth HV, Hurst JA. Corneal clouding. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 120–1.
67. El Helwe H, Samuel S, Gupta S, Neeson C, Chachanidze M, Solá-Del Valle DA. Case Report: Reversal and subsequent return of optic disc cupping in a myocilin (MYOC) gene-associated severe Juvenile Open-Angle Glaucoma (JOAG) patient. F1000Res. 2022;11:1361. doi:10.12688/f1000research.127871.1. PMID: 38868171. PMC11167334.
68. Abu-Amero K, Sesma G. PCG with biallelic CYP1B1 and CPAMD8 variants: a longitudinal case report. Front Ophthalmol (Lausanne). 2026;6:1829921. doi:10.3389/fopht.2026.1829921. PMID: 42528638. PMC13414795.
69. Jones KL, Jones MC, del Campo M. Peters-plus syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 838–9.
70. Jones KL, Jones MC, del Campo M. Sturge-Weber syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 698–9.
71. Jones KL, Jones MC, del Campo M. Microphthalmia–linear skin defects syndrome (MIDAS syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 758–9.
72. Jones KL, Jones MC, del Campo M. SHORT syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 116–7.
73. Jones KL, Jones MC, del Campo M. Oto-palato-digital syndrome, type II. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 546–9.
74. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Melnick-Needles osteodysplasty. In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 320–5.
75. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Frank-ter Haar syndrome. In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 346–50.
76. Jones KL, Jones MC, del Campo M. Congenital Zika syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 812–5.
77. Gagrani M, Garg I, Ghate D. Surgical interventions for primary congenital glaucoma. Cochrane Database Syst Rev. 2020;8(8):CD008213. doi:10.1002/14651858.CD008213.pub3. PMID: 32816311. PMC8094178.
78. Boese EA, Alward WLM, Fingert JH. Gonioscopy-Assisted Transluminal Trabeculotomy for Myocilin-Associated Juvenile Open-Angle Glaucoma: A Case Series of 8 Eyes Over 2.2 to 4.1 Years. Ophthalmol Glaucoma. 2025;8(5):466-473. doi:10.1016/j.ogla.2025.03.011. PMID: 40188877. PMC12724723.
79. Khan AM, AlJadaan I, Al-Omair AM, Motabagani DMA, Mousa A, Sesma G. Nonpenetrating Deep Sclerectomy versus Conventional Surgery in Primary Congenital Glaucoma: a Systematic Review and Meta-analysis. Ophthalmol Ther. 2026;15(9):2897-2917. doi:10.1007/s40123-026-01443-9. PMID: 42455251. PMC13518635.
80. Kunu DM, Ucgul AY, Gonen N, Uysal BS, Atalay HT, Ozmen MC, et al. 5-year outcomes of ahmed glaucoma valve implantation after failed trabeculotomy in primary congenital glaucoma. Graefes Arch Clin Exp Ophthalmol. 2026;. doi:10.1007/s00417-026-07458-z. PMID: 42611055.
81. Abbas J, Haider F, Arooj H, Rustam Z, Srivastava S, Haider H, et al. Comparison of 360-Degree Trabeculotomy Versus Traditional Angle Surgery in Primary Congenital Glaucoma: A Systematic Review and Meta-Analysis. J Glaucoma. 2026;35(7):e86-e94. doi:10.1097/IJG.0000000000002705. PMID: 41875194. PMC13299040.
82. CHU de Lille. Prospective Study of the Diagnostic and Therapeutic Management of Congenital Glaucoma in France. ClinicalTrials.gov identifier NCT03077789. Available at: https://clinicaltrials.gov/study/NCT03077789
83. Viatris. Xalatan 50 micrograms/ml eye drops solution — Summary of Product Characteristics [Internet]. electronic medicines compendium (UK); revised 04/12/2025. Available at: https://www.medicines.org.uk/emc/product/1616/smpc
84. Viatris Specialty LLC. XALATAN (latanoprost ophthalmic solution) 0.005% [prescribing information]. NDA020597. Revised 2023-05. DailyMed setid a98595b3-9f47-48e0-b18d-550a2095f264. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=a98595b3-9f47-48e0-b18d-550a2095f264
85. European Medicines Agency. Travatan (travoprost) — EPAR, product information [Internet]. EMEA/H/C/000390. Amsterdam: EMA; EC update 22/01/2026. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/travatan
86. Novartis Pharmaceuticals Corporation. TRAVATAN Z (travoprost ophthalmic solution) 0.004% [prescribing information]. NDA021994. DailyMed setid 028455e0-ae77-4213-8819-3b58ef7d6a14. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=028455e0-ae77-4213-8819-3b58ef7d6a14
87. Nordic Pharma, Inc. TIMOPTIC in OCUDOSE (timolol maleate ophthalmic solution) preservative-free [prescribing information]. ANDA212592. DailyMed setid 16323b0e-78ae-40e5-be93-04f698dd672b.
88. Santen UK Limited. Timoptol 0.25% and 0.5% w/v eye drops, solution — Summary of Product Characteristics [Internet]. electronic medicines compendium (UK); revised 10/12/2019. Available at: https://www.medicines.org.uk/emc/product/10611/smpc
89. Sagent Pharmaceuticals. Dorzolamide hydrochloride ophthalmic solution 2% [prescribing information]. ANDA215660. DailyMed setid 0246463c-6c24-4611-8e22-2d5254ace66a.
90. Santen UK Limited. Trusopt 20 mg/ml eye drops, solution — Summary of Product Characteristics [Internet]. electronic medicines compendium (UK); revised 29/01/2026. Available at: https://www.medicines.org.uk/emc/product/1009/smpc
91. European Medicines Agency. Azopt (brinzolamide) — EPAR, product information [Internet]. EMEA/H/C/000267. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/azopt
92. Sandoz Inc. AZOPT (brinzolamide ophthalmic suspension) 1% [prescribing information]. NDA020816. DailyMed setid 0afb6589-51ba-432c-aed1-20a12d5d3455.
93. Avet Pharmaceuticals Inc. Acetazolamide tablets [prescribing information]. ANDA205530. DailyMed setid 0700a910-79aa-4431-aecf-637fe14fcae8.
94. Allergan, Inc. ALPHAGAN P (brimonidine tartrate ophthalmic solution) 0.1%/0.15% [prescribing information]. NDA021262/NDA021770. DailyMed setid 264c4494-c225-486f-888a-caaf36be46b3.
95. AbbVie Ltd. Alphagan 0.2% w/v (2 mg/ml) eye drops, solution — Summary of Product Characteristics [Internet]. electronic medicines compendium (UK); revised 08/12/2025. Available at: https://www.medicines.org.uk/emc/product/856/smpc
96. Sandoz Inc. Pilocarpine hydrochloride ophthalmic solution 1%, 2%, 4% [prescribing information]. NDA200890. Revised 2020-05. DailyMed setid 53fa0b7a-f982-48a7-92c4-6f098683c1ed. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=53fa0b7a-f982-48a7-92c4-6f098683c1ed
97. Fera Pharmaceuticals, LLC. PHOSPHOLINE IODIDE (echothiophate iodide for ophthalmic solution) 0.125% [prescribing information]. NDA011963. Revised 2025-07. DailyMed setid 8afc8951-db76-4f27-8752-7c0d6aa47244. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=8afc8951-db76-4f27-8752-7c0d6aa47244
98. Alcon Laboratories, Inc. RHOPRESSA (netarsudil ophthalmic solution) 0.02% [prescribing information]. NDA208254. DailyMed setid 7d4f0e3a-5b86-4c43-982a-813b22ae7e22.
99. Bausch & Lomb Incorporated. VYZULTA (latanoprostene bunod ophthalmic solution) 0.024% [prescribing information]. NDA207795. DailyMed setid 60ce7900-a677-4fc2-88d0-1603baa9dbd4.
100. Thomson BR, Carota IA, Souma T, Soman S, Vestweber D, Quaggin SE. Targeting the vascular-specific phosphatase PTPRB protects against retinal ganglion cell loss in a pre-clinical model of glaucoma. Elife. 2019;8. doi:10.7554/eLife.48474. PMID: 31621585. PMC6874417.
101. EyePoint Pharmaceuticals (Aerpio). Phase 2 study of AKB-9778 ophthalmic solution as an adjunct to latanoprost in OHT or OAG. ClinicalTrials.gov identifier NCT04405245. Available at: https://clinicaltrials.gov/study/NCT04405245
102. Jemmeih S, Malik S, Okashah S, Zayed H. Genetic Epidemiology of Primary Congenital Glaucoma in the 22 Arab Countries: A Systematic Review. Ophthalmic Epidemiol. 2022;29(1):1-12. doi:10.1080/09286586.2021.1883676. PMID: 33641569.
103. Souzeau E, Tram KH, Witney M, Ruddle JB, Graham SL, Healey PR, et al. Myocilin Predictive Genetic Testing for Primary Open-Angle Glaucoma Leads to Early Identification of At-Risk Individuals. Ophthalmology. 2017;124(3):303-309. doi:10.1016/j.ophtha.2016.11.011. PMID: 27993484.
104. Souzeau E, Glading J, Keane M, Ridge B, Zhou T, Burdon KP, et al. Predictive genetic testing experience for myocilin primary open-angle glaucoma using the Australian and New Zealand Registry of Advanced Glaucoma. Genet Med. 2014;16(7):558-63. doi:10.1038/gim.2013.196. PMID: 24406458.
105. Souzeau E, Glading J, Ridge B, Wechsler D, Chehade M, Dubowsky A, et al. Predictive genetic testing in minors for Myocilin juvenile onset open angle glaucoma. Clin Genet. 2015;88(6):584-8. doi:10.1111/cge.12558. PMID: 25582056.
106. Olawoye O, Young BP, Nyunt AW, Fafowora OF, Ajani M, Sarimiye T, et al. Prevalence of Myocilin Mutations in a Cohort of Patients with Juvenile Open-Angle Glaucoma from sub-Saharan Africa. Ophthalmol Glaucoma. 2025;8(5):450-456. doi:10.1016/j.ogla.2025.04.010. PMID: 40280414. PMC12459295.
107. Childhood Glaucoma Research Network [Internet]. Available at: https://www.childhoodglaucoma.com/
108. Glaucoma UK [Internet]. Available at: https://glaucoma.uk/
109. Glaucoma Research Foundation [Internet]. Available at: https://glaucoma.org/