Why Alzheimer’s breakthroughs still need better access infrastructure
Lecanemab and donanemab have changed the science of Alzheimer’s treatment, but coverage rules, registry requirements, diagnostic capacity, infusion access and health IT workflows will determine whether eligible patients can actually receive them.
The approval of lecanemab (Leqembi) in 2023 and donanemab (Kisunla) in 2024 marked the first time amyloid-targeting therapies had been shown to meaningfully slow the progression of Alzheimer's disease. Nearly two years later, the more consequential story for health systems is not the science but the plumbing: how coverage rules, registry requirements, diagnostic capacity, and infusion infrastructure determine who actually receives a therapy once it exists. That gap between approval and access has turned out to be a data and operations problem as much as a clinical one.
The real cost of treatment
Eisai priced lecanemab at $26,500 per year, a figure the company describes as below its own internal estimate of the drug's societal value. That number, however, understates the full cost of treatment considerably. Patients require an amyloid PET scan to confirm eligibility, typically running $3,000 to $4,500 out of pocket before Medicare coverage, plus a baseline brain MRI and additional monitoring MRIs before the 5th, 7th, and 14th infusions to screen for amyloid-related imaging abnormalities, or ARIA. The Institute for Clinical and Economic Review estimates the total system-wide annual cost per patient, drug plus PET, MRI, infusion visits, and genetic testing, at approximately $82,500. A UCLA-led analysis projects that lecanemab and its associated imaging and monitoring services could add $2 billion to $5 billion annually to Medicare spending, a scale of resource commitment that inevitably shapes how cautiously payers approach coverage design.
Coverage design as an administrative burden
In April 2022, the Centers for Medicare & Medicaid Services applied Coverage with Evidence Development, or CED, to the entire class of amyloid-targeting monoclonal antibodies, only the second time in the program's history CED has been applied to a prescription drug. Under this framework, broader Medicare coverage requires the prescribing physician to enroll patients in a CMS-approved registry with an appropriate clinical team and follow-up care. In practice, this registry requirement functions as unfunded administrative work: CMS mandates the registry but does not compensate physicians for the staff time needed to complete it, and clinical operations analyses note that without a blinded control group, the registry itself cannot generate rigorous safety or efficacy data, meaning the burden exists without fully delivering the evidentiary benefit it is designed to produce. Clinics report that once a treated-patient caseload exceeds roughly 50 patients, it becomes operationally necessary to hire a dedicated coordinator role solely to manage MRI and infusion scheduling, registry submission, and billing tied to CED compliance.
This cautious posture has a documented history. Following the 2021 accelerated approval of aducanumab (Aduhelm) amid substantial controversy over its clinical benefit, CMS raised the 2022 standard Medicare Part B premium by $21.60, the largest dollar-amount increase in the program's history, citing the need for a contingency reserve in case Medicare had to cover the drug at its initial $56,000 list price. CMS's own subsequent reexamination report confirmed that potential Aduhelm costs accounted for roughly half of that premium increase, even after the manufacturer cut its price. Analysts have linked that episode directly to CMS's subsequent design choice for the broader drug class: a more cautious, registry-based, evidence-generation-first approach for lecanemab and donanemab rather than unrestricted coverage. None of this reflects a judgment about any individual policymaker's decision; it reflects how one high-visibility coverage episode can shape years of subsequent administrative design for an entire therapeutic class.
The gap between eligibility and uptake
The result of this cumulative cost and administrative complexity shows up starkly in utilization data. Estimates of the eligible US population for amyloid-targeting antibody treatment range from roughly 1 million to 2.2 million people, a range corroborated by separate market analysis, yet only an estimated 4,000 to 13,500 patients had actually started lecanemab treatment as of early 2025, a small fraction of even the more conservative eligibility estimate. Eisai's own stated goal was to reach 10,000 US patients by March 2024; the company instead reported roughly 2,000 patients by January 2024, with an additional 8,000 reportedly waiting to begin treatment, evidence pointing toward diagnostic and infusion-capacity bottlenecks rather than a lack of patient demand.
Capacity modeling from the USC Brain Health Observatory projects that US infusion capacity for Alzheimer's treatment will grow from roughly 370,000 infusions in 2024 to 5.2 million by 2033, but even at that scale, researchers project a shortfall exceeding 13 million infusions in 2033, delaying treatment access for an estimated 2.2 million patients. Geographic access compounds the capacity gap unevenly: a study of the Atlanta, Chicago, and Houston metro areas found that roughly 12% of the population lives in an "infusion desert" more than five miles from the nearest infusion center, with those areas containing disproportionately higher shares of Black and Hispanic residents than non-desert areas. At the University of Alabama at Birmingham, average wait times for a memory evaluation appointment roughly doubled, from 245 to 490 days, following lecanemab's approval, illustrating how new-therapy demand strains an already thin specialist pipeline. These delays carry a clinical cost measured daily: industry commentary and policy analysis both cite an estimate that up to 2,000 people per day progress beyond the disease stage where current amyloid-targeting therapies remain appropriate, a consequence largely attributable to diagnostic and care-pathway delays that can stretch across months.
Building the operational pathway
Some health systems have already demonstrated that dedicated infrastructure can compress this pathway. The University of Kansas Alzheimer's Disease Research Center, in partnership with its affiliated health system, built a dedicated Anti-Amyloid Treatment Clinic in 2023 that condenses the normally months-long diagnostic workup, cognitive testing, biomarker confirmation, MRI, informed consent, using a fixed multidisciplinary team of neurologists, a dementia-trained nurse practitioner, an infusion team, a neuropsychologist, and dedicated care navigators, published as a case model in the Journal of the American Geriatrics Society. Clinical operations guidance for scaling this kind of program recommends a defined staffing template, a coordinator role once caseloads exceed 50 patients, an infusion nurse where drugs are administered onsite, and electronic tracking tools to automatically manage MRI scheduling, ARIA monitoring, and CMS registry documentation. Notably, much of this need not require new construction: infusion suites already built for oncology, rheumatology, or multiple sclerosis programs can often absorb modest additional Alzheimer's antibody infusion volume without new facilities, provided scheduling and documentation systems are adapted to the new registry requirements.
The infrastructure question health systems face
The clinical case for amyloid-targeting therapies is no longer the primary open question; the open question is whether health systems have built the data and operational infrastructure to translate FDA approval into equitable, timely access. That infrastructure spans several distinct layers — diagnostic capacity for confirmatory biomarker testing, registry-integrated documentation workflows that do not fall entirely on unfunded staff time, infusion scheduling systems that can absorb new volume without months-long waitlists, and geographic planning that accounts for where infusion deserts already exist. Closing the gap between the 1 to 2.2 million Americans estimated to be eligible and the fewer than 15,000 currently being treated will depend less on any single new drug or policy change than on health systems and health IT leaders building — deliberately and in advance — the operational pathway that a genuinely breakthrough therapy still requires to reach patients.
Kenneth R. Deans Jr., DHA, MBA is the president and CEO of Health Sciences South Carolina.