The impact of genetic testing for microphthalmia, anophthalmia, and coloboma: experiences from a translational approach

KEYWORDS:

• Anophthalmia
• coloboma
• genetics
• microphthalmia
• translation

1. Introduction

Microphthalmia (small eye), anophthalmia (missing eye), and coloboma (disruption of optic fissure closure) (MAC) are a spectrum of structural developmental eye anomalies. Approximately 6–30 per 100,000 children per year are affected by these disorders, accounting for up to 20% of childhood visual impairment [1,2]. These phenotypes sometimes occur in isolation, but they can be part of more complex disorders, including both additional ocular and systemic features [2,3]. While MAC can be attributable to environmental factors such as viral infections and alcohol, the majority result from genetic alterations [4,5], and it is in our understanding of the genetic architecture underpinning these disorders that our knowledge has advanced rapidly.

It has been two decades since Fantes et al. demonstrated that variants in the transcription factor SOX2 cause bilateral anophthalmia [6]. While SOX2 variants continue to be the most common cause of MAC, explaining 10–15% of cases, subsequently the number of additional genes implicated (to varying degrees of evidence) has expanded to over 100 (see Web resources) [2,7]. This has highlighted multiple pathways in these disorders including BMP signaling, Sonic Hedgehog signaling, and the retinoic acid pathway. However, while several genes account for a significant proportion of cases (particularly SOX2, OTX2, STRA6, GJA8), the majority explain only a limited number of patients.

2. Patient clinical testing

Perhaps the most obvious impact of the last 20 years of MAC genetics research has been its translation into the clinical diagnostic setting. Within the UK National Health Service this is mediated through expert reviewed gene lists for different phenotypes, each gene classified according to the strength of evidence linking it to disease (see Web resources) [7]. Currently, 50 genes meet strict diagnostic-grade criteria on the ‘Ocular coloboma’ and ‘Anophthalmia or microphthalmia’ panels, with a further 57 having more limited degrees of evidence. However, additional genes associated with these phenotypes are included on other panels, indicating the complexity of translating the genetic findings of varied research teams for such heterogeneous and phenotypically complex disorders. Despite these challenges, such translational efforts provide considerable diagnostic and counseling benefits to affected families. Thus, it is currently estimated that about 60% of severely affected individuals with MAC (reduced to 20–30% when including unilateral or mildly affected cases), but <25% of those with coloboma, receive a genetic diagnosis [2,8,9]. Such striking advances are obviously of great encouragement, but yet indicate considerable work remains to be done. Therefore, how can we seek to continue improving our knowledge of these disorders, and consequently patient care?

3. Collaboration

As indicated, the number of cases explained by any given MAC-associated gene is relatively low, making it challenging to identify new candidates and compile sufficient evidence for them to be considered diagnostic, further compounded by the rarity of the disorders themselves. Therefore, it is vital to generate and maintain collaborative networks in order to have access to enough patients and data to generate clinically impactful outcomes. This was recently exemplified in our own research on CAPN15, where a patient series was collated through ongoing relationships between researchers, conference presentations, and the uploading of a pre-publication manuscript to bioRxiv [10]. This led to the gene, previously known for little more than its role in sea slug memory, exceeding the UK health-care criteria for diagnostic status. The key point is that no single group would have been able to achieve this alone, requiring as it did the use of informal networks and the willingness to present unpublished data at conferences, itself a challenge in a world of increasingly competitive science where the temptation to protect data is strong. Yet such willingness to dare to reach out, including by sharing unpublished work, may be a necessary risk, but one which can reap significant rewards. Hence, means by which to reify informal connections through routes such as the UK Eye Genetics Group and the recently established international Genetics of Ocular Development Network are vital (see Web resources).

4. The importance of phenotyping

Second, MAC research has reaffirmed the need for careful and thorough phenotyping and therefore the importance of the relationship between geneticists and physicians to better understand the causes of these disorders, the phenotypic spectra of specific genes, and phenotype-genotype correlations. This is true not only for research, but also for translation as such careful observations have important implications for the appropriate counseling of families. For example, FZD5 has recently been implicated in ocular coloboma, but in a mixture of syndromic and non-syndromic individuals, and with variable penetrance [11,12]. However, further research indicated that variants in FZD5 are linked only with the ocular phenotypes of individuals, additional phenotypes being attributable to changes in other genes, and variable penetrance sometimes being due to the overlooking of subtle features which were identified upon re-phenotyping of carrier parents [3]. Such care in determining what is attributable, and what is not, to a specific genetic change is particularly important for these complex, often syndromic disorders, requiring the careful reporting of phenotypes over multiple publications. Unfortunately, rather than making such vital data prominent within the main text of a manuscript, it is all too easy for a researcher to focus on the phenotypes for which they have a keen interest, relegating others to a ‘secondary’ status and a small place in a supplemental table. Thus, attention to phenotypic detail will be vital to understand the extent of the effect of pathogenic variants and potential issues regarding penetrance and so forth.

5. Beyond the coding region

Third, while there has been a significant harvest of knowledge from research focussing on the coding regions of genes in relation to MAC disorders, to continue our advances we must start stepping beyond these confines to consider other mechanisms of disease. For example, PAX6, a gene vital for eye development, is in part regulated by the short (800bp) SIMO element. This sequence is located approximately 150kb downstream of PAX6 within an intron of the neighboring ELPA4, and both single nucleotide changes and deletions of SIMO cause aniridia [13,14]. Similarly, in mice Tsang et al. identified several tissue-specific enhancers flanking MAB21L2, a gene associated with microphthalmia and skeletal anomalies. These included a short 172bp sequence 3’ of the gene which drives eye-specific expression, again indicating the potential for such regions as a source of diagnoses [15]. Other interesting virgin territory for MAC research could include the role of topologically associating domains (TADs), known to be significant in mammalian development, and the role of ribosomes in modulating gene expression [16,17]. While technically challenging, such studies will likely help further prise apart the complex interactions between genes in eye development, particularly given the number of transcription factors associated with MAC phenotypes. Indeed, as our knowledge grows, it is likely that we will have to move beyond the paradigm of a ‘one patient-one variant’ understanding of MAC, to account for modifying effects such as genetic background and environment. As an example of the latter, a recent study by Plaisancié et al. has indicated that the maternal genotype of RBP4 may affect the severity of MAC phenotypes in some individuals [18]. Such advances may also be supported by the adoption of emerging technologies such as artificial intelligence-based tools [19], which are already showing promising results with respect to the analysis of genetic data, but will require careful assessment in regard to their ethical implications.

6. Our role as researchers

Finally, it is worth acknowledging how translational approaches to genetics challenge our self-understanding as researchers. Basic research has the potential to generate a form of cognitive distance from the patients who will ultimately benefit. However, historically in MAC research the link with the patient has been more direct, both in terms of the translational impact and the collaborative nature of the research, with close ties between researchers, clinicians, and families. This challenges researchers to understand their subjects as more than ‘consenting tubes of DNA,’ instead recognizing them as persons and families. This can create an additional sense of purpose for the research, allowing the appreciation of tangible impacts beyond the published papers which can so often be the focus of academia. Borrowing a term from the theologian Dietrich Bonhoeffer, the research more clearly becomes understood as ‘vicarious representative action’; action performed on behalf of others (patient and family), which they themselves are incapable of, rather than for personal gain [20]. Indeed, as hopefully has become apparent, advancement in MAC research has been dependent on such reciprocal vicarious representative action between various partners including geneticists, developmental biologists, bioinformaticians, clinicians, and families, each being vital to the success that has been enjoyed. Those who are different are bound by a mutual dependency.

However, this closeness can also confront individual scientists with ethical questions for which they are unprepared or may take for granted, such as the application of their findings to prenatal testing or the validity of experiments utilizing human embryonic tissue. Such questions are applicable to a variety of research, but are brought into crystal clarity in the face of genetic translation. A theme of MAC research is communal effort, and here collaborations beyond the scientific sphere will be important through engagement with colleagues of other specialties including philosophy, ethics, and theology.

7. Conclusion

The advances in our understanding of the genetics of MAC disorders and their successful translation into the clinical setting have made for an exciting 20 years in the field. While the task of furthering our knowledge is likely to be challenging as alternative mechanisms increasingly come to the fore, the habits of collaborative research prevalent in the field will provide a firm basis for continued success. However, it remains important that these advances are not limited to the wealthier developed countries, but are shared in partnership with developing nations where clinical genetic testing for inherited disorders remains sparse. As a significant strength of MAC research has been its underlying partnerships, the formation of such new networks will undoubtedly help contribute to an exciting future, both for basic and clinical genetics, if we continue to work with and for one another.

Declaration of interests

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Web resources

Genetics of Ocular Development (GoOD) Network: https://www.goodsoc.org/

PanelApp: https://panelapp.genomicsengland.co.uk/

UK Eye GeneticsGroup (UK-EGG): https://ukegg.com/

Additional information

Funding

This work was generously supported by a grant from Baillie Gifford.