Therapeutic plasma exchange (TPE)

What happens during plasma exchange

Blood flows from a vein into a centrifuge that separates the blood cells from the fluid. The cells are returned, the plasma is discarded, and an albumin solution flows in instead — albumin is the most abundant blood protein and is obtained from donor plasma. The procedure is established medicine: in autoimmune diseases in which antibodies in the blood are themselves the problem, it has been part of the standard of care for decades.

Related, but not the same, are procedures in which the plasma is filtered and returned. In an exchange, it is replaced — precisely this difference carries the anti-aging idea.

The idea behind it: dilution instead of young blood

In 2005, researchers at Stanford connected the circulation of an old mouse with that of a young one. The muscle stem cells of the old animal started working again, and the liver cells divided again. The obvious interpretation — that young blood contains something rejuvenating — became a business idea: a Californian company sold young donor plasma for 8,000 dollars per liter until the US Food and Drug Administration stated in 2019 that there was no proven benefit against aging or dementia, but rather a risk of infection, allergic reactions, lung injury and circulatory overload. The company stopped the treatments.

The same lab then turned the question around. In 2016, a one-time blood exchange without surgically joining the animals showed that old blood holds back young tissue more strongly than young blood promotes old tissue. In 2020 came the decisive experiment: in old mice, half of the plasma was replaced with saline containing 5 percent albumin. Nothing young was in it. Even so, after a single exchange, muscle repair, fatty liver and the formation of new nerve cells in the hippocampus improved. Not addition, then, but dilution.

The comparison with bloodletting for iron overload suggests itself and at the same time contains the sharpest criticism. With bloodletting, you know which substance is in excess, measure it before and after, and know when to stop. With aging, nobody knows the one substance, and no target value exists.

What the studies show

Kim 2022 — 8 people, no control group

From the lab that put forward the dilution hypothesis, together with an apheresis physician: 8 people, several rounds of plasma exchange at monthly intervals. The protein profile in the blood looked younger, and markers of cellular aging and DNA damage fell. There was no control group, and the physician involved owns the company that carried out the treatments.

AMBAR 2020 — 347 Alzheimer's patients

The largest study outside the classic indications: 347 patients with mild to moderate Alzheimer's disease were randomized, 322 evaluated. 6 weeks of weekly exchange, then 12 months of monthly exchange with a smaller volume, each with albumin and in some cases immunoglobulins, versus a sham treatment. In activities of daily living, the decline was 52 percent smaller, p equal to 0.03. The second primary endpoint, a scale of cognitive performance (ADAS-Cog), narrowly missed significance at p equal to 0.06. In the mild cases no effect appeared; in the moderate cases it did. The study was sponsored by the manufacturer of the albumin used.

Fuentealba 2025 — 42 healthy people, epigenetic clocks

The only study in healthy humans: participants over 50 years old, four groups — exchange twice a month, the same plus immunoglobulins, exchange once a month, and a sham treatment with a curtain in front of the device. 44 were enrolled and 42 completed, so around 10 per group. After 3 sessions, the epigenetic aging clocks were on average 2.6 years below the sham group in the immunoglobulin group and 1.3 years below in the monthly exchange group. After 6 sessions, no difference was measurable in any group anymore; the authors suspect counter-regulation and themselves call their work hypothesis-generating.

Where the data stop

The study in healthy humans does not carry the promise. Around 10 participants per group, allocation not by lot but by order of registration, carried out by the company that sells the treatment — and the measured effect had disappeared after 6 sessions. If the body readjusts after a few rounds, this looks more like a disturbance being compensated than a rejuvenation. The changes in immune cells also occurred only in the group with immunoglobulins; the authors themselves consider it possible that they are due to these alone.

Then there is the question of the endpoint. What was measured were epigenetic aging clocks — estimates that are themselves not validated as an endpoint. Physical or cognitive function was explicitly not assessed. Whether 2 years less on a clock says anything about years lived, nobody knows. And AMBAR cannot separate three interventions: plasma out, albumin in, immunoglobulins added. To this day there is no approval for the treatment of Alzheimer's.

Status, approval and legal

Therapeutic plasma exchange is an approved, established procedure with clearly defined medical indications; aging is not one of them. As an anti-aging treatment, it is purely a self-pay service. In California, around 6,000 dollars per session are charged, recommended every 2 months, which comes to about 36,000 dollars a year. A German private practice charges 1,800 euros per session with 4 sessions as the basic treatment, so a little over 7,000 euros, explicitly not covered by health insurance and advertised with anti-aging. The sharpest critic is the researcher whose lab discovered dilution: there are very few studies, she says, and people are applying the procedure to everything.

Safety

The risks are well documented from other indications. The registry of the World Apheresis Association covers more than 50,000 procedures: side effects occurred in a little over 8 percent of first sessions and in about 5 to 6 percent of subsequent ones. They are mostly mild — tingling, because the anticoagulant citrate binds calcium, and a drop in blood pressure, most frequently with albumin as the replacement solution. Serious events such as collapse or cardiac arrhythmias occurred in 0.4 percent. In the AMBAR study, about one in six sessions had a procedure-related side effect, and one in five with central venous access. Antibodies and clotting factors are also lost with the plasma, which is why immunoglobulins were partly replaced in the studies. For repeated use in healthy people over years, there are no data. This is not medical advice; the decision belongs in the hands of a physician.

Sources

• Conboy et al., Nature 2005
• Mehdipour et al., Aging (Albany NY) 2020
• Kim et al., Geroscience 2022
• Boada et al., Alzheimers Dement 2020 (AMBAR)
• Fuentealba et al., Aging Cell 2025
• Mörtzell Henriksson et al., Transfus Apher Sci 2016
• Boada, Kiprov et al., J Clin Apher 2023
• FDA, Information about Young Donor Plasma Infusions 2019