July 11, 2024

Transform Retinal Disease Management by Evaluating Structural Endpoints

How can AI-driven evaluation and monitoring of structural endpoints aid in managing complex retinal diseases? This is the topic we're exploring in this blog.

The challenge of defining relevant outcomes for clinical studies in Ophthalmology is considerable1. Traditional patient-reported outcomes often fall short in accurately measuring disease progression and treatment effects, particularly in complex eye diseases.

This gap has led to a shift towards closely monitoring structural endpoints, in some cases as primary clinical endpoints, that are being considered by regulators in some retinal diseases such as Geographic Atrophy and MacTel. 

In this blog, we delve into how structural endpoints measured and evaluated using artificial intelligence technology are helping to better monitor diseases such as neovascular Age-related Macular Degeneration (nAMD), Geographic Atrophy (GA) and Central Serous Chorioretinopathy (CSC). 
9 minutes read

The emergence of structural endpoints


How can we transcend traditional diagnostic methods to more precisely gauge the effectiveness of treatments? The answer might lie in the adoption of measuring structural endpoints, which are objective and can be monitored over time to support our evaluation of therapeutic outcomes in complex eye diseases. 


Understanding Structural endpoints

Structural endpoints refer to quantifiable, physical changes or statuses within the body that can be measured consistently through imaging or other diagnostic methods2

In Ophthalmology, these endpoints provide a more definitive, quantitative way of assessing the progression of diseases and the efficacy of treatments over time. Unlike patient-reported outcomes which are subjective and can vary greatly between individuals, structural endpoints offer a standardized method to evaluate and compare clinical outcomes.

For instance, in diseases like Geographic Atrophy (GA) and neovascular Age-related Macular Degeneration (nAMD), the measurement of lesion sizes, retinal thickness, or the volume of fluid accumulation can serve as reliable indicators of disease activity3.

Scientific Validation and advantages 

The FDA's acceptance of structural endpoints for diseases like GA highlights their growing credibility and utility, especially when evaluating diseases with little to no significant change in vision occurring within the duration of a clinical study. These endpoints not only facilitate a clearer understanding of a treatment's impact but also enhance the rigor of clinical trials by providing objective data that can be uniformly interpreted.

Moreover, the precision of imaging technologies such as Optical Coherence Tomography (OCT) allows for detailed visualization and measurement of retinal layers, making it possible to detect minute changes that might indicate the early stages of disease or subtle responses to treatment.

For example, the quantification of subretinal fluid (SRF) in nAMD using OCT provides crucial data on the disease’s response to anti-VEGF therapies, offering insights into the optimal frequency and dosage of treatment. 

The shift towards structural endpoints is already making a significant impact in clinical settings. Their application is proving particularly valuable in long-term diseases where monitoring disease progression and treatment efficacy over many years is crucial. By providing a clear, objective measure of disease status, structural endpoints allow healthcare providers to make more informed decisions, adjust treatments more precisely, and better predict patient outcomes.

Integrating AI for enhanced precision 

The integration of Artificial Intelligence (AI) with imaging technologies further amplifies the potential of measuring structural endpoints, quickly, consistently, and accurately, without human fatigue. AI algorithms can analyze vast amounts of imaging data with high precision and speed, identifying patterns and changes that may be imperceptible to the human eye.

This capability not only speeds up the diagnostic process but also enhances the accuracy of treatment monitoring, potentially leading to more personalized and effective treatment strategies.


How RetinAI Discovery’s features can help

At RetinAI, our mission is to empower clinicians with tools to transform retinal disease management. RetinAI Discovery provides comprehensive visual and quantitative assessments of disease progression and treatment response. By leveraging AI-driven analysis, clinicians can obtain actionable insights into fluid dynamics and retinal structure, aiding personalized treatment planning. 

Curious about the specific benefits of RetinAI Discovery?
Read the entire blog for more in-depth demonstrations and insights. 

Structural endpoints in Retinal Diseases

Structural endpoints are transforming the landscape of retinal disease management, offering clinicians precise tools to monitor disease progression and treatment efficacy. Let's explore their application in key retinal conditions:


Structural endpoints in Neovascular Age-Related Macular Degeneration (nAMD) 


In nAMD, treatment approaches often focus on extending dosing intervals without compromising efficacy. Understanding which patients can safely have extended intervals without risk of disease progression is then critical. Structural endpoints such as fluid volume dynamics and retinal layer thickness emerge as key indicators in this regard4

However, numerous questions persist regarding the utilization of structural endpoints in nAMD management:

  • Can we effectively inhibit the growth of macular neovascularization (MNV)?
  • Is such inhibition even desirable, or could controlled MNV growth potentially confer benefits to patients?

These questions underscore the complexity of treatment decisions in nAMD and highlight the imperative need for precise, objective assessment methods. 

The advent of automated analysis, facilitated by AI-driven platforms like RetinAI Discovery, presents a transformative opportunity for clinicians. By generating comprehensive progression graphs over the treatment course, such platforms enable clinicians to stratify patient populations and disease phenotypes, to devise novel parameters for assessing treatment response and disease progression.


Clinicians might be able to then examine patients who never had fluid ‘spikes’ above a certain threshold and compare them to others who experienced more ‘spikes’, evaluating their performance. This hypothesis, suggesting that ‘spikes’, or acute, significant increases in fluid from previous visits, may be more detrimental than persistent SRF, remains theoretical. However, with automated measurements gained through AI efficiency and long-term follow-up data, clinicians can potentially gain deeper insights into disease progression, stratification and treatment response.

Explore how RetinAI Discovery provides actionable insights into nAMD management: 

Structural endpoints in Geographic Atrophy

In Geographic Atrophy (GA), treatment strategies often emphasize the need to comprehend disease progression and the efficacy of interventions. Given the nature of GA and its propensity to cause irreversible vision loss, accurately gauging disease status and treatment response become paramount. Structural endpoints measured on OCT, such as the photoreceptor layer integrity, including the ellipsoid zone, and RPE loss emerge as crucial metrics in this context, in addition to GA area on FAF, especially in targeting treatment. 

Yet, questions remain:

  • Can we accurately predict GA progression?
  • Is even earlier intervention beneficial, such as in intermediate AMD, or should treatment initiation be delayed until later stages?
  • Whom to treat first?
  • Has the treatment been effective?
  • What can we expect from a potential treatment?
  • Could microperimetry become a standard tool alongside structural imaging to enhance the understanding and management of GA?

Objective assessment methods showcasing precise and quantifiable metrics could help clinicians in determining answers to these questions.

Automated AI-driven analysis can definitely help clinicians managing GA by delineating the extent of atrophy, tracking and predicting changes over time. This is what RetinAI Discovery is doing; the platform empowers clinicians to monitor GA progression with unparalleled accuracy.

Explore how RetinAI Discovery enhances GA treatment outcomes:  

Structural endpoints in Central Serous Chorioretinopathy 

Application of RetinAI’s Discovery platform and AI has expanded to understanding Central Serous Chorioretinopathy (CSC) better. The platform’s ability to identify and quantify subretinal fluid provides valuable insights into this disease status and treatment response.

Predicting CSC progression or regression accurately remains challenging, especially considering the variability in patient outcomes over time. Determining the optimal timing for treatment initiation poses another dilemma—is early intervention beneficial, or should it be delayed until later stages, or will the fluid regress on its own? These questions highlight the complexity of managing CSC and underscore the need for precise, objective assessment methods.

Enhancing disease treatment with RetinAI Discovery 

RetinAI Discovery leverages AI-driven analysis to enhance treatment outcomes across various retinal diseases. Let's explore how our platform supports clinicians in optimizing patient care:

How RetinAI Discovery Aids nAMD Management


RetinAI Discovery platform offers support in navigating the intricacies of nAMD management. Leveraging advanced algorithms, the platform precisely delineates areas of intraretinal (IRF), subretinal fluid (SRF) and pigment epithelium detachment (PED)  on OCTs, facilitating monitoring of treatment efficacy using OCT scans. 

Allowing OCT juxtaposition, to see the evolution of scans before and after treatment, RetinAI Discovery empowers clinicians to use both a visual comparison between scans and a quantitative measure of change as evidence of the therapeutic impact on fluid levels. This understanding enables clinicians to help educate patients, tailor treatments to individual patient needs, optimizing outcomes and minimizing potential risks.

In addition to visualizing and quantifying fluid volumes and retinal layer thickness on OCT scans, RetinAI Discovery provides longitudinal data insights. These data serve as a cornerstone for elucidating disease progression patterns and refining treatment strategies. With RetinAI Discovery, clinicians gain access to actionable insights, paving the way for enhanced patient care and improved treatment outcomes. 

How RetinAI Discovery Enhances GA Treatment

With RetinAI Discovery, clinicians can visualize and measure changes in the geographic atrophy area and retinal layer thickness over time, aiding in the assessment of treatment response and disease progression. This approach enables clinicians to optimize outcomes and preserve vision to the greatest extent possible.

The platform's predictive analytics capabilities allow for the detailed monitoring of GA progression. By analyzing the expansion of atrophic patches over time, over 54 months,  RetinAI Discovery helps in forecasting disease progression and evaluating the potential benefits of therapeutic interventions.

How RetinAI Discovery Supports CSC Treatment

The RetinAI Discovery platform provides precise measurements and quantifiable metrics. By analyzing fluid volume and location over time, clinicians gain deeper insights into disease mechanisms and treatment responses. This data-driven approach enhances clinical decision-making and enables personalized treatment strategies for patients with CSC.

Incorporating longitudinal data from follow-up examinations, sometimes spanning 10 to 15 years, with quick analysis possible, allows for a comprehensive analysis of disease progression and treatment outcomes. By quantifying fluid volume in different compartments and tracking changes over time, clinicians can better understand the impact of subretinal fluid on visual function and disease progression for instance. This long-term analysis lays the foundation for future research aimed at elucidating the underlying mechanisms of CSC and refining treatment approaches.

Conclusion

Structural endpoints, coupled with AI-driven analysis, provide an avenue for enhancing clinical decision-making and improving treatment outcomes in complex retinal diseases, including nAMD, GA, and CSC. By providing objective, quantifiable measures of disease activity and treatment response, these advancements empower clinicians to make more informed decisions, optimize treatment strategies, including personalized ones, and ultimately improve patient outcomes.

As we continue to refine our understanding of disease mechanisms and treatment responses, longitudinal data insights become increasingly crucial. By incorporating follow-up examinations spanning years, and working with tools allowing for progression data visuals, clinicians gain deeper insights into disease progression and treatment outcomes, paving the way for more personalized and effective treatment plans.

Thinking of harnessing the combined force of OCT and AI to enhance diagnostic precision in retinal disease management? Connect with our experts now.

This article was written by Léa Errog, MSc



References

1. Schmetterer, L., Scholl, H., Garhöfer, G., et al. Endpoints for clinical trials in ophthalmology. Progress in Retinal and Eye Research (2023), volume 97,101160, ISSN 1350-9462. https://doi.org/10.1016/j.preteyeres.2022.101160.

2. Schmetterer, L., Scholl, H., Garhöfer, G., et al. Endpoints for clinical trials in ophthalmology. Progress in Retinal and Eye Research (2023), volume 97,101160, ISSN 1350-9462. https://doi.org/10.1016/j.preteyeres.2022.101160

3. J Bakri, S., Bektas, M., Sharp, D., et al. Geographic atrophy: Mechanism of disease, pathophysiology, and role of the complement system. J Manag Care Spec Pharm. (2023), 29(5-a Suppl). https://doi.org/10.18553/jmcp.2023.29.5-a.s2

4. Metrangolo C, Donati S, Mazzola M, et al. OCT Biomarkers in Neovascular Age-Related Macular Degeneration: A Narrative Review. J Ophthalmol. 2021. https://doi.org/10.1155/2021/9994098