We investigated the retinal phenotype of ADD, both cross-sectionally and longitudinally. This represents the largest cohort of affected patients reported to date. The retinal phenotype has been well described in previous reports and is in keeping with our findings.
7 We were able to broadly characterize the clinical and retinal imaging features. An unexpected finding was an apparent sex imbalance in our cohort, with a male-to-female ratio of 1:3.4, which has been recently reported by our group in a range of genotypes, including
EFEMP1.
24 Interestingly, some studies in AMD (to which
EFEMP1 overexpression may contribute to neovascularisation
16) have pointed out a higher likelihood in women of early AMD, and a potentially higher burden and risk for the neovascular form of the disease if compared to men.
17–19 The reason of this apparent imbalance in the present cohort is unclear and may have been subject to selection bias.
From a clinical perspective, ADD remains a relatively late-onset disease with symptoms starting typically in the 40s, most of which are mild, such as central vision distortion, albeit 70% of the cohort presented with decreased vision. Our data suggest that ADD is not associated with any specific type of refractive errors. Similarly, most patients maintain functional vision throughout the disease course. Clinically, two unrelated patients had optic nerve drusen, which are likely to be incidental findings. Surprising, though, was the relatively high prevalence of CNV formation, which was much higher than previously reported.
7 Reactivation of CNV can happen even in late-stage disease with concurrent subretinal fibrotic tissue. Although these data are incomplete, given the uncertain number and frequency of anti-angiogenic injections, there were significantly fewer injections performed than would be expected for the neovascular form of AMD, which could suggest a milder behavior of the CNV associated with ADD. The presence of CNV is associated with a significantly worse natural history and overall vision loss. In light of this, we recommend regular follow-up of affected individuals (e.g., annually), as well as adequate patient counseling with thorough explanation of symptoms suggestive of CNV and an immediate visit to an eye service unit.
We performed in-depth retinal imaging, which is imperative for patient prognostication. The qualitative classification used in this study does carry some degree of subjectiveness. Therefore, for example, grouping OCTs of affected individuals is to be taken as a “soft” recommendation, albeit it does help in distinguishing disease stages, particularly given its direct association with BCVA. Similarly, quantitative analyses of OCTs are not straightforward in the context of ADD. Establishing measurements of PED area in early disease phases (i.e., OCT group 1) is an achievable task, which tends to become progressively more difficult, particularly when the drusen begin to coalesce and in the presence of concurrent outer retinal/RPE abnormalities. Nevertheless, with the advances in imaging device capabilities, such as follow-up mode—allowing the exact same retinal loci to be imaged longitudinally—and faster scanning speeds (OCT2, 85 kHz), it is feasible to quantify with accuracy, at least, the ONL thickness, which is inversely correlated with BCVA as shown in our results. Indeed, the structural retinal changes appear to follow a specific natural history pattern: (1) well-defined PED, which (2) coalesce and eventually lead to loss of outer retinal lamination and subretinal fibrosis, after which (3) RPE atrophy develops.
Future prospective studies would be essential to further establish ideal prognostic measurements and endpoints for future clinical trials. The universal presence of the point variant R345W in affected individuals opens up avenues for treatment such as, for example, (non-vector delivered) gene editing. Given the chorioretinal pathogenesis of ADD, it is not unfeasible that suprachoroidal delivery of gene therapy products could provide an attractive route for a more posterior and circumferential distribution of investigational medical products.
20,21 It remains to be established what benefits patients might experience with therapy, given the relatively slow disease progression and relatively good prognosis. The use of high-resolution imaging techniques, such as adaptive optics scanning light ophthalmoscopy,
22 and devices to assess retinal sensitivity in detail, such as fundus-guided perimetry,
23 may provide further insights into the structural and loci-specific functional integrity of the photoreceptor mosaic, ideally in prospective, robust, long-term natural history studies.
This study has limitations, particularly given its retrospective nature. Follow-up intervals and protocols for BCVA acquisition were not standardized, nor were the imaging techniques (i.e.,. FAF and OCT were unavailable in some visits). Moreover, no further functional data were available, which remains a gap that must be addressed.