Suramin - Why do Clinical Trials in Autism Struggle to be Convincing? And Oxytocin fails in a large trial.
For me, Bumetanide for Autism is now ten-year-old news, for us it has been working since 2012; the next interesting drugs in the pipeline include Suramin and Leucovorin.
It is extremely difficult to trial Suramin at home, or indeed anywhere, and this makes it ever more desirable to many parents.
Leucovorin (calcium folinate) is easy to obtain; you can even buy liquid calcium folinate from iHerb. You can find out pretty quickly if it produces a profound benefit on your child’s type of autism.
I wish Dr Frye and Professor Ramaekers good luck with the phase 3 trial of Leucovorin. It certainly works for our adult reader Roger, but not for my 18 year old son, Monty. Our reader SB’s child recently joined the group of confirmed responders.
After I started writing this post, the results came in of a large (250 children) trial of intranasal oxytocin. This trial failed to show any benefit, over the placebo, in increasing social behaviors in autistic children. As I have mentioned previously, there is an inherent problem with intranasal oxytocin, the hormone has a very short action, its half-life is 2-6 minutes. It would be much more effective to provide a sustained release of oxytocin, which can indeed be achieved via adding a specific bacterium to the gut. The other problem with intranasal delivery is that you are not supposed to inhale the drug into your lungs, it has to stay in upper part of your nose. How likely is it that parents/children use the spray correctly? There is even a special dispenser developed for drug delivery to the brain, but did they use it?
In my trials of L. reuteri DSM 17938 it was obvious that the oxytocin improved social behaviors, but I concluded that this was not such a big deal and certainly was not a treatment priority. How would you assess the effect? Very simple, you just count how many times your child is shaking boys’ hands and kissing the girls. I don’t suppose that was the measurement that Duke University used.
Many parents do use Syntocinon nasal spray and this failed trial does not mean they are imagining the effects. If I was them, I would try L. reuteri DSM 17938 and compare the effect and use whichever is the most beneficial.
Suramin
Suramin is moving towards its Phase 3 clinical trials and, very unusually, two different companies are trying to commercialize the same drug. One company is PaxMedica and the other is Kuzani, who are ones that cooperate with Dr Naviaux.
In the background is Bayer, the German giant, who have been making Suramin for a hundred years as a therapy for African sleeping sickness and river blindness. We are told that making Suramin is quite difficult, it is a large molecule; but if they could make it a century ago, how difficult can it really be? The reality appears to be that Bayer do not want to supply PaxMedica or Kuzani and so they will have to figure out how to make it. Suramin is sold as a research chemical, but there seem to be questions about its purity. The very cheap Suramin sold on the internet is very likely to be fake.
Today we will look at the data from the South African trial carried out by PaxMedica and take a look at their patent for their intranasal formulation.
We have heard very positive anecdotal reports from the very small initial trials carried out by Professor Naviaux. Naviaux himself is very interesting, because even though he is not an autism researcher, he is far more knowledgeable than almost all of them on the subject of autism. If you read his papers, they show a rare global understanding of the subject. This “big picture” is what you need to understand such a heterogenous condition as autism.
In the PaxMedica trial, 44 children completed the trial, so that should be enough to tell us something insightful about whether this drug is effective.
A recurring problem in all autism trials is how well the placebo performs. Here again in the Paxmedica data we have a very impressive blue line – the placebo. It is just salt and water and yet it is nearly as good as the trial drug (the orange line).
A big part of clinical trials is the statistics used to validate them.
Although I do have a mathematical background, I believe in “seeing is believing”. The data should be crying out to you what it means. If it is so nuanced that it needs a statistician to prove the effect, there likely is no effect.
In the above chart we want to see a decreasing slope that would possibly level off as the drug achieved its maximum effect.
What we see are two apparently effective therapies, blue and orange.
The problem is that blue line is just water, with a bit of salt.
Show me the data
What we really want to see are results of each of the 44 participants, not the average.
There are likely groups:
· Super responders
· Responders
· Partial responders
· Non-responders
No statistician is needed.
The data from the Suramin trial needs to be presented in the kind of form used in the stem cell trial below:-
Since many hundreds of different biological conditions can lead to an autism diagnosis, we really should not expect there to be any unifying therapy that works for everyone. Indeed, we should perhaps be suspicious of any therapy claimed to work for everyone.
We always get to hear about the super-responders in anecdotal reports.
We heard great things about Memantine/Namenda, but the phase 3 trial was a failure. We heard great things about Arbaclofen (R-Baclofen), but the phase 3 trial failed. In Romania our reader Dragos is currently seeing great benefits from the standard version of Baclofen (a mixture of R-Baclofen and S-Baclofen).
My son is a super-responder to Bumetanide, but I know that most people are not. However, when I came across the “bumetanide has stopped” working phenomena, it became clear that the situation is more complex than a single one-time evaluation. We know why bumetanide can “stop working” and how to make it “start working again”. An increase in inflammatory cytokines from the periphery (i.e. outside the brain) further increases the expression of NKCC1 in the brain and negates the effect of bumetanide; reduce the inflammation and bumetanide will start to work again.
Why does the placebo always do well in autism trials?
The assessments used to measure outcome are all observational, they are not blood tests or MRI scans. They are highly subjective.
It has been suggested that just being in an autism trial improves symptoms of autism. The parents give more attention to the child and this then skews the results.
My way round this problem in my n=1 trials was always to tell nobody about the new trial I was making and wait for unprompted feedback. This works really well.
Who chooses the trial goal (the primary endpoint)?
I like the fact that in the Leucovorin trial the goal is speech. It is a very simple target and relatively easy to measure.
For Bumetanide, I did suggest to the researchers that they used change in IQ as an endpoint. Nice and simple, start with kids with IQ<70 and then recruit those who have a negative reaction (paradoxical response) to Valium/diazepam. Then expect an increase in measured IQ of 10 to 40 points. Then you would have a successful phase 3 trial.
In many previous trials that ultimately failed, some people did see a benefit, but they were different benefits. I did get a reader telling me how great Memantine (Namenda) had been for her child, when I asked why she told me that it was the only therapy that had ever solved her child GI problems. That certainly was never considered as a trial goal/endpoint.
In my trial of Pioglitazone, I read the research about both the mechanism of action and the observed effects listed in the phase 2 trial:
"improvement was observed in social withdrawal, repetitive behaviors, and externalizing behaviors as measured by the Aberrant Behavior Checklist (ABC), Child Yale-Brown Obsessive Compulsive Scale (CY-BOCS), and Repetitive Behavior Scale–Revised (RBS-R)."
I was targeting something entirely different. Based on the mechanism of action, specifically the reduction of the inflammatory cytokine IL-6, I expected a reduction in summertime raging. It worked exactly as hoped for. This is the second summer we have used it.
Our reader Sara’s initial assessment of the effect of Pioglitazone is focused on the improvement in sleeping patterns. This is great, assuming the benefit is maintained, but it is an entirely different benefit.
Was the trial drug actually taken?
I suspect in the bumetanide trial, many parents did not give the trial drug every day, as per their instructions, because the diuresis was too much bother. I know from reader comments and emails that many parents stop giving bumetanide, even though their child is a responder. Some schools refuse to allow bumetanide because of the disruption caused by frequent toilet breaks.
Because Suramin is given once a month by infusion, there is 100% certainty that the drug or placebo was actually taken. This is a big plus.
Was the intranasal oxytocin correctly administered in the recent trial? I doubt it.
The problem with Leucovorin is that in a minority of children is causes aggression, even if you follow Prof Ramaeker’s advice and very slowly increase the dosage. In the phase 3 trial parents should be informed of this possibility and told to report it and be invited to withdraw from the trial. If they just stop the therapy to halt the aggression, but their data remains included in the study, the results are invalidated.
Intranasal Suramin
Patents are often a good source of information and they do also tell you something about the people who wrote them.
Here below is PaxMedica's patent for intranasal suramin:-
Compositions and methods for treating central nervous system disorders
These results demonstrate that an antipurinergic agent such as suramin can be delivered intranasally to achieve plasma and brain tissue levels and that variations in the brain tissue to plasma partitioning ratio can be observed. These results demonstrate that an antipurinergic agent such as suramin can be delivered to the brain of a mammal by intranasal (IN) administration.
The following Table 1 provides the averaged accumulated amount, in mg, of suramin that has penetrated as a function of time
But how can the accumulated level after 6 hours be less than after 5 hours?
The results of the study are also shown graphically in FIG. 1 where the cumulative amount (mg) of drug permeated was plotted versus time in hours. These data demonstrate that Formulation B containing methyl β-cyclodextrin (methyl betadex) provides significantly better penetration, versus Formulations, A , C, and D in the tissue permeation assay. Also, as is seen from a comparison of Formulations A and D, having a higher drug concentration can be advantageous to increasing permeation.
Formulation A - suramin hexa-sodium salt at 100 mg/mL in water (no excipients) Formulation B - suramin hexa-sodium salt at 100 mg/mL in water, with 40% methyl β-cyclodextrin (methyl betadex)Formulation C - suramin hexa-sodium salt at 100 mg/mL in water, with 40% HP (hydroxyl propyl) -cyclodextrin Formulation D - suramin hexa-sodium salt at 160 mg/mL in water (no excipients)
FIG. 7 shows a plot comparing the total percentage of suramin in plasma in mice when administered by intraperitoneal (IP) injection once weekly for 4 weeks (28 days), intranasally (IN) daily for 28 days, intranasally (IN) every other day for 28 days, and intranasally (IN) once per week for 4 weeks (28 days).
FIG. 8 shows a plot comparing the total percentage of suramin in brain tissue in mice when administered by intraperitoneal (IP) injection once weekly for 4 weeks (28 days), intranasally (IN) daily for 28 days, intranasally (IN) every other day for 28 days, and intranasally (IN) once per week for 4 weeks (28 days).
Does anyone think the above chart makes any sense?
The mice were maintained in group cages (6 mice per cage based on treatment group) in a controlled environment (temperature: 2 1.5 ± 4.5 °C and relative humidity: 35-55%) under a standard 12-hour light/1 2-hour dark lighting cycle (lights on at 06:00). Mice were accommodated to the research facility for approximately a week. Body weights of all mice were recorded for health monitoring purposes.
The mice were divided into the following 5 test groups, with 6 mice per group.
Group 1: Intraperitoneal (IP) injection of suramin, 20 mg/kg, administered weekly to animals beginning at 9 weeks of age and continuing for four weeks (i.e. given at Age Weeks 9 , 10 , 11 and 12). The suramin was formulated in Normal saline solution.
Group 2 : Intraperitoneal (IP) injection of saline, 5 mL/g, administered weekly to animals beginning at 9 weeks of age and continuing for four weeks (i.e. given at Age Weeks 9 , 10 , 11 and 12). This was a control group.
Group 3 : Intranasal (IN) administration of a formulation, described below, of suramin, at a concentration of 100 mg/mL x 6 mL per spray, administered as one spray per nostril, one time per day, (interval of each application is around 2 minutes to ensure absorption) for 28 days (total of 56 sprays over 28 day period) beginning at 9 weeks of age (i.e. given daily during Age Weeks 9 , 10 , 11 and 12).
Group 4 : Intranasal (IN) administration of a formulation, described below, of suramin, at a concentration of 100 mg/mL x 6 mL per spray, administered as one spray per nostril, one time every other day, for 28 days (total of 28 sprays over 28 day period) beginning at 9 weeks of age (i.e. given once every other day during Age Weeks 9 , 10, 11 and 12).
Group 5 : Intranasal (IN) administration of a formulation, described below, of suramin, at a concentration of 100 mg/mL x 6 ml_ per spray, administered as one spray per nostril, one time every week, for 4 weeks (28 days) (total of 8 sprays over 28 day period) beginning at 9 weeks of age (i.e. given once weekly during Age Weeks 9 , 10 , 11 and 12).
This question was posed to me:-
A nasal spray in a human is about 0.1 ml, how do you give a tiny mouse 6 ml per nostril? Even 0.6 ml looks implausible.
Conclusion
Will Suramin pass a phase 3 trial? I think if it is trialed on a random group of 400 young people with moderate or severe autism, it will very likely fail.
Professor Naviaux believes Suramin may be a unifying therapy, one that works in all autism. The results from the PaxMedica study do not support this.
PaxMedica has the data showing the individual results. Are there super-responders? Are there non-responders? Does Suramin perhaps make some people's autism worse? All we can see is the average response, which is marginally better than the placebo; not what we expected after seeing the initial study.
Expecting Suramin to work well for everyone is raising the bar too high. Try and identify markers for the responders and super-responders and then limit the phase 3 trial to these people.
Is intranasal delivery of Suramin going to achieve a therapeutic level inside the human brain? Hopefully yes, but it may not work.
Is long term use of Suramin going to be safe? Will it require ever-increasing doses? Nobody knows, and note that safety was the original concern when Suramin’s use was proposed by Naviaux.
Intranasal administration has the best chance of being totally safe. Spend a little extra money on the clever dispenser covered in this old post, that keeps 100% of the drug in the right place.
https://epiphanyasd.blogspot.com/2015/09/opn-300-oxytocin-and-autism.html
Maybe get someone other than a lawyer, to proof read your patent.
Comments
Post a Comment