ARTICLE IN BRIEF
Lower levels of the neuropeptide arginine vasopressin in the cerebrospinal fluid distinguished children with autism from those who did not have the condition and predicted symptom severity in social behavior in boys. Still, its precise role in autism spectrum disorder is still a question mark that calls for further exploration.
Using what is thought to be the largest collection of cerebrospinal fluid (CSF) samples to study neuropeptides from children with autism, researchers found that lower levels of the neuropeptide arginine vasopressin (AVP) in the CSF distinguished children with autism from those who did not have the condition. In addition, vasopressin predicted symptom severity in social behavior in boys, according to a paper published online on August 28 in Annals of Neurology.
The findings demonstrate the relevance of the vasopressin hormone in autism spectrum disorder (ASD), the study authors wrote. But they said its precise role in ASD is still a question mark that calls for further exploration.
Researchers who performed the CSF study are evaluating intranasal vasopressin for the treatment of social ability in children with autism, even as Roche works on an opposite approach — a vasopressin receptor-blocker for adults, for which it received breakthrough status from the US Food and Drug Administration (FDA) earlier this year, allowing it to proceed through the development process more quickly. (The data on the Roche-sponsored vasopressin receptor antagonist was first presented at the International Society for Autism Research in 2017.)
Karen J. Parker, PhD, associate professor of psychiatric and behavioral sciences at Stanford University, the lead author on the Annals of Neurology study, acknowledged that the conflicting findings are somewhat vexing. “It'll be very interesting to see where this heads, especially since I haven't seen preclinical data on their part that would mirror the work that we've done, or clinical data for that matter,” Dr. Parker said.
The findings on the predictive value of AVP in the CSF of children is a follow-up to a study by Dr. Parker and her colleagues published in May in Science Translational Medicine. In that study, they found that AVP in the CSF was a key marker of group differences in the social behavior of rhesus monkeys, findings that were then replicated in an independent monkey cohort. In that same paper, they showed that CSF levels of vasopressin were lower in a small cohort of children with autism than in age-matched controls.
Interest in the role of hormones in autism — especially oxytocin, which has been thought to have a more prominent role in females, and vasopressin, which has been thought to have a more prominent role in males — has grown in recent years, but the research has been hindered by the difficulty of obtaining CSF samples to analyze.
For the current study, researchers from Stanford and the University of California, San Francisco (UCSF) collaborated with the National Institute of Mental Health, which provided CSF samples from children with autism, whose parents or guardians had granted permission for lumbar puncture. Samples from children seen at Stanford and UCSF, who required lumbar puncture for a medical condition other than autism, were used as age- and sex-matched controls. Levels of AVP and oxytocin were assessed in the CSF samples from a total of 72 children, age 1 to 9.
The average AVP concentration was significantly lower in the autism group compared to the control group (p=.0004), but no difference in oxytocin concentration was seen between the two groups. Researchers found no differences in AVP concentration levels by sex.
Lower CSF AVP concentrations predicted greater symptom severity in males, but not in females, researchers found. These differences were driven by social behavior, with lower AVP levels predicting more social impairment, as measured by the Social Affect portion of the Calibrated Severity Score (SA-CSS), in boys (p=.0063).
“We thought that it was possible that we would see that the girls with autism would have lower CSF oxytocin levels, and the boys would have lower CSF vasopressin levels,” Dr. Parker said. “But what we in fact saw was that regardless of whether you were male or female — in general on the group level but also when we did individual predictions — we saw that it was the vasopressin, not the oxytocin, that was lower.”
Dr. Parker said that the findings elevate the importance of CSF vasopressin in ASD, but the roles of oxytocin and vasopressin need to be clarified further. “I think that the picture of what each molecule does isn't very well understood,” she said.
The accuracy of low CSF AVP levels in predicting autism in the study, researchers noted, was “robust” but “imperfect,” with eight children with autism and nine controls misclassified.
Although this particular study suggests that vasopressin replacement, and not oxytocin replacement, might be a therapeutic option, Dr. Parker said there is still likely a role for oxytocin replacement in some patients. Indeed, patients with certain syndromic forms of autism — including Prader-Willi syndrome and fragile X syndrome — have been found to have very low levels of oxytocin, and mouse models treated with oxytocin have been found to improve in social functioning.
“Autism is a very heterogeneous disorder and so I think the conventional wisdom by researchers in the field is that there's probably lots of different causes for the behavioral symptoms that we see,” Dr. Parker told Neurology Today. “And so, could it be that there are some children that have oxytocin deficits that can benefit from oxytocin treatments? I think that's very likely.”
Dr. Parker said she couldn't discuss her findings from her vasopressin treatment trial in children yet, but a conference abstract from a 2017 scientific meeting presentation of her preliminary data found that it was well-tolerated and improved social abilities.
Roche's findings that led to the FDA breakthrough status suggest there may be people with autism who have high vasopressin levels, she said, even though that runs counter to her group's pre-clinical and clinical data and “all of the work that we've transparently published.”
“I think that's something that sort of remains to be seen,” she said.
Jeremy Veenstra-VanderWeele, MD, the Ruane Professor for the implementation of science for child and adolescent mental health at Columbia University Medical Center, said the very acquisition of the data makes the CSF paper interesting.
“It is incredibly difficult to do this sort of work, and it is impressive that they were able to get these data,” he said.
Dr. Veenstra-VanderWeele cautioned that the findings don't necessarily indicate that vasopressin is related to the cause of ASD, although that could be the case.
“These data could also potentially indicate that lower vasopressin levels are a result of lower social interaction due to autism spectrum disorder or might even be a compensatory mechanism in ASD,” he said. “It would be necessary to manipulate the vasopressin system in some way in order to evaluate whether it has a causal role in social function in ASD. Regardless, it suggests that the central vasopressin system is relevant to ASD, at least in males.”
As for the two therapeutic paths being pursued, he said the CSF data don't preclude either one.
“I do not think that these data necessarily mean that increasing vasopressin would be beneficial in ASD. Likewise, I do not think that they necessarily mean that an antagonist at a specific vasopressin receptor, such as Roche is studying, would be a bad idea.”
He added: “The data do suggest, however, that manipulations of the vasopressin system — with medications or otherwise — would be a logical next step to understand vasopressin's specific role in autism and evaluate this system as a potential avenue for treatment.”
Michael L. Platt, PhD, professor of psychology and neuroscience at the University of Pennsylvania, who has studied vasopressin's role in the social behavior of rhesus monkeys, said that up until now the field has been limited to peripheral markers, and that because of the hormones' varied roles in the body, that approach tends to lead to a “mixed bag of results.”
The CSF findings on vasopressin help to bring a bit more clarity to vasopressin's role, Dr. Platt said. “It's starting, at least, to come together to tell a sensible story.”
He said the findings raise the possibility of using CSF screens to identify children who could benefit from therapy, he said. “The kids with the lowest levels those would be the ones you'd predict that if you tried to boost their levels — at least get them higher — that you might have more of a behavioral response,” he said.
Reconciling the twin pursuits of both raising and lowering vasopressin for social behavior benefit effects is a challenge, he said, but the conflicting findings suggest that there could be “a kind of optimum bandwidth, if you will, for the circuitry to function appropriately.”
“Really what we'd like to see, as scientists, is to understand the underlying mechanism, and that, of course, is very challenging to do in humans,” he said. “I would plug non-human primate models where the biology and the behavior is much more highly shared with humans. That model system can give us greater insights into the mechanism of action and the efficacy, and potentially the safety, compared to what you can achieve in a small animal model like a mouse or rat.”