Retinal vein occlusion is a challenging health condition that significantly reduces patients’ quality of life, despite available treatments. Dr Adrian Hunt reviews new treatment options available for this condition.

WRITER Dr Adrian Hunt

Retinal vein occlusion (RVO) is a common retinal vascular disease characterised by venous dilatation, retinal haemorrhages, and oedema in the distribution of retina drained by the obstructed vein. Variants include branch (BRVO), central (CRVO), and hemiretinal (HRVO). RVO commonly presents with acute unilateral loss of vision due to macular oedema (MO).

It would be simplistic to consider the pathology of RVO merely an issue of plumbing, as multiple angiogenic and inflammatory cytokines are involved.1-3Chief among these is vascular endothelial growth factor (VEGF). The discovery of these molecules, and the potential for intravitreal therapies, no doubt prompted enthusiasm 20 years ago. At that time, that meant triamcinolone. However, it was the newly developed VEGF inhibitors that went on to dominate treatment of RVO. More recently, the therapeutic appeal of targeting Ang-2has been heightened by reports of particularly high levels in RVO.4

For many years, management of RVO was reactive, involving observation and laser. Early anecdotal experience with bevacizumab, backed up by trial evidence of VEGF inhibitors designed for ocular use, shifted practice to a proactive approach involving injections guided by visual acuity, clinical examination, and optical coherence tomography (OCT). Nevertheless, RVO remains a challenging condition causing significant morbidity, consuming considerable health resources, and decreasing quality of life, despite the safety and efficacy of available treatments.

Faricimab (6 mg Vabysmo, F. Hoffmann-La Roche AG, Basel, Switzerland) was added to the treatment armamentarium for RVO in Australia in March 2025, based on positive results from BALATON and COMINO trials.5 Aflibercept 8 mg (Eylea HD, Bayer) may become available in future. Given this, it is worth reviewing what we have learnt over the past few years, from trials and real-world studies, and considering the likely impact of new agents becoming available in routine care.

FIRST-GENERATION VEGF INHIBITOR RVO TRIALS

Trials have described the outstanding safety and efficacy of ranibizumab, aflibercept or bevacizumab delivered frequently to select cohorts.6-11 Treatment, typically with around nine or more injections in the first year, led to roughly +12 to +17 letter gains in visual acuity (VA).

The need for prompt referral to ensure early commencement of treatment of RVO was highlighted by studies with sham arms. A sixmonth delay in treatment meant the outcomes in the initially sham-treated eyes never really caught up to eyes treated from the beginning of the studies.

Beyond the initial stages of therapy, treatment outcomes tend to decay when injections are delivered less frequently. This is particularly the case in eyes with CRVO, when adopting a pro-re-nata (PRN) regimen either too early, injudiciously, or with three-monthly reviews. Some evidence suggests potential for good longer-term outcomes in RVO if macular oedema is well controlled.

Holistic management continues to include consideration of associated systemic disease, and panretinal photocoagulation (PRP) and sectoral laser remain standard of care for neovascular complications. Intravitreal steroids – because of their limited efficacy and side effect profile – have a second-line role in select eyes that are unresponsive to VEGF therapy according to treatment guidelines.

TRIALS OF NEW AGENTS FOR RVO

The BALATON (BRVO) and COMINO (CRVO/HRVO) double-masked, multicentre, randomised, Phase 3 studies assessed the safety, efficacy, and pharmacokinetics of faricimab for RVO. In part one of the study, from baseline to week 24, the primary endpoint was reached; monthly faricimab was found non-inferior to monthly aflibercept 2 mg, and all groups gained around +17 letters. The adjusted changes in central subfield thickness (CST) were also similar by drug at 24 weeks. Fluorescein angiography at 24 weeks, which is not typically part of routine care, differentiated the agents. Significantly more eyes achieved an absence of macular leakage when treated with monthly faricimab than with monthly aflibercept for both BRVO and CRVO.5

In part two of the BALATON and COMINO studies, from week 24 to 72, patients received faricimab only (mean, five to six injections).12 The injection interval, ranging from four to 16 weeks, was determined using a personalised treatment interval (PTI) based on disease activity. Visual and anatomical outcomes at week 24 were maintained through week 72 on the flexible faricimab regimen. A final treatment interval of 12 or 16 weeks was reached by over 57% of eyes with BRVO, and over 45% of eyes with CRVO at week 68. Some eyes that were stable at the maximum 16-week interval might have been extended further if allowed.

However, at the other end of the spectrum, 23% of eyes with BRVO and 32% of eyes with CRVO still required treatment every four weeks at 68 weeks.

The higher dose formulation of aflibercept (8 mg Eylea HD, Bayer) has not yet been evaluated for RVO by any regulatory bodies, nor have the results of the QUASAR Phase 3, multicentre, randomised, double-masked, active-controlled clinical study been formally presented or published. The results will presumably be reported in due course, following a press release in December 2024 from Regeneron.13 This release announced that the QUASAR study had met its primary endpoint of non-inferior visual gains in patients with RVO treated with Eylea HD every eight weeks compared with Eylea 2 mg every four weeks, through 36 weeks.

LIMITATIONS OF TRIAL EVIDENCE

The design of trials can limit the utility of the results they produce for patients in the real world. Trials are like an experiment aiming to provide evidence of safety and efficacy required for approval by regulatory bodies. Trials apply strict selection criteria to ensure that the only thing impacting outcomes is the treatment that was randomly assigned. This serves the aims of the trial, but each exclusion chips away at how representative the studied sample is of the full range of patients we encounter in routine practice. For example, eyes with ischaemic CRVO are typically excluded from trials, despite around a quarter to a third of eyes having it at presentation. Some RVO trials even exclude patients with blepharitis, let alone other common ocular pathologies that may influence outcomes such as cataract, age-related macular degeneration, or diabetic retinopathy.

The patients in routine practice are diverse and the treatment they receive is very unlikely to match the frequency of injections delivered to participants of trials. Randomised control trials (RCTs) are well funded and conducted within academic environments, ensuring strict adherence to protocol and follow-up. Treatment in routine care is more fluid, with fewer support staff to ensure adherence and persistence with treatment. This means significant burden falls on the patients and caregivers, and there is often the added pressure of paying for treatment.

REAL-WORLD EVIDENCE

Real-world evidence (RWE) collectively contributes information regarding “effectiveness” of an intervention that complements the “efficacy” data of pivotal trials. Prospectively designed clinical ophthalmic registries, such as the Save Sight Institute’s Fight Retinal Blindness! (FRB!) registries, can provide particularly high-quality data that lifts the standing of RWE in the hierarchy of evidence.14 (Research outputs from the FRB! registries can be viewed at: savesightregistries.org/research-outputs.)

The trial-like data captured using FRB! have informed management of RVO when treated with bevacizumab, ranibizumab or aflibercept 2 mg. The RVO analyses through 12 months found that:

• Lower treatment frequency in routine care, compared with trials, affects outcomes with the greatest impact on eyes with CRVO.15,16

• Aflibercept has greater efficacy than ranibizumab in reducing macular thickness in BRVO and CRVO and in improving visual acuity in CRVO in routine care through 12 months.15,17,18

• Similar outcomes can be achieved using bevacizumab.18

• Caution should be exercised by clinical trialists when including eyes with HRVO in BRVO or CRVO cohorts, as particularly good outcomes in HRVO with VEGF inhibitors may skew results.19

The RVO analyses through three years from FRB! found:

• Aflibercept continued to be more efficacious in reducing macular thickness than ranibizumab and bevacizumab.16,20 In CRVO one-third of eyes had good vision (≥ 6/12) and one-third poor vision (≤ 6/60) at three years,16 and in BRVO two-thirds of eyes had good vision and very few had poor vision at three years.20

• Typical injection frequency in routine care in the first, second, and third years was around eight, five, and four injections respectively, with half of all eyes still requiring injections at three years.16,20

• Cataract surgery is safe and efficacious during VEGF inhibition of MO due to RVO, though more injections are required in the 12 months after surgery.21

• A range of outcomes is achievable in routine care (BRVO: +9 to +16 letters and in CRVO: +5 to +14 letters) with treatment frequency of practitioners ranging from six to 12-weekly on average across 24 months.22

LIKELY IMPACT OF NEW AGENTS FOR RVO

The findings from the BALATON and COMINO trials are significant, as they suggest that patients may require fewer injections over time, perhaps helping to bridge the gap between the outcomes achieved in routine care and trials. The highly select subset of patients enrolled in trials, compared with the broad cohort managed in routine practice, along with different levels of treatment intensity, also needs to be kept in mind when trying to translate evidence. Nevertheless, if patients have a limited tolerance for treatment burden in routine care, a new agent may offer better disease control within that interval or achieve similar outcomes with a longer treatment interval.

The main lesson learnt from past experience with availability of new agents is that, almost by reflex, we tend to think of our most treatment dependant patients as the most likely beneficiaries. These patients may certainly benefit from better levels of disease control. However, expectations regarding long intervals should be tempered, as these patients likely resemble the patients in the BALATON and COMINO trials that required four-weekly treatments at 68 weeks. The reality is that when clinical trials report that a new agent may have greater durability and efficacy, while maintaining non-inferior visual and anatomical outcomes compared with existing therapies, it is on average rather than being the case for every patient.

RVO can be a challenging condition to treat. We certainly welcome reported maintenance of treatment outcomes with injection intervals of 12 or more weeks in around half of trial patients using faricimab. The true impact on the broader cohort, which is typically managed with less frequent injections in routine care and includes many eyes not eligible for trials, will only become clear through observational research using high-quality data as new agents are integrated into practice.

Dr Adrian Hunt MBBS MBioMedE FRANZCO is a general ophthalmologist and medical retina sub-specialist.

Since completing a retinal fellowship at Westmead Hospital, he has been a prominent contributor to the field, focussing on the management of age-related macular degeneration, diabetic eye disease, and vascular retinopathies. He continues to research this important field as a University of Sydney PhD candidate under the supervision of Professor Mark Gillies at the Save Sight Institute at Sydney Eye Hospital. His thesis examines real world outcomes of intravitreal treatment for retinal vein occlusion. He has authored, presented, and reviewed international and national peer reviewed research.

Dr Hunt is also a Senior Staff Specialist at Westmead Hospital and has a practice at Miranda, in Sydney’s south. He also consults out of Sydney’s CBD and Bathurst, in the Central Tablelands of NSW.

References available at mivision.com.au.