UA-45667900-1
Showing posts with label Verapamil. Show all posts
Showing posts with label Verapamil. Show all posts

Wednesday, 6 November 2019

Metformin to raise Cognition in Fragile X and some other Autisms?




I started to write this post a long time ago, when Agnieszka first highlighted an interview with Dr Hagerman from UC Davis.  Hagerman is experimenting in using Metformin to treat Fragile-X.

Having again be reminded about Metformin, I realized that I never finished my post on this subject. With some extras about autophagy and a nice graphic courtesy of Ling’s excellent paper, here it is. 

Metformin has already been covered in 5 previous posts.


One interesting point is that the researchers at UC Davis are using the measurement of IQ as one of the outcome measures in their trial of Metformin.  I have been suggesting the French Bumetanide researchers do this for a long time.

It is my opinion that simple medical interventions can have a profound impact on the IQ of some people with severe autism. I mean raising IQ not by 5-10 points as at UC Davis, but by 20-50 points.  IQ can be measured using standardized tools and is far less subjective than any autism rating scale.

The big-time potential IQ enhancers we have seen in this blog include: -

·        Bumetanide/Azosemide
·        Statins (Atorvastatin, Lovastatin, Simvastatin, but they are not equivalent and the effect has nothing to do with lowering cholesterol)
·        Micro-dose Clonazepam
·        Clemastine
·        It seems DMF, in n=2 trial

The good news is that these drugs are all off-patent cheap generics (except DMF), as is metformin.  No need for drugs costing $50,000 a year.

For those that do not know, metformin is the first line medication for type-2 diabetes. It was introduced as a medication in France in 1957 and the United States in 1995.  In many countries Metformin is extremely cheap, with 30 x 500 mg tablets costing about $2 or Eur 2. In the US it costs about $10 for generic, so not expensive. 

There are sound reasons why Metformin could increase IQ in someone with autism or Fragile-X. In the case of idiopathic autism is there a likely biomarker to identify a likely responder? One has not yet been identified.

Clearly Metformin will not work for all people with autism and MR/ID, but even if it only works for 10% that would be great.

Are all parents going to notice an increase in IQ of 5-10 points?  You might think so, but I doubt it.  I would hope therapists, teachers and assistants would notice.

I think basic mental maths is the best way to notice improved cognitive function in people with IQ less than 70.  You can easily establish a baseline and then you can notice/measure improvements.

Improved cognitive function does not just help with maths, it helps with learning basic skills like tying shoe laces, brushing teeth and later shaving.  This does also involve many other types of skill.





In the study, researchers from the UC Davis Medical Investigation of Neurodevelopmental Disorders Institute in California tested the long-term effects of metformin, delivered at 1,000 milligrams (mg) twice a day, for one year in two male patients, 25 and 30 years old. Genetic analysis confirmed that both patients had mutations in the FMR1 gene, confirming their fragile X syndrome diagnoses.

The younger patient had autism and was also diagnosed with generalized anxiety disorder. First prescribed metformin at 22, he is currently taking 500 mg of metformin twice a day and 10 mg per day of simvastatin — used to lower the level of cholesterol in the blood.
The second patient was also diagnosed with anxiety and exhibited socially nervous behaviors, including panic attacks. He had severe limitations in language use, and communicated in short sentences and by mumbling. He had been on an extended-release formulation of metformin, taking 1,000 mg once a day for one year.

Both patients showed significant cognitive and behavioral improvements. After one year of treatment with metformin, test results revealed an increase in the patients’ IQ scores, from 53 to 57 in the younger patient and from 50 to 58 in the second patient.

Verbal and nonverbal IQ — the ability to analyze information and solve problems using visual or hands-on reasoning — were also improved in both patients. Non-verbal IQ increased from 50 to 52 in the younger patient and from 47 to 51 in the other. Verbal IQ went from 61 to 66 in the first patient, and from 58 to 68 in the second.

                                                              

Researcher Randi Hagerman is a big proponent of metformin — a diabetes drug that helps people manage their weight. In fact, Hagerman takes the drug herself as a preventive measure against cancer.
Metformin has also unexpectedly shown promise for improving cognition in people with fragile X syndrome, a leading genetic cause of autism characterized by severe intellectual disability.

A study published in 2017 linked impaired insulin signalling in the brain to cognitive and social deficits in a fruit fly model of fragile X, and the flies improved on metformin. A second paper that year showed that metformin reverses abnormalities in a mouse model of the syndrome, including the number of branches the mice’s neurons form. It also improved seizures and hyperactivity in the mice — issues we also see in people with fragile X.
I began prescribing metformin to people with fragile X syndrome to help curb overeating. Many of the people I treat are overweight because of this habit — it’s one of the symptoms of a subtype of fragile X called the Prader-Willi phenotype, not to be confused with Prader-Willi syndrome.
I was surprised when the families of these individuals told me they could talk better and carry out conversations, where they couldn’t before. That really gave us impetus to conduct a controlled clinical trial.
It’s not a cure-all, but we do see some positive changes. It doesn’t resolve intellectual disability, but we have seen IQ improvements of up to 10 points in two boys who have been treated with metformin. We are very excited about that.

Individuals on metformin tend to start eating less, and often lose weight as a result. I could kick myself, because metformin has been approved to treat obesity for many years, but I never thought to use it in fragile X syndrome. Oftentimes children with fragile X syndrome have so many problems that you aren’t thinking about obesity as the top priority.
We’ve also seen a gradual effect on language, which we can detect after two to three months. Sometimes there are improvements in other behaviors too; I’ve seen mood-stabilizing effects. Many people with fragile X syndrome have issues with aggression, and it’s possible these could be moderated with metformin too. 

Individuals with fragile X syndrome (FXS) have both behavioral and medical comorbidities and the latter include obesity in approximately 30% and the Prader‐Willi Phenotype (PWP) characterized by severe hyperphagia and morbid obesity in less than 10%. Metformin is a drug used in individuals with type 2 diabetes, obesity or impaired glucose tolerance and it has a strong safety profile in children and adults. Recently published studies in the Drosophila model and the knock out mouse model of FXS treated with metformin demonstrate the rescue of multiple phenotypes of FXS.

Materials and Methods

We present 7 cases of individuals with FXS who have been treated with metformin clinically. One case with type 2 diabetes, 3 cases with the PWP, 2 adults with obesity and/or behavioral problems and, a young child with FXS. These individuals were clinically treated with metformin and monitored for behavioral changes with the Aberrant Behavior Checklist and metabolic changes with a fasting glucose and HgbA1c.

Results

We found consistent improvements in irritability, social responsiveness, hyperactivity, and social avoidance, in addition to comments from the family regarding improvements in language and conversational skills. No significant side‐effects were noted and most patients with obesity lost weight.

Conclusion

We recommend a controlled trial of metformin in those with FXS. Metformin appears to be an effective treatment of obesity including those with the PWP in FXS. Our study suggests that metformin may also be a targeted treatment for improving behavior and language in children and adults with FXS.

Recruiting: Clinical Trial of Metformin for Fragile X Syndrome


While a growing number of families are trying metformin and reporting mixed results, metformin has not yet been systematically studied in patients with Fragile X syndrome. This open-label trial is designed to better understand the safety and efficacy of this medicine on behavior and cognition, and to find the best dosages for children and adults.

20 children and adults with Fragile X syndrome will take metformin 250mg twice a day for the first week, followed by metformin 500mg twice a day for the next 8 weeks.
The study will measure changes in the total score on the Aberrant Behavior Checklist-Community (ABC-C) after 9 weeks of metformin treatment. The ABC-C is a 58-item behavior scale which is filled out by a caregiver. In addition, Transcranial Magnetic Stimulation (TMS) will be used to look for changes in cortical excitability and Electroencephalography (EEG) will assess levels of synaptic plasticity.
Participants in this study must be Canadian residents and be able to travel to the University of Sherbrooke in Quebec, Canada, for several visits. If you are interested in metformin but this trial is not convenient, there are two alternatives. FRAXA is funding a new trial of metformin in New Jersey, and Dr. Randi Hagerman is currently recruiting for metformin trial at the University of California at Davis MIND Institute.



Metformin has emerged as a candidate drug for the targeted treatment of FXS based on animal studies showing rescue of multiple phenotypes in the FXS model. Metformin may contribute to normalizing signalling pathways in FXS in the central nervous system, which may include activities of mTOR and PI3K, both of which have shown to be pathogenically overactive in FXS. In addition, metformin inhibits phosphodiesterase, which would lead to correction of cAMP levels, and MMP9 production, which is also elevated in FXS. Looking at the potential signalling pathways, metformin appears to be a good candidate for targeting several of the intracellular functions in neurons disrupted in FXS and, therefore, has potential to rescue several types of symptoms in individuals with FXS. The researchers have utilized metformin in the clinical treatment of over 20 individuals with FXS between the ages of 4 and 58 years and have found the medication to be well tolerated and to provide benefits not only in lowering weight gain and normalizing appetite but also in language and behavior. In this controlled trial, the researchers hope to further assess metformin's safety and benefits in the areas of language and cognition, eating and weight loss, and overall behavior.


mTOR and P13K

Hagerman highlights Metformin’s effects on mTOR and P13K pathways.

This is a highly complex subject and the graphic below from an early post shows how interconnected everything is.  If mTOR is not working correctly you can expect many things not to work as nature intended.

Numerous things can cause an imbalance in mTOR and so there are numerous ways to re-balance it.

Not surprisingly much of this pathway plays a role in many types of cancer.

Hagerman herself is taking Metformin to reduce her chances of developing cancer. I think that is a good choice, particularly if you are overweight.  My anticancer choice, not being overweight, is Atorvastatin which targets inhibition of PI3K signalling through Akt and increases PTEN.

Hagerman is 70 years old and I think many cancers actual initiate years before they are large enough to get noticed and to be effective any preventative therapy needs to be started before that initiation has occurred. Hopefully she started her Metformin long ago. 

Given that 50% of people are likely to develop one cancer or another, I am with Dr Hagerman on the value of prevention, rather than treatment/cure.







The Wrong Statin for Fragile-X?

In the first article highlighted in this post, there is a case history of a man with FX being treated by a Statin, it looks to me that he has the wrong prescription (Simvastatin). Perhaps Dr Hagerman should read this old post from this blog:-


Choose your Statin with Care in FXS, NF1 and idiopathic Autism







   Simvastatin does not reduce ERK1/2 or mTORC1 activation in the Fmr1-/y hippocampus.
So  ? = Does NOT inhibit

The key is to reduce Ras. In the above graphic it questions does Simvastatin inhibit RAS and Rheb.
                                                                                                     

For anyone really interested, the following graphic from a previous post shows the fragile X mental retardation protein, FMRP.  Lack of FMRP goes on increase neuroligins (NLFNS) this then creates an excitatory/inhibitory imbalance which cause mental retardation and features of autism.





This all suggests that the 25 year-old young man with Fragile X treated at UC Davis (case study above) should switch from Simvastatin to Lovastatin.




Metformin and Autophagy

I also think Dr Hagerman is less likely to get dementia now that she is talking metformin.  If she takes vigorous exercise at least once a week, I think that is also going to keep her grey cells ticking over nicely. Like Dr Ben Ari, Hr Hagerman is working way past normal retirement.  If you love your job, then why not?  As with many things, in the case of neurons, “use them or lose them”.

Autophagy in Dementias


Dementias are a varied group of disorders typically associated with memory loss, impaired judgment and/or language and by symptoms affecting other cognitive and social abilities to a degree that interferes with daily functioning. Alzheimer’s disease (AD) is the most common cause of a progressive dementia, followed by dementia with Lewy bodies (DLB), frontotemporal dementia (FTD), vascular dementia (VaD) and HIV associated neurocognitive disorders (HAND).
The pathogenesis of this group of disorders has been linked to the abnormal accumulation of proteins in the brains of affected individuals, which in turn has been related to deficits in protein clearance. Autophagy is a key cellular protein clearance pathway with proteolytic cleavage and degradation via the ubiquitin-proteasome pathway representing another important clearance mechanism. Alterations in the levels of autophagy and the proteins associated with the autophagocytic pathway have been reported in various types of dementias. This review will examine recent literature across these disorders and highlight a common theme of altered autophagy across the spectrum of the dementias.

Below is an excellent graphic from a paper highlighted by Ling. Note metformin, above AMPK.


Autophagy Activator Drugs: A New Opportunity in Neuroprotection from Misfolded Protein Toxicity









I would highlight the presence of IP3R, the calcium channel proposed by Gargus as being a nexus in autism, for where multiple types of autism meet up, to do damage.

Verapamil, in Monty’s Polypill, increases autophagy independently of mTOR in a complicated mechanism  involving IP3R and likley calpain.  It is proposed as a therapy for Huntington’s Disease via this mechanism. At the lower right of the chart below we see calpain, a group of calcium dependent enzymes, not well understood.  ROS can activate calpains via L-type calcium channels.





I would not worry about the details.  The take home point is that if you have autism, dementia or many other neurological conditions, you might well benefit from increasing autophagy.  There are very many ways to do this.      
                                                           
Conclusion

Fortunately, I am not a doctor.  I do recall when my doctor father was out visiting his sick patients at their homes, he did have not only his medical bag, but also some useful gadgets always kept in his car, that might come in handy.

The autism equivalent is the personalized Polypill therapy for daily use and the autism toolbox to delve into to treat flare-ups in autism as and when they arise.

I do think some people should have metformin in their daily Polypill therapy.

I think we can safely call Fragile-X a type of autism, so we already know it works for at least some autism.  Metformin is a very safe old drug, with minimal side effects and it is cheap.  It ticks all the boxes for a potential autism therapy.  Will it work for your case?  I can tell you with certainty that it does not work for everyone.

Metformin has been trialled to treat people with obesity and autism, since it can reduce appetite.

Metformin forTreatment of Overweight Induced by Atypical Antipsychotic Medication in YoungPeople With Autism Spectrum Disorder: A Randomized Clinical Trial.


INTERVENTIONS:

Metformin or matching placebo titrated up to 500 mg twice daily for children aged 6 to 9 years and 850 mg twice daily for those 10 to 17 years.

MAIN OUTCOMES AND MEASURES:

The primary outcome measure was change in body mass index (BMI) z score during 16 weeks of treatment. Secondary outcomes included changes in additional body composition and metabolic variables. Safety, tolerability, and efficacy analyses all used a modified intent-to-treat sample comprising all participants who received at least 1 dose of medication.

RESULTS:

Of the 61 randomized participants, 60 participants initiated treatment (45 [75%] male; mean [SD] age, 12.8 [2.7] years). Metformin reduced BMI z scores from baseline to week 16 significantly more than placebo (difference in 16-week change scores vs placebo, -0.10 [95% CI, -0.16 to -0.04]; P = .003). Statistically significant improvements were also noted in secondary body composition measures (raw BMI, -0.95 [95% CI, -1.46 to -0.45] and raw weight, -2.73 [95% CI, -4.04 to -1.43]) but not in metabolic variables. Overall, metformin was well tolerated. Five participants in the metformin group discontinued treatment owing to adverse events (agitation, 4; sedation, 1). Participants receiving metformin vs placebo experienced gastrointestinal adverse events during a significantly higher percentage of treatment days (25.1% vs 6.8%; P = .005).

CONCLUSIONS AND RELEVANCE:

Metformin may be effective in decreasing weight gain associated with atypical antipsychotic use and is well tolerated by children and adolescents with ASD.

My guess is that a minority will be responders, the benefit will manifest itself in different ways and so it will be a useful part of polytherapy for some people, but it will not be a silver bullet.  Other than via an IQ test, I think the benefit will be hard to measure, even when it is very evident. 

In the end there will be a clever way to predict who will respond to which therapy.  Today’s post actually replaces one that will look into genetic testing and DEGs (differentially expressed genes). Most likely testing for DEGs will be the best predictor of what drugs work for whom.

Intelligent, cautious trial and error using safe drugs is an alternative strategy.  It is available today; it is cheap and it does work.

I have not tried Metformin yet, in recent years I have had most success with my own ideas. I have some of Dr Frye's calcium folinate sitting at home waiting for a trial.  Both Metformin and calcium folinate should be trialled.  The other obvious thing to trial is that Japanese PDE4 inhibitor Ibudilast (Ketas).  Thanks to Rene we now know you can acquire this is via any international pharmacy in Germany, with a prescription. It also reappeared on the website of a Japanese online pharmacy. The Western PDE4 inhibitors, like Daxas/Roflumilast are not selective enough and so are emetic (they make you want to vomit). Low dose Roflumilast has been patented as a cognitive enhancer, but you may need to have a bucket with you at all times.




     






Thursday, 24 May 2018

An Autism Case History - EpiphanyASD in a Pill





It is not quite that easy!


Initials:                        LT
Age:                           14 years old

Year
of Birth:              2003

Sex:                            Male

Date:                           24 May 2018

Diagnosis      
LT was diagnosed with autism in January 2007, at a multi-disciplinary assessment in London, at the age of 3 years 6 months.  At that time, LT was non-verbal but had some emerging vocalization. No tics, no seizures, no unusual physical features, no self-injury, no sleep disorder, no feeding disorder. Toilet trained. Very limited attention span. No imaginative play. Liked to jump.

IQ not tested.
No CARS (Childhood Autism Rating Scale) assessment.
TEACCH and PECS were recommended as therapy.
Further medical testing or referrals – none recommended (standard practice in the UK) 

LT has an older brother who is intelligent, multilingual and highly social.
Comorbidities
GI disease:                   None
Epilepsy:                       None
Asthma:                        Yes, mild asthma from early childhood
Allergy:                         Pollen
Sleep disorder:             None 

General Health          
Very healthy and almost never ill. When visiting his GP at the age of 14 the doctor commented how she had not seen him for three years, whereas she has seen his older brother twice a year.

Born via a planned caesarean section, without complications, APGAR score was 10.


Growth                      
Body is well proportioned, no obvious macro/microcephaly. No physical features of any syndromes/metabolic anomalies.

However, LT was initially on the 90th percentile for height and dropped to the 20th by the time he was 5 years old. He was a very muscular baby.  At the age of 10 his bone age (X ray of left hand) was estimated to be two years delayed.  IGF-I was normal, FT3 was slightly above the reference range.

At birth he fitted the research description of hyperactive pro-growth signaling pathways, even though there was no macrocephaly.

Regression at age 8              
Aged 8, a big regression took place with self-injurious behavior (SIB) and aggression to others. He would slam his head into walls, other people, car windows, punch himself etc, but he was still small enough to be physically controlled/restrained by larger adults. He could not be controlled by smaller/older adult family members.

This aggression could occur immediately on waking until finally falling asleep at night, it was not predictable.  At that time in the afternoons, LT had a male 1:1 assistant with experience from a school for severe autism and in the mornings a very firm-minded tall female 1:1 assistant. LT’s father imposed a policy of zero acceptance of any SIB, to avoid it becoming a permanent acquired behavior. SIB was physically blocked.
The regression was triggered by the departure of his long time full-time 1:1 female assistant. It was an emotional trauma.  Occasional visits from her just made the situation worse.  In response no drugs were used, just a consistent firm behavioral approach. Over a ten month period the situation slowly stabilized, but skills were lost and bad habits (SIB) were acquired.  LT subsequently did see his assistant again and sees her regularly to this day.
Throughout this time his classmates and teacher at school were remarkably understanding. He was never excluded from school. His assistant ensured nobody at school got hurt.
Since assistants will inevitably come and go, from the age of 8 LT has had two part-time assistants rather than one full time.  As and when subsequent assistants have left, he has not had any troubling emotional reaction. 

Summer-time raging and loss of cognitive function
Summertime raging with self injurious behavior and aggression to others developed from the age of 9.

Later it became clear that in addition there was a loss of cognitive function during the summer months. This became evident once it was possible to teach mental math, from aged 9 onwards.  For example, at the age of 11, simple verbal tasks like 7 x 8 = ?, that had previously been mastered, could not be answered in the summer months.

The raging and cognitive loss were ultimately treatable.

Winter-time raging

Summertime raging was resolved and then winter-time raging developed. This was traced back to the cytokines released to signal reabsorption of milk teeth roots (a proves that takes months) and the eruption of permanent teeth. It was not tooth ache, i.e. pain. LT has retarded bone age and apparently this applied to his teeth development as well.

He lost his later milk teeth always in the winter.

The winter time raging did not respond to his summertime therapy, but responded very well to a low dose of ibuprofen. Summertime raging does not respond to Ibuprofen 

PANS-like episode aged 13
At Christmas time, following a minor viral infection, LT developed acute onset profound verbal tics. LT does not have Tourette’s type autism and had never exhibited such behavior previously. The tics were treated as a PANS/PANDAS flare-up with 5 days of prednisone. Over a two week period the tics faded away and have never returned.

Intellectual disability 
IQ was never measured, LT’s ABA consultant said there was no point, but the very much more rigorous ABBLS was completed, see below. Evidently, prior to pharmacological treatment at the age on 9, there was a 5 year developmental delay.
With hindsight, IQ pre-treatment was probably in a similar range to Down Syndrome (DS) meaning less than 70.

At the age of 14, LT’s academic performance now puts him in the top half of his class of 12 year old neurotypical peers. His grades are mainly As, with maths and computing being particular strengths. 

Other testing:    No genetic testing, MRI or EEG.

Family History:          

LT has a 7 years younger, very distant cousin who is non-verbal with autistic disorder. They have shared great great great maternal grandparents. The cousin has parents who are both doctors and were high academic achievers as medical students.

The father’s family has a large number of Cambridge-educated doctors on both the grandmother's and grandfather's side; one gave his name to the scale still used to assess severity of Ulcerative Colitis and helped develop the first H2 anti-histamine drug. The father and uncle are engineering graduates from top universities. One distant cousin was a math’s protégé at Oxford University.  One distant cousin has bipolar. One uncle has type-1 diabetes.

The mother is an academic alpha female in a stressful creative profession. The maternal grandmother was a teacher and grandfather was an army Colonel.
The maternal grandmother and her children all had premature hair greying, which may be linked to Bcl-2 expression and Wnt signaling, both implicated in autism. Thickness and greying of hair share biological mechanisms, which overlap with those controlling development of dendritic spines. LT and his father have very dense hair, mother has thin hair.
Maternal grandparents both smoked and the grandfather has COPD (Chronic Obstructive Pulmonary Disease). Oxidative stress is a core feature of COPD, because anti-oxidant genes are silenced; these effects are known to be heritable via epigenetic tags. 
The family fits the high IQ  type of autism (some autism genes are linked to intelligence and some bipolar genes are linked to creativity, which helps explain why some actors/artists are bipolar) with oxidative stress raised during pregnancy, anti-oxidant response possibly weakened, no oxytocin surge during delivery and no microbiota transfer at birth (C-section delivery).  No pets at home during pregnancy (a good source immuno-stabilizing bacteria).  No obesity in the family.

Education
LT has attended the same mainstream international school, following the English curriculum, since the age of 3. Class sizes are very small, about 12 pupils. From the age of 4 he has had a 1:1 assistant eight hours a day, throughout the year.
LT commenced a parent-managed ABA (Applied Behavioral Analysis) inspired home program shortly after diagnosis.  Both parents attended a 2 day training program to learn the use of PECS (Picture Exchange Communication System).  PECS was applied and shortly thereafter LT became partly verbal at the age of 4, speaking single words.
1:1 assistants were recruited mainly from the local University and trained to apply ABA, with elements from Floortime and the Canadian Hanen Program. There was some supervision from US-trained Behavioral Consultants that would fly in for training. A large collection of specialist training material was acquired from the US. 
Extensive use was made of professional (i.e. expensive) special needs language teaching software (Laureate Learning) from the age of 4 until 8 years old.
Later, web-based reading software (Headsprout) was used and years later special maths teaching software (Math Wizz). Neither are made for special needs, but both are very compatible with an ABA approach.
LT spent an extra year in kindergarten and in primary/junior school was held back 2 years at the age of 9, following a request from the parents.
In primary school (English system) he went Year 1, Year 2, Year 3 (started bumetanide) then back to Year 2, then Year 3, Year 4, Year 5, Year 6 and currently attends Year 7 in secondary/high school
The equivalent in the US system would have been, he went K, 1st, 2nd, then 1st, 2nd 3rd, 4th etc.
From the age of 13, LT attended school full time, prior to that he attended only the morning and then went home after lunch to work 1:1 with his assistant for three hours.
During school holidays LT has a 1:1 home learning program.
LT learnt to read and write at home as result of the unrelenting efforts of his assistant. He started to learn maths from the age of 8, prior to that he could not master the basic concepts, or understand the relevant vocabulary.
From the age of 9, LT has been able to keep up with his new peer group at school, two years his junior.
At the age of 14, in a class with 12 year old neurotypical children, LT takes the same assessments as the rest of the class and his grades currently place him in the top half of the class. He is now particularly good at things like arithmetic, algebra, coordinates, spelling and has neat handwriting (very unusual in autism). He is still clearly autistic and his speech is limited to what he wants to say; there is no small talk.
LT started to learn the piano aged 8. He progressed from an extremely basic level and a desire to hit his teacher to his current level 4 of the popular Faber Music piano course (there are just 5 levels). When he plays in public people are very surprised, he does not play like someone with any cognitive impairment. His peers as school have asked “how can he play like that?” 

Motor Skills
Like many people with classic autism LT had problems with both fine and gross motor skills as an infant. After a great deal of 1:1 therapy, motor skills are now normal.
LT started to learn to ski at the age of 5 with a special needs instructor. Progress was initially slow, but 9 years and one broken collar bone later, LT can confidently ski on red slopes and deal with all the various types of lifts you encounter in the Alps.
Stamina improved considerably after starting to take Agmatine, which is evident at school where they are timed to run 2.5 km (1.5 miles) and when swimming.

Behavioral Treatment (age 3- 8)
From diagnosis aged three, until nine years old, therapy was exclusively based on behavioral interventions. Extensive use of ABA (Applied Behavioral Analysis) and VB (Verbal Behavior) with 40 hours a week with a 1:1 Assistant.
At the aged of 9, LT had mastered almost all the skills in the very extensive ABBLS (Assessment of Basic Language and Learning Skills) assessment. The language skills and other basic learner skills that are tracked by this tool are those that are acquired by most typically developing children by the time they reach four to five years of age. LT’s elder brother had acquired these while he was three years old.
LT’s skill acquisition to the age of eight was seen by the ABA consultants as nothing unusual in someone with classic autism. There was slow but continuous progress. 
All learning was taking place at home with school attended mainly for socialization.

Pharmacological Treatment (age 9 onwards)
In late 2012 a small clinical trial was published by Ben Ari and his clinical associate Lemmonier; it showed the benefit of the NKCC1/2 blocker bumetanide in autism. This paper was studied by LT’s father and contact has been maintained for several years with Dr Ben Ari, who originated and patented this therapy.
Bumetanide (1mg per day) was commenced just before Christmas December 2012, unknown to the school, or LT’s assistants.
On returning to school in January 2013 the Head Teacher summoned LT’s father and asked what had happened to LT. He was “so joyous” and “like a different child”.
At the suggestion of his original ABA consultant, LT’s father had been asking LT every school day for 5 years “what did you have for lunch at school today?”. The usual answer would be no answer, the wrong answer, but sometimes a brief correct answer. From now on LT would say precisely what he had eaten “peas, potatoes and chicken – cake for dessert”. The assistant was there to confirm what had really been eaten for lunch. 
LT’s 1:1 assistant at that time described the effect of bumetanide as making him “more present”. Since his assessment at the age of 3, it was always noted that LT had a very short attention span and would not be able to focus on the class teacher for more than a couple of minutes. LT was never hyperactive, quite the opposite. He was physically present but not mentally.
Later on it would be realized that the most potent effect of long term bumetanide use in strictly defined autism (SDA) is enhanced cognition, which leads to accelerated acquisition of new skills.  IQ has long been seen as the best predictor of more favorable outcomes in autism.  
Bumetanide use has continued for five years, with occasional pauses to confirm it still works.  Different doses were tested and currently the dose is 2mg once a day.
When stopping bumetanide for a week and returning to his web-based maths learning program, LT was unable to complete previously mastered tasks, no matter how many times he tried. Having recommenced bumetanide, the same maths problems were attempted a week later and could be solved. 
Blood potassium levels were checked regularly at the beginning, but were always high normal (5.0 mmol/L).  Bumetanide is taken with 250mg of K+ per 1mg of bumetanide. Diet is rich in potassium, with bananas and other fruit.
Dehydration, another potential problem, is entirely self-regulated with LT drinking more water. Total consumption is 2.5 to 3 liters per day.
Diuresis occurs mainly within one hour of taking bumetanide and has never caused a problem at home or school. LT takes his bumetanide at least an hour before leaving home for school.
Bumetanide’s suggested mode of action is lowering intracellular chloride via blocking NKCC1 cotransporters in the brain.  Bumetanide crosses the blood brain barrier very poorly and many researchers are dubious it can have any effect. Bumetanide is a partial solution.
A new drug is being developed by Dr Ben Ari that will cross the blood brain barrier more effectively than bumetanide and have less effect on NKCC2, so producing less diuresis.
An alternative strategy discussed in the literature is to improve the pharmacokinetics of bumetanide, by slowing its excretion via OAT3 (organic anion transporter 3) and thus increasing plasma concentration. There are many OAT3 inhibitors, the best known and most potent is probenecid, used to treat gout by increasing the excretion of uric acid. Some foods are OAT3 inhibitors. One readily available substance is chlorogenic acid (more precisely 1,3- and 1,5-dicaffeoylquinic acid) which is sold as a coffee-based weight loss supplement. Interestingly, coffee, but not caffeine, has been shown to reduce the risk of gout.
Little is known about exactly how bumetanide is transported/excreted across the blood brain barrier.
Bumetanide’s autism benefit appears to be from lowering intracellular chloride and hence making GABAA become more inhibitory. Excitatory-Inhibitory (E/I) imbalances are widely believed to be at the core of autism.  An E/I imbalance during so-called Critical Periods, will result in permanent changes to the developing brain, nonetheless it appears that correcting an E/I imbalance in later years can still be highly beneficial, though not curative. 
Another experimental therapy also makes GABAA become more inhibitory. This uses very low doses of clonazepam to modify the behavior of GABAA receptors that contain the α3 sub unit.  In LT the effective dose of clonazepam is just 0.03mg, which might be considered sub-clinical, but as predicted by Professor Catterall, it does have a beneficial effect (a bumetanide-like effect). It has no side effects and there is no tolerance develops at this tiny dose, after four years of use.
At the time low dose clonazepam was introduced, LT would go swimming at 5pm most days. He was not really interested to do much independently in the water, he was very passive. This passive behavior was notably changed once the effective clonazepam dose had been found. He became more like a typical child playing in a swimming pool. Instead of sitting on the steps he wanted/demanded interaction/play with the attending adult.  The effect was not as profound as that seen in the first months of bumetanide, but noticeable nonetheless.
After 4 years of bumetanide the effect was still there, but there was a desire to accelerate skill acquisition to keep up with neurotypical school peers.
A new strategy was adopted to further reduce intracellular chloride, this time using a method first documented in the 1850s, when potassium bromide (KBr) was used to treat epilepsy. Reading old case studies from Great Ormond Street Hospital in London it appeared to LT's father that some children with epilepsy, MR/ID and undiagnosed autism improved behaviorally and developed age-appropriate play when treated with KBr. Lack of age-appropriate play is a hallmark of autism.  Modern research shows that bromide ions compete with chloride ions to enter cells and the result is a lower intracellular concentration of Cl-. The limiting factor in the use of KBr is that it increases mucous secretions and so causes acne (and can make asthma worse), in a dose dependent fashion. At a low dose of 400mg per day there is a cognitive gain without significant spots. KBr is still used at high doses to treat pediatric epilepsy in Germany and Austria. Some leading US neurologists regret they cannot prescribe it; technically they could ask the FDA for permission on a patient by patient basis.

Another strategy to reduce intracellular chloride is to target chloride ions that enter neurons via the AE3 exchanger, this is possible using Acetazolamide (Diamox). This therapy does seem to work for some people, but was not tolerated by LT, it caused reflux.
KBr has a very long half-life and so it takes 4-5 weeks to reach the maximum effect. 
Bumetanide took about two weeks to lower chloride and show behavioral and cognitive improvements.
Low dose clonazepam takes three days, as was predicted by its half-life.
The cognitive loss in severe autism has parallels with that in Down Syndrome (DS). Bumetanide has been patented as a therapy for DS by Ben Ari, based on the results from mouse studies.
In mouse models of Down Syndrome both a negative allosteric modulator and a selective inverse agonist of α5 sub-unit of the GABAA receptor improve cognition. 
Mouse research has shown that poor learners have greater GABRA5 expression than good learners and that in mice GABRA5 expression can be normalized by eating cinnamon, or its metabolite sodium benzoate (NaB); this makes a poor learner become a good learner, at least in mice.
So it may be that increasing the effect of α3 sub-unit of the GABAA and reducing the effect of the α5 sub-unit of the GABAA can both improve cognition. For the moment the latter remains unproven. NaB is an approved food additive, E211. Ceylon cinnamon, which is safe for long term consumption, is metabolized to NaB. People who are histamine intolerant have to avoid DAO inhibitors such as cinnamon and NaB. 

Summertime raging and loss of cognitive gains
From the aged of 8 it became apparent that summer provoked behavioral deterioration. At this point there was no obvious allergy, but behavior improved when moving to the mountains in summer. At first, OTC mast cell stabilizers were investigated; some common H1 antihistamines are partial mast cell stabilizers. Rupatadine, azelastine, ketotifen, loratadine and cetirizine were all tried, as was the flavonoid quercetin.
Some of the above did indeed help reduce the summertime self injury, but not to a satisfactory level.
A final solution was found in a small dose of the Cav1.2 blocker, verapamil. 
When mast cells degranulate, one step requires activation of an L-type calcium channel. This is why most mast cell stabilizers are actually calcium channel blockers.
It should be noted that mutation in the CACNA1C gene, which encodes the Cav1.2 ion channel, leads to a severe kind of autism called Timothy Syndrome. Because Cav1.2 is widely expressed in the heart those affected have a very poor prognosis.
In addition, verapamil blocks the potassium ion channel Kv1.3.  Potassium channels, Kv1.3 and KCa3.1, have been suggested to control T-cell activation, proliferation, and cytokine production. Kv 1.3 is widely regarded as a therapeutic target for immunomodulation in autoimmune diseases.  Research has shown that peptides from parasitic worms that suppress the body's immune response do so by blocking Kv1.3. A drug therapy based on these peptides is being developed.
Verapamil also upregulates autophagy, which is impaired in many neurological disorders, such as Huntington’s. Lack of autophagy has been linked to the synaptic pruning deficits found in autism.
Verapamil has a short half-life of about 3 hours. Only a small dose is required to prevent the onset of SIB and the preceding agitation (described by LT as “spray the fire in my head”).
From the age of 10, LT’s summertime raging has been treated with 40-80 mg of Verapamil split into 2-3 doses from May until late November.
On the occasions that he has missed his 1pm dose in the peak allergy period, he has repeatedly developed aggression and self-injury by 4 or 5pm.
When he has taken verapamil there has never been any aggression and or self-injury.
Once self-injury was removed as a concern, learning progressed during the long summer school holidays. It became clear that during summer cognition was reduced as if bumetanide was no longer working.
It has been shown that the expression KCC2, the cotransporter that allows Cl- to leave neurons is affected by inflammatory cytokines like IL-6. It therefore appears plausible that the histamine and IL-6 released directly and indirectly by mast cell degranulation was causing an increase in neuronal Cl- and thus undoing the good work being done by bumetanide. Inflammation also increases α5 GABAA receptor activity and can thus reduce cognitive function.
At this point, the bumetanide dose was raised from 1mg once a day to 2mg in the morning and on occasion 1mg in the late afternoon.
The combination of an increased dose of bumetanide and the use of verapamil, cetirizine and azelastine has produced a very favorable result (no SIB and minimal summertime cognitive decline). Perhaps of note is that cetirizine is an eosinophil stabilizer, which may also be helpful and not just for asthma.
OTC therapies that have a helpful effect in summer are L-histidine, curcumin and L. reuteri DSM 17938 (sold as Biogaia Protectis). The amino acid histidine is a precursor to histamine and it seems that the body’s feedback loops can be tricked into not degranulating mast cells by slightly increasing the level of circulating histidine. The immunomodulatory effects of L. reuteri DSM 17938 have been well studied; the effect however does not continue after prolonged use. Curcumin is a very widely studied natural substance that performs much better in vitro than in vivo, due to very poor bioavailability. Modified versions of curcumin have been developed and there is a marginal benefit. Histidine is extremely cheap and easy to administer. Modified curcumin and L. reuteri are quite expensive.
It is reported by others that at a higher dose verapamil is as effective as an H1 antihistamine in treating allergy. 

IPR3
It appears that aberrant calcium channel signaling is a key feature of much autism. Gargus has suggested that IP3R is a nexus for different dysfunctions that lead to autism. IP3R controls the release of calcium stored within cells (the endoplasmic reticulum).
Excessive calcium within cells is known to be damaging. L-type calcium channels that remain open will raise intracellular calcium and the same is true with IP3R. Caffeine can be used to inhibit calcium release via IP3R.
Gargus has not proposed an IP3R therapy.  


RORα

RORα is another proposed nexus where different dysfunctions  that lead to autism may converge. One potential RORα agonist is estradiol.  We know that in much autism there is elevated testosterone and reduced estradiol; we also know that estrogen receptor beta is under-expressed. Estradiol is known to be highly neuroprotective and may help protect females from developing autism. Females lacking in estradiol, for example in Turner Sydrome, may exhibit features of autism. A logical therapy would be to either use estrogens, or reduce testosterone (effectively the same thing). Ideally you would do this just in the brain; a brain selective pro-drug of estradiol, called DHED, actually exists. Less ideal therapies range from estradiol itself, to phytoestrogens or a high soy diet, to drugs reducing testosterone, like spironolactone; these will have effects beyond the brain.

Wintertime raging
Having solved summertime raging, wintertime raging appeared. As expected, verapamil had no effect.
Ultimately the likely trigger was traced back to the very slow loss of milk teeth and eruption of permanent teeth. Both reabsorption of roots and the eruption new teeth is signaled using pro-inflammatory cytokines.
Moderate use of Ibuprofen, as and when behavior began to deteriorate, resolved the problem. Ibuprofen has no effect on summertime raging.

PANS-like episode aged 13
PANS (Pediatric Acute-onset Neuropsychiatric Syndrome) and PANDAS (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections) are infection-induced autoimmune conditions that disrupt a patient’s normal neurologic functioning, resulting in a sudden onset of Obsessive Compulsive Disorder (OCD) and/or tics and cognitive loss.
The import part is acute-onset; behavior changes overnight.
LT exhibits the classic traits of autism including stereotypy/stimming but never tics, which are a feature of Tourette’s-type autism.
Just before Christmas LT was recovering well from what presented as mild viral infection that had not warranted any medical intervention. He suddenly developed very loud verbal tics.
It is well known in PANS that delayed treatment severely affects prognosis. The sooner the patient is treated, the more complete recovery will be. Diagnosis is based on a very specific set of laboratory tests, only available in the US.
LT was treated from the third day of the tics as if he had PANS flare-up. He was treated with 40mg of prednisone for 5 days, requiring no taper.
Over a two week period the tics faded away. There have been no more tics.

Use of antioxidants
A recurring feature in autism research is oxidative stress. Two clinical trials have shown the benefit of the antioxidant NAC (N-acetylcysteine) in autism.
In LT the effect of NAC is the immediate disappearance of stereotypy and a type of anxiety. Without NAC, LT always wants to know what is happening next, to the point of obsession.
Oxidative stress has been shown to vary throughout the day and LT’s therapy is tailored to match it. Oxidative stress causes a cascade of further disruptions and causes many of the side effects of type-1 diabetes, for example.
LT takes 2,400 mg of NAC per day (a dose slightly lower than in the clinical trials). He has 600mg immediate release NAC at 7am, 600mg sustained release at 7am and then 600 mg sustained release at 1pm and 5pm. 
There have been no side effects after more than 4 years. 

Anti-inflammatory
Numerous studies (e.g. Ashwood) show elevated pro-inflammatory cytokines and reduced  anti-inflammatory cytokines as a feature of autism; but specific subgroups exist. Activated microglia is another feature of autism, which also suggests chronic inflammation.
Numerous anti-inflammatory strategies have been researched.
Atorvastatin has potent anti-inflammatory effects that are very well studied. It also affects the autism/cancer proteins RAS, PTEN and BCL2.
RASopathies are associated with MR/ID and indeed autism. Mutations in PTEN generally cause loss of function in PTEN and are associated with macrocephaly, enlarged corpus callosum, MR/ID and autism. Loss of function of PTEN is also found in some cancers, for example prostate cancer.
Because autism is polygenic and hundreds of genes are over/under expressed, it is not necessary to have a mutation to have misexpression. The mutation is just the extreme case (be it Cav1.2 or PTEN).
The effect of Atorvastatin is visible from the first dose and fades away the next day if therapy is stopped. The effect is very specific, it releases cognitive inhibition; it is as if the person with autism has the desire and capability to do something, but some barrier prevents him from doing it.
In broader severe autism, this is very important, Why does a child with autism who can verbalize never speak?
At the age of 9, LT was having piano lessons at home twice a week. He would practice the piano only if his assistant or father sat beside him. He never played independently.
After taking 10mg Atorvastatin for the first time, the next day LT went himself to his piano and started playing, without any prompting of any kind. He then began to practice on a daily basis.
As a child aged 3, LT had the habit of coming to the entry of the room with the television and watching from around the corner of the wall. He wanted to watch but could not enter the room. At the time it was thought he somehow just liked the visual sensation of peering around corners.
When he later moved to a multi-level house, LT would not come downstairs by himself; he would wait at the top of the stairs for someone to lead him down, every morning.  With atorvastatin not only did this behavior disappear, but it reappeared the day after Atorvastatin was withdrawn.
During one test withdrawal of the treatment, he got “stuck” in the kitchen and could not leave the room.

Sulforaphane Nrf2 and HDAC
In 2014, and again in 2017, Talalay/Zimmerman published research that sulforaphane from broccoli showed a benefit in autism. Sulforaphane is an HDAC inhibitor and thus has potential epigenetic properties, like some cancer drugs. Sulforaphane may also activate the Nrf2 redox “switch” and so be protective in conditions associated with oxidative stress.
LT’s father did contact the researchers and shortly after the first research was published LT started to take a broccoli sprout supplement. It did produce a very obvious effect and within 30 minutes; LT was laughing so much, be went to look at himself in the bathroom mirror. The more general effect was an unmissable increase in speech.
After three years of use the positive effect of sulforaphane/broccoli is no longer visible, even trying alternative brands.
In the 2017 clinical trial the authors found one responder retained the benefit of sulforaphane after the trial ended. They suggest an epigenetic switch may have been activated.  

Mitochondria and Microvasculature
A distinct type of autism has been characterized by Kelley at Johns Hopkins, Autism Secondary to Mitochondrial Disease (AMD). Kelley suggests that almost all regressive autism is caused by mitochondrial dysfunction and usually deficiency of the rate-limiting complex 1.
By stabilizing the mitochondria with antioxidants and then trying to stimulate more complex 1, a gradual improvement can occur.
Mitochondrial disease effectively starves the brain and body of energy (ATP), so lack of exercise endurance is exhibited in people with a genuine mitochondrial dysfunction.
One feature of autism is that growth factors (BNDF, IGF-1, NGF, VEGF etc) are disturbed, but the disturbance varies greatly by the type of autism.  Vascular endothelial growth factor (VEGF) in particular and its receptors are known to be disturbed and this has implications for microvasculature. Studies suggest that unstable, rather than reduced blood flow occurs in autistic brains.
In sports medicine, exercise endurance is a key target and it can be raised by improving the energy production from mitochondria and by improving the circulation of blood throughout the body by targeting eNOS (Endothelial Nitric Oxide Synthase) and NO (Nitric Oxide).
In Mild Cognitive Impairment (MCI) studies have shown the benefit of improved cerebral blood flow using cocoa flavanols to indirectly affect NO and hence improve memory.
Studies show that eNOS and NO can be safely increased by Agmatine and NO can be increased  using L-citrulline, which then produced more L-arginine. These supplements are widely used by sportsmen and women.
A small dose of Agmatine (1 g) has a near immediate substantial effect on LT, making him far more energetic.  It moved him from being rather passive physically, to being active. This has been very evident from his performance at school during physical activities, where it has been widely noted. At home LT started trampolining before breakfast and late in the evening.

Sensory Overload and Sensory Gating
An apparent over-sensitivity to sensory stimuli is a common observation in autism and is often the precursor to behavioral problems. In some younger children these can be trivial, but in more severe autism it can produce profound behavioral problems that never fade away.
Hypokalemic sensory overload and hypokalemic periodic paralysis are described in the literature. LT had sound sensitivity as a young child, in particular an inability to cope with the sound of crying. Tests were carried out to establish whether LT’s tolerance to the sound of crying improved after oral potassium. He consistently tolerated a high volume of a recording of this sound, when played 20 minutes after 250mg of potassium. Following ABA, he was purposefully exposed to this sound and taught to understand why people cry and modify his response, to the extent that his response changed to laughter, which again has to be modified towards empathy. 
Aged 10, LT developed a phobia to traveling in elevators/lifts. This was because the elevator he regularly used to visit his Grandparents was the old-fashioned type, with an internal sliding gate that you close by hand, which is extremely noisy.  He refused to use the elevator from that point on.  People with autism very easily form habits, or are allowed to form them, following the path of least resistance.  Elevators are a part of modern life and hard to avoid.
After a few weeks of this behavior, LT was given 500mg of potassium and half an hour later willingly entered the elevator and coped with the ride. The behavior has never recurred.
Sensory gating is another common issue in autism and schizophrenia, the individual is not able to filter out repetitive background sounds, like a clock ticking or the sound of a noisy eater. Sensory gating can be measured by looking at the P50 response on an EEG. α7 nicotinic acetylcholine receptor (α7 nAChR) agonists, like nicotine, can correct impaired P50 gating. A low dose of a PDE4 inhibitor is another suggested therapy
LT does exhibit was presents as impaired P50 gating. It is really only evident when his pharmacological therapy is halted for a few days. Then he finds all kinds of unavoidable noises very annoying, even the sound of a person sitting next to him eating. 

Typical Psychiatric Drugs
LT has never been treated with any of the usual antipsychotics, stimulants, anti-depressants, or anti-anxiety drugs sometimes prescribed in autism. His use of clonazepam is at a dose far below its standard clinical use.

Current status
In September 2017 LT moved to secondary/high school where some of the teachers recall how he used to be 10 years previously. Initially there was some trepidation and the view by some that a boy with classic autism should not be there. The school does have a boy with Asperger’s. However, LT surprised his new teachers, achieving grades placing him in the top half of his class. He is now extremely attentive in class, no attention deficit anymore, and has clearly not reached his intellectual limit. He has likely already far surpassed his intellectual limit, had he remained untreated.
As the end of the first year of high school approaches, LT continues to keep up academically with his peers. His agmatine-boosted physical performance has been maintained and he competes very well in long distance running and swimming.
LT is still intellectually far away from the trajectory followed by his older brother, but LT is keeping up academically with many of his classmates who are neurotypical, with average IQs.
A significant number of people diagnosed very young with autism do indeed make dramatic progress by the age of 6.  Zappella proposed his Dysmaturational Syndrome that he says applies to about 6% of early childhood autism, but they all have Tourette’s type autism (with tics).   There is an additional group without tics that also achieve what Fein calls Optimal Outcome, essentially they lose their autism diagnosis. In total it is 10-15% of cases that seem to “get better” all by themselves, regardless of intervention. As more diagnosis takes place even before 2 years of age and autism threshold grows ever wider, Optimal Outcome may become even more common.  
The definition of autism has been greatly watered down in recent years (DSM3 to DSM5). LT started with DSM3-type autism and by the age of 8 he still had it. DSM5 autism includes very much milder variants, some of which are trivial.
Each therapy used by LT has been found to be reversible based on careful withdrawal trials.



People with strictly defined autism (SDA) start to acquire skills with a delay compared to NT peers and thereafter acquire skills at a slower rate and hence fall ever further behind, making inclusion at school a delusion. The aim is to have similar skills to NT peers to make inclusion effective.
People with SDA often leave high school with an educational level of a 7 to 10 year old.

From the age of 12, LT ceased having any autism-specific learning curriculum; he just follows the curriculum of his mainstream school.  

Anecdotal Evidence
LT’s piano teacher exclusively teaches people with disabilities (mainly severe autism and a few with Asperger’s) and so has great experience of the disorder. She says while she has taught people who learnt to play as well as LT does today, this has never happened before with a child who started in his kind of condition at 8/9 years old.
The American ABA consultant (with Ph.D. and 20 years of experience) knowing LT from the age of 8, before he started bumetanide, told the family that of all her clients, LT is the one she sees the least but has improved the most and how strange that is. 


Current Therapy

The current therapy, called the Autism PolyPill, may be found in the link below.  

https://epiphanyasd.blogspot.com/p/polypill-for-autism.html

Autism is a highly heterogeneous condition, but there appear to be broad sub-types. At least some people with an autism diagnosis respond to each individual therapy in the PolyPill. Some people respond to almost the entire combination of therapies; other people respond to none.


Future Therapy

Some other interesting therapies remain to be investigated and it is clear that more improvement is possible because short term therapy with the flavones nobiletin and tangeretin produces a marked change in cognition and behaviour. The effect only lasts two or three days.  Tangeretin is a PPAR gamma agonist, among other properties. It reduces cholesterol when used long term, but its autism benefit is transient.  

The ketone Beta-Hydroxy Butyrate (BHB) also looks interesting; it has epigenetic properties amongst its other effects.