Over the past several months, I have been spending considerable time investigating a treatment called immunoadsorption, commonly abbreviated as IA. It is a therapy that has been used in Germany and other countries for certain antibody-mediated diseases, and researchers have recently been exploring whether it could also benefit people with Long COVID, ME/CFS, autonomic dysfunction, and other complex conditions in which the immune system may be contributing to persistent symptoms.
What first caught my attention was the possibility of physically removing potentially harmful antibodies from someone’s bloodstream. Most people are familiar with medications that suppress inflammation or interfere with the immune system, but immunoadsorption works differently. Instead of primarily trying to change how immune cells behave, the procedure attempts to remove antibodies that have already been produced and are circulating in the blood.
This is an interesting concept, especially when we consider how many chronic illnesses involve abnormal immune findings that can persist for years. However, the more deeply I investigated the research, the more I realized that removing antibodies is only one part of a much larger biological problem. The immune system contains specialized cells that manufacture those antibodies, memory cells that preserve information about previous immune responses, and survival mechanisms that can keep antibody-producing cells alive for remarkably long periods. Many of these cells reside in tissues that immunoadsorption does not directly affect.
This distinction has become particularly important in my work because I have been investigating clients with combinations of neurological symptoms, small-fiber neuropathy, dysautonomia, post-infectious illness, and abnormal autoantibody results. Across different cases, I have examined antibodies directed against adrenergic receptors, muscarinic acetylcholine receptors, dopamine D1 receptors, tubulin, glycine receptors, and trisulfated heparin disaccharide, commonly called TS-HDS. I have also investigated anticardiolipin antibodies, thyroid peroxidase antibodies, thyroglobulin antibodies, and anti-vinculin antibodies in the context of gastrointestinal motility dysfunction.
These antibodies do not all represent the same disease, and their clinical significance varies considerably. Some antibodies are well-established causes of particular autoimmune disorders, while others remain poorly understood or are associated with conditions without being proven to cause them. Nevertheless, studying these findings has led me to spend more time investigating B-cell biology, plasma-cell survival, and the possibility that persistent antibody production may be maintained by immune mechanisms operating far upstream of the antibodies we measure in blood.
Before explaining immunoadsorption, I think it is important to understand what antibodies actually do and why the body produces them in the first place.
Our immune system is designed to recognize potential threats, including viruses, bacteria, and other foreign substances. When an infection occurs, specialized immune cells can produce proteins called antibodies, which recognize particular molecular structures associated with the invading organism. These antibodies can attach to a pathogen, interfere with its ability to infect cells, or help recruit other components of the immune system to eliminate it. This is an essential part of how we recover from infections and develop long-lasting immune protection.
The immune system also has mechanisms intended to prevent it from attacking the body’s own tissues. During immune-cell development, many cells that strongly recognize the body’s own structures are eliminated, inactivated, or controlled. However, these protective mechanisms are not perfect. In certain circumstances, the immune system can produce antibodies that recognize proteins, receptors, or other molecules belonging to the body itself. These are called autoantibodies.
The presence of autoantibodies does not automatically mean someone has an autoimmune disease. Some autoreactive antibodies can be detected in otherwise healthy people, and not every positive laboratory result reflects a biologically harmful process. The important distinction is whether an antibody merely recognizes a particular molecule or whether it actually interferes with that molecule’s function, damages tissue, or contributes to disease.
For example, some of the antibodies I have been investigating are directed against receptors involved in the autonomic nervous system. This is the part of the nervous system responsible for regulating many functions that occur automatically, including heart rate, blood vessel constriction, blood pressure, digestion, and glandular secretion.
Under normal conditions, these receptors respond to chemical messengers released by the body. When norepinephrine binds to certain adrenergic receptors, for instance, it activates intracellular signaling pathways that allow cells to adjust their behavior. Different adrenergic receptors produce different effects depending on the tissue in which they are expressed. Some influence how strongly the heart contracts, while others help regulate the diameter of blood vessels or relax smooth muscle.
The same principle applies to muscarinic acetylcholine receptors, which are important in parasympathetic nervous system signaling and gastrointestinal physiology.
If an antibody genuinely interferes with one of these receptors, the consequences could theoretically be significant. An antibody might act like an abnormal activating signal, prevent the receptor from responding normally, or alter how the receptor is regulated by the cell. Because these receptors participate in so many physiological processes, persistent interference could potentially contribute to symptoms involving circulation, digestion, or autonomic function.
However, this is where I think some discussions of autoimmunity become too simplistic. Finding an antibody against a receptor does not establish that the antibody is activating or blocking that receptor. Many available tests only measure binding, meaning they detect whether something in the blood recognizes a particular target. They do not necessarily tell us whether that interaction changes cellular signaling in a meaningful way. To establish that, researchers need functional experiments that examine what happens inside cells when those antibodies are present.
This becomes even more important with antibodies against intracellular structures. Tubulin, for example, is a major component of the microtubules that help maintain cell structure and transport materials inside neurons and other cells. These internal transport systems are important for moving mitochondria, proteins, and other cellular components to locations where they are needed. But detecting anti-tubulin antibodies in blood does not demonstrate that those antibodies can enter intact cells or disrupt these processes. The biological hypothesis may be interesting, but the actual mechanism must be demonstrated rather than assumed.
Similarly, TS-HDS antibodies have attracted interest in the investigation of small-fiber neuropathy, a condition involving damage or dysfunction of small sensory and autonomic nerve fibers. Some researchers have proposed relationships between these antibodies and peripheral nerve dysfunction, but the evidence remains insufficient to conclude that a positive TS-HDS result is necessarily responsible for a person’s symptoms. Antibodies against glycine receptors can have more established clinical significance in certain neurological autoimmune syndromes, although proper laboratory validation and a compatible clinical presentation remain essential.
This is the background that makes immunoadsorption interesting to me. If a particular antibody has genuinely been demonstrated to contribute to disease, reducing the amount of that antibody might help interrupt the harmful process.
Immunoadsorption is designed to accomplish exactly that at the level of the circulating antibody pool.
During the procedure, blood is withdrawn from the patient and passed through specialized medical equipment. The liquid portion of the blood, called plasma, is separated from the blood cells. Plasma contains many different substances, including proteins, hormones, nutrients, and antibodies. The separated plasma is then passed through a column containing materials capable of binding certain immunoglobulins.
Immunoglobulin is simply the scientific term for an antibody. Different categories of immunoglobulins exist, including IgG, IgM, and IgA, and each has somewhat different biological functions.
As plasma moves through the immunoadsorption column, susceptible antibodies bind to the material inside the column and are retained. The processed plasma is then returned to the patient’s circulation together with the blood cells. Depending on the system and the number of sessions, this procedure can produce substantial reductions in circulating antibody concentrations.
The process is sometimes confused with therapeutic plasma exchange, but the two procedures are not identical. During conventional plasma exchange, a portion of the patient’s plasma is removed and replaced with another fluid, such as albumin solution or donor plasma. Immunoadsorption instead processes the patient’s separated plasma through an adsorption system and returns it after selected molecules have been removed.
Both procedures can reduce circulating antibodies, but their effects on different plasma components depend on the technology and treatment conditions.
One important limitation is that most immunoadsorption systems do not distinguish perfectly between harmful autoantibodies and beneficial antibodies that protect against infections. They often remove broader immunoglobulin populations. Some columns are highly effective at removing IgG but less effective against IgM, while other systems have different binding characteristics.
That distinction matters because the antibody class involved in a particular disease can influence which removal technology is appropriate. An intervention that substantially reduces IgG cannot automatically be assumed to eliminate an IgM autoantibody.
When researchers in Germany began investigating immunoadsorption for post-COVID illnesses, one of their interests involved antibodies directed against G-protein-coupled receptors. These receptors are found throughout the body and help cells respond to hormones, neurotransmitters, and other chemical signals.
In an earlier prospective study, researchers examined 20 patients with post-COVID ME/CFS who had elevated antibodies against the β2-adrenergic receptor. Participants underwent five immunoadsorption sessions. The investigators reported reductions of approximately 79% in total IgG and 77% in β2-adrenergic receptor antibodies. They also observed improvements in physical functioning in a proportion of the participants.
Those findings understandably attracted attention because the procedure appeared to accomplish what researchers intended at the biological level. The antibody concentrations decreased substantially, and some patients reported improvement.
However, that study did not include a randomized sham-treatment comparison. This is an important limitation because conditions such as ME/CFS and Long COVID can fluctuate considerably. A person may experience periods of improvement or worsening for reasons unrelated to an intervention. Without a comparison group undergoing a procedure that appears similar but does not provide the active treatment, it is difficult to determine how much of the improvement was truly caused by immunoadsorption.
A later small observational study also demonstrated substantial antibody reductions after immunoadsorption, but the researchers found that many receptor antibody concentrations had returned toward their original levels within approximately one month. Some physiological and cognitive measurements improved, while important symptom and fatigue measures showed no clear significant improvement.
That result illustrates one of the problems I have been investigating. A procedure may remove a large quantity of circulating antibodies, but the body can subsequently replenish them.
More recently, a randomized, patient-blinded, sham-controlled crossover study involving 40 patients with post-COVID syndrome provided a stronger test of the treatment hypothesis. Researchers successfully reduced circulating GPCR autoantibodies and serum IgG concentrations through immunoadsorption, confirming that the intervention had achieved its immediate biological objective.
However, the treated patients did not demonstrate significant clinical improvements compared with those receiving the sham procedure on the measured outcomes. The investigators also recorded more adverse events associated with immunoadsorption than with sham treatment.
I think this result deserves considerable attention because it separates two things that are often treated as though they are equivalent: changing a biomarker and changing the disease.
The treatment reduced the measured antibodies, but that reduction did not translate into a demonstrated clinical advantage in the study population.
It would be incorrect to dismiss immunoadsorption entirely on the basis of this result because the procedure already has clinical applications in selected established antibody-mediated diseases. It is also possible that future research could identify more specific groups of patients who respond differently. Nevertheless, the study substantially weakens the argument that removing circulating GPCR antibodies is, by itself, an effective general treatment for Long COVID.
It also raises the possibility that some of the measured antibodies are not the principal cause of the symptoms, or that the disease involves additional mechanisms not adequately addressed by antibody removal.
This is why my investigation has increasingly shifted toward understanding how antibodies are produced, maintained, and replaced.
Most people think of the immune system as a collection of white blood cells moving through the bloodstream, but some of its most important activities occur inside the bone marrow, lymph nodes, spleen, and other tissues.
The bone marrow is particularly important because it contains hematopoietic stem cells. These are specialized cells that can produce the different blood and immune-cell lineages. They are essentially the upstream source from which many immune cells develop.
Through several developmental stages, some of these cells become B lymphocytes, commonly called B cells.
A B cell is not automatically an antibody-producing cell. It first develops a receptor that allows it to recognize particular molecular structures. When it encounters an appropriate antigen and receives the additional signals necessary for activation, it may begin a process of differentiation.
Some activated B cells enter structures called germinal centers, which are commonly found inside lymph nodes and other lymphoid tissues. Within these structures, B cells interact with specialized helper T cells and undergo processes that can improve antibody affinity and change the class of antibody being produced.
Molecules such as CD40, CD40L, and IL-21 participate in these interactions. They help coordinate the development of more mature antibody responses.
Eventually, activated B cells can follow different developmental paths. Some become memory B cells, while others differentiate into plasmablasts and plasma cells.
The distinction between these populations is central to understanding why autoimmune activity can persist.
Memory B cells are essentially a biological record of previous antigen exposure. Their presence allows the immune system to respond more rapidly when it encounters a familiar antigen again. This is extremely useful for protecting us against infections we have previously experienced.
Plasma cells perform a different function. They are highly specialized cells designed to manufacture large quantities of antibodies.
Inside a plasma cell, the cellular machinery responsible for producing and processing proteins becomes particularly active. The cell continuously synthesizes immunoglobulins and releases them into the surrounding environment, where they can eventually enter circulation.
Some plasma cells are short-lived and disappear after a relatively brief period. Others develop into long-lived plasma cells capable of surviving for years, sometimes much longer, while continuing to secrete antibodies.
This is one reason certain vaccine-induced antibody responses can persist long after vaccination. The immune system does not necessarily need to encounter the original antigen every day to maintain circulating antibodies against it. Long-lived plasma cells can continue producing them.
The same mechanism can become problematic when the antibodies being produced are pathogenic.
If a person has a population of long-lived plasma cells producing antibodies that interfere with an important physiological process, removing the existing circulating antibodies does not necessarily stop new ones from being manufactured.
The plasma cells may remain alive and continue functioning.
This is where the bone marrow becomes particularly relevant. The bone marrow is not merely the place where blood cells originate. It also provides specialized environments that support the survival of certain mature plasma cells.
Researchers refer to these environments as plasma-cell survival niches.
A survival niche consists of interactions between plasma cells, neighboring cells, and molecular signals that help sustain plasma-cell function. Some of these signals can prevent the cells from undergoing programmed cell death, allowing them to remain active for prolonged periods.
One important molecule involved in this biology is APRIL, which stands for a proliferation-inducing ligand. APRIL can interact with receptors such as BCMA and TACI on relevant immune-cell populations and contributes to plasma-cell survival.
Another molecule, BAFF, helps regulate the survival and maturation of B cells, although its effects depend on the developmental stage and cellular context.
IL-6 can also contribute to plasma-cell biology, while a signaling pathway involving CXCL12 and its receptor CXCR4 helps guide and retain certain cells within supportive tissue environments.
These molecular interactions are important because long-lived plasma cells can sometimes maintain antibody production independently of the event that originally stimulated the immune response.
A viral infection, for example, may have triggered a particular immune response months or years earlier, yet an established plasma-cell population may continue producing antibodies after the original infection has resolved.
This does not establish that persistent post-infectious symptoms are caused by those antibodies. It simply explains how long-term antibody production can occur biologically.
When immunoadsorption removes circulating antibodies, it does not directly destroy the plasma cells residing inside these survival niches. It also does not eliminate memory B cells or permanently correct the immune processes that allowed those populations to develop.
Consequently, if the antibody-producing cells remain active, the concentration of antibodies in the blood can begin increasing again.
There is another important part of this process that is often overlooked. Antibodies are not confined exclusively to the bloodstream. Many are distributed throughout the extracellular spaces of tissues, and antibodies outside blood vessels can move back into circulation when the blood concentration falls.
This redistribution can contribute to an early increase in measured antibody levels after treatment, even before considering continued production.
The immune system also possesses a recycling mechanism that helps certain antibodies survive longer.
One of the most important proteins involved is called the neonatal Fc receptor, or FcRn.
Despite its name, FcRn remains functionally important throughout adult life. Its role includes protecting IgG antibodies from being broken down inside cells.
Normally, proteins taken up by cells can be directed toward compartments where they are degraded. FcRn binds IgG in the acidic environment of particular intracellular compartments and helps recycle it back to the cell surface, where it is released instead of destroyed.
Because of this recycling process, many IgG antibodies have a circulating half-life of approximately three weeks, although the exact duration varies.
This means the amount of IgG detected in someone’s blood depends on several processes occurring simultaneously. New antibodies are being produced by plasma cells, antibodies are moving between blood and tissues, existing antibodies are being recycled, and some antibodies are being broken down and eliminated.
Immunoadsorption primarily changes the existing circulating pool. It does not directly shut down antibody production or the FcRn recycling system.
This has led to the development of another category of therapy called FcRn inhibitors. These medications interfere with the normal recycling process that protects IgG, allowing more IgG to be degraded.
They have demonstrated clinical benefit in selected IgG-mediated autoimmune diseases, including certain forms of myasthenia gravis.
However, FcRn inhibition also lowers protective IgG and does not eliminate the cells producing the antibodies. It does not have an equivalent direct effect on IgM because IgM is not maintained through the same FcRn-dependent recycling mechanism.
The distinction between removing antibodies, increasing their degradation, and stopping their production is extremely important.
These interventions may all reduce a circulating antibody concentration, but they operate at different locations within the immune system.
This is also why I have been looking more closely at therapies that target B cells themselves.
One example is rituximab, a medication that targets CD20, a protein expressed on many mature B cells. By binding to CD20, rituximab can promote the depletion of these cells, interfering with parts of the immune system responsible for generating new antibody responses.
Rituximab has established clinical applications in certain autoimmune diseases, but there is an important limitation. Most mature plasma cells do not express CD20.
Therefore, a person can experience substantial depletion of circulating CD20-positive B cells while still having long-lived plasma cells that continue secreting antibodies.
This helps explain why some autoantibody concentrations may persist even after B-cell depletion.
Another medication, belimumab, targets BAFF, a molecular survival factor involved in B-cell biology. By reducing BAFF signaling, belimumab can alter the survival and activity of certain B-cell populations. However, it does not directly eliminate every mature plasma cell.
More advanced treatments attempt to target additional components of the antibody-producing system.
CD19 is another surface protein expressed by many B-lineage cells. Because its expression differs from CD20, therapies directed against CD19 can affect a broader range of relevant cell populations.
This has become especially interesting with the development of CAR-T-cell therapy.
CAR-T therapy involves collecting a person’s T cells and genetically modifying them so they can recognize a selected target on other cells. The modified T cells are then returned to the patient, where they can identify and destroy cells expressing that target.
Researchers have begun investigating CD19-directed CAR-T therapy for severe autoimmune diseases, including systemic lupus erythematosus, systemic sclerosis, and inflammatory myositis.
Early studies in carefully selected patients with severe treatment-resistant disease have reported striking improvements, with some individuals achieving sustained remission after deep B-cell depletion.
These findings are important because they suggest that changing the underlying autoreactive B-cell population can sometimes produce more durable clinical effects than merely removing antibodies already present in circulation.
However, CAR-T therapy is a highly specialized intervention with substantial risks, including severe inflammatory reactions, cytopenias, and infections. Its promising early results in particular autoimmune diseases do not establish effectiveness for Long COVID, ME/CFS, MCAS, or unexplained neurological autoantibodies.
There are also treatments being investigated that target plasma cells more directly.
Daratumumab, for example, targets CD38, which is expressed strongly on many plasma cells. Other experimental strategies involve BCMA, a receptor that plays an important role in plasma-cell survival.
These approaches are interesting because they act closer to the source of ongoing antibody production.
However, their biological effects are not restricted to harmful antibodies. Normal plasma cells are responsible for maintaining much of our long-term protective immunity. Depleting these cells can reduce antibodies against infections and increase the risk of serious complications.
This is one of the reasons I think it is essential to establish a convincing disease mechanism before considering aggressive immune-directed treatment.
There is an enormous difference between demonstrating an abnormal antibody result and proving that the person’s symptoms are being sustained by a particular autoreactive plasma-cell population.
Another area that I have been investigating is hematopoietic stem-cell transplantation, particularly in the context of severe autoimmune disease.
This is often misunderstood because people hear the words stem-cell therapy and assume that all stem-cell interventions work in the same way.
Hematopoietic stem cells and plasma cells are not interchangeable. Hematopoietic stem cells are upstream cells capable of generating new blood and immune-cell lineages. Plasma cells are mature antibody-producing cells that developed through the B-cell lineage.
Autologous hematopoietic stem-cell transplantation involves collecting a patient’s own hematopoietic stem cells, administering intensive treatment to suppress or deplete much of the existing immune system, and then reinfusing the collected cells to support blood and immune-system recovery.
The objective in selected autoimmune conditions is to alter the existing immune architecture sufficiently that the disease process may enter prolonged remission.
This treatment has demonstrated benefits in carefully selected patients with severe systemic sclerosis, but it also carries major risks, including serious infections, organ toxicity, infertility, and treatment-related mortality.
It is not equivalent to commercial stem-cell infusions or mesenchymal stromal-cell therapies, which involve different cells and mechanisms. It also does not guarantee that every autoreactive memory cell or long-lived plasma cell will be eliminated.
I mention these treatments because they illustrate how many distinct levels of the immune system are now being investigated.
At one level, we have therapies such as immunoadsorption that physically remove circulating antibodies. At another level, FcRn inhibitors increase the rate at which certain antibodies are degraded. B-cell-directed medications interfere with populations involved in producing new antibody responses. Plasma-cell-directed therapies attempt to affect mature cells responsible for sustained antibody secretion. More intensive cellular therapies and transplantation approaches attempt broader changes in the immune system.
These are fundamentally different interventions, even though all may influence autoantibody concentrations.
The concept that interests me is whether a future treatment strategy for a well-defined, genuinely antibody-mediated disease might need to account for several of these biological compartments rather than focusing exclusively on the amount of antibody circulating at one particular moment.
There is already clinical experience with combining antibody-removal procedures and immune-directed treatments in selected established autoimmune diseases. In those settings, rapidly reducing circulating antibodies while also treating the underlying immune process can sometimes be useful.
However, this cannot be automatically extended to post-infectious illnesses or conditions involving poorly validated autoantibody tests. We currently lack adequate evidence that combining immunoadsorption with B-cell depletion, plasma-cell targeting, or other advanced interventions would improve outcomes for those populations.
The risks of combining immune-directed treatments could also be much greater than the risks of any single intervention.
For that reason, I am more interested in understanding the mechanism than in assuming a more aggressive treatment must be more effective.
An important lesson from my work with complex cases is that several abnormal findings can coexist without all being responsible for the same symptoms.
A person may have an elevated antibody against an adrenergic receptor, a positive TS-HDS result, thyroid autoantibodies, and evidence of autonomic dysfunction. It may be tempting to combine all of these findings into a single explanation involving widespread autoimmune injury.
But that interpretation can be misleading.
Thyroid antibodies may indicate thyroid autoimmunity without explaining the person’s neurological symptoms. TS-HDS positivity may not establish immune-mediated small-fiber injury. An adrenergic receptor binding assay may not demonstrate altered receptor signaling. And autonomic dysfunction can arise through several mechanisms unrelated to circulating autoantibodies.
The correct interpretation depends on whether each finding has been independently validated and whether it fits the person’s actual physiology.
In my view, one of the most valuable developments in this field would be the wider availability of reliable functional antibody testing.
For example, if an antibody is suspected of interfering with an autonomic receptor, researchers should ideally demonstrate how it affects that receptor under controlled conditions. They can investigate whether the antibody changes intracellular cyclic AMP, calcium signaling, receptor activation, or other measurable cellular responses.
They should also establish whether those effects occur at antibody concentrations relevant to the patient’s biology.
An even stronger experiment would involve selectively removing the suspected antibody from a biological sample and examining whether the abnormal cellular response disappears. If adding the antibody back restores that response, the evidence for a causal relationship becomes stronger.
These experiments are not routine clinical tests for most exploratory autoantibodies, but they illustrate the level of evidence necessary to move from an association to a convincing mechanism.
This distinction is particularly relevant when dealing with small-fiber neuropathy, dysautonomia, and post-infectious illness. These conditions can involve complex interactions between the nervous system, immune signaling, circulation, metabolism, and tissue function.
Some patients may have genuine antibody-mediated pathology. Others may have immune abnormalities that are secondary to a different underlying process. Still others may have positive antibody tests that do not contribute meaningfully to their symptoms.
If we fail to distinguish between these situations, we risk treating the wrong biological target.
And that brings me back to the central limitation of immunoadsorption.
It is entirely possible for a procedure to successfully remove a large portion of circulating antibodies without meaningfully changing the illness.
The German sham-controlled research provides a particularly important example. The investigators achieved measurable antibody depletion, but the treated group did not experience superior outcomes on the study’s measured clinical endpoints.
That finding does not mean antibody-removal treatments are ineffective in every autoimmune disease. There are established clinical contexts, including selected cases of myasthenia gravis and pemphigus, where removing pathogenic antibodies can provide benefit.
The difference is that these diseases have much stronger evidence linking particular antibody mechanisms to clinical pathology.
For people with Long COVID or poorly characterized autoantibody findings, the same causal relationship has not been established.
I also think the safety considerations deserve more attention, particularly because people with chronic illnesses may already have complicated immune histories.
Immunoadsorption is not a simple blood-filtering wellness procedure. It is a specialist medical intervention that can require repeated sessions, specialized equipment, anticoagulation, and vascular access.
Possible complications include low blood pressure, bleeding, allergic reactions, infection, and vascular-access problems. Depending on the technique, patients may also lose protective immunoglobulins, potentially increasing susceptibility to infection. Catheter-associated thrombosis has been reported in the recent controlled research.
These concerns are especially relevant in someone who already experiences recurrent infections, has impaired protective antibody responses, or has other significant medical vulnerabilities.
The fact that a procedure is capable of lowering an abnormal laboratory value does not by itself justify those risks.
This is where I think modern autoimmune research faces one of its greatest challenges. We now have increasingly sophisticated technologies capable of removing circulating antibodies, interfering with antibody recycling, depleting B cells, targeting plasma cells, and even restructuring parts of the adaptive immune system.
But our ability to identify exactly which immune abnormality is driving a particular person’s illness has not always advanced at the same pace.
We can sometimes manipulate an immune pathway more precisely than we can establish whether that pathway is truly responsible for the disease.
This is why I have increasingly focused on distinguishing the measurable consequences of a biological process from the process itself.
An elevated antibody concentration is a measurable finding. Persistent autoreactive B-cell activation is a different process. Survival of a long-lived plasma-cell population is another. Failure of immune tolerance, chronic inflammatory signaling, and abnormal receptor function are additional mechanisms that may or may not be connected in a particular person.
Treating one of these processes does not guarantee that the others will change.
I am also interested in the possibility that persistent immune activation can alter the behavior of immune cells over time through changes in cellular metabolism and gene regulation. Researchers have increasingly investigated how inflammatory exposures can influence epigenetic regulation, including chromatin modifications that affect how readily particular genes are expressed.
Some of these mechanisms are involved in what researchers call trained innate immunity, where innate immune cells or their progenitors develop altered responses following previous inflammatory exposure.
This is a fascinating area of immunology because it suggests that some immune responses can remain altered even after the initial stimulus has disappeared.
However, trained innate immunity is not the same as conventional B-cell memory, and there is insufficient evidence to claim that a particular post-infectious autoantibody is being maintained through one specific epigenetic mechanism without directly investigating it.
I believe these distinctions matter because the term immune dysfunction is frequently used as though it describes one biological problem. In reality, the immune system consists of multiple interacting cell populations, signaling pathways, tissues, and regulatory mechanisms.
A persistent abnormality could arise through continued antigen exposure, altered immune tolerance, expansion of particular B-cell populations, long-lived plasma-cell survival, abnormal cytokine signaling, or mechanisms that have nothing to do with the antibodies detected on a laboratory panel.
Different causes require different evidence and may ultimately require very different treatments.
After examining the immunoadsorption research, I remain interested in the treatment, but my interpretation is more cautious than it would have been based on the earlier observational studies alone.
Immunoadsorption clearly has the capacity to remove circulating immunoglobulins. This makes it a legitimate therapeutic technology in appropriate clinical circumstances.
What has not been demonstrated is that broad removal of circulating GPCR autoantibodies reliably improves Long COVID or other complex post-infectious illnesses.
And even in a condition where pathogenic antibody removal provides a short-term benefit, the longer-term question remains whether the cells producing those antibodies continue to survive and function.
This is where research involving memory B cells, long-lived plasma cells, bone marrow survival niches, BAFF, APRIL, BCMA, FcRn, and advanced immune-cell-directed therapies becomes particularly relevant.
I do not consider these separate treatment technologies interchangeable, and I certainly do not believe that every person with elevated antibodies needs aggressive immune suppression.
Rather, they help us understand the different levels at which antibody-mediated disease could potentially be interrupted once the pathogenic mechanism has been established.
Ultimately, I think the future of autoimmune treatment will depend on doing a better job of determining which antibodies are genuinely harmful, identifying the cells and survival mechanisms responsible for sustaining them, and selecting interventions that match the actual disease process.
For a person with a confirmed pathogenic antibody, reducing its concentration may be clinically meaningful. For someone whose illness is maintained by persistent autoreactive immune-cell populations, a treatment aimed only at the circulating antibody pool may have limited or temporary effects. And for someone whose antibodies are not causing the illness, even complete antibody depletion may accomplish very little.
The challenge is knowing which situation we are dealing with before exposing someone to a treatment that may carry significant risks.
That is the part of immunoadsorption research I find most valuable. It forces us to distinguish between removing something that can be measured in blood and correcting the biological process responsible for the disease.
The recent German studies have made that distinction much clearer. They show that we can substantially reduce circulating autoantibodies without necessarily improving symptoms, and that successful antibody depletion should never be confused with proof of clinical efficacy.
I will continue investigating this area, particularly the relationship between circulating antibodies, B-cell memory, long-lived plasma-cell survival, and the possibility of more selective immune-directed interventions in properly characterized autoimmune conditions.
But the conclusion I keep returning to is that the immune system is not simply a collection of antibodies floating through the bloodstream. It is an interconnected biological network with specialized cellular populations, long-term memory, tissue-specific survival environments, and mechanisms capable of maintaining immune responses for years.
When we remove antibodies from circulation, we may be altering only one compartment of that network.
Understanding what is happening in the other compartments, and whether the antibodies are actually responsible for the illness in the first place, is what will determine whether a treatment offers meaningful benefit.
The future of antibody-directed therapy should not be measured simply by how many antibodies we can remove, but by whether we can identify and safely change the biological process that is actually causing the disease.