Cost impact of different treatment regimens of brolucizumab in neovascular age-related macular degeneration: A budget impact analysis

Abstract

BACKGROUND:

Brolucizumab is a new anti-vascular endothelial growth factor (anti-VEGF) approved for treating neovascular age-related macular degeneration (nAMD). Multiple treatment regimens are available for treating nAMD. These regimens include manufacturer-recommended regimens, pro re nata (PRN) regimens, and treat-and-extend (T&E) regimens, which are based on clinical practice guidelines and data observed in the real-world clinical setting, classified as real-world evidence (RWE). Most budget impact models predict the financial consequences of adding a new drug to the formulary based on the manufacturer-recommended regimen. With different anti-VEGF treatment regimens being used in nAMD by ophthalmologists, it is

OBJECTIVE:

To estimate the budget impact of different treatment regimens of brolucizumab in nAMD from a US payer perspective.

METHODS:

A Microsoft Excel-based budget impact model was developed for different treatment regimens of brolucizumab over a 1-year time frame from a US payer perspective. A separate analysis was performed to estimate the budget impact from a US patient population perspective. Model inputs included drug costs, administration costs, physician visit costs, and disease monitoring costs. Outcomes in the budget impact model included the cost per member per month, annual health plan cost, and the US patient population-based annual cost. Based on the prevalence of nAMD in public and commercial health plans, a scenario analysis was conducted on the US population to account for the differences in the drug cost to the public and commercial payers. Further, 1-way sensitivity analyses were conducted to test model assumptions and uncertainty in model inputs.

RESULTS:

The addition of brolucizumab to the formulary increased the net budgetary impact under PRN and T&E regimens. The maximum increase in expenditure for a hypothetical health plan with 1 million enrollees was associated with the PRN regimen ($824,696), followed by the T&E regimen ($163,101). In contrast, using the manufacturer-recommended and RWE regimens led to an annual saving of $93,068 and $94,170 for the health plan, respectively. In the US patient population model, the introduction of brolucizumab resulted in savings in the manufacturer-recommended ($30.99 million) and RWE regimens ($31.35 million) but led to an increase in annual expenditures for the PRN ($274.58 million) and T&E ($54.30 million) regimens.

CONCLUSIONS:

Payers need to evaluate the cost impact of different treatment regimens of existing and new anti-VEGFs when making formulary decisions in nAMD management.

Plain language summary

Neovascular age-related macular degeneration (nAMD) is the leading cause of vision loss among older adults. Injectable drugs are used to treat nAMD, and their cost depends on the type of drug, the number of injections, and the number of physician office visits. The goal of nAMD-related treatment guidelines is to reduce the number of injections and physician office visits while maintaining the treatment outcomes in patients. This study demonstrates how different treatment guidelines can affect the overall cost of these injectable drugs.

Implications for managed care pharmacy

There is no clear preference for anti-vascular endothelial growth factor (anti-VEGF) treatment regimens in nAMD among physicians. This study provides valuable information to payers on cost related to various anti-VEGF treatment regimens. The study results show that the cost of adding brolucizumab to the formulary varies with the choice of its treatment regimen. In building a budget model for nAMD, payers should consider not only the anti-VEGF injection costs but also the different treatment regimens used by the providers. imperative that payers quantify the potential budget impact of different treatment regimens of brolucizumab.

Neovascular age-related macular degeneration (nAMD) is the leading cause of vision loss among older adults in developed countries, including the United States.¹ The disease is characterized by the growth of new vessels under the macula, known as choroidal neovascularization, which in the end stage of disease results in fibrovascular scars, leading to central vision loss.^2,3 Patients with nAMD experience poor quality of life and have reported higher stress and depression and lower physical activity than similarly aged individuals without nAMD.³ The societal economic burden due to nAMD-related blindness in the United States was estimated at $16 billion in 2020, with the cost projected to more than double by 2050. The societal burden included direct medical costs (eg, treatments and procedures), indirect costs (eg, productivity loss to patients and caregivers), and intangible costs (eg, years of life lost and quality-adjusted life years).⁴

Limited treatment options were available for nAMD in the 1980s and 1990s. These treatments were only able to stabilize and limit the patient’s vision loss and did not help in vision gain.⁵ The introduction of anti-vascular endothelial growth factor (anti-VEGF) in the early 2000s brought a paradigm shift in the management of nAMD, as these treatments were able to prevent vision loss and demonstrated effectiveness in vision gain. Currently, 4 anti-VEGF agents are available: ranibizumab, approved in 20 06⁶; aflibercept in 2011⁷; brolucizumab in 2019⁸; and bevacizumab, which is used off label. Pivotal randomized clinical trials (RCTs) ANCHOR and MARINA showed significant visual gain and stability for ranibizumab compared with the early photodynamic therapy and placebo, respectively.^9,10 Aflibercept demonstrated efficacy similar to ranibizumab in the RCTs VIEW 1 and 2.^11,12 Similarly, brolucizumab has shown comparable outcomes to aflibercept in HAWK and HARRIER trials.¹³ Even though bevacizumab has not been approved to treat nAMD, it has demonstrated efficacy for visual gain similar to that of ranibizumab in the IVAN and CATT trials.^14,15 Clinical guidelines^16,17 in the United States recommend the use of anti-VEGFs as a first-line treatment for nAMD, which have now replaced older treatments, such as laser photocoagulation and photodynamic therapy.¹⁸

Anti-VEGFs are administered using a fixed- or variableinterval dosing. These drugs can be administered monthly, as needed, or at increasing intervals.¹⁹ Some RCTs used fixed-interval treatment regimens, and anti-VEGFs were administered at a regular interval, usually monthly.^9,10,20-22 However, ophthalmologists reportedly use different injection frequencies when administering anti-VEGFs. In addition to fixed monthly dosing, pro re nata (PRN) and treat-and-extend (T&E) regimens are used based on their clinical experience and on the individual patient’s disease progression. The rationale for individualized treatment is rooted in reducing the overall treatment burden due to regular patient visits, pain and discomfort of intravitreal injections, and the cost of monthly injections.^23,24 Two clinical practice guidelines adapting the injection frequency, PRN and T&E, have been robustly tested in multiple RCTs^25-34 and have become popular among ophthalmologists.^35,36

The PRN regimen includes monthly office visits, and after initial loading doses, injections are administered only when the disease is active^25-28 Although studies have shown comparable visual outcomes between fixed monthly injection and PRN dosing, the monthly visit in PRN for evaluating disease status has led to a more proactive T&E regimen being adopted by ophthalmologists.^35-37 The T&E regimen uses a more objective and formulated approach to calculate the patients’ subsequent return for injection rather than a fixed monthly follow-up.^33,34 In the T&E regimen, an initial loading dose is given every 4 weeks for 3 consecutive visits. At subsequent visits, the patient follow-up interval is increased by 2 weeks, up to a maximum of 12 weeks if the nAMD is inactive (no blood or fluid noted on the exam and optical coherence tomography [OCT]).¹⁹ Conversely, if recurrent disease activity is noted, the follow-up interval is reduced by 2 weeks, with a minimum interval of 4 weeks.

Data from real-world evidence (RWE) studies rarely match the evidence from the RCTs, irrespective of the monthly or clinical guideline-based treatment regimen. Unlike RCTs, patients in RWE studies are not limited to rigid inclusion/exclusion criteria, often resulting in diverse patients with advanced disease.³⁸ Several RWE studies in nAMD have reported poor visual improvement among patients treated with anti-VEGF.^39-44 Poor visual outcomes could be attributed to undertreatment of patients (fewer visits and/or fewer injections) or patients with advanced diseases. More importantly, RWE studies have consistently shown fewer mean injections and office visits than pivotal RCTs that corelate with poor outcomes.^41,45,46 A metaanalysis of ranibizumab involving 42 global RWE studies showed a mean of 6.3 injections and 8.4 visits in the first year, with a subsequent decrease in both injections and visits for the second and third years.⁴⁷ A positive association between injection frequency and the visual outcomes was also reported, which suggested injection frequency as an important independent indicator of visual improvement.⁴⁷

Several studies of anti-VEGFs involving aflibercept and ranibizumab have shown noninferiority of both PRN and T&E regimens compared with a monthly regimen.^25-27,48-50 A 2020 Cochrane review of 15 RCTs evaluating visual outcome based on treatment regimen showed no difference in visual acuity gain between monthly and T&E regimens at 1-year follow-up and minor but clinically insignificant gain between monthly and PRN regimens.¹⁹ Considering that the outcomes of various anti-VEGF injection patterns are similar, assessing the cost associated with these drugs becomes imperative. Cost-effectiveness studies from the United States^51-54 and international perspectives^55-62 have been published for older anti-VEGFs, which have shown bevacizumab as the most cost-effective option, followed by aflibercept and ranibizumab. A 2018 review of economic studies reported that the cost-effectiveness comparison of aflibercept and ranibizumab was inconclusive and that the economic dominance between the 2 drugs was dependent on the injection frequency and the medication cost assumptions.⁶³ Another major premise often seen in nAMD economic studies is analyzing the outcome using a 1-eye model. As nAMD is a bilateral disease and can affect both eyes, a 1-eye model underestimates per patient cost. Despite this, only a few studies have used a 2-eye model,^52,61,64,65 presumably, because of the unavailability of the needed data and the requirement of sophisticated modeling expertise.^63,66

Besides the limitation of existing economic evidence due to injection frequency and the limited use of 2-eye economic models, only 1 economic study in the United States has evaluated the cost effectiveness of brolucizumab.⁶⁷ This study showed brolucizumab to be less costly and more effective than aflibercept and ranibizumab. Furthermore, the budgetary impact of adding brolucizumab to a formulary has not been reported in the published literature. Therefore, this study will estimate the budget impact of different treatment regimens of brolucizumab in nAMD from a US payer perspective.

Methods

Two Microsoft Excel-based budget impact models (BIMs) were developed: one for a health plan with hypothetical enrollees and the other based on actual US populations using 2022 US census data. Both models estimated the budget impact of brolucizumab using 4 treatment regimens over a 1-year time frame. The 4 treatment regimens included (1) the manufacturer-recommended regimen, as prescribed in the US Food and Drug Administration (FDA) label; (2) PRN; (3) T&E; and (4) the RWE regimen using data from RWE studies. The direct medical cost associated with anti-VEGFs in the 4 regimens differed primarily on the annual ophthalmologist visits and the number of anti-VEGF injections. Figure 1 shows the schematic representation of these models.

FIGURE 1.

Schematic Representation of the Budget Impact Model for the Brolucizumab in nAMD Treatment Based on Different Treatment Regimens

TARGET POPULATION

A 1-year BIM utilized a health plan with 1 million hypothetical enrollees. The treatment eligibility was based on nAMD prevalence (1.2%)⁶⁶ and the proportion of the health plan population aged at least 50 years (36.04%), which was derived from the US census data.⁶⁸ The treatment-eligible population was further divided into 1- (83.5%) and fellow-eye (16.5%) involvement based on the literature.⁶⁹ The uptake of anti-VEGF among the eligible population in the base model was assumed to be 80%. The budget impact for the health plan was reported as total annual cost and per member per month (PMPM) cost. The BIM model with the US population used a population aged at least 50 years based on 2022 census data.⁶⁸ Besides the baseline population for the base case population model, all other eligibility criteria and model inputs were the same for both the BIM models.

TREATMENT REGIMENS

Treatment regimens reported in the literature were modeled to reflect anti-VEGF use, and the net budget impact was based on the differences in the annual dose and direct medical cost related to the administration and monitoring of the different regimens of the same drug. The first regimen was based on the manufacturer recommendation and used FDA prescribing labels^6-8 for all the drugs except bevacizumab, which is prescribed off label. For bevacizumab, the number of injections reported in the CATT trial was used.^14,28 For PRN and T&E, the brolucizumab dose was taken from the first-year data of the HAWK and HARRIER RCTs, and the number of injections was determined by accounting for the proportion of patients treated every 12 weeks if the disease was stable and every 8 weeks if not stable.^31,32,70 For the RWE regimen, the brolucizumab dose for 12 months was estimated from Bilgic and colleagues’ 2021 study that reported a mean follow-up of 10.4 months.⁷¹ The doses for other anti-VEGFs under RWE regimen were taken as a weighted means from Ciulla et al⁴¹ and Rao et al³⁹ that used claims data. The doses for the remainder of the drug regimens, not described here, were taken directly from RCTs and did not require any adjustment or calculation (Table 1). Multicenter and international studies were used when US data were not available.

TABLE 1.

Population, Treatment Regimens, and the Market Share Input

	Value	Sources, details, and assumptions
Population parameters
US population 2022, n	332,941,000	US Census Bureau68
Enrollee in a health care plan, n	1,000,000	Assumption
Population aged at least 50 years, %	36.04	US Census Bureau68
Prevalence of nAMD in people aged at least 50 years, %	1.20	Rudnicka et al (2015)81
Patients with 1-eye involvement, %	83.5	Zarbin et al (2020)69
Proportion of patients on anti-VEGF therapy before launch of brolucizumab, %	80	Assumption
nAMD population on public insurance, %	73	Assumption based on Rao et al (2018)39
Treatment regimens, dosage (packs/year for 1 eye)
Manufacturer regimen
Ranibizumab	12	0.5 mg (0.05 mL) monthly6
Aflibercept	8.02	2 mg (0.05 mL)/4 weeks for the first 3 doses, followed by the same doses once every 8 weeks7
Bevacizumab	12	1.25 mg (0.05 mL) monthly14,28
Brolucizumab	7.01	6 mg/0.05 mL/4 weeks for the first 3 doses, followed by every 8-12 weeks (10 weeks taken for calculation)8
Pro re nata
Ranibizumab	6.9	CATT RCT28,29
Aflibercept	5.1	Veritti et al (AFFIRMED study)89
Bevacizumab	7.7	CATT RCT28,29
Brolucizumab	7.13	HAWK & HARRIER RCTs30-32
Treat and extend
Ranibizumab	8.00	LUCAS RCT33
Aflibercept	8.00	ATLAS RCT34
Bevacizumab	8.90	LUCAS RCT33
Brolucizumab	7.13	HAWK & HARRIER RCTs31,32,70
Real-world evidence
Ranibizumab	7.23	Ciulla et al (2020)41; Rao et al (2018)39
Aflibercept	7.04	Ciulla et al (2020)41; Rao et al (2018)39
Bevacizumab	6.91	Ciulla et al (2020)41; Rao et al (2018)39
Brolucizumab	5.55	Bilgic et al (2021) (REBA study)71
Market shares, %
Anti-VEGF market shares before brolucizumab launch
Ranibizumab	24.56	Parikh et al (2018)72; Ciulla et al (2020)41; Rao et al (2018)39
Aflibercept	17.48	Parikh et al (2018)72; Ciulla et al (2020)41; Rao et al (2018)39
Bevacizumab	57.96	Parikh et al (2018)72; Ciulla et al (2020)41; Rao et al (2018)39
Brolucizumab uptake in 1 year after launch
Brolucizumab uptake in 1 year after launch	3.84	Zarbin et al (2020)69
Percentage of Brolucizumab-switch	92.51	Zarbin et al (2020)69
Percentage of Brolucizumab-naive	7.49	Zarbin et al (2020)69
Proportion of switch from anti-VEGF to brolucizumab
Ranibizumab to brolucizumab	15.41	Zarbin et al (2020)69
Aflibercept to brolucizumab	71.15	Zarbin et al (2020)69
Bevacizumab to brolucizumab	13.44	Zarbin et al (2020)69
Anti-VEGF market shares after brolucizumab launched (switch patients)
Ranibizumab	24.01	Calculated from input data
Aflibercept	14.95	Calculated from input data
Bevacizumab	57.48	Calculated from input data
Brolucizumab	3.56	Calculated from input data

nAMD = neovascular age-related macular degeneration; RCT = randomized clinical trial; VEGF = vascular endothelial growth factor.

MARKET SHARE (TREATMENT MIX)

The market share of anti-VEGFs was determined before and after the introduction of brolucizumab. The market share before brolucizumab was determined using 664,766 eye-level data taken from 3 RWE studies that used patient registries and health plan data.^39,41,72 A 1-year uptake of brolucizumab was assumed to be 3.84% based on a 6-month uptake of 1.92% reported in a patient registry study involving more than 1.15 million patients with nAMD.⁶⁹ Further, in 1 year, 92.51% of the brolucizumab population (3.84%), switched from existing anti-VEGFs to brolucizumab, and the remaining were considered treatment naive.⁶⁹ Treatmentnaive patients were those eligible for the anti-VEGF but who were not on anti-VEGF therapy prior to the launch of brolucizumab. The proportion of patients switching from existing anti-VEGFs to brolucizumab is provided in Table 1.

COST

The BIM included direct medical costs related to drug acquisition, office visits, administration, and disease monitoring (Table 2) . The per-unit drug costs in the base case analysis was based on the wholesale acquisitions cost (WAC) listed in the IBM Micromedex RedBook.⁷³ The cost associated with drug administration, office visits, and monitoring were based on per-visit service costs using Current Procedural Terminology codes reported in the Centers for Medicare and Medicaid Services physician fee schedule national payment.⁷⁴ The intravitreal injection cost for both eyes was estimated at 1.5 times the cost of 1 eye injection using Medicare’s multiple surgery payment guidelines.^75,76 The cost of fluorescein angiography for both eyes was estimated at twice the cost of 1 eye based on a recently published study.⁷⁴ The number of annual office visits and OCT examinations were assumed to be equivalent to the number of injections for all regimens except PRN. For the PRN regimen, both the monthly clinical visits and OCT examination at each visit were considered based on PRN treatment guidelines. The models assumed fluorescein angiography only at the initial visit.

TABLE 2.

Cost Input for the Model

Services	Rate	Cost per unit	CPT and HCPCS codes
Drug costs
Wholesale acquisitions cost
Ranibizumab 0.5 mg/0.05 mL	Manufacturer: 12, PRN: 6.9, T&E: 8, RWE: 7.23	$1,950.00	IBM Micromedex Red Book73
Aflibercept 2 mg/0.05 mL	Manufacturer: 8.02, PRN: 5.1, T&E: 8, RWE: 7.04	$1,850.00	IBM Micromedex Red Book73
Bevacizumab 1.25 mg/0.05 mL	Manufacturer: 12, PRN: 7.7, T&E: 8.9, RWE: 6.91	$73.20	IBM Micromedex Red Book73
Brolucizumab 6 mg/0.05 mL	Manufacturer: 7.01, PRN: 7.13, T&E: 7.13, RWE: 5.55	$1,850.00	IBM Micromedex Red Book73
Average sales price	Same rate as used for WAC
Ranibizumab 0.5 mg/0.05 mL		$1,376.00	HCPCS J277880
Aflibercept 2 mg/0.05 mL		$1,826.00	HCPCS J017880
Bevacizumab 1.25 mg/0.05 mL		$73.00	IBM Micromedex Red Book73
Brolucizumab 6 mg/0.05 mL		$1,872.00	HCPCS J017980
Administration costs
Cost of intravitreal injection in 1 eye	Based on the treatment regimen	$114.20	CPT 6702874
Cost of intravitreal injection in both eyes (1.5x the cost of 1 eye)	—	$171.30	Medicare multiple surgery payment guidelines75,76
Physician visits
Cost of an initial office visit	1	$169.57	CPT 9920474
Cost of subsequent office visit	PRN is monthly; other regimens rate is determined by dosages	$128.39	CPT 9201474
Lab and disease monitoring costs
Optical coherence tomography at each visit	PRN is monthly; other regimens rates is same as their dosages	$41.18	CPT 9213474
Fluorescein angiography, at initial evaluation	1	$127.70	CPT 92235,74 rate assumed to be one for the first visit
Cost of fluorescein angiography involving fellow eye (2x the cost of 1 eye)		$255.40	Assumed based on a study Yu et al (2021)67

CPT = Current Procedural Terminology; HCPCS = Healthcare Common Procedure Coding System; PRN = pro re nata; RWE = real-world evidence; T&E = treat and extend; WAC = wholesale acquisitions cost.

SCENARIO ANALYSIS (BASED ON MEDICARE ALLOWABLE PAYMENTS FOR PART B DRUGS)

A scenario analysis was conducted for the US population model by incorporating the percentage of patients with nAMD enrolled in public and commercial health plans. The reimbursement for public insurance was based on Medicare allowable payment for Part B using the average sales price (ASP). The remaining population on anti-VEGFs was considered enrolled in a commercial health plan, and the WAC was used for cost as described in the base case model. Based on Rao et al, 73% of patients were estimated to be enrolled in public insurance.³⁹ If patients had dual coverage (private and Medicare/Medicaid), as mentioned in Rao et al, their reimbursement was considered to be under the public-Medicare plan. This assumption was made as the proportion of patients enrolled in public insurance reported in Rao et al closely matched the proportion of nAMD enrollees in a Medicare fee-for-service reported in the literature.⁷⁷

The unit cost under public insurance for all drugs except bevacizumab was determined using the ASP. For bevacizumab, the WAC was used, as Medicare reimbursement varies among Medicare administrative contractors because of its off-label use, and no invariant national reimbursement cost data were available.^77,78 The actual cost to Medicare was determined by using the Medicare allowable payment for Part B drugs, as described by Glasser et al⁷⁷ and Werble et al.⁷⁹ For Part B drug payment, Medicare pays 80% to an ASP and an additional 6% of ASP (ASP + 6% × ASP). To this amount, a 2% sequester Medicare payment reduction is applied under the Budget Control Act of 2011. As an example, in our study, the ASP of $1,872 for brolucizumab will be reimbursed at $1,556 ($1,872 × 106% × 80% × 98%) by Medicare. Table 1 includes the ASP cost for the drugs based on the Healthcare Common Procedure Coding System using 2022 ASP drug pricing files from the Centers for Medicare and Medicaid Services.⁸⁰

SENSITIVITY ANALYSIS

A deterministic sensitivity analysis was conducted using the upper and lower bounds of the selected model parameters to assess their net budgetary impact. The parameters included the prevalence of nAMD in patients aged at least 50 years (base case: 1.2%; upper/lower bound: 0.8%-1.7%),⁸¹ the proportion of patients on anti-VEGFs (base case: 80%; upper/lower bound: ±10%), patients with 1- or 2-eye involvement (for 1 eye, base case: 83.5%; upper/lower bound: ±10%), brolucizumab drug costs (WAC and ASP ±10%), and the brolucizumab market share (base case: 3.84%; upper/lower bound: ±10%-20%). The changes in the net budgetary impact are presented as tornado diagrams (Figure 2).

FIGURE 2.

One-Way Sensitivity Analysis of the Budget Impact Model for the Brolucizumab in nAMD Treatment Based on Different Treatment Regimens

Results

POPULATION

For a health plan with a hypothetical 1 million enrollees, 3,460 patients (1-eye = 2,888 and fellow-eye = 572) with nAMD were estimated to be on anti-VEGFs. For the 2022 US population, 1,151,923 patients (1-eye = 961,418 and fellow-eye = 190,504) with nAMD were estimated to be on anti-VEGFs.

COSTS

Ranibizumab had the highest drug and administration cost per patient in all the regimens under the WAC. The ASP (scenario analysis) was highest for aflibercept under the T&E and RWE regimens, whereas the cost for ranibizumab and brolucizumab was highest for the manufacturer and the PRN regimens, respectively. The cost of bevacizumab was lowest for all 4 regimens, primarily because of a low WAC per unit. Although the drug administration could differ based on the clinical progress, the monitoring cost in the PRN regimen remains the same for all the drugs, as this regimen involves monitoring patients regularly at a monthly interval. The cost per patient for an annual treatment with anti-VEGFs in nAMD based on the WAC and ASP using different treatment regimens is described in Supplementary Table 1 ^{(203.2KB, pdf)} (available in online article).

BASE CASE BUDGET IMPACT

In the health plan model, introducing brolucizumab as a treatment for nAMD increased the net budgetary impact for PRN and T&E and decreased the net budgetary impact for the manufacturer-recommended and RWE regimens. The maximum increase in expenditure was associated with the PRN regimen (health plan = $824,696 and PMPM = $0.069), followed by the T&E regimen (health plan = $163,101 and PMPM = $0.014). The manufacturer-recommended regimen led to an annual savings of $93,068 (PMPM =$0.008), and the RWE regimen led to an annual savings of $94,170 (PMPM = $0.008). In the US population model, the introduction of brolucizumab resulted in a savings of $30.99 million and $31.35 million in the manufacturer and RWE regimens, respectively. In contrast, it increased annual expenditures for PRN and T&E (Table 3).

TABLE 3.

Base Case Model

	Manufacturer regimen			PRN regimen			T&E regimen			RWE regimen
	US population	Health plan	PMPM	US population	Health plan	PMPM	US population	Health plan	PMPM	US population	Health plan	PMPM
Without brolucizumab
Drug cost and administration cost
One eye	$9,750 M	$29.3 M	$2.44	$5,851 M	$17.6 M	$1.46	$7,470 M	$22.4 M	$1.87	$6,571 M	$19.7 M	$1.64
Fellow eye	$3,741 M	$11.2 M	$0.94	$2,242 M	$6.7 M	$0.56	$2,868 M	$8.6 M	$0.72	$2,528 M	$7.6 M	$0.63
Monitoring cost
One eye	$2,005 M	$6.0 M	$0.50	$2,119 M	$6.4 M	$0.53	$1,552 M	$4.7 M	$0.39	$1,306 M	$3.9 M	$0.33
Fellow eye	$422 M	$1.3 M	$0.11	$444 M	$1.3 M	$0.11	$332 M	$1.0 M	$0.08	$283 M	$0.9 M	$0.07
Total cost	$15,917 M	$47.8 M	$3.98	$10,656 M	$32.0 M	$2.67	$12,221 M	$36.7 M	$3.06	$10,687 M	$32.1 M	$2.67
With brolucizumab
Drug cost and administration cost
One eye	$9,734 M	$29.2 M	$2.44	$6,043 M	$18.1 M	$1.51	$7,510 M	$22.6 M	$1.88	$6,553 M	$19.7 M	$1.64
Fellow eye	$3,735 M	$11.2 M	$0.93	$2,317 M	$7.0 M	$0.58	$2,884 M	$8.7 M	$0.72	$2,521 M	$7.6 M	$0.63
Monitoring cost
One eye	$1,997 M	$6.0 M	$0.50	$2,125 M	$6.4 M	$0.53	$1,550 M	$4.7 M	$0.39	$1,300 M	$3.9 M	$0.33
Fellow eye	$420 M	$1.3 M	$0.11	$445 M	$1.3 M	$0.11	$331 M	$1.0 M	$0.08	$282 M	$0.8 M	$0.07
Total cost	$15,886 M	$47.7 M	$3.98	$10,930 M	$32.8 M	$2.74	$12,275 M	$36.9 M	$3.07	$10,656 M	$32.0 M	$2.67
Budget impact	−$30.99 M	−$93,068	−$0.008	$274.58 M	$824,696	$0.069	$54.30 M	$163,101	$0.014	−$31.35 M	−$94,170	−$0.008

PMPM budget impact cost in the last row is presented to 3 decimals for accuracy of interpretation, and rest are presented up to 2 decimals. A negative value indicates a net budgetary decrease (cost saving), whereas a positive value indicates a net budgetary increase with the introduction of brolucizumab in the market.

M = million; PMPM = per member per month; PRN = pro re nata; RWE = real-world evidence; T&E= treat and extend.

SCENARIO ANALYSIS BUDGET IMPACT (MEDICARE PAYMENT FOR PART B)

In this analysis, 840,903 individuals received anti-VEGFs under the Medicare allowable payment for Part B drugs using ASP and 311,019 individual received anti-VEGFs under commercial health plans based on the WAC. Introducing brolucizumab as a treatment for nAMD increased the net budgetary impact for all the treatment regimens except RWE, which resulted in a savings of $3.14 million. The maximum increase in expenditure was associated with the PRN regimen ($267.93 million), followed by T&E ($76.41 million), and manufacturer regimen ($10.08 million). A major difference was noted in the manufacturer regimen, which showed an increased budgetary impact ($10.08 million) in the Medicare-adjusted model compared with a decreased budgetary impact (−$30.99) in the base population model.

SENSITIVITY ANALYSIS

In the health plan model, varying the brolucizumab WAC by 10% had the most significant impact for all treatment regimens except PRN, which showed the prevalence of nAMD as having the highest impact on the budget. The major impact of input variability was observed in the PRN and T&E regimens for the prevalence of nAMD and brolucizumab WAC, respectively. In the PRN regimen, an increased prevalence of nAMD resulted in a budgetary increase of $1,168,319 compared with a budgetary increase of $824,696 in the base case model. In the T&E regimen, a decrease in brolucizumab WAC by 10% resulted in a net budgetary decrease and annual savings of $41,177 compared with the base model budgetary increase of $163,101. Finally, the variability in the percentage of fellow-eye involvement had a negligible impact on all treatment regimens. One-way sensitivity analyses of the budgetary impact of the health plan are presented as tornado diagrams (Figure 2), and the sensitivity analysis of the population model, including the Medicare payment for the Part B adjusted model (scenario analysis), is presented in Supplementary Table 2 ^{(203.2KB, pdf)} .

Discussion

Anti-VEGFs have become the standard of care in the management of nAMD, as they not only prevent vision loss but also improve visual acuity.^82,83 From a payer perspective, there is also a need to estimate the financial consequences of adoption and diffusion of a new treatment in a health care setting, as it allows payers to allocate resources where needed. Most BIMs use the manufacturer-recommended regimen when seeking drug formulary approval. The manufacturer-recommended regimen uses a regular dosing interval for anti-VEGFs as commonly reported in the RCTs. However, regular treatments by ophthalmologists have used variable frequency treatment regimens, such as T&E and PRN. The outcomes related to anti-VEGFs in RCTs are also different than those reported in real-world studies, and one of the reasons for this discrepancy is attributed to the variable frequency regimens.⁸⁴

For payers, managing health care costs is critical, and they need to assess the impact of different treatment regimens for anti-VEGFs on their budgets. This is the first study in the United States or otherwise that has assessed the affordability of brolucizumab and other anti-VEGFs based on variable frequency regimens. Our results show that the annual cost per patient for treating nAMD is lowest when using off-label bevacizumab. Besides bevacizumab, in the health plan model using the WAC, brolucizumab has the lowest cost in all the regimens except PRN, which showed the lowest cost for aflibercept. However, when accounting for the market share and switching from existing anti-VEGFs to brolucizumab, the per-patient cost savings translates into a net budgetary savings for the manufacturer-recommended and RWE regimens. Although administering bevacizumab has the least cost and has comparable efficacy to other anti-VEGFs,^28,29 it is only approved for treating various cancers and is available in very high dosages. Therefore, administering bevacizumab for nAMD requires an adjustment of dosage by compounding pharmacists, which carries the potential risk of contamination.¹⁹

The differences in the outcome of the existing nAMD economic studies are primarily due to the differences in the cost and/or frequency of injections, as anti-VEGFs have shown similar outcomes in RCTs.^61,63 Besides the cost and frequency of injections, a difference in the economic outcomes could arise because of the 1-eye vs 2-eye model, with 2-eye models generating a more accurate estimation of economic outcomes.^63,65 Although the majority of the economic studies have used a 1-eye model, the studies performed using 2-eye models have shown mixed results for the economic outcomes.

When looking at the cost component from the 2-eye model studies, our results for bevacizumab and monthly ranibizumab are similar to other studies as the least and most costly anti-VEGFs, respectively, while considering the WAC.^52,61 When considering the WAC for drugs, our findings differ from the study by Ghosh et al that showed ranibizumab T&E as less costly than the manufacturer-recommended guideline for aflibercept. Our study found a slightly higher cost associated with the ranibizumab T&E.⁶⁴ However, when considering the ASP, our findings were similar to the study by Ghosh et al.⁶⁴ No economic studies have been conducted using the 2-eye model for brolucizumab. The results of the sensitivity analysis suggest that even after varying the data input ranges for different regimens, results were stable for most variables except for the WAC and disease prevalence.

The study findings will assist payers in adjusting their budgets by taking into consideration the different variable treatment regimens and their impact on nAMD-related spending. Although the study highlights cost savings and additional expenditures associated with the treatment regimens, the decision to include new drugs in the formulary should not be solely based on their budgetary impact. Instead, the health plan should consider treatment outcomes, providers’ opinions, and patients’ preferences when adding a drug to the formulary. In the management of nAMD, an ophthalmologist may not choose the same regimen for all their patients. Rather, they may tailor the treatment based on the individual patient’s age, fear and anxiety for intravitreal injections, the convenience of monthly followup, and, most importantly, disease progression. Adherence to treatments also needs to be emphasized, as it has been shown to correlate with maintaining or gaining visual acuity.⁸⁵

There are some recent developments in nAMD treatment, including a new drug delivery system (Susvimo), an ocular implant port delivery system with ranibizumab,⁸⁶ and a new drug (faricimab-svoa). Susvimo was approved in October 2021 and is surgically inserted in the eye and continuously delivers ranibizumab, requiring refilling every 6 months.^86,87 Faricimab-svoa received FDA approval in January 2022 and is a bispecific antibody and inhibits angiopoietin 2 in addition to VEGF.⁸⁸ We excluded Susvimo and faricimab-svoa from the analysis because of a lack of data on market penetration and the number of injections for faricimab-svoa under different regimens.

LIMITATIONS

Although our study is the first to conduct a budget impact analysis of variable frequency treatment regimens for anti-VEGFs in nAMD, there are some limitations to consider. First, we used the data from multicenter (including the United States) and international studies when data from the United States were not available. The HAWK and HARRIER RCTs were performed in a multicenter setting, and the LUCAS and AFFIRM RCTs were conducted in Europe. In addition, the RWE study for the brolucizumab dose was conducted in Germany and India, with the majority of patients (87%) being from Germany. Second, the study did not include adverse event-related cost differences among anti-VEGFs. Costs associated with side effects were excluded, as the study compared different anti-VEGF treatment regimens and comparing the side effects from the clinical setting (RWE) and research setting (RCTs) was not feasible because of different objectives of the 2 settings. Third, an assumption was made for the market share of brolucizumab, and its impact on the existing anti-VEGFs in the market was extrapolated based on 1-year data. Finally, 100% adherence to treatments and the stable prevalence of nAMD were assumed.

Conclusions

The cost impact of adding brolucizumab to the formulary in treating nAMD is dependent on the treatment regimens. Although manufacturer-recommended and RWE regimens resulted in cost savings, payers should be wary of the additional cost they could incur when accounting for clinical guidelines and real-world settings. The study provides additional economic inputs for payers to consider when evaluating the impact of a new treatment that may have a variable frequency treatment regimen, such as anti-VEGFs in the management of nAMD. The budget impact of new nAMD treatments will assist payers in developing treatment and monitoring guidelines to improve nAMD-related care.