Study setting
Patients with primary open-angle glaucoma will be consecutively recruited from Chinese University of Hong Kong Eye Clinic, Hong Kong Eye Hospital, and Prince of Wales Hospital eye clinic.
Eligibility criteria
Inclusion criteria are as follows: age ≥18 years; best corrected VA ≥20/40; IOP ≤28mmHg; and evidence of progressive RNFL/GCIPL thinning detected by Guided Progression Analysis (GPA) or Trend-based Progression Analysis (TPA) in one or both eyes.17-20 Exclusion criteria are as follows: pathological myopia; diseases that may cause visual field loss or optic disc abnormalities other than glaucoma; inability to perform reliable visual field; suboptimal quality of OCT images; diabetic retinopathy/maculopathy; and a history of abnormal liver function within 12 months.
Who will take informed consent?
The principal investigator will take informed consent after explaining the potential benefits and downsides of participation, the details of clinical investigations performed during the study follow-up, and the right to withdraw from the study. Patients will be given at least one week to consider the participation and raise any questions at their concern.
Additional consent provisions for collection and use of participant data and biological specimens
Not applicable
Interventions
Explanation for the choice of comparators
Patients randomized to the control arm will be provided a placebo containing corn starch. The placebo capsule has identical color, appearance, and packaging as the capsule containing NR.
Intervention description
During the 24-month study period, patients randomized to the NR treatment group will have oral administration of 3 capsules of NR every morning, each containing 100mg NR; patients randomized to the controlled group will have oral administration of 3 capsules of placebo every morning. Patients will be followed up at 1 month, 4 months, and then every 2 months for clinical examination, VF testing, and OCT imaging of the RNFL.
Criteria for discontinuing or modifying allocated interventions
The safety endpoints are: (1) development of visual field progression (defined by the Early Manifest Glaucoma Trial criteria);21 (2) decrease in visual acuity ≥2 lines; and (3) IOP>30mmHg on 2 consecutive visits. Patients will exit the study and receive appropriate treatment if any of the safety end-points is reached. Patients are allowed to withdraw from the study for any reasons at any time.
Strategies to improve adherence to interventions
The capsules (NR or placebo) are packed in weekly packets (each packet contains 21 capsules – 3 capsules per day for 7 days) labelled with patient’s name. Patients are requested to bring back the packets at each follow-up visit. Capsules left in the packets will be counted to check and remind patients’ compliance to the treatment.
Relevant concomitant care permitted or prohibited during the trial
Patients will continue the existing IOP-lowering regimen as prescribed before study enrollment. No additional IOP-lowering intervention will be provided unless a safety end-point is reached (i.e. detection of VF progression; decrease in visual acuity ≥2 lines; or IOP>30mmHg on 2 consecutive visits). They are allowed to take other medications for medical conditions during the trial. Details of the medications used during the study period will be recorded and reported.
Provisions for post-trial care
The no-observed-adverse-effect-level of NR was 300mg/kg/day.14 We do not expect to observe any adverse effect with an oral intake of NR at 300mg/day.
Outcomes
The primary outcome measure is the rate of change of RNFL thickness measured by OCT over 24 months of study follow-up. Secondary outcome measures include (1) time to VF progression, and (2) time to progressive RNFL/GCIPL thinning, and (3) the rate of change of VF sensitivity over 24 months (to investigate neuroprotection) and 1 month (to investigate neuroenhancement). Change in inner retinal density over the parapapillary region and the macula measured by OCT angiography over 24 months will be evaluated as an exploratory outcome measure.
Participant timeline
Study enrollment started on 9 March 2020 and the expected completion date of recruitment is 8 March 2021. Patients will be follow-up at 1 month, 4 months, and then every 2 months until 24 months for clinical examination, IOP measurement, OCT imaging of the RNFL, and VF testing (Figure 2).
Sample size
In our pilot study following 236 glaucoma patients every 4 months for ≥3 years, the mean rate of RNFL thinning was 1.2µm/year (P<0.001) for eyes with evidence of progressive RNFL thinning detected by TPA; the slope variance was 1.39(µm/year)2 and the residual variance was 3.83µm2.20 We aim to detect a difference of at least 0.70µm/year in the rate of RNFL thinning between treatment groups, a rate slightly greater than the mean rate of age-related RNFL thinning, which has been reported to be approximately 0.52µm/year.22 Taking reference from the method of sample size calculation for linear mixed modeling to compare the rates of change of a parameter of interest between two treatment groups described by Ard and Edland,23 53 patients per study arm are needed for serial RNFL thickness measurements at baseline, 1 month, 4 months, and then every 2 months for 24 months (3 separate measurements will be collected at the baseline, 1-month, and 24-month follow-up visits) with a power of 80% and an alpha of 5%. Assuming a default rate of 15% during study follow-up, a total of 125 patients will be recruited.
Recruitment
Eligibility for study inclusion will be reviewed by the attending ophthalmologists. Eligible patients will be referred to the investigators to cross check if the admittance criteria are met. The principal investigator will then explain the study procedures to the patients. Those who agree to participate in the study will come back on a scheduled visit for baseline investigation.
Assignment of interventions: allocation
Sequence generation
Patients were randomly allocated (1:1) at the baseline visit to receive NR or placebo within 1 month of enrollment. Randomization with minimization by age (<60 years vs ≥60 years), gender (male vs female), VF MD (<-6 dB vs ≥-6 dB), spherical equivalent (<-6.0D vs ≥ -6.0D), mean IOP measurement over the past 3 years (≤21mmHg vs >21mmHg), and number of glaucoma medications (0,1,2,3, or ≥4) will be performed using an open source computer program MinimPy.24 A random element was introduced in the minimization procedure (i.e. weighted randomization of 0.8) to make the randomization more unpredictable.25
Concealment mechanism
Patients and investigators are unaware of treatment allocations and have no access to the randomization sequence and codes, which are stored in a password-protected file in a computer. The NR and placebo tablets are in identical appearance and uniformly packaged.
Implementation
The principal investigator will enroll participants. Two masked research technicians will generate the allocation sequence with randomization by minimization and assign participants to interventions.
Assignment of interventions: Blinding
Who will be blinded
The attending ophthalmologists, the participants, the outcome assessors, and data analysts will be blinded after assignment to intervention.
Procedure for unblinding if needed
The blinding will be maintained until serious adverse event occurs. The unblinded participant will exit the trial and the ocular and/or medical conditions will be managed accordingly. The management results will be recorded on the clinical report form and reported to the Kowloon Central / Kowloon East Cluster Research Ethics Committee.
Data collection and management
Plans for assessment and collection of outcomes
Patients will receive clinical examination, OCT imaging of the RNFL, and VF testing at baseline, 1 month, 4 months, and then every 2 months for 24 months. Three repeated OCT and VF measurements will be collected at the baseline, 1-month, and 24-month visits.
Clinical examination
Clinical examination includes slit-lamp biomicroscopy of the anterior and posterior segments, and Goldmann applanation tonometry. Two IOP readings will be obtained to calculate the mean. A third reading will be obtained if the difference between the first two is >2mmHg and the median is recorded. Axial length will be measured with partial coherence laser interferometry (IOLMaster, Carl Zeiss Meditec). Subjective and objective refraction will be performed at the screening and the last follow-up visits. Central corneal thickness will be measured with ultrasound pachymetry. Dilated fundus examination for color optic disc stereophotography will be performed at the screening visit and then yearly.
Optical coherence tomography imaging
The RNFL will be imaged with the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, USA) for comparison of the rates of RNFL thinning (primary outcome measure) between treatment groups. The RNFL/GCIPL will also be imaged by the Triton OCT (Topcon, Tokyo, Japan) to determine the time to progressive RNFL/GCIPL thinning detected by TPA for an individual eye (secondary outcome measure). RNFL-GCIPL thickness data of individual pixels in serial RNFL/GCIPL thickness maps collected over 24 months will be exported for TPA. The algorithm of TPA has been described.17,18,20 In brief, TPA performs pixel-by-pixel linear regression analysis between RNFL/GCIPL thickness and time for evaluation of progressive RNFL/GCIPL thinning after registering and aligning serial OCT scans in corresponding retinal locations of an eye. To minimize type I errors consequential to multiple testing in an eye, the RNFL-GCIPL thickness maps will be condensed from 512x256 pixels to 128x64 superpixels with a false discovery rate (FDR) controlled at 5%. Low test-retest variability of Fourier-domain OCT measurements has been shown.26,27 Images with a poor signal-to-noise ratio, motion artifact, poor centration, segmentation error or missing data (e.g. blinking) will be discarded; re-scanning will be performed in the same visit. Peripapillary and macular inner retinal vessel density will be measured with OCT angiography.
Visual field test
Visual field test will be performed with the Humphrey Field Analyzer II 24-2 SITA standard strategy (Carl Zeiss Meditec). A reliable VF test has fixation losses <20% and false positive rate <15%. Unreliable tests will be repeated on the same day. Visual field progression is identified when three test locations show significant reduction in VF sensitivity greater than the test-retest variability observed on at least two (possible progression) or three (likely progression) consecutive tests, according to the Early Manifest Glaucoma Trial (EMGT) criteria.21
Plans to promote participant retention and complete follow-up
Investigators will phone contact or send text message to remind the participants for follow-up before each study visit. Participants who default a scheduled appointment will be contacted by the investigators to re-arrange another appointment within 2 weeks.
Data management
Clinical data will be collected and recorded by two designated technicians. The data will be stored in password-protected computers. The study will be conducted in compliance with Good Clinical Practices to ensure the rights and well-being of the participants and that the data collected are complete and verifiable from source documents. Data validation is performed with cross-system consistency checks by another technician.
Confidentiality
No personal information will be recorded on the data sheets or electronic data files. A study code will be assigned to each participant. The document containing the information of the study code and the identity of the patient will be kept separate from the study data files and data sheets.
Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use
There are no plans for collection of biological specimens in this trial.
Statistical methods
Statistical methods for primary and secondary outcomes
(1) The rates of RNFL thinning (primary outcome measure) between treatment groups will be compared with linear mixed modeling after adjusting for covariates including age, axial length, glaucoma severity (baseline RNFL thickness), IOP levels during following-up, OCT signal strength, multiple testing, and clustering between fellow-eyes. (2) Changes in VF sensitivity (measured in dB) at (A) one month and (B) 24 months between treatment groups will be compared with linear mixed models after adjusting for covariates. (3) Survival analysis will be performed to compare the differences in (A) time from baseline to VF progression (determined by EMGT criteria) and (B) time from baseline to progressive RNFL/GCIPL thinning (determined by TPA) between treatment groups using Cox proportional hazards models with shared-frailty to adjust for clustering between fellow eyes.
Interim analyses
Interim analyses will be performed by a statistician after data collection at (A) 4 months and (B) 12 months. The rates of change of VF sensitivity and RNFL thickness will be compared between treatment groups. The participants and investigators will remain masked to the study groups. The trial steering committee will stop the trial if the rate of VF sensitivity decline and the rate of RNFL thinning were significantly worse in the NR-treated group than the placebo-treated group.
Methods for additional analyses (e.g. subgroup analyses)
Detection of progressive RNFL thinning and VF sensitivity decline has been shown to be more difficult in moderate to advanced glaucoma compared with early glaucoma. Subgroup analysis will be performed to determine if the rates of change of RNFL thickness and VF sensitivity would be different between treatment groups in patients with early glaucoma versus those with moderate to advanced glaucoma. Subgroup analysis will also be performed to investigate if the responses of NR supplementation would be different between old and young patients.
Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data
Data will be analyzed on an intention-to-treat basis to include all patients who receive the randomized treatment and have at least two follow-up visits. Multiple imputation by chained equations (MICE) algorithm will be applied for estimation of missing data.28
Plans to give access to the full protocol, participant level-data and statistical code
No later than 3 years after the completion of data collection, the protocol and deidentified dataset will be available from the principal investigator upon reasonable request.
Oversight and monitoring
Composition of the coordinating centre and trial steering committee
The coordinating center is located at Hong Kong Eye Hospital. The trial steering committee is directed by the principal investigator and includes two independent researchers and one statistician, who are not part of the same institution as any of the members of the study team. The trial steering committee is responsible to ensure the protocol is implemented as planned, review study progress on a regular basis, and uphold good clinical practice at all times. The coinvestigators, research assistants and technicians of the trial will be responsible for all aspects of logistics and organization of the trial.
Composition of the data monitoring committee, its role and reporting structure
An independent data monitoring committee comprises two independent experts in statistics, who are not part of the same institution as any of the members of the study team. They will be responsible to inspect clinical data collected during the study period, review interim analysis, and report back to the investigators if any action is required.
Adverse event reporting and harms
Any adverse event will be recorded whether they are related to the study or not. A serious adverse event will be reported to the Kowloon Central / Kowloon East Cluster Research Ethics Committee within 24 hours of the event. The principal investigator will be responsible to follow the management of the serious adverse event until resolution or conclusion. The investigators and the trial steering committee will determine whether an adverse event or serious adverse event is related to the study drug.
Frequency and plans for auditing trial conduct
A team of clinical research coordinators independent of the investigators will monitor the data reported in the clinical research forms are complete and accurate, ensure all adverse events and serious adverse events are recorded and reported, and confirm study drugs to be stored and distributed according to the good clinical practice every 3 months.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees)
Any amendment to the protocol will be submitted by the principal investigator to be approved by the Kowloon Central / Kowloon East Cluster Research Ethics Committee and Hong Kong Food and Health Bureau Research Secretariat before implementation.
Dissemination plans
Trial results will be presented at international scientific conferences and published in peer-reviewed scientific journals.