Adjuvant mecobalamin for anti-VEGF therapy in retinal vein occlusion: protocol of a single-centre, randomised, double-blinded controlled trial

Abstract

Introduction

Retinal vein occlusion (RVO), a common retinal vascular disease, is frequently treated with anti-vascular endothelial growth factor (anti-VEGF) agents as first-line therapy. However, anti-VEGF monotherapy lacks neuroprotective effects, primarily targets vascular leakage and neovascularisation, and requires frequent long-term injections that impose substantial economic burdens. Therefore, combined therapeutic strategies that address both vascular pathology and neural damage are being explored. This article describes the protocol for evaluating mecobalamin (a widely used neuroprotective drug) in combination with anti-VEGF for the treatment of macular edema (ME).

Methods and analysis

The study is a randomised, double-blind, placebo-controlled clinical trial that will enrol 120 patients with RVO from the First Affiliated Hospital of Chongqing Medical University. Participants will be randomly assigned (1:1) to an experimental group and a control group. The experimental group will receive intravitreal injections of conbercept plus oral mecobalamin for 6 months, while the control group will receive the same conbercept regimen plus a placebo for 6 months. All patients will undergo 1 year of follow-up after initial treatment, with visits at 1, 3, 6, 9 and 12 months. The primary outcome is the change in central subfield thickness from baseline to 1 year post initial treatment. Secondary outcomes consist of change in the best-corrected visual acuity from baseline over time, capillary density, cone photoreceptor distribution characteristics, mean light sensitivity and fixation stability, serum vitamin B12 levels, the number of treatments, frequency of injection (times per year), interval time, incidence and severity of adverse events (AEs) and serious AEs. Statistical analyses will use appropriate parametric and non-parametric tests for group comparisons, correlation analyses, and multivariate modelling.

Ethics and dissemination

This study has been approved by the Research Ethics Committee of The First Affiliated Hospital of Chongqing Medical University (No. 2025–387-01). The results will be disseminated through conference presentations and publication in peer-reviewed international journals.

Trial registration number

NCT07133438.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study uses a rigorously designed randomised, double-blind, placebo-controlled trial methodology, providing high-quality evidence for evaluating the combined efficacy of mecobalamin and anti-vascular endothelial growth factor therapy in retinal vein occlusion (RVO).

• The trial adopts a novel dual-target strategy integrating neuroprotection and vascular intervention, which may better address the multifactorial pathogenesis of RVO-related macular oedema compared with conventional monotherapy.

• A comprehensive set of functional and structural outcome measures—including best-corrected visual acuity, central subfield thickness, capillary density, photoreceptor metrics and microperimetry—enables multidimensional assessment of treatment response and potential neurovascular unit repair.

• As a single-centre study with a relatively limited sample size, the generalisability of the findings may be constrained, and longer-term outcomes beyond the 12-month follow-up period remain to be investigated.

Introduction

Retinal vein occlusion (RVO) is the second leading cause of vision loss resulting from retinal vascular disease, with a prevalence of approximately 0.6%–1.6%.^{1 2} Macular edema (ME) secondary to RVO (RVO-ME) is a common cause of vision loss. RVO induces hypoxia and ischaemia, leading to neuronal loss and increased vascular permeability.³ Disruption of endothelial barrier function allows fluid leakage into the parenchyma, resulting in oedema. Hypoxia elevates vascular endothelial growth factor (VEGF) levels, which promotes vasodilation and increases vascular permeability.

Therefore, anti-VEGF injections are widely used as first-line treatment for RVO-ME, achieving remarkable therapeutic effects.⁴ The landmark phase III Ranibizumab for the Treatment of Macular Edema following Branch Retinal Vein (BRAVO)⁵ and Ranibizumab for the Treatment of Macular Edema after Central Retinal Vein Occlusion (CRUISE)⁶ trials validated the efficacy of ranibizumab for RVO-related ME in distinct disease subtypes: the BRAVO (Bevacizumab Ranibizumab Aflibercept Versus Observation) trial focused on branch retinal vein occlusion (BRVO)-related ME, while the CRUISE (Central Retinal Vein Occlusion Study of Intravitreal Eylea) trial targeted central retinal vein occlusion (CRVO)-induced ME. Both multicentre, double-masked controlled trials demonstrated robust visual and anatomical benefits of ranibizumab. Specifically, ranibizumab treatment yielded mean best-corrected visual acuity (BCVA) gains of approximately 16–18 letters in BRVO patients and 12–15 letters in CRVO patients at 6 months, alongside substantial macular thickness reduction, with average central foveal thickness decreasing by over 300 µm and 400 µm in BRVO and CRVO cohorts, respectively, far exceeding the improvements observed in sham groups. However, VEGF monotherapy has inherent therapeutic limitations despite its well-documented long-term vascular efficacy in RVO-ME management. Although anti-VEGF agents effectively resolve ME and improve short-term and long-term visual outcomes in most patients, clinical outcomes are compromised by high re-treatment burden and suboptimal neural functional recovery. Importantly, anti-VEGF therapy primarily targets vascular hyperpermeability but does not directly address the underlying neurodegenerative cascades secondary to RVO, including photoreceptor impairment and Müller cell dysfunction, which may persist and constrain full visual restoration. Consequently, combined strategies synchronising vascular leakage repair and neuroprotection have become an international research priority.⁷ To clarify and standardise the trial intervention, the present study adopts conbercept as the exclusive anti-VEGF regimen. Conbercept is a novel, long-acting recombinant fusion protein with broad-spectrum anti-angiogenic properties, which simultaneously inhibits VEGF-A, VEGF-B and placental growth factor (PlGF). Differing from single-target anti-VEGF drugs, its multi-target mechanism comprehensively suppresses pathological retinal angiogenesis and vascular hyperpermeability, exerting sustained anti-oedema effects with a prolonged half-life, which optimises clinical treatment outcomes and reduces injection burden.

As a typical neurotrophic drug, mecobalamin can repair nerve tissue and accelerate axonal regeneration and myelination.⁸ Mecobalamin is a derivative of vitamin B12 and is more easily absorbed by nerve cells than vitamin B12 itself.⁹ As a methyl donor, it facilitates homocysteine (Hcy) to methionine conversion, supporting nucleic acid, protein and phospholipid synthesis critical for myelin formation. This process is essential for the formation and repair of nerve myelin. Mecobalamin accelerates lipid, protein and nucleic acid metabolism in neurons, enhancing nerve function recovery and alleviating neurological symptoms. A previous study found that oral vitamin B1 and mecobalamin could improve corneal nerve fibre density.¹⁰ A randomised controlled trial (RCT) showed that intramuscular injection of mecobalamin 0.5 mg/day significantly improved corneal nerve length and density in diabetic patients, which may be an effective treatment for diabetic peripheral neuropathy and can significantly improve nerve function and quality of life in patients.¹¹ Thus, mecobalamin has been shown to attenuate oxidative stress in animal models, but clinical evidence for its translation into RVO therapy remains insufficient. Notably, VEGF exhibits dual roles as both a permeability factor and neuroprotectant. Excessive VEGF inhibition may exacerbate retinal ganglion cell apoptosis. Although mecobalamin promotes neural repair, its spatiotemporal synergy with anti-VEGF in neurovascular unit (NVU) recovery remains uncharacterised. Critically, there is a direct mechanistic link between mecobalamin-mediated neural repair and reduced central subfield thickness (CST), which stems from the integrated structure and function of the retinal NVU in RVO. RVO-induced vascular injury disrupts NVU homeostasis, impairing the crosstalk between retinal neurons, glial cells and vascular components. As core NVU constituents, Müller glial cells closely interact with neurons and maintain blood–retinal barrier (BRB) stability. Neural and glial damage in RVO triggers inflammation, mitochondrial dysfunction and persistent BRB breakdown, which sustains ME and elevates CST. Adjuvant mecobalamin complements anti-VEGF therapy by targeting neural injury to break the NVU dysfunction-oedema cycle, validating CST as an appropriate primary outcome for this trial.

This prospective, double-blind, RCT evaluates the synergistic effects of mecobalamin combined with anti-VEGF therapy on NVU repair in patients with ME secondary to both BRVO and CRVO using dual-targeted ‘vascular intervention+neuroprotection’ to overcome limitations of current approaches. Current assessment of therapeutic efficacy for RVO-ME relies primarily on optical coherence tomography (OCT)-derived CST, with few tertiary centres using OCT angiography (OCTA), multifocal electroretinography (mfERG) or adaptive optics (AO). Existing metrics inadequately characterise NVU function or anti-VEGF effects on non-vascular components (glia and photoreceptors). The absence of standardised NVU repair quantification further impedes cross-study comparisons. To address this, we propose a multimodal ‘OCT structural tomography+OCTA flow stratification+AO cellular morphology+microperimetry functional mapping+mfERG functional localisation’ assessment to establish a new therapeutic standard. The primary objectives of the registry are to:

1. Evaluate the effect of mecobalamin in combination with anti-VEGF therapy on the improvement of CST and BCVA in patients with RVO-ME.

2. Validate mecobalamin’s synergistic contribution to NVU repair using multimodal imaging.

Therefore, this trial aims to establish high-level evidence for RVO management by combining mecobalamin with anti-VEGF therapy. Our results, alongside advances in multimodal therapeutics, will inform future multi-pathway treatments aimed at reducing recurrence rates and improving long-term visual prognosis.

Trial design

This study is a randomised, double-blind, placebo-controlled clinical trial with two parallel arms. The 120 eligible participants with RVO will be randomly assigned in a 1:1 ratio to either the experimental group or the control group. To ensure balance of the key prognostic factor, randomisation will be stratified by RVO subtype (CRVO vs BRVO). To ensure blinding, all study medications and placebo are identical in packaging, appearance and smell. Participants and investigators (including clinicians, outcome assessors, imaging technicians and data analysts) will remain blinded to treatment assignment throughout the trial.

The approved protocol is identified as V.1.0, dated 22 April 2025. The present manuscript incorporates minor amendments made in response to peer review, such as clarifications to statistical methods and updates to references, which do not alter the study design or participant risk profile. Any future amendments affecting study conduct, participant safety or scientific validity to this protocol will be documented, submitted for ethical approval and result in an updated version number and date. Researchers must ensure that they are using the most recent approved version of the protocol.

Methods and analysis

Study type and study design

Prospective, single-centre, two-arm, double-blind, placebo-controlled clinical trial registered under the ClinicalTrials.gov Identifier: NCT07133438.

The full trial protocol and detailed statistical analysis plan can be accessed. Any amendments to these documents will be updated in the trial registry in a timely manner.

Protocol compliance

This protocol is developed in accordance with the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines to ensure transparent and standardised reporting. The SPIRIT 2013 checklist is shown in online supplemental file 1.

Study setting and participants

This study is an ongoing prospective, double-blind, randomised controlled intervention study of RVO, conducted at the First Affiliated Hospital of Chongqing Medical University between 1 August 2025 and 31 July 2027. The purpose of this study is to evaluate the synergistic effect of mecobalamin combined with the anti-VEGF drug (conbercept) on the repair of retinal NVU in patients with RVO-ME. The study adheres to the Declaration of Helsinki and Ethical Guidelines for Medical and Biological Research Involving Human Subjects. All participants will provide written informed consent. An opt-out approach will be implemented where applicable. Primary outcomes will be assessed at 12 months post-initial treatment.

Eligibility

Inclusion criteria

Patients will be included in the study if they satisfy all the following criteria:

1. Diagnosis of RVO meeting the international diagnostic criteria,^{4 12} age 18–80 years, gender not limited.

2. Treatment-naive RVO-ME (no prior anti-VEGF, glucocorticoid or laser therapy).

3. The CST (measured from retinal pigment epithelium (RPE) to the inner limiting membrane inclusively) is confirmed to be ≥300 µm by OCT.

4. The baseline BCVA (Early Treatment Diabetic Retinopathy Study (ETDRS) letter count) was 20/400 to 20/40 (34–78 letters).

5. Signed informed consent and ability to comply with follow-up.

Exclusion criteria

Patients will be excluded from the study if any of the following criteria apply:

1. Combined with other eye diseases that cause ME (such as diabetic retinopathy (DR) and uveitis).

2. Media opacities affecting imaging (such as severe cataract and vitreous haemorrhage).

3. Prior anti-VEGF, steroid or macular laser therapy.

4. Systemic use of glucocorticoids or immunosuppressants within 3 months.

5. Uncontrolled systemic disease (hypertension, diabetes and hepatic/renal dysfunction); pregnant or lactating women.

6. Allergy to mecobalamin or conbercept.

7. Unable to cooperate with examinations or follow-up or participation in other clinical trials within 1 year.

Eligibility criteria for intervention providers

Surgeons/injectors:

1. Must be board-certified ophthalmologists with retinal specialisation.

2. Minimum of 2 years experience performing intravitreal injections.

3. Demonstrated proficiency with anti-VEGF injection techniques.

4. Performed at least 100 intravitreal injections in the past 12 months.

5. Certified in Good Clinical Practice (GCP) guidelines.

6. Trained on and compliant with study-specific injection protocol.

Patient withdrawal

Participants will be automatically withdrawn from the study if they meet any of the following criteria: (1) miss three or more consecutive follow-up visits; (2) have poor treatment compliance or refuse to continue participation; and (3) develop a new systemic disease that can affect the study outcome. Additionally, patients will be excluded from the final per-protocol data analysis if they violate the treatment regimen or lack essential data preventing the assessment of the trial’s efficacy and safety. Participants have the right to withdraw from the study at any time, for any reason, without affecting their future medical care. Should a participant decide to withdraw consent for continued participation or use of their personal data, the investigator will make reasonable efforts to contact the participant to discuss the primary reasons for withdrawal. This information will be documented confidentially in the study records.

Recruitment procedure

Potential eligible patients will be identified during the first RVO medical visit at the ophthalmology outpatient department. Recruitment will be performed by ophthalmologists. After the assessment of eligibility and the provision of informed consent, standard medical treatment will commence immediately. The reasons for refusal will be documented for subjects who decline to participate. Meanwhile, participants will be randomised to corresponding follow-up treatments.

Screening procedures and baseline evaluation

Basic information of potential participants will be collected through interview records, including age, gender, medical history, current medications and details regarding ocular condition.

The following assessments will be completed in the programme to assess eligibility and baseline levels:

1. BCVA and intraocular pressure (IOP): following manifest refraction, BCVA will be assessed for both eyes according to the ETDRS visual acuity protocol¹³ . IOP will also be measured in both eyes.

2. Ophthalmic examination: after measuring BCVA and IOP, investigators will perform a standard slit-lamp biomicroscopy examination and evaluate the fundus under mydriasis induced by topical eye drops.

3. Swept Source Optical Coherence Tomography (SS-OCT): SS-OCT images will be acquired using the TowardPi system (TowardPi Medical Technology, Beijing, China).

4. OCTA: OCTA images will be obtained using the TowardPi swept-source OCTA system.

5. Microperimetry: this will be performed using the Nidek MP-3 microperimeter (Nidek Co, Japan). Examinations will be conducted under standard ambient light conditions with undilated pupils. Mean light sensitivity (in dB) and fixation stability (defined as the percentage of fixation points within the 2° and 4° diameter circles) will be assessed.

6. Electroretinography (ERG): the Roland RETIport/scan 21 system (Roland Consult, Germany) will be used to perform ERG in accordance with the standards of the International Society for Clinical Electrophysiology of Vision (ISCEV). The testing protocol will include the dark-adapted 0.01 ERG, dark-adapted 3.0 ERG, dark-adapted 3.0 oscillatory potentials, light-adapted 3.0 ERG and light-adapted 3.0 flicker ERG. Amplitude and implicit time will be recorded for all responses.

7. AO: macular cone cell mosaic metrics will be performed using the rtx1 AO fundus camera (Imagine Eyes, Orsay, France). Images will be acquired centred on the fovea. Cone photoreceptor metrics, including density, spacing, dispersion and regularity, will be analysed within the central 2° retinal area using the manufacturer’s software (AO Detect, V.1.3.1).

Each participant will be assigned a unique subject identification number on providing written informed consent. This identifier will serve as the primary reference for all data collected throughout the study period.

Randomisation and masking

Once an eye is enrolled in the study, the research centre coordinator will assign an alphanumeric ID code to it. Each study eye will be randomly assigned to a treatment group 2 days before the initiation of treatment. Randomisation will be stratified by RVO subtype (CRVO vs BRVO) to ensure balanced distribution of this key prognostic factor between the two treatment arms. To ensure prognostic balance, randomisation will be stratified by RVO subtype and performed using block randomisation within each stratum via a computer programme (Statistical Package R, V.3.6; R Foundation for Statistical Computing). The system will automatically assign the next available treatment code based on the predetermined sequence. All study participants will be blinded to their intervention assignment. The randomisation list will be securely stored in a password-protected database accessible only to the independent statistician and the unblinded pharmacy personnel. The drug and the placebo are identical in packaging, appearance and smell. Mecobalamin and placebo were assigned by an independent pharmacy according to random codes, and evaluators were blinded to group allocation. All study medications and placebo will be identically labelled to ensure adequate blinding of investigators and patients. Both participants and study personnel (including principal investigators, assessor physicians, imaging technicians, central reading centre graders, study coordinators and BCVA examiners) will remain blinded to treatment assignment throughout the study. The allocation concealment process will be monitored by the data safety and monitoring board. Emergency unblinding will be permitted only in medical emergencies where knowledge of the treatment assignment is essential for clinical management. All unblinding events will be documented and reported to the ethics committee. This randomisation methodology ensures allocation concealment, minimises selection bias and maintains the scientific integrity of the trial throughout the study period.

Interventions

Patients will be randomly assigned (1:1) to either the experimental group or the control group. Patients in the experimental group will receive conventional anti-VEGF therapy (intravitreal injections of conbercept 0.5 mg initially once a month for 3 months, followed by on-demand therapy based on recurrence criteria) along with oral mecobalamin capsules 0.5 mg three times daily for 6 months. Patients in the control group will receive the same anti-VEGF treatment as the experimental group and will take a pharmacologically inert placebo three times daily for 6 months (figure 1). Recurrence criteria include: an increase in CST measured by OCT of ≥50 µm compared with the minimum recorded value; a decrease of ≥5 ETDRS letters in BCVA; or the presence of new or increased intraretinal and subretinal fluid.

Figure 1. Work flow. AO, adaptive optics; BCVA, best-corrected visual acuity; IOP, intraocular pressure; mfERG, multifocal electroretinography; OCT, optical coherence tomography; OCTA, optical coherence tomography angiography.

Follow-up visits

All eyes will be examined at baseline (the first medical visit), on the day of initial treatment, and at 1, 3, 6, 9 and 12 months following the initial treatment. The information required at each follow-up visit is included in table 1.

Table 1. Assessment protocol for the current study from V1 to V6.

Visit of data entry	V1	V2	V3	V4	V5	V6
Meets enrolment/exclusion criteria	X
Month		1	3	6	9	12
Days	X	30 (± 7) days after first injection	90 (±7) days after first injection	180 (±7) days after first injection	270 (±7) days after first injection	360 (±7) days after first injection
Consent form	X
Demographic data	X
Systemic and ocular disease history	X
BCVA	X	X	X	X	X	X
IOP	X	X	X	X	X	X
OCT	X	X	X	X	X	X
OCTA	X	X	X	X	X	X
AO fundus camera	X	X	X	X	X	X
mfERG	X	X	X	X	X	X
Microperimetry	X	X	X	X	X	X
Serum vitamin B12	X	X	X	X	X	X
Accompanying medication	X	X	X	X	X	X
Data record	X	X	X	X	X	X
Treatment programme records		X	X	X	X	X
Adverse events		X	X	X	X	X
Serious adverse events		X	X	X	X	X

AO, adaptive optic; BCVA, best-corrected visual acuit; IOP, intraocular pressure; mfERG, multifocal electroretinography; OCT, optical coherence tomography; OCTA, optical coherence tomography angiography.

Timeline

Study start date: 1 August 2025.

Patient follow-up: 12 months after inclusion.

Planned end date: 31 July 2027.

Drug and storage

Conbercept is supplied as a sterile, preservative-free solution in single-use glass vials. Each vial contains 0.5 mg of conbercept in a 0.2 mL iso-osmotic buffer solution (pH ~7.0), intended for intravitreal injection. Mecobalamin capsules are provided as hard gelatin capsules, each containing 0.5 mg of mecobalamin as the active ingredient. The placebo capsules are identical in appearance, packaging and excipient composition, lacking only the active mecobalamin component.

The storage and supervision of investigational products will comply with the drug characteristics and ICH-GCP. The designated investigator will be responsible for the receipt of the investigational products, ensuring that they are handled safely and appropriately, and stored in a designated secure location. A continuous temperature monitoring system will be implemented at the study site to record storage conditions, with logs readily available for inspection by study monitors. All investigational products must be stored in accordance with manufacturer specifications. Throughout the study, the designated personnel will maintain a drug accountability log documenting the transportation, distribution and return of all investigational products.

Concomitant and rescue treatment

All medications taken by participants from the date of registration (screening visit) until the end of the study must be recorded on the concomitant medication case report form (CRF) page, with the exception of standard medications required per protocol for ophthalmic procedures (eg, fluorescein, mydriatic eye drops, topical antibiotics, topical anaesthetics and ocular corticosteroids). Each entry shall include the start and end dates of medication use, as well as the reason for prescription. If during the study a participant requires additional treatment due to severe visual deterioration or safety concerns, study participation must be discontinued prior to initiating such treatment. Any additional ocular treatments will be captured in the electronic CRF (eCRF). Monitoring of the fellow eye must be performed in accordance with routine clinical practice, and any adverse events (AEs) will be recorded in the eCRF.

Monitor adherence

These strategies are designed to maintain high adherence rates while ensuring participant safety and data quality throughout the trial period.

Participant education and engagement:

1. Comprehensive explanation of study procedures and importance of adherence during the informed consent process.

2. Provision of written materials outlining dosing schedules and visit requirements.

3. Regular reminders for upcoming visits via telephone calls or text messages.

4. Designated research coordinator available to address participant concerns throughout the study.

Medication adherence monitoring:

1. Participants will receive medication diaries to record daily intake of study capsules.

2. Monthly pill counts will be conducted by research staff during study visits.

3. Participants will be instructed to return all used and unused medication packaging at each visit.

Visit compliance:

1. Flexible scheduling options for study visits to accommodate participant availability.

2. Reminder system for upcoming visits initiated 1–2 weeks prior to scheduled appointment.

3. Transportation assistance or reimbursement if needed to reduce barriers to participation.

Study completion and post-study treatment

For patients who complete the 1-year follow-up, the study reaches its endpoint at 1 year after the initial treatment. No additional procedures or treatments will be conducted beyond those outlined in the study protocol. Participants will be advised to resume their regular follow-up care with their private physicians. Additionally, patients will be excluded from the final per-protocol data analysis if they violate the treatment regimen or lack essential data preventing the assessment of the trial’s efficacy and safety.

Outcomes

Primary outcomes

The primary outcome is the change in CST from baseline to 1 year after initial treatment.

Secondary outcomes

The secondary outcomes include:

1. The changes in BCVA over time.

2. Capillary density, the changes in capillary density will be assessed as vessel density (%) in the superficial and deep retinal capillary plexuses using OCTA. Scans will be acquired with a standardised 3×3 mm or 6×6 mm pattern centred on the fovea, processed with projection artefact removal, and quantified within ETDRS grid subfields. All analyses will undergo quality control and centralised grading.

3. Cone photoreceptor distribution characteristics, the changes in cone photoreceptor mosaic metrics including density (cones/mm²), inter-cone distance (µm) and hexagonal packing (%) will be assessed using AO ophthalmoscopy centred at the fovea. Images will be processed with semi-automated cone identification algorithms and compared against normative databases.

4. Mean light sensitivity and fixation stability, the changes in mean light sensitivity (dB) and fixation stability (%P1, BCEA 63%) will be assessed using microperimetry (68-point central 10° grid, 4–2 staircase strategy) under standardised conditions. Testing requires pupil dilation ≥6 mm, with validity criteria including <15% false-positive rate and <20% fixation losses. All analyses will undergo central reading centre verification.

5. Serum vitamin B12 levels.

6. Number of treatments.

7. Frequency of injection (times per year), interval time, and incidence and severity of AEs and serious AEs (SAEs). Cumulative intravitreal injections over 12 months, injection frequency (annualised rate, injections/year) and mean injection interval (average days±SD between consecutive injections, excluding non-protocol treatments) will also be evaluated.

Safety assessment

Safety evaluations will include monitoring of vital signs, ophthalmic examinations, laboratory assessments, and the type, frequency, and severity of all AEs. AEs and SAEs are defined according to established criteria used in large-scale clinical trials of intravitreal anti-VEGF injections.^14–16 Safety outcomes include the incidence and severity of ocular and non-ocular AEs, such as elevated IOP, systemic AEs (eg, headache and gastrointestinal symptoms), and local ocular complications (eg, endophthalmitis and vitreous haemorrhage). The number and percentage of AEs and SAEs in the overall study population will be calculated. All SAEs, complications and unexpected events will be reported. Any detected AE will be followed until resolution or until judged to be permanent. Periodic assessments will be conducted to evaluate the severity of AEs, their suspected relationship to the study intervention, required treatments and changes in outcome. If any, all of them will be extracted, kept detailed records, and reported to the ethics committee and the trial steering committee in regular progress reports.

To ensure participant safety, all SAEs, regardless of causality, must be reported within 24 hours of the investigator becoming aware of their occurrence, after the participant has provided informed consent. Any SAE occurring within 30 days after the final study visit should be reported only if the investigator suspects a causal relationship with the study treatment. Any additional details regarding an SAE—including complications, progression or recurrence of the initial event—must be submitted as a follow-up to the initial report within 24 hours of the investigator becoming aware of the update. If an SAE occurs at a different time or is considered entirely unrelated to a previously recorded event, it shall be documented and reported as a new incident.

Data collection and management

Data collection is the responsibility of the clinical trial staff under the supervision of the site investigator. Researchers are responsible for ensuring the accuracy, completeness, readability and timeliness of the reported data. Source data will be recorded in the patient’s medical record and subsequently entered into a secure, web-based electronic data capture (EDC) system. The EDC system will include built-in validation checks to reduce data entry errors and ensure consistency. All eCRF entries will be backed by source documentation, and any data modifications will be tracked via an audit trail. Trained study personnel will perform regular source data verification (SDV) to confirm the accuracy and completeness of the collected data. A monitoring plan will be implemented to review data quality periodically, with queries resolved in a timely manner. All study data will be de-identified using a unique participant identification number to protect confidentiality. Electronic data will be stored on encrypted, password-protected servers with regular backups, while any paper-based records will be maintained in locked filing cabinets with restricted access.

All personnel involved in data collection will undergo protocol-specific training and certification for standardised procedures. All data will be kept strictly confidential in compliance with applicable regulations.

To maximise participant retention and ensure complete follow‐up throughout the trial, a multi-faceted strategy will be implemented. This includes sending reminders via telephone, SMS or email prior to each scheduled visit, offering flexible appointment times, including weekends and evenings, to accommodate the schedules of participants, and providing transportation reimbursement or assistance to reduce barriers to attendance. A dedicated research coordinator will maintain regular contact with participants to address concerns and build rapport. Additionally, financial compensation for time and effort will be offered at each completed visit, and culturally appropriate engagement strategies such as festival greetings will be used to maintain participant motivation and commitment to the study.

Data analysis

Sample size

The target sample size is 120 patients (1 eye per patient) with ME secondary to RVO (including both BRVO and CRVO subtypes). In this trial, all enrolled BRVO and CRVO patients will be pooled for the primary analysis, with randomisation strictly stratified by RVO subtype to balance the subtype distribution between the two study groups. This pooled analysis strategy is methodologically feasible and clinically reasonable. Additionally, mecobalamin-mediated neuroprotective, anti-inflammatory and antioxidant effects are universal across RVO subtypes, with no published evidence indicating divergent treatment responses between BRVO and CRVO. Therefore, pooling the two subtypes ensures adequate statistical power for the primary endpoint while avoiding underpowered subgroup analyses caused by excessive subtype stratification. The calculation is primarily based on detecting a clinically meaningful difference in the primary anatomical outcome, CST measured by OCT.^17–19

SD: based on pivotal RVO-ME trials, the SD of CST change from baseline at 12 months ranges from 40 to 60 μm in recent phase III trials in BALATON/COMINO¹⁸ to 60–100 μm in the SCORE2 studies. To ensure adequate power under a conservative assumption that accounts for the heterogeneity of RVO subtypes and the exploratory nature of our novel combination therapy, we adopted a pooled SD of 100 μm.

Clinically meaningful difference: a between-group difference of 50 μm in CST is considered clinically relevant. This threshold is grounded in the latest evidence from SCORE2 Report 24,¹⁷ which demonstrated a non-linear relationship between CST and visual acuity in CRVO/HRVO patients treated with anti-VEGF therapy. The estimated inflection points at which the CST–visual acuity association changes from positive to negative range from 217 to 256 μm. A 50 μm difference—exceeding the entire width of this optimal range (39 μm)—is sufficient to shift a patient’s CST from a prognostically unfavourable category (either >300 µm (‘thick’) or ≤216 µm (‘thin’), as shown in SCORE2 AJO 2024)¹⁹ into the optimal range associated with the best expected visual acuity.

Using a two-sided α=0.05, power=80%, δ=50 µm and σ=100 µm, the required sample size is 63 patients per group (126 total). Accounting for a 10% dropout rate over 12 months, the target would be 140 patients. However, a total of 120 patients (60 per group) provides sufficient power (>80%) under the stratified, covariate-adjusted analysis framework, considering that:

• Randomisation will be stratified by RVO subtype (CRVO vs BRVO).

• Analysis of covariance (ANCOVA) with baseline CST and BCVA as covariates will increase statistical efficiency by 10%–15%.

• Recent high-quality RVO-ME trials (BALATON/COMINO, n=553–729; SCORE2, n=350) have successfully detected significant differences with similar or smaller effect sizes.

This sample size also provides >80% power to detect a 0.1 logMAR (5-letter) difference in BCVA (SD=0.2 logMAR), consistent with the clinically meaningful visual gains observed in recent trials.

Statistical methods

Outcomes will be analysed on trial completion. All randomised patients will be included in the efficacy analysis based on the intention-to-treat principle, grouped according to their original treatment assignment. The change in CST from baseline to month 12 (primary outcome) will be compared between the two treatment groups using an ANCOVA model. The model will include the treatment group as a fixed effect, with adjustment for the baseline CST value and the randomisation stratification factor and RVO subtype (CRVO vs BRVO). Adjusted mean differences between groups, along with their 95% CIs and two-sided p-values, will be reported. A p value <0.05 will be considered statistically significant.

Longitudinal continuous measures (BCVA, capillary density, AO metrics, microperimetry, serum B12) will be analysed using linear mixed-effects models (LMM). Models will include fixed effects for treatment group, visit time, RVO subtype and the treatment-by-time interaction, with the baseline value of the respective measure as a covariate and participant as a random effect.

Count and time-to-event measures (number of injections and injection interval) will be analysed as follows: the cumulative number of injections over 12 months will be compared using a negative binomial regression model, adjusting for RVO subtype; injection intervals will be analysed using survival analysis methods (eg, Kaplan-Meier estimates with log-rank test or Cox proportional hazards model).

Safety outcomes (AEs and SAEs) will be summarised by incidence and severity for each treatment group and compared using Fisher’s exact test.

To control the inflated type I error risk due to multiple secondary outcomes and time points, a hierarchical testing strategy will be used. Statistical inference for secondary outcomes will proceed only if the primary outcome analysis is statistically significant. Within key secondary outcomes measured at multiple time points (eg, BCVA at months 1, 3, 6, 9 and 12), the family-wise error rate will be controlled using the Bonferroni method or a pre-specified closed testing procedure.

Pre-specified subgroup analyses will be performed to explore the consistency of the treatment effect within the strata of the randomisation factor: RVO subtype (CRVO vs BRVO). This will be tested by including a treatment-by-subgroup interaction term in the respective statistical models (ANCOVA for the primary outcome and LMM for longitudinal outcomes). Results from subgroup analyses will be interpreted as exploratory and hypothesis-generating.

Baseline characteristics will be summarised descriptively. Between-group differences for continuous and categorical variables will be compared using independent samples t-tests (or Mann-Whitney U tests if non-normal) and χ² tests (or Fisher’s exact test), respectively, for descriptive purposes only.

All statistical analyses will be performed using IBM SPSS Statistics (V.27.0 or higher; IBM Corp.). A two-sided significance level of α=0.05 will be used unless otherwise adjusted for multiple comparisons.

Access to documents

Any deviations from the pre-specified statistical analysis plan will be reported in the final study report. Throughout the study, monitoring will emphasise the following aspects:

1. Adherence to patient inclusion and exclusion criteria and proper completion of informed consent forms (ICFs).

2. Completeness of documentation throughout the participant’s involvement in the study.

3. Storage, dispensing and accountability of investigational products in accordance with protocol specifications.

4. Documentation of AEs and conduct of safety assessments.

5. Accuracy of all entries in the CRFs.

6. Overall compliance with the study protocol and GCP guidelines.

Discussion

RVO is a vascular occlusive disease caused by thrombus formation within the retinal venous system. Characteristic features include retinal venous tortuosity and obstruction, increased vascular permeability and inflammatory activation. Following occlusion, retinal ischaemia induces extensive capillary non-perfusion, upregulating inflammatory cytokines such as VEGF, tumour necrosis factor-alpha and interleukin-1β.²⁰ VEGF drives neovascularisation and enhances capillary permeability, resulting in fluid leakage and ME. The proangiogenic effects of VEGF can be blocked by antagonistic manipulations targeting any stage of its signalling pathway. Current therapeutic strategies focus on pharmacological VEGF suppression or inhibition of VEGF/VEGFR interactions using monoclonal antibodies. Various anti-VEGF drugs for the treatment of ME secondary to RVO such as ranibizumab, bevacizumab, aflibercept and conbercept have emerged in recent years. Conbercept has attracted wide attention for its considerable therapeutic potential.²¹

However, the short intraocular half-life of these agents and the need for frequent injections to maintain effective drug concentrations impose a substantial psychological and economic burden on patients.²² Intraocular injection of anti-VEGF agents also carries procedural risks, including endophthalmitis, vitreous haemorrhage, retinal detachment, elevated IOP and thrombosis.²³ Additionally, some patients develop tolerance to long-term anti-VEGF therapy and experience suboptimal treatment outcomes. Long-term follow-up (>12 months) has revealed that approximately 40% of patients require continuous on-demand injections to sustain therapeutic effects, with some exhibiting anatomic-functional dissociation.^{24 25} This indicates that suppression of vascular leakage alone is insufficient to reverse retinal neural damage.²⁶

Vascular injuries in RVO disrupt communication among endothelial cells, pericytes, glial cells and neurons. The retinal NVU comprising neurons, RPE, macroglia (Müller cells and astrocytes), microglia, vascular endothelial cells and mural cells (pericytes and vascular smooth muscle cells) collectively regulates retinal energy metabolism and vascular homeostasis.²⁷ Disruption of the NVU leads to breakdown of the BRB, oedema, inflammation, retinal neuronal atrophy and visual dysfunction. Within the NVU, Müller cells serve as principal glial elements, maintaining intimate contact with retinal neurons to provide metabolic support, signal transduction and structural stabilisation of the BRB.²⁸ Concurrently, pericytes and vascular smooth muscle cells cooperate with endothelial cells on the basement membrane to preserve vascular integrity.²⁹ When the structure or function of the retinal NVU is compromised, microangiopathy, oxidative stress and neurologic damage occur, contributing to retinal degenerative diseases. Consequently, therapeutic strategies that simultaneously target vascular repair and neuroprotection are urgently needed. Vitamin B12 (mecobalamin) has emerged as a promising candidate for retinal neuroprotection through multiple mechanistic pathways. Elevated Hcy—a neurotoxic and vasculotoxic metabolite—is a well-established risk factor for various retinal disorders, including age-related macular degeneration, glaucoma and DR.^30–32 The metabolism of Hcy to methionine is dependent on vitamin B12 and folate; deficiency of vitamin B12 leads to hyperhomocysteinaemia, which induces oxidative stress, endothelial dysfunction and RPE injury, and promotes pro-inflammatory cascades.^{33 34} In the retina, hyperhomocysteinaemia has been shown to reduce retinal blood flow, diminish central retinal artery diameter and compromise the BRB, ultimately triggering retinal capillary endothelial cell apoptosis and neuronal degeneration.³⁵ Vitamin B12 supplementation effectively lowers circulating Hcy levels, thereby mitigating these deleterious effects. Beyond Hcy-lowering, vitamin B12 exerts direct neuroprotective and vasoprotective actions within the retina. As an essential cofactor in one-carbon metabolism, vitamin B12 regulates DNA methylation and epigenetic expression of genes involved in oxidative stress response and inflammation.³⁰ Experimental studies have demonstrated that B vitamins (including B12) are critical for mitochondrial bioenergetics in highly metabolically active retinal cells such as photoreceptors and RPE; deficiency impairs mitochondrial function and accelerates neurodegeneration.^{36 37} Furthermore, vitamin B12 deficiency has been linked to increased release of inflammatory cytokines from RPE cells, disruption of the BRB and retinal neuronal apoptosis—pathological hallmarks directly relevant to NVU injury in RVO. Protective effects of vitamin B12 on retinal neurons are further supported by clinical observations in glaucoma: a prospective study reported that oral B12 supplementation stabilised visual fields in normal-tension glaucoma patients over 4 years of follow-up.³⁸ In DR, meta-analyses have confirmed that lower serum vitamin B12 levels are significantly associated with more severe retinopathy and peripheral neuropathy, suggesting its role in maintaining retinal neural integrity.^{39 40} Emerging evidence strongly supports its therapeutic potential for retinal NVU repair. By targeting vascular pathology (via Hcy reduction, endothelial protection and BRB stabilisation) and neural injury (via mitochondrial support, anti-apoptotic and anti-inflammatory mechanisms), vitamin B12 offers a rational adjunctive strategy to complement anti-VEGF therapy in RVO-ME.

Our study combines mecobalamin with anti-VEGF therapy under a dual-targeted strategy of ‘neuroprotection+vascular intervention’. Through multimodal dynamic assessment (OCT structural tomography+OCTA flow stratification+AO cellular morphology+microperimetry functional mapping+mfERG functional localisation), we reveal for the first time the temporal regulation of cone photoreceptor activation and vascular endothelial barrier restoration by mecobalamin during the anti-VEGF treatment window (baseline to 12 months). This approach overcomes the limitations of traditional single-modality assessments by capturing spatiotemporal patterns of NVU repair. We further clarify mecobalamin’s regulatory mechanisms on ganglion cells, Müller cells and vascular endothelium, providing: (1) clinical-grade evidence for targeted RVO-ME therapy; and (2) a novel efficacy evaluation standard through quantitative NVU profiling.

Ethical and dissemination

Ethical compliance

This study has been approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University (No. 2025–387-01). The trial will be conducted in accordance with the ethical principles originating in the Declaration of Helsinki, consistent with ICH E6 GCP guidelines and applicable regulatory requirements. The study has been registered at ClinicalTrials.gov (NCT07133438).

Informed consent

The process of obtaining informed consent must be thoroughly documented in the subject’s source documents. The investigator is responsible for ensuring that each participant is fully informed of the nature and purpose of the study, as well as the potential risks associated with participation. Each subject shall sign and personally date the written ICF to indicate their voluntary agreement to take part in the study. The investigator will retain the original signed ICF and provide a copy to the participant.

Confidentiality

The confidentiality of all participant data will be strictly protected throughout the trial. Personal information will be collected only when essential, and each participant will be assigned a unique identification number to de-identify all research records. Electronic data will be stored on encrypted, password-protected servers with access restricted to authorised study personnel. Paper documents containing identifiable information will be kept in locked cabinets within secured areas. No personally identifiable information will be shared with third parties unless required by regulatory authorities or for safety reporting, and any shared data will be fully anonymised. All research staff will receive training in data protection measures.

Safety considerations and monitoring

Safety monitoring will instead be performed by the principal investigator and the study team, with oversight by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University. All SAEs will be reported to the ethics committee within 24 hours of awareness.

Given that this trial involves a marketed drug (mecobalamin) used in a new combination regimen and follows a well-established safety profile for both interventions, an independent data monitoring committee will not be constituted. The risks associated with the study interventions are considered low and well-characterised based on extensive clinical experience. Trial conduct will be monitored through regular on-site and centralised reviews by an independent clinical research associate, including verification of informed consent, source data, protocol compliance, drug accountability and safety reporting, in accordance with GCP guidelines.

Patient and public involvement

Patients or the public were not involved in the design, conduct, reporting, or dissemination plans of this research.

Publication policy

On completion of the study and finalisation of the statistical report, the results of this trial will be prepared for submission to a peer-reviewed scientific journal. The study findings will also be presented at relevant national and international conferences to share insights with the scientific community.

Data sharing statement

Anonymised participant data and related study documents will be made available for reasonable research purposes upon request to the corresponding author after the primary trial results are published. Access will be granted following review and under a data sharing agreement. Data will be shared after article publication with researchers who provide a methodologically sound proposal. Proposals should be directed to the corresponding author (dengwenli@hospital.cqmu.edu.cn). Data requestors will need to sign a data access agreement. Data will be shared via a secure data repository (eg, Figshare and Dryad) or upon direct request. In compliance with BMJ’s Tier 2 data policy, the detailed data sharing plan has been incorporated into the trial’s registration record on ClinicalTrials.gov (Identifier: NCT07133438), which was updated on 24 February 2026.

Responsibilities of the investigator and institutional review board/independent ethics committee

Prior to study initiation, the principal investigator must obtain approval or a favourable written opinion from the institutional review board/independent ethics committee (IRB/IEC) regarding the trial protocol, the ICF, any amendments to the consent documents, participant recruitment materials and all other written information provided to subjects.

The principal investigator and the clinical research team are responsible for conducting the study in strict compliance with the approved protocol and GCP guidelines. The investigator must ensure the protection of participants’ rights, safety and well-being throughout the trial, and provide all necessary documentation and updates to the IRB/IEC in a timely manner, including reports of SAEs and protocol deviations. Investigators ascertain that they will apply due diligence to avoid protocol deviations.

Ancillary and post-trial care

During the trial, necessary medical care for any AEs related to the study intervention will be provided free of charge. Participants will receive appropriate treatment and follow-up for any trial-related injuries in accordance with the insurance coverage secured by the sponsor. No long-term post-trial care or additional financial compensation is planned beyond the provisions described in the protocol and ICF. Participants will retain the right to access their anonymised data and may inquire about aggregate study results after the trial’s conclusion.

Financing and insurance

This is an investigator-initiated clinical trial. The China Health Promotion Foundation will provide financial support for the study. No additional funding will be provided to patients for treatment expenses, as the trial involves conventional therapeutic interventions without extra financial subsidies. In accordance with national regulations, the sponsor will ensure that insurance is in place to cover any potential injuries to participants that may be directly associated with the trial’s procedures and/or treatments.

Data deposition and curation

All study data will be handled and retained in compliance with GCP guidelines and applicable regulatory requirements. Source documents, including clinical records, laboratory reports and imaging outputs, will be maintained securely both electronically and in hard copy. Electronic data will be stored in a password-protected database with regular backups and encryption. Hard copies will be kept in locked cabinets within access-controlled premises. CRFs, both eCRF and paper-based, will be completed accurately and promptly by authorised personnel. Any data entry or modification will be tracked via an audit trail. SDV will be performed periodically to ensure consistency between original records and CRFs. All records will be de-identified using a unique subject identification number to protect participant confidentiality. The data will be retained for at least 5 years after study completion or as required by national regulations. Access to the final dataset will be limited to authorised investigators and statutory auditors. Data may be used for further research following ethical approval and provided that participant anonymity is preserved. All research outputs will adhere to the FAIR (Findable, Accessible, Interoperable, Reusable) principles. The curation and preservation of the data will comply with funder policies and ethical standards set by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University.

Protocol amendments

Any modifications to the research protocol that may affect the study’s conduct, participant safety or scientific validity will be classified as substantial amendments. Such amendments include, but are not limited to, changes to study design, intervention procedures, eligibility criteria, outcome measures or sample size. All substantial amendments must be submitted to the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University for review and approval prior to implementation. Minor amendments, such as administrative or logistical adjustments that do not affect participant safety or scientific integrity, will be documented and notified to the ethics committee as required. All approved amendments will be updated in the trial registry (ClinicalTrials.gov) in a timely manner. Participants will be informed of relevant changes where appropriate, and consent will be re-obtained if necessary.