Comparison of vascular parameters between normal cynomolgus macaques and healthy humans by optical coherence tomography angiography

The retina plays a vital role in vision. The metabolic activity of the retina is higher than other human tissues [1, 2]; therefore, the vascular circulation of the retina is complex. A variety of techniques have been used to measure retinal perfusion. Fundus fluorescence angiography (FFA) is routinely used to evaluate retinal vascular retinopathy [3] as it can analyse the choroid-retinal vasculature and show vascular leakage and neovascularization [4]. However, this method has limitations, such as the intravenous injection of contrast agents that can lead to some side effects [5, 6]. Additionally, further details of the deeper capillary plexus are unable to be extracted from FFA images due to limited depth perception. In contrast, optical coherence tomography angiography (OCT-A) is a novel imaging method [7‐9]that detects the flow of blood via intrinsic signals without requiring an intravenous agent. OCT-A visualizes and divides the retina into various layers, clearly showing the capillary vessels in each layer. The ability of OCT-A to quantify the blood flow and vascular density of the retinal choroidal circulation is essential to the study of ophthalmology worldwide.

Due to the close similarity to humans in terms of optical structure and function, nonhuman primates such as cynomolgus macaques play a crucial role in ophthalmic disease research [10‐13]. The FFA and OCT-A vascular parameters for normal humans have been previously reported. [1, 14, 15] Researchers have reported flow perfusion parameters for normal rhesus monkeys [16] and chorioretinal structure and stratification of cynomolgus macaques and rhesus monkeys by SD-OCT. [17, 18] However, very little literature is available on the vascular density parameters for the macular and optic disc of ocularly normal adult cynomolgus macaques, researchers have not clearly determined how OCT-A may be used for disease management, particularly in animal models, and the comparison of VD between healthy humans and cynomolgus macaques using OCT-A has not been conducted. Our study aimed to evaluate macular vascular circulation parameters in normal adult cynomolgus macaques using OCT-A. For a better use of nonhuman primate models in studies of ophthalmic disease, we compared the vascular parameters between healthy humans and cynomolgus macaques.

OCT-A is a novel imaging method [6, 19]. OCT-A was developed as an extension of OCT that clearly shows the vascular circulation in the retina and choroid [20, 21]. FFA is a conventional, qualitative technology used to diagnose retinal vascular diseases, such as diabetic retinopathy (DR) and central retinal vein occlusion (CRVO) [22, 23]. Using a special intravenous dye, FFA is one of the most valuable methods for evaluating the retinal vasculature. FFA records the filling process of the choroid-retinal vasculature and has a wide field of view.

Unlike FFA, OCT-A detects blood flow via an intrinsic signal without the use of any intravenous agents, thereby avoiding any serious side effects arising from fluorescent dyes. OCT-A saves time with quick scanning speeds. FFA unquestionably provides two-dimensional (2D) images of vascular circulation [3], but with a limited depth perception and detailed investigation of vessels [4]. The development of OCT-A solves these problems. OCT-A visualizes blood flow and divides the retina into various layers, clearly showing capillary vessels in each layer, not just transparent vessels. Therefore, deeper capillary plexuses, which are often mistaken for background choroid fluorescence in FFA, have been observed using OCT-A [4, 24]. Interestingly, the OCT-A image in which the RPC network is readily visible and the fluorescein angiographic image of the same region in our study showed considerable similarities with research by Richard F in humans [4]. No previous study has shown the RPC network using fluorescein angiography [25].

Researchers can quantitatively analyse vessel parameters using high-resolution imaging with OCT-A. Therefore, in the present study, we further compared the VD of the macular, optic disc and surrounding regions using OCT-A. two levels of choroid-retinal capillary networks are present in the macular of cynomolgus macaques, including the superficial layer, the deep retinal layer. Because the VD of choroid capillary can be disturbed by the inner vascular network, so we only compare the VD of superficial and deep retinal layer and find out that the deep layer had the higher VD in the two groups. These results are consistent with Florence Coscas’ research in humans [1]. One possible explanation for this finding is that the deep layer is formed by a homogenous capillary vortex [26], whereas the superficial layer is formed by transverse capillaries alone. Additionally, the superficial layer may artificially influence the VD assessment of the DCP [1]. The lowest VD is observed in the foveal area than in other sections of the retinal layer, probably due to the foveal avascular zone (FAZ) [21].

The VD of the peripapillary region was much higher in the RPC layer. In RPC layer of optic disc, the VD in the N quadrant was lower than the IT quadrant. Overall, OCT-A provides better images of the RPC network. The blood supply associated with early optic disc lesions in patients with glaucoma is derived from the microcirculation of the posterior ciliary artery; hence, an observation of the RPC layer is beneficial in the early diagnosis of glaucoma [27].

Despite the novelty of OCT-A, particularly in assessing different diseases [28‐32], researchers have not clearly determined how OCT-A may be used for disease management, particularly in animal models. Currently, nonhuman primates, particularly cynomolgus macaques, play a very important role in the research of ophthalmic diseases due to their similarities to humans in terms of the optical structure and function [8‐11]. Therefore, a better understanding of novel OCT-A parameters and a comparison of differences between healthy humans and cynomolgus macaques will help establish these techniques, particularly OCT-A, as methods for the diagnosis of diseases in animal models.

We further compared the VD of normal humans and cynomolgus macaques in the area of the RPC network and the layers of the superficial and deep retina, and all of the data showed some similarities and significant differences between two groups. In macular, the VDs of the superficial and deep layer are different in both cynomolgus macaques and healthy humans. In the fovea area of superficial and deep layer, VD of healthy human is much higher. Significant differences in the RPC layer of the optic disc were observed between the two groups, healthy humans present higher VDs in various sections. Differences between two groups of VD are obvious in the macular and optic discs. These meaningful similarities and differences should take into account in the animal researches about human optical diseases, especially vascular diseases in macular and optic disc, such as glaucoma.

In conclusion, we evaluated macular vascular circulation parameters in normal adult cynomolgus macaques using OCT-A and analysed the VD parameters of the choroid-retinal vasculature and compared the vascular parameters between healthy humans and cynomolgus macaques. Obviously, the retinal structure of cynomolgus macaques was very similar to healthy humans; thus, we can use this animal model to better study the development of human optical diseases. However, some differences in the VD were observed between the two groups, indicating that when we use animal models to study optical diseases, we should also consider the functional differences between animals and humans and take these into account in the animal researches about human optical diseases, especially vascular diseases in macular and optic disc. Overall, our research provides the normal vascular parameters of cynomolgus macaques and healthy humans, promoting the establishment of an eye parameter database for nonhuman primates and animal models of ophthalmic diseases. However, our study has certain limitations that must be addressed. For example, we used cynomolgus macaques as the experimental animal and we did not clearly determine whether these results are applicable to other species. Our future goal is to conduct an extensive study across different ages, genders and species, comparing the similarities and differences in OCT-A parameters between healthy humans and nonhuman species to facilitate the research of ophthalmic vascular diseases.