Machine learning and AI advances show promise with AMD mapping and tracking

By Janice English

Long-awaited treatments for late-stage dry age-related macular degeneration (AMD; geographic atrophy) are a year underway in the United States, and the rest of the world is watching. The future of care appears promising, with a new era of sight-saving intervention in the pipeline.

Enhanced diagnostic imaging and analysis capabilities, as well as customized treatment pathways, require an efficient approach to patient management and engagement with costly, lifelong therapies for optimum use of resources and successful patient outcomes.

AMD is predicted to affect 288 million people by 2040, with the condition causing 3 times the sight loss of any other ocular morbidity.¹ Earlier diagnosis and treatment options are a reason for optimism among patients and their families.

Artificial intelligence (AI) is showing success in identifying patients with the greatest opportunity for effective treatment, furthering optimism within the field.

Ursula Schmidt-Erfurth, MD, Chair of the Department of Ophthalmology and Optometry at the Medical University of Vienna, has a keen eye on the subject. She is a pioneer of machine learning and AI for diagnostic image analysis.

“Severe vision loss has increased by 24% over the past 15 years², yet we now have the most advanced screening technologies,” she said. “We need to ensure that doctors are maximizing the potential of these solutions.”

Dr. Schmidt-Erfurth suggested that the development of the CE-marked RetInSight GA Monitor is set to have a profound effect on care. Launched in October 2023³, the application is accessible through Heidelberg AppWay and is being used by clinicians in Europe, as well as researchers globally, as a gateway to further analysis.

“My group has analyzed all the clinical trial data from the 3 Apellis trials, which encompasses more than 2,000 patients over a 2-year period,” she said.⁴ “Our focus is on analyzing gold-standard optical coherence tomography (OCT) images to see biomarkers reflecting disease activity. With advanced analysis of OCT images, we can examine every pixel and investigate the 2 layers of interest. Measuring the retinal pigment epithelial (RPE) can be fully automated, and it is reproducible and reliable in real time. The photoreceptor layer is the first to be affected, and we can visualize this with RetInSight’s AI-based OCT tools. Within a minute, we can have a report showing the detailed loss of RPE and photoreceptors. We do not want to wait and observe; we want to offer treatment as soon as possible. But how do we know which patients will benefit from therapy? If we only see loss of pigment, we know the lesion is not currently active. Studies indicate this is around 25% of patients, so we review them every 3 months to catch the lesions when they become active.”⁵

Millions of OCT images are being utilized to train deep-learning algorithms to identify new biomarkers.⁶ As part of this, the Vienna clinic’s work in advanced image analysis enables disease diagnosis; monitoring of disease progression; and identification of the need for personalized, precision medicine. Working in real-time, AI is becoming an assistant to the physician, providing fast analysis for informed decision-making.

Frank Holz, MD, Chair of the Department of Ophthalmology at the University of Bonn in Germany, sees the tremendous value of utilizing high-quality imaging to streamline care for this growing patient group resulting from an aging population.

“AI is essential,” he said. “We are taking a sequence of images and generating a huge amount of data. Reading them can be time-consuming, and a human grader cannot extract the degree of data that machine learning can provide.”

Referrals can filter patients in the community and, once in the referral pathway, they can be assessed faster, with decisions made on monthly or bimonthly treatment. For many patients, injections every 2 months suffice, as was the case for 80% of patients in the U.S. trial.⁷ Personalized medicine is important, as injections can induce very variable reactions amongst patients; it is important to reduce overdosing and wasting resources.

AI solutions have the potential to optimize therapeutic pathways for the diagnosis and monitoring of neovascular AMD. Recent innovations in therapy have seen the introduction of long-duration agents and surgically implantable devices proving sustained slow delivery of anti-angiogenic drugs. Even with new treatment approaches, the burden on clinical services remains high. The addition of therapy for atrophic AMD will only confound the problem and innovations like AI-based clinical support tools and service modernization will be the key to success.

RetInSight’s MDR-approved, CE Marked Fluid Monitor for wet AMD cases,⁸ now used by consultants in 30 clinics internationally, is enabling decision-making based on rapid or slowly progressing disease, according to Dr. Schmidt-Erfurth.

“The AI tool, used in conjunction with HEYEX 2 and Heidelberg AppWay, produces an immediate report that visualizes where the fluid is and measures the volume; everything is precise in real-time. This brings an enormous improvement in workflow, and early adopters say their productivity in clinics has improved by 40%,” she said.

“We have moved on considerably since central retinal thickness was the single measurement. Distinct types of fluid have different impacts on visual acuity—intra-retinal fluid is the most dangerous, while sub-retinal fluid can be tolerated to a degree. We need to know if the fluid is dynamic, and the profession needs to know when to treat and how much. A personalized approach leads to faster, more efficient treatment without wasting resources,” she added.

The simple 1-page Fluid Monitor Report indicates the treatment pathway.⁹ “It is objective, reproducible, and enhances overall understanding. Clinicians can feel confident that they are giving bespoke treatment at the right time; the retinal specialist is still pivotal in decision-making,” added Dr. Schmidt-Erfurth.

The report is also a valuable aid to patient compliance and, working together with the OCT images, provides unrivaled visualization. It can show the full history of 3 key biomarkers of intra-retinal fluid, sub-retinal fluid, and retinal pigment epithelium detachments—from 1 clinic visit to the next.

With an array of breakthrough drugs in the pipeline, clinicians are up-skilling, and healthcare managers are preparing patient pathways. The price of treatment may be considerably less than residential care for a patient who is no longer able to look after themselves and requires round-the-clock care.

Visualization of a condition with no obvious manifestation in the early stages is perhaps the greatest challenge among an older population. Patients may be keenly aware of the symptoms of cancer, rheumatoid and osteoarthritis, gout, diabetes, cardiovascular, pulmonary disease, and even cataracts. Patient engagement with lifelong monitoring and treatment for AMD is much easier with high-resolution imaging.

It is time to move the profile of this disease up the healthcare agenda and to give patients optimism with a new era of sight-saving intervention becoming more widely accessible to all.

Janice English is a UK-based medical journalist who, since 2012, has worked with ophthalmic industry leaders to promote diverse technologies and its users. She is also Director of PR for the UK Optical Supplier’s Association with a keen interest in device interoperability and standardization of e-health device outputs. Prior to moving into the ophthalmology arena, Ms. English trained as a journalist on weekly titles, DOCTOR and HOSPITAL DOCTOR, and launched and co-edited the UK practice management journal, THE DENTIST.

References

1. Klaver CC, Wolfs RC, Vingerling JR, et al. Age-specific prevalence and causes of blindness and visual impairment in an older population: the Rotterdam Study. Arch Ophthalmol. 1998;116(5):653-658. doi:10.1001/archopht.116.5.653
2. Vos T, Allen C, Arora M, et al. Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016;388(10053):1545-1602. doi:10.1016/S0140-6736(16)31678-6
3. RetInSight secures the first MDR CE-certification for its AI-based GA monitor for the management of geographic atrophy. November 9, 2023. Accessed July 12, 2024. https://retinsight.com/retinsight-secures-the-first-mdr-ce-certification-for-its-ai-based-ga-monitor-for-the-management-of-geographic-atrophy/
4. Schmidt-Erfurth U, Mai J, Reiter GS, et al. Therapeutic effect of pegcetacoplan on retinal pigment epithelium (RPE) and photoreceptor (PR) integrity in geographic atrophy (GA) in the phase III OAKS and DERBY trials. Presented at the ARVO 2023 Annual Meeting, April 23-27, New Orleans, Louisiana.
5. Schmidt-Erfurth U, Mai J, Gregor S, et al. Disease activity and therapeutic response to pegcetacoplan for geographic atrophy identified by deep learning-based analysis of OCT. Ophthalmology. 2024. In press.
6. Schmidt-Erfurth U, Reiter GS, Riedl S, et al. AI-based monitoring of retinal fluid in disease activity and under therapy. Prog Retin Eye Res. 2022;86:100972. doi:10.1016/j.preteyeres.2021.100972
7. Syfovre Injection Prescribing Information. Drugs.com. April 15, 2024. Accessed July 12, 2024. https://www.drugs.com./pro/syfovre-injection.html#s-34068
8. RetInSight Fluid Monitor provides next evolutionary step in monitoring of neovascular age-related macular degeneration. RetInSight. May 18, 2022. Accessed July 12, 2024. https://retinsight.com/retinsight-fluid-monitor-provides-next-evolutionary-step-in-monitoring-of-neovascular-age-related-macular-degeneration/
9. The RetInSight fluid monitor for neovascular age-related macular degeneration (nAMD). RetInSight. 2024. Accessed July 12, 2024. https://retinsight.com/fluid-monitor/