Abstract
We combined data from 121 amblyopic children enrolled in two prospective open-label pilot studies and a randomized trial of a binocular digital therapeutic to identify factors associated with response to amblyopia treatment. Visual acuity improved ≥1 line in 81% of participants after 12 weeks of therapy. Treatment response was not found to be associated with age, severity of amblyopia, or prior treatment status. While these findings may suggest broad efficacy for this treatment approach, further investigation in larger cohorts is needed to identify factors associated with treatment response. The potential role of binocular therapy in populations where standard amblyopia treatments have been less effective warrants a dedicated clinical trial.
Understanding the factors associated with treatment success and failure of dichoptic therapy delivered through a binocular digital therapeutic1 may help identify children who would benefit from this novel intervention for amblyopia. Poor response to patching and atropine eyedrops for amblyopia therapy is associated with older age and history of failed prior treatment;2 however, factors associated with response to binocular digital therapy are not yet well described. The goal of this investigation is to evaluate outcomes from previously conducted clinical trials of a binocular digital therapeutic (Luminopia) to examine factors associated with treatment response. These findings may help guide patient selection for dichoptic treatment in actual clinical practice where therapeutic options are expanding.
Methods:
This retrospective cohort study combined data from two prospective open-label pilot studies and the treatment arm of a randomized trial of a binocular digital therapeutic for amblyopia. This study adhered to the tenets of the Declaration of Helsinki. Institutional review board or ethics committee approval was obtained at all participating sites. Written informed consent was obtained from the parents or guardians of each participant, as described previously.3,4
The prospective open-label pilot studies were performed between 08/2017 and 10/2018 and included 90 children aged 4-12 years.3 The treatment arm of the randomized control trial was performed between 01/2019 and 04/2020 and included 51 children aged 4-7 years.4 For this analysis, we included all children between 4 and 12 years of age with best corrected visual acuity measurements at the initial and 12 week follow-up visits. All children in these two studies were prescribed a 12-week, one hour/day, 6 days/week treatment regimen involving a digital therapeutic (Luminopia; Luminopia Inc.) that delivered dual-acting dichoptic amblyopia therapy consisting of signal rebalancing through fixed contrast reduction in the fellow eye, and fusion encouragement through complimentary dichoptic masking (signal suppression) between the two eyes.1
The primary outcome was treatment response, defined as improvement in the visual acuity of the amblyopic eye by ≥1 line between the initial and 12-week follow-up visits. The secondary outcome was any improvement in stereoacuity (>0 arcsec) between the initial and 12-week follow-up visits.
We collected baseline demographic, diagnostic, and treatment data, including age, sex, amblyopia laterality, best corrected visual acuity, presence of strabismus, prior treatment, and adherence to treatment. Visual acuity was measured monocularly in each eye in current refractive correction using an electronic visual acuity system with the ATS-HOTV protocol.5 Age at enrollment, in years ranging from 4 to 12, was treated as a continuous variable and further categorized into 4 to ≤7 years old and 8 to ≤12 years old. Details on amblyopia diagnosis collected included the laterality and severity. Baseline visual acuity was categorized as 20/40 or better vs. 20/50 to 20/200. Intraocular visual acuity difference (<4 vs. ≥4 line difference), and stereoacuity (<400 arcsec vs. ≥ 400 arcsec) were dichotomized. The presence of strabismus was treated as a dichotomous variable based on the diagnosis of strabismic amblyopia or mixed (strabismic and refractive) amblyopia. All the children with strabismus included in the studies had 5 prism diopters or less at distance by simultaneous prism and cover test. Details of prior treatment including the type of prior treatment (patching or atropine eyedrops) and the duration of prior treatment (≤6 months or >6 months) were recorded. Adherence to treatment was recorded by the therapeutic device and defined as the actual usage divided by the prescribed usage.
Logistic regression models were used to evaluate the association between treatment response and each baseline factor. Multivariable models were used to adjust for treatment adherence which may confound the relationship between the baseline factors and improvement in visual acuity and stereoacuity. We report adjusted odds ratios (OR) and 95% confidence intervals from these models. All statistical analysis was performed using R, version 4.2.0, with two-tailed tests and statistical significance defined as p<0.05.
Results:
This study included 121 children with a median age of 6.0 years (interquartile range [IQR], 5.0-7.0 years); 56 (46%) were female. Median adherence to the prescribed treatment regimen was 86% (IQR, 68% to 99%). Visual acuity in the amblyopic eye improved by one line in 35 (29%), two lines in 33 (27%) and at least three lines in 30 (25%) of children following 12 weeks of treatment. Median improvement in visual acuity was 2 lines (IQR, 1 to 2 lines; mean ± SD, 1.6 ± 1.4 lines). No baseline factor, including those associated with a poor response to patching (age, severity of amblyopia, and duration of prior treatment),5 was associated with improvement in visual acuity in models before and after adjusting for treatment adherence (Table 1). After 12 weeks of treatment, 47 (39%) children had any improvement in stereoacuity. The median improvement in stereoacuity was 0 log arcsec (IQR, 0 to 0.57 log arcsec; mean ± SD, 0.15 ± 0.65 log arcsec). No demographic, diagnostic, or treatment factor was associated with stereoacuity improvement (Table 2).
Table 1:
Factors associated with visual acuity improvement ≥1 line in the amblyopic eye following 12 weeks of binocular dichoptic therapy.
| Improvement in amblyopic eye visual acuity (≥1 line) |
Model 1: Unadjusted | Model 2: Adjusting for treatment adherence |
||||
|---|---|---|---|---|---|---|
| Yes, N = 98 | No, N = 23 | OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Age (years) 1 | 6.00 (5.00, 7.00) | 6.00 (5.00, 7.00) | 1.11 (0.85 to 1.52) | 0.47 | 1.09 (0.83 to 1.50) | 0.55 |
| Age group | ||||||
| 4 to ≤ 7 years | 84 (81%) | 20 (19%) | — | — | ||
| 8 to ≤ 12 years | 14 (82%) | 3 (18%) | 1.11 (0.32 to 5.15) | 0.88 | 1.00 (0.28 to 4.72) | >0.99 |
| Sex | ||||||
| Female | 48 (86%) | 8 (14%) | — | — | ||
| Male | 50 (77%) | 15 (23%) | 0.56 (0.21 to 1.40) | 0.22 | 0.58 (0.21 to 1.46) | 0.26 |
| Laterality | ||||||
| Right eye | 42 (86%) | 7 (14%) | — | — | ||
| Left eye | 54 (77%) | 16 (23%) | 0.56 (0.20 to 1.45) | 0.25 | 0.57 (0.20 to 1.47) | 0.26 |
| Strabismus 2 | ||||||
| Absent | 52 (83%) | 11 (17%) | — | — | ||
| Present | 46 (79%) | 12 (21%) | 0.81 (0.32 to 2.02) | 0.65 | 0.78 (0.31 to 1.95) | 0.59 |
| Prior treatment duration | ||||||
| None | 20 (83%) | 4 (17%) | — | — | ||
| ≤6 months | 31 (76%) | 10 (24%) | 0.62 (0.15 to 2.14) | 0.47 | 0.59 (0.14 to 2.04) | 0.42 |
| >6 months | 45 (85%) | 8 (15%) | 1.13 (0.27 to 4.02) | 0.86 | 1.08 (0.26 to 3.90) | 0.91 |
| Unknown | 2 | 1 | ||||
| Prior treatment type | ||||||
| None | 20 (83%) | 4 (17%) | — | — | ||
| Patching | 51 (81%) | 12 (19%) | 0.85 (0.22 to 2.78) | 0.80 | 0.79 (0.20 to 2.61) | 0.71 |
| Patching + Atropine | 16 (76%) | 5 (24%) | 0.64 (0.14 to 2.80) | 0.55 | 0.58 (0.12 to 2.58) | 0.47 |
| Unknown | 11 | 2 | ||||
| Baseline visual acuity of amblyopic eye | ||||||
| 20/40 or better | 39 (78%) | 11 (22%) | — | — | ||
| 20/50 to 20/200 | 59 (83%) | 12 (17%) | 1.39 (0.55 to 3.47) | 0.48 | 1.32 (0.52 to 3.33) | 0.55 |
| Baseline stereoacuity | ||||||
| ≤400 arcsec | 23 (74%) | 8 (26%) | — | — | ||
| >400 arcsec | 75 (83%) | 15 (17%) | 1.74 (0.63 to 4.55) | 0.27 | 1.80 (0.65 to 4.75) | 0.24 |
| Baseline IOD | ||||||
| <4 lines | 46 (79%) | 12 (21%) | — | — | ||
| ≥4 lines | 52 (83%) | 11 (17%) | 1.23 (0.49 to 3.10) | 0.65 | 1.17 (0.47 to 2.95) | 0.74 |
Median (IQR); N (%)
Strabismus was defined as heterotropia during baseline examination or history of strabismus surgery. All patients included in this study at 5 prism diopters or less of horizontal misalignment during the baseline exam.
OR = Odds Ratio, CI = Confidence Interval, IOD= Interocular difference
Table 2:
Factors associated with any stereoacuity improvement (>0 arcsec) following 12 weeks of binocular dichoptic therapy.
| Improvement in stereoacuity (>0 arcsec) |
Model 1: Unadjusted | Model 2: Adjusting for treatment adherence |
||||
|---|---|---|---|---|---|---|
| Yes, N = 47 | No, N = 74 | OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Age (years) 1 | 6.00 (5.00, 7.00) | 6.00 (5.00, 7.00) | 1.01 (0.81 to 1.26) | 0.90 | 1.00 (0.80 to 1.25) | 0.99 |
| Age group | ||||||
| 3 to ≤ 7 years | 41 (39%) | 63 (61%) | — | — | ||
| 8 to ≤ 12 years | 6 (35%) | 11 (65%) | 0.84 (0.27 to 2.38) | 0.75 | 0.77 (0.24 to 2.23) | 0.64 |
| Sex | ||||||
| Female | 27 (48%) | 29 (52%) | — | — | ||
| Male | 20 (31%) | 45 (69%) | 0.48 (0.22 to 1.00) | 0.051 | 0.50 (0.23 to 1.05) | 0.070 |
| Laterality | ||||||
| Right eye | 17 (35%) | 32 (65%) | — | — | ||
| Left eye | 28 (40%) | 42 (60%) | 1.25 (0.59 to 2.71) | 0.56 | 1.31 (0.61 to 2.87) | 0.49 |
| Strabismus 2 | ||||||
| Absent | 26 (41%) | 37 (59%) | — | — | ||
| Present | 21 (36%) | 37 (64%) | 0.81 (0.39 to 1.68) | 0.57 | 0.77 (0.36 to 1.64) | 0.50 |
| Prior treatment duration | ||||||
| None | 9 (38%) | 15 (62%) | — | — | ||
| ≤6 months | 19 (46%) | 22 (54%) | 1.44 (0.52 to 4.13) | 0.49 | 1.31 (0.47 to 3.81) | 0.61 |
| >6 months | 19 (36%) | 34 (64%) | 0.93 (0.35 to 2.59) | 0.89 | 0.83 (0.30 to 2.35) | 0.72 |
| Unknown | 0 | 3 | ||||
| Prior treatment type | ||||||
| None | 9 (38%) | 15 (62%) | — | — | ||
| Patching | 25 (40%) | 38 (60%) | 1.10 (0.42 to 2.97) | 0.85 | 1.04 (0.39 to 2.83) | 0.94 |
| Patching + Atropine | 6 (29%) | 15 (71%) | 0.67 (0.18 to 2.32) | 0.53 | 0.62 (0.17 to 2.18) | 0.46 |
| Unknown | 7 | 6 | ||||
| Baseline visual acuity of amblyopic eye | ||||||
| 20/40 or better | 23 (46%) | 27 (54%) | — | — | ||
| 20/50 to 20/200 | 24 (34%) | 47 (66%) | 0.60 (0.28 to 1.26) | 0.18 | 0.54 (0.25 to 1.15) | 0.11 |
| Baseline stereoacuity | ||||||
| ≤400 arcsec | 13 (42%) | 18 (58%) | — | — | ||
| >400 arcsec | 34 (38%) | 56 (62%) | 0.84 (0.37 to 1.96) | 0.68 | 0.89 (0.38 to 2.12) | 0.79 |
| Baseline IOD | ||||||
| <4 lines | 24 (41%) | 34 (59%) | — | — | ||
| ≥4 lines | 23 (37%) | 40 (63%) | 0.81 (0.39 to 1.69) | 0.58 | 0.74 (0.35 to 1.56) | 0.43 |
Median (IQR); N (%)
Strabismus was defined as heterotropia during baseline examination or history of strabismus surgery. All patients included in this study at 5 prism diopters or less of horizontal misalignment during the baseline exam.
OR = Odds Ratio, CI = Confidence Interval, IOD = Interocular difference
Discussion:
Visual acuity improved in 81% of children with amblyopia treated for 12 weeks with a dual-acting dichoptic therapeutic. The treatment benefit was observed across several clinical factors known to reduce efficacy of traditional amblyopia treatments, which suggests broad efficacy for this approach; however, currently available sample sizes may limit the power to detect clinically significant factors relevant to treatment response.
Amblyopia treatments, including patching and atropine penalization, are known to be less effective in older children.2 We did not find a difference in the improvement of visual acuity or stereoacuity following 12 weeks of treatment, based on age (range of 4 to 12 years), severity of amblyopia, or history of prior treatment. However, the cohort included relatively few children (14%) older than 7 years of age and did not include children with visual acuity worse than 20/200. We also did not find a difference in the treatment response based on additional demographic factors including sex, which suggests success with the digital therapeutic device across a range of patient populations.
Limitations of this study include the cohort size, which was limited to available trial data, potentially obscuring meaningful patient factor-outcome relationships. We did not analyze the magnitude of improvement relative to these factors because of the small sample. Also, all children included in the study received 12 weeks of treatment, so we cannot comment on the utility of additional weeks of treatment or durability of improvement following treatment cessation. Finally, the children included in this study had 5 prism diopters or less of horizontal misalignment and the results may not be generalizable to children with larger manifest deviations.
In conclusion, this cohort of all children enrolled in clinical trials using the Luminopia therapeutic to date did not demonstrate differences in visual acuity or stereoacuity improvement based on demographic patient factors, including those known to hinder responses to patching. Further investigation in larger cohorts is needed to identify factors predicting benefit from dual-acting dichoptic therapy that may not have been appreciated in this study.
Sources of funding:
Oke – Agency for Healthcare Research and Quality (T32HS000063)
Gaier – National eye institute (K08EY030164)
Footnotes
Conflict of interest:
E.D.G.: Scientific advisor and Equity owner Luminopia, Inc; Patent pending novel amblyopia treatment technology that does not directly relate to this work
M.X.R.: a faculty member at the Johns Hopkins University School of Medicine, he engaged in this research as a private consultant and not in his capacity as a Johns Hopkins faculty member and was compensated for the consulting service in income
Presentation: This work was presented at the 2023 AAPOS Conference in New York.
References:
- 1.Xiao S. et al. Improved adherence and treatment outcomes with an engaging, personalized digital therapeutic in amblyopia. Sci Rep 10, 8328 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Holmes JM et al. Effect of age on response to amblyopia treatment in children. Arch Ophthalmol 129, 1451–1457 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Xiao S. et al. Digital therapeutic improves visual acuity and encourages high adherence in amblyopic children in open-label pilot study. J AAPOS 25, 87.e1–87.e6 (2021). [DOI] [PubMed] [Google Scholar]
- 4.Xiao S. et al. Randomized Controlled Trial of a Dichoptic Digital Therapeutic for Amblyopia. Ophthalmology 129, 77–85 (2022). [DOI] [PubMed] [Google Scholar]
- 5.Jost RM et al. Randomized clinical trial of streaming dichoptic movies versus patching for treatment of amblyopia in children aged 3 to 7 years. Sci Rep 12, 4157 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]