Spectacle Lenses With Aspherical Lenslets for Myopia Control vs Single-Vision Spectacle Lenses

This randomized clinical trial aimed to evaluate whether spectacle lenses with aspherical lenslets slow myopia progression over 2 years and whether the level of lenslet asphericity affects myopia control efficacy in a dose-dependent manner.

The analysis will be presented in a report, which will be used as the basis of the 91 primary research publications according to the study publication plan. This SAP 92 describes the statistical methods for the primary, secondary, and exploratory 93 outcomes of the study as defined in the protocol, as well as additional subgroup 94 analysis. 95 96 97 4.1 Preface 98 Currently, myopia is managed mainly by inducing hyperopic defocus by either using 99 contact lenses like orthokeratology and multifocal soft contact lenses and 100 spectacles lenses like progressive lenses, bifocals, and peripheral defocus lenses. 101 Other methods like atropine are also used to control myopia. The methods 102 mentioned above have various efficacy in controlling myopia, with low dose 103 atropine being the most effective at 59%, followed by orthokeratology with about 104 43% of myopia control and multifocal soft contact lenses at 49% of myopia control. 105 These methods are invasive and may not be suitable for all young children. 106 Moreover, the rebound effect for atropine was not desirable upon cessation of drug 107 use. For spectacle correction method, bifocals have the best outcome of about 45% 108 myopia control. However, with more studies with animal to understand the effect of 109 optical effect has on myopia progression, better myopia can be achieved with a new 110 spectacle lens design (Wildsoet et al., 2019). 111 112 It has been shown that a persistent myopic defocus constantly applied on the whole 113 retina has a positive effect on myopia progression, i.e., decrease in myopia 114 progression (Anstice and Phillips, 2011). Based on the above background and 115 outcome of several studies on optical treatment on myopia progression, we 116 developed 2 FIN spectacle lenses described in Section 6.1.1 of the Protocol. 117 118 These analyses will assess the efficacy and safety of SAL and HAL compared to 119

Purpose of the analyses
Single Vision Spectacle Lenses (SVL) and will be included in the clinical study report. 120 SAP The study aims to evaluate the efficacy of spectacle lenses with aspherical lenslets  124  to reduce the progression of myopia by either reducing the myopia progression rate  125 per year and/or reducing the elongation of eyeball through myopic defocus 126 compared with SVL group. 2 embodiments of the spectacle lenses with aspherical 127 lenslets will be tested and compared with single vision spectacles lenses (SVLs) as a 128 control. 150 healthy children aged 8 to 13 years old will be recruited in a double-129 masked randomized clinical trial over a wearing period of at least two years. 130 Cycloplegic autorefraction and axial length will be the primary measure for myopia 131 progression. Other measures like peripheral autorefraction will be done to evaluate 132 the effect of myopic defocus on peripheral retina. Finally, visual acuity and contrast 133 sensitivity will also be compared between spectacle lenses with aspherical lenslets 134 and SVL to quantify the quality of vision using myopia control lenses like FIN. 135 0.75D to -4.75D (spherical equivalent) and meeting the specific inclusion and 142 exclusion criteria will be recruited. Consent will be taken from guardians and 143 children who are eligible for the study, and they will be randomized using Randola 144

Endpoints
to wear either of the SAL, HAL (test arm), single vision lenses (control arm). All 145 children will be followed up 6 monthly for 2 years to monitor changes in cycloplegic 146 autorefraction and axial length. At the end of the study, the above measures will be 147 compared between test and control arm for objective and subjective performance. 148 Each subject will be followed for at least 2 years and 3 weeks. 149 The ophthalmic lenses will be worn by subjects and allocated according to a 150 randomization ratio of 1:1:1 in each arm (150/3 in each arm). 151 The study will be conducted at the  Subjects will be included in the study if they satisfy the following criteria: 174 175

Inclusion-Exclusion Criteria and General Study Population
General inclusion criterion: 176 177 -Volunteer subject and guardian, fluent Chinese spoken, willing to follow the 178 protocol, and able to read, comprehend and sign the informed consent form. 179 180 Study related inclusion criteria: 181 182 • Age: equal to or greater than 8 years and not older than 13 years. 183 • Spherical refractive error of -0.75 to -4.75 D in each eye (spherical 184 equivalent), as measured by cycloplegic autorefraction. 185 • Astigmatism of not more than 1.50 D. 186 • Anisometropia of not more than 1.00 D. 187 • Best-corrected visual acuity of equal or better than 0.05 LogMAR (≥ 0.9 188 as Snellen) 189 • No strabismus by cover test at near and distance. 190 • Have the ability to comply with the protocol to get the reliable study 191 measurements. 192 •  198 • No history of PALs or bifocals use and no prior use of contact lenses or 199 any treatment for myopia control. 200 201

Exclusion Criteria 202
Subjects presenting with any of the following exclusion criteria will not be included 203 in the study: 204 205 General exclusion criteria: 206 207 -Vulnerability of the subject, 208 -Participation in another study which might have an influence on vision or 209 interfere with study assessments, 210 211 Study related exclusion criteria: 212 213 • Age: less than 8 years old or greater than 13 years. 214 • Spherical refractive error: less than -0.75 D or greater than -4.75 D in each 215 eye (spherical equivalent), measured by cycloplegic autorefraction. 216 • Astigmatism of more than 1.50 D. 217 • Anisometropia of more than 1.00 D. 218 • Best-corrected visual acuity of less than 0.05 LogMAR (≤ 0.8 as Snellen) 219 • With strabismus by cover test at near or distance. 220 • Without the ability to comply with the protocol to get the reliable study 221 measurements 222 • Presence of any ocular disease that would influence refractive development, 223 such as retinal disease, cataract, and ptosis. Presence of systemic or 224 neurodevelopmental conditions that may influence refractive development. 225 Use of ocular or systemic medicine, which might affect myopia progression 226 or visual acuity through known effects on retina, accommodation, or 227 significant elevation of intraocular pressure. 228 • Prior use of progressive adaptive lenses, bifocals use, use of contact lenses, 229 or any treatment for myopia control. 230

231
Scheduled randomization will be generated by the Study Manager in charge of lens 232 logistics using online application Randola and forwarded only to the person in 233 charge of fabrication in Essilor (France or Singapore). The Investigators will never 234 have access to the randomization list. Randomization is based on spherical 235 equivalent of cycloplegic autorefraction in right eye, age, and gender. 236 237 Each time a subject becomes eligible for the study, he/she will be randomly 238 assigned to an arm in a ratio of 1:1:1, then the Investigator will receive the study 239 device from the person in charge of fabrication accordingly. 240 241 Once the database has been locked, and populations for analysis agreed, all codes 242 (subject identification and randomization) will be re-checked. 243 244 o Double -masking 245 The Investigator, subjects, and study personnel of the study center will be masked 246 to the trial equipment worn. Examiners measuring primary outcomes have no 247 contact with study equipment from the study or the participants. 248 249 The pockets containing the lenses will be the same for both Reference and Test 250 lenses. The lens packaging design will be created in order to dispense the correct 251 equipment to each subject according to the randomization list. 252 253 In the event of early discontinuation of the study or in case of emergency, the 254 Investigator will ask the study manager in charge of lens logistics to break the 255 masking, but only if it is considered mandatory for a further major reason. The fact 256 that a subject reaches an endpoint is not a reason for breaking the masking. 257 258 The Investigator should record the reason for the date and time of breaking, his/her 259 name, and function in the study together with his/her signature. The date and 260 reason for the unmasking must also be documented in the source document and 261 the Case Report Form. 262

386
The interim analysis at 1 year and a final analysis at 2 years will be performed 387 when 150 subjects have completed visit 6 or when 33 has dropped out prior to 388 visit 6 and when data transferred to the file FIN Data.xlsx, having been 389 documented as meeting the cleaning and after the finalization and approval of 390 this SAP document. 391 The intention-to-treat approach will be used in the analysis for all subjects who 394 received any study device. The principle of intention-to-treat will be the main 395 strategy of analysis adopted for primary and secondary endpoints. These analyses 396 will be conducted on all patients assigned to the treatment groups as randomized 397 regardless of the study treatment received. 398

Safety Population 399
For the analysis of safety outcomes using exploratory endpoint, all subjects who 400 received any study treatment (including control) but excluding subjects who drop 401 out prior to receiving any treatment are included as in the intention-to-treat 402 approach. 403

Covariates and Subgroups
Provide a general comment identifying the covariates (continuous or categorical, 405 including subgroups) that are expected to have an important influence on specific 406 endpoints (e.g., demographic or baseline measurements, concomitant therapy). 407 Document any model selection procedures (e.g., forward stepwise selection). 408 Subgroup analysis will only be done for variables that were not randomized equally. 409 The main variables for randomization were cycloplegic autorefraction, age, and 410 gender, which are important variables for primary endpoint. The other variable that 411 will be included in subgroup analysis will be axial length should there be any 412 between group difference during baseline. 413 Other demographic variables like age of myopia onset, parental myopia, height, and 414 weight will be analyzed for their effect on myopia progression. 415

416
No missing data imputation will be done. 417

418
Interim analysis will be done after 1year follow-up. Blind analysis will be carried out 419 by a third-party statistician that is independent with no interest and involvement in 420 the clinical trial. Results from the blind analysis will be shared with the Data and 421 Safety Monitoring Committee (DSMC). DSMC will advise if the clinical trial is safe to 422 be continued. 423 Blind analysis is done by removing all identifiable variables like subject 424 identification and birth dates after including the masked concepts variable by the 425 one personnel with the masked information. The data will be scrambled before it is 426 passed to the third-party statistician for analysis. 427

Purpose of Interim Analyses 428
The purpose of Interim Analysis is to have early decisions for safety, endpoint, and 429 for future planning of resources. 430 The safety of treatment will be assessed based on adverse events reporting, 431 changes from baseline through cycloplegic autorefraction, and axial length. 432 Efficacy of treatment will be based on differences in cycloplegic spherical equivalent 433 and axial length with control group. 434

Planned Schedule of Interim Analyses 435
There will only be 1 interim analysis at 1 year follow-up. 436

Scope of Adaptations 437
Change in study design like switching inferior treatment arm to a more superior one 438 or control will be employed should there be a safety issue in the interim analysis. 439

Stopping Rules 440
Trial will be stopped if cycloplegic autorefraction or axial length of treatment 441 progressed 10% worse than control.  The summary statistics will be produced in accordance with section 10. 510

511
This will be reported in the summary statistics. 512

513
Compliance with intervention is monitored using questionnaires collected during 514 Visit 3, 4, 5, and 6. In addition, phone interviews will be done 3 days, 2 weeks, and 515 3 months post dispensing after every visit. The main measure of compliance is the 516 number of days the investigative device was worn and the number of hours worn 517 each day. Total compliance is defined as 7 hours a day for 7 days a week, 518 equivalent to 49 hours of wearing a week. Based on total hours of wearing in a 519 week, Actual compliance will be actual wearing duration in a week, and Individual 520 compliance is defined below: 521 522 Individual compliance = (Actual compliance / Total compliance) * 100% 523 524 Study compliance will be the percentage of study participants that manage to 525 comply at least 80% of Individual compliance, defined below: 526 527 Study compliance = (Individual compliance => 80% / Total study population) * 100% 528

529
All efficacy variables will be listed by subject. Data will be summarized by 530 treatment group. N, Mean, Standard Deviation, Standard Error, Minimum, and 531 Maximum will summarise continuous efficacy variables, whereas number and 532 percent will summarise categorical efficacy variables. 533 All analyses of the continuous efficacy variables (e.g., cycloplegic autorefraction 534 and axial length) will be performed as analysis of variance with treatment group. 535 Treatment groups will be tested at the 2-sided 5% significance level. 536 All assumptions for regression models will be assessed by viewing plots of the 537 residual values 538 All analyses of categorical efficacy measures will be performed using logistic 539 regression with the treatment group and adjustments for variables statistically 540 different in each group. 541

542
Spherical equivalent will be calculated from sphere and cylinder measured in 543 Diopters (D) on an interval scale and computed as sphere + cylinder /2. Axial length 544 will be measured in millimeters (mm) on an interval scale. 545 Progression of spherical equivalent and axial length will be defined as the change in 546 spherical equivalent from the baseline within each crossover stage and will be 547 computed for each participant-eye. Progression of spherical equivalent and axial 548 length will be analyzed within a stage in a grouped format and between stages in a 549 paired format. 550 For the progression variables, normality of the underlying mean distribution will be 551 assumed based on central limit theorem as spherical equivalent, and axial length 552 are measured on an interval scale and the study sample being >30. Graphical 553 representation of the raw progression data between groups and stages will be done 554 using Box and Whisker plots. Descriptive statistics such as mean, median, standard 555 deviation, standard error, minimum, maximum will also describe the raw 556 progression data between groups and stages. 557 A linear mixed model that accounts for both fixed and random factors will analyze 558 progression in a grouped format. Study group will be modelled as a fixed factor. 559 Subject intercepts will be used to account for the correlation of 2-eye data. 560 Confounders such as age, gender, parental myopia, and baseline refractive error 561 could be used as covariates in the model. Model-based estimated means for each 562 study group that is adjusted for confounders with its 95% confidence limits will be 563 reported. 564 Spherical equivalent and axial length between-group will be analyzed ANOVA with 1 565 treatment comparing to 1 control group. The order of assignment will be accounted 566 for in the model and tested for significance. Confounders such as age, gender, 567 parental myopia, and baseline refractive error could be used as covariates in the 568 model. 569 Model-based estimated means for each study group adjusted for confounders with 570 its 95% confidence limits will be reported. 571 572 These analyses will be performed on the (1) Intent to treat analysis datasets and (2)  573 Per-protocol analysis dataset. 574 575 Individual data points that are missing will be excluded from analysis involving only 576 those specific variables. A participant's complete visit data will not be excluded if 577 some of the observations are missing. Inclusion of outliers in the analysis will be 578 based on the magnitude of change in test statistics with and without the outliers. 579 Outliers will preferably be retained unless there is a significant change in test 580 results. Any outliers removed from analysis will be reported along with reasons for 581 exclusion. In conjunction with the PI, the study optometrist may label a study visit 582 as "non-evaluable" on any protocol deviations as deemed appropriate. This would 583 then be used to exclude the study visit from the analysis. 584 Two-tailed distributions will be used for inferential statistics. Level of significance 585 will be set at 5%. Any post hoc multiple comparisons will be adjusted using 586 Bonferroni correction to ensure the family wise error rate is set at 5%. 587

588
Proportion of population without myopia progression will be compared with control 589 group. Proportion with 80% compliance with device will also be compared with 590 control group using Fisher exact test base on outcome of questionnaires. 591 Wearing time is measured using 6-monthly questionnaire. Average wearing hours 592 was calculated by taking the weekly average between 6 month and 24 month 593 questionnaires divide by 7 days in 1 week. 594 Group comparison of near, distance visual acuity and wearing time will be done 595 using ANOVA, which includes study group. means for each study group with its 95% 596 confidence limits will be reported. 597 598 Graphical representation of visual acuity data between groups will be done using 599 line graph or Box and Whisker plots. Descriptive statistics such as mean, median, 600 standard deviation, minimum, maximum will also be used to describe the raw 601 accommodation data between groups. 602 603 604

606
Exploratory outcomes like peripheral autorefraction and axial length and choroidal 607 thickness will be compared with control group using ANOVA after adjusting for 608 confounders. 609

610
It will be based on adverse events and serious adverse events reporting. 611

612
Changes in cycloplegic autorefraction and axial length will be plotted against 613 wearing period to understand the extent of exposure. 614