This blog is seen by many as being bumetanide-inspired, so it is only fair to highlight the recently published Bumetanide for autism Phase IIb results. This study probably will not get the attention it should get in the media, in part because it is French and not American.
Effects of bumetanide on neurobehavioral function in children and adolescents with autism spectrum disorders
Nonetheless, there are some interesting things in the study, most importantly is that you can see the guiding hand of the drug approving agency (the European Medicines Agency). They have specified how to measure the results (the CARS rating scale), the level of severity of autism to test the drug on (severe) and the age group (2-18 years old).
In this blog we know that autism is very heterogeneous and that while bumetanide can be highly effective, there is a large group that do not respond. This trial did not use any kind of biomarker and so it contains a broad mix of responders and non-responders.
The good news is that even though only about 30-50% of people with an autism diagnosis seem to be responders, this was more than enough to make the drug look effective in a trial of just 88 people.
The small sample size does throw up anomalies, like giving the impression that 0.5mg twice a day is more effective than 1mg twice a day, on some measures.
The most effective treatment, judged by the authors, was 2mg twice a day, but it was ruled out for the Phase III trial because of side effects.
“Clearly this trial must be viewed as a source of data on the safety and dose-ranging usage of bumetanide and it provides further support to justify a large multisite European Phase III trial.
We report here the results of a multicenter phase II dose-ranging study conducted according to a pediatric investigation plan approved by the EMA to determine the optimum bumetanide dose in ASD children and adolescents aged 2–18 years. To achieve these aims, the study population was divided into four subgroups receiving three doses of bumetanide (0.5, 1.0, 2.0 mg twice a day) or placebo. The distribution of age was the same in the four groups (2–4, 5–9, 10–13, 14–18 years old). The results of this trial confirm and extend our earlier observation that bumetanide improves the symptoms of ASD and that too in the full age range targeted.
EMA imposed the age range from 2 to 18.
CARS has been selected by the EMA in their recent (2016) guidelines for evaluation of ASD core symptoms (Guidelines on the clinical development of medicinal products for the treatment of ASD, 25 December 2015).
The mean initial CARS scores were similar across all treatment groups in the six clinical centers and were above the cut-off for severe autism as required by EMA (>34). “
In the CARS scale, a score greater than 30 is the threshold for autism. The higher the score, the more severe the autism. The regulator required that people in the trial had a CARS score greater than 34 to be eligible for the trial. The average CARS score for those in the trial was 41, so you would need an average improvement of 11 to potentially make them all “better” (this is a simplification). This just shows the meaning of improving by 8 CARS points. It is a big deal.
Here you see what happens when half a study group are not responders to bumetanide and you see a large group who either show minor improvement or get worse. This is just the expected background noise in all autism trials. There is even one person who improved greatly (>8 on the CARS scale) just by taking the placebo.
Change in the completers of the total CARS score from screening to day 90 after bumetanide (blue bars; n=52) and placebo (orange bars; n=21). All changes were calculated from the initial values for each individual participant at screening. Note a significant amelioration of the CARS scale after the treatment period (>4) is almost entirely restricted to the bumetanide-treated patients (only placebo). CARS, Childhood Autism Rating Scale.
As we have seen in feedback on this blog the harmful side effects of bumetanide are dehydration and low levels of potassium. These side effects are entirely manageable, but that job is for the parents.
I was interested in the chart below which measures the potassium levels during the trial of the four groups. I assume green (placebo), blue (0.5mg), red (1.0) mg, black (2.0mg)
Points perhaps unbeknown to the authors
An important point we have seen on this blog is that intracellular levels of chloride vary under the influence of inflammation (which affects KCC2 expression). This results in some people responding well to bumetanide at some times and apparently not at others.
We also noted that bumetanide does not cross the blood brain barrier very well and so in some people with high levels of intracellular chloride they may appear not to respond to bumetanide, not because they do not have elevated Cl-, but because it is just too high for bumetanide to show effect.
Taken together the point is that if at any one time say 40% of people with autism are responders, there may be another x% who are potential responders to a similar but more potent therapy.
If you trial bumetanide in the summer and have a pollen allergy, it may appear not to work, as in the case of my son. Fortunately I did my trial in the winter.
Dr Ben-Ari told Medscape Medical News that the phase 3 study, which is approved by the European authorities, will be performed in about 400 children in five EU countries. The patients will receive 1 year of treatment, and they will reflect the entire pediatric autism population.
"We hope to get the drug on the market in Europe for autism by the end of 2021," he said.
As some wise older person told me many years ago, "things take time". In the world of autism, things seem to take forever.