Randomized Clinical Trial of Diffusion Optics Technology Spectacle Lenses in a Chinese Population (CATHAY): 12-Month Results

Abstract

Objective

To evaluate 12-month myopia control efficacy of diffusion optics technology (DOT) spectacle lenses in a Chinese population (CATHAY study).

Design

Interim analysis of 6-month and 12-month data from a 24-month, randomized controlled clinical trial across 5 Chinese hospital sites, evaluating DOT spectacle lenses versus control standard single-vision (SV) spectacle lenses.

Participants

One hundred ninety-five myopic children aged 6-13 years were enrolled across 5 sites in China. Participants were dispensed DOT lenses in a 2:1 ratio versus SV.

Methods

Participants were randomized to DOT or SV spectacles. Axial length (AL) and cycloplegic spherical equivalent refraction (cSER) were measured at baseline, 6 months, and 12 months.

Main Outcome Measures

Twelve-month changes in AL and cSER from baseline.

Results

One hundred eighty-six children (DOT 128, SV 58) completed the 12-month visit (modified intent-to-treat cohort) with a mean age at prerandomization eligibility visit (baseline) of 9.6 ± 1.8 years. No severe adverse events were observed in either group. For DOT wearers, the mean observed change ± standard deviation in AL was 0.09 ± 0.22 (95% confidence interval [CI]: 0.05 to 0.13), and cSER change was –0.16 ± 0.51 D (95% CI: –0.25 to –0.07). For the SV group, mean observed change from baseline at 12 months for AL was 0.35 ± 0.19 mm (95% CI: 0.30–0.40), and cSER change was –0.64 ± 0.42 D (95% CI: –0.76 to –0.53). The least squares mean differences (±standard error) in AL elongation and cSER progression were significantly lower for DOT vs. SV at 12 months. The difference in AL change between DOT and SV was –0.26 ± 0.03 mm (P < 0.0001), and the difference in cSER progression for DOT versus SV was 0.48 ± 0.08 D (P < 0.0001).

Conclusions

The 12-month interim results from this ongoing trial demonstrate the effectiveness of DOT spectacles for reducing myopic progression and axial elongation in Chinese children. A lens design based on the modulation of contrast on the retina works favorably for young Chinese children.

Financial Disclosure(s)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

The prevalence of myopia is growing worldwide, with an estimated 22.9% of the world's population currently myopic, a number predicted to rise to close to 50% by 2050.¹ In some East Asian countries and regions, myopia has reached epidemic proportions, with prevalence values of 65.5% to 87.8% reported in urban Chinese teens and over 81.3% of Koreans in their 20s.1, 2, 3, 4, 5, 6 The rates of myopic progression are also higher for East Asian groups versus other ethnic groups (White/European/American), even when ethnic groups are studied outside of Asia.^6,7 These factors have created urgency to develop and test new designs and treatment modalities. Several approaches to control the progression of myopia have been studied, including spectacle lenses, rigid and soft contact lenses, pharmacological interventions such as atropine, and, more recently, red light therapy.^8,9 However, perhaps due to differences in costs or regulatory pathways, the vast majority of randomized clinical trials on new devices for control of myopia progression have been conducted in countries and regions in East Asia with limited ethnic diversity (mainland China, Hong Kong, Taiwan, etc.).8, 9, 10 Two exceptions are the early clinical trials conducted in predominantly North American and European White populations: the study of a dual-focus contact lens developed for myopia control, and another for a spectacle lens for myopia control featuring diffusion optics technology (DOT).11, 12, 13, 14 Single-site studies or those with very ethnically homogeneous populations naturally raise questions about how generalizable results would be across ethnic groups not studied.

There is also a question as to whether different mechanisms and designs would have similar impacts across more diverse populations. The dual-focus contact lens design is based on the defocus theory, where the optics create myopic defocus in the peripheral retina.¹³ The dual-focus contact lens study included 1 site in Singapore in addition to the North American and European sites, and the authors tested for interaction between efficacy and investigational site. They found no interaction, suggesting that the Singaporean Chinese cohort did not behave appreciably differently from the predominantly White populations at the other sites.¹³ The clinical trials of contemporary myopia-control spectacle lenses with designs based on the peripheral defocus theory were first conducted with ethnically Chinese children in mainland China and Hong Kong.15, 16, 17 Recently, data have become available for these designs used in White participants from European countries.¹⁸

The contrast theory of myopia is a relatively recent concept from research into polymorphisms in the long (L) and middle (M) wavelength-sensitive cone opsin genes associated with myopia. These genetic variations cause the retina to signal high contrast where none actually exists.¹⁹ According to the theory, the combination of genetically encoded spurious contrast and exposure to high-contrast stimuli in our modern environment is responsible for myopia.¹⁹ Reducing the formation of high-contrast images from environmental myopia genetic stimuli, such as screens and reading material, is hypothesized to decrease contrast burden and potentially slow the progression of myopia.¹⁹ A proprietary pattern of thousands of microscopic diffusers was applied across the surface of an ophthalmic lens in the DOT lens design. One 4-year prospective randomized clinical trial using this design demonstrated significant efficacy in a North American population (Control of Myopia Using Peripheral Diffusion Lenses Efficacy and Safety Study [CYPRESS study]), though fewer than 10% of participants reported any Asian ethnicity.¹²

The current study (the “CATHAY” study) was designed to evaluate the safety and efficacy of DOT spectacle lenses in Chinese children compared to single-vision (SV) spectacle lenses over 2 years. This report summarizes 6-month and 12-month results from a planned 12-month interim analysis of the CATHAY study.

Methods

Study Design

The CATHAY study was a randomized, controlled, multicenter, observer-masked, 2-arm parallel group clinical trial lasting 24 months, modeled after the original CYPRESS study.^11,12 Interim analyses were planned at 6, 12, and 18 months.

Institutional review board/ethics committee approval was secured from the Ethics Committee of Fudan University Eye and ENT Hospital, Shanghai, China (No. 2022062), and the clinical trial was conducted in accordance with CONSORT guidelines and the Declaration of Helsinki. Informed assent from participants and written informed consent from parents were obtained for all participants. The clinical trial (NCT05562622) was registered at www.clinicaltrials.gov on September 28, 2022.

Sample Size

The targeted enrollment was 195 participants (130 test, 65 control) across 5 ophthalmology hospitals in China. The sample size was calculated to ensure sufficient power at an alpha level of 0.05 for treatment comparisons in axial length (AL) change from baseline at 12 months, assuming a 2:1 randomization ratio, a one-sided type I error of 0.025, 80% statistical power, a mean change in AL of 0.324 mm (standard deviation [SD] 0.35 mm) for the test group, and 0.5 mm (SD 0.35 mm) for the control group.²⁰ The required sample size at 12 months was 144 (96 in the test group, 48 in the control group). For the secondary cycloplegic spherical equivalent refraction (cSER), the expected mean change in the test arm was –0.65 D (SD 0.6 D) and –1.00 D (SD 0.6 D) in the control arm. To account for potential 35% attrition due to COVID-19 and other factors, the final sample size was increased to 195.¹²

Participants

Myopic children aged 6 to 13 years at baseline with refractive sphere between –0.75 D and –5.25 D (cSER by cycloplegic autorefraction), astigmatism less than or equal to 1.50 D, and anisometropia less than or equal to 1.00 D were enrolled.²¹ Participants were required to meet the age range and refractive criteria in each eye, have best-corrected visual acuity by manifest refraction of 0.10 logarithm of the minimum angle of resolution (6/7.5, 0.8 Snellen) or better in each eye, and agree to wear assigned study spectacles for ≥10 hours per day, at all times except when sleeping, swimming, or engaging in activities where the use of spectacles would be dangerous or otherwise not possible. They also needed to comply with all study procedures, be willing to participate in the full 24-month trial and sign relevant assent and parental consent forms. Participants were excluded if they had ever used atropine, contact lenses, or spectacle lenses for myopia control, failed to meet refractive criteria, or had amblyopia in either eye or strabismus by cover test at far (4 m) or near (40 cm) while wearing best correction. Those with any other ocular or systemic conditions that could influence refractive development or status, had a known allergy to proparacaine, tetracaine, cyclopentolate, or had participated in any investigational clinical study within 30 days prior to the screening visit were also excluded.

Randomization and Masking

After the screening visit, eligible participants were randomly assigned to the test (DOT) or control (SV) group in a 2:1 ratio using a permuted block design to assign spectacle lenses. The randomization sequence was generated by Microsoft Excel and stratified by age and degree of myopia within each site. ClinFlash Electronic Data Capture was used to conceal randomization and interventions. The DOT and SV lenses were identical except for the placement of the DOT design on the test lenses. The baseline visit was conducted 7 to 28 days after the screening visit and randomization, allowing time for the study spectacles to be received. Study visits over the full 24-month study were scheduled at 6, 12, 18, and 24 months ± 21 days following the baseline visit. All lenses were replaced every 6 months, even if a prescription update was not required, to maintain surface quality and frame integrity. Trained study staff technicians who measured AL and cSER were masked to treatment assignment, and participants were instructed to keep their glasses out of sight and not discuss their glasses with the staff taking the measurements.

Procedures

Evaluations of cSER and AL took place at baseline, month 6, and month 12. Participants received 1 drop of 0.5% proparacaine or tetracaine in each eye, followed by 2 drops of 1.0% cyclopentolate in each eye, 5 minutes apart, to achieve adequate cycloplegia before autorefraction. The cSER was measured 25 min after the final drop of cyclopentolate was instilled using open-field auto refractometers: WAM-5500 Binocular Auto Refractometer or WR-5100K/Shin-Nippon NVision-K 5001 (Grand Seiko Co, Ltd). Measurements of AL were taken using the IOLMaster 700 or 500 (Zeiss). Parents completed questionnaires to monitor compliance and the visual impact of the study spectacles. Adverse events were assessed at each study visit.

Outcome Measures

The main outcome measures of the current analysis included changes from baseline for AL and cSER at 12 months for DOT and SV groups. Twelve-month safety endpoints included adverse events, device deficiencies, best-corrected visual acuity, and symptoms, problems, and complaints.

Statistical Analysis

Statistical analyses were conducted using SAS version 9.4. Data from both eyes of each participant were averaged for analysis. Poolability analyses across the 5 sites were performed for age at baseline, age group distribution, baseline AL, baseline cSER, sex, change in AL, and change in cSER at 6-month and 12-month visits. Descriptive summaries included mean, standard deviation, median, and range for continuous variables, and counts and percentages for categorical variables. Two-sided 95% confidence intervals were calculated for means and percentages, as this represented a more rigorous criterion than the one-sided intervals used for sample size calculations. For key outcome measures, differences between the test and control groups were analyzed, and two-sided 95% confidence intervals were computed. To evaluate the treatment effect over time, a linear mixed-effects model for repeated measures was fitted to the data.^22,23 Diagnostics were assessed to ensure the validity of the model, including assessments of normality, homoscedasticity, and independence of residuals. An unstructured covariance matrix was used to model within-participant correlations and variability across time points. Sex, baseline age, treatment, visit, and the interaction of treatment and visit were included as fixed effects; the baseline value of the outcome variable was included as a covariate; participant was included as a participant-level effect; and site was included as a random effect. Significance was based on least squares means with P < 0.05 for each endpoint. Observed data are reported as mean ± SD. Least squares mean estimates derived from the linear mixed model used to account for repeated measures are reported as mean ± standard error. Safety data were listed or summarized descriptively. Additional subgroup analyses for change in AL and cSER were conducted by age group at study entry to determine if there were any early indications of differences in efficacy by age. Data were stratified into the age groups 6 to 7, 8 to 10, and ≥11 (Tables S1–S6, available at www.ophthalmologyscience.org).

Results

Study Population

A total of 215 potential participants were screened beginning in February 2023, with 195 enrolled. Of these, 130 were randomized to DOT, 65 to SV. One was randomized but did not receive the SV control lens due to withdrawal of consent. Enrollment was completed in August 2023, with 6-month visits occurring between August 2023 and March 2024 and 12-month visits between February and August 2024. The 2-year full study was scheduled to conclude in September 2025. Study flow and participant status are summarized by lens type in Table 1 and Figure 1. The modified intent-to-treat analysis included 186 participants (128 DOT, 58 SV) for whom complete AL data were available through the 12-month visit. Participants ranged in age from 6 to 13 years at baseline and were all of Chinese ethnicity. The modified intent-to-treat cohort was 48.4% (90/186) male, and 51.6% (96/186) female, and 19% (28/186) were in the 6 to under 8-year age group. Data for cSER were missing in 5 participants due to lack of participant cooperation or an equipment error. Two participants with missing data were in the DOT group (participant ages 8 and 13) and 3 were in the SV group (1 participant age 7 and 2 partcipants age 10).

Table 1.

Demographic and Clinical Characteristics (mITT Population) of Participants Who Completed 12-Mo Visit

Variable	Total	DOT Group (Test)	SV Group (Control)
Number of subjects	186	128 (69%)	58 (31%)
Integer age∗, n (%)
6 yrs	7 (3.8)	5 (3.9)	2 (3.5)
7 yrs	20 (10.8)	12 (9.4)	8 (13.8)
8 yrs	23 (12.4)	18 (14.1)	5 (8.6)
9 yrs	34 (18.3)	26 (20.3)	8 (13.8)
10 yrs	44 (23.7)	28 (21.9)	16 (27.6)
11 yrs	27 (14.5)	19 (14.8)	8 (13.8)
12 yrs	22 (11.8)	13 (10.2)	9 (15.5)
13 yrs	9 (4.8)	7 (5.5)	2 (3.5)
Mean (SD)	9.62 (1.78)	9.59 (1.77)	9.69 (1.82)
Sex, n (%)
Male	90 (48.4)	64 (50.0)	26 (44.8)
Female	96 (51.6)	64 (50.0)	32 (55.2)
Myopic parents†, n (%)
Two	64 (34.4)	51 (39.8)	13 (22.4)
At least one	6 (3.2)	2 (1.6)	4 (6.9)
One	74 (39.8)	46 (35.9)	28 (48.3)
Zero	35 (18.8)	24 (18.8)	11 (19.0)
Unknown	7 (3.8)	5 (3.9)	2 (3.4)
Axial length (mm)
Mean	24.45	24.49	24.37
SD	0.80	0.78	0.82
Min	22.72	22.83	22.72
Max	26.61	26.61	26.37
cSER (D)
Mean	–2.28	–2.36	–2.12
SD	0.85	0.86	0.79
Min	–4.92	–4.92	–4.51
Max	–1.10	–1.10	–1.13
Missing‡	5	2	3

cSER = cycloplegic spherical equivalent refraction; D = diopter; DOT = diffusion optics technology; mITT = modified intent to treat; SD = standard deviation; SV = single vision.

^∗Integer age is the subject's age rounded down to the nearest whole year at baseline.

^†Subjects/parents were given the options: unknown, 0, 1 (1 parent myopic and the other not), at least 1 (1 parent myopic and the other unknown), or 2.

^‡Missing data due to poor subject cooperation or equipment error.

Figure 1.

Study flow detailing numbers screened, randomized, dispensed study lenses, and discontinued and attending study visits. DOT = diffusion optics technology; mITT = modified intent-to-treat; SV = single vision.

Discontinuations totaled 8 (4.1%, 8/195) over 12 months, including 2 (1.5%, 2/130) in the DOT group: one withdrawal of consent at baseline with no AL or cSER data collected, and one between the 6-month and 12-month visits due to long-standing, persistent strabismus not detected at enrollment but recorded at later study visits. This was considered an adverse event, and the investigator discontinued the participant. Discontinuations in the SV group totaled 6 (9.2%, 6/65): 3 participants were discontinued at baseline with no AL or cSER data collected, and 3 more prior to the 6-month visit due to withdrawal of consent because of rapid myopia progression and a desire to explore alternative treatment options. These participants were excluded from the modified intent-to-treat analysis because no data were available. One discontinuation in the DOT group was associated with an adverse event (exotropia diagnosis previously described), and no discontinuations were related to a device deficiency in either group.

Efficacy

Table 2 shows the unadjusted observed means (±SD) for change from baseline at 6 and 12 months, along with the least squares mean estimate of the mean difference between the DOT and SV groups. Significantly less axial elongation (Fig 2A) and progression of cSER (Fig 2B) is observed for the DOT cohort (both P < 0.0001). The SV rates of change are approximately linear over the 6-month and 12-month visits. However, there is a visibly different slope between baseline and 6 months and between the 6-month and 12-month interval for both AL and cSER.

Table 2.

Change from Baseline for Axial Length and Cycloplegic Spherical Equivalent Refraction: Observed and LSM Estimate of Difference (DOT-SV)

	DOT Observed (±SD)	SV Observed (±SD)	LSM Difference Estimate (±SE)	Significance (DF, t, P)
Axial length – change from baseline (mm)
Visit	n = 128	n = 58
6 mos	0.01 ± 0.16	0.18 ± 0.12	–0.171 ± 0.023	181, t = –7.45, <0.0001
12 mos	0.09 ± 0.22	0.35 ± 0.19	–0.26 ± 0.03	181, t = –7.94, <0.0001
cSER – change from baseline (D)
Visit	n = 126	n = 56
6 mos	0.04 ± 0.41	–0.31 ± 0.35	0.36 ± 0.06	179, t = 5.82, <0.0001
12 mos	–0.17 ± 0.51	–0.64 ± 0.42	0.48 ± 0.08	179, t = 6.33, <0.0001

cSER = cycloplegic spherical equivalent refraction; D = diopter; DF = dual focus; DOT = diffusion optics technology; LSM = least squares mean; SD = standard deviation; SE = standard error; SV = single vision.

Figure 2.

Change in (A) AL in mm and (B) cSER in diopters at 6 and 12 months. The change from baseline between DOT and SV was significant at 6-month and 12-month visits (P < 0.001). Error bars are standard error; ∗ = significant difference for change from baseline between DOT and SV, P < 0.001. AL = axial length; cSER = cycloplegic spherical equivalent refraction; D = diopter; DOT = diffusion optics technology; SE = standard error; SV = single vision.

Data by age group (6–7 years, 8–10 years, and ≥ 11 years) were analyzed for both AL and cSER, as age has been identified as a factor in the speed of progression.²⁴ Figure 3 shows changes from baseline comparisons for DOT versus SV by age group for AL (3A) and cSER (3B), respectively, for the 6-month and 12-month visits. Greater variability was observed in the younger cohorts, particularly for AL, where several outliers were present, more noticeably in the DOT group, with 2 participants progressing at a much higher rate. When change from baseline data are plotted by visit (Fig 4), differences in the rates of change between visits and age groups become more obvious. The DOT lenses resulted in significantly less axial elongation than SV lenses at both 6-month and 12-month visits across all age groups (all P < 0.05). For cSER, significant differences were only confirmed for the 8 to 10-year and ≥11-year analyses (Tables S1–S6).

Figure 3.

Box-and-whisker plot by age group for least squares mean estimate of change from baseline at 12-month visit for (A) AL and (B) cSER showing mean (x), median (line), and interquartile range (box); † = significant difference between DOT and SV, P < 0.05. AL = axial length; cSER = cycloplegic spherical equivalent refraction; DOT = diffusion optics technology; SV = single vision.

Figure 4.

Rate of progression of (A) AL and (B) cSER between 6-month and 12-month visits by age group (Y = years). Dotted lines represent data for the DOT lens cohort, and solid lines data for SV lens cohort. AL = axial length; cSER = cycloplegic spherical equivalent refraction; D = diopter; DOT = diffusion optics technology; SV = single vision.

Safety

Participants tolerated the lenses well throughout the study, with average daily wear times of over 11 hours per day, 7 days per week, as per protocol, for both groups. Reportable events totaled 184 across 558 study visits (33.0%), involving 70 participants (37.6%). Additional details are in Table 3. None of these were considered serious (vision-threatening). There were 8 nonocular events classified as significant: 7 in 2 participants in the DOT group: fatigue (1), dizziness (3), flu (1), headache (1), and 1 event (virus) in the SV group. Of the 70 ocular events classified as potentially device-related, 54 (14.1%) occurred in the DOT group and 16 (9.2%) in the SV group. The events classified as possibly device-related included blurred vision, halos, glare, and exophoria, but none were classified as significant. Binocular high-contrast distance visual acuity was similar between the DOT and SV groups across visits and reflected changes associated with progressing myopia (Table 4).

Table 3.

Adverse Events (mITT Population)

	Total n = 186	Test n = 128	Control n = 58
Subjects with any event, n (%)	70 (37.6)	51 (39.8)	19 (32.8)
Subjects with any ocular event, n(%)	58 (31.2)	44 (34.4)	14 (24.1)
Total events across all visits
All event visits (n × 3 visits)	558	384	174
Any event, n (%)	184 (33.0)	142 (37.0)	42 (24.1)
Ocular AEs, n (%)	104 (18.6)	81 (21.1)	23 (13.2)
Blurred vision	29 (5.2)	22 (5.7)	7 (4.0)
Halo	25 (4.5)	19 (4.9)	6 (3.4)
Glare	22 (3.9)	16 (4.2)	6 (3.4)
Dizziness	16 (2.9)	14 (3.6)	2 (1.1)
Headache	2 (0.4)	2 (0.5)	0 (0.0)
Itchy eyes	2 (0.4)	1 (0.3)	1 (0.6)
Pink eye	1 (0.3)	1 (0.3)	0 (0.0)
Dry eye	3 (0.5)	3 (0.8)	0 (0.0)
Exophoria	1 (0.2)	1 (0.3)	0 (0.0)
Exotropia	3 (0.5)	2 (0.5)	1 (0.6)
Nonocular, nondevice-related AEs, n (%)	80 (14.3)	61 (15.9)	19 (10.9)
Ocular events – possibly related to study device, n, (%)
Events	70 (12.5)	54 (14.1)	16 (9.2)
Subjects	38 (20.4)	29 (22.7)	9 (15.5)

AE = adverse event; mITT = modified intent-to-treat.

Table 4.

Summary of logMAR Visual Acuity Binocular High-Contrast Visual Acuity through Study Spectacle Lenses

Visit	Lens	N	Mean ± SD	Range	Change from Baseline
Dispensing	DOT	129	–0.05 ± 0.08	0.1 to –0.3	N/A
	SV	63	–0.04 ± 0.07	0.2 to –0.2	N/A
6 mos	DOT	128	–0.06 ± 0.09	0.3 to –0.3	–0.01 ± 0.10
	SV	59	–0.02 ± 0.07	0.2 to –0.2	0.02 ± 0.09
12 mos	DOT	126	–0.04 ± 0.08	0.2 to –0.18	0.01 ± 0.10
	SV	57	0.00 ± 0.09	0.4 to –0.2	0.04 ± 0.13

DOT = diffusion optics technology; logMAR = logarithm of the minimum angle of resolution; SD = standard deviation; SV = single vision.

Discussion

The results of this interim analysis showed that the DOT spectacle lenses slowed myopia progression and axial elongation over 12 months in a group of Chinese children. Significant reductions in change from baseline were observed for both AL and cSER compared to the SV control spectacle condition at the 6-month and 12-month visits. When stratified by age, similar results were observed, although no significant difference for cSER change was found in the 6-7-year-old group. This group size was very small, and included significant outliers for both AL and cSER, as well as missing cSER data for 1 participant. Although the sample size did not permit between-group analyses by age, the expected trend toward faster progression in younger age groups was evident. The large standard deviations observed in this study are consistent with both historical and contemporary studies of myopia progression.^7,13 Therefore, while differences between groups were statistically significant, the large individual variability seen in the observed data is a reminder that individual results may vary. Figure 3 illustrates the substantial overlap among groups in cSER interquartile ranges and error bars. Greater separation is observed for the AL interquartile ranges, although significant overlap remains between the error bars.

The retention rate and wearing time data for both groups throughout the first year suggest that the DOT lens design was well tolerated by these young participants. Due to the visibility of the design elements of the DOT lenses, participants were not masked to treatment. Other published reports confirm that children adapted well to the DOT spectacle lenses.²⁵

When visually comparing the data from this study with other studies involving spectacle lens designs for myopia control in ethnically Chinese participants, there is good consistency in the 12-month change data in the SV control group for both AL and cSER (Fig 5). The mean changes from baseline ranged from 0.32 to 0.36 mm for AL and from –0.55 to –0.81 D for cSER in the SV control group. This consistency was observed, despite differences in the timing of the studies relative to the COVID-19 pandemic, site location, single or multisite designs, or refractive error range at entry.26, 27, 28 The current study enrolled children as young as 6 years old, similar to the study of cylindrical annular refractive elements, whereas studies of defocus incorporated multiple segments and highly aspheric lenslet technologies began recruitment at 8 years of age. Fortunately, these 3 clinical trials were initiated and completed their 1-year data analyses outside the peak COVID-19 years, thereby eliminating a major confounding factor.²⁹ The current study's interim analysis showed trends previously observed in other studies: a change in the rate of progression between the 6-month and 12-month visits in the treatment group,26, 27, 28 which manifests as a “jog” in the curve, where the slope increases (axial elongation and/or myopic refractive error accelerates in the treatment condition; see Figs 2 and 4).

Figure 5.

Comparison of 12-month changes in (A) AL and (B) cSER across spectacle lens studies conducted in Asia. Data reported as mean ± SE in mm for AL and diopters for spherical equivalent refraction. AL = axial length; CARE = cylindrical annular refractive elements (dark blue); cSER = cycloplegic spherical equivalent refraction; DOT = diffusion optics technology (green); DIMS = defocus incorporated multiple segments (light blue); HAL = highly aspherical lenslets (yellow); SE = standard error; SV = single vision (orange).

While systematic literature reviews suggest that year-1 efficacy results can be doubled to accurately estimate 3-year efficacy, without a complete rigorous analysis of the primary and secondary endpoints, it may not be prudent to predict such a result.³⁰ As seen with nearly all well-controlled clinical trials, the treatment effects are not linear, and myopia progression tends to slow with age, even without clinical intervention. The enrollment of younger participants is a potential confounding factor when comparing AL change with rates reported in other studies that evaluated older cohorts.31, 32, 33 The loss of 3 participants in the SV group due to rapid myopia progression who chose to seek alternative treatment options could introduce bias, especially given the 2:1 randomization scheme that amplifies the impact of losses from the SV control group. Additionally, questions arise about continuing the control group using SV lenses, given the evidence that many new spectacle and contact lens designs can significantly impact axial elongation and refractive progression.

While most reportable adverse events in this study were mild, the prevalence of vision-related symptoms and complaints raises concerns about the tolerability of the DOT lenses in certain patients and their impact on ongoing adherence to wearing schedules, particularly as the children enter their teenage years. Reports of glare, hazy vision, and haloes could be driven by the design features of the DOT lenses, though these symptoms were reported in both DOT and SV groups. The more rapid progression of myopia in the SV group makes it difficult to separate the optical effects of uncorrected myopia from the impact of the light scattering effect of the design elements of the DOT lenses. The complaints appeared to be patient-specific in both DOT and SV groups. However, as other studies have noted, diminished efficacy due to noncompliance with prescribed wear schedules, practitioner awareness of issues with tolerability is important.^26,34 Future analyses should include assessment of the impact of lens wear compliance and adherence on efficacy.

Although the study designs of the CYPRESS and CATHAY studies were similar, this interim analysis was not intended to provide an in-depth comparison of device performance between a Western White population and an East Asian population. Additionally, the CYPRESS study straddled the peak of the COVID-19 pandemic, which introduced many confounding factors, and limited enrollment to children aged 6 to 10 years. A visual review of the 1-year data from the 2 interim analyses (Fig 6) suggests a higher rate of axial elongation in the SV group in the CATHAY compared to the CYPRESS study. This difference could exaggerate the relative treatment effect, but whether it reflects a higher baseline rate of progression in the CATHAY cohort or an actual biological variation cannot be determined with the current data. In addition, even limiting the comparison to the same age group, the current study had different refractive criteria (spherical equivalent refraction and astigmatism), and the sample size for this subgroup is slightly smaller than required for good statistical power.

Figure 6.

Mean change from baseline (± standard deviation) at 12 months for (A) AL in millimeters (mm) and (B) cSER in D for SV controls and DOT test subjects for the current study in China and the CYPRESS study run in the United States for an age-matched sample (aged 6 to 10 years at study entry). Orange bars = SV control; green = DOT test lenses; AL = axial length; cSER = cycloplegic spherical equivalent refraction; D = diopter; DOT = diffusion optics technology; SV = single vision.

Notably, the consistency in data from several studies of myopia control spectacle designs may support contrast theory as a possible mechanism behind the success of peripheral defocus spectacle lens designs. While these designs incorporate plus-powered optics in the periphery of the lens to pull out-of-focus peripheral rays from behind the retina to in front, the amount of plus power in the defocus theory concepts often exceeds that required to compensate for peripheral hyperopia. It is therefore possible that lowering the contrast on the retina with the out-of-focus image drives the effect rather than the fact that the peripheral defocus has moved in front of the retina. More research is needed to validate each theory and to fully understand the mechanisms of action of the array of designs for myopia control.

In conclusion, the current study demonstrated that spectacle lenses with DOT effectively slow axial elongation and myopic progression in Chinese children. The multicenter design of the trial also increases the likelihood that the data will be more generalizable across the region. While speculation about the mechanism of action and its merits versus peripheral defocus designs will continue, this contrast-modulation design worked well for young Chinese children, supporting the hypothesis that exposure to high-contrast environments may underlie the fast myopia progression in Chinese children, as well as in North American children.

Manuscript no. XOPS-D-25-00719.