Therapeutic Plasma Exchange dosage and evidence
Written by Aaron CuhaReviewed Sep 2026
The short answer
Therapeutic Plasma Exchange is a standard hospital procedure that takes out plasma and puts a blood protein called albumin back. A 347-person trial in Alzheimer disease hit one of its two main goals. Two small 2025 trials of ageing markers came to opposite conclusions.
Evidence grade for Therapeutic Plasma Exchange: Moderate. The rule: at least one finished study in people, and at least 100 people who took it. The sum on this page: 5 completed trials, 389 people given it. A grade counts how much has been done to people, not whether it works.
- People given it
- 389
- Human studies
- 5
- Found nothing
- 1
Dose
Approved medicine. No human dose established for unlabelled uses
Labeled amounts sit on the prescribing information. This page records what trials measured, not a personal dose.
Therapeutic plasma exchange is an established medical procedure. A machine cleared for the job removes plasma, the liquid part of blood, and replaces it. It has accepted uses in some nerve, blood and kidney diseases. It is not approved anywhere for ageing, for lowering biological age or for Alzheimer disease.
6 effects reported, 4 unwanted, 2 wanted, by people using it, not a trial
Goal pages: Brain, memory and mood
In plain English
In plain English, from Aaron
This is how I would explain it to a friend. The evidence, with the numbers, is further down the page.
This is a standard hospital procedure. A machine takes out some of your plasma and puts back a blood protein called albumin. Hospitals have done it for decades for brain, blood and kidney disease. Using it against ageing is the new part.
What people take it for
- Lowering their biological age on a clock test.
- Slowing Alzheimer disease.
- Clearing out whatever builds up in blood with age.
What studies in people showed
The biggest trial took 347 people with mild to moderate Alzheimer disease. They got plasma exchange or a fake procedure, weekly for six weeks then monthly for a year. At 14 months, the treated group lost 52 percent less of their everyday function. That was one of two main goals, and it held. Thinking skills declined 66 percent less, but that was not strong enough to count. People who started with moderate disease got better on both. People with mild disease did not change. The albumin maker paid for it. Two small 2025 trials tested ageing clocks and disagreed. One, in 42 adults over 50 with albumin put back, found 15 clocks reading younger. One, in blood donors with nothing put back, found three clocks reading older.
What people report
Reports, not study results
The good
- It is available today, done routinely, with a long known safety record.
- In the Alzheimer trial, people with moderate disease improved on both everyday function and thinking.
The bad
- Two people in the 42 person trial stopped because of side effects. One of those was blamed on the added immune protein drip.
- Your blood work changes. In the donor trial, cholesterol, triglycerides, total protein and albumin all fell.
- It removes clotting factors and antibodies along with everything else.
- Clinics sell a clock reading as the result, and that is exactly what the two 2025 trials disagree about.
Where these come from: The side effects and blood test changes come from the published trials on this page. How it is sold reflects the clock measures those trials name. No clinic or patient account is quoted.
My bottom line
The strongest human file in this whole group, and still no proof it does anything for a healthy older person. Get your blood work first.
An opinion, not a finding. I am a coach, not a doctor.
What it has been measured to do
Measured in people
Goals Therapeutic Plasma Exchange has been tested for in people, biggest studies first. We count only what the studies on this page say, so the real numbers may be higher.
- Memory and thinking347 people · 2 studies · 1 found nothing
What people using it report
Real people telling you what happened to them. Nobody checked it, and the people who post are not a fair sample, so it cannot prove anything works. It is still the best guide on this page to what people actually do with Therapeutic Plasma Exchange, what they expect, and what goes wrong. The full account, including the negative reports, is below.
34 of the 36 indexed goals have no study of any kind behind Therapeutic Plasma Exchange
No study in people, animals or cells on this page reports a result for these. That does not prove Therapeutic Plasma Exchange does nothing for them. It does mean anything you have heard about Therapeutic Plasma Exchange and one of these did not come from a study cited here.
Show full evidence
What is Therapeutic Plasma Exchange?
Therapeutic Plasma Exchange is a standard hospital procedure that takes out plasma and puts a blood protein called albumin back. A 347-person trial in Alzheimer disease hit one of its two main goals. Two small 2025 trials of ageing markers came to opposite conclusions.
Therapeutic plasma exchange is not new as a procedure. It is a standard treatment in brain, blood and kidney medicine, done with approved machines. Some of the patient's plasma is taken out and replaced with albumin, a blood protein, sometimes with an immune protein drip added. What is still being tested is using it against ageing rather than against a named disease.
The biggest controlled trial is AMBAR, run by Instituto Grifols in mild to moderate Alzheimer disease. 347 people were split by chance into three plasma exchange groups, each with different amounts of albumin and immune protein, or a fake procedure. Treatment ran weekly for six weeks, then monthly for a year. At 14 months, treated people lost less of their everyday function than the fake group: 52 percent less decline. That was the first of two main goals, and it held up. On thinking skills, the second main goal, decline was 66 percent less, but that comparison was not strong enough to count. People who started with moderate disease got better on both scales. People with mild disease did not change. Grifols, which makes the albumin, paid for the trial.
Two small trials published in 2025 tested plasma exchange against ageing markers head on, and they disagree. In Aging Cell, 42 healthy adults over 50 were split by chance between exchange every two weeks, with or without an immune protein drip, exchange once a month, or a dummy procedure. The authors report that long-term exchange was safe, with two people stopping because of side effects. They also report that 15 different epigenetic ageing clocks read younger than in the dummy group, and that the two-week schedule with the immune protein did best. In Scientific Reports, healthy blood donors had either 8 or 4 plasma sessions over 18 weeks, each person trying both schedules in turn, with no young plasma and no albumin put back. The clocks did not read younger. Three of them, DNAmGrimAge, the Hannum clock and the Dunedin Pace of Aging, all went the wrong way. The two trials used different methods, and the second one points out that it put no fluid back.
How does Therapeutic Plasma Exchange work?
Two rival explanations, neither settled in people. The thinning idea, worked out in mice, says that taking out some old plasma waters down proteins that build up with age and resets the body's signals, and that what you put back barely matters. Mouse work in 2020 found that salt water plus albumin matched the effect of young blood. The replacement idea says the albumin itself is what counts. It may bind and carry away unwanted molecules, or restore the blood's carrying capacity. That is the thinking behind the AMBAR method, where albumin is the replacement fluid. The 2025 Scientific Reports trial took plasma out without putting albumin back, and the ageing clocks moved the wrong way. That is evidence against thinning being enough on its own, though one small trial does not settle it.
What have studies in people found about Therapeutic Plasma Exchange?
A lot, by this site's standards. One 347-person controlled trial in Alzheimer disease with published results, and two small 2025 trials of ageing markers that came to opposite conclusions. That is more evidence in people than almost anything else here claimed to affect ageing.
- AMBAR (NCT01561053, Alzheimer's and Dementia 2020): 347 people with mild to moderate Alzheimer disease, split by chance into three plasma exchange schedules or a fake procedure. Patients, their care teams and the people scoring them were all kept in the dark. Weekly for six weeks, then monthly for a year. One of the two main goals was met: everyday function declined 52 percent less than in the fake group. The thinking-skills goal showed 66 percent less decline, but was not strong enough to count. Paid for by Instituto Grifols, which makes the albumin.
- AMBAR group detail, which cuts both ways. People who started with moderate disease did better on both everyday function and thinking, with 61 percent less decline on each. People with mild disease did not change. Results in small groups, from a trial that missed one of its two main goals, are exploring, not proving.
- AMBAR also reported better scores on two other dementia scales, one measuring how much help a person needs across several areas and one a doctor's overall impression.
- Aging Cell 2025 (NCT06534450): 42 healthy adults over 50, split by chance, with a dummy procedure and only the patients kept in the dark. Exchange every two weeks with or without an immune protein drip, exchange monthly, or dummy. Reported safe over the long run, with two people stopping for side effects. 15 epigenetic ageing clocks read younger than the dummy group, and the two-week plus immune protein group did best.
- Scientific Reports 2025: healthy blood donors had 8 or 4 plasma sessions over 18 weeks, each person trying both schedules in turn, with at least 14 days between sessions and no young plasma or albumin put back. The clocks did not read younger. DNAmGrimAge, the Hannum clock and the Dunedin Pace of Aging all went the wrong way. Total cholesterol, non-HDL cholesterol, triglycerides, apolipoprotein A, total protein and albumin all fell.
What this does not tell you: AMBAR tested people with a disease, not healthy ageing. It missed one of its two main goals. And it was paid for by the company that makes the albumin. The two 2025 marker trials are small: 42 people in one, a group of blood donors in the other. They used different methods and different replacement fluids, which is the most likely reason they disagree. Epigenetic clocks are markers that tend to track with health, not health itself. No study in people has linked a clock change from plasma exchange to a health outcome, an illness or living longer. Plasma exchange also removes clotting factors, antibodies and blood fats along with whatever is being targeted. That is why the Scientific Reports trial saw albumin and total protein fall.
What do people who took Therapeutic Plasma Exchange report?
This is what real people say happened to them, good and bad. Nobody measured it. A report cannot tell a real effect from the lift people get just from expecting it to work, from a diet or training change made at the same time, or from what was really in the vial. We still put it near the top. It is the next best thing to a study in people: what happens when real people use it, including what goes wrong.
- Longevity clinics sell plasma exchange under names like plasmapheresis, albumin exchange and blood cleansing. They usually sell it as a course of several sessions, and usually quote an epigenetic ageing clock reading as the result. That clock reading is what is being sold, and it is exactly the thing the two 2025 trials disagree about.
- The procedure itself is routine, and what goes wrong is recorded in the trials rather than in stories. In the Aging Cell trial of 42 adults over 50, two people stopped because of side effects. One of those was put down to the added immune protein drip, not to the exchange.
- The procedure changes your blood work, and that is worth knowing before booking one. In the published blood donor trial, total cholesterol, non-HDL cholesterol, triglycerides, apolipoprotein A, total protein and albumin all fell. Two red blood cell measures rose.
- Where the immune protein drip is added, that is a separate product with its own risks. It is the part blamed for the one person who stopped in the Aging Cell trial.
Sources: The published trial reports themselves, which is where the side-effect and lab-change numbers above come from: the AMBAR primary results in Alzheimer's and Dementia 2020, the Aging Cell 2025 study and the Scientific Reports 2025 study. The description of how the procedure is marketed reflects the measures those clinics advertise, which are the epigenetic clock measures named in the trials. No clinic, forum post or individual patient account is quoted or relied on.
The state of the research
This page carries a live disagreement, and this site does not settle it. The Aging Cell trial says 15 ageing clocks read younger than the dummy procedure. The Scientific Reports trial says the clocks moved the wrong way. The differences that could explain it are concrete. The Aging Cell method put albumin back in place of the plasma taken out, and in its strongest group added an immune protein drip. The Scientific Reports method took plasma from healthy blood donors and put nothing back, and the authors say so plainly. If simply thinning old plasma were enough on its own, the second trial should have shown a benefit. It did not. That is a real argument against the thinning idea, made by a small trial, which is how this field usually works.
The funding runs the way you would expect in one case, and it should be named. AMBAR was paid for by Instituto Grifols, a plasma products company whose albumin is the replacement fluid being tested. That does not make the result wrong, and company funding is normal for a trial that size. But it belongs on the page next to the fact that one of the two main measures was missed. The 2025 Aging Cell trial's registry record lists an individual researcher as sponsor rather than a company, and its status is listed as unknown. That is a different kind of weakness.
One last point about what this actually is. Unlike almost everything else in this batch, plasma exchange is available today. It is done routinely, and decades of use in brain and blood disease have shown how safe it is. The question is not whether it can be done. It is whether doing it in a healthy older person changes anything that matters. On that, there are two small trials pointing opposite ways, and no data at all on health outcomes.
The evidence, in charts
Fig. 1a · how many people
776people, across 4 studies in people cited here
| Study | People | Share |
|---|---|---|
| Alzheimer's and Dementia 2020 | 347 | 45% |
| AMBAR, ClinicalTrials.gov 2012 | 347 | 45% |
| Aging Cell 2025 | 42 | 5% |
| ClinicalTrials.gov 2022 | 40 | 5% |
Counts are the number of people each study on this page says it signed up. That includes studies that only watched people and gave nobody the compound. 3 more studies in people give no headcount and are not counted here. Studies still signing people up are left out, and only studies cited on this page are counted.
Fig. 1b · evidence mix
All 7 sources here are studies in people.
- Study in people · done in people7100%
Counted from the list of sources on this page. It counts only what we cite, so there may be more studies out there. A review counts once, however many studies it sums up.
Fig. 2 · evidence over time
The studies cover 5 different years, 2012 to 2025, counted from the list of sources on this page.
Fig. 3 · legal status at a glance
US
Not approved
UK
Not approved
AU
Not approved
CA
Not approved
Not approved in any of the four countries shown. This is what the law says, not a verdict on the science. See Legal status below for the full text and any notes.
Fig. 5 · what the molecule is
- Kind of compound
- Procedure
- Molecule weight
- Not on file
- Half-life
- None on file
- Chain length
- None on file
No amino acid sequence is on file for Therapeutic Plasma Exchange, which is expected: a procedure is not built from amino acids.
The studies we cite
- Study in people2020A randomized, controlled clinical trial of plasma exchange with albumin replacement for Alzheimer's disease: Primary results of the AMBAR Study
Plasma exchange slowed decline in Alzheimer disease, and the company that makes the albumin paid for the trial. 347 people were split by chance between three plasma exchange schedules and a sham procedure. Treatment ran weekly for six weeks, then monthly for a year. At 14 months, treated patients lost about half as much everyday function. Thinking scores declined about two thirds less, but that comparison was not strong enough to call. People with moderate disease improved on both measures. People with mild disease did not change.
347 patients randomised 1:1:1:1 from 496 screened into three plasma exchange arms with different albumin and immunoglobulin replacement doses or sham, with six weeks of weekly conventional exchange then twelve months of monthly low-volume exchange. At month 14, treated patients had 52 percent less decline on activities of daily living (p = 0.03) and 66 percent less decline on the cognitive scale (p = 0.06, not significant). Moderate-disease patients improved on both; mild-disease patients did not change. Funded by Instituto Grifols, which manufactures the albumin used.
- Study in people2019Plasma exchange for Alzheimer's disease Management by Albumin Replacement (AMBAR) trial: Study design and progress
This paper is the plan, not the result. It describes how AMBAR was going to be run: four groups, a sham procedure for the comparison group, and which measures would count. It was published before the outcomes were reported, which is the point of it. Saying what you will measure before you look is what stops a team picking the flattering number afterwards. There is no result in this paper to report.
The published design paper for AMBAR, describing the four-arm structure, the sham procedure and the endpoint plan before results were reported.
- Study in people2022Neuropsychological, neuropsychiatric, and quality-of-life assessments in Alzheimer's disease patients treated with plasma exchange with albumin replacement from the randomized AMBAR study
The extra measures from AMBAR moved in the same direction as the main ones. All 347 people in the trial were assessed on memory, language, mood and quality of life alongside the main scores. People with mild disease improved on word fluency and on how quickly they processed information. People with moderate disease improved on short term verbal memory. Mood and behaviour changed no differently from the sham group. This is a second look at one trial, and Grifols, which makes the albumin, employed several authors.
Secondary analysis of the AMBAR randomised trial reporting the neuropsychological, neuropsychiatric and quality-of-life measures alongside the primary cognitive and functional endpoints.
- Study in people2025Multi-Omics Analysis Reveals Biomarkers That Contribute to Biological Age Rejuvenation in Response to Single-Blinded Randomized Placebo-Controlled Therapeutic Plasma Exchange
Plasma exchange moved the ageing clocks in a small trial, and the clocks are not health. 42 healthy adults over 50 were split by chance between plasma exchange every two weeks, with or without an immune protein drip, exchange once a month, or a dummy procedure. Only the patients were kept in the dark. Fifteen different epigenetic ageing clocks read younger than the dummy group, with the two-week schedule plus the immune protein doing best. Two people stopped for side effects. Nobody measured whether anyone got healthier.
42 healthy adults over 50 randomised to biweekly plasma exchange with or without intravenous immunoglobulin, monthly exchange, or placebo. Long-term exchange was reported safe, with two adverse events requiring discontinuation and one related to immunoglobulin. Fifteen epigenetic clocks showed rejuvenation versus placebo at a false discovery rate below 0.05, with biweekly exchange plus immunoglobulin performing best. Registered as NCT06534450.
- Study in people2025Human clinical trial of plasmapheresis effects on biomarkers of aging (efficacy and safety trial)
Removing plasma without replacing it made the ageing clocks read older, not younger. Healthy blood donors had either 8 or 4 plasma sessions over 18 weeks, each person trying both schedules in turn, with no young plasma and no albumin given back. No rejuvenation appeared. Three separate ageing measures all went the wrong way, and the authors write that this protocol may speed epigenetic ageing up. Blood fats, total protein and albumin all fell. This is the closest test of taking plasma out on its own.
A randomised crossover trial in healthy blood donors comparing 8 versus 4 plasmaphereses over 18 weeks, without young plasma or albumin replacement. No epigenetic rejuvenation was observed. DNAmGrimAge, the Hannum clock and the Dunedin Pace of Aging all increased, and the authors write that this protocol may accelerate epigenetic ageing. Serum lipids, total protein and albumin fell; red cell distribution width and mean corpuscular haemoglobin concentration rose.
- Study in people2012A Study to Evaluate Albumin and Immunoglobulin in Alzheimer's Disease (AMBAR)
This is the paperwork behind AMBAR, not a separate finding. Grifols sponsored it, enrolled 347 people, started in April 2012 and posted results in July 2019. Patients, their care teams and the people scoring the outcomes were all kept in the dark about who got the real procedure. Grifols makes the albumin that was used. The result itself is in the published trial report.
Registry record for AMBAR. Sponsor Instituto Grifols S.A., phase 2/3, randomised parallel assignment, triple masked across participant, care provider and outcomes assessor, 347 participants enrolled, start April 2012, results first posted 31 July 2019.
- Study in people2022The Effects Of Therapeutic Plasma Exchange (TPE) On Age Related Biomarkers And Epigenetics
This is the registry entry for the 2025 plasma exchange trial, and it is a thin record. It enrolled 40 people, started in September 2022, and is labelled phase 3, which does not match a trial of this size. The sponsor is listed as an individual investigator rather than an institution. The registry still shows its status as unknown.
Registry record for the 2025 Aging Cell trial. Phase 3 designation, 40 participants enrolled, start 12 September 2022, sponsor listed as an individual investigator, registry status listed as unknown.
Common questions
Has Therapeutic Plasma Exchange been tested in people?
Yes. 389 people given it across 5 completed trials cited here, 1 of which found nothing. Therapeutic Plasma Exchange is a standard hospital procedure that takes out plasma and puts a blood protein called albumin back. A 347-person trial in Alzheimer disease hit one of its two main goals. Two small 2025 trials of ageing markers came to opposite conclusions.
What is the Therapeutic Plasma Exchange dosage?
Approved medicine. No human dose established for unlabelled uses. Labeled amounts sit on the prescribing information. This page records what trials measured, not a personal dose. Therapeutic plasma exchange is an established medical procedure. A machine cleared for the job removes plasma, the liquid part of blood, and replaces it. It has accepted uses in some nerve, blood and kidney diseases. It is not approved anywhere for ageing, for lowering biological age or for Alzheimer disease.
Is Therapeutic Plasma Exchange FDA-approved?
Therapeutic plasma exchange is an established medical procedure. A machine cleared for the job removes plasma, the liquid part of blood, and replaces it. It has accepted uses in some nerve, blood and kidney diseases. It is not approved anywhere for ageing, for lowering biological age or for Alzheimer disease.
Does plasma exchange reduce biological age?
Two small 2025 trials disagree. One, in 42 adults over 50, put albumin back and in one group added an immune protein drip. It reported 15 epigenetic ageing clocks reading younger than the dummy procedure. The other, in healthy blood donors with nothing put back, found no change, and reported three clocks moving the wrong way. Neither trial measured whether anyone got healthier. Both measured clocks.
What did the AMBAR trial show?
In 347 people with mild to moderate Alzheimer disease, plasma exchange with albumin put back meant 52 percent less decline in everyday function than the fake procedure at 14 months. That was one of the trial's two main goals, and it held up. The thinking-skills goal showed 66 percent less decline but was not strong enough to count. People with moderate disease got better on both measures. People with mild disease did not change. Grifols, which makes the albumin, paid for the trial.
Is plasma exchange approved?
The procedure is an established treatment with recognised uses in brain, blood and kidney disease, done with approved machines. It is not approved anywhere for ageing, for lowering biological age or for Alzheimer disease. A procedure being approved is not the same as being approved for this use.
Is it risky?
It is a routine procedure with a known profile rather than an unknown one, but it is not nothing. It removes clotting factors, antibodies and blood fats along with everything else. The published blood donor trial recorded falls in total protein, albumin, cholesterol and triglycerides. In the 42-person trial, two people stopped because of side effects, one of them put down to the added immune protein drip.
How is this different from young plasma infusion?
They point in opposite directions. Young plasma infusion adds plasma from a young donor. Plasma exchange takes out the person's own plasma and replaces the volume with albumin, sometimes with an immune protein drip added, so nothing from a young donor goes in. The mouse work behind the second approach found that salt water plus albumin matched the effect of young blood. That is the thinning idea.