Brain Structure Disrupted in Autistic Kids
Medpage Today: The Gupta Guide
Published: Mar 26, 2014 | Updated: Mar 27, 2014
Autopsied brains of autistic children showed "patches of disorganization" not seen in those from non-autistic children, a small study indicated.
Among 11 children diagnosed with autism who had died unexpectedly (most by drowning), 10 had "focal disruption of cortical laminar architecture," whereas 10 of 11 non-autistic children who also had died suddenly did not, according to Eric Courchesne, PhD, of the University of California San Diego, and colleagues.
The specific locations of these patches varied among the autistic children, but in general they corresponded to "regions mediating the functions that are disturbed in autism: social, emotional, communication, and language functions," the researchers wrote in the March 27 issue of the New England Journal of Medicine.
"Such abnormalities may represent a common set of developmental neuropathological features that underlie autism and probably result from dysregulation of layer formation and layer-specific neuronal differentiation at prenatal developmental stages," Courchesne and colleagues concluded.
Jonathan Mink, MD, PhD, of the University of Rochester in New York, who was not involved with the study, told MedPage Today that the findings could eventually lead to better understanding of the specific structural and neurological abnormalities that produce the symptoms associated with autism.
Although the results need to be replicated in a larger sample, he said, it opens the door to a new line of research.
"This is certainly going to raise a lot of questions -- not because we're doubting the results, but because it points to a whole new way of thinking about brain organization in autism."
Study Details
At the study's heart were in situ hybridization analyses of gene expression in brain tissue samples, focusing on a large set of genetic markers that Courchesne and colleagues had determined in previous research to be associated with brain function. These were used to identify specific types of excitatory and inhibitory neurons as well as other cell types such as microglia and oligodendrocytes. The markers also included candidate autism-related gene variants.
The autistic children ranged in age from 2 to 15. All but two had died by drowning; one of the others fell from a ninth-floor dwelling and the other died from a rapidly progressing cancer. Causes of death were more variable among the 11 controls, whose ages ranged from 4 to 15: two died in vehicle accidents, two suffered fatal asthma attacks, three had sudden cardiac deaths, two drowned, and two died of other causes. Only one of the controls had a neuropsychiatric diagnosis (attention deficit-hyperactivity disorder).
On the other hand, the autistic children varied considerably in their disabilities before death. Four of the 11 were not considered intellectually disabled. Scores for the social domains in the revised Autism Diagnostic Interview ranged from 14 to 26; for communication, scores ranged from from 9 to 20; and for restrictive/repetitive behaviors, scores ranged from 3 to 10. Eight of the 11 were boys.
Marker patterns in the control children were mostly similar to those seen in normal adults in earlier research, Courchesne and colleagues reported. In contrast, 10 of the 11 autistic children had well-defined areas of reduced marker expression or showing "unusual patterns." These patches spanned 5 to 7 mm and were like islands in otherwise normal-appearing cortical tissue.
"No two patches were identical in presentation" between autistic children, the researchers wrote. "We observed heterogeneity between case samples with respect to the layers and the cell types that were most abnormal." However, one consistent finding was the neuron density was not reduced in the autistic children relative to controls -- in fact, it was slightly but significantly increased within the patches.
The implications of these results for etiology and treatment remain to be worked out, Courchesne and colleagues indicated. "Although our data suggest a novel pathological mechanism in autism, they do not identify the mechanism," they wrote.
Possibilities include defects in migration during neurodevelopment, such that cells fail to reach their proper destinations and instead accumulate elsewhere, the researchers suggested. Alternatively, the cells themselves may fail to develop normally in the patch regions.
But the variability in patch geography seen among the autistic children in the study is encouraging in one respect, the researchers argued: it could explain the heterogeneity in clinical manifestations.
"Disorganized patches in different locations could disrupt disparate functional systems in the prefrontal and temporal cortexes and potentially influence symptom expression, response to treatment, and clinical outcome," Courchesne and colleagues reasoned.
The fact that one of the control children had some disorganized patches is not necessarily inconsistent with this model, they noted: it could be indicative of a "subclinical patch phenotype." Likewise, the failure to identify disorganized patches in one of the autistic children could simply mean the researchers didn't look in the right place or at the right markers.
Strengths and Limitations
Mink told MedPage Today that these are plausible conclusions. Eventually, he said, the study could lead to new ways to evaluate children with autism spectrum disorders.
"It may be that the specific parts of the brain where the abnormality is may be the best predictor of the types of [clinical] abnormalities they'll have," he said.
Other researchers contacted by MedPage Today also found the study very important.
Sophie Molholm, PhD, of Albert Einstein College of Medicine in New York City, noted that most of the patches were in cortical layers IV and V, which correspond to sensory processing and autonomic processing, respectively. At the same time, "these patches were not found in the primary visual cortex," she said in an email. "This is particularly important because it indicates that these abnormalities are not widespread in the brain in autism, but rather are focused pathology."
The findings also tend to reinforce the notion that autism is "prenatal in origin," said Max Wiznitzer, MD, of UH Hospitals and UH Rainbow Babies and Children's Hospital in Cleveland.
The disorganization might either "directly cause the communication disturbances between parts of the brain," he said, "or [it] predisposes the child to another, yet to be determined trigger (if it exists)."
Although the study sample was small, these experts didn't think it was a major limitation.
"This was actually a good sample size for a postmortem study on this population," Molholm said in an email, "and the age range was as focused as it could be for the amount of cases available."
But other limitations were more concerning. Mink told MedPage Today that the brains available for autopsy -- both from autistic children and from normal controls -- may not be representative of their respective populations just because of the nature of the children's deaths and the parents' decisions to donate the organs for research.
Wiznitzer said it would have been useful to also include a control group with non-autistic neuropsychiatric diagnoses, to determine whether the disorganized patches are unique to autism spectrum disorders versus other types of behavioral abnormalities.
Curtis Deutsch, PhD, of the University of Massachusetts Medical School in Worcester, Mass., said he didn't think that was a serious problem. But the approach taken by Courchesne and colleagues limited the amount of data gathered in the study, he said.
"The authors used an exploratory design, one that employed sparse sampling of tissue across a wide area of whole brain. In other words, they cast a wide net in search of neuroanatomic abnormalities, and they found an abundance of neuropathologic 'patches,'" Deutsch wrote in an email.
"It will be of interest in the future to perform [higher resolution] analyses to investigate the extent and distribution of these pathologic features, and to study their statistical properties."
Similarly, Molholm said she wished the authors had been more specific about the patch locations and their correspondence in individual cases to behavioral abnormalities, if any.
But, she added, the faux-3D illustrations Courchesne and colleagues included in the NEJM paper were "very cool."
The study was funded by the Simons Foundation, the Peter Emch Family Foundation, Cure Autism Now/Autism Speaks, the Thursday Club Juniors, the Allen Institute for Brain Science, and the University of California San Diego.
One author reported equity ownership in Stereology Resource Center, which manufactured equipment used in the study. One author reported payments from Austism Speaks. Other authors declared they had no relevant financial interests other than research funding from the organizations listed above.
Primary source: New England Journal of Medicine
Source reference: Stoner R, et al "Patches of disorganization in the neocortex of children with autism" N Engl J Med 2014; 370: 1209-1219.
Medpage Today: The Gupta Guide
Published: Mar 26, 2014 | Updated: Mar 27, 2014
Autopsied brains of autistic children showed "patches of disorganization" not seen in those from non-autistic children, a small study indicated.
Among 11 children diagnosed with autism who had died unexpectedly (most by drowning), 10 had "focal disruption of cortical laminar architecture," whereas 10 of 11 non-autistic children who also had died suddenly did not, according to Eric Courchesne, PhD, of the University of California San Diego, and colleagues.
The specific locations of these patches varied among the autistic children, but in general they corresponded to "regions mediating the functions that are disturbed in autism: social, emotional, communication, and language functions," the researchers wrote in the March 27 issue of the New England Journal of Medicine.
"Such abnormalities may represent a common set of developmental neuropathological features that underlie autism and probably result from dysregulation of layer formation and layer-specific neuronal differentiation at prenatal developmental stages," Courchesne and colleagues concluded.
Jonathan Mink, MD, PhD, of the University of Rochester in New York, who was not involved with the study, told MedPage Today that the findings could eventually lead to better understanding of the specific structural and neurological abnormalities that produce the symptoms associated with autism.
Although the results need to be replicated in a larger sample, he said, it opens the door to a new line of research.
"This is certainly going to raise a lot of questions -- not because we're doubting the results, but because it points to a whole new way of thinking about brain organization in autism."
Study Details
At the study's heart were in situ hybridization analyses of gene expression in brain tissue samples, focusing on a large set of genetic markers that Courchesne and colleagues had determined in previous research to be associated with brain function. These were used to identify specific types of excitatory and inhibitory neurons as well as other cell types such as microglia and oligodendrocytes. The markers also included candidate autism-related gene variants.
The autistic children ranged in age from 2 to 15. All but two had died by drowning; one of the others fell from a ninth-floor dwelling and the other died from a rapidly progressing cancer. Causes of death were more variable among the 11 controls, whose ages ranged from 4 to 15: two died in vehicle accidents, two suffered fatal asthma attacks, three had sudden cardiac deaths, two drowned, and two died of other causes. Only one of the controls had a neuropsychiatric diagnosis (attention deficit-hyperactivity disorder).
On the other hand, the autistic children varied considerably in their disabilities before death. Four of the 11 were not considered intellectually disabled. Scores for the social domains in the revised Autism Diagnostic Interview ranged from 14 to 26; for communication, scores ranged from from 9 to 20; and for restrictive/repetitive behaviors, scores ranged from 3 to 10. Eight of the 11 were boys.
Marker patterns in the control children were mostly similar to those seen in normal adults in earlier research, Courchesne and colleagues reported. In contrast, 10 of the 11 autistic children had well-defined areas of reduced marker expression or showing "unusual patterns." These patches spanned 5 to 7 mm and were like islands in otherwise normal-appearing cortical tissue.
"No two patches were identical in presentation" between autistic children, the researchers wrote. "We observed heterogeneity between case samples with respect to the layers and the cell types that were most abnormal." However, one consistent finding was the neuron density was not reduced in the autistic children relative to controls -- in fact, it was slightly but significantly increased within the patches.
The implications of these results for etiology and treatment remain to be worked out, Courchesne and colleagues indicated. "Although our data suggest a novel pathological mechanism in autism, they do not identify the mechanism," they wrote.
Possibilities include defects in migration during neurodevelopment, such that cells fail to reach their proper destinations and instead accumulate elsewhere, the researchers suggested. Alternatively, the cells themselves may fail to develop normally in the patch regions.
But the variability in patch geography seen among the autistic children in the study is encouraging in one respect, the researchers argued: it could explain the heterogeneity in clinical manifestations.
"Disorganized patches in different locations could disrupt disparate functional systems in the prefrontal and temporal cortexes and potentially influence symptom expression, response to treatment, and clinical outcome," Courchesne and colleagues reasoned.
The fact that one of the control children had some disorganized patches is not necessarily inconsistent with this model, they noted: it could be indicative of a "subclinical patch phenotype." Likewise, the failure to identify disorganized patches in one of the autistic children could simply mean the researchers didn't look in the right place or at the right markers.
Strengths and Limitations
Mink told MedPage Today that these are plausible conclusions. Eventually, he said, the study could lead to new ways to evaluate children with autism spectrum disorders.
"It may be that the specific parts of the brain where the abnormality is may be the best predictor of the types of [clinical] abnormalities they'll have," he said.
Other researchers contacted by MedPage Today also found the study very important.
Sophie Molholm, PhD, of Albert Einstein College of Medicine in New York City, noted that most of the patches were in cortical layers IV and V, which correspond to sensory processing and autonomic processing, respectively. At the same time, "these patches were not found in the primary visual cortex," she said in an email. "This is particularly important because it indicates that these abnormalities are not widespread in the brain in autism, but rather are focused pathology."
The findings also tend to reinforce the notion that autism is "prenatal in origin," said Max Wiznitzer, MD, of UH Hospitals and UH Rainbow Babies and Children's Hospital in Cleveland.
The disorganization might either "directly cause the communication disturbances between parts of the brain," he said, "or [it] predisposes the child to another, yet to be determined trigger (if it exists)."
Although the study sample was small, these experts didn't think it was a major limitation.
"This was actually a good sample size for a postmortem study on this population," Molholm said in an email, "and the age range was as focused as it could be for the amount of cases available."
But other limitations were more concerning. Mink told MedPage Today that the brains available for autopsy -- both from autistic children and from normal controls -- may not be representative of their respective populations just because of the nature of the children's deaths and the parents' decisions to donate the organs for research.
Wiznitzer said it would have been useful to also include a control group with non-autistic neuropsychiatric diagnoses, to determine whether the disorganized patches are unique to autism spectrum disorders versus other types of behavioral abnormalities.
Curtis Deutsch, PhD, of the University of Massachusetts Medical School in Worcester, Mass., said he didn't think that was a serious problem. But the approach taken by Courchesne and colleagues limited the amount of data gathered in the study, he said.
"The authors used an exploratory design, one that employed sparse sampling of tissue across a wide area of whole brain. In other words, they cast a wide net in search of neuroanatomic abnormalities, and they found an abundance of neuropathologic 'patches,'" Deutsch wrote in an email.
"It will be of interest in the future to perform [higher resolution] analyses to investigate the extent and distribution of these pathologic features, and to study their statistical properties."
Similarly, Molholm said she wished the authors had been more specific about the patch locations and their correspondence in individual cases to behavioral abnormalities, if any.
But, she added, the faux-3D illustrations Courchesne and colleagues included in the NEJM paper were "very cool."
The study was funded by the Simons Foundation, the Peter Emch Family Foundation, Cure Autism Now/Autism Speaks, the Thursday Club Juniors, the Allen Institute for Brain Science, and the University of California San Diego.
One author reported equity ownership in Stereology Resource Center, which manufactured equipment used in the study. One author reported payments from Austism Speaks. Other authors declared they had no relevant financial interests other than research funding from the organizations listed above.
Primary source: New England Journal of Medicine
Source reference: Stoner R, et al "Patches of disorganization in the neocortex of children with autism" N Engl J Med 2014; 370: 1209-1219.
