A Randomized Trial Took 2.6 Years Off Biological Age by Filtering the Blood. It Ran Three Months in 42 People and Costs About as Much as a Car.
Every few years the longevity field produces an intervention that is genuinely interesting and genuinely oversold at the same time, and therapeutic plasma exchange is the current one. The procedure itself is not speculative. Apheresis machines have been separating plasma from blood cells in hospitals for decades, and plasma exchange is standard care for a specific list of autoimmune and haematological conditions. What is new is the claim that running the same procedure on a healthy fifty-five year old man makes him biologically younger. In May 2025 a randomized placebo-controlled trial put a number on that claim, and the number was 2.61 years. This article is about where that number came from, what it is measuring, what the largest randomized plasma exchange trial in history found when it looked at something harder than a biomarker, and what the whole record looks like once the marketing is stripped out.
Key Takeaways: What the Therapeutic Plasma Exchange Evidence Actually Shows
Therapeutic plasma exchange removes a person's plasma and replaces it with albumin and saline, and it has been standard hospital care for conditions like Guillain-Barre syndrome and myasthenia gravis for decades. A single-blinded randomized placebo-controlled trial published in Aging Cell in May 2025 by Fuentealba and colleagues enrolled 42 adults over the age of 50 across four arms and reported an average biological age reduction of 2.61 years in the biweekly arm that also received intravenous immunoglobulin, against 1.32 years for plasma exchange alone. That trial ran roughly eleven people per arm across six sessions in about three months, sampled the epigenome, proteome, metabolome, glycome and immune cells, and measured no clinical outcome of any kind. AMBAR, the largest randomized plasma exchange trial ever run in an aging-related condition, randomized 347 Alzheimer's patients and produced 52 percent less decline on daily function at p equals 0.03 but only a non-significant trend on the cognitive co-primary endpoint at p equals 0.06. The mechanistic case does not require young blood: Mehdipour and Conboy showed in 2020 that simply diluting old mouse plasma with saline and 5 percent albumin reproduced the rejuvenation seen in parabiosis experiments. The common adverse event is citrate-induced hypocalcemia, running about 9.1 percent without calcium prophylaxis and around 1 percent with it. No regulator has approved plasma exchange for aging, insurance does not cover the elective use, and United States clinics charge roughly 5,000 to 12,000 dollars per session.
What is therapeutic plasma exchange and what does the machine actually do?
Therapeutic plasma exchange separates whole blood into plasma and cells, discards the plasma, and returns the cells to the patient along with a replacement fluid, usually human serum albumin in saline. The procedure runs through an apheresis machine, takes a couple of hours, and typically processes about one plasma volume per session. Blood leaves through one line, a centrifuge or membrane filter splits the lighter plasma from the heavier red cells, platelets and white cells, and the cellular fraction goes back in with fresh albumin standing in for what was removed. Citrate is infused as an anticoagulant to keep blood from clotting in the circuit. This is not an experimental technology and it is not a wellness gadget. Plasma exchange has been used in hospitals since the 1970s and sits in Category I of the American Society for Apheresis guidelines, meaning first-line accepted therapy, for conditions including Guillain-Barre syndrome, myasthenia gravis and thrombotic thrombocytopenic purpura. In those diseases the rationale is direct: something harmful is circulating in the plasma, usually an autoantibody, and taking the plasma out takes the harmful thing with it. The longevity application borrows the same machine and the same procedure and applies it to a different premise, which is that the plasma of an older person accumulates something that is worth removing even when no disease has been diagnosed. That premise is the part under test, not the hardware.
Where did the idea that filtering blood reverses aging come from?
The idea traces to heterochronic parabiosis, a set of experiments in which an old mouse and a young mouse are surgically joined so they share a circulation, and the old mouse's tissues start behaving younger. Conboy and colleagues published the foundational muscle work in Nature in 2005, showing restored regenerative capacity in aged muscle stem cells, and Villeda and colleagues later extended the finding to the brain with reports of improved hippocampal neurogenesis and cognition in old mice exposed to young blood. The obvious commercial reading was that young blood contains a rejuvenating factor, and a startup called Ambrosia proceeded to sell young donor plasma infusions until the United States Food and Drug Administration issued a safety notification in February 2019 warning that infusions of plasma from young donors for aging or dementia had no proven benefit and carried real risks. The scientific reading turned out to be different and more useful. In 2020, Mehdipour, Conboy and colleagues published a study in Aging showing that replacing half of an old mouse's plasma with saline containing 5 percent albumin, a procedure they called neutral blood exchange, matched or exceeded the effects of parabiosis. A single exchange improved muscle repair, reduced liver fat and fibrosis, and increased hippocampal neurogenesis. The implication reframes the whole field. If dilution works as well as young blood, then the active principle is the removal of accumulated inhibitory factors from old plasma rather than the addition of anything youthful. That reframing is what makes therapeutic plasma exchange, which is fundamentally a dilution procedure, a plausible human translation instead of a marketing stretch.
Did the randomized trial really cut biological age by 2.6 years?
The trial reported an average 2.61 year reduction in a composite biological age estimate in one arm out of four, in a cohort of about eleven people, over roughly three months, with no clinical outcome measured. Fuentealba and colleagues published the study in Aging Cell on 28 May 2025 under the title describing a multi-omics analysis of biomarkers contributing to biological age rejuvenation in response to single-blinded randomized placebo-controlled therapeutic plasma exchange. The design enrolled 42 participants over the age of 50 and allocated them across four arms: plasma exchange monthly, plasma exchange biweekly, plasma exchange biweekly with intravenous immunoglobulin added, and a placebo control. Each active participant received six sessions. The headline 2.61 year figure belongs specifically to the arm that combined plasma exchange with IVIG. Plasma exchange without IVIG averaged 1.32 years, roughly half as much. Three things deserve emphasis before that number gets repeated. The first is arithmetic: dividing 42 participants across four arms leaves roughly eleven people in each, and a group average drawn from eleven people carries wide uncertainty regardless of how the randomization was done. The second is that the larger effect required adding IVIG, which is a separate and expensive biologic drug with its own risk profile, so the 2.61 year result is not a result about plasma exchange on its own. The third is that this was a single-blinded trial, meaning participants were blinded but the investigators were not, which is a weaker control against expectation effects than a double-blind design provides.
Two Point Six Years, and Where That Number Comes From
Reported average change in composite biological age, in years, from the single-blinded randomized placebo-controlled therapeutic plasma exchange trial published in Aging Cell in May 2025 by Fuentealba and colleagues. The trial enrolled 42 participants over the age of 50 and split them across four arms: plasma exchange monthly, plasma exchange biweekly, plasma exchange biweekly with intravenous immunoglobulin, and a placebo control, working out to roughly eleven people per arm across six sessions in about three months. The biweekly arm that also received IVIG averaged a 2.61 year reduction. Plasma exchange without IVIG averaged 1.32 years. The placebo arm is the reference the other bars are measured against. Both figures are group averages from cohorts of about eleven people, which is the single most important thing to hold in mind when reading them.
What did the multi-omics readouts actually move?
The trial reported movement on about 15 of roughly 35 epigenetic clocks against placebo, shifts in immune cell composition, and modulation of senescence-associated proteins. The multi-omics design was the trial's genuine strength. Rather than picking one favourable clock and reporting it, the investigators sampled the epigenome across a wide panel of clocks alongside the proteome, metabolome, glycome and immune cell profiles, then looked for which molecular signatures tracked with the biological age changes. Reporting that roughly 15 of 35 clocks moved is considerably more honest than reporting one, and it is a better standard than most of this field manages. The immune findings were the most mechanistically interesting: treated participants showed shifts in immune cell populations in the direction associated with less age-related immune decline, which is at least consistent with the dilution hypothesis, since many of the factors implicated in immune aging circulate in plasma. The investigators also reported that most of the benefit arrived within the first three sessions and that additional sessions produced diminishing returns, which is a useful practical detail and an awkward one for any clinic selling ongoing subscriptions. Set against all of that is what the trial did not collect. There is no cognitive testing, no incident disease, no mortality, and nothing measured beyond the roughly three-month treatment window. An epigenetic clock reading is a statistical estimate built to correlate with chronological age and, in the better clocks, with mortality risk across large populations. Moving one is a hypothesis about future health, not an observation of it.
A Great Deal of Molecular Data and No Clinical Outcomes
A schematic map of what the 2025 plasma exchange trial measured. Teal marks readouts that moved in the treated arms, amber marks findings that are real but narrower than they sound, and dark grey marks outcomes the trial never collected. The multi-omics design sampled the epigenome across roughly 35 clocks, of which about 15 showed rejuvenation against placebo, alongside the proteome, metabolome, glycome and immune cell composition. Immune profiles shifted toward patterns associated with less age-related decline and senescence-associated proteins were modulated. The trial also reported that most of the effect arrived in the first three sessions, with diminishing returns afterward. What it did not measure is the entire clinical layer: cognition, incident disease, mortality, or anything at all past the roughly three-month treatment window. A molecular signal is a hypothesis about future health, not a demonstration of it.
What happened in the largest randomized plasma exchange trial ever run?
AMBAR randomized 347 Alzheimer's patients to plasma exchange with albumin replacement or a sham procedure and produced a significant result on daily function and a near-miss on cognition. Boada and colleagues published the primary results in Alzheimer's and Dementia in 2020. The trial screened 496 people with mild to moderate Alzheimer's disease and randomized 347 across three plasma exchange regimens at increasing albumin doses plus one sham arm, running for 14 months. On the co-primary endpoint of daily function, measured by the Alzheimer's Disease Cooperative Study Activities of Daily Living scale, treated patients declined 52 percent less than placebo at p equals 0.03. On the co-primary cognitive endpoint, the Alzheimer's Disease Assessment Scale Cognitive Subscale, treated patients declined 66 percent less but the result landed at p equals 0.06 and did not clear conventional significance. In the moderate Alzheimer subgroup, defined by a baseline Mini-Mental State Examination score of 18 to 21, cognitive decline was 61 percent lower. AMBAR matters to the longevity conversation for a reason that has nothing to do with dementia. It is the only place where plasma exchange has been tested at scale, against a sham procedure, for long enough to measure something a patient would actually notice. The answer it gave was genuinely encouraging on function, genuinely ambiguous on cognition, and dependent on a subgroup analysis for its most quoted number. Anyone citing the 2.61 year biological age result as though plasma exchange is an established intervention should be asked why the much larger trial, in a population with far more room to improve, produced a p value of 0.06.
The Big Trial Hit the Function Endpoint and Missed the Cognitive One
Percent reduction in decline against placebo at month 14 in the AMBAR trial, reported by Boada and colleagues in Alzheimer's and Dementia in 2020. AMBAR screened 496 people with mild to moderate Alzheimer's disease and randomized 347 of them to three plasma exchange with albumin replacement regimens or a sham procedure, which makes it by a wide margin the largest randomized plasma exchange trial ever run in an aging-related condition. Treated patients declined 52 percent less on the ADCS-ADL daily function scale at p equals 0.03, a result that cleared conventional significance. On the co-primary cognitive endpoint, ADAS-Cog, treated patients declined 66 percent less but at p equals 0.06, which did not. The moderate Alzheimer subgroup with a baseline Mini-Mental State Examination score of 18 to 21 showed 61 percent less cognitive decline. Amber marks the endpoints that are suggestive rather than statistically established.
Is therapeutic plasma exchange safe?
Plasma exchange has a well-characterised safety profile dominated by citrate-induced hypocalcemia, which is manageable but not trivial, and the risk calculus changes when the patient is healthy. Weinstein's work in the Journal of Clinical Apheresis established hypocalcemic toxicity as the most common adverse effect, driven by the citrate anticoagulant that binds calcium in the circuit and by the albumin in the replacement fluid, which binds calcium as well. Symptomatic hypocalcemia ran about 9.1 percent without prophylactic calcium administration and fell to roughly 1 percent with it. The other recurring complications in the apheresis literature are hypovolemia and hypotension from the fluid shifts, and allergic or anaphylactoid reactions to the protein-containing replacement fluid. Vascular access carries its own risk when a central line is required. There is a second consideration specific to the elective use. Plasma exchange with albumin replacement does not selectively remove the factors somebody wants removed. It removes immunoglobulins, clotting factors and fibrinogen along with everything else in the plasma, which is precisely why the IVIG arm in the 2025 trial existed: adding immunoglobulin back is a way of compensating for what the procedure strips out. In a patient with myasthenia gravis, depleting circulating antibodies is the therapeutic goal and the risk is worth it. In a healthy man in his fifties, the same depletion is a side effect with no disease to justify it, and there is no long-term data on what repeated elective courses do to immune competence. This article is educational and is not medical advice.
What does a course actually cost and what are clinics selling?
United States private clinics advertise elective plasma exchange at roughly 5,000 to 12,000 dollars per session, which puts a six-session course reproducing the trial protocol somewhere in the region of 30,000 to 70,000 dollars. Published clinic price pages through 2025 show 5,000 dollars per session at the low end, around 8,000 to 9,000 dollars per session inside multi-session packages, 10,000 dollars per session at some New York providers, and figures reaching 12,000 dollars at the upper end. Reporting in 2024 described packages priced near 36,000 dollars. None of this is covered by insurance, because insurance covers plasma exchange for the ASFA-recognised medical indications and no regulator has approved an aging indication. The commercial structure is worth noticing on its own terms. The trial found that most of the effect arrived in the first three sessions, with diminishing returns after that, which sits badly with a business model built on recurring treatment. It is also worth separating the two products being sold. Therapeutic plasma exchange, the full procedure that removes and replaces a plasma volume, is what the trial tested. Plasma donation and various partial apheresis procedures are not the same thing and do not have the same evidence behind them, though the distinction is sometimes blurred in marketing. Anyone evaluating a clinic should ask which procedure is actually being performed, what replacement fluid is used, whether IVIG is included, and what the total course price is rather than the per-session figure.
What the Longevity Clinics Charge Per Session
Advertised per-session pricing for elective therapeutic plasma exchange at United States private clinics, in dollars, drawn from published clinic price pages and reporting through 2025. The range runs from roughly 5,000 dollars to 12,000 dollars per session depending on the city and whether the session is sold inside a multi-session package. The trial that produced the 2.61 year figure used six sessions, so reproducing its protocol at a mid-range clinic price of around 8,000 dollars per session costs on the order of 48,000 dollars, and reporting in 2024 described packages priced near 36,000 dollars. Insurance does not cover elective plasma exchange for aging, because no regulator has approved that indication. The cost is worth stating plainly next to a group-average effect estimated in about eleven people over three months.
What should a man over 40 take from the plasma exchange evidence?
The plasma exchange record supports continued research and does not yet support paying tens of thousands of dollars for it as a healthy person. The honest summary is that a real medical procedure with fifty years of hospital use has a plausible mechanism behind it, drawn from parabiosis and confirmed in the mouse dilution experiments, and one randomized placebo-controlled human trial reporting movement on epigenetic clocks and immune markers in a cohort of 42 people over three months. That is a legitimate scientific finding and it is also, in evidence terms, an early one. The counterweight is AMBAR, where 347 patients over 14 months produced a significant benefit on daily function and a p value of 0.06 on cognition, which is what a real clinical signal looks like when it is genuinely borderline rather than genuinely established. For men specifically there is a further gap. Neither trial was designed or powered to answer questions about male physiology, and the published reports of the biological age trial do not include sex hormone outcomes, so anyone wondering what a plasma volume exchange does to testosterone, sex hormone binding globulin or fertility is asking a question the evidence has not addressed. The practical position that follows is not dismissal. It is that plasma exchange belongs in the category of interventions worth watching closely and worth funding properly, alongside the rapamycin and senolytic trials, rather than the category of things to buy. Anyone considering it should discuss it with a physician who understands apheresis, ask specifically about calcium prophylaxis and immunoglobulin depletion, and be clear that no regulator has approved plasma exchange as a treatment for aging or any disease.
The Plasma Exchange Record on One Axis
Every strand of the therapeutic plasma exchange evidence arranged as a single schematic map. Teal marks what the record supports, amber marks findings that are real but narrower or more qualified than the marketing suggests, slate marks the costs and risks that count against elective use, and dark grey marks claims nobody has tested. Plasma exchange is a genuine procedure with a real mechanistic story behind it, running from heterochronic parabiosis through the mouse dilution experiments to a randomized human trial that reported movement on epigenetic clocks. It is also a procedure whose best aging data comes from 42 people over three months, whose largest randomized trial missed its cognitive co-primary endpoint, and which is being sold at five figures a course for an indication no regulator has approved.
Frequently Asked Questions
Have more questions? Contact us or explore our full peptide intelligence platform.
Related Articles
Does Telomere Length Determine Lifespan? Short Telomeres Track With Earlier Death. The Genetics Say Longer Ones Raise Cancer Risk.
Does telomere length determine lifespan? Short telomeres track with earlier death, yet genetic studies show longer ones raise the risk of several cancers.
Low-Dose Aspirin Added 8 Percent to Male Mouse Lifespan. In 19,114 Healthy Older Adults, It Added No Healthy Years and Deaths Ran Higher.
Does low-dose aspirin help longevity? It extended male mouse lifespan 8%, but the ASPREE trial in 19,114 healthy older adults found no gain and more bleeding.
Is NAD+ a Peptide? No. It Is a Coenzyme Built From Two Nucleotides, and the Peptide Label Comes From the Clinic, Not the Chemistry
Is NAD+ a peptide? No. NAD+ is a coenzyme built from two nucleotides. What NAD+ is used for, what IV NAD+ studies found, and how it differs from SS-31.