CADTH Reimbursement Review

Faricimab (Vabysmo)

Sponsor: Hoffmann-La Roche Ltd.

Therapeutic area: Macular degeneration, age-related

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Introduction

Age-related macular degeneration (AMD) is a chronic eye disease caused by degeneration of the macula.¹ It is a leading cause of central vision loss in people 50 years of age or older.² The disease is classified into dry or wet forms. Wet AMD, also known as neovascular AMD (nAMD), is characterized by abnormal formation of new blood vessels beneath the macula,² leading to accumulation of blood and other factors, and resulting in severe and irreversible impairment of central vision, and potentially blindness. In Canada, it is estimated that nAMD affects more than 150,000 individuals.²

The standard of care for nAMD in Canada is anti–vascular endothelial growth factors (VEGFs). Such drugs, which include ranibizumab, aflibercept, brolucizumab, and (off-label) bevacizumab,³ have been shown to limit the progression of nAMD, and stabilize or reverse vision loss.⁴ Anti-VEGF drugs are administered as intravitreal injections on an ongoing basis at intervals of between 1 and 3 months after completion of loading doses.^3,5-7

Faricimab is a bispecific form of immunoglobulin G1 that inhibits human vascular endothelial growth factor-A (VEGF-A) and angiopoietin-2 (Ang-2), the key mediators in the development of nAMD.² Faricimab is indicated for the treatment of nAMD.⁸ The recommended dose of faricimab is 6 mg (0.05 mL) administered by intravitreal injection every 4 weeks (approximately every 28 ± 7 days) for the first 4 doses, followed by anatomic and visual acuity evaluations at week 20 and week 24 to inform dosing faricimab at intervals of 8, 12, or 16 weeks through week 60.⁸

The objective of this report was to perform a systematic review of the beneficial and harmful effects of faricimab 6 mg intravitreal injections for the treatment of nAMD in adults. The systematic review protocol for the current review was developed before the issuance of the Health Canada Notice of Compliance for faricimab on May 27, 2022.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient and clinician groups that responded to CADTH’s call for input and from a clinical expert consulted by CADTH for the purpose of this review.

Patient Input

CADTH received 1 joint input from Fighting Blindness Canada, the Canadian Council of the Blind, the CNIB Foundation, and Vision Rehabilitation Canada. Canadians with AMD (wet or dry) indicated that AMD had physical, psychological, and social impacts on their daily lives. Patients indicated that vision loss could affect daily activities such as their ability to use electronic devices, read, administer self-care, recognize or meet people, and drive. The patients reported that having to think about their disease frequently and worry about their condition worsening in the future created psychological stress. Patients indicated that the need for assistance and feelings of isolation and loneliness also had social impacts.

Most patients receiving anti-VEGF injection indicated that the treatment had helped them avoid losing more eyesight, while some noted no beneficial effect or were unsure if the treatments had any effect. Missing injection appointments, most commonly due to the lack of assistance, was reported to be a challenge by some patients. Anxiety and fear about the injections were noted by the patients to be the most difficult parts of the appointment. Most patients reported experiencing some pain into the evening after the appointment. Post-injection visual complications (e.g., blurry vision) and the resulting need for more frequent assistance were also reported.

Most patients expressed a preference for new treatments that can be taken less frequently.

Clinician Input

Input From Clinical Experts Consulted by CADTH

The clinical expert consulted by CADTH indicated the treatment goals of nAMD are to delay and/or reverse disease progression, reduce symptom severity, minimize adverse events (AEs), preserve and/or enhance health-related quality of life (HRQoL), and maintain patient independence. Given that most patients are currently required to attend treatment visits once every 1 to 3 months, the clinical expert noted that there is an unmet need for treatments that can be given at longer treatment intervals without recurrence of disease to reduce the burden of care associated with frequent treatment visits. This must be balanced with a promising safety profile.

The clinical expert noted that faricimab is expected to have a place of therapy similar to that of other anti-VEGFs, as a first-line treatment in patients with nAMD. The clinical expert pointed out that, if faricimab is reimbursed, a shift in the treatment paradigm will be likely, as faricimab is the first anti-VEGF therapy approved for administration at an extended interval of 16 weeks, which could potentially address the unmet need related to frequent treatment visits.

The clinical expert noted that patients with nAMD who have early and small neovascular lesions and 1 or more signs of active choroidal neovascularization (CNV) are suitable candidates for faricimab. The clinical expert indicated that faricimab can be used in patients who are treatment-naive or require a change in therapy due to inadequate response to other anti-VEGF drugs. The clinical expert indicated that patients with extensive subretinal fibrosis and macular tissue damage, very poor baseline visual acuity, long disease duration, or a history of unsuccessful therapy with an anti-VEGF drug for more than 2 years may not be suitable for treatment.

The clinical expert added that a clinical evaluation and optical coherence tomography (OCT) should be performed every 6 to 8 weeks at follow-up visits after the completion of loading doses to assess treatment response. Key assessment outcomes include a change in visual acuity and the presence of intraretinal fluid (IRF) or subretinal fluid (SRF), and blood accumulation in the macula. The clinical expert reported that an optimal response to anti-VEGFs is generally achieved at least 4 to 6 months after initiation of therapy.

The clinical expert indicated that faricimab should be discontinued in patients with extensive subretinal fibrosis (disciform scarring) accompanied by vision loss to counting fingers or worse, those in whom disease progression could not be modified with faricimab therapy, or those with end stage disease.

Clinician Group Input

CADTH received input from 1 clinician group, the Canadian Retina Society.

The clinician group noted that the current intravitreal anti-VEGF therapies used to treat nAMD are subject to limitations. The clinician group indicated that there has been an efficacy gap in real-world treatment outcomes compared to the outcomes observed in clinical trials due to the intense treatment burden associated with anti-VEGF therapy. The clinician group reported that, the visual outcomes in real-world practice are suboptimal, and the recent COVID-19 pandemic and other limitations on health care delivery have made it more difficult for patients to receive regular and intense treatment.

The clinician group described durability and reduced treatment frequency as the most important unmet needs in nAMD treatment. Because patients need to be treated with current anti-VEGF therapy every 7 to 8 weeks after an intensive monthly loading treatment cycle, the clinician group reported that this puts a high burden on patients and their caregivers in the form of time off work to attend appointments. The clinician group indicated that new treatments that require less-frequent injections would help reduce the treatment burden for these patients. Another unmet need associated with nAMD treatment identified by the clinician group is development of fibrosis and atrophy due to poor disease control in the long-term, which can eventually result in vision loss. According to the clinician group, improving long-term visual outcomes will require drugs that can effectively dry the retina for a longer period and reduce the nAMD treatment burden. Last, the clinician group noted the need for drugs that can also maintain a high safety profile to minimize the risk of ocular complications.

The clinician group mentioned that the dual mechanism of faricimab, which targets both the VEGF-A and Ang-2 pathways, which are critical in the development of retinal and choroidal vascular disease, is different from that of other available drugs. The clinician group agreed that this drug can be considered a first-line treatment or a rescue therapy for patients not responding to current nAMD treatment, while potentially reducing the treatment burden with the option of using longer treatment intervals than existing drugs.

Drug Program Input

The drug programs expressed an interest in understanding the appropriateness of study design (comparator, masking, and inclusion and exclusion criteria) of the pivotal trials of faricimab, the proportion of patients on extended intervals of anti-VEGFs in clinical practice, and whether there is experience with using faricimab beyond every 16 weeks. The clinical expert consulted by CADTH did not identify any particular concerns with regard to the study design elements noted by the drug programs. The clinical expert was unable to comment on the proportion of patients on extended treatment intervals as it varies by drug and is affected by multiple factors. The clinical expert noted that clinicians are generally reluctant to extend the treatment interval for existing anti-VEGF drugs beyond 16 weeks due to concerns with disease recurrence.

Other considerations of interest to the drug programs included the use of faricimab in treatment-experienced patients, the appropriate dosing interval for faricimab, and pricing. The clinical expert indicated that they would not hesitate to switch patients from another anti-VEGF drug to faricimab if a switch was deemed medically necessary. The expert expected that most patients would receive faricimab every 8 weeks to every 16 weeks in the maintenance phase if reimbursed.

Clinical Evidence

Pivotal Studies and Protocol-Selected Studies

Description of Studies

The TENAYA and LUCERNE studies met the inclusion criteria for the systematic review. They were identically designed, phase III, multi-centre, randomized, double-blind, active-controlled, noninferiority trials that evaluated the use of faricimab in comparison with aflibercept in treatment-naive nAMD patients (TENAYA, n = 671; LUCERNE, n = 658) for 112 weeks. Patients were randomized to either the faricimab or aflibercept arm on a 1:1 ratio. Patients in the faricimab arm were given faricimab 6 mg intravitreally every 4 weeks for 4 loading doses followed by maintenance doses every 8, 12, or 16 weeks. Patients in the aflibercept arm received aflibercept 2 mg intravitreally every 4 weeks for 3 doses then at a fixed maintenance interval of 8 weeks.

Both pivotal studies aimed to establish the noninferiority of faricimab to aflibercept through the primary outcome, which was the change from baseline in best corrected visual acuity (BCVA) as measured by Early Treatment Diabetic Retinopathy Study [ETDRS] charts averaged over weeks 40, 44, and 48 in the intention-to-treat (ITT) population. The noninferiority margin was specified as 4 letters on the ETDRS chart. Secondary outcomes included the frequency of administration of faricimab, retinal thickness, retinal fluids, pigment epithelial detachment (PED), and vision-related function, all of which were measured without control for multiplicity. At the time of this review, the studies were ongoing and data from the primary analysis at week 48 were available.

In both studies, the mean age of patients at baseline was between 74 and 77 years, and the majority were female (57.2% to 62.6%) and White (82.6% to 90.7%). The mean time since the diagnosis of nAMD was between 1.5 and 3.2 months and the majority had a baseline BCVA of 73 to 55 letters on ETDRS charts.

Efficacy Results

A summary of the key efficacy results is provided in Table 2.

Change in Visual Acuity

The primary outcome of both studies was the change from baseline in BCVA (ETDRS letters) averaged over weeks 40, 44, and 48 in the ITT population. The mean difference between the faricimab and aflibercept arms was 0.7 ETDRS letters (95% C, −1.1 to 2.5) in the TENAYA study, and 0.0 ETDRS letter (95% confidence interval [CI], −1.7 to 1.8) in the LUCERNE study in the ITT populations, both of which fell within the noninferiority margin.

The proportion of patients gaining 15 or more ETDRS letters in BCVA from baseline averaged over weeks 40, 44, and 48 was a secondary outcome. The proportions were 20.0% and 15.7% in the faricimab arm and aflibercept arm, respectively, of the TENAYA study, and 20.2% and 22.2% in the respective arms of the LUCERNE study. In the TENAYA study, the proportions of patients avoiding a loss of 15 or more ETDRS letters in BCVA from baseline averaged over weeks 40, 44, and 48 were 95.4% and 94.1% in the faricimab and aflibercept arms, respectively, while in the LUCERNE study, the proportions were 95.8% and 97.3% in the respective arms.

Frequency of Faricimab Injections

In the TENAYA study, the mean numbers of treatment injections given through week 48, a secondary outcome, were 6.9 (standard deviation [SD] = 0.63) in the faricimab arm and 7.8 (SD = 0.45) in the aflibercept arm. In the LUCERNE study, the mean numbers of study treatment injections given through week 48 were 7.0 (SD = 0.53) in the faricimab arm and 7.9 (SD = 0.32) in the aflibercept arm.

The studies measured the proportion of patients in the faricimab arm on an injection interval of 8, 12, or 16 weeks as a secondary outcome. The proportions of patients who received faricimab every 8, 12, or 16 weeks at week 48 were 20.3%, 34.0% and 45.7%, respectively, in the TENAYA study, and 22.2%, 32.9%, and 44.9%, respectively, in the LUCERNE study.

Vision-Related Function

In the TENAYA study, the change from baseline in National Eye Institute Visual Functioning Questionnaire 25 (NEI VFQ-25) composite score at week 48 (exploratory outcome), was 4.82 points (SD = 10.81) and 2.54 points (SD = 10.93) in the faricimab and aflibercept arms, respectively. In the LUCERNE study, the composite scores increased from baseline at week 48 by 4.35 points (SD = 10.65) and 5.55 points (SD = 11.17) in the faricimab and aflibercept arms, respectively.

Legal Blindness

The proportions of patients with legal blindness averaged over weeks 40, 44, and 48 (a secondary outcome) were 6.4% and 7.0% in the faricimab and aflibercept arms, respectively, in the TENAYA study, while in the LUCERNE study, the proportions were 7.9% and 7.5% in the respective arms.

Anatomic Outcomes

The change from baseline in central subfield thickness (CST), which is the distance between the internal limiting membrane (ILM) and retinal pigment epithelium (RPE), was a secondary end point. A reduction in CST (ILM-RPE) was observed in both treatment arms in both trials.

The proportions of patients with absence of IRF, SRF, and PED (secondary outcomes) at week 48 were in the ranges of 74.4% to 84.1%, 62.1% to 75.7%, and 3.0% to 7.7%, respectively.

Harms Results

A summary of the key harms results is provided in Table 2.

As of the primary analysis (follow-up until week 48), ocular AEs were reported in 36.2% and 38.1% of patients who received faricimab and aflibercept, respectively, in the TENAYA study. In the LUCERNE study, 40.2% of patients who received faricimab and 36.2% of patients who received aflibercept reported at least 1 ocular AE. The most common ocular AE was conjunctival hemorrhage (5.7% to 8.9%). The frequencies of ocular serious adverse events (SAEs) were 1.2% and 1.8% in the faricimab and aflibercept arms, respectively, in the TENAYA study, and 2.1% in both arms in the LUCERNE study. The most common ocular SAE was worsening of nAMD (up to 0.9%). In both studies, approximately half of the study populations reported non-ocular AEs. Nasopharyngitis (4.9% to 8.3%) was the most frequently reported non-ocular AE. The frequency of non-ocular SAEs in either arm of the studies was within a range of 9.0% to 14.7%, with cardiac failure being the most frequently reported SAE. The frequency of treatment discontinuation due to AEs was 2.4% in the faricimab arm and 0.3% in the aflibercept arm of the LUCERNE study, and 0.9% in both treatment arms of the TENAYA study. Death was reported in 0.3% to 2.1% of patients across treatment arms.

In terms of notable harms, 1 patient reported endophthalmitis in the aflibercept arm of the LUCERNE study, while none reported this AE in the faricimab arms of either trial. No incidence of retinal vasculitis was reported. Intraocular inflammation was reported in 0.6% to 2.4% of patients, and vitreous floaters were reported in 1.2% to 3.9% of patients. Arterial thromboembolic events were reported in 0.9% to 1.2% of patients.

Table 2: Summary of Key Results From the Pivotal Studies

AE = adverse event; BCVA = best corrected visual acuity; CI = confidence interval; CMH = Cochran-Mantel-Haenszel; ETRDS = Early Treatment Diabetic Retinopathy Study; ITT = intention-to-treat; MMRM = mixed model for repeated measures; NR = not reported; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; SAE = serious adverse event; SD = standard deviation; SE = standard error.

^aAdjusted mean. The primary end point was analyzed using an MMRM, with the change from baseline in BCVA as the dependent variable, considering all available observations of BCVA score at all visits. The model was adjusted for treatment group, visit, visit-by-treatment-group interaction, baseline BCVA (continuous), baseline BCVA (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), baseline low luminance deficit (< 33 letters and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world).

^bCMH-weighted estimate. The observed proportions and the differences in observed proportions were obtained by applying CMH weight, stratified by baseline BCVA score (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), low luminance deficit (< 33 letters and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world).

^cAnti-Platelet Trialists’ Collaboration–defined arterial thromboembolic events, defined as nonfatal strokes or nonfatal myocardial infarctions, or vascular deaths (including deaths of unknown causes).

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Critical Appraisal

The overall designs of the TENAYA and LUCERNE studies were appropriate for the objectives of the studies. There was no particular concern with the methods of randomization, allocation concealment, or blinding. While imbalances were identified in 2 baseline characteristics in the LUCERNE study, including time since diagnosis of nAMD and proportion of patients with occult CNV lesions, these imbalances were unlikely to bias the results in favour of the faricimab arm according to the clinical expert consulted by CADTH. The conclusion of noninferiority of faricimab to aflibercept was based on an ITT analysis of the primary outcome. It is generally preferred that a claim of noninferiority be based on agreement between the ITT population and the per-protocol population for a more conservative approach in the context of noninferiority studies. Nonetheless, the results of a supplementary per-protocol analysis in the studies, and several sensitivity analyses conducted by the sponsor and the FDA, were consistent with those of the primary ITT analysis. The noninferiority margin of 4 ETDRS letters was justified, and the clinical rationale was considered reasonable by the clinical expert. The studies were adequately powered for the assessment of the primary outcome. The dropout rate of 4.3% in both studies was acceptable according to the clinical expert. Intercurrent events (ICEs), most of which were related to COVID-19, were reported in approximately 10% of patients in both studies, and the approach used to handle ICEs was considered appropriate by the CADTH review team. A key limitation in the statistical analysis was the lack of adjustment for multiplicity for secondary outcomes and subgroup analyses. As such, the findings were considered exploratory.

In terms of generalizability, because the studies included only treatment-naive patients, the applicability of the trial results to treatment-experienced patients is unclear. In addition, aflibercept was given at a fixed dosing interval in the maintenance phase, which does not align with the “treat-and-extend” protocol commonly used in clinical practice, further limiting the generalizability of the results. The outcome of the frequency of faricimab injections is also uncertain, given that the method of interval assignment for faricimab in the maintenance phase until the primary analysis at week 48 was more rigid than what is seen in clinical practice. However, keeping the dosing consistent may have helped reduce internal validity issues in the studies. The clinical expert consulted by CADTH anticipated that later analysis may have more generalizability value, given the implementation of a personalized treatment-interval algorithm from week 60 and onward involves routine adjustment of intervals based on disease activity. Furthermore, while the length of assessment in the primary analysis was adequate for assessing the efficacy and safety of faricimab in the context of a noninferiority trial, the clinical expert expected that at least 2 to 3 years of clinical data would be required to assess the durability of faricimab. Last, the lack of direct evidence comparing faricimab to brolucizumab or bevacizumab represents an important evidence gap in the evaluation of anti-VEGFs.

Indirect Comparisons

Description of Studies

One indirect treatment comparison (ITC) was submitted by the sponsor and included in this review. No additional ITCs were identified in the literature. The sponsor performed a Bayesian network meta-analysis (NMA) to estimate the efficacy of faricimab in patients with nAMD compared to other anti-VEGFs.

Efficacy Results

For the outcome of BCVA at 1 year, 35 trials were analyzed in a random-effects model. In the ITC, faricimab 6 mg intravitreal every 8 to 16 weeks was not different (95% credible intervals [CrIs] include the null) to comparators for BCVA. For the outcome of number of injections at 1 year, 27 trials were analyzed in a random-effects model. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| For the outcome of retinal thickness, at 1 year, 25 randomized controlled trials (RCTs) were analyzed using a random-effects model. The ITC suggests that faricimab may be favourable (95% CrIs exclude the null) to bevacizumab regimens, ranibizumab 0.5 mg intravitreal every 8 weeks for the outcome of mean change in retinal thickness (CST). Additionally, brolucizumab 6 mg intravitreal every 8 to 12 weeks may be favourable (95% CrI excludes the null) to faricimab for this outcome. The outcome of proportion of patients gaining or losing 10 or 15 EDTRS letters at 1 year was analyzed, but poor model fit precludes making conclusions about the effect of faricimab versus comparators for this outcome.

Harms Results

Limited data were available for the NMAs conducted for ocular AEs and for discontinuation. Fixed-effects models were therefore used for these end points, and there was a high degree of uncertainty in these models.

Critical Appraisal

Limitations to the sponsor’s ITC include considerable heterogeneity in the study and some baseline characteristics (most notably the heterogeneity in the methods to assess retinal thickness and in the method of assessing retinal thickness) and the availability of information about prognostic factors such as presence of SRF or IRF. Additionally, there was a weak connection between faricimab and the rest of the network through aflibercept via the LUCERNE and TENEYA trials, and through ranibizumab in a phase II trial.

The results of the analysis related to the number of injections may have been affected by protocol-driven administration of therapies with fixed intervals in clinical trials. Limitations to the NMA preclude making conclusions about the proportion of patients gaining or losing 10 or 15 EDTRS letters and retinal thickness.

As limited data were available for the NMAs conducted for ocular AEs and treatment discontinuation, fixed-effects models were used for these end points, and there was a high degree of statistical uncertainty in these models. Limited data are therefore available to draw conclusions about the effect of faricimab versus comparators on ocular AEs and treatment discontinuation.

Other Relevant Evidence

Description of Studies

The STAIRWAY (N = 76) and AVENUE (N = 273) studies were phase II, multi-centre, randomized, double-blind, active-controlled trials that did not meet the inclusion criteria of the systematic review. However, because they are the only head-to-head comparisons between faricimab and ranibizumab to date, they are summarized and critically appraised in this review.

In the STAIRWAY study, patients were assigned to either faricimab 6 mg (4 monthly loading doses then 1 maintenance dose every 12 weeks), faricimab 6 mg (4 monthly loading doses then 1 maintenance dose every 16 weeks.), or ranibizumab 0.5 mg (every 4 weeks) in a 2:2:1 ratio over a 48-week double-blind period. The primary outcome was the mean change from baseline in BCVA (ETDRS letters) at week 40 in the ITT population. Data were analyzed descriptively.

In the AVENUE study, patients were assigned to either ranibizumab 0.5 mg (every 4 weeks; arm A), faricimab 1.5 mg (every 4 weeks; arm B), faricimab 6 mg (every 4 weeks; arm C), faricimab 6 mg (4 monthly loading doses then 1 maintenance dose every 8 weeks.; arm D), or ranibizumab 0.5 mg for 3 monthly doses then faricimab 6 mg every 4 weeks (arm E), in a 3:2:2:2:3 ratio over a 32-week double-blind period. Arm B does not align with the recommended dose in the product monograph and therefore was not summarized in this review. The primary objective of the AVENUE study was to evaluate the efficacy of faricimab compared to ranibizumab monotherapy in treatment-naive patients from baseline to week 36, and in treatment-experienced patients (switched from ranibizumab to faricimab in the study after an incomplete response) from week 12 to week 36. The primary outcome was the mean change in BCVA (ETDRS letters) from baseline to week 36 in the comparisons of arms A, C, and D (treatment-naive population). In the comparison of arms A and E (treatment-experienced population), the primary outcome was the mean change in BCVA (ETDRS letters) from week 12 to week 36 in patients with a BCVA of less than or equal to 68 ETDRS letters at week 12.

Efficacy Results

STAIRWAY

In the STAIRWAY trial, the mean differences between the faricimab and ranibizumab arms in BCVA were −2.1 ETDRS letters (80% CI, −6.8 to 2.6) for faricimab every 12 weeks, and 1.1 letters (80% CI, −3.4 to 5.5) for faricimab every 16 weeks, at week 40, and 0.5 letters (80% CI, −4.3 to 5.3) for faricimab every 12 weeks, and 1.8 letters (80% CI, −2.7 to 6.4) for faricimab every 16 weeks, at week 52. Because the trial was not designed to test a hypothesis, the results were considered exploratory.

AVENUE

In the AVENUE trial, the mean differences between the faricimab and ranibizumab arms in the change in BCVA from baseline to week 36 were −1.6 letters (80% CI, −4.9 to 1.7) in arm C, and −1.5 letters (80% CI, −4.6 to 1.6) in arm D, of the treatment-naive population, The mean difference between the faricimab and ranibizumab arms in the change in BCVA from week 12 to week 36 was −1.7 (80% CI, −3.8 to 0.4) in the treatment-experienced population C in AVENUE trial. No statistically significant difference between the faricimab and ranibizumab treatment groups was identified for the primary end point in either analysis subpopulation. Overall, superiority of faricimab to ranibizumab was inconclusive, based on the results of the primary outcomes.

Harms Results

The harms and notable harms reported in both phase II trials were generally similar and consistent with the TENAYA and LUCERNE trials. The proportion of patients who experienced at 1 or more AEs in the STAIRWAY trial was 81.3% in the ranibizumab arm compared to between 74.2% and 75% in the faricimab arms. However, the proportions of patients who experienced SAEs were between 9.7% and 16.7%, and the proportions of patients who died were between 4.2% and 6.5% in the faricimab arm, with no events occurring in the ranibizumab arm. In the faricimab arms, the SAEs were non–ocular-related (cardiac disorders), and the 3 deaths were associated with ischemic stroke, sepsis, and metastatic neoplasm.

In the AVENUE trial, the proportions of patients with AEs were 84.8% and 84.4% in arms D and E, respectively, compared with 76.1% in arm A and 79.5% in arm C. One death (in arm E) was related to cardiorespiratory arrest.

Critical Appraisal

In both studies, trial eligibility criteria were appropriate for the indication, and the trial populations were generally representative of the Canadian patient population, based on baseline characteristics. While the dropout rates due to an AE were similar for all arms in the STAIRWAY trial, it was highest (7.8%) in arm E in the AVENUE trial. In terms of harms and notable harms, all these groups had small sample sizes and few events, making it difficult to draw any conclusions in either trial.

The study designs and planned analyses were the key limitations to the studies. The phase II designs were not appropriate for testing the superiority of faricimab versus ranibizumab. As the STAIRWAY trial was designed as an exploratory study that did not test a hypothesis, no conclusions can be drawn by the CADTH review team regarding the relative efficacy and safety of faricimab compared to ranibizumab based on this study. The AVENUE study was designed to test an a priori hypothesis that faricimab was superior to ranibizumab. The primary objective of AVENUE was not met as no statistically significant difference between the faricimab treatment groups and ranibizumab were identified for the primary end point in either analysis subpopulation.

Conclusions

Based on evidence from the pivotal trials, faricimab is noninferior to aflibercept in the change in BCVA from baseline over 48 weeks of treatment in treatment-naive adult patients with nAMD. The evidence regarding comparative efficacy in other BCVA outcomes, anatomic outcomes, vision-related function, and HRQoL was supportive of noninferiority but associated with some uncertainties due to limitations with the design of the studies and analyses. Neither the reviewed phase II studies nor the NMA submitted by the sponsor provides clear evidence that faricimab is superior to other anti-VEGFs for BCVA outcomes. Most patients received faricimab at an extended interval of 12 or 16 weeks at week 48 in the pivotal studies. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||; however, the heterogeneity in study design and patient characteristics may limit the usefulness of conclusions that can be made about these results. Overall, the safety profile of faricimab was similar to that of aflibercept in the pivotal trials, with a low occurrence of intraocular inflammation and no reports of retinal vasculitis or endophthalmitis.

Introduction

Disease Background

AMD is a chronic eye disease caused by degeneration of the macula.¹ It is a leading cause of central vision loss in people aged 50 years or older² and is classified into dry or wet forms. Wet AMD, also known as nAMD, is characterized by CNV, which refers to abnormal formation of new blood vessels underneath the macula.^2,3 The accumulation of fluid, lipids, and blood resulting from leakage from these new blood vessels can lead to impairment of central vision. The symptoms of nAMD include visual distortion, a scotoma (blind spot), and blurred vision.² Due to the progressive nature of nAMD, severe and irreversible vision loss can occur. In Canada, it is estimated that nAMD affects more than 150,000 individuals.²

The diagnosis of nAMD is based upon the presence of characteristic findings (e.g., SRF and/or IRF, retinal and subretinal hemorrhage, retinal thickening, or PED) on eye examination using imaging techniques such as OCT, OCT-angiography, and fundus fluorescein angiography.^1,2

Standards of Therapy

The standard of care for patients with nAMD is intravitreal injection of an anti-VEGF drug.^11,12 Ranibizumab, aflibercept, and brolucizumab are anti-VEGF drugs approved by Health Canada, and bevacizumab is commonly used off-label.³ A “treat-and-extend” protocol is commonly used in the management of nAMD in Canada, where clinicians adjust the interval of anti-VEGF injections based upon clinical findings after initial treatments and stabilization.^1,3 Interval reduction is considered in the presence of disease activity (e.g., retinal fluid or a hemorrhage) in the macula, and interval extension or maintenance in the absence of disease activity.¹

Other therapies include photodynamic therapy with verteporfin, which may be indicated in patients with a suitable CNV lesion from rare forms of nAMD, but it is not routinely prescribed for most forms of nAMD.³ Supplementation with zinc and antioxidant vitamins is also encouraged.³

The clinical expert consulted by CADTH indicated that the treatment goals of nAMD are to delay and/or reverse disease progression, reduce symptom severity, minimize AEs, preserve and/or enhance HRQoL, and maintain patient independence.

Drug

Faricimab is a humanized bispecific immunoglobulin G1 that selectively blinds to and neutralizes VEGF-A and Ang-2, which are mediators in the pathogenesis of nAMD.⁸ Endothelial cell proliferation is promoted by VEGF-A, leading to increased neovascularization and vascular permeability, whereas Ang-2 promotes endothelial destabilization, pericyte loss, and pathological angiogenesis and sensitizes blood vessels to the activity of VEGF-A. Through the inhibition of Ang-2 and VEGF-A, faricimab is expected to reduce vascular permeability and inflammation, inhibit pathological angiogenesis, and restore vascular stability.

This is the first CADTH review for faricimab. Faricimab was granted a Health Canada Notice of Compliance for the indication of treatment of nAMD on May 27, 2022, at a recommended dose of 6 mg (0.05 mL) by intravitreal injection every 4 weeks (approximately every 28 ± 7 days) for the first 4 doses, followed by anatomic and visual acuity evaluations at week 20 and week 24 to inform dosing at intervals of 8, 12, or 16 weeks through week 60. Patients should be assessed regularly and monitored between dosing visits, which should be scheduled based on the patient's status and at the physician's discretion. Faricimab has received FDA approval for the treatment of nAMD and diabetic macular edema.

The sponsor is seeking reimbursement of faricimab according to the approved indication, which is for the treatment of nAMD.

The key characteristics of faricimab and the comparator drugs are summarized in Table 3.

Table 3: Key Characteristics Of Faricimab, Aflibercept, Ranibizumab, Brolucizumab, and Bevacizumab

Ang-2 = angiopoietin-2; ATE = arterial thromboembolic events (includes nonfatal stroke, nonfatal myocardial infarction, or vascular death); IOP = intraocular pressure; mAb = monoclonal antibody; nAMD = neovascular age-related macular degeneration; PIGF = placental growth factor; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; VEGF = vascular endothelial growth factor.

^aBevacizumab is used off-label in the treatment of nAMD.

^bHealth Canada–approved indication.

Source: Vabysmo product monograph,⁸ Eylea product monograph,⁶ Lucentis product monograph,⁷ Beovu product monograph,⁵ Avastin product monograph,¹³ Canadian Pharmacists Association: Therapeutic Choices – Age-related Macular Degeneration.³

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

Four patient advocacy groups (Fighting Blindness Canada, the Canadian Council of the Blind, the CNIB Foundation, and Vision Loss Rehabilitation Canada), provided 1 joint input for the treatment of AMD. These groups gathered information from Canadians living with wet or dry AMD through an online survey during the first months of 2020. Overall, 337 participants responded to this survey.

Patients reported that AMD had physical, psychological, and social impacts on their daily lives. Sight loss due to AMD affected daily activities of patients, such as using phones and tablets, reading books and newspapers, looking after their appearance, recognizing or meeting people, and driving. In addition, AMD was reported to impart significant psychological burden on patients, such as frequently thinking about their disease and its impacts, and worrying that their condition might worsen. When asked about the social implications of AMD, patients mentioned the need for assistance due to limited eyesight and feelings of isolation and loneliness.

The majority of respondents (75.4%) were currently being treated with anti-VEGF injections at the time of the survey. While describing their experiences with injections to treat the AMD, 72.7% participants expressed their satisfaction by stating that “they helped me avoid losing more eyesight.” Although most patients were satisfied with their injections, almost 20% of respondents indicated that they thought the injections have no beneficial effect or were unsure if there is an effect. Some patients reported missing injection appointments, with “unable to find someone to take me to the appointment” the most common reason. The most difficult part of the eye injection appointments was reported to be anxiety and fear about the injection, and approximately 4 out of 5 patients reported experiencing at least some pain into the evening after their appointments. In addition, visual complications (e.g., blurry vision) were reported post-injection, as well as the need for more frequent assistance due to post-injection complications.

When asked about their preferences for a new treatment for AMD, most patients indicated they would prefer a treatment that can be taken less frequently. However, the survey did not capture information on which outcomes were important to patients.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 1 clinical specialist with expertise in the diagnosis and management of nAMD.

Unmet Needs

The clinical expert consulted by CADTH noted that the treatment goals of nAMD are to delay and/or reverse disease progression, reduce symptom severity, minimize AEs, preserve and/or enhance HRQoL, and maintain patient independence. The clinical expert reported that there is an unmet need for treatments that can be given at longer treatment intervals without recurrence of disease. According to the clinical expert, while existing anti-VEGFs are effective in managing nAMD, they are given intravitreally on an ongoing basis by trained clinicians every 4 to 12 weeks, in most cases. The clinical expert reported that in some cases, the duration of treatment can be as long as a decade and treatment adherence is crucial to achieving favourable outcomes in the treatment of nAMD. In the clinical expert’s experience, some patients who cannot adhere to the regimen due to the burden of frequent visits ultimately lose vision after discontinuing treatment. The clinical expert suggested that having new treatments with an extended injection interval would imply a reduction in the number of treatment visits, which could alleviate the burden on patients and caregivers. The clinical expert suggested an extended injection interval could also mean that clinicians may have spare capacity to provide timely diagnosis and treatment for new patients.

Place in Therapy

The clinical expert noted that, unlike other currently available treatments, and in addition to VEGF inhibition, faricimab also targets Ang-2. The clinical expert expected faricimab to be a first-line therapy in the treatment of nAMD, similar to existing anti-VEGF drugs. The expert added that a shift in the current treatment paradigm will be likely with the introduction of faricimab, given that faricimab is the first anti-VEGF with an approved maintenance interval of up to every 16 weeks, which can potentially fulfill unmet needs with regard to frequency of visits.

In the clinical expert’s opinion, it is not appropriate to recommend that patients try other treatments before initiating faricimab. Faricimab is expected to be prescribed as a first-line treatment for nAMD, and, as with any of the existing treatments, the earliest initiation is crucial to achieve the best clinical outcomes.

Patient Population

The clinical expert noted that patients with nAMD who have early and small neovascular lesions and signs of active CNV (evidence of IRF or SRF and blood accumulation in the macula) based on clinical assessment, OCT, and OCT-angiography are suitable candidates for faricimab. Faricimab can be used in patients who are treatment-naive or require a change in therapy due to inadequate response to other anti-VEGF drugs.

The clinical expert noted that patients with acute symptoms of visual loss and signs of active nAMD have the greatest need for treatment. However, based on the evidence available currently, the expert did not expect that a specific subpopulation of nAMD patients would be more likely to benefit from faricimab compared with other populations. Patients with extensive subretinal fibrosis and macular tissue damage, very poor baseline visual acuity, long disease duration, or unsuccessful therapy with an anti-VEGF for more than 2 years may not be suitable for treatment.

The clinical expert noted that misdiagnosis and underdiagnosis can occur in clinical practice, more often in non-subspecialty clinics where resources may be limited.

Assessing Response to Treatment

According to the clinical expert consulted by CADTH, after completion of the loading doses, follow-up assessments are performed every 6 to 8 weeks with a treat-and-extend approach to achieving the longest sustainable interval without recurrence as determined by clinical and OCT evaluations. Key assessment outcomes include change in visual acuity, as well as presence of IRF or SRF, and blood accumulation in the macular.

The clinical expert noted that, when assessing the magnitude of change in visual acuity, it is crucial to keep in mind that patients with better vision at baseline generally have less room for improvement than those with poor baseline vision. The clinical expert reported that there is no agreed-upon threshold that is indicative of a clinically meaningful change in visual acuity in all patients with nAMD. The clinical expert also noted that a realistic goal in patients who have structural tissue damage in the central macula from nAMD is to achieve visual acuity stabilization, rather than improvement.

The clinical expert indicated that the presence of IRF or SRF, and blood accumulation in the macula are indicators of active disease that prompt modification of treatment plans, and often involve reduction in injection interval.

The clinical expert noted that it generally takes at least 4 to 6 months to achieve an optimal response to therapy. Based on the clinical expert’s experience, the majority of patients can achieve stabilized vision and improved quality of life, and about 30% to 40% of patients can achieve visual acuity improvement.

Discontinuing Treatment

The clinical expert indicated that faricimab should be discontinued in patients with extensive subretinal fibrosis (disciform scarring) that occurred with vision loss to counting fingers or worse, in patients in whom disease progression could not be modified with faricimab therapy, or with end stage disease.

Prescribing Conditions

The clinical expert indicated that it would be most appropriate for retina specialists to prescribe and administer faricimab. However, in rural settings where access to a retinal specialist is challenging, trained ophthalmologists with experience and expertise in managing nAMD will allow for timely access to care and reduce caregiver burden.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups.

CADTH received input from 1 clinician group, the Canadian Retina Society.

The clinician group noted that there have been limitations in the current intravitreal anti-VEGF therapies used to treat nAMD. The clinician group identified an efficacy gap in real-world treatment outcomes compared to the outcomes observed in clinical trials due to the intense treatment burden associated with anti-VEGF therapy. The clinician group reported that visual outcomes in real-world practice are suboptimal, and the recent pandemic and other limitations to health care delivery have made it more difficult for patients to receive regular and intense treatment.

The clinician group identified durability and reduced treatment frequency as the most important unmet needs in nAMD treatment. Because patients need to be treated with current anti-VEGF therapy every 7 to 8 weeks after an intensive monthly loading treatment cycle, the clinician group reported that this puts a high burden on patients and their caregivers, who need time off work to attend appointments. The clinician group indicated that new treatments that require less-frequent injections would help reduce the treatment burden for these patients. Another unmet need for nAMD treatment identified by the clinician group is the development of fibrosis and atrophy due to poor disease control, which can result in vision loss in the long-term. To improve the long-term visual outcomes, the clinician group mentioned the need for drugs that can effectively dry the retina for a longer period and reduce the nAMD treatment burden. Last, the clinician group mentioned the need for drugs that could also maintain a high safety profile to minimize the risk of ocular complications.

The clinician group mentioned that the dual mechanism of faricimab, which targets both the VEGF-A and Ang-2 pathways that are critical in the development of retinal and choroidal vascular disease, differs from that other available drugs. The clinician group agreed that this drug can be considered a first-line treatment or rescue therapy for patients not responding to current nAMD treatment, while potentially reducing the treatment burden and providing an option to use longer treatment intervals than those associated with existing drugs.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may affect their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Response

AMD = age-related macular degeneration, VEGF = vascular endothelial growth factor; BCVA = best corrected visual acuity; CDEC = CADTH Canadian Drug Expert Committee; CEDAC = Canadian Expert Drug Advisory Committee; nAMD = neovascular aged-related macular degeneration; q.8.w. = every 8 weeks; q.10.w. = every 10 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; QALY = quality-adjusted life-year.

Clinical Evidence

The clinical evidence included in the review of faricimab is presented in 3 sections. The first section, the systematic review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review. The third section includes additional relevant studies that were considered to address important gaps in the evidence included in the systematic review.

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of faricimab 6 mg intravitreal injections for the treatment of nAMD in adults.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 5. Outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

The systematic review protocol was established before the granting of a Notice of Compliance from Health Canada.

Table 5: Inclusion Criteria for the Systematic Review

AE = adverse event; ATE = arterial thromboembolic event; CRT = central retinal thickness; HRQoL = quality of life; IOP = intraocular pressure; nAMD = neovascular age-related macular degeneration; NEI VFQ-25 = National Eye Institute Visual Functioning Questionnaire 25; SAE = serious adverse event; WDAE = withdrawal due to adverse event.

^aBevacizumab is an off-label treatment for nAMD.

The literature search was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist.¹⁴

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946 to February 21, 2022) via Ovid and Embase (1974 to February 21, 2022) via Ovid. All Ovid searches were run simultaneously as a multi-file search. Duplicates were removed using Ovid deduplication for multi-file searches, followed by manual deduplication in EndNote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concept was Vabysmo (faricimab). Clinical trials registries searched included the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform (ICTRP) search portal, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit the retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. Appendix 1 provides detailed search strategies.

The initial search was completed on February 22, 2022. Regular alerts updated the search until the meeting of the CADTH Canadian Drug Expert Committee on June 22, 2022.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist.¹⁵ Included in this search were the websites of regulatory agencies (US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. Appendix 1 provides more information on the grey literature search strategy. These searches were supplemented by reviewing bibliographies of key papers and through contacts with appropriate experts. In addition, the manufacturer of the drug was contacted for information regarding unpublished studies.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the pre-determined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

A focused literature search for NMAs dealing with Vabysmo (faricimab) and AMD was run in MEDLINE All (1946–) on February 22, 2022. No limits were applied.

Findings From the Literature

Three reports presenting data from 2 unique studies were identified from the literature for inclusion in the systematic review (Figure 1). The included studies are summarized in Table 6. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Table 6: Details of Included Studies

ADA = anti-drug antibody; AE = adverse event; Ang-2 = angengenesis-2; BCVA = best corrected visual acuity; BM = Bruch’s membrane; CFP = colour fundus photograph; CNV = choroidal neovascularization; CST = central subfield thickness; ETDRS = Early Treatment Diabetic Retinopathy Study; ILM = internal limiting membrane; LLD = low luminescence deficit; nAMD = neovascular age-related macular degeneration; NEI VFQ-25 = National Eye Institute Visual Functioning Questionnaire 25; OCT-A = optical coherence tomography–angiography; PED = pigment epithelial detachment; PK = pharmacokinetic; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 week; RPE = retinal pigment epithelial; VEGF-A = vascular endothelial growth factor A.

^aCNV activity was defined as showing evidence of subretinal fluid, subretinal hyperreflective material, or leakage.

^bCorresponds to 20/32 to 20/320 approximate Snellen equivalent.

^cExcept for appropriately treated carcinoma in situ of the cervix, non-melanoma skin carcinoma, and prostate cancer with a Gleason score of 6 or lower and a stable prostate-specific antigen for more than 12 months.

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Description of Studies

Two studies were included in the systematic review, TENAYA and LUCERNE.¹⁶ They were identically designed, phase III, multi-centre, randomized, double-blind, active-controlled, noninferiority trials that aimed to evaluate the efficacy, safety, durability, and pharmacokinetics of faricimab compared with aflibercept in treatment-naive patients with nAMD.

The TENAYA study (N = 671) was conducted in 149 sites in 15 countries (9 sites in Canada) and the LUCERNE study (N = 658) was conducted in 122 sites in 20 countries (Canada not included). Both trials consisted of a 28-day screening period, followed by a 112-week double-blind treatment period. On day 1 of the double-blind phase, eligible patients were assigned a randomization identification letter by an interactive web-based response system, then randomized to either the faricimab or aflibercept arm (1:1). Randomization was stratified by 3 baseline factors: baseline BCVA ETDRS letter (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), low luminance deficit (LLD; < 33 letters, and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world). Patients in both arms received assigned treatment up to and including week 108, and returned for a final visit at week 112 in the double-masked period. Study visits occurred every 4 weeks until the end of the study in both studies. The study design of both studies is illustrated in Figure 2. Note that both studies are ongoing and the primary analysis that included data up to week 48 (data cut-off for the TENAYA study was October 26, 2020, and for the LUCENRE study, October 5, 2020) was reviewed in this submission. The final analysis is planned to occur after week 112 (anticipated study completion date for the TENAYA study was August 2022, and for the LUCERNE study, September 2023).

Figure 2: Study Design Schematic for TENAYA and LUCERNE

IVT = intravitreal; Q8W = every 8 weeks; Q16W = every 16 weeks; PTI = personalized treatment interval; W = week.

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Populations

Inclusion and Exclusion Criteria

The key inclusion criteria of both studies included patients aged 50 years or older with treatment-naive, active CNV lesions (of any subtype) secondary to nAMD; BCVA scores of 78 to 24 letters using the ETDRS protocol (20/32 to 20/320 Snellen equivalent); a total CNV lesion size of less than or equal to 9 disc areas; and a CNV component area of greater than or equal to 50% of the total lesion size. The key inclusion and exclusion criteria of the trials are shown in Table 6. Only 1 eye was assigned as the study eye in the studies. If both eyes were eligible, the eye with the worse BCVA at baseline was selected.

Baseline Characteristics

A summary of baseline characteristics of the ITT population in both studies is shown in Table 7. The baseline characteristics of patients were balanced overall between the treatment arms within each study, except in the LUCERNE study, in which the mean time since diagnosis of nAMD was longer in the faricimab arm (3.2 months [SD = 14.5]) than in the aflibercept arm (1.7 months [SD = 4.5]), and the proportion of patients with an occult CNV lesion was higher in the faricimab arm (51.7%) than in the aflibercept arm (42.8%). The baseline characteristics were generally similar across the studies. Patients had a median age of 74 to 77 years, and the majority were female (> 57%), and White (> 82%). Most patients had occult CNV lesions (42.8% to 52%) in the subfoveal area (55.2% to 63.1%), with baseline BCVA scores of 73 to 55 letters (54.7% to 59.9%). Evidence of SRF and PED was present at baseline in the majority of patients (approximately 65% and 90%, respectively), while IRF was present in about half of patients.

Table 7: Summary of Baseline Characteristics in the TENAYA and LUCERNE Studies (ITT Population)

BCVA = best corrected visual acuity; CNV = choroidal neovascularization; ETRDS = Early Treatment Diabetic Retinopathy Study; IRF = intraretinal fluid; ITT = intention-to-treat; nAMD = neovascular aged-related macular degeneration; PCV = polypoidal choroidal vasculopathy; PED = pigment epithelial detachment; RAP = retinal angiomatous proliferation; SRF = subretinal fluid; SD = standard deviation.

^aMeasured by fundus fluorescein angiography.

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Interventions

In the TENAYA and LUCERNE studies, eligible patients were randomized in a 1:1 ratio to faricimab 6 mg or aflibercept 2 mg for a duration of 112 weeks.

In the faricimab arm, patients received faricimab 6 mg intravitreally every 4 weeks for 4 doses (day 1, week 4, week 8, and week 12) followed by maintenance doses of 6 mg at up to 16-week intervals (either every 8, 12, or 16 weeks). The maintenance doses between week 20 to week 60 were given at a fixed dosing interval, which was determined by disease severity assessments pre-determined to occur at week 20 and 24. Patients were considered to have active disease if any of the following criteria was met:

• increase of greater than 50 µm in CST compared with the average CST value over the previous 2 scheduled visits (weeks 12 and 16 for the week 20 assessment and weeks 16 and 20 for the week 24 assessment)

• increase equal to or greater than 75 µm in CST compared with the lowest CST value recorded at either of the previous 2 scheduled visits

• decrease of 5 or more ETDRS letters in BCVA compared with the average BCVA value over the previous 2 scheduled visits, due to nAMD disease activity

• decrease of 10 or more ETDRS letters in BCVA compared with the highest BCVA value recorded at either of the previous 2 scheduled visits, owing to nAMD disease activity

• presence of new macular hemorrhage, due to nAMD activity.

At week 20, patients with active disease received faricimab at that visit then every 8 weeks until week 60. At week 24, patients with active disease (excluding those with active disease at week 20) received faricimab at that visit then every 12 weeks until week 60. Some exceptions were given to patients who had significant nAMD disease activity at week 24 but did not meet the criteria of active disease, if the investigator deemed that treatment was warranted. A definition of “significant nAMD disease activity” was not reported. These patients received a faricimab injection at week 24 and continued with a fixed dosing interval of every 12 weeks until week 60. Patients who did not have active disease at weeks 20 or 24 received faricimab at week 28 and continued with a 16-week dosing interval until week 60. From week 60, faricimab was dosed according to a personalized treatment-interval algorithm (Table 8). Dosing intervals were maintained, extended, or reduced based on OCT, BCVA, and clinical assessment at drug dosing visits.

In the aflibercept arm, patients received aflibercept 2 mg intravitreally every 4 weeks for 3 doses (day 1, week 4, and week 8), followed by maintenance doses of 2 mg at a fixed interval of every 8 weeks until the end of the study.

Treatment assignments were masked to all patients, assessors, and investigators, except treatment administrators. A sham procedure, which involved pressing a needle-free syringe against the anesthetized eye to mimic an intravitreal injection, was performed on patients in both treatment arms at study visits (every 4 weeks) when no treatment was scheduled to preserve masking.

The following therapies were prohibited during both studies: systemic anti-VEGF therapy; systemic drugs known to cause macular edema (fingolimod and tamoxifen); intravitreal anti-VEGF drugs in the study eye; intravitreal, periocular (subtenon), steroid implants, chronic topical (ocular) corticosteroids in the study eye; concurrent use of any macular photocoagulation or photodynamic therapy with verteporfin in the study eye; and other experimental therapies (except vitamins and minerals).

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 9. These end points are further summarized in the following section. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 4.

Table 8: Personalized Treatment-Interval Algorithm (Week 60 to Week 108) in the TENAYA and LUCERNE Studies

BCVA = best corrected visual acuity; CST = central subfield thickness; ETRDS = Early Treatment Diabetic Retinopathy Study.

^aWhere stability is defined as a change of CST of less than 30 µm.

^bChange in BCVA should be attributable to nAMD disease activity (as determined by investigator).

^cRefers to macular hemorrhage owing to nAMD activity (as determined by investigator).

Note: Patients whose treatment interval is reduced by 8 weeks from every 16 weeks to every 8 weeks will not be allowed to return to a every 16 weeks interval during the study.

Source: TENAYA Study Protocol² and LUCERNE Study Protocol.²

Table 9: Summary of Outcomes of Interest Identified in the CADTH Review Protocol

AE = adverse event; BCVA = best corrected visual acuity; CRT = central retinal thickness; CST = central subfield thickness; ETRDS = Early Treatment Diabetic Retinopathy Study; ILM = internal limiting membrane; IRF = intraretinal fluid; NEI VFQ-25 = National Eye Institute Visual Functioning Questionnaire 25; PED = pigment epithelial detachment; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; RPE = retinal pigment epithelium; SAE = serious adverse event; withdrawal due to adverse event.

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Efficacy Outcomes

Change From Baseline in Visual Acuity

The change from baseline in BCVA (ETDRS letters) averaged over weeks 40, 44, and 48 was the primary end point in both studies. Secondary end points of visual acuity change included: proportion of patients gaining greater than or equal to 15, 10, 5, or 0 ETDRS letters in BCVA from baseline; and proportion of patients avoiding loss of greater than or equal to 15, 10, or 5 ETDRS letters in BCVA from baseline, all of which were reported as an average over weeks 40, 44, and 48. These outcomes were also assessed over time (i.e., at all assessment time points through week 48).

The BCVA score was measured based on the ETDRS visual acuity chart assessed at a starting distance of 4 m. ETDRS charts consist of 70 letters that are distributed across 14 rows. Each row contains 5 uniformly spaced characters of the same size. The level of difficulty increases between successive rows as the size of characters decreases. The BCVA score corresponds to the number of letters an individual can read from the ETDRS chart. The higher the BCVA score the better the visual acuity.¹⁷ Clinical trials supporting regulatory approval of previous anti-VEGF treatments for nAMD had the proportion of patients with a loss of fewer than 15 letters on the ETDRS charts (considered to be vision maintenance) as the primary end point.^18,19

Frequency of Injection

The mean number of injections received in the study eye through week 48, as well as the proportion of patients in the faricimab arm on an 8-, 12- or 16-week treatment interval among patients completing weeks 20 and 24, and week 48, were secondary outcomes of the studies.

Health-Related Quality of Life and Vision-Related Function

Both studies measured the change from baseline in NEI VFQ-25 composite scores over time as an exploratory end point. The NEI VFQ-25 measures vision-targeted quality of life and was administered by the interviewer in both studies. The questionnaire consisted of 25 items relevant to 11 vision-related constructs, as well as a single-item, general-health component. The overall composite score ranges between 0 to 100, with 0 representing worst vision-related function and 100 representing best vision-related function. A 15-letter change in visual acuity in the study eye (typically the worse-seeing eye) corresponds to a change of 3.90 to 4.34 points in the composite score.²⁰ For the better-seeing eye, the clinically relevant difference for the NEI VFQ-25 composite score based on a 3-line change is 7.35 to 8.18 points.²⁰

Blindness (Legal)

Legal blindness is defined as a BCVA of 20/200 or less measured with a Snellen chart.²¹ The proportion of patients with a BCVA Snellen equivalent of 20/200 (BCVA ETDRS ≤ 38 letters) or worse averaged over weeks 40, 44, and 48, and over time was measured in both studies.

Change in Central Retinal Thickness

The change from baseline in CST (ILM-RPE) averaged over weeks 40, 44, and 48 was a secondary outcome in both studies. The change from baseline in CST (ILM-RPE) over time (up to week 48) was also reported as a secondary outcome. This outcome was measured based on the distance between the ILM and RPE, whereas the change from baseline in CST (ILM-BM) over time, as measured by the distance between the ILM and Bruch’s membrane (BM), was an exploratory outcome of the studies. A reduction in CST is considered a favourable outcome in the treatment of nAMD; however, a minimal important difference has not been established.

Subretinal Fluid, IRF, and Pigment Epithelial Detachment

The proportions of patients with an absence of IRF, SRF, or PED over time (through week 48) were measured as secondary end points. The presence of IRF, SRF, and PED specifically in the central subfield (within 1 mm diameter centre of macular) were of interest. Both IRF and SRF are indicators of active disease that are routinely measured in clinical practice to evaluate clinical responses, according to the clinical expert consulted by CADTH. An absence of IRF has been shown to be associated with a better prognosis in nAMD,^22-24 while the prognostic values of SRF and PED have not been established.

Harms Outcomes

The safety analysis included AEs (ocular and non-ocular), SAEs (ocular and non-ocular), and AEs of special interest (infection transmitted by study drug and sight-threatening AE) that occurred through week 48. The occurrence of AEs was assessed at all assessment time points.

Statistical Analysis

Noninferiority Margin

In the TENAYA and LUCERNE studies, a noninferiority margin of 4 ETDRS letters was used in the primary outcome analysis, in which noninferiority would be demonstrated in the lower limit of the 95% CI of the difference in change in BCVA from baseline between the 2 treatment groups was greater than −4 ETDRS letters. The noninferiority margin of 4 ETDRS letters was determined based on data from nAMD studies comparing ranibizumab with a sham control (the MARINA study), verteporfin photodynamic therapy (the ANCHOR study), and aflibercept (the VIEW1 and VIEW2 studies). In the MARINA study, a noninferiority margin of 4 preserved about 70% of the least estimated benefit of ranibizumab over sham. The investigators also considered a loss of 5 letters (1 ETDRS line) between treatments to be generally clinically relevant, and therefore inferred that the noninferiority margin was fewer than 5 ETDRS letters.

Sample-Size Calculation

A sample-size calculation determined that 320 patients per treatment arm was required to demonstrate noninferiority between faricimab and aflibercept with respect to the change in BCVA from baseline averaged over weeks 40, 44, and 48 at an 1-sided significance level of 0.025 with a power of 90% using a 2-sample t-test, assuming a noninferiority margin of 4 ETDRS letters, equal efficacy between faricimab and aflibercept, an SD of ETDRS 14 letters, and a 10% dropout rate.

Statistical Analysis for Efficacy Outcomes

The primary outcome was change from baseline in BCVA averaged over weeks 40, 44, and 48. The primary analysis was based on the ITT population and performed using an MMRM, which included the change from baseline at weeks 4 to 48 as the response variable and was adjusted for treatment group, visit, visit-by-treatment group interaction, baseline BCVA (continuous), and stratification factors (baseline BCVA [≥ 74 ETDRS letters, 73 to 55 letters, and ≤ 54 letters], baseline LLD < 33 ETDRS letters and ≥ 33 letters], and region [US and Canada, Asia, and the rest of the world]), assuming an unstructured covariance structure. In the primary estimand, ICEs due to COVID-19 (study drug discontinuation, use of prohibited therapy, missed dose, and death) were handled using a hypothetical strategy in which all values were censored after the ICE, while ICEs not due to COVID-19 (study drug discontinuation due to AEs or lack of efficacy and use of prohibited therapy) were handled with a treatment policy strategy in which all observed values were used regardless of the occurrence of the ICE. Missing data for continuous outcomes were implicitly imputed by the MMRM model based on the assumption that data were missing at random (MAR). Missing data for categorical outcomes were not imputed.

Pre-specified subgroup analyses were conducted with respect to the primary outcome. The baseline BCVA subgroup (≥ 74 ETDRS letters, 73 to 55 ETDRS letters, and ≤ 54 ETDRS letters) was relevant to this review. No statistical testing was performed for treatment-by-subgroup interactions.

A pre-specified sensitivity analysis was performed using the same estimand and analysis method as the primary analysis, with the exception that a last observation carried forward imputation approach was used to account for missing BCVA data, as well as BCVA assessments that were censored after COVID-19–related ICEs. Six supplementary analyses using the per-protocol population, a multiple imputation method, and different analysis methods (analysis of covariance and trimmed mean) and strategies for handling ICEs (treatment policy strategy only and hypothetical strategy only) were performed to further evaluate the robustness of the evidence from the primary analysis. A summary of the statistical analyses of efficacy end points in both studies is shown in Table 10.

The analysis for secondary outcomes assessed data in the ITT population through week 48. Continuous secondary end points of interest in this review were analyzed using the same approach as the primary analysis. Binary secondary end points that assessed the proportion of patients in each treatment group and the difference in proportions between treatment groups were calculated by applying Cochran-Mantel-Haenszel (CMH) weights, stratified by baseline BCVA (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), baseline LLD (< 33 letters and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world). Change from baseline in NEI VFQ-25 composite scores over time (through week 48), an exploratory outcome, was analyzed using descriptive statistics (mean, SD, median, and range).

No adjustments were made for multiplicity for secondary end points and subgroup analyses.

Statistical Analysis for Harms Outcomes

The safety analysis was based on AEs recorded through week 48 and were summarized using descriptive statistics.

Table 10: Statistical Analysis of Efficacy End Points in the TENAYA and LUCERNE Studies

BCVA = best corrected visual acuity; CMH = Cochran-Mantel-Haenszel; CST = central subfield thickness; ETRDS = Early Treatment Diabetic Retinopathy Study; ILM = internal limiting membrane; ITT = intention-to-treat; LLD = low luminance deficit; LOCF = last observation carried forward; MMRM = mixed model for repeated measures; NA = not applicable; NEI VFQ-25; RPE = retinal pigment epithelium; SD = standard deviation.

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Analysis Populations

Results are reported for the following populations in the TENAYA and LUCERNE studies:

• ITT: All patients who were randomized in the study were included. Patients were assessed according to the treatment assigned at randomization. This analysis population served as the primary analysis set for all efficacy analyses.

• Per-protocol: All patients randomized in the study who received at least 1 dose of study treatment and who did not have a major protocol violation that affected the efficacy evaluation or the treatment-interval determination were included. Patients were assessed according to the actual treatment received. This analysis population was used for supplementary analysis for the primary efficacy end point.

• Safety-evaluable population: All patients who received at least 1 injection of either faricimab or aflibercept in the study eye were included. Patients were assessed according to the actual treatment received. This analysis population was used for safety analyses.

For this review, noninferiority will be assessed using the results from both the ITT and per-protocol analyses.

Results

Patient Disposition

A summary of patient disposition is shown in Table 11.

Of 989 screened patients in the TENAYA study, 671 patients were randomized: 334 patients (49.7%) and 361 patients (50.1%) to the faricimab and aflibercept arms, respectively. In the LUCERNE study, of 1,012 screened patients, 658 patients were randomized; 331 patients (50.3%) and 327 patients (49.7%) to the faricimab arm and aflibercept arm, respectively.

In the TENAYA study, the proportions of patients who discontinued the study before week 48 were comparable between treatment arms (4.5% in faricimab arm versus 4.2% in aflibercept arm), while in the LUCERNE study, the proportion was numerically higher in the aflibercept arm (5.5%) than in the faricimab arm (3.0%). The proportion of patients who discontinued study treatment before week 48 was numerically higher in the faricimab arm (7.8%) than in the aflibercept arm (4.5%) in the TENAYA study, while in the LUCERNE study, the proportions were similar in both arms. Withdrawal by patient was the most frequently reported reason for study discontinuation and study treatment discontinuation in both studies.

Table 11: Patient Disposition

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Exposure to Study Treatments

The mean duration of exposure was similar between treatment arms in both studies, ranging from 46.0 weeks to 46.4 weeks. The mean number of injections, and the proportion of faricimab-treated patients who were on an injection interval of 8, 12, or 16 weeks at week 48 were secondary efficacy outcomes of the studies. Table 13 provides additional details on frequency of injections.

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported below. Appendix 3 provides detailed efficacy data.

Change From Baseline in Visual Acuity

Change From Baseline in BCVA (Primary End Point)

The change from baseline in BCVA averaged over weeks 40, 44, and 48 was the primary outcome in both studies. The mean differences between the faricimab and aflibercept arms were 0.7 ETDRS letters (95% CI, −1.1 to 2.5) in the TENAYA study and 0.0 ETDRS letters (95% CI, −1.7 to 1.8) in the LUCERNE study (Table 12), both of which met the primary end point of noninferiority. A supplementary analysis in the per-protocol population found that the results aligned with those of the ITT analysis. The results of a sensitivity analysis and other supplementary analyses were also consistent with those of the primary analysis (Table 41).

A pre-specified subgroup analysis based on baseline BCVA suggested a similar mean change in visual acuity from baseline averaged over weeks 40, 44, and 48 between the faricimab and aflibercept arms in all subgroups of baseline BCVA (≥ 74 ETDRS letters, 73 to 55 ETDRS letters, and ≤ 54 ETDRS letters), as outlined in Table 42 in Appendix 3.

Proportion of Patients Who Gained 15 or More ETDRS Letters in BCVA From Baseline

The between-group differences in the proportions of patients who gained 15 or more ETDRS letters in BCVA from baseline over weeks 40, 44, and 48 with faricimab versus aflibercept were 4.3% (95% CI, −1.6% to 10.1%) in the TENAYA study and −2.0% (95% CI, −8.3% to 4.3%) in the LUCERNE study (Table 12).

Proportion of Patients who Avoided Losing 15 or More ETDRS Letters in BCVA From Baseline

The between-group differences in the proportions of patients who avoided losing 15 or more ETDRS letters in BCVA from baseline over weeks 40, 44, and 48 with faricimab versus aflibercept were 1.3% (95% CI, −2.2% to 4.8%) in the TENAYA study and −1.5% (95% CI, −4.4% to 1.3%) in the LUCERNE study (Table 12).

Other BCVA Outcomes

Other secondary outcomes of visual acuity change including the proportion of patients who gained 10 or more, 5 or more, or 0 or more ETDRS letters in BCVA; and the proportion of patients who avoided losing 10 or more, or 5 or more ETDRS letters in BCVA from baseline averaged over weeks 40, 44, 48, were generally comparable between treatment arms (Table 12).

Table 12: BCVA Outcomes Averaged Over Weeks 40, 44, and 48

BCVA = best corrected visual acuity; CI = confidence interval; CMH = Cochran-Mantel-Haenszel; ETDRS = Early Treatment Diabetic Retinopathy Study; ITT = intention-to-treat; LLD = low luminance deficit; MMRM = mixed model for repeated measures; SE = standard error.

^aAdjusted mean. The primary end point was analyzed using an MMRM, with the change from baseline in BCVA as the dependent variable, considering all available observations of BCVA score at all visits. The model was adjusted for treatment group, visit, visit-by-treatment group interaction, baseline BCVA (continuous), baseline BCVA (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), baseline LLD (< 33 letters and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world).

^bCMH-weighted estimate. The observed proportions and the differences in observed proportions were obtained by applying the CMH weight, stratified by baseline BCVA score (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), LLD (< 33 letters and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world).

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Frequency of Injection

In the TENAYA study, the mean number of treatment injections given through week 48, a secondary outcome, was 6.9 (SD = 0.63) in the faricimab arm and 7.8 (SD = 0.45) in the aflibercept arm. In the LUCERNE study, the mean number of treatment injections given through week 48 was 7.0 (SD = 0.53) in the faricimab arm and 7.9 (SD = 0.32) in the aflibercept arm.

In the TENAYA study, the proportions of patients in the faricimab arm on an 8-, 12-, or 16-week interval at week 48, a secondary outcome, were 20.3%, 34.0%, and 45.7% respectively, while in the LUCERNE study, the respective proportions were 22.2%, 32.9%, and 44.9% (Table 13).

Between-group analyses were not reported in either study.

Health-Related Quality of Life and Vision-Related Function

The change from baseline in NEI VFQ-25 composite scores was an exploratory outcome and no between-group comparisons were reported. As summarized in Table 14 at weeks 24 and 48, the mean composite scores were increased from baseline in both treatment groups in both studies.

Blindness (Legal)

The difference between treatment groups in the proportion of patients who progressed to legal blindness (BCVA Snellen equivalent of 20/200 or worse [BCVA ETDRS less than or equal to 38 letters] averaged over weeks 40, 44 and 48) (Table 15) was −0.5% (95% CI, −4.2% to 3.3%) in the TENAYA study, and 0.4% (95% CI, −3.6% to 4.4%) in the LUCERNE study.

Change in Central Retinal Thickness

The mean change from baseline in CST (ILM-RPE) was a secondary outcome, while the mean change from baseline in CST (ILM-BM) was an exploratory outcome.

As outlined in Table 16, the difference between treatment groups in the mean change from baseline in CST (ILM-RPE) averaged over weeks 40, 44, and 48 was −7.4 µm (95% CI, −15.7 to 0.8) in the TENAYA study and −6.4 µm (95% CI, −14.8 to 2.1) in the LUCERNE study. The differences between treatment groups in the mean change from baseline in CST (ILM-BM) averaged over weeks 40, 44, and 48 were −9.7 µm (95% CI, −22.7 to 3.4) and −6.9 µm (95% CI, −21.1 to 7.4) in the TENAYA and LUCERNE studies, respectively.

Subretinal Fluid, IRF, and Pigment Epithelial Detachment

The proportions of patients with absence of IRF, SRF, and PED were secondary outcomes of the studies (Table 17).

In the TENAYA study, the differences between treatment groups in the proportion of patients with absence of IRF at weeks 40, 44, and 48 between-treatment groups were 4.9% (95% CI, −1.5% to 11.3%), −9.5% (95% CI, −15.9% to −3%), and 8.7% (95% CI, 1.0% to 14.4%) respectively, while in the LUCERNE study, the differences between treatments groups were 7.0% (95% CI, 0.6% to 13.3%), −6.0% (95% CI, −12.4% to 0.4%), and 6.4% (95% CI, −0.0% to 12.9%), respectively.

In the TENAYA study, the differences between treatment groups in the proportions of patients with absence of SRF at weeks 40, 44, and 48 between-treatment groups were 11.2% (95% CI, 4.1% to 18.3%), −8.4% (95% CI, −15.6% to −1.2%), and 10.0% (95% CI, 2.6% to 17.3%) respectively, while in the LUCERNE study, the differences in proportions between treatment groups were 13.3% (95% CI, 6.1% to 20.6%), −10.7% (95% CI, −18.0% to 3.4%), and 10.4% (95% CI, 2.9% to 17.9%) respectively.

In the TENAYA study, the differences between treatment groups in the proportion of patients with absence of PED at weeks 40, 44 and 48 between-treatment groups were −1.8% (95% CI, −6.3% to 2.8%), −5.2% (95% CI, −9.2% to −1.2%), and −4.6% (95% CI, −8.3% to −1.0%) respectively, while in the LUCERNE study, the differences between treatment groups were −1.1% (95% CI, −5.3% to 3.0%), −5.0% (95% CI, −8.9% to −1.0%), and −3.2% (95% CI, −6.7% to 0.2%), respectively.

Table 13: Frequency of Injection Outcomes (Intention-to-Treat Population)

CI = confidence interval; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks.

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report¹⁰

Table 14: Change from Baseline in NEI VFQ-25 Composite Score Over Time (ITT Population)

ITT = intention-to-treat; NEI VFQ-25 = National Eye Institute Visual Functioning Questionnaire 25; SD = standard deviation.

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Table 15: Proportion of Patients With BCVA Snellen Equivalent of 20/200 (38 or Fewer BCVA ETDRS Letters) or Worse Averaged Over Weeks 40, 44 and 48 (ITT Population — CMH)

BCVA = best corrected visual acuity; CI = confidence interval; CMH = Cochran-Mantel-Haenszel; ETRDS = Early Treatment Diabetic Retinopathy Study; ITT = intention-to-treat; LLD = low luminance deficit.

^aCMH-weighted estimate. The observed proportions and the differences in observed proportions were obtained by applying Cochran-Mantel-Haenszel weight, stratified by baseline BCVA score (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), LLD (< 33 letters and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world)

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Table 16: Change From Baseline in CST (ILM-RPE) and CST (ILM-BM) Averaged Over Weeks 40, 44, and 48 (ITT Population — MMRM)

BM = Bruch’s membrane; CI = confidence interval; CST = central subfield thickness; ILM = internal limiting membrane; ITT = intention-to-treat; LLD = low luminance deficit; MMRM = mixed model for repeated measures; RPE = retinal pigment epithelium; SD = standard deviation; SE = standard error.

^aAdjusted mean. This secondary end point was analyzed using an MMRM, with the change from baseline in CST as the dependent variable, considering all available observations of BCVA score at all visits. The model was adjusted for treatment group, visit, visit- by-treatment-group interaction, baseline CST (continuous), baseline BCVA score (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), LLD (< 33 letters and ≥ 33 letters), and region (US and Canada, Asia, and the rest of the world).

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Table 17: Proportion of Patients With Absence of Intraretinal Fluid, Subretinal Fluid, or Pigment Epithelial Pigment Over Time (Intention-to-Treat Population — CMH)

CI = confidence interval; CMH = Cochran-Mantel-Haenszel; ITT = intention-to-treat.

^aCMH-weighted estimate. The observed proportions and the differences in observed proportions were obtained by applying Cochran-Mantel-Haenszel weight, stratified by baseline BCVA score (≥ 74 letters, 73 to 55 letters, and ≤ 54 letters), LLD (< 33 letters and ≤ 33 letters), and region (US and Canada, Asia, and the rest of the world).

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Harms

Only those harms identified in the review protocol are reported in the following section. Table 18 provides detailed harms data.

Adverse Events

In the TENAYA study, the proportions of patients reporting at least 1 ocular AE were 36.3% in the faricimab arm and 38.1% in the aflibercept arm, while in the LUCERNE study, the proportions were 40.2% and 36.2%, respectively. The most common ocular AEs were conjunctival hemorrhage, worsening nAMD, and dry eye. The proportions of patients who reported at least 1 non-ocular AE in the TENAYA study were 52.3% in the faricimab arm and 51.8% in the aflibercept arm. In the LUCERNE study, the proportions were 52.0% and 58.0%, in the respective treatment arms. The most common non-ocular AE was nasopharyngitis (4.9% to 8.3%) and urinary tract infection (3.0% to 5.1%)

Serious Adverse Events

Ocular SAEs were infrequently reported in both trials. In the TENAYA study, 1.2% of patients in the faricimab arm and 1.8% of patients in the aflibercept arm reported ocular SAEs, while in the LUCERNE study, 2.1% of patients in both arms reported ocular SAEs. The most common ocular SAE was worsening nAMD. In the LUCERNE study, non-ocular serious AEs were reported in 11.5% of patients in the faricimab arm and 14.7% of patients in the aflibercept arm, in comparison to 9.0% in the faricimab arm and 10.1% in the aflibercept arm in the TENAYA study. Congestive cardiac failure (0.3% to 1.2%) was reportedly the most common non-ocular SAE in the trials.

Treatment Discontinuation Due to Adverse Events

In the TENAYA and LUCERNE studies, 0.9% and 2.5% of patients treated with faricimab discontinued treatment due to AEs, respectively. In the TENAYA and LUCERNE studies, 2.4% and 0.3% of patients treated with aflibercept discontinued treatment due to AEs, respectively. No specific AE was identified to account for the majority of treatment discontinuations.

Mortality

Six deaths occurred in the TENAYA study, 5 of which involved patients who received faricimab and 1 involved a patient who received aflibercept. Eleven deaths occurred in the LUCERNE study: 4 in the faricimab arm and 7 in the aflibercept arm. No specific AE was identified to account for the majority of deaths in either arm.

Notable Harms

One patient (0.3%) reported endophthalmitis in the aflibercept arm of the LUCERNE study and none in the faricimab arms of either study. There was no report of retinal vasculitis in either study. Intraocular inflammation was reported in 1.5% and 0.6% in the faricimab and aflibercept arms, respectively, of the TENAYA study; and 2.4% and 1.8% in the respective arms of the LUCERNE study. Vitreous floaters were reported in 3.9% and 2.1% in the faricimab and aflibercept arms, respectively, of the TENAYA study; and 2.1% and 1.2% in the respective arms of the LUCERNE study. Arterial thromboembolic events were reported in 0.9% to 1.2% of patients.

Table 18: Summary of Harms (Safety-Evaluable Population)

NR = not reported.

^aFrequency of 2% or greater in at least 1 treatment arm.

^bExperienced by 1 or more patients in at least 1 treatment arm.

^cRefers to worsening of neovascular age-related macular degeneration.

^dAnti-Platelet Trialists’ Collaboration–defined arterial thromboembolic events, defined as nonfatal strokes or nonfatal myocardial infarctions, or vascular deaths (including deaths of unknown causes).

Source: TENAYA Clinical Study Report⁹ and LUCERNE Clinical Study Report.¹⁰

Critical Appraisal

Internal Validity

The TENAYA and LUCERNE studies were identically designed, randomized, double-blind, active-controlled, noninferiority, phase III trials comparing faricimab and aflibercept. The overall trial designs of the TENAYA and LUCERNE studies were appropriate for the objectives of the studies. There was no concern with regard to the method of randomization, which involved stratification by baseline BCVA, LLD, and geographic region, as well as the use of an interactive web-based response system for randomized assignment. The baseline characteristics of the study population were generally balanced between treatment arms and across studies, except in the LUCERNE study, in which longer mean times since diagnosis and higher proportions of patients with occult CNV lesions were observed in the faricimab arm than in the aflibercept arm. The clinical expert consulted by CADTH commented that the distribution of the CNV lesion type is not expected to confound the results. According to the clinical expert, the difference in mean time since diagnosis could cause results to favour the aflibercept arm, rather than the faricimab arm, and is therefore unlikely to affect the interpretation of the outcomes in the context of the studies. The methods of allocation concealment and blinding were appropriate. The use of sham injections to preserve blinding in patients was likely successful, according to the clinical expert, considering the procedure was done on anesthetized eyes.

The studies concluded faricimab was noninferior to aflibercept based on primary outcome analyses of the ITT population. It is generally preferred that a claim of noninferiority be based on agreement between both the ITT population and the per-protocol population for a more conservative approach in the context of noninferiority trials. However, a supplementary per-protocol analysis confirmed the conclusion of noninferiority in the primary ITT population. In addition, several sensitivity analyses by the sponsor and by the FDA confirmed the findings of each study.

The noninferiority margin of 4 ETDRS letters, which was chosen by the sponsor based on clinical trial data and clinical reasoning, aligned with recommended approaches. The clinical rationale was considered reasonable by the clinical expert consulted by CADTH. The enrolled sample sizes were adequate for the assessment of the primary outcome. Subgroup analyses were pre-specified; however, due to the lack of sample-size considerations, control for multiplicity, and statistical testing for treatment-by-subgroup interaction, no conclusions related to subgroup effects can be drawn. Similarly, the secondary and exploratory end points should be interpreted based on the lack of sample-size considerations and control for type I error.

According to the protocol, patients in the faricimab arms were assigned a maintenance treatment interval of every 12 weeks, if they had protocol-defined active disease at week 24. Exceptions were made for patients who did not fulfill the criteria of active disease but were deemed by the investigators to have significant nAMD disease activity. A fixed maintenance dosing interval of 12 weeks was assigned to these patients. The definition of “significant nAMD disease activity” was not reported in the studies and the proportion of patients in this category in each group was not specified, making it impossible to determine the impact on the results.

The proportion of patients who discontinued before week 48 was 4.3% in the TENAYA and LUCERNE studies, and the proportions were generally balanced between treatment arms within each study. Intercurrent events occurred in approximately 10% of patients in both studies, and the majority involved a missed dose related to COVID-19. Treatment policy strategy and hypothetical strategy were used to address non–COVID-19–related and COVID-19–related ICEs, respectively, in the primary estimand of the primary efficacy end point. The strategies were consistent with the approaches recommended by the FDA for ICEs.²⁵ Although the hypothetical strategy (i.e., ICEs due to COVID-19 were censored and imputed using the MMRM modelling and assuming missing data were MAR) is 1 of the approaches identified in the FDA guidance, the treatment policy strategy (i.e., including all data regardless of ICEs) would be the preferred approach to ensure all data were used and because it is not clear that the MAR assumption would be met. The FDA statistical review of the studies for faricimab likewise considered the treatment policy strategy to be the better of the 2 that were used.²⁶ However, the results of the supplementary analyses confirmed those of the primary estimand, suggesting the approach to handling ICEs was unlikely to have introduced bias.

The studies used implicit imputation by MMRM that assumed an MAR mechanism to account for missing data for continuous outcomes, while observed data with no imputation were used on missing categorical outcomes. No sensitivity analyses were conducted to assess the impact of ICEs and missing data on the secondary outcomes, resulting in unsubstantiated assumptions about the secondary analyses. The FDA statistical review also noted this as a limitation and conducted additional analyses.²⁶ The results of these additional analyses confirmed the original secondary results.

External Validity

Nine of 149 study sites included in the TENAYA study and none in the LUCERNE study were in Canada. The clinical expert consulted by CADTH considered the inclusion and exclusion criteria reflective of the eligibility criteria for anti-VEGF treatment in clinical practice. The studies included only treatment-naive patients, whom the clinical expert noted would likely be the majority of patients eligible for faricimab; however, a switch in anti-VEGF may be warranted in some cases. The efficacy of faricimab in treatment-experienced patients was not explored in these studies and is therefore unclear. The clinical expert commented that the baseline characteristics of the study populations were similar to those of patients with nAMD in Canada.

In consultation with the clinical expert, aflibercept was considered an appropriate comparator, representing the most commonly prescribed on-label anti-VEGF in Canada. The dosing regimen of aflibercept in the studies aligns with the product monograph dosing. However, aflibercept was given at a fixed interval of 8 weeks in the maintenance phase, which does not align with the treat-and-extend approach used by most clinicians for the treatment of nAMD in Canada. The clinical expert added that most patients receive aflibercept at 8-week intervals in clinical practice, which may lessen the uncertainty of the generalizability of study results.

In terms of clinical relevance of the outcomes assessed in the studies, BCVA change and disease activity indicated by the presence of IRF, SRF, and/or PED, were routinely assessed to evaluate treatment response in clinical practice, according to the clinical expert. Other outcomes such as CST and NEI VFQ-25 were infrequently measured outside of clinical research.

An outcome of key interest to clinicians and patients was the frequency of injections, which was measured as the proportion of patients in the faricimab arm on an 8-, 12-, or 16-week injection interval. The generalizability of this outcome in the primary analysis was uncertain because the method of interval assignment for faricimab in the maintenance phase has limited resemblance to the approach in clinical practice. The maintenance doses for faricimab were given at a fixed interval of either 8, 12, or 16 weeks between week 20 and week 60. Although the treatment interval was assigned based on disease activity at week 20 and week 24 using an approach that is intended to mimic the treat-and-extend protocol, the fact that intervals were fixed until week 60 is more rigid than would be expected in practice, according to the clinical expert. The dosing of faricimab was informed by phase II studies, according to the sponsor. The lack of flexibility to adjust the dosing interval increases the uncertainty of the generalizability of the frequency of injection outcomes presented, although it may have helped reduce internal validity issues by keeping the dosing consistent.

The clinical expert noted that the length of assessment in the primary analysis (48 weeks) was adequate for assessing the efficacy and safety of faricimab in the context of a noninferiority trial. However, the clinical expert indicated that at least 2 to 3 years of results are needed to gain confidence on the durability of faricimab.

The TENAYA and LUCERNE studies were the only phase III trials available to date that provide direct evidence comparing faricimab to other anti-VEGF drugs in patients with nAMD. There is no direct evidence comparing faricimab to some of the other anti-VEGF drugs currently used in Canadian practice (i.e., brolucizumab or bevacizumab), which represents an evidence gap.

Indirect Evidence

Objectives and Methods for the Summary of Indirect Evidence

An ITC was required because of a lack of studies directly comparing faricimab with treatments other than aflibercept (as discussed in the Systematic Review section) and ranibizumab (Other Relevant Evidence) currently in use in the Canadian setting for nAMD.

Search Methods

A focused literature search for NMAs dealing with faricimab and AMD was run in MEDLINE All (1946–) on February 22, 2022. No limits were applied. No published NMAs were identified.

Description of Indirect Comparison

One report that included ITCs was supplied by the sponsor.

An overview of the submitted ITC is presented in Table 19.

Table 19: Study Selection Criteria and Methods for the Sponsor-Submitted ITC

AE = adverse event, BCVA = best corrected visual acuity, CNV = choroidal neovascularization, CPT = centre point thickness, CST = centre subfield thickness, EDTRS = Early Treatment Diabetic Retinopathy Study, nAMD = neovascular age-related macular degeneration, HRQoL = health-related quality of life; Mac-TSQ = Macular Disease Treatment Satisfaction Questionnaire; NEI VFQ-25 = National Eye Institute Visual Functioning Questionnaire 25; NICE = National Institute for Health and Care Excellence; SAE = serious adverse event, VEGF = vascular endothelial growth factor.

Source: Sponsor-submitted NMA.²⁷

Methods of Sponsor-Submitted Indirect Treatment Comparison

Objectives

The objective of the ITC was to assess the comparative efficacy and safety of faricimab with the interventions listed in Table 20.

Study Selection Methods

A strategy was developed, and the search was conducted in MEDLINE, Embase, the Cochrane library, and abstracts of relevant conferences, and an internet search was made of relevant health technology assessment agencies, clinical trial registries, and key government or international bodies. The population of interest was adult patients (> 18 years of age) with nAMD. Studies of both treatment-naive and -experienced patients were included. The main intervention was defined as faricimab 6 mg by intravitreal injection every 8 to 16 weeks.

Table 20: Treatment Doses and Regimens Considered in the Indirect Treatment Comparison

IVT = intravitreal; PDS = port delivery system; PRN = as needed; PRNX = as needed and extend; q.4.w. = every 4 weeks; q.6.w. = every 6 weeks; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; q.24.w. = every 24 weeks; ST = standard treatment of photodynamic treatment with verteporfin for predominantly classic-type neovascular age-related macular degeneration, or intravitreal pegaptanib or sham treatment for occult or minimally classic-type neovascular age-related macular degeneration, pooled with sham arms. TREX = treat and extend.

Source: Sponsor-submitted NMA.²⁷

The criteria included studies published before the cut-off date of July 1, 2020; this search was updated in September of 2021 (no new trial data were added). Two reviewers independently screened the retrieved reports at 2 stages (titles and abstracts, and then full papers), and any disagreement was adjudicated by a third party. Final citations were verified by the project lead. Reasons for exclusion were documented. Data extraction was conducted by a single reviewer and quality was checked by a second reviewer. Details of the methods used to extract data from the included studies were described.

Quality assessment of the selected studies was carried out by 2 reviewers, with any disagreements resolved by discussion or additional referees. The quality (risk of bias) assessment of studies was conducted using the 7-criteria checklist provided in section 2.5 of the National Institute for Health and Care Excellence single-technology appraisal user guide.

Outcomes of interest are listed in Table 19. Visual acuity outcomes were defined as mean change from baseline in BCVA according to ETDRS letters. In addition, the proportion of patients gaining or losing 10 or more, or 15 or more letters on the ETDRS scale from baseline was captured. This was defined as the proportion of patients in mutually exclusive categories (≥ −15, > −15 to ≥ −10, > −10 to ≥ −5, > −5 to < 5, ≥ 5 to < 10, ≥ 10 to < 15, and ≥ 15). Anatomic outcomes included the mean change in retinal thickness as measured by CST. If the CST was missing from a study, but 1 or more anatomic outcomes were reported, the other value was used in the following order: CST, centre point thickness (CPT), and central retinal thickness (CRT). |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Safety outcomes included overall ocular AEs and ocular SAEs and discontinuation. No specific definitions for overall discontinuation, injection frequency, and AEs (overall ocular AEs and SAEs) were specified beyond the standard reporting in each trial.

The sponsor’s ITC reported outcomes at 2 end points: 12 months and 24 months. The 12-month outcome included any results reported between weeks 48 and 56, with 12 months or 1 year classified as 12 months. The 24-month outcome included any result reported between week 96 and week 104, 24 months, or 2 years.

Indirect Treatment Comparison Analysis Methods

An overview of the submitted ITC analysis methods is presented in Table 21.

The ITC compared faricimab with comparators for the available end points of visual acuity (BCVA and proportion of patients gaining or losing 10 or more, or 15 or more ETDRS letters), anatomic outcomes (retinal thickness), number of injections, ocular AEs, and all-cause treatment discontinuation at 12 months. The feasibility of pooling to create a network for analysis was pre-assessed based on study and patient characteristics.

The ITC was an NMA performed with a Bayesian approach. The model was a Bayesian comparison using a generalized linear model framework, with BCVA, retinal thickness, and number of injections using the identity link and proportion of patients gaining or losing 10 or more, or 15 or more ETDRS letters modelled using the probit and all other end points modelled using the logit link. Non-informative (vague) priors were planned for all parameters and alternative priors were considered if the planned priors did not produce sensible results or were too informative. A total of 10,000 iterations were used as burn-in followed by 40,000 iterations with a thinning parameter of 10; however, this was increased by a factor of 10 for ocular AEs, serious ocular AEs, meta-regressions on patient characteristics for BCVA, scheduled meta-regressions for CST, and number of injections to improve convergence. At least 2 parallel chains were run in all model fits. Convergence of the model was assessed using Brooks-Gelman-Rubin diagnostics.

Model fit was assessed using the deviance information criterion (DIC), for which a 5-point difference was considered meaningful, and residual deviance. However, random-effects models were planned for all end points given the anticipated heterogeneity between included studies. Ranibizumab 0.5 mg by intravitreal injection every 4 weeks was the reference treatment.

Sensitivity analyses were conducted with Bayesian fixed-effects models and standard NMA models.

Two types of meta-regression were conducted to investigate the effect of drug administration schedule and patient characteristics. The impact of pooling arms with the same treatment and dose for the following dose schedules were investigated using:

• as needed versus routine monthly schedules

• loading yes versus no for as-needed schedules (loading)

• treat-and-extend versus a monthly schedule

• as needed with potential to extend versus routine monthly schedule

• the interval between injections (coded to a continuous month variable)

• combinations of this list.

Different dosing schedules were treated as different treatment arms in the initial NMA (“split model”), and the impact of pooling arms with the same treatment and dose and applying shared effects for dosing schedules was investigated in a pooled model. Patient age, gender, race, and BCVA at baseline were investigated using standard network meta-regression methods to determine if the treatment effect varied according to covariate. These were modelled in a split model; they were planned to be investigated in the pooled model along with chosen schedule covariates.

As a sensitivity analysis, a standard NMA was conducted for BCVA, CST, and proportion of patients gaining or losing 10 or 15 letters.

Every dosing regimen explored was considered as a separate node in the evidence network.

Table 21: Analysis Methods for Indirect Treatment Comparison

BCVA = best corrected visual acuity, CST = central subfield thickness; ETRDS = Early Treatment Diabetic Retinopathy Study; ITC = indirect treatment comparison; NICE = National Institute for Health and Care Excellence.

Source: Sponsor-submitted NMA.²⁷

Results of Indirect Treatment Comparison

Summary of Included Studies

A total of 67 RCTs were included in the feasibility assessment; 14 were excluded due to treatment-related reasons. Five studies were excluded as they did not report outcomes at the time points of interest, 1 was not a RCT, and 2 included polypoidal choroidal vasculopathy. Therefore, 45 studies were considered for inclusion in the NMA. A further 8 trials were excluded for reasons including not reporting outcomes at the time point of interest (1 trial), dosing outside the approved range (4), a dose schedule that was not of interest (1), and a study end point that was not of interest (1), for a total of 37 trials that were included in the NMA. Additionally, 1-year and 2-year analyses of the CATT trial were considered separate studies in some analyses, and a combined analysis of the VIEW 1 and VIEW 2 trials was considered a separate study in some analyses, for a total of 39 separate studies in the NMA.

An overview of the assessment of homogeneity for the ITC is presented in Table 22. Most (32 of the 37 trials) were head-to-head trials, except 5 trials that compared active treatments to sham intravitreal injections (1 study included standard therapy of photodynamic therapy with verteporfin, pegylated intravitreal injection or sham). Most of the trials (18 of 37) were double-masked, 11 were open-label, 5 were single-masked, and 3 did not report masking. Most (17 of 37) were phase III studies, 3 were phase II, 6 were phase IV, 1 was a phase I and II study, and 10 studies did not report phase. Most (23 of 37) were multi-centre trials, with another 8 muti-centre international trials and 6 single-centre trials. Study size ranged from 23 to 1,240 patients; there were 9 trials with fewer than 100 patients and 5 trials with more than 1,000 patients. The study year ranged from 2003 to 2022, with 4 trials not reporting dates.

Mean age at baseline in a study treatment arm ranged from 63.5 years old to 84 years old. The proportion of males across the study treatment arms ranged from 25% to 66%. The proportion of study participants who were White ranged from 0% to 100%. Mean BCVA at baseline ranged from 32.7 to 75.5 ETDRS letters. There was heterogeneity in the way retinal thickness was measured and defined in the included trials (central foveal thickness; CRT; CST, CPT, retinal thickness, and foveal thickness). The retinal thickness at baseline varied widely, from 176 μm (CPT) to 533 μm (CRT). Time since diagnosis was available for only 6 trials. Information about the presence of IRF data was available for 12 trials (37.7% to 100%) and information about the presence of SRF was available for 11 trials (58% to 90.9%). Prior therapy for nAMD was reported as 0 for 21 trials, not reported in 10 trials, and included various therapies (e.g., laser photocoagulation, supplements, and photodynamic therapy) for 4 trials and anti-VEGF medication for 2 trials of the ranibizumab port delivery system. It was noted in the systematic literature review that many studies reported the mean number of injections, although limited variance data were available.

Table 22: Assessment of Homogeneity for Sponsor-Submitted ITC

BCVA = best corrected visual acuity; CRT = central retinal thickness; CST = central subfield thickness; ITC = indirect treatment comparison; nAMD = neovascular age-related macular degeneration; VEGF = vascular endothelial growth factor; vs. = versus.

Source: Sponsor-submitted NMA.²⁷

Results

Comparators

Overall, the ITC included trials with relevant comparators. However, results for trials that included the ranibizumab port delivery system are not reported in this summary because it is not a treatment that was pre-specified as relevant to the review of faricimab. In addition, pairwise results between faricimab and sham/placebo or “standard therapy” as a comparator are not reported in this summary.

Risk of Bias

The quality of the 69 studies considered for inclusion in the feasibility assessment was reported. Overall, included studies were rated to be of moderate to high quality. However, a quality assessment specifically for the 37 studies that were included in the NMA was not reported.

Best Corrected Visual Acuity

For the outcome of BCVA at 12 months, 35 trials were included in the analysis, which was conducted under a random-effects model as this model had a lower DIC than the fixed-effects model (DIC not reported). A graphic representation of the evidence network is presented in Figure 3. Most trials compared active treatments; the network did include closed loops involving faricimab as an intervention.

All comparisons included the null in the 95% CrI. The comparative results are outlined in Table 23. There was no evidence of inconsistency from the comparison of the consistency and inconsistency model fits (DIC and mean residual deviance were compared).

Number of Injections at 12 Months

For the number of injections at 12 months, 27 RCTs were included in the analysis, which was conducted with a random-effects model. A graphic representation of the evidence network is presented in Figure 4. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Results are outlined in Table 24.

Table 23: Mean Change in BCVA — Indirect Treatment Comparison Results

AFL = aflibercept; BCVA = best corrected visual acuity score mean change from baseline at 12 months; BEV = bevacizumab; BRO = brolucizumab; CrI = credible interval; IVT = intravitreal; NMA = network meta-analysis; PRN = as needed; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; RAN = ranibizumab; TREX = treat and extend; vs. = versus.

Note: The sponsor reports that sensitivity analyses with standard NMA showed similar results; no evidence of inconsistency was observed from comparison of consistency and inconsistency models

Source: Sponsor-submitted NMA.²⁷

Table 24: Number of Injections at 12 Months — Indirect Treatment Comparison Results [Redacted]

AFL = aflibercept; BCVA = best corrected visual acuity score mean change from baseline at 12 months; BEV = bevacizumab; BRO = brolucizumab; CrI = credible interval; FAR = faricimab; IVT = intravitreal; NMA = network meta-analysis; PBO = placebo; PRN = as needed; PRNX = as needed and extend; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; q.4.w. = every 12 weeks; q.16.w. = every 16 weeks; q.24.w. = every 24 weeks; RAN = ranibizumab; TREX = treat and extend.

Note: Redacted rows have been deleted.

Source: Sponsor-submitted NMA.²⁷

Figure 3: Network Diagram for the Outcome Mean Change in BCVA at 12 Months

AFL = aflibercept; BEV = bevacizumab; BRO = brolucizumab; FAR = faricimab; IVT = intravitreal; PDS = port delivery system; PBO = placebo; PRN = as needed; PRNX = as needed and extend; Q4W = every 4 weeks; Q6W = every 6 weeks; Q8W = every 8 weeks; Q12W = every 12 weeks; Q16W = every 16 week; Q24W = every 24 weeks; RAN = ranibizumab; ST = standard treatment of photodynamic treatment with verteporfin for predominantly classic-type neovascular age-related macular degeneration, or intravitreal pegaptanib or sham treatment for occult or minimally classic-type neovascular age-related macular degeneration, pooled with sham arms. TREX = treat and extend.

Source: Sponsor-submitted NMA.²⁷

Figure 4: Network Diagram for the Outcome Mean Number of Injections at 12 Months

AFL = aflibercept; BEV = bevacizumab; BRO = brolucizumab; FAR = faricimab; IVT = intravitreal; PDS = port delivery system; PBO = placebo; PRN = as needed; PRNX = as needed and extend; Q4W = every 4 weeks; Q6W = every 6 weeks; Q8W = every 8 weeks; Q12W = every 12 weeks; Q16W = every 16 weeks; Q24W = every 24 weeks; RAN = ranibizumab; TREX = treat and extend.

Source: Drug Reimbursement Review submission: Vabysmo (faricimab single-use pre-filled syringe, 6 mg/0.05 mL solution for intravitreal injection) [CONFIDENTIAL sponsor’s submission].²

Retinal Thickness

For retinal thickness at 12 months, the sponsor included 25 RCTs in the analysis, which was conducted with a random-effects model. A graphic representation of the evidence network is presented in Figure 5. The results showed faricimab 6 mg intravitreal every 8 to 16 weeks may have had greater effect on retinal thickness (the 95% CrI did not include the null) when compared to bevacizumab regimens and ranibizumab 0.5 mg intravitreal every 8 weeks. Brolucizumab 6 mg intravitreal every 8 or 12 weeks was found to have a mean difference in retinal thickness of 32.86 (95% CrI, 9.71 to 55.61), which was greater than that of faricimab. Results are outlined in Table 25.

Patients Gaining or Losing at least 10 or 15 ETDRS Letters

For the outcome of patients gaining or losing at least 10 or 15 ETDRS letters at 12 months, 31 trials were included for this analysis, which was conducted with a random-effects model. A graphic representation of the evidence network is presented in Figure 6. The results showed faricimab 6 mg intravitreal every 8 to 16 weeks was associated with vision change that was not different to all regimens (the 95% CrI included the null). Results are outlined in Table 26.

The technical report noted that various assessments showed the model for this outcome had poor fit, likely due to limited data and heterogeneity. Various methods were used to adjust for the limitations, but these did not improve the model fit. This precludes using these data to make conclusions about the effect of faricimab versus comparators on patients gaining or losing at least 10 or 15 EDTRS letters.

Table 25: Retinal Thickness — Indirect Treatment Comparison Results

AFL = aflibercept; BCVA = best corrected visual acuity score mean change from baseline at 12 months; BEV = bevacizumab; BRO = brolucizumab; CrI = credible interval; IVT = intravitreal; NMA = network meta-analysis; PRN = as needed; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; RAN = ranibizumab; TREX = treat and extend; vs. = versus.

Notes: Bolded numbers indicate that the 95% CrI excludes the null.

Sensitivity analysis with various definition of retinal thickness was not reported; sensitivity analyses with standard NMA showed similar results; as the inconsistency assessment for BCVA did not show any inconsistency, an inconsistency assessment was not reported for this end point

Source: Sponsor-submitted NMA.²⁷

Table 26: Patients Losing or Gaining at Least 10 or 15 Letters ETDRS — ITC Results

AFL = aflibercept; BCVA = best corrected visual acuity score mean change from baseline at 12 months; BEV = bevacizumab; BRO = brolucizumab; CrI = credible interval; IVT = intravitreal; NMA = network meta-analysis; PRN = as needed; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; RAN = ranibizumab; TREX = treat and extend; vs. = versus.

Notes: Sensitivity analyses with standard NMA showed similar results; as the inconsistency assessment for BCVA did not show any inconsistency, an inconsistency assessment was not reported for this end point.

The technical report indicated the posterior for tau is truncated at the upper limit of the uniform prior (5) suggesting it is not well estimated; this was not resolved by increasing the upper limit, suggesting that there may be insufficient data to estimate between-study heterogeneity and there is uncertainty in these results

Source: Sponsor-submitted NMA.²⁷

Figure 5: Network Diagram for the Retinal Thickness Outcome at 12 Months

AFL = aflibercept; BEV = bevacizumab; BRO = brolucizumab; FAR = faricimab; IVT = intravitreal, loading = with loading phase; PDS = port delivery system; PRN = as needed; PRNX = as needed and extend; Q4W = every 4 weeks; Q6W = every 6 weeks; Q8W = every 8 weeks; Q12W = every 12 weeks; Q16W = every 16 weeks; Q24W = every 24 weeks; RAN = ranibizumab; ST = standard treatment of photodynamic treatment with verteporfin for predominantly classic-type nAMD, or intravitreal pegaptanib or sham treatment for occult or minimally classic-type neovascular age-related macular degeneration, pooled with sham arms. TREX = treat and extend.

Source: Sponsor-submitted NMA.²⁷

Figure 6: Network Diagram for the Outcome of Proportion of Patients Losing or Gaining at Least 10 or 15 Letters ETDRS at 12 Months

AFL = aflibercept; BEV = bevacizumab; BRO = brolucizumab; FAR = faricimab; IVT = intravitreal; PDS = port delivery system; PBO = placebo; PRN = as needed; PRNX = as needed and extend; Q4W = every 4 weeks; Q6W = every 6 weeks; Q8W = every 8 weeks; Q12W = every 12 weeks; Q16W = every 16 weeks; Q24W = every 24 weeks; RAN = ranibizumab; ST = standard treatment of photodynamic treatment with verteporfin for predominantly classic-type neovascular age-related macular degeneration, or intravitreal pegaptanib or sham treatment for occult or minimally classic-type neovascular age-related macular degeneration, pooled with sham arms. TREX = treat and extend.

Source: Sponsor-submitted NMA.²⁷

Ocular Adverse Events

For the outcome of all ocular AEs at 12 months, 14 trials were included in the analysis, which was conducted with a fixed-effects model as this model had a lower DIC compared with the random-effects model, despite not being considered meaningful in the technical report (DIC not reported). A graphic representation of the evidence network is presented in Figure 7. The results showed that faricimab 6 mg intravitreal every 8 to 16 weeks had ocular AEs that were not different to comparators (95% CrIs included 1 for the odds ratio of ocular AEs), although credible intervals were wide for some of the comparisons. It was reported that, due to events resulting in sparse data, a normal likelihood model on the odds ratio scale with continuity correction was applied for all AE analysis. The comparative results are outlined in Table 27.

An NMA for serious ocular AEs was not reported because limited data were available for these events.

Treatment Discontinuation

For the outcome of treatment discontinuation at 12 months, 25 trials were included in the analysis, which was conducted with a fixed-effects model as this model had a lower DIC than the random-effects model, despite not being considered meaningful in the technical report (DIC not reported). A graphic representation of the evidence network is presented in Figure 8. The results showed that faricimab 6 mg intravitreal every 8 to 16 weeks had discontinuation rates that were not different (95% CrIs include 1 for odds of discontinuation) to comparators, although the CrIs were wide for some comparisons. The comparative results are outlined in Table 28. A normal likelihood model was applied to this analysis due to sparse data from infrequent events.

Table 27: Ocular Adverse Events at 12 Months — Indirect Treatment Comparison Results

AFL = aflibercept; BCVA = best corrected visual acuity score mean change from baseline at 12 months; BEV = bevacizumab; BRO = brolucizumab; CrI = credible interval; FAR = faricimab; IVT = intravitreal; PRN = as needed; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; RAN = ranibizumab; TREX = treat and extend.

Note: As the inconsistency assessment for BCVA did not show any inconsistency, an inconsistency assessment was not reported for this end point.

^aOdds ratio greater than 1 favours faricimab.

Source: Sponsor-submitted NMA.²⁷

Figure 7: Network Diagram for the Outcome Ocular Adverse Effects at 12 Months

AFL = aflibercept; BEV = bevacizumab; FAR = faricimab; IVT = intravitreal; PDS = port delivery system; PBO = placebo; PRN = as needed; Q4W = every 4 weeks; Q6W = every 6 weeks; Q8W = every 8 weeks; Q12W = every 12 weeks; Q16W = every 16 weeks; Q24W = every 24 weeks; RAN = ranibizumab; TREX = treat and extend.

Source: Sponsor-submitted NMA.²⁷

Figure 8: Network Diagram for Outcome Treatment Discontinuation at 12 Months

AFL = aflibercept; BEV = bevacizumab; BRO = brolucizumab; FAR = faricimab; IVT = intravitreal; PDS = port delivery system; PBO = placebo; PRN = as needed; PRNX = as needed and extend; Q4W = every 4 weeks; Q6W = every 6 weeks; Q8W = every 8 weeks; Q12W = every 12 weeks; Q16W = every 16 weeks; Q24W = every 24 weeks; RAN = ranibizumab; ST = standard treatment of photodynamic treatment with verteporfin for predominantly classic-type neovascular age-related macular degeneration, or intravitreal pegaptanib or sham treatment for occult or minimally classic-type neovascular age-related macular degeneration, pooled with sham arms. TREX = treat and extend.

Source: Sponsor-submitted NMA.²⁷

Table 28: Treatment Discontinuation at 12 Months — ITC Results

AFL = aflibercept; BEV = bevacizumab; BRO = brolucizumab; CrI = credible interval; IVT = intravitreal; PRN = as needed; PRNX = as needed and extend; q.4.w. = every 4 weeks; q.6.w. = every 6 weeks; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; RAN = ranibizumab; TREX = treat and extend.

Note: As the assessment for BCVA did not show any inconsistency, an inconsistency assessment was not reported for this end point; a normal likelihood on the odds ratio scale with continuity correction was applied due to rare events; this may be a source of bias.

^aOdds ratio greater than 1 favours faricimab.

Source: Sponsor-submitted NMA.²⁷

Critical Appraisal of Sponsor-Submitted Indirect Treatment Comparison

The research question and inclusion criteria for the systematic review were reported in the ITC and feasibility assessment. The ITC NMA was based on a systematic literature review that identified studies according to pre-specified inclusion criteria. A comprehensive and transparent approach to the systematic review was provided, including the search strategy and the use of several databases. The literature search was comprehensive, involving multiple databases (i.e., Embase, the Cochrane Register of Controlled Trials, MEDLINE, clinical trial registries (e.g., clinicaltrials.gov), and conferences). The literature search was well reported, with a complete copy of the search strategy included in the report. Overall, study selection and data extraction methods were appropriate. Study selection was performed by 2 reviewers. Data extraction was performed by 1 reviewer and verified by a second reviewer independently. Any disagreements were resolved through consensus. A comprehensive list of inclusion and exclusion criteria for the studies in the systematic literature review was included. A risk of bias evaluation for the studies included in the systematic literature review was performed, based on tools that considered the appropriateness of randomization and allocation concealment, similarity at baseline across treatment groups in prognostic factors, masking, imbalances in dropouts, outcomes reporting, and ITT analysis. Whether risk of bias assessments were performed in duplicate was not reported. However, the risk of bias assessment for the studies ultimately included in the NMA was not presented and how the results of the quality appraisal factored into the NMA (e.g., sensitivity analyses excluding studies rated with a high risk of bias) was also not reported.

The inclusion criteria would allow for a population that is relevant to Canadian settings. The comparisons reported in this ITC have generally incorporated relevant treatments for Canadian settings, including treatments that have extensive clinical use but lack a formal review from Health Canada, such as bevacizumab, which is commonly used in Canada. However, the ranibizumab port delivery system was included, which was not considered relevant to the review of faricimab.

The degree of heterogeneity between the included studies was difficult to assess because of incomplete reporting of study and patient characteristics. Some study characteristics were inadequately reported in the ITC. Description of trial design, sample size and duration, and country were reported. However, the ITC failed to report on allocation concealment, methods used for handling missing data, and eligibility (inclusion and exclusion). There was considerable variability in study design, year of conduct, and sample size. Four phase II trials were included in the NMA; these trials included arms that were pooled with phase III trials. The degree of similarity between these trials was not reported, nor were the results of a sensitivity analysis examining the effects of pooling reported.

Baseline patient characteristics were reported in the ITC for all 37 studies that were included in the ITC. Clinical trial eligibility criteria were not described for the trials ultimately included in the NMA. Many individual studies failed to report or inadequately reported patient characteristics, resulting in gaps in the extracted ITC data. Information about key baseline characteristics, such as the presence of IRF and SRF, duration of nAMD and use of prior therapies, was lacking. Of those factors that were reported, baseline values were heterogeneous across studies. There was also heterogeneity in baseline patient characteristics, including age, sex distribution, race, and geographic location, and in the reporting of methods for measuring and in results of changes in retinal thickness. The apparent heterogeneity of the factors that were reported combined with the inability to assess those that were not reported means there is considerable uncertainty as to whether the assumptions related to homogeneity were met. The clinical expert consulted by CADTH agreed that patient populations and study methodologies were heterogenous. The technical report notes that there was no evidence that the treatment effect differed by patient characteristics (age, sex, ethnicity, and baseline BCVA) or that model fit was improved by patient characteristic meta-regressions. However, meta-regression was not performed for all relevant patient and study characteristics. In addition, the technical report states that “random-effects models were used as the principal analyses for all end points as absence of heterogeneity is implausible.” Furthermore, the technical report notes that, for some outcomes, such as the proportion of patients gaining or losing at least 10 or 15 ETDRS letters at 12 months, there were insufficient data to estimate between-study heterogeneity, making the models unstable and the results uninterpretable. Despite acknowledging the degree of heterogeneity, the technical report did not provide sufficient information of assessments of heterogeneity (e.g., graphic representation of baseline characteristics, and statistical tests) to fully understand the sources of heterogeneity. Heterogeneity may therefore have influenced the comparative efficacy and safety estimates, and it is not possible to quantify or identify the direction of the bias.

The technical report notes that there was no evidence that the treatment effect differed by patient characteristics or that model fit was improved with the patient characteristic meta-regressions, although these results were not provided. The schedule meta-regression using all schedule covariates had a lower DIC compared to the base-case separate-treatments model, but a poorer fit. Consequently, analyses of other outcomes used both the separate-treatments model (each treatment and schedule were treated as a separate node) and (for efficacy outcomes) the meta-regression with all schedule covariates. The meta-regression NMA is the base case for efficacy outcomes. The model fit was good in most cases, and where fit was poor (patients gaining or losing at least 10 or 15 ETDRS letters at 12 months), it is described in the relevant section.

Additional limitations to the ITC include the following:

• The connection between faricimab and the rest of the network was weak: faricimab is only connected to the network through aflibercept via the LUCERNE and TENEYA trials, and through ranibizumab in a phase II trial.

• The networks include a large number of interventions, as every dosing regimen explored was considered a separate node. Although there are some closed loops for some networks, overall, the nodes were connected by few trials.

• The geometry of the networks likely contributed to uncertainty in the estimates for models the level of imprecision in certain comparisons as evidenced by wider credible intervals.

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| However, the results of the analysis related to number of injections may have been affected by protocol-driven administration of therapies with fixed intervals in clinical trials. In addition, there were minimal data on the variance for the number of injections.

The ITC suggests that faricimab 6 mg intravitreal every 8 to 16 weeks may be favourable (the 95% CrIs exclude the null) to bevacizumab regimens and ranibizumab 0.5 mg intravitreal every 8 weeks for the outcome of mean change in retinal thickness (CST). Additionally, brolucizumab 6 mg intravitreal every 8 to 12 weeks may be favourable (the 95% CrI excludes the null) to faricimab for this outcome. It is not clear from the sponsor’s report how the values for this outcome compared across the trials. This variability in how retinal thickness was defined, measured, and reported may contribute considerable heterogeneity to the ITC. However, heterogeneity in the methods to assess retinal thickness across studies adds considerable uncertainty to the results for this analysis and limits conclusions about the relative effect of faricimab on CRT.

As limited data were available for the NMAs conducted for ocular AEs and for treatment discontinuation, fixed-effects models were used for these end points, and there was a high degree of statistical uncertainty in these models. Limited data are therefore available from which to draw any conclusions about the effect of faricimab versus comparators on ocular AEs and treatment discontinuation.

Other Relevant Evidence

This section includes 2 phase II studies, which are summarized and appraised because they provide the only head-to-head comparisons between faricimab and ranibizumab.

Sponsor-Submitted Phase II Studies (STAIRWAY and AVENUE)

Description of Studies

The characteristics of the STAIRWAY and AVENUE studies are presented in Table 29. The STAIRWAY trial had 3 treatment arms, whereas the AVENUE trial had 5 arms. Results for arm B of the AVENUE study will not be presented in this report because the dosing regimen is outside the Health Canada–recommended dose for faricimab. The primary objective of the STAIRWAY trial was to evaluate the efficacy of faricimab compared to ranibizumab on visual acuity from baseline to week 40. The primary objective of the AVENUE study was to evaluate the efficacy of faricimab compared to ranibizumab monotherapy in treatment-naive patients from baseline to week 36, and in anti-VEGF incomplete-responder patients from week 12 to week 36.

Overviews of the study designs for the STAIRWAY and AVENUE trials are shown in Figure 9 and Figure 10, respectively.

Table 29: Details of STAIRWAY and AVENUE Studies

ADA = anti-drug antibody; AE = adverse event; AMD = age-related macular degeneration; BCVA = best corrected visual acuity; CFT = central foveal thickness; CNV = choroidal neovascularization; CST = central subfield thickness; ETDRS = Early Treatment Diabetic Retinopathy Study; FFA = fundus fluorescein angiography; IVT = intravitreal; OCT = optical coherence tomography; q.4.w. = every 4 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; nAMD = neovascular age-related macular degeneration; VEGF = vascular endothelial growth factor.

^aCNV activity was defined as showing evidence of subretinal fluid, subretinal hyperreflective material, or leakage.

^bCorresponds to 20/40 to 20/320 approximate Snellen equivalent.

^cOther causes include ocular histoplasmosis, pathological myopia, trauma, choroidal rupture, angioid streaks, or uveitis.

^dOnly 1 eye was chosen as the study eye in each trial.

Source: STAIRWAY Clinical Study Report³³ and AVENUE Clinical Study Report.³⁴

Figure 9: Study Design Schematic for STAIRWAY Trial

IVT = intravitreal; Q4W = every 4 weeks; Q12W = every 12 weeks; Q16W = every 16 weeks; W = week.

Source: STAIRWAY Clinical Study Report.³³

Figure 10: Study Design Schematic for AVENUE Trial

E = exit visit; IVT = intravitreal; Q4W = every 4 weeks; Q8W = every 8 weeks; RBZ = ranibizumab.

Source: AVENUE Clinical Study Report.³⁴

Baseline Characteristics

The baseline characteristics of the studies are presented in Table 30 and Table 31. Characteristics were generally similar between treatment groups in each study and between studies.

Table 30: Summary of Baseline Characteristics in STAIRWAY (ITT Population)

BCVA = Best corrected visual acuity; CNV = choroidal neovascularization; ETRDS = Early Treatment Diabetic Retinopathy Study; IRF = intraretinal fluid; ITT = intention-to-treat; PED = pigment epithelial detachment; q.4.w. = every 4 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; SRF = subretinal fluid; SD = standard deviation.

^aMeasured by fundus fluorescein angiography.

Source: STAIRWAY Clinical Study Report.³³

Table 31: Summary of Baseline Characteristics in AVENUE (All Patients)

AMD = aged-related macular degeneration; BCVA = best corrected visual acuity; CNV = choroidal neovascularization; ETRDS = Early Treatment Diabetic Retinopathy Study; PCV = polypoidal choroidal vasculopathy; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; RAP = retinal angiomatous proliferation; SD = standard deviation.

^aMeasured by fundus fluorescein angiography.

Source: AVENUE Clinical Study Report.³⁴

Statistical Analysis

The analysis for the AVENUE study was based on a hypothesis test of superiority between faricimab and ranibizumab, with a null hypothesis of no difference between each of the faricimab treatment groups and the ranibizumab control group. The STAIRWAY study was not hypothesis-based, but focused on estimation of treatment effects between the 2 faricimab treatment schedules and ranibizumab. Both trials used an MMRM for primary efficacy analyses of the change in BCVA and generalized estimating equations for the difference in the proportion of patients who gained or lost 15 or more letters in BCVA. All measures of treatment effect were reported with 80% CIs in both trials.

Analyses in the AVENUE study were conducted on 2 populations: the treatment-naive population (population A, comprising treatment arms A, B, C, and D) and the anti-VEGF incomplete-responder population, also referred to as the treatment-experienced population (population C, comprising treatment arms A and E).

Results

Patient Disposition

Table 32 summarizes the disposition of patients for the STAIRWAY trial. A total of 137 patients were screened, and 76 (55.5%) were randomized into the trial in 3 arms. Table 33 summarizes the disposition of patients for the AVENUE trial. A total of 507 patients were screened, and 68 (25.9%), 42 (16.0%), 47 (17.9%), and 69 (26.2%) patients were randomized to arms A, C, D, and E, respectively.

Efficacy

Given the designs and planned analyses of the trials, the summary of results in this report focuses on the studies’ primary objectives to assess the effects of treatment on BCVA (Table 34, Table 35, and Table 36).

In the STAIRWAY trial, the mean differences between the faricimab and ranibizumab arms in BCVA were −2.1 ETDRS letters (80% CI, −6.8 to 2.6) in arm A and 1.1 letters (80% CI, −3.4 to 5.5) in arm B at week 40, and 0.5 letters (80% CI, −4.3 to 5.3) in arm A and 1.8 letters (80% CI, −2.7 to 6.4) in arm B at week 52. In the AVENUE trial, the mean differences between the faricimab and ranibizumab arms in BCVA were −1.6 letters (80% CI, −4.9 to 1.7) in arm C and 1.5 letters (80% CI, −4.6 to 1.6) in arm D in the treatment-naive population A, and −1.7 (80% CI, −3.8 to 0.4) in the treatment-experienced population C in the AVENUE trial.

Table 32: Patient Disposition for the STAIRWAY Trial

q.4.w. = every 4 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks.

Source: STAIRWAY Clinical Study Report.³³

Table 33: Patient Disposition for the AVENUE Trial

q.4.w. = every 4 weeks; q.8.w. = every 8 weeks.

Source: AVENUE Clinical Study Report.³⁴

Table 34: BCVA Outcomes in STAIRWAY Trial

BCVA = best corrected visual acuity; CI = confidence interval; ITT = intention-to-treat; MMRM = mixed model for repeated measures; q.4.w. = every 4 weeks; q.12.w. = every 12 weeks; q.16.w. = every 16 weeks; SD = standard deviation.

^aMean BCVA change from baseline generated using a mixed model for repeated measures. Model includes categorical covariates of treatment group, visit, and visit-by-treatment-group interaction, and the continuous covariate of baseline BCVA.

Source: STAIRWAY Clinical Study Report.³³

Table 35: BCVA Outcomes in Treatment-Naive Population A of the AVENUE Trial (ITT Population)

BCVA = best corrected visual acuity; CI = confidence interval; ITT = intention-to-treat; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; SD = standard deviation.

^aProportions were estimated using a longitudinal generalized estimating equation model. Categorical time (visits), study arm, and their interaction were used as predictors in the model.

Source: AVENUE Clinical Study Report.³⁴

Table 36: BCVA Outcomes in the Treatment-Experienced Population C for AVENUE Trial (ITT population)

BCVA = best corrected visual acuity; CI = confidence interval; ITT = intention-to-treat; q.4.w. = every 4 weeks; SD = standard deviation.

Note: Baseline is defined as the last non-missing assessment before week 12 dosing. P values for differences of proportions computed using the Fisher exact test.

^aProportions were estimated using a longitudinal generalized estimating equation model. Categorical time (visits), study arm, and their interaction were used as predictors in the model.

Source: AVENUE Clinical Study Report.³⁴

Harms

The frequencies of AEs in the STAIRWAY and AVENUE studies are shown in Table 37 and Table 38, respectively. The harms and notable harms were similar across both phase II trials and in the pivotal trials. The total percentage of patients who experienced an AE in the STAIRWAY trial was higher in the ranibizumab arm than in the faricimab arms. However, the percentages of patients who experienced SAEs and who died were higher in the faricimab arms, with no events occurring in the ranibizumab arm. In the faricimab arms, the SAEs were non–ocular-related (cardiac disorders) and the 3 deaths were associated with ischemic stroke, sepsis, and metastatic neoplasm.