Fuchs’ endothelial corneal dystrophy: fostering change in clinical care using observational data

Keywords::

• clinical monitoring
• corneal thickness
• epidemiology
• Fuchs’ dystrophy
• genetic association

Fuchs’ endothelial corneal dystrophy (FECD), a progressive disease of the corneal endothelium that results in the development of corneal edema, is estimated to have a prevalence of approximately 4% in the USA and is a leading cause for corneal transplantation, with 12,493 of the 59,431 (21%) keratoplasties (penetrating or endothelial) in 2011 performed for this condition according to the Eye Bank Association of America [1,2]. While clinical studies have led to a remarkable replacement of penetrating keratoplasty by endothelial keratoplasty for the surgical management of FECD over the past 5 years [3–6], efforts to determine the genetic risk factors underlying FECD have been pursued concurrently. Despite the prevalence of FECD, however, only recently has research focused on examining disease characteristics and comorbidities within the general population. The assembly of cohorts to advance studies of genetic variations predisposing to FECD has provided a wealth of additional epidemiological findings beyond genetic risks and has begun to provide insight into the pathophysiology of FECD.

Definitive treatment of FECD consists of corneal transplantation, although symptomatic management is used in early stages of the disease with sodium chloride drops and ointment. With long-term data, Descemet stripping endothelial keratoplasty has become the accepted procedure of choice [3,4]. A new procedure, Descemet membrane endothelial keratoplasty, may offer even better visual results than Descemet stripping endothelial keratoplasty with a lower rejection rate [5,6]. This procedure is not yet widely employed because of difficulty in donor preparation and a higher detachment rate in the immediate postoperative period. Hope for improved donor preparation and delivery techniques may lead to wider acceptance.

Surgical interventions have typically been reserved until a patient experiences a significant, persistent decrease in vision throughout the day, not simply in the morning when the cornea is most swollen. Originally, this was thought to occur when FECD reached a threshold where compensatory mechanisms such as the upregulation of endothelial cell pumping action were no longer able to offset the declining barrier function of the endothelium. More recently, examination of clinical characteristics in a cohort assembled for a large-scale genetic study of FECD demonstrated an increase in central corneal thickness with increasing severity of FECD, including at stages of FECD where stromal and/or epithelial edema was not evident on slit-lamp examination [7]. This suggests there may be clinical benefit to measuring corneal thickness even at early stages of the disease, to monitor progression and subsequent timing of surgical intervention. This approach may prove particularly useful given the individual variation in baseline corneal thickness even prior to the occurrence of any disease-related changes. Tracking changes in corneal thickness over time may better identify surgical candidates than the use of preoperative thresholds and could change standards of care to include interventions before structural changes to the corneal stroma occur and associated visual problems develop.

Recent research has also suggested an increased need to monitor FECD patients for the development of ocular hypertension/glaucoma in addition to changes in corneal thickness. Previous work suggested an association between shallow anterior chamber and angle-closure glaucoma with FECD [8]. Newer findings demonstrate an association of ocular hypertension/glaucoma with late-stage FECD, although the causal nature of the relationship remains to be elucidated [9]. The prevalence of ocular hypertension/glaucoma was not increased in subjects with early stages of FECD. These findings suggest that as FECD patients develop more severe disease, close monitoring of intraocular pressure and possible inclusion of optical coherence tomography and additional glaucoma assessments should be periodically undertaken.

Beyond these clinical observational studies, efforts are being made to identify heritable and environmental risk factors for FECD. There is a preponderance of FECD in females, although the biological mechanism is not established [10]. One study estimated the odds of FECD increased by 51% in females [11]. This same study also identified smoking as a risk factor, increasing the odds of FECD by 25%. This novel finding suggests that particular efforts should be made to encourage smoking cessation in patients with FECD or those with a strong family history. To date, smoking represents the only modifiable risk factor identified for FECD.

FECD is an inherited disease, likely following an autosomal dominant inheritance pattern [12]. Earlier research efforts utilized family-based methods and linkage mapping to identify genetic regions likely to harbor FECD susceptibility variants. From this work and, more recently, case–control association studies, several genetic risk factors for FECD have been identified. Missense mutations in the collagen type VIII, alpha 2 (COL8A2) gene, which is expressed in Descemet’s membrane, have been associated with early-onset FECD in several familial studies [13,14], although not with late-onset disease. Variations in transcription factor 4 (TCF4), originally identified through genome-wide association testing [15], have been consistently associated with late-onset FECD, although the mechanism by which TCF4 affects FECD pathology remains under study. The strongest association signal localizes to a noncoding region, suggesting a role in the regulation of TCF4 expression. The gene product of TCF4 is known to be expressed in the developing cornea and to affect the expression of zinc finger E-box-binding homeobox 1 (ZEB1), another gene harboring variants associated with increased FECD risk, although less consistently [15]. Variants in TCF4 have also been associated with changes in central corneal thickness, although this appears to be mediated through the development of FECD, rather than a direct effect [16].

Additional genome-wide association projects are underway to further clarify the genetic underpinnings of FECD. With this knowledge, a better understanding of the cellular pathways affected can be developed and tested both at the molecular and physiological levels. Such work has started in examining COL8A2 and TCF4. Knock-in mice containing a COL8A2 point mutation identified in early FECD patients demonstrate endothelial cell loss and develop guttae in a fashion analogous to the human disease process [17]. Endothelial cell loss in this model appears to be mediated by apoptotic mechanisms. With TCF4, a recent study tested for changes in endothelial cell density in young, otherwise healthy adult carriers of the TCF4 risk allele, but was unable to identify a correlation [18]. Whether corneal structural differences exist or structural changes progress more rapidly based on genetic risk requires examination. Efforts to characterize the 3D corneal shape at different stages of FECD recently identified progressive deformation of the posterior corneal surface with increasing disease severity [19], and similar studies should be possible with respect to risk genotypes.

Equally important will be understanding the interaction of genetic variations with epidemiological factors such as smoking or with other genetic variations such as those directly impacting corneal thickness, as this information will allow for improved risk stratification and clinical management. Some evidence for potential interaction between genetic loci has already been observed in a family-based study in which family members with two affected parents demonstrated more rapid progression of FECD [20]. As additional genetic and environmental risk factors are identified, it may be possible to create a composite risk score for patients, allowing for the identification of those who may benefit from more frequent clinical monitoring. Current evidence shows the TCF4 risk variants are common in the general population (frequency of 18–19% in those of northern European ancestry) [16], indicating the presence of this risk factor alone is not sufficient for the development of FECD and that other factors must also contribute. Prospective studies are needed to determine the proportion of those with TCF4 risk variants that proceed to FECD and to examine this population for additional characteristics, hereditary and environmental, predisposing them to the disease.

The continued efforts to identify genetic risk factors for FECD and the subsequent assembly of cohorts to this end have created remarkable opportunities to explore the FECD disease process beyond identifying heritable risk modifiers. As more clinical and epidemiological information is integrated with findings from genetic studies, broader conclusions regarding the origins of FECD will be able to be drawn. In light of the current findings, clinicians may wish to consider several modifications to their present clinical monitoring including:

• • Asking patients with early signs of FECD about their family history;

• • Measuring corneal thickness changes longitudinally by ultrasonic pachymetry in addition to usual monitoring of progression by slit-lamp biomicroscopy;

• • Monitoring patients for ocular changes, such as glaucoma/ocular hypertension, associated with FECD severity;

• • Counseling patients to cease smoking cigarettes;

• • Using genetic testing as it becomes available (proxy markers at TCF4 or COL8A2) to see if younger family members are vulnerable to future disease.

Counseling of patients should include that FECD is an inherited disease that is passed on from generation to generation, with women at greater risk. Additional counseling may incorporate that scientists have discovered several genes that cause disease, including COL8A2 and TCF4, and more risk factors likely remain to be discovered. Some people only inherit mutations at one gene, while others may inherit more than one change, which may be why some people have severe disease, and others remain with fewer symptoms.

Prospective studies to evaluate the effect of integrating these practices into improving clinical outcomes are still needed. We hope, through continued efforts such as those highlighted above, refinements to the understanding of FECD continue and clinical care further advances.

Financial & competing interest disclosure

This work was supported by the National Eye Institute (grant numbers: R01EY16482, R21 EY015145 and P30 EY11373), Research to Prevent Blindness and the Ohio Lions Eye Research Foundation. The authors have no other 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 apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.