Bumetanide - Biomarkers from Shanghai for Autism responders

 

“three cytokine levels, namely the IFN-γ, MIG and IFN-α2  … These cytokine levels at the baseline could improve the prediction of the bumetanide responders”

“… cytokines had a potential to construct a blood signature for predicting and monitoring the bumetanide treatment in young children with ASD.”

“a significant part of the clinical heterogeneity in the treatment effect of bumetanide for ASD is associated with the differences in the immune system of patients”

 

Autism is a very heterogeneous family of conditions and this is a big part of the reason why all clinical trials to date have failed.  Ideally, there would be a diagnostic test to identify which person will respond to which therapy.  Then you can have a successful clinical trial, because you are only including people likely to respond.

Researchers from China have just published their results that suggest that a blood test measuring three inflammatory markers can predict who will respond to bumetanide.  This is good news and where it is coming from is also very notable.

Autism research has been very fragmented, some of it is very sophisticated and insightful but much is very amateur and some is quite trivial.  There is usually a real lack of common sense among these people and no sense of urgency whatsoever.

China is a very organised country; plans are made and then they are implemented.  Forget political correctness.

This kind of approach is what is required to move along with autism treatment.

In addition, there is also another new study from China, this time on the microbiota in autism that compared those with and without GI problems (it found it is equally disturbed in both groups). Hopefully, that Chinese group will do the next common sense step and compare the microbiota of autistic people with and without restrictive diets. To what extent to people give themselves a microbiota problem through poor diet.

 

Disentangling the relationship of gut microbiota, functional gastrointestinal disorders and autism: a case–control study on prepubertal Chinese boys

The altered gastrointestinal microbiota composition in ASD appeared to be independent of comorbid functional gastrointestinal disorder

 

The bumetanide researchers are from Fudan University in Shanghai, one of the 3 ultra-selective Chinese Universities alongside Tsinghua University and Peking University in Beijing.

The paper, not surprisingly, may look complicated, but there are a great deal of interesting things in it.

In their words:-

An immuno-behavioural covariation was identified between symptom improvements in the Childhood Autism Rating Scale (CARS) and the cytokine changes among interferon (IFN)-γ, monokine induced by gamma interferon and IFN-α2. Using this covariation, three groups with distinct response patterns to bumetanide were detected

The three groups were: best responders, least responders and medium responders.

It should be noted that the dosage used in their trials was 0.5mg of bumetanide twice a day.

Chinese children tend to be smaller than Western children and this might help explain why the results were more positive than in Servier’s failed phase 3 clinical trial in Europe. I also imagine the Chinese children were more severely autistic than the European group.

The dosage used is selected to minimize the diuresis rather than to maximize the impact on the autism. This is understandable, but I think it is a mistake.

 

The immuno-behavioural covariation associated with the treatment response to bumetanide in young children with autism spectrum disorder 

Bumetanide, a drug being studied in autism spectrum disorder (ASD) may act to restore gamma-aminobutyric acid (GABA) function, which may be modulated by the immune system. However, the interaction between bumetanide and the immune system remains unclear. Seventy-nine children with ASD were analysed from a longitudinal sample for a 3-month treatment of bumetanide. The covariation between symptom improvements and cytokine changes was calculated and validated by sparse canonical correlation analysis. Response patterns to bumetanide were revealed by clustering analysis. Five classifiers were used to test whether including the baseline information of cytokines could improve the prediction of the response patterns using an independent test sample. An immuno-behavioural covariation was identified between symptom improvements in the Childhood Autism Rating Scale (CARS) and the cytokine changes among interferon (IFN)-γ, monokine induced by gamma interferon and IFN-α2. Using this covariation, three groups with distinct response patterns to bumetanide were detected, including the best (21.5%, n = 17; Hedge’s g of improvement in CARS = 2.16), the least (22.8%, n = 18; g = 1.02) and the medium (55.7%, n = 44; g = 1.42) responding groups. Including the cytokine levels significantly improved the prediction of the best responding group before treatment (the best area under the curve, AUC = 0.832) compared with the model without the cytokine levels (95% confidence interval of the improvement in AUC was [0.287, 0.319]). Cytokine measurements can help in identifying possible responders to bumetanide in ASD children, suggesting that immune responses may interact with the mechanism of action of bumetanide to enhance the GABA function in ASD.

 

The use of bumetanide as a potential drug to improve symptoms in ASD is based on a hypothesised pathoetiology of ASD, namely the delayed developmental switch of the gamma-aminobutyric acid (GABA) functioning from excitatory to inhibitory [10,11,12]. In the valproate and fragile X rodent models of autism, this GABA-switch can be facilitated by the reduction of intracellular chloride concentration, which is mediated by a sequential expression of the main chloride transporters, such as the potassium (K)-Cl co-transporters 2 (KCC2) and the importer Na-K-Cl cotransporter 1 (NKCC1) [12]. Therefore, bumetanide as an NKCC1 inhibitor has been tested for its ability to restore GABA function in ASD [5,6,7, 13, 14]. However, these transporters can also be influenced by other molecules, such as cytokines, which are a number of small cell-signalling proteins closely interacting with each other to modulate the immune reactions. The cytokines have been implicated not only in brain development [15], but also in GABAergic transmission [16,17,18]. It has been reported that the interferon (IFN)-γ can decrease the levels of NKCC1 and the α-subunit of Na⁺-K⁺-ATPase, contributing to the restore of inhibitory GABA function [16]. In mice subjected to maternal deprivation, the interleukin (IL)-1 has also been found to reduce the expression of KCC2, delaying the developmental switch of the GABA function and thereby possibly contributing to the pathophysiology of developmental disorders such as ASD [17, 18]. Therefore, a question naturally arises that whether the treatment effect of bumetanide for ASD can be affected by the immune responses in the patients.

Indeed, compared with healthy controls, changes of the cytokine levels have already been reported in patients with ASD [19,20,21,22]. Recent meta-analyses showed that the levels of anti-inflammatory cytokines IL-10 and IL-1 receptor antagonist (Ra) were decreased [20], while proinflammatory cytokines IL-1β, IL-6 and anti-inflammatory cytokines IL-4, IL-13 were elevated in blood of patients with ASD [21]. The levels of IFN-γ, IL-6, tumour necrosis factor (TNF)-α, granulocyte-macrophage colony-stimulating factor (GM-CSF) and IL-8 were observed to be elevated [22] in postmortem brain tissues of ASD patients, and increased level of IFN-γ, monocyte chemotactic protein (MCP)-1, IL-8, leukaemia inhibitory factor (LIF) and interferon-gamma inducible protein (IP)-10 were found in another study [23]. These widely spread changes suggest that the cytokine signalling in ASD may be better characterised by multivariate patterns of cytokines. In literatures, many associations had been reported between the levels of cytokines (e.g., MCP-1, IL-1β, IL-4, IL-6, etc.) and both core symptoms and adaptive functions in children with ASD [24,25,26]. Therefore, it has been suggested that cytokines may be used as biomarkers to identify different subsets within ASD. In each of these subsets the patients with ASD may share a commonly immune-related pathoetiology and therefore may have similar profiles of response to treatment [27].

Based on these previous findings, we analysed data acquired through the Shanghai Xinhua ASD registry, China, that began in 2016 to test the hypothesis that the immune activity of patients might help to identify the best responders to bumetanide in ASD.

 

Between May 1st, 2018, to April 30th, 2019, a total of 90 ASD children, aged 3–10 years old, under a 3-month stable treatment of bumetanide without behavioural interventions and any concomitant psychoactive medications had both blood draws and behavioural assessments. Among these patients, 11 of them were further excluded due to the lack of the follow-up data at month 3. A group of 37 children, under 3-month stable treatment of placebo without behavioural interventions and any concomitant psychoactive medications had both blood draws and behavioural assessments. Therefore, the current analysis used a subsample of 116 young children with ASD, whose blood samples were available both before and after the treatment. The blood samples were sent in three batches (Discovery Set: n = 37 on December 4, 2019; Validation Set: n = 42 on May 22, 2019; and Control Set: n = 37 on January 5, 2022) to measure the serum levels of 48 cytokines for the immune response (Table S1), and the clinical symptoms were assessed using CARS, ADOS and the Social Responsiveness Scale (SRS). 

In this study, we observed a significant improvement of clinical symptoms with bumetanide treatment in children with ASD, and such improvement was associated with a pattern of changes in three cytokine levels, namely the IFN-γ, MIG and IFN-α2 (r = 0.459 in the Discovery Set and r = 0.316 in the Validation Set). These cytokine levels at the baseline could improve the prediction of the bumetanide responders compared with using the behavioural assessments alone, and the best predictor achieved an AUC of 0.83 in the independent test data set (Table S8). The implications of these findings may be twofold: (1) a significant part of the clinical heterogeneity in the treatment effect of bumetanide for ASD is associated with the differences in the immune system of patients, and (2) the component score of cytokines had a potential to construct a blood signature for predicting and monitoring the bumetanide treatment in young children with ASD.

Following the protocols of previous studies [8], bumetanide treatment consisted of two 0.5 mg tablets per day for three months, given at 8:00 a.m. and 4:00 p.m. The tablet size is 8 mm diameter x 2 mm thickness, which is quite small. Each time, the patient took half of a tablet, which was not difficult for most of the patients. However, the careers were recommended to grind the half-tablet into powder and give the powder in water, if necessary. Possible side effects were closely monitored during the treatment. Blood parameters (serum potassium and uric acid) were monitored via laboratory tests (Table S2) and symptoms (thirst, diuresis, nausea, vomiting, diarrhoea, constipation, rash, palpitation, headache, dizziness, shortness of breath, and any other self-reported symptoms) were telephone interviewed (Table S3), and both of them were reported to the research team by telephone at 1 week and 1 month after the initiation of treatment and at the end of the treatment period. The cytokine levels of the children with gastrointestinal problems were compared with those without such problems (Table S4).

The supplemental table S4 shows that GI problems had no effect on cytokine levels.

Changes after the administration of bumetanide

Seventy-nine patients were treated with bumetanide for 3 months, and the CARS total score decreased after the treatment (effect size Cohen’s d = 1.26, t₇₈ = 11.21, p < 0.001). The treatment effect showed no difference between the Discovery Set and the Validation Set (ΔCARS_total: mean(±SD): 1.54 (±1.40) vs. 1.90 (±1.34)). Consistent to the previous studies of the low-dose bumetanide for ASD, the side effects were rarely reported (Tables S2 and S3). No significant difference in the cytokine levels between the children with and without the gastrointestinal problems at the baseline (Table S4). A number of cytokine levels were changed significantly after the treatment of bumetanide, but none of them was changed significantly after the treatment of placebo (Table S6). No significant pairwise association could be identified in the Discovery Set, the Validation Set and the Control Set among four groups of variables, including the baseline CARS total score, the baseline cytokine levels, the change of CARS total score, and the changes of cytokine levels (Fig. S2).

 

In this study, we observed a significant improvement of clinical symptoms with bumetanide treatment in children with ASD, and such improvement was associated with a pattern of changes in three cytokine levels, namely the IFN-γ, MIG and IFN-α2 (r = 0.459 in the Discovery Set and r = 0.316 in the Validation Set). These cytokine levels at the baseline could improve the prediction of the bumetanide responders compared with using the behavioural assessments alone, and the best predictor achieved an AUC of 0.83 in the independent test data set (Table S8). The implications of these findings may be twofold: (1) a significant part of the clinical heterogeneity in the treatment effect of bumetanide for ASD is associated with the differences in the immune system of patients, and (2) the component score of cytokines had a potential to construct a blood signature for predicting and monitoring the bumetanide treatment in young children with ASD.

Accumulating evidences support that IFN-γ can inhibit chloride secretion [38] and down-regulate both the NKCC1 expression [16, 38] and the Na⁺-K⁺-ATPase expression [16], which had been implicated in the GABAergic dysfunction in ASD [10, 39].

 

The cytokine-symptom association was identified in the changes after the treatment of bumetanide but not before the treatment, suggesting that bumetanide might interact with the cytokines and the changes of which contributed to the treatment effect of bumetanide. Animal studies showed a rapid brain efflux of bumetanide, but a number of clinical trials have shown a significant treatment effect for neuropsychiatric disorders, including ASD, epilepsy and depression [41, 42]. These findings may suggest the possible systemic effects of bumetanide as a neuromodulator for these neuropsychiatric disorders. Considering its molecular structure, bumetanide has been recently identified by an in vitro screen of small molecules that can act as an anti-proinflammatory drug via interleukin inhibition [43]. This anti-proinflammatory activity of bumetanide might alter the blood levels of cytokines outside the brain-blood-barrier (BBB).

Our findings may suggest that the identified canonical score of cytokines had a potential to construct a blood signature for predicting and monitoring the bumetanide treatment in young children with ASD. Accurately identifying patients who are likely to respond positively to bumetanide can facilitate the precision medicine for ASD. Our prediction model based on the cytokine levels before the treatment may provide a potentially new tool for the precision medicine of ASD. 

 

In summary, we identified an association between the changes of the cytokine levels and the improvements in symptoms after the bumetanide treatment in young children with ASD, and found that the treatment effect of bumetanide can be better characterised by an immuno-behavioural covariation. This finding may provide new clinically important evidence supporting the hypothesis that immune responses may interact with the mechanism of bumetanide to restore the GABAergic function in ASD. This finding may also have relevance for determining enriched samples of ASD children to participate in novel drug treatment studies of drugs with a similar mode of action to bumetanide, but with potentially greater efficacy and fewer side effects.

 

Conclusion

I think we can give the Shanghai researchers 10 out of 10 for their paper.

Monty, aged 18 with ASD, has been to Shanghai twice. It is a vast city, but well worth a visit. With the high speed train network it is now very easy to travel around China, quite different to when I visited as a teenager.

Hopefully the Chinese will continue in their pursuit of precision medicine for autism. They do not have much competition.

My perspective is a little different because I know that a bumetanide responder can cease to be a responder when affected by an inflammatory condition like allergy, which increases pro-inflammatory cytokines like IL-6. This suggests that some people with elevated cytokines are potential responders, you just have to use an anti-inflammatory therapy before you start bumetanide therapy. The inflammatory cytokines shift the balance between NKCC1 and KCC2 towards NKCC1 and so increasing intracellular chloride.  We also know that some people need a dose higher than 0.5mg twice a day to see a large benefit; I have been using 2mg once a day for several years.

The Chinese researchers have established biomarkers for who is likely now to respond to bumetanide. This certainly is a big step forward, if it can be replicated. This is not the same as identifying who could respond to bumetanide, if their current inflammatory condition was moderated. The levels of specific cytokines might indeed mark someone as both a current non-responder, but also as a potential future responder.

Autism is all about n=1, it is about the exceptions being more important than the average.

Unlike the Shanghai researchers, I do not really see Bumetanide as an anti-inflammatory therapy in my son’s Polypill, but I do have therapies included that are.

Understanding inflammation will be a key to treating autism using precision medicine.  That is less simple that it sounds. When it comes to preventing autism, inflammation in the mother is a key part of the equation. This also gets complicated, maternal antibodies damage the brain of the fetus, no genetic mutations were needed.