Heterogenous Thinning of Peripapillary Tissues Occurs Early During High Myopia Development in Juvenile Tree Shrews

Abstract

Myopia is an independent risk factor for glaucoma, but the link between both conditions remains unknown. Both conditions induce connective tissue remodeling at the optic nerve head (ONH), including the peripapillary tissues. The purpose of this study was to investigate the thickness changes of the peripapillary tissues during experimental high myopia development in juvenile tree shrews. Six juvenile tree shrews experienced binocular normal vision, while nine received monocular −10D lens treatment starting at 24 days of visual experience (DVE) to induce high myopia in one eye and the other eye served as control. Daily refractive and biometric measurements and weekly optical coherence tomography scans of the ONH were obtained for five weeks. Peripapillary sclera (Scl), choroid-retinal pigment epithelium complex (Ch-RPE), retinal nerve fiber layer (RNFL), and remaining retinal layers (RRL) were auto-segmented using a deep learning algorithm after nonlinear distortion correction. Peripapillary thickness values were quantified from 3D reconstructed segmentations. All lens-treated eyes developed high myopia (−9.8±1.5 D), significantly different (P<0.001) from normal (0.69±0.45 D) and control eyes (0.76±1.44 D). Myopic eyes showed significant thinning of all peripapillary tissues compared to both, normal and control eyes (P<0.001). At the experimental end point, the relative thinning from baseline was heterogeneous across tissues and significantly more pronounced in the Scl (−8.95 ± 3.1%) and Ch-RPE (−16.8 ± 5.8%) when compared to the RNFL (−5.5 ± 1.6%) and RRL (−6.7±1.8%). Furthermore, while axial length increased significantly throughout the five weeks of lens wear, significant peripapillary tissue thinning occurred only during the first week of the experiment (until a refraction of −2.5±1.9 D was reached) and ceased thereafter. A sectorial analysis revealed no clear pattern. In conclusion, our data show that in juvenile tree shrews, experimental high myopia induces significant and heterogeneous thinning of the peripapillary tissues, where the retina seems to be protected from profound thickness changes as seen in Ch-RPE and Scl. Peripapillary tissue thinning occurs early during high myopia development despite continued progression of axial elongation. The observed heterogeneous thinning may contribute to the increased risk for pathological optic nerve head remodeling and glaucoma later in life.

1. Introduction

Myopia, or nearsightedness, has been increasing dramatically (Dolgin, 2015; Holden et al., 2016; Lin et al., 2004; Matsumura et al., 2020; Vitale, 2009; Yusufu et al., 2021) and it is expected that around half of the world population will be myopic by 2050 (Holden et al., 2016). High myopia, defined in humans as spherical equivalent ≤ −6 diopters (D) and/or axial length (AL) ≥ 26 mm, is a leading cause of legal blindness in the developed world (Holden et al., 2016). While the refractive component of myopia is treatable, the increased risk of developing blinding comorbidities, including myopic macular degeneration and glaucoma, remains a major concern, particularly with high myopia. Glaucoma is a leading cause of irreversible blindness, and its prevalence is increasing along with the epidemic rise of myopia (Chang, 2011; Czudowska et al., 2010; Mitchell et al., 1999; Qiu et al., 2013; Tham et al., 2014; Xu et al., 2007), making both conditions a global health concern.

While intraocular pressure (IOP) is the major risk factor, myopia is an independent risk factor for glaucoma. The risk of developing glaucoma increases by approximately 20% for each diopter of myopia (Ha et al., 2022). Glaucoma in myopic eyes is of particular interest due to overlapping structural remodeling and functional changes caused by both conditions (Chang and Singh, 2013; Nicholas Y Q Tan et al., 2019). Furthermore, glaucoma in myopic eyes can occur with normal IOP, posing challenges for detection and follow-up of glaucoma in myopic eyes (Chang and Singh, 2013; Nicholas Y Q Tan et al., 2019). Nevertheless, the link between both conditions remains largely unknown.

Connective tissue remodeling at the optic nerve head (ONH), which includes the lamina cribrosa (LC) and peripapillary tissues, is of particular interest due to its potential role in the development and progression of both myopia and glaucoma (Burgoyne, 2011; Burgoyne et al., 2005, 2004; Grytz et al., 2020; Roberts et al., 2009; Yang et al., 2011, 2007). Myopia induces structural changes at the ONH, including tilting, torsion, area increase, and shearing between connective and neural tissues, as well as stretching and thinning of different ocular layers due to axial elongation (Abbott et al., 2011; Jeoung et al., 2020; Khalafallah et al., 2023; Lee et al., 2018; Park et al., 2019; Nicholas Y.Q. Tan et al., 2019). Growing evidence indicates that tissue thinning and other deformations of the posterior sclera in myopic eyes may be associated with microvascular dropout, LC deformations, retinal nerve fiber layer (RNFL) thinning, visual field defects, and development of normotension glaucoma (NTG) (Jeon et al., 2020; Park et al., 2019, 2012; Sawada et al., 2017; Sung et al., 2018). Consequently, connective tissue remodeling at the ONH during myopia development early in life might increase the risk for pathologic ONH remodeling and glaucoma later in life (Grytz et al., 2020). The mechanistic and diagnostic intersection between myopia and glaucoma raises the need for a detailed investigation of the morphological ONH changes in these conditions.

Several studies have investigated the morphological changes of the peripapillary tissues in myopic eyes. Optical coherence tomography (OCT) (Akagi et al., 2013; Shinohara et al., 2016) and histological studies (Jonas et al., 2014a; Jonas and Xu, 2014; Ren et al., 2009) have shown thinning and protrusion of the peripapillary sclera (Scl) in highly myopic eyes. Similarly, myopia has been associated with thinning of the peripapillary RFNL (Choi and Lee, 2006; Hwang et al., 2012; Kim et al., 2010; Lee et al., 2019; Mohammad Salih, 2012; Tai et al., 2018), ganglion cell layer complex (Sezgin Akcay et al., 2017; Xiao et al., 2022), and choroid (Abdolrahimzadeh et al., 2017; Gupta et al., 2015; Jiang et al., 2015; Read et al., 2015), in addition to choroidal microvascular dropout (Na et al., 2020; Park et al., 2018). However, other studies reported no correlations between axial length and the thicknesses of peripapillary RNFL (Hoh et al., 2006; Kim et al., 2022; Lee et al., 2019; Sezgin Akcay et al., 2017), total retina (Garcia-Valenzuela, 2000; Garcia-Valenzuela et al., 2002) or peripapillary choroid (Kim et al., 2022). When examined closely, most of the above mentioned studies investigated the effect of myopia in adult subjects (Akagi et al., 2013; Choi and Lee, 2006; Garcia-Valenzuela, 2000; Garcia-Valenzuela et al., 2002; Hwang et al., 2012; Jonas et al., 2014a; Jonas and Xu, 2014; Kim et al., 2010; Lee et al., 2019; Mohammad Salih, 2012; Na et al., 2020; Park et al., 2018; Ren et al., 2009; Shinohara et al., 2016; Tai et al., 2018), where myopic thickness changes may be influenced by additional changes due to aging and/or preperimetric glaucoma (Jeoung et al., 2020). Kim et al. (Kim et al., 2022) reported no significant thickness changes of RNFL and choroid over four years of myopia progression in Korean children that were already myopic at the time of enrollment. Consequently, it remains unclear if the observed thinning seen in adult myopes (Akagi et al., 2013; Choi and Lee, 2006; Garcia-Valenzuela, 2000; Garcia-Valenzuela et al., 2002; Hwang et al., 2012; Jonas et al., 2014a; Jonas and Xu, 2014; Kim et al., 2010; Lee et al., 2019; Mohammad Salih, 2012; Na et al., 2020; Park et al., 2018; Ren et al., 2009; Shinohara et al., 2016; Tai et al., 2018) occurred during the development of myopia or afterwards. Finally, no study has investigated the thickness changes of the retina, choroid, and sclera in the same cohort of subjects, limiting our understanding of how thickness changes differ across tissues (Cui et al., 2021; Shin et al., 2019). Therefore, translational studies are needed to fill in those gaps and provide a better understanding of ONH remodeling in myopia.

Tree shrews are small mammals closely related to primates with well-developed visual systems that undergo rapid maturation (Norton, 1999). With an extensive database available for normal visual development and experimental myopia, the tree shrew serves as a reliable model for investigating tissue changes associated with development of myopia (Norton, 1999; Siegwart and Norton, 2010, 2005, 1998). Additionally, the recently established tree shrew model of glaucoma provides a unique opportunity to investigate the interplay between myopia and glaucoma and the underlying ONH remodeling in both conditions (Samuels et al., 2018). In a previous study, experimental high myopia development induced relative deformations between the Bruch’s membrane opening (BMO) and the anterior scleral canal opening (ASCO) along with border tissue configuration changes from internally to externally oblique (Khalafallah et al., 2023). Notably, the change to an externally oblique configuration was not uniform across the ONH. Instead, it exhibited a distinct asymmetric pattern, with a greater prevalence in sectors that were close to the posterior pole (Khalafallah et al., 2023). In the current study, we aim to quantify and characterize the longitudinal thickness changes of the peripapillary tissues during experimental high myopia development in juvenile tree shrews. In so doing, we expand our understanding of the remodeling mechanisms underlying high myopia development.

2. Methods

2.1. Experimental Groups

Northern tree shrews (Tupaia Belangeri) bred and raised at The University of Alabama at Birmingham (UAB) were used for this study. Tree shrews were housed in individual cages with continuous access to water and dry food. The study protocol was approved by the Institutional Animal Care and Use Committee of the UAB and followed the Association of Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research. Juvenile tree shrews were randomly assigned to one group with binocular normal visual experience (n=6 animals) and another group with monocular −10D lens treatment (n=9). Lens treatment occurred from 24 to 59 days of visual experience (DVE) to induce experimental high myopia in one eye with the other eye serving as a control. These groups represent a subset of a larger cohort of animals used in a previous study (Khalafallah et al., 2023). Insufficient OCT quality, typically caused by eye movements during OCT imaging, hindered reliable quantification of peripapillary thicknesses and led to the exclusion of six animals (three myopic and three normal animals) from the original cohort. Details regarding the lens treatment were reported previously (Khalafallah et al., 2023). Briefly, a dental acrylic pedestal was surgically implanted on the cranium of all animals at 21 DVE with subsequent fixation of an aluminum goggle frame to the pedestal at 24 DVE.

2.2. Refractive and Biometric Measurements

Early studies by Norton et al. established the standards for reliable refractive and biometric measurements in tree shrews (Norton et al., 2010, 2003; Siegwart and Norton, 2010). Since cycloplegic and non-cycloplegic refractions have been reported to be essentially identical, non-cycloplegic refractive (Nidek ARK-700A infrared auto-refractor, Marco Ophthalmic, Jacksonville, FL) and biometric (Lenstar LS-900 optical biometer, Haag-Streit USA, Mason, OH) measurements were conducted on conscious animals every day around 10:00 AM in a dark room. Installation of a dental acrylic pedestal on the skull allows for control of the head during these conscious measurements. The raw Lenstar data were analyzed using tree shrew-specific refractive indices as detailed in prior work (El Hamdaoui et al., 2019).

2.3. Optical Coherence Tomography (OCT) Imaging

This study utilized the Spectralis OCT2 (Heidelberg Engineering Inc., Germany) and its enhanced-depth imaging mode to acquire 48 radial B-scans through the center of the optic nerve head (ONH). Baseline scans were obtained at 24 DVE, and subsequent scans were acquired on a weekly basis using the follow-up mode of the Spectralis OCT. To minimize eye movement during imaging, the animals were anesthetized with a combination of isoflurane inhalation (1–3%) and intramuscular xylazine injection (7.5 mg/kg). A custom-built bed with an attached heating pad was used to position the animal in front of the OCT. Pupillary dilation was achieved through topical administration of 2.5% phenylephrine hydrochloride (Akorn, Inc., Lake Forest, IL, USA) and 1% tropicamide (Akorn, Inc., Lake Forest, IL, USA). Rigid contact lenses (plano sphere, base curve: 4.0 mm, diameter: 6.3 mm) were placed on the cornea, and the OCT camera was aligned perpendicular to the limbal plane. The camera position was adjusted so that the peripapillary tissues appeared flat in a circular scan pattern indicating a perpendicular alignment of the OCT, and the radial scan pattern was centered on the ONH. The camera focus and reference arm length were adjusted so that the LC was in best focus during imaging.

To account for lateral magnifications, nonlinear distortion corrections of each OCT scan were employed (Grytz et al., 2022). After that, the peripapillary Scl, choroid-retinal pigment epithelium (Ch-RPE) complex, RNFL, and remaining retinal layers (RRL) were automatically segmented using a deep learning algorithm (Girard et al., 2011). Thickness values of different tissues were obtained from 3D reconstructed tissue segmentations and averaged over a 50-μm band starting at 1000 μm distance from the anterior scleral canal opening centroid (Figure 1). In addition to these global values, sectorial thickness profiles have been investigated. We positioned the animal’s head and OCT camera so that the horizontal B-scan would be parallel to the horizontal nasal plane. Using the horizontal axis of the enface image (Figure 1A) as nasal/temporal axis, eight 45° sectors were defined for thickness quantification (starting from the temporal side as 0° and going in a clockwise fashion): temporal (T), superotemporal (ST), superior (S), superonasal (SN), nasal (N), inferonasal (IN), inferior (I), and inferotemporal (IT). Note that the tree shrew retina has no fovea that can be used as a landmark to define the sectors.

Figure 1:

Thickness calculation for peripapillary tissues. A: Enface confocal scanning laser ophthalmoscopy image of a right eye showing the 48 radial scans centered on the optic nerve head. B: Distortion-corrected B-scan (marked by the thick green line in Figure 1A) showing the auto-segmented interfaces of the peripapillary tissues. The yellow circle and band in 1A and 1B, respectively, refer to the 50-μm band location used for thickness quantification. C: Three-dimensional reconstruction of the peripapillary tissue interfaces used to calculate the thickness values. Abbreviations: S, superior; I, inferior; N, nasal; T, temporal; ST, superotemporal; IN, inferonasal; RNFL, retinal nerve fiber layer; RRL, remaining retinal layers; Ch-RPE, choroid-retinal pigment epithelium complex; Scl, sclera; ASCO, anterior scleral canal opening.

2.4. Statistical Analysis

Statistical analyses were performed using SAS 9.4 (SAS Institute, Cary, NC). To determine the significance of continuous outcomes, a generalized linear mixed-effects model (GLMM) with Bonferroni correction was employed for all statistical comparisons. The GLMM incorporated main effects (eye group and time effects), along with interaction terms (eye group*time). Apart from comparing various groups, we also investigated whether the changes in tissue thickness occurred in concert with axial elongation. For this purpose, we compared weekly AL and thickness changes from one time point to the next one within individual groups. Furthermore, we explored whether the thinning of different tissues would exhibit uniformity within each group globally and in individual sectors. Data are presented as mean ± standard deviations and all statistical tests were two-tailed with the level of statistical significance set at a P value of less than 0.05.

3. Results

3.1. Refractive and Biometric Measurements

A significant group interaction with time (P<0.001) has been observed for AL, vitreous chamber depth (VCD) and refractive measurements. At 59 DVE, all lens-treated eyes developed significantly high levels of myopia in terms of AL (7.61 ± 0.06 mm), VCD (3.00 ± 0.11 mm), and refraction (spherical equivalent (SE) of −9.8 ± 1.5 D), compared to normal (AL: 7.47 ± 0.11 mm; VCD: 2.78 ± 0.04 mm; and SE: 0.69 ± 0.45 D; P<0.002) and control (AL: 7.42 ± 0.15 mm; VCD: 2.82 ± 0.09 mm; and SE: 0.76 ± 1.44 D; P<0.001) eyes. No significant differences were detected between normal and control eyes for AL, VCD, and refraction (Figure 2). Compared to baseline, lens-treated eyes exhibited an average increase in AL of 0.53 ± 0.05 mm (7.4 ± 0.72%), in contrast to 0.28 ± 0.06 mm (3.9 ± 0.87%) and 0.26 ± 0.06 mm (3.7 ± 0.85%) in normal and control eyes, respectively. The relative increase in AL in lens-treated eyes was 250 μm (3.5%) and 266 μm (3.8%) compared to normal and control eyes, respectively. These relative changes in AL and refraction align well with previously published measurements in tree shrews (Siegwart and Norton, 2010).

Figure 2:

Development of (A) axial length (AL), vitreous chamber depth (VCD) and (B) spherical equivalent (SE) of normal, control and experimental high myopic eyes from 24 to 59 days of visual experience. Error bars represent standard error of mean.

3.2. Global Thickness Profiles

A significant group interaction with time was observed for the thickness changes of all peripapillary tissues (P<0.002) with significant group (P<0.001) and time effects (P<0.001). Experimental high myopic eyes exhibited significantly thinner Scl, Ch-RPE, and RRL compared to both normal and control eyes at all follow-up time points (P <0.002). In contrast, significant RNFL thinning has been observed in myopic eyes from 38 to 59 DVE when compared to control eyes and only at 52 DVE when compared to normal eyes (P <0.002). No significant differences were observed between normal and control eyes except for a significantly thicker Scl in the control group at 38 DVE (P <0.002). Table 1 summarizes the absolute thickness values for all peripapillary tissues at baseline (24 DVE) and the end of the study (59 DVE). Plots in Figure 3 show the relative thickness changes from baseline for all tissues.

Table 1:

Absolute thickness values for all peripapillary tissues at baseline (24 DVE) and the experimental endpoint (59 DVE). Abbreviations: RNFL, retinal nerve fiber layer; RRL, remaining retinal layers; Ch-RPE, choroid-retinal pigment epithelium complex; Scl, sclera.

Tissue	Normal		Control		Experimental High Myopia
	Baseline	Endpoint	Baseline	Endpoint	Baseline	Endpoint
RNFL (μm)	100.12 ± 7.06	96.17 ± 7.40	104.98 ± 7.09	102.84 ± 8.76*	104.93 ± 8.51	98.40 ± 9.07*
RRL (μm)	158.78 ± 5.30	150.15 ± 3.91*	160.44 ± 5.07	153.50 ± 4.55*	158.45 ± 5.14	146.90 ± 6.10*
Ch-RPE (μm)	49.22 ± 4.43	47.59 ± 4.05	50.88 ± 2.58	48.32 ± 3.46	50.11 ± 5.09	41.88 ± 5.10*
Sclera (μm)	159.78 ± 14.87	161.19 ± 17.82	158.93 ± 14.43	164.27 ± 20.83	159.26 ± 9.82	141.58 ± 7.82*

^*indicates a significant thickness change from 24 to 59 DVE (P <0.001).

Figure 3:

Global thickness changes of the peripapillary retinal nerve fiber layer (A), remaining retinal layers (B), choroid-retinal pigment epithelium complex (C) and sclera (D). Thickness values are plotted as percentage change from baseline at each measurement point. Error bars represent standard error of mean.

To explore if longitudinal thickness changes occurred in concert with axial elongation, we compared weekly AL and thickness changes from one time point to the next one. Furthermore, thickness profiles were compared against each other to explore the uniformity of thickness changes across various tissues (Figure 4). In normal (Figure 4A) and control (Figure 4B) eyes, AL increased significantly over time between 24 DVE (baseline) and 52 DVE. However, this significant axial elongation did not induce significant thickness changes over time in these groups except for a significant thickening of the sclera in control eyes during the first week of the experiment. Furthermore, among-tissue comparisons revealed a homogeneous thinning profile in normal and control eyes with no significant differences among RNFL, Chr-RPE, and RRL. Surprisingly, the peripapillary sclera was significantly thicker compared to the other tissues within the normal and control groups (Figure 4A, B). In myopic eyes, all tissues exhibited significant thinning only during the first week of the experiment, while AL increased significantly throughout the entire experiment (Figure 4C). Furthermore, the observed thinning was heterogeneous among tissues, where significantly greater thinning (P<0.003) was observed in the Scl (−8.95 ± 3.1%) and ChRPE (−16.8 ± 5.8%) compared to the RNFL (−5.5 ± 1.6%) and RRL (−6.7 ± 1.8%). Additionally, the thinning of the Ch-RPE was significantly greater compared to the Scl (P<0.002), while no significant differences were observed between RNFL and RRL (Figure 4C).

Figure 4:

Relative percentage changes from baseline at each measurement point for AL and peripapillary tissue thicknesses in normal (A), control (B) and experimental high myopic eyes (C). Abbreviations: AL, axial length; RNFL, retinal nerve fiber layer; RRL, remaining retinal layers; Ch-RPE, choroid-retinal pigment epithelium complex; Scl, sclera; DVE, days of visual experience. Solid and broken lines represent significant and non-significant changes from one follow-up point to the next one, respectively.

3.3. Sectorial Thickness Profiles

For accurate interpretation of the sectorial findings, it is important to note that the posterior pole is located nasal and slightly inferior to the optic nerve head in tree shrews (Abbott et al., 2009; Sajdak et al., 2019), which is opposite to the temporal location of human fovea. Figure 5 illustrates the sectorial thickness changes from baseline for all tissues at all follow-up points. In myopic eyes, the Scl and Ch-RPE thinned significantly across all sectors and time points compared to normal and control eyes (P<0.002). The RRL of the myopic eyes thinned significantly (P<0.002) across most DVEs and sectors except the temporal sector when compared to control eyes. Significant RRL changes were also seen between myopic and normal eyes but in fewer sectors. Lastly, the RNFL of myopic eyes exhibited a variable sectorial thinning profile that changed over time in myopic eyes. At 59 DVE, significant RNFL thinning (P<0.002) was identified in myopic eyes in three sectors (ST, S, ad N) when compared to both, normal and control eyes, and in two more sectors (T and SN) when compared to control eyes only. Normal and control eyes did not show any significant differences except for a thicker sclera in the nasal and temporal sectors in control eyes at 38 DVE. Overall, no clear or consistent asymmetric thinning pattern was observed during myopia development and progression.

Figure 5:

Sectorial thickness changes from baseline of peripapillary Retinal Nerve Fiber Layer (RNFL), Remaining Retinal Layers (RRL), Choroid-Retinal Pigment Epithelium complex (Ch-RPE) and Sclera. Solid lines and shaded areas represent mean changes from baseline and standard error of mean, respectively. The compass at the bottom right indicates the different sectors and the direction towards the posterior pole (inferonasal in tree shrews). Abbreviations: T, temporal; ST, superotemporal; S, superior; SN, superonasal; N, nasal; IN, inferonasal; I, inferior; IT, inferotemporal; PP, posterior pole.

4. Discussion

The present study investigated the longitudinal thickness changes of peripapillary tissues during experimental juvenile high myopia development and progression in tree shrews. Our results show that all peripapillary tissues underwent significant thinning that occurred predominantly during the first week of myopia development leading to a refractive error of −2.5±1.9 D at 31 DVE. Despite the continued progression of axial elongation and myopia development from 31 to 59 DVE reaching −9.8 ± 1.5 D at the end of the experiment, no significant thickness changes occurred over time during this period. Additionally, the observed thinning was heterogeneous among tissues, with the Scl and Ch-RPE exhibiting more profound thinning compared to the RNFL and RRL. The observed heterogeneous thickness changes may predispose myopic eyes to pathological ONH remodeling and glaucoma later in life.

Previous studies have widely reported scleral and chorioretinal thinning in myopic eyes. Scleral thinning is a hallmark in myopic eyes with a myriad of clinical (Akagi et al., 2013; Cheng et al., 1992; Jonas et al., 2014a; Jonas and Xu, 2014; Ohno-Matsui et al., 2012; Park et al., 2014) and experimental (McBrien et al., 2001; Zi et al., 2020) evidence for a negative correlation between AL and scleral thickness. Consistent with these findings, our study showed significant scleral thinning in all sectors in the peripapillary region. There is ongoing debate regarding the correlation between AL and chorioretinal thickness changes in myopia. Choroidal thickness has been often reported to negatively correlate with AL (Gupta et al., 2015; Wei et al., 2013; Yang et al., 2019). It has been estimated that 1 mm increase in AL correlates with a 32 μm reduction in subfoveal choroidal thickness in human adult eyes (Wei et al., 2013). Similarly, several studies (Leung et al., 2006; Lim and Chun, 2013; Tai et al., 2018) have reported a negative correlation between AL and the thicknesses of RNFL, estimating a 2.2 μm decrease in mean peripapillary RNFL thickness for every 1 mm increase in AL (Budenz et al., 2007). Our results show a similar trend, where the choroid thinned significantly more than the RNFL or RRL during experimental high myopia. However, other studies have found no or even positive correlations between AL and choroidal/RNFL thickness (Deng et al., 2019; Hoh et al., 2006; Kang et al., 2010; Kim et al., 2022; Saito et al., 2023). One potential explanation for these variable results might reside in the different age range of participants across those studies. Since the choroid (Chirco et al., 2017; Wakatsuki et al., 2015) and RNFL (Celebi and Mirza, 2013; Feuer et al., 2011; Lee et al., 2019; Repka and Quigley, 1989) thin with aging, myopic eyes may exhibit amplified age-related chorioretinal thinning. A similar difference may not be seen at adolescent or childhood age. Speaking to this, greater thinning of the RNFL (Lee et al., 2019) and ganglion cell-inner plexiform layer (GCIPL) (Lee et al., 2020) has been reported in myopic eyes of elderly people (50–59 years old) compared to younger adults (20–29 and 30–39 years old). Similarly, significant thinning of the RNFL was observed during five years of myopia progression in a cohort of Chinese adolescents (mean ± SD baseline age and SE were 10.97 years and −1.89 ± 0.92 D, respectively at baseline) (Xiao et al., 2022). In contrast, Kim et al. (Kim et al., 2022) reported no significant thickness changes of RNFL and choroid over four years of myopia progression in Korean children (mean ± SD baseline age and SE were 9.6 years and −4.26 ± 2.34 D, respectively).

In addition to age, another potential reason for the contradictory findings might be the refractive status at the time of enrollment. To date, all clinical studies have included children who were already myopic at the time of enrollment. Our results suggest that peripapillary tissue thinning occurs only during the very early stage of myopia development until a refraction of −2.5±1.9 D is reached at 31 DVE. After this early phase, the thickness of all peripapillary tissues remains stable despite continued myopia progression in tree shrews. The children of the Boramae Myopia Cohort Study (Kim et al., 2022, 2018) had moderate myopia (−4.26±2.34 D) at baseline and may have already reached this stable phase at the onset of the study. Consequently, the thickness changes might have happened during the early phase and remained undetected (Kim et al., 2022, 2018).

Lastly, the AL-dependent magnification effect in OCT imaging can bias the location of the obtained thickness measurements, which may contribute to the observed controversy (Öner et al., 2013; Xiao et al., 2022). The negative correlation between AL and RNFL thickness can be reduced after magnification correction (Öner et al., 2013). Since most of the studies did not report whether it has been accounted for the magnification, it is unclear if the reported thicknesses were obtained from the same location within each study and across the studies. Note that the OCT images used here were corrected for nonlinear distortions including AL-dependent magnification effects (Grytz et al., 2022).

Pertinent to the global thinning of all peripapillary tissues seen here, two observations need careful interpretation. The first is the mismatch between the time-dependent thinning and axial elongation process. The sclera is thought to remodel and stretch during axial elongation and myopia development. Soft tissues such as the sclera and the other peripapillary tissues are nearly incompressible and, therefore, thin to preserve the tissue volume when they are stretched or remodeled in-plane (tangential to the scleral surface). Consequently, we would have expected that time-dependent tissue thinning would mirror the axial elongation process, which was not the case as shown in Figure 4. Accelerated tissue growth has been proposed as a compensatory mechanism to offset the thinning induced by accelerated axial elongation in myopic eyes. However, scleral and choroidal volumes were reported to not change after two years of age (Shen et al., 2016) or in myopia (Jonas et al., 2014a). In addition, the areal density of different retinal cells has been shown to decrease in form deprivation-induced myopia in tree shrews (−15.9 ± 2.3D) while the volumetric cellular density was preserved, indicating a potential early growth cessation for the retina too (Abbott et al., 2011). The existing evidence for cessation of ocular tissue growth early in life (Abbott et al., 2011; Jonas et al., 2014a; Shen et al., 2016) and the simultaneous cessation of thinning of all tissues make a compensatory growth response an unlikely explanation. Another potential explanation is that the cessation of thinning occurs only at the peripapillary region while thinning continues at more peripheral regions. In a prior report, we demonstrated a similar trend of early thinning cessation near the posterior pole in tree shrews (El Hamdaoui et al., 2020), but other peripheral regions have not been investigated so far. Wide-angle OCT imaging may be employed to gain insights into the thickness changes in the periphery during myopia development and progression.

The second observation is the heterogeneous nature of the thinning among the peripapillary tissues as shown in Figure 4. We hypothesize that scleral remodeling underlies axial elongation and drives ONH change during myopia development and progression (Grytz et al., 2020). Consequently, we expected that the other peripapillary tissues would follow the “lead” of the sclera and thin in a similar manner as the sclera remodels. However, the relative thinning from baseline was significantly lower in retinal layers (RNFL and RRL) compared to the sclera and choroid. One potential explanation of this observation is that an intermediate structure segregates the ocular tissues into two compartments: the sclera and choroid on one side and the retina on the other side. The Bruch’s membrane is a thin tri-lamellar structure of collagen and elastin located between the choroid and RPE (Curcio and Johnson, 2013). Given its collagenous structure and intermediate location, the Bruch’s membrane might indeed shield the retina from the remodeling deformations that occur at the sclera, while the choroid may act as a shear-band between the sclera and the Bruch’s membrane. In addition, translational (Dong et al., 2019) and clinical (Bai et al., 2017; Jonas et al., 2014b) studies reported that the thickness of Bruch’s membrane remained unchanged despite axial elongation, pointing to its potential ability to grow during myopia development. Lee et al. proposed preferential retinal growth at the equator as a potential compensatory mechanism to preserve the retinal structure near the posterior pole during myopia development (Lee et al., 2021). While it remains unclear if the retina can still grow at ages at which myopia develops, the idea that the retina changes at the equator (it may be growth, remodeling or just stretching) to protect its integrity and prevent retinal thinning at the posterior pole is consistent with the heterogenous tissue thinning observed in this study. Further studies are needed to unravel retinal changes at the equator and the potentially important role of the Bruch’s membrane during myopia development. Furthermore, the functional consequences of the here observed heterogeneous thinning remains unknown. While it might be argued that the retina is protected from excessive thinning during myopia development, abnormal electroretinogram (ERG) responses have been reported in myopic eyes without coexisting ocular pathologies in children (Ho et al., 2012, 2011; Li et al., 2017; Wang et al., 2013) and adults (Gupta et al., 2022; Sachidanandam et al., 2017). Incorporating functional testing (ERG) with OCT imaging in longitudinal translational myopia studies may add valuable functional insights to the here observed structural changes.

One key aspect yet to be considered is the potential influence of aging on the observed peripapillary thickness changes in myopic eyes. Lee et al. investigated the rates of RNFL (Lee et al., 2019) and GCIPL (Lee et al., 2020) thinning in myopic and emmetropic eyes of different age groups. In people aged 20–29 and 30–39 years old, the RNFL (Lee et al., 2019) and GCIPL (Lee et al., 2020) thickness remained stable. In contrast, people aged 40–49 years old exhibited significant RNFL thinning (Lee et al., 2019) and people aged 50–59 years old showed significant thinning of both, RNFL (Lee et al., 2019) and GCIPL (Lee et al., 2020). Furthermore, the rates of RNFL (Lee et al., 2019) and GCIPL (Lee et al., 2020) thinning were more pronounced in the high myopic group compared to the control group. Combining juvenile findings from our tree shrews with longitudinal clinical data in children (Kim et al., 2022) and adults (Lee et al., 2020, 2019) might allow the construction of a hypothetical model for peripapillary thickness changes. We hypothesize that peripapillary tissue remodeling incorporates three phases of tissue thinning in myopic eyes: (i) an early phase of thinning during the early development of myopia, (ii) a relatively stable phase without thinning during myopia progression and adulthood, and (iii) a late thinning phase, where myopia interacts with aging leading to accelerated age-dependent retinal thinning.

In addition to the global thickness changes, we have quantified the sectorial thinning pattern. In contrast to our previous work, where we found very clear asymmetric relative deformations between the anterior scleral canal opening and Bruch’s membrane opening positions during myopia development (Khalafallah et al., 2023), no clear asymmetric thinning patterns were observed in the present study. The peripapillary Scl, Ch-RPE, and RRL thinned across all sectors, while the asymmetric thinning pattern of the RNFL was inconsistent across time points. Abott et al. (Abbott et al., 2011) investigated retinal thickness changes in form-deprived tree shrews in four sectors (IN, SN, IT, ST). When compared to the control eyes, the retina of the form-deprived eye was found to be significantly thinner in the nasal but not the temporal sectors. The sectorial thinning patterns of the RRL and RNFL observed here do not match the results by Abott et al. (Abbott et al., 2011), which may suggest that form-deprivation myopia and negative lens myopia lead to different thickness changes at the ONH. Prior clinical studies have investigated the sectorial thickness changes in myopic eyes (Gupta et al., 2015; Kim et al., 2010; Song et al., 2020). Kim et al. (Kim et al., 2010) reported a significantly thinner peripapillary RNFL in the non-temporal sectors of highly myopic eyes compared to low myopic eyes. Gupta et al. (Gupta et al., 2015) and Song et al. (Song et al., 2020) found the peripapillary choroid to be thinnest inferiorly. Other studies reported relatively stable peripapillary chorioretinal thickness (Kim et al., 2022) and ONH vasculature (Lee et al., 2018) during juvenile myopia progression. The aforementioned findings highlight the need for additional research to better understand the determinants for such sectorial differences.

There is a growing interest in deciphering the mechanism by which myopia increases the risk for glaucoma. The here observed thinning of individual ocular tissues may lead to long-term functional consequences and contribute to the future risk for glaucoma. Scleral thinning is a shared structural feature in both, myopia (Akagi et al., 2013; Cheng et al., 1992; Jeon et al., 2020; Jonas et al., 2014a; Jonas and Xu, 2014; Ohno-Matsui et al., 2012; Saito et al., 2023) and glaucoma (Jeon et al., 2020; Kim et al., 2019; Park et al., 2014; Ren et al., 2009; Wang et al., 2022). Acting as a mechanical anchor for the LC, scleral thinning might increase the vulnerability of the LC to IOP-induced deformations. Furthermore, scleral thinning and altered curvature have been associated with choroidal microvascular dropouts and RNFL defects (Ohno-Matsui et al., 2012; Shin et al., 2019; Shinohara et al., 2016). Park et al. (Park et al., 2014) reported a significantly thinner sclera in myopic eyes with NTG compared to myopic eyes with or without primary open angle glaucoma. Similarly, choroidal thinning was significantly associated with the development of NTG in myopic eyes, which may have involved regional ONH hypoxia (Akagi et al., 2013; Asao et al., 2014; Hirooka et al., 2012; Na et al., 2020). Furthermore, the presence of axial myopia in glaucomatous eyes was associated with a significant increase in both, the prevalence and area of microvascular dropouts, and correlated with a higher degree of glaucoma severity (Micheletti et al., 2023). Since thinning of RNFL and GCIPL is a hallmark of structural abnormality in glaucoma (Chang and Singh, 2013; Nicholas Y Q Tan et al., 2019), retinal thinning during myopia development might predispose the retinal ganglion cells to increased risk for mechanical insult even at normal IOP levels. The here observed thinning of peripapillary tissues along with other ONH remodeling changes during juvenile myopia development may also predispose myopic eyes to accelerated age-related remodeling changes. Normal aging can lead to scleral stiffening, compromised choroidal circulation and retinal thinning (Grytz et al., 2020; Guedes et al., 2011; Quigley, 2011). These age-related remodeling changes may become more profound and occur earlier in myopic eyes, where ONH remodeling including the tissue thinning established early in life may increase the risk for pathologic ONH remodeling and glaucoma at much younger ages compared to emmetropic eyes (Grytz et al., 2020; Guedes et al., 2011; Quigley, 2011; Shim et al., 2017).

Potential limitations of our study include the inability to determine the nature of the observed thinning whether it is attributed to stretching or degeneration or both. Added to that, the observed scleral thickening in the control eyes is surprising. The underlying mechanism remains unclear and warrants additional investigation. However, this observation stresses the importance of including a normal group as the response of the control eye may differ from normal eyes. High-resolution OCT holds promise for investigating micro-scale (cell-level) growth and remodeling changes (Reche et al., 2023). Pertinent to the imaging methods, our OCT scans in this study were limited to a 30° view centered on the ONH, and it might be argued that the changes in such a relatively small zone cannot be generalized to other regions of the eye including the posterior pole and the periphery. Wide-angle OCT may provide additional information at peripheral regions. While the tree shew is a reliable model of myopia, differences from human myopia should be acknowledged, necessitating caution when interpreting our findings in relation to the human condition. The first difference is the relatively smaller magnitude of axial elongation when compared to humans for the same level of myopia. To reach −10D myopia, a human eye needs to elongate by approximately 20% (Jonas et al., 2023; Xu et al., 2010) compared to 3.5–3.8% observed here in tree shrews. It is unclear if the relatively smaller change in AL in tree shrews has any implication on the observed heterogeneous thickness changes. Additionally, the rapid rate of axial elongation and myopia development in our model deviates from the slower progression in human myopia. Recently, Khanal et al. (Khanal et al., 2023) proposed a novel slowly progressing myopia model in tree shrews, which may better mimic the slowly progressing changes in humans. One more limitation lies in the definition of sectors. The horizontal axis of the enface confocal scanning laser ophthalmoscopy image (Figure 1A) has been used to define the nasal/temporal axis. Consequently, the sectorial orientations may vary across animals based on variations in the head position during the OCT scan, which may have masked a potential sectorial pattern of thickness changes. To overcome this potential bias, the fovea-BMO axis has been proposed as an anatomical landmark to define sectorial orientations (He et al., 2014). This approach could not be used here as tree shrews lack a fovea. Lastly, our study included only juvenile animals, and a longer follow-up period into adulthood and older age is essential to clarify the interactions of early myopic changes with aging and the increased risk of glaucoma.

In summary, experimental high myopia induces significant thinning of all peripapillary tissues. The observed thinning occurs early during myopia development despite continued axial elongation. Furthermore, peripapillary tissues exhibited a heterogenous thinning pattern, where retinal tissues were protected from profound thickness changes. Longer follow-up studies into adulthood may reveal the potential impact of aging on the observed thinning pattern and its potential impact on ONH remodeling and glaucoma risk later in life.

• Peripapillary tissues undergo significant thinning during high myopia development.

• In contrast to the sclera and choroid, retina seems to be protected from profound thinning.

• Thinning occurs early during high myopia development until −2.5±1.9 D is reached.

• After that, thickness changes cease despite continued myopia progression.