The AMBAR Trial in Plain Language

A lot of people with Alzheimer’s feel there’s nothing they can do and they’re stuck with a limited menu of medications, but a large randomized study suggests there may be more you can do alongside regular care.

  • Randomized, placebo-controlled trial
  • 347 people randomized
  • About 14 months
  • Evidence status: Studied
  • 8090 N. 85th Way, Suite 100, Scottsdale, AZ 85258
  • (623) 294-5577
  • Monday to Friday, 9 a.m. to 5 p.m.
Illustration of a brain and blood vessels. Not a test result.

The key numbers

AMBAR at a glance: how much less scores declined than with placebo

  • 52%Daily activitiesLess decline than placebo in the combined plasma exchange groups at 14 months. Statistically significant (p = 0.03).
  • 66%Memory and thinkingLess decline in the combined groups. A trend that fell just short of statistical significance (p = 0.06).
  • 61%Moderate Alzheimer’sLess decline on both measures in people with moderate disease (MMSE 18 to 21).
  • 71%Clinician-rated dementia scaleLess decline in the combined groups on a secondary measure (p = 0.002).

These are differences in decline, not shares of people who improved. In mild Alzheimer’s, no significant effect was seen on the two main measures.

Source: Boada M, et al. Primary results of the AMBAR Study. Alzheimer’s & Dementia. 2020;16(10):1412–1425.

These figures describe slower decline, not improvement. In people with mild Alzheimer’s, no significant effect was seen on the two main measures.¹

Talk to our clinical team

The trial in brief

The trial tested a plasma exchange schedule with albumin replacement. In AMBAR, people were randomized to one of four groups. Plasma exchange was given weekly for 6 weeks and then monthly for 12 months, with outcomes assessed at 14 months.¹

Earlier, smaller studies of plasma exchange in Alzheimer’s gave mixed results, and AMBAR is the larger randomized trial that followed.

Plasma exchange is not an approved treatment for Alzheimer’s disease. The Brain Health Program’s study-based protocol is described on the Brain Health Program page.

Talk to our clinical team

How the trial was designed

  • 496 screened
  • 347 randomized
  • Four groups (three plasma exchange groups, one placebo group)
  • 41 sites in Spain and the United States
  • About 14 months

The placebo group had a sham (simulated) procedure instead of plasma exchange, so the trial compared plasma exchange against having a procedure, not against no procedure at all.

In AMBAR, people were randomized to one of four groups. Plasma exchange was given weekly for 6 weeks and then monthly for 12 months, with outcomes assessed at 14 months.

Talk to our clinical team

Who was enrolled

AMBAR randomized 347 patients diagnosed with mild-to-moderate Alzheimer's disease.¹

  • Mild disease: MMSE 22 to 26.
  • Moderate disease: MMSE 18 to 21.

Its subgroup findings do not establish a benefit for other dementias, general cognitive optimization or people without symptoms.

Talk to our clinical team

What the trial found, and what it did not

How much did scores decline? Slower decline, not improvement.
ResultWhat the trial reportedBadge
Daily activitiesIn the combined plasma exchange groups, ability to manage daily activities declined about 52% less than in the placebo group. This result was statistically significant (p = 0.03).Significant
ThinkingOn the memory and thinking test, the combined groups declined about 66% less, but this fell just short of statistical significance (p = 0.06). It is best read as a trend.Trend
Moderate Alzheimer’sIn a planned look at people with moderate disease (MMSE 18 to 21), the difference was larger on both measures: about 61% less decline.Subgroup
Mild Alzheimer’sIn people with mild disease (MMSE 22 to 26), no significant effect was seen.No effect seen

The wording here says “slower decline,” never “improvement.” The “no effect seen” result carries the same weight as the others, because it is part of what the trial found.

Talk to our clinical team

Who it appeared to help

The difference was larger in people with moderate disease (MMSE 18 to 21). In people with mild disease (MMSE 22 to 26), no significant effect was seen. This is why the Brain Health Program speaks about moderate disease, and why it is not a screening service or a way to predict future risk.

Talk to our clinical team

Limits of the trial

  • Plasma exchange is not an approved treatment for Alzheimer’s disease.
  • The moderate-disease result came from a subgroup analysis, which needs confirmation in further research.
  • The memory and thinking result fell just short of statistical significance (p = 0.06) and is best read as a trend.
  • These findings do not establish a benefit for other diagnoses or people without symptoms, and no individual improvement is promised.
  • The American Society for Apheresis lists plasma exchange for Alzheimer’s disease as Category III (role unclear; decide case by case).³

Talk to our clinical team

The brain-sink working hypothesis

The “brain sink”: a working hypothesis

Lowering amyloid-beta in the blood may draw it out of the brain.

What is still unknown. Researchers have not shown that this happens in people, and other effects of plasma exchange, such as on inflammation, may matter as much or more.

Working hypothesis diagram: lowering blood amyloid may draw it out of the brain. Source: a published perspective on the peripheral sink hypothesis (Frontiers in Neurology, 2020).

The “brain sink” idea (a working hypothesis). Amyloid-beta is a protein that builds up in Alzheimer’s disease. Small, soluble forms of it move between the brain and the blood, and most of the amyloid-beta in blood is carried on albumin. Plasma exchange replaces a person’s plasma with albumin that contains no detectable amyloid-beta. The hypothesis is that the lower level in the blood could act as a “sink,” encouraging soluble amyloid-beta to move out of the brain. Researchers have not shown that this happens in people, and other effects of plasma exchange, such as on inflammation, may matter as much or more. We describe it as a working hypothesis because that is what it is.²

Talk to our clinical team

Talk to our clinical team

Wondering whether this could apply to you or your family member? Talk to our clinical team.

Our clinical team will talk through your or your family member’s situation, explain what the AMBAR trial does and does not show, and tell you honestly whether the Brain Health Program may be a fit. There is no pressure to decide.

Reading list

Plasma exchange is not an approved treatment for Alzheimer's disease. This page summarizes a published trial for general information; it is not medical advice and does not promise an individual result. Continue care with your own neurologist or physician.

Questions worth asking any clinic, including ours.

Before you decide, ask us these. We’ll answer them plainly.

  • What is the evidence for this use of TPE, and how strong is it?
  • What will my Focused panel measure, what would an Expanded panel add, and how will we know what changed?
  • What are the risks, and who should not have this procedure?
  • What will it cost in total? (The Focused panel is included; the Expanded panel is an additional fee.)
  • How will my own physician or clinician be kept informed?

Call us or send a request online, and our clinical team will answer your questions and set up a consultation.

Visit or call the clinic

Blood Rehabilitation Centers, Scottsdale, Arizona. Call us or send a request online, and our clinical team will answer your questions and set up a consultation.

Call (623) 294-5577
  • 8090 N. 85th Way, Suite 100, Scottsdale, AZ 85258
  • Monday to Friday, 9 a.m. to 5 p.m.
Talk to our clinical teamCall (623) 294-5577