Over the last several years, there have been multiple studies published evaluating the efficacy of optical lenses for slowing myopia progression in children. In addition to bifocal spectacles, optical devices studied have included orthokeratology and soft bifocal/multifocal contact lenses. As described in multiple previous Refractive Focus columns, there have been promising results with contact lens modalities.
While there is great momentum in myopia intervention research, clinical studies can differ from one another in important areas including entry criteria, primary endpoint, study length, the way outcomes are reported, etc. These differences, even if subtle, can make it difficult to directly compare statistics and results across two or more studies.
Despite a heavy focus on myopia control in the literature, the use of any optical device in the United States to slow the progression of myopia is an off-label use; no device has an indication for myopia control from the U.S. Food and Drug Administration (FDA).
In September 2016, a workshop titled “Controlling the Progression of Myopia: Contact Lenses and Future Medical Devices” was hosted at the FDA’s White Oak campus in Silver Spring, MD (full transcript can be found online at www.fda.gov/MedicalDevices/NewsEvents/WorkshopsConferences/ucm500404.htm ). The event was cosponsored by the FDA and multiple optometry and ophthalmology organizations.
The one-day workshop began with talks to set the stage regarding our current understanding of myopia and previous intervention studies, the current and predicted future prevalence of myopia, adverse events in children, and potential complications associated with contact lens wear.
The workshop included three panels, each composed of two moderators, three to four myopia researchers (both from the United States and international), and two representatives from the FDA who provided clarification and insight into FDA policies and concerns.
Each panel was charged with coming to a consensus on key areas of clinical trial design for studies aimed at using a medical device to slow the progression of myopia. The consensus from these panels will be used by the FDA to aid in establishing guidelines for study requirements to obtain a myopia control indication for a medical device.
Panel 1 (Selection of Trial Participants) was tasked with reaching a consensus on issues surrounding eligibility criteria for clinical trials. After comments and discussion, the group reached the consensus that the range of myopia to include should be from either –0.75D or –1.00D spherical equivalent to up to –4.00D of myopia at the start of the study.
Additionally, cycloplegic autorefraction should be the method by which refractive error is measured. After considering both cyclopentolate 1% and tropicamide 1% as the cycloplegic agent, tropicamide 1% was chosen because it has been shown to provide sufficient cycloplegia for the measurement of refractive error outcomes (Egashira et al, 1993; Manny et al, 2001).
It was agreed that cyclopentolate is the best agent for cycloplegia; however, the panel also acknowledged that tropicamide 1% was a nice balance due to the shorter cycloplegic and pupil dilation effects given the importance of subject retention in studies for which children must return for multiple visits.
When considering the age of children, 7 to 12 years of age at enrollment was recommended by the panel to ensure that children are still progressing at the end of an up to three-year study. No children should have used a previous myopia control treatment. The panel noted that the race and ethnicity of the children enrolled should match the demographics of the United States, as opposed to over-representing any one race or ethnicity.
The panel also recommended that there be no requirement that children have a certain level of recorded myopia progression (i.e., a certain rate of progression over the previous year) to be eligible, as that would influence study generalizability and exclude newly diagnosed myopic children.
The panel on which I served (Panel 2: Study Design & Clinical Outcomes) covered topics ranging from study endpoints and duration to acceptable rates of microbial keratitis (MK). A single-vision soft contact lens was believed to be the best control group (as opposed to spectacles) due to the ability to maintain masking in studies of other soft contact lenses and for retention purposes. There was also discussion during the open comment period that the ideal control lens would be a soft lens with no spherical aberration so that it does not change peripheral defocus; many commercially available soft contact lenses change peripheral defocus in part due to the lens design’s manipulation of spherical aberration (Moore et al, 2017).
After discussing the best primary endpoint of a clinical trial, the panel agreed that axial length should be the primary outcome due to its better repeatability versus cycloplegic autorefraction. The panel’s rationale was that the purpose of any myopia control device is to slow eye growth, though it was noted that refractive error should still be measured in each study as a secondary outcome, as it is easier for a clinician to explain to a patient. Utilizing axial length also has the advantage of allowing for easier comparisons across studies of soft multifocal designs and orthokeratology.
The recommended length of a clinical trial was three years (possibly two years) to ensure that the treatment continues to accumulate over time. The panel was also asked to address the possibility of a rebound (faster growth) once children stopped using a treatment. Based on two clinical trials, one of progressive addition lenses and one of a novel soft lens for myopia control, evidence indicates that there is no rebound effect when treatment effects are clinically small (Berntsen et al, 2012; Cheng et al, 2016).
The question remains as to whether there might be a rebound with devices that produce a larger treatment effect. For this reason, the panel recommended a minimum of six months, but preferably one year, of follow up after children stop wearing the myopia control device to assess for a rebound effect. One year was preferable due to known seasonal variation in the progression of myopia (Gwiazda et al, 2014).
The acceptable rate of MK in children was discussed at length. The ultimate consensus was that a uniform definition of MK that is consistent with previous studies of MK rates is necessary to allow for cross-study comparisons. The MK definitions used by Schein et al (2005) when determining the rate of MK with 30-day silicone hydrogel extended wear were suggested.
The panel noted that the number of children needed to establish an MK rate in children far exceeded the sample size necessary for a myopia control clinical trial. It was suggested that data could potentially be combined across multiple trials or a post-market surveillance study could be performed to establish an MK rate in children. Overall, the importance of determining the rate of MK in children was stressed given that any lens with a myopia control indication would likely lead to more children being fitted with contact lenses.
Panel 3 (Patient-Centric Factors & Outcomes) discussed patient-centric outcomes in clinical trials. The panel generally felt that these outcomes should center on patient symptoms related to contact lens wear and visual quality of life and that additional patient-centric outcomes should be determined via focus groups of lens wearers.
It was noted that it is best to survey the children regarding symptoms, while parents would be better to survey regarding observable behaviors related to contact lens wear. Common strategies to improve clinical trial enrollment and retention were also discussed based on experiences from previous and current trials. These suggestions included the utilization of social media, providing comprehensive eye care to those enrolled, and maintaining engagement with parents and children, among others.
This workshop brought together a variety of myopia researchers. The topics discussed are timely, as new studies are being designed. The standardization of clinical trials would allow for more straightforward comparisons across studies. We look forward to guidelines for optical devices being issued by the FDA, which will help bring us a step closer to the potential for a device with a myopia control indication. CLS
For references, please visit www.clspectrum.com/references and click on document #258.