A clinical and molecular genetics study of primary congenital glaucoma in South Korea
- Department of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
- Correspondence to Professor Changwon Kee, Department of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul 135-710, Korea;
- Accepted 4 July 2012
- Published Online First 1 September 2012
Objectives To investigate the clinical manifestations associated with the mutation spectrums of the human cytochrome P450 (CYP1B1) and myocilin (MYOC) genes in South Korean patients with primary congenital glaucoma (PCG).
Methods Eighty-five unrelated PCG patients and their family members of South Korean origin were screened for mutations in the CYP1B1 and MYOC genes by sequencing with the PCR. We analysed phenotypes related to the presence, number and types of CYP1B1 mutations. In addition, the phenotype associated with the MYOC gene mutations was evaluated.
Results There was no statistically significant difference in clinical studies between PCG patients with CYP1B1 mutations (N=63) and those without mutations (N=22), although the mutation group manifested disease earlier, with greater severity, and frequency in both eyes (p>0.05). However, the response to treatments was statistically different between groups and tended to be poor according to the number of mutant alleles (p=0.000, 0.0017). Patients with MYOC mutations (N=2) showed various phenotypic features.
Conclusions No consistent correlation was observed between the initial clinical manifestations and the CYP1B1 genotype. However, the response to treatment was associated with the CYP1B1 mutant alleles. This is the first report discussing the phenotypes of South Korean PCG patients associated with CYP1B1 mutations.
Primary congenital glaucoma (PCG) is a type of childhood glaucoma, in which a developmental abnormality within the anterior chamber angle leads to the obstruction of aqueous outflow.1 PCG is usually managed surgically, but even when intraocular pressure is well controlled, a significant number of children never achieve good vision.1 Previous studies have reported that approximately one half of the patients had visual acuities of less than 20/50, and this was related to persistent corneal changes, irreversible optic nerve damage or amblyopia from the induced anisometropia.2 ,3 Therefore, early diagnosis and appropriate management are critical for this disease.
The genetic heterogeneity of PCG was confirmed using genetic linkage studies conducted during the 1990s and three different genetic loci for PCG—GLC3A linked to the 2p21 region, GLC3B linked to 1p36 region and GLC3C linked to 14q24.3 region—were identified.1 In particular, the first gene to be directly implicated in the pathogenesis of PCG, that is, the cytochrome P450 1B1 (CYP1B1) gene, was identified by positional cloning and was mapped to the 2p21 region.4 These mutations were observed in Saudi Arabia, Turkey, the UK, Canada, and Slovakia, thus indicating CYP1B1 as the congenital glaucoma gene.4
However, the frequency of CYP1B1 mutation varies, depending on the geographical location and ethnic background, and the association with the phenotype of PCG is still unknown. Yuichiro et al reported that age at the initial onset of glaucoma and sex preponderance were related to CYP1B1 mutation.5 Several studies have shown a genotype-phenotype correlation.6–9 In contrast, other studies failed to report a definite genotype-phenotype correlation.10 ,11 In addition, the CYP1B1 mutation was present in different types of glaucoma: primary open angle glaucoma, juvenile open angle glaucoma, Rieger's anomaly, and Peters’ anomaly.12–15 Moreover, mutations in the myocilin gene (MYOC) have been identified in approximately 5.55% of PCG patients and Vincent et al reported that simultaneous mutations in both CYP1B1 and MYOC may lead to the early onset of juvenile open angle glaucoma.16 ,17
In previous studies, we reported molecular genetic findings regarding the mutation spectrums of CYP1B1 and MYOC in unrelated South Korean patients with PCG.18 In this sequential study, we further investigated the genotype-phenotype correlation with the expectation that this work could contribute to predicting the severity or prognosis of PCG by genetic analysis.
Patients and methods
We investigated 85 unrelated PCG patients recruited from seven hospitals in South Korea. The study was conducted in compliance with the tenets of the Declaration of Helsinki for the use of human subjects in biomedical research and was approved by the Institutional Review Board. Written informed consent was obtained from all participants or from parents of participating children. The method of genomic analysis was presented in a previous report.18
All PCG patients were diagnosed by glaucoma specialists. Slit lamp biomicroscopy, tonometry, gonioscopy, fundus examination, refraction and axial length measurements were performed when available. Intraocular pressure was measured using either Goldmann tonometry or Perkins applanation tonometry. Primary congenital glaucoma was diagnosed when intraocular pressure was higher than 21 mm Hg in at least one eye before treatment, accompanied with the compatible corneal lesion, increased axial length and optic nerve head changes suggestive of glaucomatous damage, when possible. Patients with other ocular or systemic anomalies, such as aniridia or Peter's anomaly were excluded. Family members were interviewed about the medical history and their pedigree was evaluated. All patients were unrelated, not relavant to other racial groups and none was the offspring of consanguineous marriages.
Patients were classified into two groups according to the CYP1B1 mutation and clinical characteristics were compared, including age at onset, the involvement of both eyes, gender and severity. The severity index based on the following five parameters were introduced in the previous study: corneal diameter (up to 10.5, >10.5–12, >12–13, >13 mm), intraocular pressure (up to 16, >16–20, >20–30, >30), cup to disc ratio (0.3–0.4, >0.4–0.6, >0.6–0.8, >0.8), visual acuity (20/20, <20/20–20/60, <20–60–20/200, <20/200) and corneal clarity (no oedema, mild oedema, severe oedema, severe oedema and Haab's striae), when available.11 According to the extent of each parameter, severity index were classficied into normal, mild, moderate and severe/very severe groups.
All surgical interventions were performed as quickly as possible, usually within 1 month after a visit, by glaucoma specialists in one of the seven included medical centres. The number and types of surgery, the response to treatment and visual outcomes were recorded. The response to treatment was classified into four categories: intraocular pressure was (1) controlled with anti-glaucoma medication, (2) controlled with surgery, (3) controlled with surgery and additional medication and (4) uncontrolled despite current treatments. We further divided the subjects into subgroups according to the number or type of mutations and compared clinical manifestations.
Statistical analysis was conducted using PASW software V.17.0 (SPSS, Inc., Chicago, Illinois, USA). Pearson's χ2 test or Fisher's exact test were used for comparing the clinical parameters between groups. The changes of severity indices and the responses to treatment according to the number of mutations were analysed using the Jonckheere-Terpstra test. The correlation between the response to treatments and number of CYP1B1 mutations was analysed with the Spearman's rank correlation coefficient, using the analysis software R.19 A p value of less than 0.05 was considered statistically significant.
A total of 61.1% of PCG patients manifested symptoms or signs within 6 months. There was no sex predominance (male to female ratio 46 : 49) and one case was family-related. Clinical and molecular pathology in PCG patients with mutations in the CYP1B1 gene (N=22) and MYOC gene (N=2) are summarised in table 1.
Phenotype analysis: primary congenital glaucoma without CYP1B1 mutation versus with CYP1B1 mutation
A comparison of phenotypic features is summarised in table 2. The age range of presentation were from 0 to 3 years in the mutation group, whereas from 0 to 6 years in the no mutation group. The proportion of early onset, the involvement of both eyes and severe/very severe index were higher in the mutation group than the no mutations group. Females were more prevalent in the mutation group. However, in all clinical parameters mentioned, there were no statistically significant differences between the two groups (p>0.05).
The number of surgical interventions was not statistically different between the groups (p=0.067). In the types of surgical interventions, 64.4% in the no mutation group had angle surgery (goniotomy or trabeculotomy), whereas 68.7% in the mutation group had trabeculectomy or valve surgery (p=0.027). The majority of the no mutation group (58.7%) achieved intraocular pressure less than 21 with surgery, whereas half of the cases with CYP1B1 mutations (50%) needed supplementary medication after the surgery. As a result, the response to treatment was statistically and significantly different between the two groups (p=0.008). In the final visual outcome, there was no significant difference regardless of the mutations (p=0.991). There was no statistically significant difference in the follow up periods between two groups (p=0.931).
Phenotype analysis depending on the number of CYP1B1 mutations: no mutation group versus one CYP1B1 mutant allele group versus two mutant alleles group
All PCG subjects were classified again into three groups (figure 1; see table 3). There was no statistically significant difference in all the clinical parameters between the groups, although PCG patients with two mutant alleles manifested the disease earlier than the other groups and the involvement of both eyes increased as the number of mutations increased (no mutation group 61.9% versus one mutation group 75% versus two mutations group 90%, respectively; data not shown). Females subjects were more prevalent among subjects with two mutations, however, the difference was not statistically significant (p=0.264). When analysing the changes in the severity index according to the number of mutations, we were unable to find a statistically significant result (p=0.335; see table 3).
The number of surgeries was not different among the groups (p=0.124). However, the responses to treatment were statistically different among the groups and it tended to be poor as the number of mutant alleles increased (p=0.001; see table 3). The response to treatment was significantly correlated with the number of mutations (Spearman's rank correlation coefficient=0.4076) (p=0.0017). The types of surgeries were also different among the groups (p=0.002). However, there was no statistically significant difference in recent visual outcomes among the groups (p=0.567). There was also no statistically significant difference in the follow up periods among the three groups (p=0.826).
Phenotype analysis depending on types of CYP1B1 mutations
The types of CYP1B1 mutant alleles are listed in table 1. PCG subjects with the T325SfsX104 mutations manifested clinical features within 6 months, in both eyes, were mostly female, and 60% who were in the severe/very severe severity index, and 60% who were in the fourth stage of treatment response. However, there was no statistically significant difference detected between those with mutations and those without mutation (p>0.05). In other types of mutant alleles (V364M, G329S, Y81X, A330F, V419GfsX11, V320L, V364M, R444Q, and R390H), there were also no characteristic clinical manifestations related to the specific types of mutants.
Analysis of MYOC gene mutation in PCG
MYOC mutation was detected in only 2 (2.53%) of the 79 patients with PCG. These two patients carried possible novel mutations: c.683T>C (p.L228S) and c.719A>G (p.E240G), as we previously reported.18 These two patients showed various phenotypic features (age of onset, gender, the involvement of both eyes, severity, the response to therapy and recent visual outcome; see table 1).
The genetic studies in various studies have reported several alterations in the coding region of CYP1B1 and phenotypic heterogeneity. There are contrasting opinions about the phenotypic features associated with the CYP1B1 mutations. Several molecular genetic studies have reported that patients with mutant alleles or specific types of mutants presented aggressive features of the disease, whereas in a German population study related to mutations, no significant difference in the clinical findings was detected.20 ,21 In our investigation, albeit not all, the symptoms or signs tended to occur earlier, with greater severitya, and in both eyes of the PCG patients with CYP1B1 mutations, compared to the no-mutation group. These results are similar to the study in a Japanese population.22 However, in our study, these differences were not statistically significant, which is similar to the German study.21 This discrepancy may result from the difference in: (1) the number of investigated subjects, (2) the method of comparing clinical parameters, (3) the harbouring CYP1B1 mutations, or (4) the ethnic and racial background.
Interestingly, in our study, 11 patients, which make up half of the mutation group, had a heterozygoous mutant allele. We were unable find any consistent clinical characteristics in this group. This may be due to (1) the existence of undetected variations, such as variations in the promoter region of the affected gene, (2) potential new genes involved in the pathogenesis of PCG, (3) various activities at the enzymatic level of CYP1B1, or (4) the influence of environmental agents at the CYP1B1 level. Jansson et al23 found that the stability of the haemoprotein complex was reduced by G61E, but was increased by R469W, although both mutations impaired the normal functioning of the CYP1B1 haemoprotein. To some extent, PCG patients with various CYP1B1 mutations may manifest phenotypic heterogeneity. Therefore, to find the phenotypic variability associated with different combinations of CYP1B1 mutations, it is essential to demonstrate the enzymatic properties of CYP1B1 as well as its genetic features.
There was a meaningful difference in the response to treatment depending on the presence or number of CYP1B1 mutations. In the mutation group, especially in patients with mutant alleles (homozygote or compound heterozygote), the proportion of patients with uncontrolled intraocular pressure despite maximal current treatments was high. In addition, the mutation group required additional trabeculectomy or valve surgery after angle surgery compared to the no mutation group. Since the CYP1B1 mutation is known to influence the angle structure, the reason why response to the treatment was poor in the mutation group is not clearly understood, even considering the variations in the skills of surgeons from multiple centres (all were skillful glaucoma specialists) or in the follow-up period. The final visual outcome was not significantly associated with CYP1B1 mutations, which may be related to the finding that involvement of one eye, that is, amblyopia, was more frequently found in patients without the CYP1B1 mutation.
Regarding the MYOC mutations, we were unable to analyse the genotype—phenotype correlation because only two patients had MYOC mutations and their phenotypes were varied. Vincent et al17 proposed that CYP1B1 may act as a modifier of MYOC expression and a mutation in both genes was related to the early onset of juvenile open angle glaucoma. However, in our study, there were no patients with simultaneous mutations in CYP1B1 and MYOC.
This is the first report discussing the phenotypes associated with CYP1B1 mutations in PCG among South Korean patients. No consistent correlation was observed between clinical findings at the initial presentation and genotype in South Korean PCG patients. However, the treatment response was associated with the presence and numbers of CYP1B1 mutations. The molecular genetic study in PCG still remains an unsolved problem and continues to be challenging. However, we believe that the research will aid in the understanding of the pathophysiology of the disease and ultimately identify personalised treatment options that will be applicable to families with PCG and defined mutations in the CYP1B1 gene.
Contributors (1) Conception and design (CK, WS), acquisition of data, or analysis and interpretation of data (WS); (2) drafting the article or revising it critically for important intellectual content (WS, CK); and (3) final approval of the version to be published (CK).
Funding Heart to heart foundation.
Competing interests None.
Ethics approval This study was conducted with the approval of seven medical centres (Severance Hospital of Yonsei University, Seoul National University Hospital, Asan Medical Center, Korea University Hospital, Chungnam National University Hospital, Catholic Medical Center, Samsung Medical Center).
Provenance and peer review Not commissioned; externally peer reviewed.