Lytenava (bevacizumab-vikg)

• To review the evidence for the safety and effectiveness of Lytenava (bevacizumab-vikg) in adults with neovascular (wet) age-related macular degeneration.
• To establish the clinical conditions under which Lytenava is authorized.
• To set out what the evidence shows about Lytenava relative to the comparator used in its trials, ranibizumab, and relative to off-label repackaged bevacizumab.
• The scope is limited to Lytenava and its own trials. Other intravitreal VEGF inhibitors appear only as the comparator arms of those trials.

• Wet age-related macular degeneration (wet AMD) is an eye disease of older adults in which abnormal blood vessels grow at the back of the eye and damage the macula, the part of the retina used for central vision. It is a less common form of late AMD, and it causes faster vision loss. (see 6)
• About 2 out of 100 Medicare beneficiaries have wet AMD, a share that was 2.14% in 2021. Without treatment the disease ends in scarring and blindness. (see 6)
• Before Lytenava, no form of bevacizumab was approved for use in the eye. Eye doctors used bevacizumab made for cancer treatment, repackaged into small syringes by compounding pharmacies, which carries a risk of contamination and of inconsistent strength. (see 6)
• Lytenava is bevacizumab made and approved for injection into the eye. The Food and Drug Administration (FDA) approved it for wet AMD in July 2026. A physician injects 1.25 mg into the eye once a month, about every 28 days, using a separate vial for each eye. (see 7.3, 7.4)
• In the main trial of 228 people, about 4 in 10 people given Lytenava every month gained 3 or more lines on an eye chart after 11 months, against about 2 in 10 people given ranibizumab, another approved eye injection, which was given 5 times in the year against 12 times for Lytenava. (see 8.1)
• In a second trial of 400 people, in which both medicines were given monthly, vision improved by 3.3 letters with Lytenava and 4.5 letters with ranibizumab after 8 weeks. That result did not meet the trial's goal of showing that Lytenava was not worse than ranibizumab. (see 8.2)
• The most common side effect was bleeding on the surface of the eye, in 4 out of 100 people. Any injection into the eye can cause an infection inside the eye, a detached retina, or a rise in eye pressure, so eye pressure is checked before and after each injection. (see 7.3, 10)
• The evidence is of low quality, because the main trial compared Lytenava given 12 times against ranibizumab given 5 times. No trial compared Lytenava with repackaged bevacizumab. The list price is $485 per vial. (see 4, 7.5, 8.1)

1. What is the effectiveness of Lytenava (bevacizumab-vikg) in adults with neovascular age-related macular degeneration?
2. What are the harms of Lytenava in adults with neovascular age-related macular degeneration?
3. Are there subgroups, defined by age, baseline visual acuity, prior anti-VEGF treatment, or demographics, in which Lytenava is more effective or associated with fewer harms?
4. Is there evidence that Lytenava differs in effectiveness or safety from off-label repackaged bevacizumab?

• There is low-quality evidence from one randomized trial, NORSE TWO, that Lytenava 1.25 mg monthly increased the proportion of participants gaining 15 or more ETDRS letters at month 11 compared with ranibizumab 0.5 mg given monthly for 3 doses and then every 3 months: 41.7% against 23.1%, risk difference 0.1859 (95% CI 0.0442, 0.3086), P = 0.0052. Mean change in BCVA was 11.2 against 5.8 letters. (see 8.1)
• The grade is held at low because the two arms received different numbers of injections, 12 against 5. The trial authors state that both arms showed a similar initial response while dosed monthly and that the ranibizumab dosing regimen is the likely reason the arms separated later. (see 8.1)
• There is low-quality evidence that Lytenava did not meet noninferiority to ranibizumab when both were given every 4 weeks: in NORSE EIGHT the mean BCVA gain at week 8 was 3.3 against 4.5 letters, LS mean difference -2.3 (95% CI -4.0, -0.5), against a prespecified margin of -3.5 letters. The sponsor reports a week 12 difference of -1.009 letters (95% CI -2.865, 0.848), a secondary timepoint. (see 8.2)
• In NORSE ONE, a 61-participant trial including previously treated participants, 2 of 25 (8%) in the Lytenava arm and 5 of 23 (22%) in the ranibizumab arm gained more than 15 letters at month 11. The trial is too small and too sparsely reported to support a conclusion. (see 8.3)
• There is insufficient evidence on quality of life, function, or patient-reported outcomes, which no Lytenava trial reports. (see 8.1, 8.2)

• There is low-quality evidence that ocular adverse reactions were similar to ranibizumab across five clinical trials: conjunctival hemorrhage 4% against 3%, eye pain 2% against 1%, and vitreous floaters 2% against 1%, in 601 participants given Lytenava and 345 given ranibizumab. (see 10)
• In NORSE TWO, treatment-related adverse events were more frequent with Lytenava, 21 (18.6%) against 9 (7.8%), with 12 injections against 5. Serious adverse events (14 [12.4%] against 16 [13.9%]) and discontinuations due to adverse events (2 [1.8%] against 5 [4.3%]) did not favor ranibizumab. The one study-related serious ocular event was elevated intraocular pressure. (see 8.1)
• Across three trials including 486 participants given Lytenava, one case of ocular inflammation (iritis) was reported, and no case of retinal vasculitis was reported in either arm of NORSE EIGHT. (see 8.1, 8.2)
• There is insufficient evidence on harms beyond 12 months and on arterial thromboembolic events, which the labeling lists as a potential risk of intravitreal VEGF inhibitors. (see 7.3, 8.1)

• There is insufficient evidence for any subgroup in which Lytenava is more effective or safer. The labeling states that no overall differences in efficacy or safety were seen with increasing age in the three completed Phase 3 studies, in which 70% of participants given Lytenava were 65 years or older. (see 8.1)
• NORSE TWO enrolled treatment-naive participants with BCVA of 25 to 67 letters, and 97.3% of the Lytenava arm was White, so the result does not establish effectiveness in previously treated eyes or in other racial groups. (see 8.1)

• There is insufficient evidence that Lytenava differs from repackaged bevacizumab in effectiveness or safety. No trial compared the two. (see 6, 8.1)
• The NORSE TWO authors compare their result with the monthly bevacizumab arm of the CATT study across trials, 41.7% against 31.3% gaining 15 or more letters, and attribute the difference either to lower potency of repackaged bevacizumab or to worse baseline visual acuity in NORSE TWO (52.1 against 60.2 letters). A cross-trial comparison of this kind does not support a conclusion. (see 8.1)

• Apply the clinical criteria at 11, which confirm the diagnosis, the prescriber, the absence of a contraindication, the labeled dose, and intraocular pressure monitoring. (see 11)
• Authorize one vial per treated eye per injection, since the labeling states that each vial is for one eye only. (see 7.3)
• Read the NORSE TWO result as evidence of effectiveness against a comparator arm that received fewer injections; NORSE EIGHT, the trial in which both arms were given every 4 weeks, did not meet its noninferiority margin. (see 8.1, 8.2)
• Review the cost of therapy when an average sales price and a permanent HCPCS code are published; the wholesale acquisition cost is $485 per vial. (see 7.5)
• Reassess when NORSE EIGHT is published in full and when NORSE SEVEN, the vial and prefilled syringe study, reports. (see 8.2, 8.5)

Abbreviations: AUC = area under the concentration-time curve; Cmax = maximum serum concentration.

See the current prescribing information for full details.

• The comparator arm received 5 injections against 12 in the Lytenava arm, and the trial authors attribute the separation of the arms after the first 3 months to the ranibizumab dosing regimen (see 8.1).
• This is the only trial in which Lytenava met its primary endpoint, and the FDA cited a lack of substantial evidence of effectiveness in two of its three complete response letters (see 7.4, 8.1).
• Follow-up was 12 months for a chronic disease that requires continued injections (see 8.1).
• Differential discontinuation, 20 participants against 9, and the pandemic-related withdrawals the authors describe add uncertainty to the primary analysis (see 8.1).
• The trial was funded by the manufacturer and two authors are its employees (see 8.1).
• No quality of life or patient-reported outcome was measured (see 8.1).

• The trial did not meet its prespecified noninferiority margin of -3.5 letters at week 8, the primary endpoint agreed with the FDA under a special protocol assessment (see 7.4, 8.2).
• Follow-up ended at week 12, which shows nothing about maintenance of vision with continued monthly dosing (see 8.2).
• The week 12 result presented as noninferior is a secondary timepoint, analyzed with the parameters of the week 8 endpoint (see 8.2).
• The week 8 figures differ between the labeling and the sponsor's announcements, and no peer-reviewed publication exists (see 8.2).
• Participant disposition, most baseline characteristics, and arm-level adverse event rates other than conjunctival hemorrhage are unreported (see 8.2).

The trial is reported only in a sponsor announcement, so a full assessment could not be performed.

• With 25 and 23 participants analyzed, the trial cannot detect a difference between arms, and the result numerically favored ranibizumab, 22% against 8% (see 8.3).
• The reported results account for 48 of 61 enrolled participants (see 8.3).
• The trial is unpublished and reported only by the sponsor (see 8.3).

• The study is single-arm and open label, so its adverse event rate of 10% cannot be compared with any control (see 8.4).
• Exposure lasted 3 months, which says nothing about harms of continued monthly dosing (see 8.4).
• The population mixed three retinal diseases, so the rate is not specific to wet AMD (see 8.4).

The study has reported no results, so only its design can be assessed.

• The study is ongoing, with an estimated primary completion in 2027-12, and has reported no results (see 8.5).
• Its purpose is to compare two presentations of Lytenava, so it will add no comparative effectiveness evidence (see 8.5).

• Visual acuity gain: the proportion of participants gaining 15 or more ETDRS letters of best corrected visual acuity from baseline, the primary endpoint of NORSE TWO and NORSE ONE. (see 8.1, 8.3)
• Visual acuity change: the mean change from baseline in best corrected visual acuity in ETDRS letters, the primary endpoint of NORSE EIGHT, judged against a noninferiority margin of -3.5 letters, and a secondary endpoint of NORSE TWO. (see 8.1, 8.2)
• Vision loss and blindness: the proportion losing fewer than 15 letters and the proportion with a Snellen equivalent of 20/200 or worse, both secondary endpoints of NORSE TWO; untreated neovascular AMD progresses to disciform scarring and blindness. (see 6, 8.1)
• Treatment burden: repeated intravitreal injections and frequent clinic visits, often with caregiver support; no Lytenava trial reports treatment burden as an endpoint, and the NORSE TWO arms differed in dosing frequency by design. (see 6, 8.1)
• Serious adverse events: at least one serious treatment-emergent adverse event in 14 (12.4%) participants given Lytenava and 16 (13.9%) given ranibizumab in NORSE TWO, with one study-related serious ocular adverse event, elevated intraocular pressure. (see 8.1)
• Withdrawal due to adverse events: treatment-emergent adverse events leading to discontinuation in 2 (1.8%) participants given Lytenava and 5 (4.3%) given ranibizumab in NORSE TWO. (see 8.1)
• Ocular inflammation: NORSE EIGHT reported no case of retinal vasculitis in either arm. (see 8.2)

• NORSE TWO: the proportion of participants who gained 15 or more BCVA letters from baseline to month 11, measured by ETDRS letter score. (see 8.1)
• NORSE EIGHT: the mean change from baseline in BCVA by ETDRS score at week 8, with a noninferiority margin of -3.5 letters. (see 8.2)
• NORSE ONE: the proportion of participants who gained 15 or more letters of BCVA, reported at month 11. (see 8.3)
• NORSE THREE: a safety study; the frequency and incidence of treatment-emergent adverse events and changes in safety parameters. (see 8.4)
• NORSE SEVEN: a safety study; the frequency and incidence of treatment-emergent adverse events after intravitreal injections from vials or prefilled syringes. (see 8.5)

Abbreviations: ARR = absolute risk reduction, or for a harm the absolute risk increase; BCVA = best corrected visual acuity; CI = confidence interval; ETDRS = Early Treatment Diabetic Retinopathy Study; ITT = intent-to-treat; mg = milligram; mmHg = millimeters of mercury; N = number of participants; nAMD = neovascular age-related macular degeneration; NNH = number needed to harm; NNT = number needed to treat; P = probability value; PP = per-protocol; SAE = serious adverse event; TEAE = treatment-emergent adverse event; VEGF = vascular endothelial growth factor

Abbreviations: AMD = age-related macular degeneration; BCVA = best corrected visual acuity; ETDRS = Early Treatment Diabetic Retinopathy Study; LS = least squares; mg = milligram; NA = not applicable; nAMD = neovascular age-related macular degeneration; NI = non-inferiority; p = probability value; PCV = polypoidal choroidal vasculopathy

Abbreviations: AMD = age-related macular degeneration; BCVA = best corrected visual acuity; mg = milligram; NA = not applicable; PIER = the ranibizumab dosing regimen of the PIER study; VEGF = vascular endothelial growth factor

Abbreviations: AMD = age-related macular degeneration; BRVO = branch retinal vein occlusion; DME = diabetic macular edema; mg = milligram; NA = not applicable; VEGF = vascular endothelial growth factor

Abbreviations: AMD = age-related macular degeneration; BRVO = branch retinal vein occlusion; DME = diabetic macular edema; mg = milligram; NA = not applicable; NR = not reported; PFS = prefilled syringe; VEGF = vascular endothelial growth factor

Abbreviations: mg = milligram; N = number of participants in the population

• To set out the clinical conditions under which Lytenava is authorized for neovascular (wet) age-related macular degeneration, in line with the labeling. (see 7.3, 7.4)
• To confine use to the labeled dose and to the population in which the trials observed an effect. (see 7.3, 8.1)

• Initial authorization covers 12 months, the duration of monthly treatment studied in NORSE TWO. (see 8.1)
• Renewal covers 12 months at a time. (see 8.1)

• No criterion sets a visual acuity range, because the two randomized trials enrolled different ranges, 25 to 67 letters and 35 to 75 letters, and the labeling sets none. (see 8.1, 8.2)
• No criterion requires the patient to be treatment-naive. NORSE TWO enrolled treatment-naive participants, but the labeling does not restrict the indication, and NORSE ONE enrolled previously treated participants. (see 8.1, 8.3)
• No criterion requires a trial of another VEGF inhibitor first. Whether to require one is a coverage decision for the plan, and no quote in this document supports a clinical requirement.