Therapeutic Plasma Exchange for Alzheimer's Disease

The AMBAR trial demonstrated that plasma exchange with albumin replacement slowed cognitive and functional decline in mild and moderate Alzheimer's disease — with some mild-stage patients showing improvement from baseline.

Alzheimer's disease is driven in part by the accumulation of toxic proteins and inflammatory compounds in the blood and brain. Therapeutic plasma exchange addresses this directly — removing amyloid-β, inflammatory cytokines, and other neurotoxic factors from circulation, and replacing them with protective albumin. This approach has been validated in the largest randomized controlled trial ever conducted on plasma exchange for Alzheimer's disease.

What Happens in Alzheimer's Disease

Alzheimer's disease is characterized by the progressive accumulation of two abnormal proteins in the brain: amyloid-β plaques and neurofibrillary tangles made of tau protein. These deposits damage neurons, disrupt synaptic communication, and trigger chronic neuroinflammation — a sustained immune response that compounds the injury over time.

What many people don't realize is that the brain does not exist in isolation. Amyloid-β and other toxic proteins circulate in the blood as well as the brain, and the two compartments are in constant equilibrium across the blood-brain barrier. When plasma concentrations of amyloid-β are high, the brain has a harder time clearing its own toxic load. This is the insight behind the “peripheral sink hypothesis” — by reducing amyloid-β levels in the blood, you create a concentration gradient that helps pull amyloid out of the brain.

Beyond amyloid, Alzheimer's patients carry elevated levels of inflammatory cytokines (IL-6, TNF-α, IL-1β), oxidized lipids, complement proteins, and other factors in their plasma that sustain neuroinflammation. These compounds cross the blood-brain barrier and amplify the damage caused by amyloid and tau. Standard drug treatments do not address this plasma-borne inflammatory burden.

Therapeutic plasma exchange removes the full spectrum of these harmful circulating factors in a single procedure — amyloid-β, inflammatory cytokines, oxidized proteins, autoantibodies, complement — and replaces them with 5% albumin, which itself binds and neutralizes amyloid-β and has potent anti-inflammatory and antioxidant properties. No other intervention addresses the peripheral compartment this comprehensively.

The AMBAR Trial — Landmark Evidence

The AMBAR study (Alzheimer's Management By Albumin Replacement) was a Phase 2b/3, multicenter, randomized, double-blind, placebo-controlled clinical trial conducted across 41 sites in Spain and the United States. It is the largest randomized controlled trial ever performed on therapeutic plasma exchange for Alzheimer's disease.

322 patients with mild-to-moderate Alzheimer's disease were enrolled. Patients were randomized into three active treatment arms (plasma exchange with different doses of albumin, with or without intravenous immunoglobulin) and one placebo arm (sham plasma exchange). Treatment involved two phases: 6 weeks of weekly conventional TPE followed by 12 months of monthly low-volume plasma exchange.

52% less functional decline
compared to placebo
ADCS-ADL · p = .03
66% less cognitive decline
compared to placebo
ADAS-Cog · p = .06 (trend)
71% less clinical dementia decline
compared to placebo
CDR-SB · p = .002
100% less global clinical decline
compared to placebo
ADCS-CGIC · p < .0001
Mild Alzheimer's (MMSE 22–26)

Stabilization observed — no significant decline from baseline during the treatment period, with some patients showing improvement.

The AMBAR trial's author list includes Dr. Dobri Kiprov — one of the founders of modern therapeutic apheresis and Dr. Green's colleague at Global Apheresis. This direct connection to the landmark study informs the clinical protocols used at Global Apheresis today.

PublicationBoada M, López OL, Olazarán J, et al. Alzheimer's & Dementia. 2020;16(10):1412–1425.

Real-World Validation

In 2025, an independent research group in Argentina published results from a real-world cohort study evaluating TPE with albumin replacement in 32 patients with mild-to-moderate Alzheimer's disease, compared to a historical control group of 194 patients. The protocol was based on the AMBAR trial design.

Language, executive function, and attention were also better preserved in the treatment group. The intervention was well-tolerated with manageable side effects.

This study provides independent validation of the AMBAR results in a real-world clinical setting, outside the controlled environment of a sponsored trial. It demonstrates that the benefits of TPE for Alzheimer's are reproducible.

45% less cognitive declinecompared to controlsMMSE · p < .001
88% less immediate memory declinecompared to controlsp < .001
74% less delayed memory declinecompared to controlsp = .04
PublicationTaragano F, Seinhart D, Epstein P, et al. Journal of Alzheimer's Disease. 2025;108(1):129–141.

Read the complete analysis: The AMBAR Series is a four-part deep dive into the trial results, independent validation, drug comparisons, and the systemic barriers to adoption. Explore the series →

How TPE Compares to Other Alzheimer's Treatments

The Alzheimer's treatment landscape has expanded significantly. Understanding how therapeutic plasma exchange fits alongside FDA-approved therapies helps patients and families make informed decisions.

TreatmentMechanismKey EvidenceAdministrationARIA RiskStage Targeted
Therapeutic Plasma Exchange (TPE)Removes full spectrum of neurotoxic and inflammatory factors from blood (amyloid-β, cytokines, oxidized proteins); replaces with protective albuminAMBAR trial — 52–71% less cognitive and functional decline; 2025 Argentina real-world study confirming resultsIn-clinic procedure, 2–4 hours per session. Protocol: 6 weekly sessions followed by monthly maintenanceNo ARIA risk. Most common side effects are mild (tingling from citrate, temporary fatigue)Mild to moderate Alzheimer’s disease
Lecanemab (Leqembi)Anti-amyloid monoclonal antibody; targets soluble amyloid-β protofibrils and plaquesClarity AD trial — 27% less decline on CDR-SB at 18 months (p < .001). FDA fully approved July 2023IV infusion every 2 weeks. Requires amyloid PET or CSF confirmation. Regular MRI monitoring~13% ARIA-E (brain edema/effusions); ~17% ARIA-H (microhemorrhages). Higher risk in ApoE4 carriers. Rare fatal cases reportedEarly-stage AD (MCI and mild dementia) with confirmed amyloid pathology
Donanemab (Kisunla)Anti-amyloid monoclonal antibody; targets deposited amyloid-β plaques (N3pG form)TRAILBLAZER-ALZ 2 — ~35% less decline on CDR-SB at 18 months. FDA approved July 2024. Treatment can stop when amyloid clearedMonthly IV infusion. Requires amyloid PET confirmation. Regular MRI monitoring~24% ARIA-E; ~31% ARIA-H. Significantly higher than lecanemab. ApoE4 homozygotes at greatest risk. Deaths reportedEarly-stage AD with confirmed amyloid pathology
Standard Symptomatic Therapies (Donepezil, Rivastigmine, Galantamine, Memantine)Symptomatic — increases acetylcholine levels or modulates glutamate to improve neural signaling. Does not modify disease courseModest, temporary improvement in cognition and function. Decades of clinical use with well-understood safety profilesOral daily pill or transdermal patch. No imaging requirementsNo ARIA risk. Common side effects include nausea, diarrhea, insomniaMild to severe AD (symptom management only)
Therapeutic Plasma Exchange (TPE)
MechanismRemoves full spectrum of neurotoxic and inflammatory factors from blood (amyloid-β, cytokines, oxidized proteins); replaces with protective albumin
Key EvidenceAMBAR trial — 52–71% less cognitive and functional decline; 2025 Argentina real-world study confirming results
AdministrationIn-clinic procedure, 2–4 hours per session. Protocol: 6 weekly sessions followed by monthly maintenance
ARIA RiskNo ARIA risk. Most common side effects are mild (tingling from citrate, temporary fatigue)
Stage TargetedMild to moderate Alzheimer’s disease
Lecanemab (Leqembi)
MechanismAnti-amyloid monoclonal antibody; targets soluble amyloid-β protofibrils and plaques
Key EvidenceClarity AD trial — 27% less decline on CDR-SB at 18 months (p < .001). FDA fully approved July 2023
AdministrationIV infusion every 2 weeks. Requires amyloid PET or CSF confirmation. Regular MRI monitoring
ARIA Risk~13% ARIA-E (brain edema/effusions); ~17% ARIA-H (microhemorrhages). Higher risk in ApoE4 carriers. Rare fatal cases reported
Stage TargetedEarly-stage AD (MCI and mild dementia) with confirmed amyloid pathology
Donanemab (Kisunla)
MechanismAnti-amyloid monoclonal antibody; targets deposited amyloid-β plaques (N3pG form)
Key EvidenceTRAILBLAZER-ALZ 2 — ~35% less decline on CDR-SB at 18 months. FDA approved July 2024. Treatment can stop when amyloid cleared
AdministrationMonthly IV infusion. Requires amyloid PET confirmation. Regular MRI monitoring
ARIA Risk~24% ARIA-E; ~31% ARIA-H. Significantly higher than lecanemab. ApoE4 homozygotes at greatest risk. Deaths reported
Stage TargetedEarly-stage AD with confirmed amyloid pathology
Standard Symptomatic Therapies (Donepezil, Rivastigmine, Galantamine, Memantine)
MechanismSymptomatic — increases acetylcholine levels or modulates glutamate to improve neural signaling. Does not modify disease course
Key EvidenceModest, temporary improvement in cognition and function. Decades of clinical use with well-understood safety profiles
AdministrationOral daily pill or transdermal patch. No imaging requirements
ARIA RiskNo ARIA risk. Common side effects include nausea, diarrhea, insomnia
Stage TargetedMild to severe AD (symptom management only)

TPE and the anti-amyloid antibodies (lecanemab, donanemab) take fundamentally different approaches to the same problem. The antibodies target amyloid-β specifically with high precision but carry the risk of ARIA — brain swelling and microbleeds that require regular MRI monitoring and can be serious, especially in ApoE4 carriers. TPE takes a broader approach, removing not just amyloid-β but the full inflammatory environment that sustains neurodegeneration. The two approaches are not mutually exclusive — some clinicians are exploring combination protocols. The cholinesterase inhibitors and memantine remain appropriate for symptom management but do not address underlying disease processes.

What Treatment Looks Like

CANDIDACY

Who Is a Candidate

TPE for Alzheimer's disease is typically most appropriate for patients with mild-to-moderate disease who are motivated and have caregiver support. Patients should be in generally good health with adequate vascular access. Dr. Green evaluates each patient individually during the consultation to determine candidacy.

PHASE 01

Intensive Treatment (Weeks 1–6)

Based on the AMBAR trial design, the initial phase consists of weekly TPE sessions over approximately 6 weeks. Each session involves a full plasma volume exchange — your plasma is removed and replaced with 5% albumin solution. This intensive phase achieves a rapid reduction in circulating amyloid-β, inflammatory cytokines, and other neurotoxic factors. For patients traveling to our Mill Valley clinic, we can arrange compressed or flexible scheduling to accommodate logistics — the specific timing is tailored during your consultation.

PHASE 02

Maintenance Treatment (Monthly)

After the intensive phase, patients transition to monthly maintenance sessions. These lower-volume exchanges sustain the benefits achieved during the intensive phase, preventing re-accumulation of harmful plasma factors. The AMBAR trial maintained patients for 12 months in this phase and showed continued benefit.

MONITORING

Assessment & Adjustment

Dr. Green tracks treatment response through regular clinical assessments and laboratory markers. Cognitive testing is performed at baseline and at regular intervals to document objective changes. Adjustments to the protocol are made based on individual response.

Explore Your Options

If you or a loved one has been diagnosed with Alzheimer's disease, a discovery call with Dr. Green can help you understand whether TPE is appropriate for your situation. There is no obligation, and the conversation is complimentary.