Improving the Blood–Brain Barrier and Cognitive/Mitochondrial Function in Alzheimer’s, and some Autism: Linking TNAP, GPLD1, Vitamin B6 and Exercise

 

 

Scientists Find a Mechanism for How Exercise Protects the Brain

UCSF study finds that an exercise-induced liver protein strengthens the blood-brain barrier, improving memory and slowing age-related decline.

Researchers at UC San Francisco have discovered a mechanism that could explain how exercise improves cognition by shoring up the brain’s protective barrier of blood vessels.

With age, this network of blood vessels — called the blood-brain barrier — gets leaky, letting harmful compounds enter the brain. This causes inflammation, which is associated with cognitive decline and is seen in conditions like Alzheimer’s disease.

Six years ago, the team identified a brain-rejuvenating enzyme called GPLD1 that mice produced in their livers when they exercised. But they couldn’t understand how it worked, because it can’t get into the brain.

The new study reveals that GPLD1 works through another protein called TNAP. As the mice age, the cells that form the blood-brain barrier accumulate TNAP, which makes it leaky. But when mice exercise, their livers produce GPLD1. It travels to the vessels that surround the brain and trims TNAP off the cells.

“This discovery shows just how relevant the body is for understanding how the brain declines with age,” said Saul Villeda, PhD, associate director of the UCSF Bakar Aging Research Institute.

 

Every few months Alzheimer’s research produces another “breakthrough.” Most focus narrowly on the brain — amyloid, tau, synapses.

Recent Alzheimer’s drugs, like lecanemab and donanemab represent a scientific advance, but their real-world impact remains modest. They cost about $30,000 a year, require intensive monitoring, and typically slow decline by only a few months.

A growing body of research is pointing somewhere else entirely.

Not just the brain, but the interface between the body and the brain.

At the center of this shift are three players:

• TNAP (tissue-nonspecific alkaline phosphatase)
• GPLD1 (an exercise-induced blood protein)
• Vitamin B6 (PLP)

Together, they connect:

• the blood–brain barrier (BBB)
• neurotransmitters
• mitochondrial function
• inflammation

This same network appears not only in Alzheimer’s disease, but also in subsets of autism.

 

The Blood–Brain Barrier: The Overlooked Gatekeeper

The blood–brain barrier is not just a passive wall. It is an active, living system that determines what reaches the brain.

When functioning properly, it:

• keeps out inflammatory molecules
• regulates nutrient delivery
• protects neurons from toxins

With age — and in many neurological conditions — this barrier begins to fail.

It becomes leaky.

This allows:

• cytokines
• immune cells
• metabolic toxins

to enter the brain.

The result is chronic low-grade inflammation, one of the key drivers of cognitive decline.

 

TNAP: A Double-Edged Enzyme

TNAP sits at a critical junction.

Inside the brain:

TNAP helps regulate vitamin B6 availability, which is essential for:

• GABA (the calming neurotransmitter, but excitatory in 30% of severere  autism)
• dopamine
• serotonin

Without sufficient active B6 (PLP), neurons become more excitable and unstable.

 

At the blood–brain barrier:

TNAP plays a different role.

With aging, TNAP accumulates in the BBB, where it begins to:

• weaken barrier integrity
• increase permeability
• promote inflammation entering the brain

So TNAP is both

• necessary for neurotransmitters
• but potentially harmful in excess at the BBB

This dual role is key to understanding the system.

 

GPLD1: The Exercise Signal

Recent research from the University of California in San Francisco has identified a protein called GPLD1, released into the bloodstream during exercise.

Its function is remarkable.

GPLD1 appears to:

• remove excess TNAP from the blood–brain barrier
• restore barrier integrity
• reduce inflammation entering the brain

In animal models this led to:

• improved cognition
• reduced amyloid pathology
• better overall brain function

This is one of the clearest mechanisms yet showing how exercise protects the brain.

 

Vitamin B6: The Neurochemical Link

Vitamin B6 (in its active form, PLP) sits downstream of TNAP.

It is essential for:

• converting glutamate → GABA
• stabilizing neuronal firing
• supporting mitochondrial enzymes

In some individuals — including subsets of autism — B6 metabolism appears to be impaired.

This can lead to:

• low GABA
• excess excitation
• sensory sensitivity
• tics or seizures

Correcting B6 availability can sometimes produce significant functional improvements.

 

Mitochondria: The Energy Perspective

All of this sits on top of a deeper requirement: energy

Neurons are extremely energy-dependent.

If mitochondrial function is impaired:

• ion gradients fail
• signaling becomes unstable
• excitability increases

Both Alzheimer’s disease and autism frequently show signs of:

• mitochondrial dysfunction
• impaired energy metabolism

Vitamin B6 supports mitochondrial enzymes.

Exercise increases mitochondrial number and efficiency.

Again, the same network appears.

Exercise is not just “burning calories.”

It is activating PGC-1α, the master regulator of mitochondrial production, effectively increasing the brain’s energy-generating capacity.

A brain with more mitochondria is more stable, more resilient, and less vulnerable to both degeneration and developmental disruption.

  

Why This Matters for Autism

At first glance, Alzheimer’s and autism may seem unrelated.

But both conditions often involve:

• neuroinflammation
• mitochondrial dysfunction
• synaptic instability
• blood–brain barrier disruption

The difference is timing:

• Alzheimer’s → degeneration of an aging system
• Autism → altered development of the system

Understanding one can illuminate the other.

If BBB dysfunction drives inflammation in Alzheimer’s, it may also contribute to instability in developing brains.

If mitochondrial support improves cognition in aging, it may improve resilience in autism.

 

Exercise: The Overlooked Multi-System Therapy

Exercise is unique because it affects all parts of this network simultaneously.

• increases GPLD1 → strengthens the BBB
• increases BDNF → improves synaptic plasticity
• improves mitochondrial function
• reduces inflammation
• enhances brain blood flow

It is not a single-target intervention.

It is a system-wide regulator.

Many autism interventions (e.g. Pentoxifylline, Agmatine and even beetroot juice) converge on improving cerebral blood flow.

Better blood flow → more oxygen and glucose delivered to the brain.

This supports mitochondrial ATP production, improving brain energy and stability.

Exercise complements this by increasing mitochondrial number via PGC-1α and strengthening the BBB (GPLD1/TNAP).

Together, these interventions enhance neurovascular–metabolic function, leading to more stable cognition and behavior.

 

A Unifying Model

We can now sketch a simple framework:

• TNAP → Vitamin B6 → neurotransmitter balance (GABA)
• Excess TNAP (BBB) → barrier breakdown → inflammation
• Exercise → GPLD1 → removes excess TNAP → restores BBB
• B6 + exercise → support mitochondria and brain stability

This links:

vascular function + metabolism + neurotransmitters + inflammation

into a single system.

 

The Bigger Insight

For years, Alzheimer’s research has tried to isolate single causes:

• one gene
• one protein
• one drug target

But the brain does not work that way.

It is a network.

TNAP is not “the cause.”
GPLD1 is not “the cure.”

They are control points in a larger system.

Conclusion

This emerging biology suggests that:

• protecting the blood–brain barrier
• supporting vitamin B6 metabolism
• improving mitochondrial function
• and maintaining regular physical activity

may all be part of the same therapeutic strategy.

Not just for Alzheimer’s disease, but for understanding — and in some cases improving — aspects of autism.

The most sophisticated and expensive interventions may still lie in the future, but one of the most powerful has been available all along.

Exercise is not just good for the body. It is a direct regulator of brain biology.

  

A Final Thought: The Brain Is Only as Protected as Its Barriers

One of the more surprising directions in Alzheimer’s research is not a new drug or gene, but a shift in perspective.

The brain is not as isolated as we once thought.

It is protected by multiple biological barriers — and when these begin to fail, risk increases.

We have already looked at the blood–brain barrier, but this is not the only route.

There is also a direct pathway from the nose to the brain via the olfactory nerve — effectively bypassing the blood–brain barrier altogether. Animal studies have shown that certain bacteria can use this route, especially when the nasal lining is damaged, triggering immune responses in the brain that resemble early Alzheimer’s pathology.

(Note to self, don’t pick your nose!)

The gut can influence the brain through immune signaling and inflammation, particularly when the intestinal barrier is compromised.

Individually, these findings may seem unrelated — blood vessels, nasal tissue, gut bacteria.

But they point to the same underlying principle:

The brain depends on the integrity of the body’s protective barriers.

When those barriers are strong:

• inflammatory signals are controlled
• harmful agents are excluded
• neuronal function remains stable

When they weaken:

• the brain becomes exposed
• immune responses increase
• long-term damage may follow

This brings us back to the central theme of this article.

Exercise is not just improving fitness — it is helping to restore control over these systems:

• strengthening the blood–brain barrier (via GPLD1)
• reducing systemic inflammation
• improving metabolic function
• supporting mitochondrial health

In other words, it helps the body maintain the boundaries that protect the brain.

The emerging biology — TNAP, GPLD1, vitamin B6, mitochondria — is complex.

 

Oral bacteria and its link to brain function

Alzheimer’s and Parkinson’s research has also looked at the effect of the oral microbiome.

Tooth decay and gum disease are not just local problems — they influence whole-body inflammation.

·        Harmful oral bacteria (e.g. Porphyromonas gingivalis) increase with poor oral hygiene.

·        These bacteria can enter the bloodstream, especially when gums bleed.

·        This can contribute to systemic inflammation and stress the brain.

·        Inflammation may weaken the blood–brain barrier (BBB).

·        A weaker BBB allows more harmful molecules to reach the brain.

·        This links oral health to cognitive decline and dementia risk.

·        At the same time, some oral bacteria are highly beneficial.

o   These bacteria convert dietary nitrates into nitric oxide (NO).

o   Nitric oxide improves cerebral blood flow and brain function.

o   Overuse of strong antiseptic mouthwash can reduce these beneficial bacteria.

o   The goal is balance, not complete sterilization of the mouth.

·        Good oral hygiene reduces harmful bacteria without eliminating beneficial ones.

·        Healthy gums act as a barrier, preventing bacterial entry into blood.

·        Diet plays a major role in shaping the oral microbiome.

·        High sugar promotes tooth decay and harmful bacteria.

·        Nitrate-rich foods (e.g. vegetables, beetroot) support beneficial bacteria.

·        Maintaining teeth and gums is therefore part of protecting long-term brain health.

 

 

Dihexa, Telmisartan (Candesartan, Losartan), PEPITEM, Cognitive Enhancement and the example of Pitt-Hopkins

 

A reader recently left an interesting comment on my earlier post about telmisartan. They wrote that they had been using Dihexa for a couple of months and had noticed new vocalizations and unexpected progress with toilet training in their child. They also mentioned another peptide, PEPITEM, which they had come across while reading about bone metabolism and inflammation.

This comment prompted me to look more closely at how several topics might intersect biologically: Dihexa, angiotensin receptor blockers such as telmisartan, the peptide PEPITEM, and conditions like Pitt-Hopkins syndrome.

 

What is a peptide?

Peptides are short chains of amino acids, the same building blocks that make up proteins. They act as signaling molecules in the body and regulate many biological processes. Examples of natural peptide hormones include insulin and oxytocin.

Scientists often design synthetic peptides to mimic or modify these natural signals. Some peptides have become successful medicines. A well-known example is semaglutide, used to treat diabetes and obesity.

In recent years peptides have become very popular in longevity and biohacking circles. This is partly because modern biology has discovered many new peptide signaling systems. These regulate metabolism, immune responses, tissue repair, and brain plasticity.

Another reason is that peptide manufacturing has become much cheaper. Automated peptide synthesis now allows laboratories to produce peptides easily. As a result, some research peptides are now sold online.

Many of these compounds are marketed as “research chemicals”. Examples often discussed online include BPC-157 and Dihexa. These compounds originated in laboratory research.

However, most of them have not gone through proper human clinical trials. Their long-term safety and effectiveness are therefore unknown.

Social media and podcasts have amplified interest in these substances. This has created a large grey market for experimental peptide therapies.

Scientists remain cautious because peptides can have strong biological effects. Problems can include uncertain purity, incorrect dosing, and lack of safety data.

Despite these concerns, peptides remain an important area of medical research. Many future medicines are likely to be peptide-based.

The reason is simple, biology uses peptides as a major language of cellular communication. They control processes ranging from metabolism to immune function and brain signaling.

This is why peptides sometimes appear in discussions of neurological conditions.

Many pathways involved in brain development and synaptic plasticity are regulated by peptide signals.

  

Dihexa and the angiotensin system

Dihexa was originally developed from angiotensin IV, a fragment of angiotensin II that belongs to the renin–angiotensin system. While this system is best known for regulating blood pressure, it also has important roles in the brain.

In the 1990s researchers noticed that angiotensin IV could improve learning and memory in animal experiments. This led scientists to design molecules that could mimic these effects but cross the blood–brain barrier and remain stable in the body. One of these molecules was Dihexa.

Interestingly, Dihexa does not appear to work primarily through classical angiotensin receptors. Instead it activates the HGF/MET pathway, which regulates neuronal growth, dendritic branching and synapse formation. In laboratory experiments Dihexa has shown very strong synaptogenic effects, meaning it can promote the formation of new synaptic connections between neurons.

This is why it sometimes appears in discussions of cognitive enhancement or experimental neurological treatments. However, it is important to stress that Dihexa has never undergone proper human clinical trials, so its safety profile and long-term effects remain unknown. It is somewhat surprising that it is sold online as a “supplement” or research compound, since it is really a laboratory-designed molecule rather than a traditional dietary supplement.

 

BDNF and synaptic plasticity

Some autism clinicians have experimented with approaches intended to increase brain levels of BDNF (brain-derived neurotrophic factor), a key regulator of synaptic plasticity and neuronal survival.

BDNF promotes dendritic growth, synapse formation and learning-related plasticity. In many ways it is one of the brain’s central “growth signals”. Dihexa became famous in neuroscience circles partly because some laboratory studies suggested it could stimulate synapse formation even more strongly than BDNF, although those findings were mainly from cell culture experiments.

The two pathways are different but converge on similar intracellular signaling networks that regulate synaptic growth.

Interestingly, one of the most reliable ways to increase BDNF is not a drug at all but physical exercise. Exercise stimulates BDNF production in the hippocampus through a combination of increased neuronal activity, metabolic signaling and muscle-derived molecules such as irisin. High-impact activity may also stimulate endocrine signals from bone, including osteocalcin, which can influence brain function.

 

Angiotensin receptor blockers and the brain

The drugs discussed in some of my previous articles—telmisartan, candesartan and losartan—belong to the class of angiotensin receptor blockers (ARBs). They block the AT1 receptor, which is activated by angiotensin II.

Although these drugs are prescribed for hypertension, the brain has its own local renin–angiotensin system. In the central nervous system angiotensin signaling influences neuroinflammation, oxidative stress, cerebral blood flow and neuronal excitability.

Blocking the AT1 receptor tends to reduce inflammatory signaling and shift the balance toward protective pathways.

Telmisartan is particularly interesting because it also activates the nuclear receptor PPAR-gamma, which influences mitochondrial function, metabolic signaling and inflammation in neurons.

Candesartan is often considered one of the more brain-penetrant ARBs and has shown neuroprotective effects in some experimental models.

Losartan has attracted attention because it can reduce excessive TGF-beta signaling, a pathway involved in inflammation and tissue remodeling.

Telmisartan might theoretically be more relevant in autism where metabolic stress and inflammation dominate (because of PPAR-γ activation). Losartan might be more relevant where excessive tissue-remodeling or TGF-β signaling plays a role. In the brain, tissue remodeling involves:

• synapse formation and elimination
• growth of dendrites and axons
• restructuring of the extracellular matrix around neurons
• activation of glial cells

Losartan is used to treat Marfan syndrome. Marfan syndrome is a systemic connective-tissue disorder that affects many parts of the body, particularly the heart.

Some studies have reported altered TGF-β signaling in certain forms of autism, suggesting that immune and tissue-remodeling pathways may contribute to aspects of neurodevelopment in at least some individuals. Losartan could theoretically influence these biological processes.

These mechanisms do not directly overlap with Dihexa’s synapse-forming activity, but they may influence the overall biological environment in the brain by reducing inflammatory and metabolic stress.

 

The peptide PEPITEM

The reader also mentioned PEPITEM, short for “PEPtide Inhibitor of Trans-Endothelial Migration”.

PEPITEM regulates the movement of immune cells across blood vessel walls. In simple terms, it helps control whether inflammatory immune cells leave the bloodstream and enter tissues.

This pathway has been studied mainly in inflammatory diseases. By limiting immune-cell migration, the PEPITEM pathway can reduce tissue inflammation.

Interestingly, the same pathway also influences bone metabolism because immune signaling strongly affects osteoclast activity and bone resorption.

 

Why bone biology keeps appearing

One surprising theme linking these topics is the intersection between inflammation, bone metabolism and the renin–angiotensin system.

Angiotensin II can stimulate osteoclast activity and promote bone resorption. Blocking the AT1 receptor with ARBs may therefore modestly reduce inflammatory bone loss. Some observational studies have suggested that ARB use may be associated with slightly higher bone density or lower fracture risk.

Given how closely immune signaling and bone metabolism interact, it is not surprising that peptides affecting immune-cell trafficking, like PEPITEM, also influence bone remodeling pathways.

 

Pitt-Hopkins syndrome as an example

Pitt-Hopkins syndrome is caused by mutations in the transcription factor TCF4. This gene regulates many downstream processes involved in neuronal development, synaptic maturation and network formation.

In experimental models of Pitt-Hopkins and related neurodevelopmental disorders, researchers often observe abnormalities in synaptic development and neuronal connectivity.

Because of this, some therapeutic ideas have focused on pathways that influence synaptic plasticity, neuronal growth or inflammatory signaling.

The HGF/MET pathway activated by Dihexa is one such pathway. The MET gene has also been linked to autism genetics in several studies, and reduced MET signaling has been associated with altered cortical connectivity.

This does not mean that Dihexa is a treatment for Pitt-Hopkins syndrome or autism, but it is certainly plausible.

We saw in previous posts that autism can be broadly divided in hypo/hyper (too little/much) active pro-growth signaling pathways. Pitt Hopkins would be in the hypo category, so increasing activity should be beneficial.

The unknown issue with Dihexa is that it has not be tested thoroughly in humans, so long term use might not be wise, particularly in older people.

The totally safe way to increase pro-growth signaling is via daily aerobic exercise, which comes up again in the next post, which looks at translating recent Alzheimer's research to autism. 

 

A broader pattern

What the reader’s comment illustrates is something that appears frequently in biomedical research, apparently unrelated compounds often converge on a small number of biological control systems.

In this case we see several different layers of regulation:

– the renin–angiotensin system influencing inflammation and metabolic signaling
– growth factor pathways such as HGF/MET and BDNF regulating synapse formation
– immune trafficking pathways such as PEPITEM controlling inflammatory cell migration
– transcriptional regulators such as TCF4 governing neuronal development

Each operates at a different level, but they all ultimately influence how neurons grow, connect and function.

This does not mean that compounds like Dihexa or peptides such as PEPITEM will become treatments for neurological conditions. Most remain at a very early stage of research.

But it does highlight how discoveries in cardiovascular biology, immunology, bone metabolism and neuroscience increasingly intersect.

 

Conclusion

Dihexa and telmisartan start from the same hormonal system but act very differently:

• Dihexa directly stimulates synapse formation through growth-factor signaling.
• Telmisartan reduces inflammation and metabolic stress that may impair neuronal function.

The overlap lies mainly in their potential downstream effects on neuronal plasticity, not in their primary mechanism of action.

In the case of Pitt Hopkins syndrome both might be potentially beneficial, although through very different mechanisms, but no clinical evidence exists.

Dihexa acts by activating the HGF/MET pathway, which promotes synapse formation, dendritic growth and neuronal plasticity. Since Pitt-Hopkins syndrome involves impaired neuronal network development caused by mutations in the TCF4 transcription factor, pathways that enhance synaptic growth should attract scientific interest.

Telmisartan works in a different way. By blocking the AT1 receptor of the renin–angiotensin system it reduces inflammatory signaling and oxidative stress, and it also activates the nuclear receptor PPAR-γ, which influences mitochondrial metabolism and cellular stress responses. These effects could potentially improve the cellular environment in which neurons function.

In simple terms, Dihexa attempts to directly stimulate synapse formation, whereas telmisartan may reduce biological stresses that interfere with normal neuronal signaling.

Both approaches therefore touch on biological processes that are relevant to brain development and plasticity.

Dihexa is used by some autism clinicians in the US.

 

Up to 40% of children in the “failed” phase 3 bumetanide trial were actually responders, according to AI reanalysis of the data – Treating autism in the real world

 

In some parts of the world even the words “treating autism” can still get you into trouble and some people have to go to quite extreme lengths to get their child’s developmental trajectory back on track.

I did note that in the US big changes have been made to their Interagency Autism Coordinating Committee (IACC) that coordinates all efforts within the Department of Health and Human Services (HHS) concerning autism. Now it includes some readers of this blog. Will this make a difference?

https://iacc.hhs.gov/ 

 

Over in France, the Bumetanide researchers Ben-Ari, Lemonnier and pals published their AI driven reanalysis of the “failed” phase 3 autism trial. They found that using AI they could actually predict who did actually respond; and many did. Nonetheless this large trial of all-jumbled-together kids with an autism diagnosis showed that overall bumetanide was no better than a placebo. Sounds strange to you? This is a common theme in autism trials because they do not narrow down a specific type of autism that they are trying to treat.

Over where I am, I keep getting positive reports of success. Some people are lucky and find that much of what works for my son works for theirs. There is a lot in this blog about other types of autism.

Why autism remains untreatable?

Autism is not simple to treat. Autism has no biological definition and measurement scales are all likely not fit for purpose. What would treatment success even mean?

From the perspective of severe autism with apparent ID (the old “Classic autism”) the biggest issues are to do with the slow rate of acquiring new skills. There are very well established tools to measure the skillset of such kids, such as  ABLLS (Assessment of Basic Language and Learning Skills). There are also non-verbal IQ tests. 

For young kids with classic autism you want them to add these basic skills ASAP, so that they can move on with their lives. In our case Bumetanide was the key to unlock new skill addition.

This is not what the phase 3 bumetanide trial was trying to measure.

Indeed one of the recurring comments from parents and teachers is the child has become more “present.” How do you quantify something like that?

For most children with autism in 2026, they do not have a problem with skill acquisition, they are a bit quirky, nervous, resistant to change, stim a bit, do not make friends. It is a very different condition. These issues are very real and genuinely concern some parents, but they are very different problems.

The modern cookie-cutter, protocol-driven, approach does work for most of medicine. But it will never work on an ill-defined category like autism. It actually becomes ridiculous when you look at all the varied types of autism. Even people with cerebral palsy or Down syndrome can be given an “autism” diagnosis on top, but they are completely different biological conditions.

Where to from here?

What does Ben-Ari do now?

Start again with another phase 3 trial? Paid for by who?  Will Servier come back and fund the second attempt?

In the meantime the clock keeps ticking.

I read Ben-Ari’s initial study and made my n=1 trial in 2012. My trial met its primary endpoint (Peter satisfied) and therapy started.

Academic performance went from complete basket-case to passing his high school public IGCSE exams a decade later.

Now it is 2026 and therapy still continues. No side effects,  heart ultrasound (echocardiogram) all normal.

Crazy world.

40% “disabled” at Stanford

I was surprised to read that almost 40% of undergraduates at Stanford University are claiming disability, to get extra time in exams. It does tell you a lot about the current generation of 20 year olds.

I would give them an E on their final diploma (I passed but needed Extra time). It is perfectly reasonable for a small number of clever students to need extra time, they might have a physical disability with their hands, be deaf, or blind, or dyslexic. It is perfectly reasonable to give some people extra time, but 40%?

It really is not fair on the remaining 60%. Maybe just give everyone an extra hour, those that finish early just leave early. They could get E on their results, for “I work fast and finish Early - hire me!"

What is annoying is the trivialization of the word disability.

40% of Stanford undergrads receive disability accommodations—but it’s become a college-wide phenomenon as Gen Z try to succeed in the current climate

So many people claim a disability like autism that theme parks in the US and Europe have had to roll back their privileged access schemes.

When I visited Charlotte International airport a while back and had to stand in a very long line for the passport control, I was amazed to see a never-ending procession of people appearing in wheelchairs to skip the queue. I have never seen this in Europe, but I suppose it will eventually come.

 

Back to those 40% in the Bumetanide trial.

New Analysis of the Bumetanide Phase 3 Trials: Were Responders Hidden in a “Failed” Study?

Approximately one-quarter to one-third of participants fit validated clinical profiles in which bumetanide showed statistically significant benefit on SRS-2, despite the overall trial being negative. The abstract itself says up to 40%.

Treating autism with Bumetanide: Identification of responders using Q-Finder machine learning algorithm

Bumetanide, a specific NKCC1 co-transporter inhibitor, restores deficient GABAergic inhibition implicated in various brain disorders, including Autism Spectrum Disorders (ASD). In keeping with this mechanism, nine successful phase 2 clinical trials, conducted by seven independent teams using an identical protocol, have shown significant improvements in ASD symptoms among individuals treated with Bumetanide. Despite these promising results, two large phase 3 clinical trials (over 400 children recruited in approximately 50 centers and covering age groups 2–6 and 7–17 years) failed with no significant difference between patients treated by placebo or Bumetanide. This failure may stem from the substantial heterogeneity of ASD symptom profiles across the study population, potentially diluting the overall observed treatment effect. To address this, we reanalyzed the phase 3 data using Q-Finder, a supervised machine learning algorithm, aiming to identify subgroups of patients who responded to the treatment. This analysis was based on clinical parameters collected at the baseline of trial and used the same standard endpoints and success criteria defined in the original phase 3 protocol. It enabled the identification of responder subgroups showing a statistically significant difference between placebo and Bumetanide treatment arms. We report detailed descriptions and statistical evaluations of these subgroups. The discovered responder subgroups, representing up to 40% of participants, were cross validated between the two study populations. These findings suggest that meaningful treatment responses can be uncovered within negative phase 3 trials, highlighting the limitations of a one-size-fits-all approach for heterogeneous conditions such as ASD. Machine learning appears to be a promising tool to support this precision medicine strategy.

The 2026 reanalysis published in Translational Psychiatry revisited the large Phase 3 bumetanide trials that previously failed to meet their primary endpoint.

The original Phase 3 trials included more than 400 children (ages 2–17) and found no significant overall difference between bumetanide and placebo on the primary outcome measure (CARS2).

This new study asked a different question:

Instead of “Did bumetanide work for everyone?”, could it have worked for specific subgroups that were diluted in the overall average?

To explore this, the authors used a supervised machine-learning algorithm (Q-Finder) to identify baseline clinical profiles associated with treatment response.

What They Found

The original overall result remains negative

Across the entire population:

• No significant benefit on the primary endpoint (CARS2).
• No meaningful average effect.

So the trial still officially failed.

Subgroups showing benefit were identified

When the data were stratified by symptom profiles at baseline, several subgroups showed:

• Statistically significant improvement on the SRS-2 (Social Responsiveness Scale)
• Treatment effects of roughly 12–17 points in validated groups
• Coverage of about 25–36% of participants in the largest responder profiles

Importantly, these findings were cross-validated between the younger and older trial cohorts.

A Consistent Feature of Responders

Across validated subgroups, one feature repeatedly appeared:

Mildly abnormal “adaptation to environmental changes” on CARS2

This domain reflects:

• Difficulty with transitions
• Rigidity around routines
• Stress with change

Responders were typically:

• Clearly autistic (often moderate–severe social symptoms)
• With repetitive behaviours
• But not globally or profoundly impaired across all domains

Interestingly, IQ did not emerge as a defining predictor of response.

Primary Endpoint vs Secondary Endpoint

A key nuance:

• No validated responder subgroups were found using the primary endpoint (CARS2).
• Validated subgroups were found using the secondary endpoint (SRS-2).

From a regulatory standpoint, this matters: trials are judged on their primary endpoint.

From a scientific standpoint, it suggests:

SRS-2 may have been more sensitive to the type of change bumetanide produces.

What This Means

This reanalysis does not prove bumetanide works broadly in autism.

It does suggest:

• Autism is highly heterogeneous.
• A one-size-fits-all trial design may dilute effects.
• A biologically or symptom-stratified approach may be necessary.
• Around one-quarter to one-third of participants may represent a responder subtype.

However, these findings are post hoc and exploratory.

To confirm them, a new trial would need to:

• Prospectively enroll only the identified responder phenotype.
• Use appropriate primary endpoints.
• Replicate the treatment effect.

Why This Matters for Autism Research

The study reflects a broader shift toward precision medicine:

• Rather than asking “Does this drug work for autism?”
• The better question may be:

“Which subtype of autism does it work for?”

Machine learning may help identify these subgroups, but prospective validation is essential.

The original Phase 3 trial remains negative at the population level.

This reanalysis suggests that meaningful responses may have been present in specific clinical subgroups — particularly children with:

• Mild adaptation abnormalities
• Repetitive behaviours
• Significant social impairment

Whether this represents a reproducible biological subtype remains to be tested in future trials.

Conclusion

In Rett syndrome a very expensive new drug called Trofinitide was approved, even though reports suggest it is only really effective in about 20% of these girls. I was really surprised.  It costs $300,000 to $900,00 a year depending on the girl’s weight.

It looks very odd that the large bumetanide failed, even though 25-40% were actually responders. By the way, my son’s bumetanide therapy has cost about $80 a year, for the last 13 years.

It does not fill you with great confidence.

I recently saw an article saying that “paracetamol/ acetaminophen does not, after all, increase the incidence of autism.” Well theoretically it should be harmful, by depleting glutathione, which is why it should be taken with NAC. We also know that NAC taken during pregnancy can significantly reduce the risk of miscarriage and this has been studied in a clinical trial.

N-acetyl cysteine for treatment of recurrent unexplained pregnancy loss

A controlled clinical trial studied N-acetylcysteine (NAC) in 168 pregnant women with a history of recurrent unexplained miscarriage. Women received either folic acid alone or folic acid plus NAC at 600 mg per day. In the NAC group, 52% of pregnancies continued beyond 20 weeks, compared with 27% in the control group. The take-home baby rate was 47% in the NAC group, compared with 21% in the control group. This represents more than a doubling of the live birth rate. NAC works by restoring glutathione, the cell’s main antioxidant, protecting placental and fetal tissue from oxidative stress. Oxidative stress is known to impair placental function and contribute to pregnancy loss. NAC was well tolerated, with no significant safety concerns reported. These results suggest that correcting oxidative stress can directly improve pregnancy outcomes in a defined high-risk group. This study illustrates how targeting a specific biological mechanism can dramatically change developmental outcomes.

If a professionally-managed autism trial cannot detect the 25-40% who responded to some extent, do you believe a study that effectively says nobody gets autism from pre-natal acetaminophen. Not even 1%? All you likely need to do is pair it with NAC to make the risk 0%.

For decades doctors refused to believe regressive autism existed. Once people started videoing their toddlers, it became impossible to doubt that some actually had developed speech and then lost it. Parents were not imagining it. It was just an inconvenient truth, and still is.