Long‐term value of lecanemab to individuals and families

Abstract

INTRODUCTION

An assessment of the value of lecanemab for patients living with Alzheimer's disease (AD) and their care partners provides them and their health‐care providers important information for deciding treatment initiation.

METHODS

We used data from a nationally representative sample of middle aged and older Americans combined with data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) on AD progression and data on lecanemab treatment effects from Clarity AD clinical trials. We use dynamic microsimulation modeling to quantify the long‐term health and economic value of lecanemab for persons living with AD and their care partners.

RESULTS

We quantified five measures of value: quality of life of the persons living with AD and their care partners, medical costs, caregiving costs, and earnings, and estimated that lecanemab had a value of $21,398 relative to no treatment after 4 years and $37,943 after 10 years. Extending the treatment to 48 months resulted in a value of $42,821 relative to no treatment after 4 years and $95,311 after 10 years. Forty‐eight months of a similar next‐generation therapy but with 50% efficacy in slowing cognitive and functional decline resulted in a value of $82,116 relative to no treatment after 4 years and $189,691 after 10 years.

DISCUSSION

Over time, lecanemab treatment reduced medical costs, hours of care required from care partners, and improved quality of life. There is much value to be gained with next‐generation treatments that have a larger impact on slowing decline. Considering a wider range of outcomes in future assessments will provide a more complete understanding of value to support decision making about treatment initiation and about reimbursement for payers.

Highlights

• There is significant value of lecanemab for persons with mild cognitive impairment or mild dementia.

• Over time, lecanemab reduces medical costs, caregiver hours, and improves the quality of life of persons living with Alzheimer's disease (AD) and their care partners.

• A next‐generation treatment for AD with similar features to lecanemab but higher efficacy, more than doubles the value.

• Assessing therapy value supports decision making by patients and their health‐care providers.

1. BACKGROUND

In July 2023, the first disease‐modifying, anti‐amyloid therapy for the treatment of Alzheimer's disease (AD), lecanemab (brand name leqembi), received full approval from the US Food and Drug Administration (FDA). The approval of a disease‐modifying therapy for AD is significant progress in this drug discovery field that over the last century had seen limited advances. Since then, a second drug, donanemab (brand name kisunla), was approved and today the AD drug development pipeline has > 164 clinical trials assessing 127 drugs, 76% of which are disease‐modifying therapies. ¹

AD affects > 6 million persons in the United States and > 50 million globally. These numbers do not include persons with mild cognitive impairment (MCI) due to AD, which is estimated to double current prevalence numbers. The progressive disease leads to impairment of memory, language, and executive function and over time, a loss of a person's ability to care for oneself. Persons living with AD require a significant amount of caregiving and eventually many require 24 hour care. Family and friends typically provide care and may suffer negative health and financial outcomes from the strain of caregiving. AD is the sixth leading cause of death in the United States. ² The significant financial, health, and quality of life impacts of AD on the person living with the disease, their care partner, and family, underscores the significance and importance of treatment advances and the approval of lecanemab.

Results from a large 18 month double‐blind, placebo‐controlled clinical trial, Clarity AD, of lecanemab in early‐stage AD (van Dyck et al. ³ ), showed a reduction in brain amyloid plaques and a slowing of cognitive decline, and improvement in the quality of life of patients and reduction in care partner burden. ⁴ Despite FDA approval, lecanemab has generated much debate about the clinical benefits, risks, and costs of the drug therapy. Clinical benefits were considered moderate and there remain concerns of efficacy biases based in part on caregiver input. There were adverse events of brain swelling and bleeding among trial participants, and uncertainty about the potential increase in risk of these events outside of clinical trials. ⁵

Lecanemab entered the market with a list price of $26,500 per year and generated immediate concern about the economic impact of the drug. The Centers for Medicare and Medicaid Services (CMS) approved reimbursement for the treatment and covers the drug costs for eligible persons when the treating physician and team participate in a registry for the collection of evidence about the drug's efficacy and safety. Estimates of annual costs to Medicare are between $2 billion and 5.1 billion depending upon numbers of treated patients, and annual per‐patient coinsurance costs are estimated at $6636. ⁶ Persons in traditional Medicare will pay the standard 20% coinsurance once they meet their part B deductible. Out‐of‐pocket costs may be lower for persons with supplemental plans (e.g., Medigap) or enrolled in Medicare Advantage. Although CMS does not use cost effectiveness in coverage determination, the Institute for Clinical and Economic Review (ICER) concluded that the price of $26,500 exceeded commonly used cost‐effectiveness thresholds. ⁷

Assessing the value of a pharmaceutical as cost per quality‐adjusted life years is a useful starting point for considering the value of a drug to the health‐care sector particularly in the context of high and increasing prices of new pharmaceutical treatments in the United States. This approach may also inform drug formulary inclusion as well as spur more efficient drug development. However, availability and use of new drugs may offer elements of value to patients and other impacted persons such as family members that are overlooked. ⁸ For example, a new therapy may reduce the financial impact associated with a family member reducing labor supply to provide caregiving. Moreover, benefits to financial resources and quality of life may accrue over a long period of time while drug costs may be concentrated in the short term.

In this study we assess the value of lecanemab to patients and their care partners over the long term using dynamic microsimulation. This modeling approach offer strengths in quantifying the value over time to a cohort of persons receiving treatment. Individual‐level panel data capture individual heterogeneity, and the modeling of the dynamics of health over time allow for competing risks of the development of other diseases and death in the assessment of value. We quantify five measures of value: quality of life of the person living with disease and of their care partner, medical costs, caregiving costs, and earnings. We consider a baseline scenario with no treatment as our benchmark. We quantify the value of lecanemab relative to this benchmark as well as the value of a next‐generation treatment with similar attributes of lecanemab but greater efficacy in slowing cognitive and functional decline.

2. METHODS

2.1. Overview

We use data from a nationally representative sample of middle and older age Americans combined with data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). ⁹ We use dynamic microsimulation modeling to quantify the long‐term value of lecanemab for persons living with AD and their families.

2.2. Dynamic microsimulation model overview

We use the AD and related dementias (ADRD) Future Elderly Model (ADRD FEM), a dynamic microsimulation model of health and economic outcomes for the US population aged ≥ 51. The ADRD FEM uses first‐order Markov transition models to estimate models of how individuals age, for example, changes in cognition and function, onset of dementia and other health conditions, and how these changes effect employment and earnings, medical costs and need for caregiving, and risk of death. Technical documentation of the FEM is available at https://healthpolicy.usc.edu/future‐elderly‐model/fem‐technical‐specifications/.

RESEARCH IN CONTEXT

1. Systematic review: The authors reviewed the literature using traditional (e.g., PubMed) sources. The long‐term health and economic value of lecanemab for patients and care partners has not been quantified using dynamic microsimulation methods; however, a few recent publications report minimal clinical effectiveness, cost effectiveness, and value of a more limited set of outcomes. These relevant citations are referenced.

2. Interpretation: Our findings provide information on the value of lecanemab for patients and clinicians who are tasked with making decisions about whether to take up this treatment for Alzheimer's disease (AD) and payers who make reimbursement decisions.

3. Future directions: The study develops a model for use in quantifying the value of the next generation of disease‐modifying therapies for AD. It demonstrates the non‐linear increase in value associated with improved efficacy. Other health and economic impacts may be included in future assessments for a more complete understanding of value.

2.3. Data and study population

The ADRD FEM uses data from the Health and Retirement Study (HRS), a biennial nationally representative longitudinal survey in the population with > 37,000 respondents over age 50 in the United States. The study population is a cohort of respondents over age 50 from the HRS selected to match the characteristics of participants in the Clarity AD clinical trial.²⁰ The inclusion criteria we used were (1) age 50 to 90, (2) a Mini‐Mental State Examination (MMSE) score ¹⁰ between 22 and 30, (3) a body mass index (BMI) between 17 and 35, and (4) a predicted global Clinical Dementia Rating (CDR) of 0.5 or 1. Exclusion criteria were (1) recent stroke and (2) missing Telephone Interview for Cognitive Status (TICS) score. We used entropy balancing to balance the selected HRS cohort to more closely resemble the clinical trial cohort on observable measures. This multivariate reweighting balanced on age (mean and variance), CDR Sum of Boxes (CDR‐SB; mean and variance), MMSE (mean), sex, race and ethnicity, and a global CDR mix (80% global CDR of 0.5 and 20% global CDR of 1). Descriptive statistics of the analytical sample from the 2016 wave of the HRS are in Table 1. Table S1 in supporting information reports descriptive statistics from participants of the Clarity AD trial.

TABLE 1.

Sample characteristics.

	%	SD
Age in years (mean)	71.2	7.67
Male	0.48	0.50
Race/ethnicity
Non‐Hispanic White	0.77	0.42
Non‐Hispanic Black	0.10	0.30
Hispanic	0.12	0.33
Highest education
Less than high school	0.17	0.38
High school	0.49	0.50
Some college or more	0.34	0.47
Global CDR
MCI	0.80	0.40
Mild dementia	0.20	0.40
Working for pay	0.21	0.40
Widowed	0.21	0.41
ADL limitations
0 ADLs	0.80	0.40
1 ADL	0.09	0.28
2 ADLs	0.04	0.21
3+ ADLs	0.06	0.24
IADL limitations
0 IADLs	0.79	0.41
1 IADL	0.11	0.31
2+ IADLs	0.11	0.31
Chronic conditions
Stroke	0.12	0.32
Cancer	0.18	0.38
Hypertension	0.72	0.45
Diabetes	0.33	0.47
Lung disease	0.21	0.41
Heart attack	0.02	0.15
Lives in nursing home	0.01	0.07

Note: HRS wave 2016; inclusion/exclusion criteria based on trial participants.

Abbreviations: ADL, activities of daily living; CDR, Clinical Dementia Rating; HRS, Health and Retirement Study; IADL, instrumental activities of daily living; MCI, mild cognitive impairment; SD, standard deviation.

The study also used data from the ADNI. The data include participants’ CDR‐SB, and their amyloid beta (Aβ) positivity status, the latter being an inclusion criteria for the Clarity AD clinical trial. We include ADNI participants who had a minimum of 24 months of follow‐up, an initial global CDR score of 0.5 or 1, and an initial positive initial Aβ scan. This resulted in a sample of 252 ADNI participants (608 total observations) whose data we used to estimate two models of CDR‐SB. The first model was used to predict CDR‐SB in the HRS sample of respondents and the second models estimated CDR‐SB progression in the dynamic microsimulation. The models are further described below.

2.4. Measures

The primary outcomes of interest are the CDR‐SB and the outcomes used for value assessment that included quality‐adjusted life years (QALYs) of persons with MCI or mild dementia, the informal caregiving hours provided by family or friends, medical costs, and long‐term nursing home care or paid home health care costs, earnings, and caregiver disutility. Table S2 in supporting information summarizes all outcomes, with modeling approach, data used, and whether uncertainty is incorporated or if the measure is deterministic.

The CDR‐SB is the primary endpoint of the Clarity AD trial. The CDR‐SB assesses three domains of cognition (memory, orientation, judgment/problem solving) and three domains of function (community affairs, home/hobbies, personal care). The scores of the six domains (from 0 to 3) can be summed to generate a CDR‐SB score, ranging from 0 to 18. A related measure, global CDR, is derived from these six domains and categorizes disease stages: normal, MCI, mild, moderate, and severe dementia. We used a crosswalk to assign global CDR from simulated CDR‐SB. In the simulation, CDR‐SB is the only outcome directly affected by lecanemab treatment. In models, changes in CDR‐SB impact ADLs, IADLs, mortality, QALYs, caregiving hours received, nursing home entry, employment, and global CDR. Global CDR is then used to assign medical costs, earnings, and caregiver disutility and assigned values depend on disease stage.

QALY is measured using the Health Utilities Index 3 (HUI3) from the 2000 wave of the HRS. The HUI3 measures multiple aspects of health and is mapped to a single score ranging from −0.36 to 1, where 0 represents death, negative scores worse than death, and 1 is perfect health. ¹¹ Models for each outcome are described below and parameter estimates from the models are provided in Tables S3 to S5 in supporting information.

2.5. Models of outcomes

We modeled CDR‐SB using ADNI data. The model was estimated with ordinary least squares (OLS) regression, including covariates age, age squared, male, race, education, TICS, and marital status (Table S3). CDR‐SB for HRS respondents in the dynamic microsimulation were predicted based on the estimated model. We also used ADNI data to estimate cognitive transitions using changes in CDR‐SB and OLS estimation. Models included splines of 2 year lagged CDR‐SB, age, sex, and race as covariates (Table S4). Using HRS data, we estimated models of QALYs, and caregiving hours, as a function of imputed CDR‐SB and other covariates. The HUI3 model is estimated using the 2000 HRS respondents. Caregiving receipt models use HRS respondents from 2000 to 2018 and multi‐stage models: a probit of any receipt, a probit of receiving full‐time caregiving conditional on receiving any care, and an OLS model of hours of care if receiving care but not receiving full‐time care.

Other outcomes that were used to calculate the value outcomes including a nursing home residency model and a model of working for pay (Table S5). CDR‐SB affects mortality through its effect on ADLs and IADLs. Transition models of ADLs and IADLs include as covariates CDR‐SB, ADL and IADL measures from the prior survey wave (approximately 2‐years prior), age, sex, marital status, race/ethnicity, and education.

We calculated medical costs and caregiver disutility based on simulated global CDR and other data (Table S2). For persons with simulated global CDR scores of 1, 2, or 3 (dementia) or those not working for pay, we assumed no earnings. For others, we assigned an average earnings dependent on imputed global CDR based on HRS data from 1998 to 2016, a value of $39,450 for those with MCI and a value of $68,780 for those with normal cognition. No earnings are assigned for those with more severe dementia.

2.6. Dynamic microsimulation process

The microsimulation begins with a cohort of HRS respondents with MCI or mild dementia. At each 2 year timestep the characteristics of each individual are updated based on predictions from the transition models in a Monte Carlo fashion. For example, the CDR‐SB value is updated based on the CDR‐SB transition model. After the transition models are updated (including mortality), the contemporaneous models for ADLs, IADLs, work, HUI3, and care hours are assigned. This allows for these outcomes to be determined conditional on the transitioned CDR‐SB value. All transition and contemporaneous models include stochastic elements. We then calculate the value of HUI3, value of caregiving hours, medical costs, earnings, and caregiver disutility. Individuals who survive to the next year are used to calculate outcomes in the subsequent year. The simulation is performed 50 times to produce the main estimates to produce stable results.

The uncertainty estimates reflect a non‐parametric bootstrap approach to the data and both the transition and contemporaneous models. For this, the ADNI and HRS data are each resampled 100 times, all models are re‐estimated for these samples, and the simulation is performed with each of these 100 sets of models, 25 simulations for each set of models. To construct confidence intervals, the simulation is repeated 2500 times for all individuals. The resulting variation across outcomes is used to construct a confidence interval (1.96 times the standard deviation of the outcomes) around the main estimates.

2.7. Treatment scenarios

We report results for three treatment scenarios. The first two scenarios assumed treatment effect of lecanemab as a 27% reduction in the change of CDR‐SB, based on Clarity AD clinical trial outcomes. We assume that this 27% reduction applies to all individuals and remains constant during the treatment period. The first scenario assumes 18 months of treatment, the second, 48 months of treatment. The third scenario assumes a hypothetical treatment effect that is a 50% reduction in the decline of CDR‐SB with 48 treatment months. In all treatment scenarios, we assume natural decline resumes after the end of the treatment period. We report the dollar value of QALYs, informal help hours, medical costs, caregiver disutility, and earnings, for 4 year and 10 year cumulative effects. We also report the value relative to no treatment. QALYs are valued at $150,000 per QALY, help hours at $28.16 per hour, and all costs are in 2022 USD and discounted at 3% annually. For all scenarios, we assumed treatment continuation when patients progress into moderate/severe dementia, which is consistent with the clinical trial procedure. We interpolated outcomes at each 6 month period in the 2 year simulation cycle, and mortality was assigned at the middle of a 2 year cycle.

When valuing assessments, we included the impact of amyloid‐related imaging abnormality (ARIA) adverse events from treatment. An ARIA event impacts treatment cessation, has medical costs associated with the event, and impacts QALYs. We used the value of percentage of patients experiencing ARIAs and percentage with treatment cessation due to ARIAs as reported in the Clarity AD trial and quantified ARIAs’ impact on costs and QALY based on data reported in prior studies (Table S6 in supporting information).

3. RESULTS

Table 1 reports the mean (standard deviation) of the characteristics of the analytic sample selected as described above to match the characteristics of the participants in the Clarity AD trial. Average age is 71.2 years and 48% of the sample is male. Seventy‐seven percent are non‐Hispanic White, 10% are non‐Hispanic Black, and 12% are Hispanic. Thirty‐four percent report their highest education is some college or more, 46% high school, and 17% less than high school. Twenty percent report working for pay and 21% are widowed. Eighty percent have MCI and 20% mild dementia. Twenty percent have one or more ADL limitation(s) and 22% have one or more IADL limitation(s). Table 1 also reports the prevalence of chronic conditions; the most common is hypertension (72%) followed by diabetes (34%). Less than 1% report living in a nursing home.

Table 2 reports the average CDR‐SB of the cohort over time from baseline to 10 years with 95% confidence intervals. The first column reports the changes in the CDR‐SB over time for the untreated persons and columns 2 to 4 for the scenarios. Across the three scenarios, there is a slowing of cognitive decline relative to the untreated population. At 48 months, CDR‐SB is 9.07 for the untreated and for the treated 8.52; 7.58 for scenarios 1, 2, representing lecanemab treatment for 18 months and 48 months, respectively; and 6.30 for scenario 3, which represents a drug treatment with a higher efficacy than lecanemab and 48 months of treatment.

TABLE 2.

CDR‐SB over time for cohort with MCI or mild dementia across no treatment and three treatment scenarios.

		Scenario 1	Scenario 2	Scenario 3
Months	No treatment	18 month 27%	48 month 27%	48 month 50%
0	3.20 [3.20, 3.20]	3.20 [3.20, 3.20]	3.20 [3.20, 3.20]	3.20 [3.20, 3.20]
24	5.89 [5.57, 6.21]	5.39 [5.13, 5.65]	5.23 [4.98, 5.48]	4.66 [4.47, 4.85]
48	9.07 [8.41, 9.73]	8.52 [7.92, 9.12]	7.58 [7.08, 8.08]	6.30 [5.92, 6.68]
72	12.14 [11.04, 13.24]	11.68 [10.62, 12.74]	10.76 [9.80, 11.72]	9.42 [8.66, 10.18]
96	14.43 [13.19, 15.67]	14.15 [12.89, 15.41]	13.53 [12.24, 14.82]	12.48 [11.27, 13.69]
120	15.84 [14.75, 16.93]	15.69 [14.56, 16.82]	15.36 [14.15, 16.57]	14.72 [13.44, 16.00]

Abbreviations: CDR‐SB, Clinical Dementia Rating Sum of Boxes; MCI, mild cognitive impairment.

Table 3 reports the value of outcomes in dollars with 95% confidence intervals: patient QALYs, informal help hours, medical and long‐term care costs, earnings, and caregiver disutility for the untreated cohort (column 1) and the three scenarios (columns 2–4). The last three columns show the difference in value of the scenario relative to no treatment. Panel A reports outcomes at 4 years and Panel B at 10 years. Table S7 in supporting information shows the values of each outcome in units before conversion to dollar values.

TABLE 3.

Value of treatment (2022 dollars): 4 year and 10 year dollar value [95% CI].

		Scenario 1	Scenario 2	Scenario 3	S1‐None	S2‐None	S3‐None
Outcome	No treat	18 month 27%	48 month 27%	48 month 50%	delta A	delta B	delta C
Panel A: 4 year
QALYs	471962 [466826, 477098]	477421 [472337, 482505]	482853 [477669, 488037]	497187 [492497, 501877]	5459 [5246, 5672]	10891 [10431, 11351]	25225 [24387, 26063]
Medical costs	99216 [92184, 106248]	93603 [87919, 99287]	88178 [83932, 92424]	79615 [76791, 82439]	−5613 [−7157 −4069]	−11038 [−14108, −7968]	−19601 [−24288, −14914]
Earnings	20780 [19559, 22001]	21490 [20251, 22729]	22042 [20766, 23318]	24008 [22704, 25312]	710 [611, 809]	1262 [1109, 1415]	3228 [2953, 3503]
Caregiving	87027 [77815, 96239]	78970 [71151, 86789]	70611 [64252, 76970]	58432 [53982, 62882]	−8057 [−9778, −6336]	−16416 [−19881, −12951]	−28595 [−34137, −23053]
Caregiving disutility	−12476 [−13998, −10954]	−10917 [−12199, −9635]	−9262 [−10280, −8244]	−7009 [−7643, −6375]	1559 [1284, 1834]	3214 [2655, 3773]	5467 [4521, 6413]
Total					21398 [18547, 24249]	42821 [37083, 48559]	82116 [72567, 91665]
Panel B: 10 year
QALYs	690806 [680607, 701005]	698840 [688620, 709060]	708520 [697988, 719052]	730779 [720271, 741287]	8034 [7662, 8406]	17714 [16903, 18525]	39973 [38760, 41186]
Medical costs	305983 [265177, 346789]	293531 [255229, 331833]	273116 [238690, 307542]	243422 [215221, 271623]	−12452 [−15363, −9541]	−32867 [−39671, −26063]	−62561 [−75835, −49287]
Earnings	22372 [20439, 24305]	23160 [21187, 25133]	23863 [21810, 25916]	26201 [24031, 28371]	788 [683, 893]	1491 [1297, 1685]	3829 [3449, 4209]
Caregiving	302966 [255462, 350470]	289037 [244341, 333733]	266854 [226849, 306859]	233128 [199829, 266427]	−13929 [−17330, −10528]	−36112 [−44771, −27453]	−69838 [−85764, −53912]
Caregiving disutility	−49998 [−56895, −43101]	−47258 [−53844, −40672]	−42871 [−48956, −36786]	−36508 [−41664, −31352]	2740 [2344, 3136]	7127 [6142, 8112]	13490 [11509, 15471]
Total					37943 [33082, 42804]	95311 [83455, 107167]	189691 [165747, 213635]

Note: $150,000 value of life, help hours valued at $28.16 hourly, future amounts discounted at 3% per year. S#—None represents the difference in the scenario compared to no treatment (no treat). 95% confidence intervals provided.

Abbreviations: CI, confidence interval; QALYs, quality‐adjusted life years.

Among persons with MCI or mild dementia, after 4 years, treated persons across all three scenarios have higher QALYs than untreated persons with differences between each scenario and untreated persons of $5,459, $10,891, and $25,225, respectively. Medical costs are $5,613, $11,038, and $19,601 lower for treated persons in scenarios 1 to 3 compared to untreated persons, respectively. Treatment has a small effect on earning with differences of $710, $1262, and $3228, respectively, for each scenario relative to no treatment. Treated persons received fewer help hours in each scenario compared to untreated persons and resulted in differences in dollar values of $8057, $16,416, $28,595. Caregiver disutility is reduced by $1559, $3214, and $5467 for each scenario 1 to 3 relative to the untreated group. The last row of Table 3 Panel A reports the net value of the treatment relative to no treatment across all five outcomes. Four year value is $82,116 higher for persons in scenario 3 compared to untreated persons and the highest value across all scenarios. The 4 year relative (relative to no treatment) value associated with scenario 1, the scenario that represents the lecanemab Clarity AD trial results, is $21,398.

Panel B reports the value of treatment over 10 years. The average treatment benefit for a cohort of persons with MCI or mild dementia relative to no treatment increases over time. The total value associated with treatment relative to no treatment under scenario 1 is $37,943, 77% higher than the difference treatment to no treatment value after 4 years. Under scenario 2, relative value after 10 years is $95,311 higher relative to no treatment and is double the 4 year treatment difference. Under treatment scenario 3, the value is $189,691 higher than no treatment and this is a 131% increase in value compared to 4 year value.

4. DISCUSSION

We assessed the value of lecanemab to persons with MCI or mild dementia and their care partners over the long term using dynamic microsimulation. We quantified five measures of value: quality of life of the person living with disease and of their care partner, medical costs, caregiving costs, and earnings. We found that the treatment effect of lecanemab had a value of $21,398 relative to no treatment after 4 years and $37,954 after 10 years. Extending the treatment to 48 months resulted in a value of $42,821 relative to no treatment after 4 years and $95,311 after 10 years. We compared the value of these two scenarios of lecanemab to that of a next‐generation treatment with similar attributes of lecanemab but greater (50%) efficacy in slowing decline compared to 27% as reported in Clarity AD trials. Forty‐eight months of treatment of this next‐generation hypothetical therapy resulted in a value of $82,116 relative to no treatment after 4 years and $189,691 after 10 years.

As shown in the pivotal Clarity AD randomized control trial of lecanemab, participants had less cognitive decline as measured by the CDR‐SB compared to placebo at 18 months. The mean difference was −0.45 and there was less decline in quality of life for participants and caregivers in the lecanemab‐treated group compared to placebo. The goal of the study is to evaluate how those reported outcomes from the Clarity AD trial translated into long‐term economic and health impacts of lecanemab for cohort of persons with MCI or mild dementia. Strengths of this study included assessment in the long term of various measures of value, and the use of panel data on individuals that allowed for competing risk of disease for an individual and heterogeneous outcomes across individuals as well as longitudinal data from ADNI that provides a profile of the natural history of AD.

Other types of evaluation studies for lecanemab have been conducted including minimal clinical effectiveness and cost effectiveness. Despite these treatment effects, the mean difference in CDR‐SB was less than what several studies considered to be a minimal clinically important difference. ¹² Studies evaluated cost effectiveness of lecanemab. Cost effectiveness studies primarily focus on the ratio of medical costs to health outcomes to assess efficiency from a health‐care sector perspective and found lecanemab was not cost effective. ¹³ A more recent study examined cost effectiveness from a societal perspective over a hypothetical cohort's lifetime, including lifetime QALYs as well as costs and found lecanemab's annual price exceeded commonly used cost‐effectiveness thresholds. ⁵ While this study was similar to this valuation study in terms of assessing value to treated person and care partners, their results were limited to averages of a hypothetical cohort based on cross‐sectional studies. A study by Monfared et al. used ADNI data and modeled individuals’ health outcomes (QALYs) but did not consider adverse events, nor include other economic outcomes or capture competing risk of other diseases in valuation. ¹⁴

This study has limitations. The analysis assumed adherence to treatment and efficacy rates were assumed for all treated persons although the analysis did impose discontinuation for persons with ARIA events. The efficacy of the therapy was assumed to hold constant during the treatment period. This valuation study did not include out‐of‐pocket expenses associated with the cost of the drug or treatment‐related medical care with the exception of medical care–related expenses associated with ARIAs. However, this is not a cost‐effectiveness study so inclusion was not a stated goal. There may be other aspects of value that were not assessed such as the impact of treatment on the workplace productivity of persons with MCI or mild dementia or their care partners, or the value of hope. ⁸

5. CONCLUSION

Understanding the value of lecanemab is important for myriad stakeholders such as patients and clinicians who are tasked with making decisions about whether to take up this treatment for AD and payers who make reimbursement decisions. While lecanemab offers value today, the study also revealed the opportunity for next‐generation AD treatments.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest. Author disclosures are available in the supporting information.

Supporting information

Supporting Information

Supporting Information

Tysinger B, Wei Y, Heun‐Johnson H, Zissimopoulos J. Long‐term value of lecanemab to individuals and families. Alzheimer's Dement. 2025;11:e70151. 10.1002/trc2.70151