Evaluating Lecanemab Eligibility for Patients in Skilled Nursing Facilities With Alzheimer Disease or Mild Cognitive Impairment

Approximately 3.9 million¹ older adults receive care in US skilled nursing facilities (SNFs) yearly, and nearly half are diagnosed with Alzheimer disease (AD) or other dementias. Lecanemab (Leqembi)² is the first anti-amyloid monoclonal antibody to receive full US Food and Drug Administration (FDA) approval to slow AD progression. SNF residents with AD comprise a unique population of multimorbid adults for consideration of disease-modifying therapies as they include persons receiving short-term rehabilitation after a hospitalization, and patients requiring long-term nursing and medical services. To date, the potential use of lecanemab for SNF patients with multiple chronic conditions has not been formally evaluated. We applied lecanemab FDA product labeling criteria to a cohort of Medicare fee-for-Service (FFS) beneficiaries with multiple chronic conditions and mild AD or mild cognitive impairment (MCI) receiving SNF care in 2019.

Methods

This study used FFS-Medicare claims linked with the Minimum Data Set (MDS v3.0) to identify newly admitted SNF patients with a diagnosis of AD or MCI among a 20% random sample of Medicare beneficiaries (n = 386,820). The MDS is a federally mandated clinical assessment completed by a clinician on admission and quarterly thereafter to identify care needs. Other data sources included the Medicare Provider Analysis and Review (MEDPAR), Chronic Conditions Warehouse (CCW), and Medicare Carrier files. AD and MCI diagnostic codes were identified in the MDS, MEDPAR, and Carrier files using International Classification of Diseases (ICD-10) G30.0 and G31.84 in principal and first secondary positions. Cognitive impairment severity was defined using the MDS Cognitive Function Scale (CFS).³ Prescription drug dispensings were ascertained in Part D claims.

Cohort selection was guided by the FDA-approved prescribing information, which states that lecanemab is indicated solely for patients with early-stage AD or MCI and requires serial brain magnetic resonance images (MRIs); therefore, we excluded individuals lacking an AD or MCI diagnosis (n = 259,533), those with moderate to severe cognitive impairment (defined as a score of 3-4 using the Cognitive Function Scale) (n = 20,849), and others with MRI contraindications (n = 1119). Additional exclusions included age <50 years (n = 4549); noncontinuous enrollment in Medicare Parts A, B, and D in the year prior to admission (n = 90,579); and those without a valid MDS assessment (n = 26). Common contraindications to MRI were identified using diagnostic and procedural codes (eg, pacemaker, implantable cardiac defibrillator, and cochlear implant) in the 3 years before admission; no absolute contraindications are described in lecanemab product labeling. Analyses used SAS, version 9.4 (SAS Institute). The study was approved by the Brown University IRB.

Results

Among Medicare beneficiaries receiving SNF care, 32,133 patients had an MCI or AD diagnosis. After excluding 20,849 individuals with moderate or severe cognitive impairment and 1199 with typical MRI contraindications, there remained 10,085 SNF patients (ie, 31% of the MCI/AD population) eligible for lecanemab treatment. The final study sample included 7038 patients with MCI and 3047 with mild AD. Mean age was 80.3 years, most (65.8%) were female, 83.7% were White, 8.2% were Black, and 5.4% were Hispanic ethnicity (Table 1). High rates of comorbid conditions, falls (53.9%), fracture (22.2%), and use of oral anticoagulant and antiplatelet medications (22.2%) were observed.

Table 1.

Characteristics of SNF Residents Eligible for Lecanemab Compared With Baseline Characteristics of Lecanemab-Treated Participants in the CLARITY-AD Trial

Demographics	SNF Residents Eligible for Lecanemab (n = 10,085)	Lecanemab-Treated Participants in the CLARITY-AD Trial4 (n = 898)
Age, y, mean (SD)	80.3 (9.5)	71.4 (7.9)
Age group, n (%)		Exclusion: age <50 and >90
65-74 y	2090 (20.7)	368 (41.0)
75-89 y	5756 (57.1)	*
90+ y	1567 (15.5)	*
Female sex, n (%)	6640 (65.8)	462 (51.4)
Race or ethnicity, n (%)
White	8443 (83.7)	685 (76.3)
Black or African American	828 (8.2)	22 (2.4)
Asian/Pacific Islander	131 (1.3)	153 (17.0)
Hispanic or Latino	549 (5.4)	117 (13.0)
American Indian or Alaskan Native	36 (0.4)	*
Rural location, n (%)	1576 (15.6)	*
BMI, n (%)		Exclusion: BMI <17 or >35
<17	235 (2.3)	†
17-35	8565 (85.3)	*
> 35	1239 (12.3)	†
ADL self-performance‡, n (%)
Independent or limited assistance	3932 (23.8)	*
Moderate assistance	9149 (55.5)	*
Extensive assistance	897 (5.6)	*
Falls, n (%)	5427 (53.9)§	93 (10.4)
Selected comorbid conditions, n (%)
Hyperlipidemia	5258 (52.1)	*
Hypertension	7865 (78.0)	*
Diabetes	3285 (32.6)	*
Any fracture	2235 (22.2)	*
Medication use, n (%)‖
Anticoagulants**	2116 (21.0)	328 (37.6)††
Platelet inhibitors and other anticoagulants‡‡	224 (2.2)
Antidementia medications	2554 (25.3)	466 (51.9)
Antidepressants	5686 (56.4)	*
Benzodiazepines and other sedatives	2547 (25.3)	*
Antipsychotics	1524 (15.1)	*
Opioids	4184 (41.5)	*

BMI, body mass index.

^*Characteristic not available for lecanemab-treated participants in the 18-month CLARITY AD trial.

^†CLARITY AD exclusion.

^‡Activities of daily living (ADL) measure physical function using the Minimum Data Set Morris 28-point ADL score. The ADL scores were divided into 3 categories: 0-14 (“independent or limited assistance”), 15-19 (“moderate assistance”), and >22 (“extensive assistance”).

^§Falls in the prior 6 months documented by the Medicare Minimum Data Set.

^‖1 year prior to nursing home admission.

^**Includes direct oral anticoagulants (DOACs) and warfarin.

^††Other anticoagulant includes pentoxifylline.

^‡‡CLARITY-AD antithrombotics (oral anticoagulants + antiplatelet medications); aspirin monotherapy >70% of use; n = 873.

Discussion

Applying selection criteria derived from the FDA-approved product labeling to our data set, we found that nearly one-third of the 32,133 SNF patients with MCI or AD were potential candidates for lecanemab. Compared with the phase 3 trial (CLARITY AD), our cohort is, on average, nearly a decade older, differs in its inclusion of racial and ethnic subgroups at elevated risk for AD, with lower utilization of antidementia medication, and is characterized by a higher burden of illness, including falls.^4,5 These differences highlight the need for additional research to evaluate the safety and effectiveness of lecanemab therapy in SNF populations, particularly among subgroups with conditions (eg, cerebrovascular disease) that may modify treatment response or adverse event risk.⁶ Ko et al⁷ highlight the potential for increased risk of cerebral macrohemorrhage among older adults receiving lecanemab and anticoagulants, a population already more likely to develop acute and chronic cardiovascular conditions.

Our findings have the limitation of lacking clinical information unavailable in administrative data (patient-level biomarker data, confirmation of MRI compatibility). With respect to generalizability, compared with a cohort of newly admitted long-stay residents, our patient population is younger, with higher rates of fracture, yet similar in regard to the percentage of women, racial minorities, comorbidities, and antipsychotic use.⁸ Nonetheless, these findings provide a timely opportunity to consider the implications of integrating substantial numbers of SNF patients into the overburdened and fragmented care system of clinics, imaging facilities, and infusion sites that persons with AD in the community must navigate in order to be considered for lecanemab therapy. If indeed our sample closely approximates the numbers of SNF patients who may seek dementia care and disease-modifying therapies, it raises significant concerns for the preparedness of our current health care infrastructure.

With eventual approval of other amyloid-lowering treatments, it is critical that the AD research community advocate for studies of the safety and effectiveness of current and future disease-modifying therapies in diverse populations of persons with AD residing in SNFs and other extended care settings. For instance, post-acute providers may be in an advantageous position to make appropriate and timely recommendations for potentially eligible patients; these counseling providers may be able to convey the possibility of adverse events and uncertain benefit due to incomplete reporting of trial results.⁹ Of equal importance is the development of new models of care designed to improve timely access to DMT treatment and support for persons living with AD and their caregivers.¹⁰