Cystic maculopathy is rare in the pediatric age group and has been associated with several diseases. The most frequent cause of cystoid macular edema is not genetic but due to damage associated with vascular diseases. Less often, cystoid or schitic spaces in the macula may be present even if the blood–retinal barrier is intact and uninvolved, in cases such as optic pit,1 myopic traction maculopathy,2 or genetic disorders such as retinitis pigmentosa,3 or be the primary sign of cone dystrophy,4 enhanced S-cone syndrome,5 and foveal retinoschisis.6
Familial foveal retinoschisis represents a rare autosomal recessive disorder characterized by a typical foveal cartwheel lesion due to the presence of schitic or cystoid spaces with apparently normal peripheral retina.7 A previous study by Vincent et al.6 (describing two families affected by familial foveal retinoschisis caused by CRB1 variants) associated familial foveal retinoschisis with CRB1 variants. CRB1 is expressed in the retina and brain and codes for the human orthologue of Drosophila melanogaster Crumbs protein that is involved in the photoreceptor morphogenesis, the assembly of adherens junction, and the establishment of the polarity of epithelial cells, thus playing a crucial role in retinal development. CRB1 variants have also been associated with several retinal phenotypes, including retinitis pigmentosa, Leber congenital amaurosis, preserved para-arteriole retinal pigment epithelium, and maculopathy.
We report the phenotype-genotype pattern of a pair of dizygotic twins, conceived by heterologous fertilization, who were affected by early-onset cystic maculopathy due to CRB1 pathogenic variants. We also highlight the importance of exome screening to assess the likelihood of recessively inherited genetic disorders by means of an appropriate algorithm able to perform a genetic compatibility test (genetic matching).
Two 3-year-old dizygotic twins (male and female) were referred to the Pediatric Ophthalmology Service of the University of Verona for suspected bilateral low vision. There was no familial history of eye disease, but the pregnancy was the result of oocyte donation.
Refraction revealed a hypermetropic defect in both children. Their visual acuity measured by the “tumbling E eye chart” was 20/63 in both eyes in the boy and 20/80 in both eyes in the girl. Slit-lamp examination revealed a normal anterior segment with no signs of inflammation. Lens, iris, and pupil reaction were unremarkable and intraocular pressure was within normal limits. The vitreous was clear. Fundus examination showed bilateral cystic maculopathy. The optic nerve and the remainder of the retina were ophthalmoscopically normal.
Optical coherence tomography (OCT) showed a different pattern in each of the twins characterized by a central macrocyst surrounded by multiple cystic spaces distributed in the external retinal layers with a central foveal thickness of 423 µm in the right eye and 455 µm in the left eye in the boy, whereas the girl presented with multiple central microcystic spaces in the external retinal layers with a foveal thickness of 205 µm in the right eye and 215 µm in the left eye, respectively (Figure 1). Outer photoreceptor segments were severely impaired with a visible but discontinuous attenuated ellipsoid zone in both cases.
Fundus examination and optical coherence tomography (OCT) at first evaluation, when the two dizygotic twins were (left side: A, B, C, and D) 3 years old and (right side: E, F, G, and H) at the 4-year follow-up examination. The boy showed macular cystoid edema in the (A) right and (B) left eyes. Fundus examination and OCT of the girl also showed macular cystoid edema in the (C) right and (D) left eyes, although less pronounced than in the boy. At the follow-up evaluation, fundus examination and OCT showed a different evolution of the disease in the twins. Indeed, edema persisted in both the (E) right and (F) left eyes of the boy, whereas the girl showed an atrophic evolution of the foveal region in both eyes (G and H, right and left eye, respectively).
Results of subsequent full-field electroretinography (ERG) performed under general anesthesia in accordance with the International Society for Clinical Electrophysiology of Vision protocol were normal for the photopic and combined maximal responses, whereas scotopic ERG results were slightly subnormal (Figure 2). General inhalation anesthesia with sevoflurane 4% vaporized in a mixture of 50% oxygen was administered following the protocol adopted for non-invasive ophthalmic pediatric procedures and trialed to prevent effects on ERG responses. Concomitant fundus examination confirmed the absence of schisis or pigmentary changes in the peripheral retina. These findings excluded other retinal phenotypes (Leber congenital amaurosis, retinitis pigmentosa, and cone dystrophy) associated with CRB1 mutations or X-linked retinoschisis, as initially suspected. Results of routine laboratory evaluation were negative and comprehensive infectious disease analysis ruled out any acute infection. Topical treatment with carbonic anhydrase inhibitors was administered for 4 months and subsequently suspended due to the lack of efficacy in reducing cystoid lesions and the age of the patients.
Electroretinography (ERG) results from the affected twins showing a mild amplitude reduction for scotopic response, whereas photopic and combined maximal responses are normal. Electroretinography trace from a control patient is shown for comparison.
Follow-up examination at the age of 5 years showed clinical stationary conditions for both children. Best corrected visual acuity was 20/80 and 20/100 in both eyes in the boy and girl, respectively. At the last follow-up at the age of 7 years, best corrected visual acuity was stable in the boy and 20/200 in both eyes in the girl, showing a slight decrease compared to previous examination. Spectral-domain OCT showed the persistence of the cystic lesion in the boy with a relevant increase of the foveal thickness compared to the baseline value (535 µm in the right eye and 602 µm in the left eye) and an atrophic evolution of the macular retina with a decrease of foveal thickness values (172 µm in the right eye and 183 µm in the left eye) in the girl (Figure 1). No significant differences were noted in full-field ERG recordings compared to the baseline values for both patients.
Sequencing and Bioinformatics Analysis
Genomic DNA was extracted from whole blood using a commercial kit (Blood DNA Kit E.Z.N.A.; Omega Bio-Tek Inc., Norcross, GA). Next generation sequencing was performed using the Illumina Nextera Rapid Custom Capture Enrichment method and 150 bp paired-end read sequencing on a MiSeq personal sequencer (Illumina, San Diego, CA) as the first step. Custom panel design included 234 genes known to be associated with syndromic and non-syndromic autosomal dominant, recessive, and X-linked eye disorders. The target regions for each gene were defined and consisted of all of the exons identified in National Center for Biotechnology Information RefSeq ± 15 intron flanking bp. Only RS1 was analyzed in the first step, and then we extended the analysis to the remaining genes of the panel.
Raw data in FASTQ format were analyzed to generate the final set of sequence variants using an in-house pipeline with the following modules: mapping (BWA), duplicate read removal, indel realignment, quality calibration (GATK), coverage analysis (SAMtools and BEDtools), and variant calling and annotation (UnifiedGenotyper, VarScan, and SAM-tools). Target regions with coverage less than 10 reads were also analyzed by Sanger sequencing according to the manufacturer's protocols. All variants were validated by Sanger sequencing using a Beckman Coulter CEQ 8000 sequencer (Beckmann Coulter, Milan, Italy). For sequence variant classification, we used the criteria of American College of Medical Genetics and Genomics standards and guidelines.8
The next generation sequencing genetic testing showed a mean coverage of targeted bases of 204X with 98.3% covered for at least 25X. Genetic testing revealed one new heterozygous frameshift variant: c.2562_2568del; p.(Val855*) on exon 7 of CRB1 (NM_201253.2) inherited from the father, and two heterozygous missense variants, already reported as pathogenic, c.2234C>T; p.(Thr745Met) and c.430T>G; p.(Phe144Val), on exons 7 and 2, respectively, inherited from the oocyte donor (Figure 3).
Family pedigree showing the inheritance pattern of the three CRB1 variants.
The truncating variant is considered likely pathogenic for several reasons. MutationTaster assesses the variant as “disease causing” and the Combined Annotation Dependent Depletion score predicts it to be deleterious (score: 32, the minimum threshold is 15). Because this variant is not located on the last exon, it is likely to trigger nonsense-mediated decay, as predicted also by MutationTaster, thereby preventing protein translation. The Human Gene Mutation Database lists other truncating mutations before and after the site of our mutation. This further sustains pathogenicity. Finally, in The Genome Aggregation Database, the ratio of the observed number of loss-of-function mutations to the expected number of loss-of-function mutations is 0.62, as expected for genes such as CRB1, which is associated with recessive disorders.
Given the visual loss associated with early-onset cystic maculopathy without other significant clinical signs and ERG recordings, X-linked retinoschisis was initially suspected. The only known gene matching this diagnosis is RS1 (OMIM gene 300839), which is associated with X-linked retinoschisis (OMIM disease 312700). Although one of the twins was female, this hypothesis was supported by previous reports of the RS1 gene in females with this clinical phenotype. Genetically, these females were homozygous, born to related parents,9 or heterozygous, probably due to skewed X-inactivation,10 for RS1 variants. However, the genetic test performed on our proband was negative. Based on additional clinical data and other tests/examinations, we extended bioinformatic analysis to all 234 genes in our custom-designed ophthalmic panel.
The identification of these variants associated with the early onset (3 years of age), the peculiar phenotype, and the schitic/cystoid changes at the macula led us to formulate a diagnosis of foveal retinoschisis. This is in accordance with a previous study from Vincent et al.,6 who described two families affected by foveal retinoschisis due to CRB1 variants. Interestingly, our patients seemed to present a different kind of phenotype or different stages of severity of the disease, the boy showing prevalent cystoid macular changes and the girl mainly schitic macular alteration, thus confirming the variability of clinical expression associated with CRB1 variants. Indeed, CRB1 protein plays a critical role in retinal development and CRB1 variants can determinate several forms of alteration of the retinal structure.
Furthermore, the 4-year follow-up examination showed a relatively similar macular pattern in the boy, whereas the girl presented with signs of atrophic evolution of the foveal region (Figure 1). Full-field ERG recordings were in keeping with maculopathy, but there was a slight reduction in scotopic response. Although other authors describe different electro-physiological patterns ranging from normal generalized rod and cone function6 or mild cone dysfunction11 to generalized photoreceptor dysfunction with evidence of more severe macular involvement,12 our results are in line with a prevous report where rod dysfunction was found, apparently without any peripheral lesions.7 This finding confirms that CRB1-related maculopathies can present variable electro-physiological phenotypes. Moreover, no significant effect was found in reducing schitic changes in our patients, which is different from other reports showing a positive effect of carbonic anhydrase inhibitors on cystoid lesions associated with inherited retinal disorders or other CRB1 maculopathies.
Because our patients were born from a donor oocyte, only the father was available for segregation study, but this was sufficient to reveal the biallelic configuration of the genetic variants (Figure 3). Targeted sequencing of the three variants showed that p.(Val855*) was inherited from the father and, by exclusion, p.(Thr745Met) and p.(Phe144Val) were inherited from the donor. The p.(Val855*) is a new variant described here for the first time, whereas p.(Thr745Met) and p.(Phe144Val) are already known to be associated with retinitis pigmentosa type 1213 and Leber congenital amaurosis,14 respectively. Interestingly, in Vincent et al.'s 2016 study,6CRB1-associated foveal retinoschisis was described in older patients with onset of symptoms late in the first or second decade of life. In our cases, symptoms appeared as early as 3 years of age. We therefore cannot exclude the possibility that this more severe phenotype may be due to variants in other genes that can modulate phenotype severity.15 Alternatively, the early onset of macular alteration may be due to the presence of three pathogenic variants, one on the paternal allele and two on the maternal allele.
Another important point to highlight is that the twins were born as a result of heterologous fertilization when the prediction of such a recessive eye disorder was not possible because the sequencing techniques available did not permit cost-effective screening. Such situations can only be avoided by screening of sperm and oocyte donors, and were already foreshadowed by Daar and Brzyski in 2009.16 Today, more affordable genetic screening based on massive parallel sequencing of panels of genes associated with genetic disorders (Mendeliome) is possible. This screening could be performed in cases of heterologous- and homologous-assisted procreation. Germline DNA of the sperm/oocyte donor and the legal parent who is contributing his or her gametes (heterologous fertilization), or alternatively of both parents undergoing homologous-assisted reproductive procedures, could be screened.
This screening could make it possible to determine the likelihood of recessively inherited disorders. Mendelian diseases account for 20% of infant mortality and 10% of pediatric hospitalization.17 Because the molecular basis of different genetic disorders is known, exome screening of both donor and receiver could minimize the possibility of inherited genetic diseases in newborns for people undergoing assisted reproduction. Previous studies reported the importance of genetic screening on a restricted selection of genes in avoiding transmission of genetically transmitted diseases,17,18 but better results could be achieved by analyzing the whole exome because more than 10,000 rare disorders are known to be monogenic and therefore easily avoidable by screening of both patient and donor. The development of a machine learning algorithm able to match the parents with all possible donors previously genetically characterized by exome analysis would help to exclude many diseases a priori, although difficulties may arise in excluding those diseases with poorly characterized or unknown genetic bases. In those cases, instead of excluding the possibility of the diseases, the algorithm would try to minimize the probability of disease inheritance and genetic disorders.
Finally, we think that the bank of oocytes should, theoretically, be notified for further use of the genetic material. However, in accordance with the current European legislation in the matter of privacy, we cannot do that because we are not allowed to know who the donor is and which biobank the oocytes come from. The notification could only be made by the referring clinician of the couple that undergoes the heterologous fecundation. This article highlights the necessity to modify the current legislation to speed up the process of notification.
- Chatziralli I, Theodossiadis P, Theodossiadis GP. Optic disk pit maculopathy: current management strategies. Clin Ophthalmol. 2018;12:1417–1422. doi:10.2147/OPTH.S153711 [CrossRef]
- Gohil R, Sivaprasad S, Han LT, Mathew R, Kiousis G, Yang Y. Myopic foveoschisis: a clinical review. Eye (Lond). 2015;29(5):593–601. doi:10.1038/eye.2014.311 [CrossRef]
- Kaiser PK, Friedman NJ, Pineda R II, . The Massachusetts Eye and Ear Infirmary Illustrated Manual of Ophthalmology, 2nd ed. Philadelphia: Saunders; 2004.
- Gelman SK, Gorin MB. Significant macular edema in a patient with cone dystrophy and improvement with acetazolamide treatment. Retin Cases Brief Rep. 2014;8(4):300–305. doi:10.1097/ICB.0000000000000058 [CrossRef]
- Genead MA, Fishman GA, McAnany JJ. Efficacy of topical dorzolamide for treatment of cystic macular lesions in a patient with enhanced S-cone syndrome. Doc Ophthalmol. 2010;121(3):231–240. doi:10.1007/s10633-010-9247-9 [CrossRef]
- Vincent A, Ng J, Gerth-Kahlert C, et al. Biallelic mutations in CRB1 underlie autosomal recessive familial foveal retinoschisis. Invest Ophthalmol Vis Sci. 2016;57(6):2637–2646. doi:10.1167/iovs.15-18281 [CrossRef]
- Lewis RA, Lee GB, Martonyi CL, Barnett JM, Falls HF. Familial foveal retinoschisis. Arch Ophthalmol. 1977;95(7):1190–1196. doi:10.1001/archopht.1977.04450070088006 [CrossRef]
- Richards S, Aziz N, Bale S, et al. ACMG Laboratory Quality Assurance Committee. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405–424. doi:10.1038/gim.2015.30 [CrossRef]
- Mendoza-Londono R, Hiriyanna KT, Bingham EL, et al. A Colombian family with X-linked juvenile retinoschisis with three affected females finding of a frameshift mutation. Ophthalmic Genet. 1999;20(1):37–43. doi:10.1076/opge.188.8.131.529 [CrossRef]
- Saldana M, Thompson J, Monk E, Trump D, Long V, Sheridan E. X-linked retinoschisis in a female with a heterozygous RS1 missense mutation. Am J Med Genet A. 2007;143A(6):608–609. doi:10.1002/ajmg.a.31568 [CrossRef]
- Wolfson Y, Applegate CD, Strauss RW, Han IC, Scholl HP. CRB1-related maculopathy with cystoid macular edema. JAMA Ophthalmol. 2015;133(11):1357–1360. doi:10.1001/jamaophthalmol.2015.2814 [CrossRef]
- Henderson RH, Mackay DS, Li Z, et al. Phenotypic variability in patients with retinal dystrophies due to mutations in CRB1. Br J Ophthalmol. 2011;95(6):811–817. doi:10.1136/bjo.2010.186882 [CrossRef]
- Ghofrani M, Yahyaei M, Brunner HG, et al. Homozygosity mapping and targeted Sanger sequencing identifies three novel CRB1 (Crumbs homologue 1) mutations in Iranian retinal degeneration families. Iran Biomed J. 2017;21(5):294–302. doi:10.18869/acadpub.ibj.21.5.294 [CrossRef]
- Lotery AJ, Jacobson SG, Fishman GA, et al. Mutations in the CRB1 gene cause Leber congenital amaurosis. Arch Ophthalmol. 2001;119(3):415–420. doi:10.1001/archopht.119.3.415 [CrossRef]
- Yzer S, Fishman GA, Racine J, et al. CRB1 heterozygotes with regional retinal dysfunction: implications for genetic testing of Leber congenital amaurosis. Invest Ophthalmol Vis Sci. 2006;47(9):3736–3744. doi:10.1167/iovs.05-1637 [CrossRef]
- Daar JF, Brzyski RG. Genetic screening of sperm and oocyte donors: ethical and policy implications. JAMA. 2009;302(15):1702–1704. doi:10.1001/jama.2009.1508 [CrossRef]
- Bell CJ, Dinwiddie DL, Miller NA, et al. Carrier testing for severe childhood recessive diseases by next-generation sequencing. Sci Transl Med. 2011;3(65):65ra4. doi:10.1126/scitranslmed.3001756 [CrossRef]
- Boada M, Abuli A, Clua E, et al. Genetic matching between recipients and oocyte donors. Current Trends in Clinical Embryology. 2017;4(2):52–56. doi:10.11138/cce/2017.4.2.052 [CrossRef]