Atropine Ophthalmic Solution to Reduce Myopia Progression in Pediatric Subjects: The Randomized, Double-Blind Multicenter Phase II APPLE Study

Abstract

Purpose: 

The purpose of this study was to assess the dose-response effects of low-dose atropine on myopia progression and safety in pediatric subjects with mild-to-moderate myopia.

Methods: 

This phase II, randomized, double-masked, placebo-controlled study compared the efficacy and safety of atropine 0.0025%, 0.005%, and 0.01% with placebo in 99 children, aged 6–11 years, with mild-to-moderate myopia. Subjects received 1 drop in each eye at bedtime. The primary efficacy endpoint was change in spherical equivalent (SE), while secondary endpoints included changes in axial length (AL) and near logMAR (logarithm of the minimum angle of resolution) visual acuity and adverse effects.

Results: 

The mean±SD changes in SE from baseline to 12 months in the placebo and atropine 0.0025%, 0.005%, and 0.01% groups were −0.55±0.471, −0.55±0.337, −0.33±0.473, and −0.39±0.519 D, respectively. The least squares mean differences (atropine−placebo) in the atropine 0.0025%, 0.005%, and 0.01% groups were 0.11 D (P=0.246), 0.23 D (P=0.009), and 0.25 D (P=0.006), respectively. Compared with placebo, the mean change in AL was significantly greater for atropine 0.005% (−0.09 mm, P=0.012) and 0.01% (−0.10 mm, P=0.003). There were no significant changes in near visual acuity in any of the treatment groups. The most common ocular adverse events were pruritus and blurred vision, each occurring in 4 (5.5%) atropine-treated children. Changes in mean pupil size and amplitude of accommodation were minimal.

Conclusions: 

Atropine doses of 0.005% and 0.01% effectively reduced myopia progression in children but no effect was noted with 0.0025%. All doses of atropine were safe and well tolerated.

INTRODUCTION

Myopia is a refractive condition caused by the excessive elongation of the eye so that distant objects are focused in front of the retina.¹ This condition is common, affecting ∼1.4 billion (22.9%) people worldwide.² By 2050, myopia is expected to affect ∼4.8 billion people or 49.8% of the global population.²

Myopia usually develops during childhood.¹ The risk appears to be associated with a combination of factors such as too much near-work, too little outdoor time, higher educational level, and parental history.^3,4 The increase in the prevalence of myopia in many East Asian populations over the last 3 decades is thought to be driven by environmental and lifestyle changes.^1,5 More recent changes in lifestyle during the coronavirus disease 2019 (COVID-19) epidemic have correlated with increased prevalence and accelerated progression of myopia across the world.⁶

Myopia is progressive and irreversible. Advanced myopia increases the risk for severe visual impairment, including retinal detachment, myopic retinopathy, and glaucoma.^7–12 The World Health Organization’s (WHO) Prevention of Blindness and Visual Impairment international initiative program has identified uncorrected myopia as one of the main global causes of visual impairment.¹³ The challenge now is to find safe, effective, and cost-effective treatments targeting myopic progression in children.

Atropine, a nonselective muscarinic receptor antagonist, is the most common therapeutic agent used to slow the progression of myopia.^14,15 Atropine eye drops have been shown to effectively slow myopia progression in a dose-dependent manner.^16–20 Current clinical research of atropine has focused on determining an effective dose with minimal adverse events due to safety concerns, as higher dosages have been associated with adverse effects (AEs), such as photophobia, near-blurred vision, loss of accommodation, and excessive pupil dilation.²¹ Although atropine doses as low as 0.01% have been shown effective in reducing myopia progression,²² the minimum effective dose of atropine for slowing disease progression in pediatric subjects remains undetermined.

The aim of this randomized, double-blinded, placebo-controlled, multicenter phase II trial was to evaluate the safety and efficacy of 3 concentrations of atropine, 0.0025%, 0.005%, and 0.01%, compared with placebo in pediatric subjects diagnosed with mild-to-moderate myopia.

METHODS

Study Subjects

The study cohort consisted of children aged 6–11 years with a spherical equivalent (SE) between −1.0 and −6.0 D in both eyes; myopia progression of at least −0.50 D in both eyes over the last 12 months; astigmatism of ≤ −1.50 D in both eyes; distance correction to logMAR (logarithm of the minimum angle of resolution) 0.2 or better in both eyes; and normal ocular health other than myopia. Subjects were excluded if they had past amblyopia or strabismus; ocular disorders (eg, glaucoma or lens opacity); are previous or current users of contact lenses or pharmacological treatments for myopia; past ocular surgery or ocular laser treatment; or any medical conditions (eg, cardiac or respiratory conditions) that would affect study participation or interpretation of data. The study was conducted in compliance with the Institutional Review Board and regulatory requirements, Singapore—Good Clinical Practice guidelines, International Conference on Harmonization guidelines, and the Declaration of Helsinki. The trial was registered with ClinicalTrials.gov (registration no: NCT03329638).

Participants who met all eligibility criteria were stratified by age group (ages 6–7, 8–9, and 10–11 y), and randomized in a 1:1:1:1 ratio to 0.0025%, 0.005%, or 0.01% of atropine dose or placebo. The trial medication was a sterile, aqueous, preservative-free ophthalmic solution containing atropine as the active ingredient. The study team, study participants, and their parents/legal guardians were blinded to the dose of treatment drops.

The primary endpoint outcome was the change in SE from baseline over 12 months, as determined by cycloplegic autorefraction in the study eye. The eye included in the study was the more myopic eye, as determined by baseline SE, or the right eye if both SE values were the same. Cycloplegic autorefraction was performed 30 minutes after the administration of the last 3 drops of cyclopentolate 1%, spaced 5 minutes apart. Autorefraction were tested with a table-mounted autorefractor (site-specific Canon RK, Topcon TPK-1P, or Nidek Tonoref II), which were calibrated onsite and validated. Five refractive error measurements were taken for each eye. The mean SE was calculated and used as a measure of progression in each eye.

Secondary endpoints included changes in axial length (AL) and best-corrected near logMAR visual acuity from baseline at all study visits. AL was measured by noncontact partial coherence interferometry (the Zeiss IOL master), and best-corrected near visual acuity (BCVA) was measured using near logMAR visual acuity charts.

Safety assessments included recording of adverse events. The amplitude of accommodation was measured using a Royal Air Force (RAF) ruler; photopic pupil size was measured using a table-mounted autorefractor under photopic conditions at 800±50 lux; distance vision was measured using distance logMAR visual acuity charts; and intraocular pressure was measured by noncontact tonometry. Eyes were examined by slit-lamp biomicroscopy, fundus photography, and indirect ophthalmoscopy.

Statistical Analysis

The sample size was calculated based on the assumption that the mean±SD changes in SE would be −0.40±0.5 D in the atropine groups and −0.80±0.5 D in the placebo group; based on outcomes of previous studies and clinical audit data.²² A planned sample of 96 total subjects was calculated to have 81% power to detect differences among the means of 4 groups at the 0.05 significance level.

The primary endpoint was determined using a prespecified mixed-effects model for repeated-measures (MMRM) analysis of covariance (ANCOVA) with baseline SE as the sole covariate, a nominal 2-sided significance level of 0.05 and a 2-sided confidence level of 95%. The advantages of this method include the inclusion of data from all visits, higher statistical power, and unbiased estimation.^23,24 Least squares (LS) means were determined by calculating estimated parameters to minimize the sum of squares of residuals in a linear model, a method used in both analysis of variance and ANCOVA analyses. The LS means, SDs, and differences in LS means between each atropine group and the placebo group at 12 months were calculated. SEMs, P values, and 95% CIs for comparisons of each atropine dose with placebo at 12 months were determined.

MMRM ANCOVA was also used for statistical analysis of secondary endpoints, including changes in SE and AL from baseline at all time points. Baseline AL was the sole covariate of change used to analyze AL. Means, SD, medians, and maximum and minimum values were determined for demographic data, changes in BCVA, and subgroup analyses. Changes in SE and AL in the study eye from baseline to 12 months in response to atropine doses were determined using regression analyses.

Subgroup analyses were performed to compare the effects of treatment in children grouped by age (6–7, 8–9, and 10–11 y), sex, baseline SE (<−4, ≥−4 to <−2, and ≥−2 D), and baseline AL (<24, ≥24 to <25, and ≥25 mm).

RESULTS

Of the 119 children who were screened, 100 met the inclusion/exclusion criteria and were randomized (Fig. 1). One child was excluded from the study because of issues with informed consent. Of the remaining 99 children, 87 (87.9%) were ethnic Chinese with a mean±SD baseline SE of −3.50±1.2 D and AL of 24.64±0.79 mm (Table 1). There was no significant difference in age among the 4 groups, but some differences in sex were observed.

FIGURE 1: Study design. AE indicates adverse effect; logMAR, logarithm of the minimum angle of resolution.

TABLE 1 - Demographic and Ocular Characteristics of Study Subjects at Baseline

Characteristics	Placebo (N=26)	Atropine 0.0025% (N=24)	Atropine 0.005% (N=24)	Atropine 0.01% (N=25)	P
Age (y)
Mean (SD)	8.9 (1.7)	8.9 (1.3)	9.0 (1.4)	8.8 (1.3)	0.969
Median (minimum, maximum)	9.0 (6, 11)	9.0 (7, 11)	9.0 (6, 11)	9.0 (6, 11)
Sex [n (%)]
Male	8 (30.8)	11 (45.8)	10 (41.7)	18 (72.0)	0.027
Female	18 (69.2)	13 (54.2)	14 (58.3)	7 (28.0)
Race [n (%)]
Chinese	22 (84.6)	19 (79.2)	22 (91.7)	24 (96.0)	0.239
Indian	3 (11.5)	5 (20.8)	1 (4.2)	1 (4.0)
Malay	0 (0.0)	0 (0.0)	1 (4.2)	0 (0.0)
Other	1 (3.8)	0 (0.0)	0 (0.0)	0 (0.0)
SE (D)
Mean (SD)	−3.93 (1.31)	−3.00 (1.1)	−3.8 (1.4)	−3.25 (1.1)	0.020
Median (minimum, maximum)	−4.05 (−5.9, −1.5)	−2.74 (−4.6, −1.3)	−3.96 (−5.8, −1.4)	−3.23 (−5.9, −1.3)
AL (mm)
Mean (SD)	24.79 (0.8)	24.35 (0.8)	24.64 (0.8)	24.77 (0.7)	0.174
Median (minimum, maximum)	24.69 (23.0, 26.5)	24.58 (22.4, 25.9)	24.71 (23.0, 26.5)	24.66 (23.2, 26.5)

AL indicates axial length; ANOVA, analysis of variance; D, diopter; SD, standard deviation; SE, spherical equivalent.

Fisher exact test was used to analyze sex and race; ANOVA was utilized for age and baseline SE and AL.

Efficacy

Mean±SD changes in SE from baseline to 12 months in the placebo and atropine 0.0025%, 0.005%, and 0.01% groups were −0.55±0.471, −0.55±0.337, −0.33±0.473, and −0.39±0.519 D, respectively (Fig. 2). After adjustments for baseline SE, the LS mean differences between the atropine 0.0025%, 0.005%, and 0.01% groups and the placebo group were 0.11 D (P=0.246; 95% CI: −0.08 D, 0.29 D), 0.23 D (P=0.009; 95% CI: 0.06 D, 0.41 D), and 0.25 D (P=0.006; 95% CI: 0.07 D, 0.43 D), respectively (Table 2). Statistical analysis with adjustment for baseline SE demonstrated a statistically significant linear dose-response between the percentage of atropine administered and the difference from placebo in SE at 12 months, with an estimate±SEM of 25.04±11.777 D/% (95% CI: 1.66 D/%, 48.42 D/%; P=0.036).

FIGURE 2: Change in spherical equivalent from baseline to month 12.

TABLE 2 - Analysis of SE and AL at 12 Months Via MMRM ANCOVA With Baseline Values as Covariates

Outcomes	Placebo (N=26)	Atropine 0.0025% (N=24)	Atropine 0.005% (N=24)	Atropine 0.01% (N=25)
SE (D)
LS mean	−0.6	−0.49	−0.37	−0.35
Difference from placebo (95% CI)		0.11 (−0.08, 0.29)	0.23 (0.06, 0.41)	0.25 (0.07, 0.43)
P		0.2463	0.0090**	0.0056**
Axial length (mm)
LS mean	0.36	0.3	0.27	0.25
Difference from placebo (95% CI)		−0.06 (−0.13, 0.01)	−0.09 (−0.16, −0.02)	−0.1 (−0.17, −0.04)
P		0.0839	0.0123*	0.0026**

AL indicates axial length; ANCOVA, analysis of covariance; CI, confidence interval; D, diopter; LS mean, least square mean; MMRM, mixed-effects model for repeated measures; SE, spherical equivalent.

The ANCOVA of SE adjusted for the covariate of baseline SE. For the analysis of AL, the covariate was baseline AL.

^*P<0.05.

^**P<0.01.

Mean changes in AL from baseline to 12 months are shown in Figure 3. The LS mean difference estimates (atropine−placebo) for changes in AL from baseline to month 12 in the atropine 0.0025%, 0.005%, and 0.01% groups were –0.06 mm (P=0.084; 95% CI: −0.13 mm, 0.01 mm), −0.09 mm (P=0.012; 95% CI: −0.16 mm, −0.02 mm), and −0.10 mm (P=0.003; 95% CI: −0.17 mm, −0.04 mm), respectively (Table 2).

FIGURE 3: Change in axial length from baseline to month 12.

Subgroup analysis showed that younger children had greater myopic progression than older children. There was very little difference in myopia progression between children in the placebo and 0.0025% atropine groups. However, there was a trend toward lower myopia progression in the 0.005% and 0.01% atropine groups, in the oldest age group (10–11 y), and for groups with higher baseline myopia (SE <−4 D, and AL ≥25 mm) (Table 3).

TABLE 3 - Subgroup Analysis of Changes in SE From Baseline to Month 12

	Mean (SD)
Change from baseline	Placebo (N=26)	Atropine 0.0025% (N=24)	Atropine 0.005% (N=24)	Atropine 0.01% (N=25)	P*
Full analysis set (n=99)	−0.55 (0.471)	−0.55 (0.337)	−0.33 (0.473)	−0.39 (0.519)	0.146 (0.408, 0.060, 0.046)
Sex
Male (n=47)	−0.59 (0.580)	−0.42 (0.261)	−0.16 (0.516)	−0.44 (0.509)	0.222 (0.105, 0.054, 0.099)
Female (n=52)	−0.53 (0.431)	−0.66 (0.363)	−0.46 (0.414)	−0.25 (0.562)	0.267 (0.636, 0.524, 0.115)
Age group (y)
6–7 (n=16)	−0.89 (0.546)	−0.96 (0.456)	−0.23 (0.604)	−0.73 (0.728)	0.355 (0.790, 0.101, 0.405)
8–9 (n=46)	−0.60 (0.533)	−0.56 (0.341)	−0.50 (0.569)	−0.45 (0.467)	0.675 (0.537, 0.436, 0.228)
10–11 (n=37)	−0.30 (0.372)	−0.40 (0.135)	−0.18 (0.186)	−0.21 (0.507)	0.752 (0.973, 0.526, 0.372)
Baseline SE
≥−2 D (n=15)	−0.71 (0.088)	−0.49 (0.296)	−0.81 (0.530)	−0.79 (0.531)	0.262 (0.991, 0.467, 0.176)
≥−4 to <−2 D (n=48)	−0.60 (0.403)	−0.58 (0.423)	−0.37 (0.351)	−0.36 (0.501)	0.329 (0.749, 0.178, 0.147)
<−4 D (n=36)	−0.49 (0.551)	−0.55 (0.104)	−0.19 (0.492)	−0.05 (0.32)	0.228 (0.810, 0.163, 0.105)
Baseline axial length
<24 mm (n=19)	−0.58 (0.372)	−0.73 (0.438)	−0.42 (0.549)	−0.86 (0.583)	0.673 (0.714, 0.505, 0.811)
≥24 to <25 mm (n=54)	−0.49 (0.421)	−0.47 (0.288)	−0.30 (0.491)	−0.38 (0.543)	0.531 (0.284, 0.419, 0.153)
≥25 mm (n=26)	−0.62 (0.603)	−0.49 (0.104)	−0.32 (0.455)	−0.27 (0.439)	0.384 (0.816, 0.189, 0.122)

ANCOVA indicates analysis of covariance; D, diopter; SD, standard deviation; SE, spherical equivalent.

^*Calculated by ANCOVA with baseline SE as covariate; numbers represent placebo versus atropine (placebo vs. 0.0025% atropine, placebo vs. 0.005% atropine, placebo vs. 0.01% placebo).

Safety

Near and distance logMAR BCVA showed little change during the course of this study, and there were no apparent trends in mean change in amplitude of accommodation (Table 4). Pupil size did not change in the placebo and 0.0025% atropine groups, but children treated with 0.005% and 0.01% atropine experienced changes in pupil size of 0.26 and 0.31 mm, respectively. The effects on intraocular pressure were also minimal (<1 mm Hg).

TABLE 4 - Changes in Ocular Assessments From Baseline to Month 12 (Study Eye)

	Mean (SD)
Change from baseline to month 12	Placebo (N=26)	Atropine 0.0025% (N=24)	Atropine 0.005% (N=24)	Atropine 0.01% (N=25)	P*
Best-corrected visual acuity near logMAR (D)	−0.022 (0.0696)	−0.015 (0.0478)	−0.007 (0.0637)	−0.002 (0.0351)	0.567
Best-corrected visual acuity distance logMAR (D)	−0.012 (0.0714)	−0.020 (0.0641)	−0.036 (0.0521)	−0.045 (0.0633)	0.244
Photopic pupil size (mm)	0.00 (0.815)	0.00 (0.715)	0.26 (0.609)	0.31 (0.628)	0.245
IOP (mm Hg)	−1.1 (1.61)	−1.3 (1.58)	−0.4 (2.09)	−0.6 (2.67)	0.395
Amplitude of accommodation (D)	−0.05 (1.544)	0.11 (1.735)	−0.38 (1.331)	0.46 (1.999)	0.371

D indicates diopter; IOP, intraocular pressure; logMAR, logarithm of the minimum angle of resolution; SD, standard deviation.

^*Calculated by analysis of variance (placebo vs. atropine).

Treatment with all study doses of atropine was safe and well tolerated. Ocular AEs were reported only by subjects in the atropine groups, with no apparent relationship to dose (Table 5). Eye pruritus and blurry vision were reported by 4 (4.0%) subjects each, with similar numbers of children reporting these AEs across all atropine doses. The glare was reported by 3 (3.0%) subjects and eye pain by 2 (2.0%). All other ocular AEs were reported by 1 subject each. One child in the atropine 0.0025% group experienced eye pain that resulted in discontinuation and withdrawal from the study.

TABLE 5 - Adverse Events

	n (%)
Adverse event	Placebo (N=26)	Atropine 0.0025% (N=24)	Atropine 0.005% (N=24)	Atropine 0.01% (N=25)
Ocular
Any*	0 (0.0)	6 (25.0)	2 (8.3)	6 (24.0)
Eye pruritus	0 (0.0)	2 (8.3)	0 (0.0)	2 (8.0)
Glare	0 (0.0)	1 (4.2)	0 (0.0)	2 (8.0)
Vision blurred	0 (0.0)	2 (8.3)	1 (4.2)	1 (4.0)
Conjunctivitis allergic	0 (0.0)	0 (0.0)	0 (0.0)	1 (4.0)
Ocular hyperemia	0 (0.0)	0 (0.0)	0 (0.0)	1 (4.0)
Eye swelling	0 (0.0)	0 (0.0)	1 (4.2)	0 (0.0)
Eye pain	0 (0.0)	2 (8.3)	0 (0.0)	0 (0.0)
Eye infection	0 (0.0)	1 (4.2)	0 (0.0)	0 (0.0)
Nonocular
Any*	2 (7.7)	5 (20.8)	13 (54.2)	7 (28.0)
Pyrexia	1 (3.8)	2 (8.3)	8 (33.3)	4 (16.0)
Influenza	1 (3.8)	0 (0.0)	4 (16.7)	0 (0.0)
Viral illness (varicella, hand-foot-mouth disease, upper respiratory tract infections)	1 (3.8)	0 (0.0)	2 (8.4)	1 (4.0)
Gastroenteritis	0 (0.0)	1 (4.2)	0 (0.0)	0 (0.0)
Headache	0 (0.0)	0 (0.0)	1 (4.2)	1 (4.0)
Epistaxis	0 (0.0)	0 (0.0)	0 (0.0)	1 (4.0)
Cough	0 (0.0)	1 (4.2)	4 (16.7)	0 (0.0)
Rash	0 (0.0)	1 (4.2)	1 (4.2)	(0.0)

^*Includes subjects with at least 1 adverse event.

Nonocular AEs were all unlikely to be related to atropine use. For example, there were more cases of pyrexia in the atropine 0.005% and 0.01% groups, but these were transient and related to influenza or other viral illnesses. No serious AEs (deaths/hospitalizations) were reported during the study (Table 5).

DISCUSSION

To our knowledge, this is the first randomized study to assess the efficacy of very low doses of atropine (<0.01%) in slowing the progression of myopia in children. At 1 year, the mean changes in SE in the 0.005% atropine, 0.01% atropine, and placebo groups were −0.33±0.47, −0.39±0.51, and −0.55±0.47 D, respectively, and the mean changes in AL were 0.27±0.15, 0.25±0.25, and 0.35±0.17 mm, respectively. Atropine doses of 0.0025% appeared to have no effect on myopia progression, while atropine doses of 0.005% and 0.01% led to significant differences from placebo in 12-month LS mean changes in SE (0.23 and 0.25 D, respectively) and AL (−0.09 and −0.1 mm, respectively), indicating reduced progression. Efficacy was better in older children and those with higher baseline myopia. Treatment with these doses of atropine had little effect on distance or near logMAR BCVA but was associated with a small increase in pupil size (0.3 mm) and change in intraocular pressure (<1 mm Hg). All doses of atropine were safe and well tolerated, with the most common AEs being eye pruritus and blurred vision in 4 (4.0%) subjects each, with the highest concentration of atropine, 0.01%, being associated with eye pruritus and eye glare in 2 (8.0%) subjects each. Although high-dose topical ophthalmic atropine is known to cause pyrexia in other studies,²⁵ the frequency seen in this study appears to be related to infections.

The findings of this study indicate that a lower dose of atropine (0.005%) than that used in previous studies could slow the progression of myopia (Table 6). Atropine doses of 0.01%–0.5% have previously been found to reduce myopia progression in the ATOM2, LAMP, and ATOM-J studies.^22,26,27 The changes in SE and AL in the 0.01% group in this study were very similar to those in the ATOM2 study but less than in the LAMP and ATOM-J studies, indicating better responses. However, changes in the placebo group were also less in the APPLE study. This is potentially due to differences in the study population, age, or formulation. Nevertheless, the results from the current study are in agreement with these studies in demonstrating atropine’s dose-response relationship with changes in SE and AL. Higher doses, however, have also been associated with an increase in AEs, with increased loss of accommodation and pupil dilation leading to near-blurred vision and glare. Although only 6% of children treated with 0.01% atropine in the ATOM2 study thought that they required tinted glasses with near-vision aid, these percentages increased to 61% and 70% in children treated with 0.1% and 0.5% atropine, respectively.²² Moreover, atropine-related AEs were uncommon at 0.01%, with low-dose atropine being safe and well tolerated in children. Similar results were observed in the present study, which evaluated the effects of a lower range of atropine doses, from 0.0025% to 0.01%, finding a low incidence of ocular AEs.

TABLE 6 - Changes in SE and AL in This Study and in Other Recent Studies of Atropine for Preventing the Progression of Myopia

Study characteristics	This study (APPLE)	This study (APPLE)	ATOM222	LAMP26	ATOM-J27
Age inclusion criteria (y)	6–11		6–12	4–12	6–2
Study location	Singapore		Singapore	Hong Kong	Japan
Atropine dosage	0.005%	0.01%	0.01%	0.01%	0.01%
N	24	25	84	91	84
Baseline SE (D)	−3.80 (1.35)	−3.25 (1.11)	−4.5 (1.5)	−3.99 (1.94)	−2.91 (−3.20, −2.62)
Mean change in SE (D)
1 y	−0.33 (0.47)	−0.39 (0.52)	−0.43 (0.52)	−0.64 (0.56)	−0.69 (−0.78, −0.60)
2 y	—		−0.49 (0.63)	−1.12 (0.85)	−1.26 (−1.35, −1.17)
Mean change in AL (mm)
1 y	0.27 (0.14)	0.25 (0.25)	0.24 (0.19)	0.35 (0.24)	0.35 (0.30, 0.39)
2 y			0.41 (0.32)	0.59 (0.38)	0.63 (0.59, 0.67)
Placebo	Placebo			Placebo	Placebo
N	26			80	84
Baseline SE (D)	−3.93 (1.31)			−4.31 (1.96)	−2.98 (−3.27, −2.69)
Mean change in SE (D)
1 y	−0.55 (0.47)			−0.82 (0.49)	−0.77 (−0.87, −0.68)
Mean change in AL (mm)
1 y	0.35 (0.17)			0.43 (0.21)	0.39 (0.34, 0.43)

Variability shown as reported by study (SD or 95% CI).

AL indicates axial length; CI, confidence interval; D, diopter; SD, standard deviation; SE, spherical equivalent.

The factors that influence responses to atropine treatment warrant further studies. Environmental factors and lifestyle, including fewer outdoor activities and more work involving closer proximity to the eye, have been linked to a greater risk for myopia.^1,5,28 Demographic factors may also influence responses to atropine.^29,30

The effect of baseline myopia severity on the efficacy of atropine is unclear and may be confounded by the link between older age and higher baseline myopia.³¹ The results of the present study found trends toward greater efficacy both in children with more severe myopia at baseline and in older children. In contrast, the 2-year ATOM-J study in Japan found that efficacy was unrelated to baseline myopia severity.²⁷ A retrospective study in Chinese children found that atropine 0.01% had greater efficacy in slowing myopia progression in children with more severe myopia,³² likely because the subjects in that study were older. In the LAMP study, younger age was associated with poorer response to atropine.²⁶ A later analysis of the LAMP study found that younger children required higher doses of atropine to achieve a similar reduction in myopic progression as older children receiving lower doses.³³ The current study’s results are in accordance with previous findings that older children have a better response to atropine.

Limitations of the study include the relatively small number of children (n=25) in each group. Age was the only factor by which study subjects were stratified befire randomization, resulting in baseline differences in sex distribution among the 4 groups. Although there were also differences in baseline SE, the MMRM ANCOVA analysis with SE as a covariate ensured that the results were not affected by variations in baseline myopia. Additional limitations included the relatively short follow-up period of 12 months, and that the study did not explore outcomes following study drug discontinuation. Although this follow-up period was sufficient for a dose exploration study, it does not capture longer-term differences among groups. However, it is encouraging that these statistically significant results display a dose-related response to atropine, and if this trend continues, even a small effect may become clinically significant over time. Larger studies with longer follow-up, and preferably a withdrawal period to explore rebound effects, may confirm the clinical significance of these results. There was also no attempt to factor in differences in lifestyle (ie, time spent indoors/outdoors) or family history of myopia.

The use of atropine eye drops to control myopia requires a balance between efficacy and safety. Ideally, a minimal dose should be administered to achieve the desired effect. The present study found that, although doses ≤0.0025% were not effective, atropine doses of 0.005% and 0.01% were effective in slowing myopia, as well as being safe and well tolerated. Even at these low doses, atropine showed dose-related effects, suggesting that children who do not respond to one particular dose may respond to a higher dose. Very low doses of atropine may also play a useful supplemental role when combined with other myopia-control treatments (eg, glasses and contact lenses).

In conclusion, 0.005% and 0.01% atropine doses effectively reduced myopia progression in children aged 6–11 years with mild-to-moderate myopia. Atropine at these low doses was also safe and well tolerated.