Can we predict long-term efficacy from short-term outcomes?

To express myopia control efficacy of a treatment, the terminology Cumulative Absolute Reduction in Axial Elongation (CARE) was introduced by Brennan et al¹ in their landmark 2020 paper Efficacy in Myopia Control. Designed to allow for accurate, direct comparison of treatment efficacy, CARE was proposed as an alternative to percentage efficacy, which is problematic when comparing across studies as it is influenced by multiple factors such as the study duration and control group characteristics. Instead of efficacy being expressed as a relative percentage of the progression measured in the control group, CARE expresses an absolute efficacy - the total reduction in axial growth of the treatment group compared to the control group.

This paper was the first to comprehensively compare different treatments for myopia control, and used the metric of CARE to do so. This analysis included a key chart which plotted the axial length difference between the treatment and control groups (the absolute treatment effect: y-axis) against the duration of the study (x-axis). This chart is shown below, with key pointers added to help your understanding. Note that only optical interventions with at least 10 data points were included - this could mean either at least 10 studies, or reports of multiple efficacy time points within a study. As such, spectacle interventions, soft multifocal contact lenses (SMCLs) and orthokeratology were included. Pharmacological interventions were not shown owing to "dramatically different observed efficacy across the concentration range."¹

Each solid circle represents a study data collection point from an individual study on orthokeratology. Studies on soft multifocal contact lenses (SMCLs) are indicated by open triangles and spectacles by the open squares. Note that this data only includes peer-reviewed studies published up to 2020 with statistical significance and demonstrable efficacy of at least 0.11mm at any time point.

The solid and dotted lines are the best fit curves for the studies of the different treatment types: the trend line which is closest to all data points for myopia control efficacy of that treatment, over time.

The largest effect size of treatment is noted in this first year - the curve is steepest over this time, indicating that the greatest annual CARE occurs in this first year of treatment.

• Around 31-40% of the total long-term effect occurs within the first 6 months
• Around 46-54% of the total long-term effect occurs within the first 12 months

This data analysis indicates that treatment effect is not constant over time, with these best-fit lines showing the greatest effect at the start of treatment and lesser absolute effect as treatment continues.

The best-fit curve for the orthokeratology data includes a three-year and a four-year study. Most myopia control studies, otherwise, are 1-3 years in duration. One seven year study of orthokeratology showed 0.44mm total CARE,² which Brennan et al equate to around 1D less final myopia. This 1D - although seeming modest - has been shown in another paper ³(with two authors in common) to be likely to reduce the risk of myopic macular degeneration (MMD) by around 40%, right across the refractive range of myopia. This appears to be a benefit of treatment which is independent of the degree of myopia.^1,3

When comparing multiple treatments as in this chart above, it appears that treatment efficacy is non-linear with time. There is an an initial burst of efficacy, with similar long-term outcomes over the duration of studies which are currently available for analysis. The authors noted that apparent differences in the curves fit to the different treatment categories should not be taken as superiority of one category over another, as they have not been statistically compared.

For optical methods of myopia control, there appears to be a burst of some 31-40% of the projected four year data happening in the first 6 months and 46-54% of the total efficacy occurring within the first 12 months. Thereafter, a combination of the longest-term data available and extrapolation of the best-fit curves indicates that there is no apparent superior treatment over the others.

The authors then go on to state the following caveats.¹

• Some treatments within these categories (eg. "SMCLs that prioritize clear vision, progressive addition spectacles and 0.01% atropine") may be less effective
• Side effects and potential for rebound within these categories may influence success
• The clinician should factor in their own skill set, treatment availability, patient and parent preferences and capacity, and regulatory considerations in choosing the ideal treatment.

The clinical application of this knowledge provides support for the wide application of myopia management, as follows. Note that these suggestions reflect the opinions of the author and are not intended to represent those of Johnson & Johnson Vision.

1. Understand that indications for various treatments vary by region. Know the approved indications for products available and prescribe the best treatment which is available to you, and assess the efficacy of the treatment after 12 months.¹ You are likely to then measure around half of the total efficacy effect in that first year.
2. Knowing that efficacy is similar across treatments, it is most critical that the treatment prescribed fits the patient's lifestyle, expectations, motivation and their abilities.
3. Keep in mind that these analyses, and indeed all studies of myopia control efficacy, present mean values - an average of outcomes. By definition, this means that around 50% of myopes could be expected to receive a greater treatment benefit than the mean CARE, while the similar proportion would demonstrate a lesser result. Since there is currently limited information to support individualized selection of treatments, using the treatments available and monitoring over at least 12 months for efficacy, tolerance and compliance constitutes evidence-based management.

• Considering The Individual In Myopia Management

• How Can We Set Myopia Control Expectations?