MCAS Article 3: When the Antihistamine Doesn’t Work: What It Does and Doesn’t Tell You About MCAS and ME/CFS
You did the trial. You took the antihistamine — maybe the H1 and H2 combination, maybe weeks of it — and nothing happened. Or your blood tests came back “normal,” and you’re being told you can’t have a mast-cell problem at all.
Neither of those outcomes is a verdict. This article is about why. It pulls together the hardest, most honest parts of the MCAS-and-ME/CFS picture: why your tests can look normal, why one drug can fail while mast cells are still driving symptoms, the difference between histamine intolerance and MCAS, and — soberly — why a longer illness may respond less. Then it tells you what those answers mean and what to actually do.
This is the third of three articles. For the background — what MCAS is and why it’s usually downstream of ME/CFS — see the overview. For how the medications work, see the treatment article.
1 First, a plain warning
This is an explanation, not self-medication advice, and I am not a doctor. If your symptoms feel sudden, severe, or life-threatening — especially any breathing difficulty or swelling in the throat — that is an emergency; go to a hospital. And because “MCAS” symptoms overlap many other conditions, a treatment failure is a reason to revisit the diagnosis and the label with a clinician — not a reason to conclude either that you’re not suffering or that mast cells were definitely your problem.
2 Why your tests can be “normal” — and why one drug can still fail
This is the part that frustrates patients and, frankly, baffles doctors too. You can have textbook MCAS symptoms, a normal full allergy work-up, and get a partial response to one drug but not another. Here are the three honest reasons, and they change how you should think about the trial.
Reason one: the “not enough to clean up histamine” subtype. Histamine is normally broken down by enzymes. One of them — called HNMT — is less efficient in roughly 15–20% of people due to a common genetic variation. In those people, even a normal level of mast cell firing could, in principle, overwhelm the cleanup system, so histamine builds up and causes symptoms even when standard blood tests come back normal. In this picture the blood test isn’t lying: the problem isn’t too much histamine made, but not enough cleared. This might be one reason some people with normal biochemistry still improve with antihistamines (Yao et al. 2025). The honesty boundary is real and important: the enzyme variant’s effect in ME/CFS has never been measured in a cohort, there is no routine clinical test for an individual’s HNMT function you can order today, and even the functional significance of the common variation is itself debated. This is a speculative explanation for normal tests, not an actionable diagnosis.
Reason two: “histamine intolerance” (HIT) may be a different machine than MCAS. A related but mechanically separate possibility is histamine intolerance: not mast cells firing too much, but the diet and gut delivering more histamine (aged cheeses, fermented foods, cured meats, wine) than the body’s clearing enzymes can handle — often attributed to a deficiency in the enzyme DAO (Comas-Basté et al. 2020). Fair warning: HIT as a distinct diagnosis, and DAO testing as its diagnostic, are themselves disputed in the allergy literature, and the correlation between DAO levels and symptoms is weak — so treat this as a hypothesis to test with a time-limited low-histamine diet, not as a firmly established mechanism. If your trouble seems largely diet-driven, an antihistamine may help less than reducing high-histamine foods would; but don’t build a long-term restrictive eating plan on a contested diagnosis without a clinician and preferably a dietitian on board.
Reason three: mast cells come in “flavours.” Some people’s mast cells may be mostly histamine-driven; others lean on tryptase/prostaglandin D₂; still others on leukotrienes, pushing symptoms through different receptors (Molderings et al. 2016). A formal version of this — histamine-dominant, PGD₂/tryptase-dominant, and leukotriene-dominant subtypes — is a hypothesis developed in the primary document underlying this series; it is not yet an established finding from independent groups, so treat it as a plausible idea, not a settled one (Loth 2026). If it’s right, its practical meaning is: an antihistamine only blocks the histamine flavour, so it could fail while mast cells still matter. But because the subtype idea is untested, the more parsimonious reading is equally on the table: the drug failed because mast cells were not driving your symptoms. This is one reason a single “take this antihistamine” trial is not a complete test — and why, where available, measuring urinary mast cell mediators (histamine’s breakdown product, a prostaglandin metabolite, a leukotriene) can add information before concluding “not mast cells” (Maeda et al. 2017).
The takeaway — with the honest limit that keeps this from being unfalsifiable. A normal blood test does not rule out a histamine/mast-cell problem, and a single failed drug does not prove you don’t have one. Both are clues, not verdicts. But that cuts both ways, and it must not become a trap where no possible result counts against mast cells. Concretely: if you have repeated negative results across several well-run trials and (where obtainable) a urinary mediator panel showing no elevation in any mast-cell product across multiple flares, then mast-cell involvement becomes genuinely unlikely for you, and the honest move is to stop pursuing that pathway and move to the next hypothesis. And that stop-condition must not depend on a test you can’t get: if mediator panels aren’t available to you, then a sequence of well-run trials — adequate dose, adequate length, defined stop-rules, no response — is sufficient on its own to move the probability down and call time on this pathway. Without naming a stop-condition like one of these, the framework would be unfalsifiable — and the intent here is the opposite: these explanations exist to keep you from over-concluding from one test or one drug, not to make the idea immune to evidence.
3 What it means if the medication does not work
This is where I want to be most honest, because people either feel failed or have been told (often wrongly) that a failed trial means it’s psychological. The fair version of the skeptical position is worth holding onto and is not dismissive: your symptoms are real — the question is whether the label MCAS is the right one and whether mast cells are really your lever. A failed trial is evidence about the label and the lever, not about whether you suffer.
One: it may not have been the right drug or the right dose. H1, H2, stabilisers, leukotriene blockers, and anti-IgE drugs target different parts of the mast-cell cascade, and they fail in different people (Molderings et al. 2016). Trying one and stopping forever is not a real negative test.
Two: a negative trial after one drug, even when the diagnosis is right. The ketotifen-in-fibromyalgia trial is a hard lesson: a plausible mast-cell drug, in a related condition, showed no benefit over placebo (Moldofsky et al. 2015). Two fair readings follow, and both belong here. The charitable one: the intervention was not effective enough to help the group. The more parsimonious one: mast-cell involvement in that syndrome may be weaker or rarer than its advocates assume. Either way, the trial says nothing about the patients being dishonest — drugs fail without the patients failing.
Three: the “ratchet” — a hypothesis that longer illness may mean less response. The primary document develops a specific and sobering possibility: each cycle of mast-cell activation and autonomic/barrier disturbance may leave behind a little more “priming” — extra nerve endings, more mast cells, more sensitive receptors. Past a certain point, the process might keep itself going even if the original trigger is gone. If that “amplification ratchet” is real, it would predict a weaker response to mast-cell treatment in people ill for many years (Loth 2026). Two cautions. First, it is a hypothesis, and there are simpler, equally plausible readings of the same “less response in long illness” pattern — long-ill people may simply be a subgroup in which mast cells were never the driver, or have accumulated so much else that a mast-cell win is diluted. Second, do not let it become a tool for converting failure into urgency: whether treating mast cells earlier actually helps is untested, and an early non-response is better read neutrally until data exist (Rohrhofer et al. 2025).
Four — the honest core, stated precisely. A single failed antihistamine trial does not, by itself, disprove mast-cell involvement, because the drug blocks histamine at its receptors and does not stop mast cells releasing other mediators — prostaglandins, leukotrienes, tryptase — that act outside histamine signalling (Molderings et al. 2016). And, separately, no mast-cell treatment result disproves ME/CFS, which is far larger than any one pathway. Keep those two claims distinct: the first is “one drug isn’t a complete test,” the second is “your illness is bigger than MCAS.” Distinguish them from a stronger claim nobody here should make — that repeated negative trials and panels say nothing at all. They do. See the takeaway above for where negative evidence should move the answer.
Five: ME/CFS is bigger than MCAS — and know when a failure actually is informative. Even in the best case, mast cells are one thread in a very tangled knot of energy, immune, and nervous-system dysfunction (Castells et al. 2024). Because in ME/CFS the mast-cell problem is usually later-arriving and possibly downstream, a mast-cell treatment failure says nothing against the underlying disease (see the overview). But resolve the frequent confusion honestly: a failure is non-informative about mast cells only within a properly conducted trial (right drug class, adequate dose, enough time). A failure after a single drug at an inadequate dose, or a half-kept trial, is simply a trial that didn’t happen. By contrast, a well-run sequence — adequate dose, adequate trial length, clear stop-rule, no response — genuinely lowers the probability the problem is MCAS-mediated, and that is exactly the signal you take back to decide whether to keep investing in this pathway.
Six: side effects, misdiagnosis — and giving the negative trial its proper weight. Antihistamines are not harmless. Some cause marked sedation, dry mouth, or worse. And because “MCAS” symptoms overlap with many other conditions — thyroid problems, dysautonomia, food intolerance, and more — a treatment failure is a useful moment to revisit the differential. On evidence weight: the strongest relevant data point is a negative randomized trial — a 1996 double-blind study of an H1 antihistamine in chronic fatigue syndrome found no benefit at all, and no MCAS-specific RCT since has overturned it (Steinberg et al. 1996). Read that honestly: it does not prove antihistamines can never help a mast-cell subset, but it is genuine evidence that the expected success rate of a first antihistamine trial in ME/CFS is soberingly low. In plain terms: of all the studies in this area, it is the one with the most controlled design, and its result deserves to weigh heavily on expectations.
4 The falsifiable predictions
For those who find the framework useful, here is what would confirm or refute the mast-cell story:
- If mast-cell stabilisers (cromolyn, ketotifen) reduce symptoms independently of allergen avoidance, that would be consistent with — though not proof of — non-allergen (non-IgE) activation being real; in an uncontrolled setting, the same observation could come from placebo or natural fluctuation, so it supports but does not settle the point (Roy et al. 2021). This must be read two-way: the negative ketotifen-in-fibromyalgia trial (Moldofsky et al. 2015) is exactly the result a weak-or-absent mast-cell contribution would predict, so a positive MCAS-selected result is needed because a null result already exists — not despite it.
- If antihistamines improve flushing and orthostatic symptoms but not core fatigue, that supports histamine contributing to some features while the core energy failure is elsewhere — with the caveat that the same selective pattern could arise from placebo/regression, so it is suggestive, not proof.
- If a true randomized, controlled trial of a mast-cell–targeted protocol (H1 + H2 + stabiliser) shows benefit in ME/CFS patients selected for mast-cell symptoms, that would move this from hopeful case series to genuine evidence. So far, the trials do not exist. That is the single biggest gap in the whole area (Nurmatov et al. 2015).
- Individual-level falsifier (because “subset” claims must stay open to disproof): a patient who fails several well-run drug classes, plus — where panels are obtainable — a full urinary mediator panel showing no elevation in any mast-cell product across multiple flares, is genuine evidence against mast cells being that person’s amplifier. Where panels aren’t obtainable, the sequence of adequate, well-run failed trials is enough on its own to do the same job. Either way, the framework should say it clearly rather than adding more escape hatches — this is the discipline that keeps the “subset” idea testable.
- If “flavour” matters (an untested hypothesis from the primary document, not an established finding): pre-treatment urinary mediator ratios (histamine breakdown vs. a prostaglandin-D₂ metabolite vs. a leukotriene) predict which symptom improves first with which drug [Maeda et al. (2017)](Loth 2026).
- If the ratchet is real: treatment response should drop measurably with each additional 5 years of illness duration — stronger than the effect of baseline blood markers. If response is identical regardless of illness duration, the ratchet idea loses support (Loth 2026).
5 What you can actually do
This is not medical advice; it is a map of the conversation to have with a clinician.
- Consider a symptom and trigger note — kept light. A timed multi-column diary is not realistic (and not safe) for a moderate-severe ME/CFS patient: the effort itself can trigger PEM. Use the minimal version — a single daily rating line, a photo of any flushing/hives, and a short “what time” note on bad days — or let a caregiver/proxy keep it. The diary is optional scaffolding for the conversation, not a requirement for treatment.
- Ask the honest question, and know what a structured trial can and can’t do. Your clinician should help you decide whether to pursue mast-cell treatment at all — via symptom patterns and, where available, mediator testing — because “just try antihistamines forever” is not a diagnosis [Molderings et al. (2011)](Valent et al. 2021). But be realistic about access: urinary mast-cell mediator panels are available mainly at specialist/research centres, often need 24-hour refrigerated collection and medication washout, and may be unaffordable or unobtainable in general practice. Where testing isn’t available, a time-limited, endpoint-defined drug trial is acceptable practice — that is an honest limitation, not a cheat. One clarification on the word “diagnostic”: a well-run open trial is a reasonable clinical guide to whether to keep pursuing mast-cell treatment, but because placebo response and natural fluctuation are uncontrolled in an individual, a positive result supports — it does not confirm — the diagnosis; that confirmation would need formal testing (Comas-Basté et al. 2020).
- Make the trial concrete. One drug family at a time, a named target symptom, a named dose, and a named reassessment point (e.g., reassess flushing at 6 weeks, stop if no improvement). More isn’t better, sedation is a real cost, and “weeks to months” needs a defined stop-rule or people stay on ineffective sedating treatment indefinitely.
- If diet is in play, make it time-limited and safe. A low-histamine diet can be informative, but in ME/CFS it risks PEM from shopping/cooking and malnutrition from over-restriction. Frame it as a 3–4 week trial, involve a dietitian if available, and stop restricting if weight falls or nothing changes in that window [Comas-Basté et al. (2020)](Weinstock et al. 2024).
- Manage expectations. Ask which symptom the drug is meant to lighten. Helping the flushing but not the crash is information, not failure. And if you’ve been ill for years, be aware response may be less dramatic — but do not let a hypothesis about long illness become a reason to write off treatment or be written off yourself; the evidence base is simply not settled enough for that (Loth 2026).
- Take a well-run failure as a real finding and bring it back to the team — it points you away from the mast-cell path and toward the next hypothesis. It does not mean you’re not sick, and it is not a verdict on your illness.
6 The bottom line
A normal blood test doesn’t rule out a histamine/mast-cell problem, and a single failed antihistamine trial doesn’t prove you don’t have one. The mast-cell story has real but thin evidence in ME/CFS, and the drugs are symptom-lighteners not cures. A well-run failure — adequate dose, adequate length, clear stop-rule, no response — does tell you more than “the drugs don’t work”: it makes a mast-cell diagnosis and a histamine lever progressively less likely, which is exactly the information you bring back to next steps. That is not a verdict on you, and it is not a verdict on your illness. It’s the next step.
For the comprehensive, fully-cited picture of how mast cells are weighed among the many candidate mechanisms in ME/CFS, see (Loth 2026).