In this episode of Evidence Based Retina, Rishi Singh, MD, FASRS, and Katherine E. Talcott, MD, FASRS discuss how imaging biomarkers—including ellipsoid zone integrity, low luminance deficit, and hypertransmission on en face OCT—can help clinicians identify which patients with geographic atrophy (GA) are most likely to benefit from complement inhibitor therapy, along with practical strategies for guiding the patient conversation and staging unilateral versus bilateral treatment.
Dr. Talcott is a retinal surgeon at the Cole Eye Institute, Cleveland Clinic.
OCT biomarkers in GA progression fast facts
- A pooled post-hoc analysis of the GATHER1 and GATHER2 trials found that avacincaptad pegol (ACP) 2mg significantly reduced hypertransmission progression compared with sham starting at month 12, with reductions of 17.4% at month 12, 31% at month 18, and 33% at month 24.1
- Percentage hypertransmission area excess—the extent of hypertransmission beyond established GA boundaries—was lower in the ACP group at every post-baseline visit, ranging from 17% at month 6 to 34% at month 24.1
- Baseline ellipsoid zone (EZ) integrity outside areas of existing GA and low luminance deficit both correlate with faster GA progression and may serve as predictive biomarkers for treatment candidacy.
- Automated machine learning–based en face optical coherence tomography (OCT) quantification enabled reliable, graded measurement of hypertransmission area across nearly 24 months of pooled trial data.
- These findings support en face OCT hypertransmission measurement as a viable clinical trial endpoint and an early structural biomarker for GA progression and treatment response.
Taking a deeper dive into GA imaging biomarkers
Among Dr. Talcott's findings, hypertransmission on en face OCT is the most immediately actionable. Defects 250μm or larger already serve as precursor markers for future GA formation, independent of treatment.2
In the pooled GATHER1/GATHER2 analysis, ACP reduced hypertransmission progression versus sham by 17.4% at month 12, 31% at month 18, and 33% at month 24, using fully automated machine learning–based quantification consistent with recently validated deep learning models achieving over 90% detection accuracy across OCT devices.3
This pairing of trial-derived efficacy data with an automatable measurement technique is what elevates hypertransmission from a research curiosity to a plausible clinical endpoint.
“But hopefully, with further AI technology and integration of these tools, that'll be easier to do in the future.”
Important indicators: EZ integrity and low luminance deficit
Ellipsoid zone integrity outside the atrophy border offers a complementary baseline biomarker. In a separate GATHER post-hoc analysis, eyes with greater EZ attenuation and a larger EZ-GA gap at baseline showed significantly faster 12-month lesion growth across every metric examined.4
An earlier analysis of the ReCLAIM-2 placebo arm reached the same conclusion in an entirely separate population,5 suggesting EZ health at diagnosis carries prognostic weight independent of which therapy a patient ultimately receives—precisely the kind of guidance Dr. Talcott said clinicians are "all looking for."
Low luminance deficit, meanwhile, captures functional vision loss under dim lighting rather than standard chart conditions. In the GATHER analysis, patients with greater baseline deficit were more likely to also show EZ disruption and faster progression, consistent with prior data linking low luminance deficit to more than double the relative risk of subsequent visual acuity loss.6
Because it requires no imaging equipment beyond a chart and a neutral density filter, it may be the most accessible of the three markers—though all three, as Dr. Talcott noted, stem from post-hoc analyses still awaiting prospective validation.
Incorporating imaging into the patient conversation
Dr. Talcott never initiates treatment at the first visit. Instead, she reviews serial OCT imaging to establish how quickly a patient's atrophy has been expanding, then walks the patient through what that imaging shows before treatment is even mentioned, as many patients prefer to discuss the decision with family before returning to initiate therapy.7
When selecting candidates for complement therapy, Dr. Talcott favors non-foveal involving GA in eyes with preserved vision, watching for hyperautofluorescence at the atrophy border as a sign of active progression.
Research points to family history of vision loss and concurrent wet AMD treatment as markers of a receptive patient, while cautioning that large lesion size, hospice care, and severe dementia limit realistic benefit.8 Patients who have already lost significant function in one eye to subfoveal GA are often the most motivated, since they can appreciate slowed progression even without a corresponding gain in acuity.7
“In those patients, I think they're very motivated; you just have to counsel them appropriately about the risks and benefits of treatment for bilateral disease.”
Additional considerations for complement therapy
When determining whether treatment should be unilateral versus bilateral, clinicians often favor treating the worse-seeing eye first as a functional safety test, given the low but nonzero risk of intraocular inflammation and retinal vasculitis.7,8 If that eye tolerates treatment, injections extend to the better-seeing eye with greater confidence.
For patients also on anti-VEGF therapy, Dr. Talcott alternates injection visits between agents rather than combining them same-day, though same-day dosing isn't contraindicated.
In closing
Imaging biomarkers are steadily narrowing the gap between GA diagnosis and informed treatment decisions. As hypertransmission, EZ integrity, and low luminance deficit move from post-hoc analyses toward prospectively validated tools, clinicians will be better equipped to identify the patients most likely to benefit—and to guide them through a conversation that remains, at its core, about managing uncertainty together.
