Abstract
Introduction
Low-dose atropine eye drops slow childhood myopia progression, but randomized evidence in European children and for intermediate concentrations remains limited. This study evaluated 0.02% and 0.04% atropine for 12-month myopia control in Czech children.
Methods
The MARS study was an investigator-initiated, phase IIIb, multicenter, randomized, double-masked, placebo-controlled trial. Children aged 6–11 years with documented accelerated prerandomization axial length (AL) elongation and cycloplegic spherical equivalent refraction (SER) from −0.50 to −4.75 D were randomized 2:1:1 to nightly 0.02% atropine, 0.04% atropine, or placebo. The primary endpoint was the between-group difference in AL change at Month 12 for 0.02% atropine versus placebo. Linear mixed-effects models accounted for inter-eye correlation. Repeated-measures and multiple-imputation sensitivity analyses assessed the effect of missing data.
Results
A total of 191 participants (351 eyes) were included in the primary analysis. Compared with placebo, adjusted mean between-group differences in AL change were −0.144 mm for 0.02% atropine (95% confidence interval [CI] −0.201 to −0.087; P < 0.0001) and −0.157 mm for 0.04% atropine (95% CI −0.222 to −0.092; P < 0.0001). Corresponding differences in SER progression were 0.352 D (95% CI 0.199 to 0.505; P < 0.0001) and 0.389 D (95% CI 0.214 to 0.564; P < 0.0001). No statistically significant difference was detected between 0.02 and 0.04% atropine for AL (0.013 mm; 95% CI −0.043 to 0.069; P = 0.8935) or SER (−0.037 D; 95% CI −0.189 to 0.115; P = 0.8802). Sensitivity analyses yielded consistent AL estimates.
Conclusions
Nightly 0.02% and 0.04% atropine reduced axial elongation and myopia progression over 12 months compared with placebo. The trial was not designed or powered to establish equivalence between the active doses; within this sample, no statistically significant difference was detected between them.
Trial registration
European Clinical Trials Database (EudraCT no. 2020–002046-16), Clinical Trials Information System (CTIS), EU Clinical Trial (EU CT) no. 2024–515888-54-00.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s40123-026-01506-x.
Keywords: Myopia progression, Low-dose atropine, Axial length, Spherical equivalent refraction, Pediatric ophthalmology, Randomized controlled trial, European children
Plain Language Summary
Myopia (nearsightedness) often worsens as a child’s eye grows longer, increasing the risk of eye disease later in life. Atropine eye drops can slow this progression, but evidence in European children and for intermediate doses is limited. In the MARS trial, Czech children aged 6–11 years were randomly assigned to use 0.02% atropine, 0.04% atropine, or inactive placebo drops every night. Neither families nor investigators knew the assignment. After 12 months, both atropine groups had less eye growth and less worsening of myopia than the placebo group. Analyses that accounted for missing measurements gave similar results. The study did not detect a statistically significant difference between the two atropine doses, but it was not designed to prove that the doses are equivalent. Corneal measurements did not explain the treatment effect; the difference was driven by slower eye growth. Detailed pupil, accommodation, visual-comfort, and adverse-event outcomes are reported or planned separately. These findings support both 0.02% and 0.04% atropine as effective options for slowing myopia progression in the studied population, while longer follow-up is needed to assess durability, tolerability, and rebound after treatment stops.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s40123-026-01506-x.
Key Summary Points
| Why carry out this study? |
| Myopia is increasingly common in children, and slowing its progression may reduce the risk of sight-threatening complications later in life. |
| Randomized clinical trial evidence for low-dose atropine in European children remains limited. |
| The MARS study evaluated 12-month efficacy of nightly 0.02% and 0.04% atropine in Czech children with documented rapid axial elongation. |
| What was learned from the study? |
| Both 0.02% and 0.04% atropine significantly reduced axial elongation and myopic progression compared with placebo; repeated-measures and multiple-imputation analyses supported the primary result. |
| No statistically significant difference was detected between active doses, but the study was not designed to demonstrate equivalence; comprehensive safety and tolerability outcomes are reported separately. |
Introduction
Myopia is a major public health concern because of its increasing prevalence and its association with irreversible ocular complications. Widely cited projections by Holden et al. estimated that nearly one-half of the global population will have myopia by 2050, with approximately 10% affected by high myopia [1]. However, these estimates were based largely on modeling studies that extrapolated epidemiological trends from East Asian populations. More recent European evidence has questioned the direct applicability of these projections to Europe. A systematic review and meta-analysis by Moreira-Rosário, Lanca, and Grzybowski demonstrated substantial heterogeneity in myopia prevalence across European populations and emphasized the need for age-specific, region-specific, and cycloplegic-refraction data [2, 3].
Myopia is now recognized as a progressive disorder driven primarily by excessive axial elongation rather than simply a refractive error. Progressive elongation of the eye increases the lifetime risk of vision-threatening complications, including myopic maculopathy, retinal detachment, glaucoma, and myopic traction maculopathy [4]. Consequently, the primary objective of contemporary myopia management is to slow axial elongation during childhood, thereby reducing the cumulative lifetime risk of pathologic myopia [5].
Several interventions have demonstrated clinically meaningful efficacy in slowing myopia progression. Among pharmacologic treatments, low-dose atropine has reduced axial elongation and myopic progression in randomized clinical trials, while optical interventions such as defocus incorporated multiple segments (DIMS) and highly aspherical lenslet (HAL) spectacle lenses have also shown efficacy [6–11]. Nevertheless, the magnitude of the treatment effect appears to vary between populations, suggesting that ethnicity, iris pigmentation, genetic background, environmental factors, or their interaction may influence response to atropine [12]. Most evidence for atropine originates from East Asian populations, where treatment efficacy has been demonstrated across several concentrations. In contrast, evidence from children of European ancestry remains comparatively limited.
Recent studies in European and North American populations have begun to address this gap. The MOSAIC randomized clinical trial reported benefit of 0.05% atropine during the third treatment year in a European cohort, and CHAMP-UK reported small but statistically significant treatment effects of 0.01% atropine in UK children [13, 14]. Conversely, the Pediatric Eye Disease Investigator Group trial in North America reported no significant benefit of 0.01% atropine over placebo [15]. Randomized placebo-controlled data evaluating intermediate concentrations in a well-defined Central European white population remain limited.
The MARS trial was designed to address this evidence gap. This investigator-initiated, multicenter, randomized, double-masked, placebo-controlled phase IIIb trial evaluated nightly 0.02% and 0.04% atropine eye drops in Czech children with progressive myopia. The prespecified primary hypothesis was that 0.02% atropine would reduce 12-month axial elongation compared with placebo. Comparisons involving 0.04% atropine were secondary; the trial was not designed to establish equivalence or noninferiority between the active concentrations.
Methods
Study Design
The MARS study was an investigator-initiated, multicenter, randomized, double-masked, placebo-controlled phase IIIb clinical trial conducted at five ophthalmology centers in the Czech Republic. Recruitment occurred from June 2022 through March 2024. The present report follows CONSORT reporting principles [16]. A detailed protocol summary is provided in Supplementary File 1 and Supplementary Table S1.
The trial was registered in the European Clinical Trials Database (EudraCT no. 2020–002046-16) and the Clinical Trials Information System (EU CT no. 2024-515888-54-00). It was conducted in accordance with the Declaration of Helsinki and the International Council for Harmonisation Guideline for Good Clinical Practice. The protocol complied with SPIRIT guidance [17] and was approved by the Multicentric Ethics Committee of University Hospital Brno (72/20MEK) and the local ethics committees of General University Hospital, Prague (1930/20 S); University Hospital Motol and Homolka, Prague (EK-1395/20); University Hospital Královské Vinohrady, Prague (EK-VP/23/01/2020); and Litomyšl Hospital (Studie Binocular 2020/1).
Written informed consent was obtained from the parents or legal guardians of all participants before randomization. Children aged 12 years or older additionally provided written assent, whereas younger children provided verbal assent.
Participants
Children aged 6–11 years with progressive myopia were eligible if they had documented accelerated axial length (AL) elongation during the 6–8 months preceding enrollment, cycloplegic spherical equivalent refraction (SER) between −0.50 and −4.75 D, astigmatism < 2.50 D, best-corrected distance visual acuity of 0.2 logMAR or better, normal binocular vision, and no previous pharmacologic or optical myopia-control treatment.
To enrich the cohort for children at risk of rapid progression, protocol-specific expert-consensus criteria required prerandomization AL elongation of at least 0.10, 0.11, and 0.12 mm over 6 months for children aged 6–7, 8–9, and 10–11 years, respectively, with corresponding increments for 7- and 8-month observation intervals. The criteria were informed by European longitudinal AL growth data [19] but were not direct percentile cutoffs derived from that report. Major exclusion criteria are listed in Supplementary File 1.
Race was recorded at enrollment in the electronic case report form from information provided by the participant together with a parent or legal guardian. A legacy protocol category was harmonized for publication to the journal-preferred term “white”; all participants in the analyzed cohort were reported in this category.
Randomization, Masking, and Study Treatment
Participants were randomized in a 2:1:1 ratio to receive 0.02% atropine, 0.04% atropine, or placebo using a centralized REDCap system stratified by center and age group [18]. Additional allocation was assigned specifically to the 0.02% group because this concentration was expected to retain clinically detectable efficacy while minimizing atropine-related adverse effects; the design preserved a placebo group of sufficient size for the prespecified primary comparison.
Participants, parents or legal guardians, investigators, study coordinators, outcome assessors, and all study personnel remained masked to treatment allocation throughout the trial. To preserve masking during the ongoing study, the prespecified 12-month efficacy analysis was performed by independent statisticians who had access only to coded treatment allocation and pseudonymized participant identification numbers. They had no access to participant-identifiable information. Investigators, participants, and all other study personnel remained masked pending completion of the planned 36-month study follow-up.
The investigational medicinal product consisted of compounded ophthalmic atropine solutions (0.02% or 0.04%) or matching placebo prepared according to a validated manufacturing protocol in the hospital pharmacy. Independent laboratory testing confirmed atropine concentration and other chemical and physicochemical properties before release of each production batch. Study medications were supplied in identical eye dropper bottles labeled only with the kit number, batch number, and expiry date.
Participants instilled one drop into each eye nightly for 24 months, followed by a planned 12-month washout period. No additional pharmacologic or optical myopia-control intervention was permitted during the study. Treatment adherence and treatment-related symptoms were recorded weekly using participant diaries.
The present report describes the prespecified primary efficacy analysis after 12 months of treatment.
Study Procedures and Outcome Measures
Participants underwent comprehensive ophthalmic examinations at baseline and every 6 months according to the study protocol. A detailed list of activities performed during each visit is provided in Supplementary Table 1.
Cycloplegic autorefraction was performed using a standardized protocol consisting of topical oxybuprocaine followed by two instillations of 1% cyclopentolate administered 10 min apart. An additional drop of cyclopentolate was administered if residual pupillary light reflex was present. Autorefraction measurements were obtained 30 min after initiation of cycloplegia.
The prespecified primary endpoint was the between-group difference in the change in AL from baseline to month 12 between the 0.02% atropine and placebo groups.
Prespecified secondary efficacy outcomes included changes in AL in the 0.04% atropine group, cycloplegic SER, corneal radius (CR), and the AL/CR ratio. Corneal radius was calculated from mean anterior keratometry (CR = 337.5/mean keratometry), and SER was calculated as the spherical component plus one-half of the cylindrical component.
Because different but equivalent ophthalmic instruments were used across participating centers, each participant was examined using the same instrument throughout the study. Consequently, all efficacy analyses were based on within-subject changes from baseline, thereby minimizing systematic measurement bias related to the multicenter study design. The instruments used for refractive, corneal curvature, and axial length measurements at each participating center are listed in Supplementary Table S2.
Statistical Analysis
The planned sample size was 288 participants, providing 80% power to detect a 0.10-mm difference in 12-month AL change between 0.02% atropine and placebo at a two-sided alpha of 0.05, allowing for 10% attrition. Recruitment was stopped early after effective myopia-control spectacle lenses entered routine Czech practice and increasing numbers of eligible families declined possible placebo allocation. A post hoc calculation based on the achieved sample size and observed AL values indicated greater than 90% power for the primary comparison; the study was not powered for the comparison between active doses.
The primary modified intention-to-treat analysis included randomized participants with evaluable month 12 data and excluded major protocol deviations and participants who initiated defocus incorporated multiple segments (DIMS) spectacles during the first 12 months. Participant disposition is shown in Fig. 1 and Table 1.
Fig. 1.
CONSORT flow diagram of participant enrollment, randomization, follow-up, and analysis for the 12-month primary outcome assessment
Table 1.
Participant disposition
| Overall | Placebo | Atropine 0.02% | Atropine 0.04% | |
|---|---|---|---|---|
| Screened | 238 | |||
| Randomized | 235 (100%) | 58 (100%) | 119 (100%) | 58 (100%) |
| Major protocol deviation | 6 (2.6%) | 2 (3.4%) | 4 (3.4%) | |
| DIMS spectacles used during first 12 months | 5 (2.1%) | 1 (1.7%) | 3 (2.5%) | 1 (1.7%) |
| Change at 12 months is not evaluable for any endpoint | 33 (14.0%) | 7 (12.1%) | 18 (15.1%) | 8 (13.8%) |
| Analyzed | 191 (81.3%) | 48 (82.8%) | 94 (79.0%) | 49 (84.5%) |
| Number of analyzed eyes | 351 | 90 | 170 | 91 |
Categorical variables are presented as frequencies and percentages. Continuous variables are summarized using means and standard deviations, medians, first and third quartiles, and ranges.
For 12-month outcomes, change from baseline was analyzed using linear mixed-effects models with treatment group and the corresponding baseline value as fixed effects and a random participant-level intercept. The random intercept induces a compound-symmetric covariance structure between fellow eyes, with equal within-participant variance and a single constant inter-eye correlation. Least-squares mean changes and between-group differences are reported with 95% confidence intervals.
The single prespecified confirmatory comparison was 0.02% atropine versus placebo for AL change. Tukey–Kramer adjustment was applied to pairwise treatment-group comparisons within each efficacy endpoint. No formal multiplicity adjustment was applied across endpoints; secondary analyses are therefore interpreted descriptively.
Two sensitivity analyses assessed the primary AL endpoint. Both used a mixed-effects model for repeated measures with change from baseline at Months 6 and 12 as the dependent variable; treatment, visit, treatment-by-visit interaction, and baseline AL as fixed effects; a random participant intercept; and an unstructured covariance for visits within eye nested within participant. Measurements obtained after DIMS initiation were treated as missing.
The first sensitivity analysis included randomized participants with at least one nonmissing post-baseline AL measurement (N = 198) without imputation. The second used multiple imputation among all 235 randomized participants. Intermittent missing values were imputed within treatment arm using Markov chain Monte Carlo, followed by monotone regression using preceding AL, sex, and age at randomization. Fifty datasets were generated; model estimates were pooled using Rubin’s rules.
All tests were two-sided, and P < 0.05 was considered statistically significant. Analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA); figures were generated using R version 4.5.1 (R Foundation for Statistical Computing, Vienna, Austria).
Results
Study Population
Between June 2022 and March 2024, 238 children were screened and 235 were randomized to placebo (n = 58), 0.02% atropine (n = 119), or 0.04% atropine (n = 58). Six participants had major protocol deviations, five initiated DIMS spectacles during the first 12 months, and 33 had no evaluable month 12 data. The primary analysis therefore included 191 participants (81.3% of those randomized; 351 eyes): 48 placebo, 94 atropine 0.02%, and 49 atropine 0.04% (Table 1; Fig. 1).
Baseline characteristics were comparable across groups (Table 2). Mean age was 10.13 years (SD 1.42) overall: 10.11 (1.44) in placebo, 10.16 (1.48) in 0.02% atropine, and 10.11 (1.31) in 0.04% atropine. All analyzed participants were reported as white.
Table 2.
Demographic characteristics and iris color
| Characteristics | Statistics | Placebo N = 48 |
Atropine 0.02% N = 94 |
Atropine 0.04% N = 49 |
Overall N = 191 |
|---|---|---|---|---|---|
| Sex | n/n miss | 48/0 | 94/0 | 49/0 | 191/0 |
| Female | n (%) | 32 (66.7) | 53 (56.4) | 30 (61.2) | 115 (60.2) |
| Male | n (%) | 16 (33.3) | 41 (43.6) | 19 (38.8) | 76 (39.8) |
| Age category at randomization | n/n miss | 48/0 | 94/0 | 49/0 | 191/0 |
| 6–7 | n (%) | 4 (8.3) | 8 (8.5) | 5 (10.2) | 17 (8.9) |
| 8–9 | n (%) | 16 (33.3) | 32 (34.0) | 14 (28.6) | 62 (32.5) |
| 10–11 | n (%) | 28 (58.3) | 54 (57.4) | 30 (61.2) | 112 (58.6) |
| Age at randomization, years | n/n miss | 48/0 | 94/0 | 49/0 | 191/0 |
| Mean (SD) | 10.11 (1.44) | 10.16 (1.48) | 10.11 (1.31) | 10.13 (1.42) | |
| 95% CI for mean | 9.76; 10.46 | 9.90; 10.41 | 9.79; 10.42 | 9.96; 10.30 | |
| Q1/Median/Q3 | 9.44/10.22/11.34 | 9.22/10.28/11.37 | 9.24/10.45/10.96 | 9.24/10.32/11.33 | |
| Min; Max | 6.15; 12.00 | 6.53; 12.00 | 7.10; 11.99 | 6.15; 12.00 | |
| Parent/guardian-reported race | n/n miss | 48/0 | 94/0 | 49/0 | 191/0 |
| White | n (%) | 48 (100) | 94 (100) | 49 (100) | 191 (100) |
| Iris color | n/n miss | 48/0 | 94/0 | 49/0 | 191/0 |
| Lighter blue | n (%) | 8 (16.7) | 6 (6.4) | 6 (12.2) | 20 (10.5) |
| Darker blue | n (%) | 5 (10.4) | 14 (14.9) | 2 (4.1) | 21 (11.0) |
| Blue with a brown pupillary border | n (%) | 7 (14.6) | 24 (25.5) | 7 (14.3) | 38 (19.9) |
| Green | n (%) | 3 (6.3) | 8 (8.5) | 5 (10.2) | 16 (8.4) |
| Green with a brown pupillary border | n (%) | 5 (10.4) | 7 (7.4) | 5 (10.2) | 17 (8.9) |
| Peripherally green, centrally brown | n (%) | 5 (10.4) | 4 (4.3) | 3 (6.1) | 12 (6.3) |
| Brown with a green outer border | n (%) | 3 (6.3) | 12 (12.8) | 5 (10.2) | 20 (10.5) |
| Brown | n (%) | 6 (12.5) | 14 (14.9) | 11 (22.4) | 31 (16.2) |
| Dark brown | n (%) | 6 (12.5) | 5 (5.3) | 5 (10.2) | 16 (8.4) |
Primary Outcome: Axial Length
Changes in axial length are summarized in Tables 3 and 4 and Figs. 2 and 3. Descriptive mean (SD) AL changes at Month 6 were 0.13 (0.128) mm with placebo, 0.07 (0.182) mm with 0.02% atropine, and 0.04 (0.110) mm with 0.04% atropine. At month 12, the placebo group had the greatest model-estimated increase (least-squares mean, 0.265 mm; 95% CI 0.219–0.311).
Table 3.
Axial length and 6-month and 12-month change from baseline
| Visit | Type of measurement | Statistics | Placebo Neyes = 90 (Nsub = 48) |
Atropine 0.02% Neyes = 170 (Nsub = 94) |
Atropine 0.04% Neyes = 91 (Nsub = 49) |
Total Neyes = 351 (Nsub = 191) |
|---|---|---|---|---|---|---|
| Baseline | Value at visit | n/n miss | 90/0 | 170/0 | 91/0 | 351/0 |
| Mean (SD) | 24.39 (0.795) | 24.35 (0.826) | 24.47 (0.867) | 24.39 (0.828) | ||
| Q1/Median/Q3 | 23.80/24.37/24.93 | 23.72/24.36/24.87 | 23.87/24.34/25.31 | 23.80/24.34/24.93 | ||
| Min; Max | 23.0; 26.6 | 22.0; 26.0 | 22.4; 26.3 | 22.0; 26.6 | ||
| 6 months | Value at visit | n/n miss | 90/0 | 170/0 | 91/0 | 351/0 |
| Mean (SD) | 24.52 (0.823) | 24.42 (0.808) | 24.50 (0.874) | 24.47 (0.828) | ||
| Q1/Median/Q3 | 23.96/24.43/25.11 | 23.85/24.43/24.89 | 23.90/24.37/25.29 | 23.88/24.42/25.03 | ||
| Min; Max | 23.0; 26.8 | 22.1; 26.1 | 22.5; 26.3 | 22.1; 26.8 | ||
| Change from baseline | n/n miss | 90/0 | 170/0 | 91/0 | 351/0 | |
| Mean (SD) | 0.13 (0.128) | 0.07 (0.182) | 0.04 (0.110) | 0.08 (0.157) | ||
| Q1/Median/Q3 | 0.04/0.13/0.20 | −0.03/0.04/0.12 | −0.02/0.02/0.10 | 0.00/0.05/0.14 | ||
| Min; Max | −0.3; 0.5 | −0.3; 1.1 | −0.2; 0.5 | −0.3; 1.1 | ||
| 12 months | Value at visit | n/n miss | 90/0 | 170/0 | 91/0 | 351/0 |
| Mean (SD) | 24.66 (0.856) | 24.48 (0.815) | 24.57 (0.871) | 24.55 (0.841) | ||
| Q1/Median/Q3 | 24.03/24.62/25.24 | 23.86/24.48/24.96 | 23.94/24.40/25.30 | 23.92/24.50/25.11 | ||
| Min; Max | 23.0; 27.0 | 22.3; 26.3 | 22.7; 26.4 | 22.3; 27.0 | ||
| Change from baseline | n/n miss | 90/0 | 170/0 | 91/0 | 351/0 | |
| Mean (SD) | 0.27 (0.191) | 0.13 (0.170) | 0.11 (0.136) | 0.16 (0.180) | ||
| Q1/Median/Q3 | 0.13/0.24/0.39 | 0.01/0.12/0.20 | 0.02/0.09/0.17 | 0.04/0.13/0.26 | ||
| Min; Max | −0.1; 0.8 | −0.3; 1.1 | −0.2; 0.5 | −0.3; 1.1 |
Max maximum, Min minimum, Neyes total number of treated eyes, Nsub total number of treated subjects, n number of eyes with available data at visit, n miss number of eyes with missing data at visit, Q1 first quartile, Q3 third quartile, SD standard deviation
Table 4.
Linear mixed-effects model of 12-month change in axial length from baseline
| Treatment group | LSM of change from BL (95% CI) |
Difference from 0 P value |
Treatment effect P value |
|---|---|---|---|
| Placebo [N participants = 48/N eyes = 90] | 0.265 (0.219; 0.311) | < 0.0001 | < 0.0001 |
| Atropine 0.02% [N participants = 94/N eyes = 170] | 0.121 (0.088; 0.154) | < 0.0001 | |
| Atropine 0.04% [N participants = 49/N eyes = 91] | 0.108 (0.062; 0.154) | < 0.0001 | |
| Post hoc difference between treatment armsa | |||
| Atropine 0.02%—Atropine 0.04% | 0.013 (−0.043; 0.069) | 0.8935 | |
| Atropine 0.02%—Placebo | −0.144 (−0.201; −0.087) | < 0.0001 | |
| Atropine 0.04%—Placebo | −0.157 (−0.222; −0.092) | < 0.0001 |
Estimates and P values were obtained using a linear mixed-effects model with change from baseline as the dependent variable, treatment group as a fixed effect, and baseline axial length as a covariate. Inter-eye correlation was modeled using a random participant-level intercept, inducing a compound-symmetric covariance structure between fellow eyes
BL baseline, CI confidence interval, LSM least-squares mean, N eyes number of treated eyes with an available change from baseline, N participants number of treated participants with an available change from baseline
apost hoc differences between treatment arms. Pairwise P values were adjusted using the Tukey–Kramer procedure
Fig. 2.
Change in axial length from baseline (observed mean ± standard error of the mean)
Fig. 3.
Model-estimated change in axial length from baseline (marginal mean ± standard error)
Compared with placebo, 0.02% atropine reduced 12-month AL elongation by 0.144 mm (95% CI −0.201 to −0.087; P < 0.0001), and 0.04% atropine reduced it by 0.157 mm (95% CI −0.222 to −0.092; P < 0.0001). No statistically significant difference was detected between the active concentrations (0.013 mm; 95% CI −0.043 to 0.069; P = 0.8935). This comparison does not establish equivalence.
Sensitivity analyses were consistent with the primary result (Supplementary Tables S4 and S5). In the repeated-measures analysis without imputation, treatment differences versus placebo were −0.141 mm (95% CI −0.202 to −0.081) for 0.02% and −0.154 mm (95% CI −0.224 to −0.085) for 0.04%. With multiple imputation, the corresponding differences were −0.141 mm (95% CI −0.190 to −0.093) and −0.157 mm (95% CI −0.212 to −0.101); all placebo comparisons had P < 0.0001.
Secondary Outcome: Cycloplegic Spherical Equivalent Refraction
Changes in cycloplegic SER are presented in Tables 5 and 6 and Figs. 4 and 5. Myopic progression was greatest with placebo (least-squares mean, −0.501 D; 95% CI −0.625 to −0.377). Cycloplegic SER was not collected at Month 6 according to the prespecified visit schedule.
Table 5.
Cycloplegic spherical equivalent refraction and 12-month change from baseline
| Visit | Type of measurement | Statistics | Placebo N eyes = 90 (N participants = 48) |
Atropine 0.02% N eyes = 170 (N participants = 94) |
Atropine 0.04% N eyes = 91 (N participants = 49) |
Total N eyes = 351 (N participants = 191) |
|---|---|---|---|---|---|---|
| Baseline | Value at visit | n/missing n | 87/1 | 170/0 | 90/1 | 347/2 |
| Mean (SD) | −2.97 (1.230) | −2.80 (1.178) | −2.90 (1.266) | −2.87 (1.213) | ||
| Q1/Median/Q3 | −3.88/−3.00/−2.13 | −3.63/−2.69/−1.88 | −3.75/−2.75/−2.13 | −3.75/−2.75/−2.00 | ||
| Min; Max | −5.5; −0.5 | −5.8; −0.5 | −5.9; 0.0 | −5.9; 0.0 | ||
| 12 months | Value at visit | n/missing n | 90/0 | 170/0 | 90/1 | 350/1 |
| Mean (SD) | −3.47 (1.299) | −2.97 (1.269) | −3.02 (1.314) | −3.11 (1.302) | ||
| Q1/Median/Q3 | −4.38/−3.50/−2.75 | −4.00/−2.81/−2.00 | −4.00/−3.00/−2.00 | −4.13/−3.13/−2.13 | ||
| Min; Max | −6.8; −0.4 | −5.9; 1.0 | −5.9; −0.6 | −6.8; 1.0 | ||
| Change from baseline | n/missing n | 89/1 | 170/0 | 89/2 | 348/3 | |
| Mean (SD) | −0.51 (0.496) | −0.17 (0.435) | −0.12 (0.424) | −0.24 (0.474) | ||
| Q1/Median/Q3 | −0.88/−0.50/−0.13 | −0.50/−0.13/0.00 | −0.38/−0.13/0.13 | −0.50/−0.25/0.00 | ||
| Min; Max | −1.8; 0.5 | −1.4; 2.3 | −1.3; 0.9 | −1.8; 2.3 |
Max maximum, Min minimum, N eyes total number of treated eyes, N participants total number of treated participants, n number of eyes with available data at visit, n miss number of eyes with missing data at visit, Q1 first quartile, Q3 third quartile, SD standard deviation
Table 6.
Linear mixed-effects model of 12-month change in cycloplegic spherical equivalent refraction from baseline
| Treatment group | LSM of change from BL (95% CI) |
Difference from 0 P value |
Treatment effect P value |
|---|---|---|---|
| Placebo [N participants = 48/N eyes = 89] | −0.501 (−0.625; −0.377) | < 0.0001 | < 0.0001 |
| Atropine 0.02% [N participants = 94/N eyes = 170] | −0.149 (−0.238; −0.060) | 0.0012 | |
| Atropine 0.04% [N participants = 49/N eyes = 89] | −0.112 (−0.235; 0.011) | 0.0740 | |
| Post hoc difference between treatment arms | |||
| Atropine 0.02%—Atropine 0.04% | −0.037 (−0.189; 0.115) | 0.8802 | |
| Atropine 0.02%—Placebo | 0.352 (0.199; 0.505) | < 0.0001 | |
| Atropine 0.04%—Placebo | 0.389 (0.214; 0.564) | < 0.0001 |
BL baseline, LSM least-squares mean, N eyes number of eyes with an available change from baseline, N participants number of participants with an available change from baseline. Pairwise P values were adjusted using the Tukey–Kramer procedure. Change from baseline was the dependent variable, treatment group was a fixed effect, and baseline spherical equivalent was a covariate. Inter-eye correlation was modeled using a random participant-level intercept, inducing a compound-symmetric covariance structure between fellow eyes
Fig. 4.
Change in cycloplegic spherical equivalent refraction from baseline (observed mean ± standard error of the mean)
Fig. 5.
Model-estimated change in cycloplegic spherical equivalent refraction from baseline (marginal mean ± standard error)
Compared with placebo, 0.02% atropine reduced SER progression by 0.352 D (95% CI 0.199–0.505; P < 0.0001), and 0.04% atropine reduced it by 0.389 D (95% CI 0.214–0.564; P < 0.0001). No statistically significant difference was detected between active concentrations (−0.037 D; 95% CI −0.189 to 0.115; P = 0.8802). The within-arm change for 0.04% atropine was not significantly different from zero (−0.112 D; 95% CI −0.235 to 0.011; P = 0.0740), whereas its treatment contrast with placebo was significant; these tests address different questions.
Secondary Outcome: Corneal Parameters
Changes in corneal biometric parameters are summarized in Supplementary Fig. S1. No statistically significant between-group differences were detected for mean anterior keratometry, mean anterior corneal radius, mean posterior keratometry, or white-to-white corneal diameter at Month 12.
The AL/corneal radius ratio increased in all groups, but its smaller increase in the atropine groups reflected reduced AL elongation rather than altered corneal geometry. Corneal measures were evaluated as mechanistic outcomes and were not intended to represent a comprehensive safety assessment.
Overall, both atropine concentrations reduced axial elongation and SER progression compared with placebo, and sensitivity analyses supported the primary finding. No statistically significant difference was detected between active doses, but equivalence was not tested.
Discussion
The MARS trial showed that nightly 0.02% and 0.04% atropine reduced 12-month axial elongation and cycloplegic SER progression compared with placebo in Czech children with documented rapid prerandomization eye growth. Repeated-measures and multiple-imputation analyses produced closely similar AL estimates, supporting robustness to the evaluated missing-data assumptions.
The cohort differs importantly from other European trials. MARS randomized 235 children aged 6–11 years and analyzed a cohort with mean age 10.13 years, all reported as white, and selected for rapid pre-randomization AL growth. CHAMP-UK randomized 289 children aged 6–12 years (mean age 9.3 years; 72% white), whereas MOSAIC randomized 250 children aged 6–16 years (mean age approximately 11.8 years; 83% white) without the same accelerated-AL enrichment criterion [13, 14]. These design and population differences limit direct effect-size comparisons but place MARS within the spectrum of Western pediatric atropine trials.
Recruitment ended before the planned sample size after DIMS and HAL spectacle lenses became available in routine Czech practice and families became less willing to accept placebo randomization. Enrollment had begun before these options were widely used, and stopping recruitment limited further change in the source population; nevertheless, selection bias cannot be excluded. Individual-level baseline data for eligible children who declined participation were not collected, so representativeness could not be assessed directly. Five randomized children initiated DIMS during follow-up; they were excluded from the primary analysis, and post-DIMS measurements were treated as missing in sensitivity analyses.
The observed effects are consistent with randomized evidence that atropine can slow myopia progression, although magnitude varies with dose, age, baseline progression, population, and follow-up [6, 13, 14, 20–23]. The accelerated-growth eligibility criterion may partly explain the comparatively rapid placebo progression and the apparent treatment effect in MARS relative to broader or older European cohorts.
The absence of a statistically significant difference between 0.02% and 0.04% atropine should not be interpreted as equivalence. The study was designed and powered for 0.02% atropine versus placebo, no equivalence margin was prespecified, and the 95% CI for the active-dose AL difference (−0.043 to 0.069 mm) includes differences that may be clinically relevant. A post hoc equivalence test was therefore not performed. The data support the statement that no active-dose difference was detected in this sample, not that the concentrations are interchangeable.
For SER, the 0.04% within-arm estimate did not differ significantly from zero, but the 0.04%-versus-placebo contrast was highly significant. The former evaluates absolute progression within one arm and is sensitive to uncertainty around that estimate; the latter is the randomized treatment contrast and is the more relevant efficacy comparison.
Corneal stability indicates that the SER treatment effect was not explained by altered corneal curvature. These corneal measures were mechanistic, not comprehensive safety endpoints. Pupil and accommodation findings from 127 MARS participants have been reported separately, including treatment-related pupil dilation with no significant between-group difference in near accommodation facility [24]. Detailed near visual acuity, visual comfort, ocular irritation, and adverse-event analyses are being prepared separately to avoid fragmentary reporting before the planned longer-term safety dataset is complete.
Strengths include the randomized, double-masked, placebo-controlled design, documented prerandomization progression, standardized within-participant measurements, and consistent sensitivity analyses. Limitations include early recruitment termination, unavailable characteristics for eligible nonparticipants, 18.7% exclusion from the primary analysis, limited power for active-dose comparisons, lack of prespecified equivalence margins, and restriction to a Czech white cohort. The present report covers the prespecified 12-month efficacy analysis; longer-term efficacy, rebound, and comprehensive safety and tolerability will be reported separately.
Conclusions
Nightly 0.02% and 0.04% atropine reduced 12-month axial elongation and myopia progression compared with placebo in Czech children with documented rapid eye growth. No statistically significant difference was detected between the active concentrations, but the study was not designed to establish equivalence or noninferiority. Longer follow-up is needed to assess durability, tolerability, and rebound after treatment cessation.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors thank the children and families who participated in the MARS trial and the complete MARS study team listed in Supplementary File 2.
Medical Writing/Editorial Assistance
During revision, the authors used ChatGPT (OpenAI) for language editing and organization of responses. All AI-assisted text was critically reviewed and approved by the authors, who take full responsibility for the content. No external funding was received specifically for AI assistance.
Author Contributions
Martin Hložánek: conceptualization, investigation, methodology, project administration, and writing—original draft and review and editing. Miroslav Dostálek: conceptualization, investigation, methodology, and writing—review and editing. Rudolf Autrata: conceptualization, funding acquisition, investigation, supervision, and writing—review and editing. Radka Štěpánová: formal analysis, methodology, visualization, and writing—review and editing. Jana Unar Vinklerová: clinical monitoring, methodology, validation, and writing—review and editing. Jarmila Heissigerová and Pavel Studený: investigation, supervision, and writing—review and editing. Barbora Čáslavská, Lucie Jeřábková, Nina Zelenayová, Inka Krejčířová, Renáta Brunnerová, Marta Korbasová, and Martin Komínek: investigation, data acquisition, and writing—review and editing. All authors approved the final manuscript and accept accountability for the work.
Funding
The clinical trial was supported by the Ministry of Health of the Czech Republic (grant no. NU21-07-00189) and was conducted in collaboration with the Faculty of Medicine, Masaryk University, through LRI CZECRIN (LM2023049), supported by the Ministry of Education, Youth and Sports from the national budget. No funding or sponsorship was received for the publication of this article.
Data Availability
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request, subject to applicable consent, ethics, and data-protection requirements.
Declarations
Conflict of Interest
Martin Hložánek, Miroslav Dostálek, Rudolf Autrata, Radka Štěpánová, Jana Unar Vinklerová, Pavel Studený, Barbora Čáslavská, Lucie Jeřábková, Nina Zelenayová, Inka Krejčířová, Renáta Brunnerová, Marta Korbasová, and Martin Komínek have nothing to disclose. Jarmila Heissigerová is an editorial board member of Ophthalmology and Therapy. Jarmila Heissigerová was not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial decisions.
Ethical Approval
The protocol was approved by the Multicentric Ethics Committee of University Hospital Brno (72/20MEK) and the local ethics committees of General University Hospital, Prague (1930/20 S); University Hospital Motol and Homolka, Prague (EK-1395/20); University Hospital Královské Vinohrady, Prague (EK-VP/23/01/2020); and Litomyšl Hospital (Studie Binocular 2020/1). The study was conducted in accordance with the Declaration of Helsinki and ICH-GCP. Written informed consent was obtained from parents or legal guardians, with age-appropriate assent from participants.
Footnotes
Prior presentation: Partially presented at World Ophthalmology Congress, 26–29 June 2026, Prague, Czech Republic.
Prior publication: The MARS study protocol was previously published: Komínek M, Autrata R, Krejčířová I, et al. Myopia Control in European Children—Study Protocol and Methodology. J Neonatol Clin Pediatr. 2022;9:096. doi:10.24966/NCP-878X/100096.
Accommodation and pupillary outcomes from a subset of MARS participants were reported previously: Dostálek M, Žajdlíková B, Komínek M, et al. Evaluation of Accommodation and Pupillary Reaction in Children with Myopia Treated by Highly Diluted Atropine in the MARS Trial. J Clin Case Stud Rev Rep. 2024;6:327. doi:10.47363/JCCSR/2024(6)327.
MARS study collaborators (other than the authors): Pavel Diblík; Marcela Michaličková; Kamila Víchová; Katarína Hanáková; Lenka Cilečková; Kateřina Kapičáková; Katarína Krivosheev; Michal Janů; Monika Golianová; Denisa Šikulová; Šárka Matyšková; Barbora Hráčková; Lucie Petříčková; Martina Krejzová; Martina Čiháková; Michaela Ducháčková; Marcela Kudová; Tereza Kašparová; Lenka Vejsová; Jitka Dulajová; Iva Malcová; Jaroslava Čeledová; Jana Kubátová; Ludmila Búdová; Lucie Havlíčková; Pavlína Pokorná; Martina Kyseláková; Martina Nováková; Jana Vaňková; Hana Ježková; Denisa Bobková; Kamila Součková; Gabriela Jirsáková; Petr Prášek; Marcela Doušová; Patrik Pluhovský.
References
- 1.Holden BA, Fricke TR, Wilson DA, Jong M, Naidoo KS, Sankaridurg P, et al. Global prevalence of myopia and high myopia and temporal trends from 2000 through 2050. Ophthalmology. 2000;123(5):1036–42. [DOI] [PubMed] [Google Scholar]
- 2.Moreira-Rosario A, Lanca C, Grzybowski A. Prevalence of myopia in Europe: a systematic review and meta-analysis of data from 14 countries. Lancet Reg Health Eur. 2025;54:101319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lanca C, Morgan I, Grzybowski A. Myopia trends in Europe: a reappraisal of current evidence. Ophthalmol Ther. 2026;15(7):2107–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Haarman AEG, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJM, Klaver CCW. The complications of myopia: a review and meta-analysis. Invest Ophthalmol Vis Sci. 2020;61(4):49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bullimore MA, Ritchey ER, Shah S, Leveziel N, Bourne RRA, Flitcroft DI. The risks and benefits of myopia control. Ophthalmology. 2021;128(11):1561–79. [DOI] [PubMed] [Google Scholar]
- 6.Yam JC, Jiang Y, Tang SM, Law AKP, Chan JJ, Wong E, et al. Low-Concentration Atropine for Myopia Progression (LAMP) study: a randomized, double-blinded, placebo-controlled trial of 0.05%, 0.025%, and 0.01% atropine eye drops in myopia control. Ophthalmology. 2019;126(1):113–24. [DOI] [PubMed]
- 7.Lam CSY, Tang WC, Tse DY, Lee RPK, Chun RKM, Hasegawa K, et al. Defocus incorporated multiple segments (DIMS) spectacle lenses slow myopia progression: a 2-year randomised clinical trial. Br J Ophthalmol. 2020;104(3):363–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bao J, Huang Y, Li X, Yang A, Zhou F, Wu J, et al. Spectacle lenses with aspherical lenslets for myopia control vs single-vision spectacle lenses: a randomized clinical trial. JAMA Ophthalmol. 2022;140(5):472–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Walline JJ, Lindsley KB, Vedula SS, Cotter SA, Mutti DO, Ng SM, et al. Interventions to slow progression of myopia in children. Cochrane Database Syst Rev. 2020;1:CD004916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ha A, Kim SJ, Shim SR, Kim YK, Jung JH. Efficacy and safety of 8 atropine concentrations for myopia control in children: a network meta-analysis. Ophthalmology. 2022;129(3):322–33. [DOI] [PubMed] [Google Scholar]
- 11.Lawrenson JG, Shah R, Huntjens B, Downie LE, Virgili G, Dhakal R, et al. Interventions for myopia control in children: a living systematic review and network meta-analysis (Review). In: Collaboration C, editor., et al., Cochrane Database of Systematic Reviews. London: Cochrane Library; 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Biswas S, Lanca C, Kynman-Sprinks D, Grzybowski A. Ethnic and dose-dependent differences in atropine efficacy for myopia control: a systematic review and meta-analysis. Br J Ophthalmol. 2026;110(5):483–91. [DOI] [PubMed] [Google Scholar]
- 13.Loughman J, Lingham G, Nkansah EK, Kobia-Acquah E, Flitcroft DI. Efficacy and safety of different atropine regimens for the treatment of myopia in children: three-year results of the MOSAIC randomized clinical trial. JAMA Ophthalmol. 2025;143(2):134–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Azuara-Blanco A, Logan NS, McConnell E, Kearney S, Kirk G, Jones S, et al. Low concentration atropine eye drops and progression of myopia in children: multicentre placebo controlled, double masked, randomised trial in the UK (CHAMP-UK). BMJ. 2026;393:e086698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Repka MX, Weise KK, Chandler DL, Wu R, Melia BM, Manny RE, et al. Low-dose 0.01% atropine eye drops vs placebo for myopia control: a randomized clinical trial. JAMA Ophthalmol. 2023;141(8):756–65. [DOI] [PMC free article] [PubMed]
- 16.Schulz KF, Altman DG, Moher D, Group C. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chan AW, Tetzlaff JM, Altman DG, Laupacis A, Gotzsche PC, Krleza-Jeric K, et al. SPIRIT 2013 statement: defining standard protocol items for clinical trials. Ann Intern Med. 2013;158(3):200–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tideman JWL, Polling JR, Vingerling JR, Jaddoe VWV, Williams C, Guggenheim JA, et al. Axial length growth and the risk of developing myopia in European children. Acta Ophthalmol. 2018;96(3):301–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hvid-Hansen A, Jacobsen N, Møller F, Bek T, Ozenne B, Kessel L. Myopia control with low-dose atropine in european children: six-month results from a randomized, double-masked, placebo-controlled, multicenter study. J Pers Med. 2023;13(2):325. 10.3390/jpm13020325. [DOI] [PMC free article] [PubMed]
- 21.Simonaviciute D, Gelzinis A, Kapitanovaite L, Grzybowski A, Zemaitiene R. Myopia control in Caucasian children with 0.01% atropine eye drops: 1-year follow-up study. Medicina (Kaunas). 2024. 10.3390/medicina60071022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Simonaviciute D, Grzybowski A, Gelzinis A, Zemaitiene R. Efficacy and safety of 0.03% atropine eye drops in controlling myopia progression: a one-year prospective clinical study. J Clin Med. 2024. 10.3390/jcm13113218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Loughman J, Kobia-Acquah E, Lingham G, Butler J, Loskutova E, Macley DA, et al. Myopia outcome study of atropine in children: two-year result of daily 0.01% atropine in a European population. Acta Ophthalmol. 2024;102(3):e245–e256. [DOI] [PubMed]
- 24.Dostálek M, Zajdlíková B, Komínek M, Krejčířová I, Heissigerová J, Autrata R, et al. Evaluation of accommodation and pupillary reaction in children with myopia treated by highly diluted atropine in the MARS trial. J Clin Case Stud Rev Rep. 2024;6:327. 10.47363/JCCSR/2024(6)327. [DOI]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request, subject to applicable consent, ethics, and data-protection requirements.