12 July 2007. The loss of cortical gray matter as an endophenotype of schizophrenia receives support from a new longitudinal study of people with a high genetic risk for schizophrenia. As reported in the July issue of Archives of General Psychiatry, Nitin Gogtay of the National Institute of Mental Health and colleagues found that siblings of people with childhood-onset schizophrenia had significantly thinner neocortical gray matter from early ages, in just the same areas as their affected siblings. However, the healthy siblings appear to compensate for their genetic hurdles by age 20, by which point these deficits had disappeared, leaving no differences between them and control subjects.

In another new prospective study of gray matter deficits, this one in patients with adult onset schizophrenia, Neeltje van Haren and colleagues at the Rudolf Magnus Institute of Neuroscience in Utrecht, the Netherlands, report that abnormal gray matter loss continues in adult patients for about the first 20 years of the illness, after which the decrement begins to resemble the gray matter loss seen in normal aging.

Cortical Thinning in Schizophrenia: A Family Trait?
Irving Gottesman, of the University of Minnesota, and colleagues have long advocated an endophenotype approach to unraveling the neurobiology and genetics of psychiatric disorders (see SRF Live Discussion; Gottesman and Gould, 2003; Gould and Gottesman, 2006). These researchers argue that the disease heterogeneity inherent in the symptom-based DSM-IV diagnostic classifications and the largely qualitative basis of these categories make it exceedingly difficult to draw clear lines from genes to neurobiology to behavior with valid animal models of human disease. According to this point of view, research in biological psychiatry will progress more quickly through fine-grained studies of heritable, stable, quantifiable markers at levels of analysis lower than the complex, variable phenotype of schizophrenia seen in the clinic.

It is widely accepted that the schizophrenia phenotype is polygenic in origin, and subject to many poorly understood epigenetic, stochastic, and environmental influences. Proponents of the endophenotype approach believe that “simpler,” quantifiable components of the schizophrenia phenotype, even those with little apparent relevance to overt symptoms, will be amenable to the creation of useful animal models and the identification of candidate genes that contribute to the disorder as a whole.

One proposed endophenotype in schizophrenia, supported by many morphometric and cortical thickness studies, is reduced gray matter (GM) volume in the cortex, hippocampus, and amygdala. However, it has been difficult to discern whether the GM deficits seen in schizophrenia are a familial/trait marker or are somehow secondary to the illness, because findings of reduced GM volume in close relatives of schizophrenia probands, most based on whole-lobe brain volumes, have been equivocal (see recent meta-analysis by Boos et al., 2007).

To address this question, Gogtay and colleagues at NIMH and Montreal Neurological Institute recently completed a prospective cortical thickness study of 52 healthy, younger full siblings of probands diagnosed with childhood-onset schizophrenia (COS), a rare form of the disorder that appears before puberty and is associated with poor outcomes (see, e.g., a recent German follow-up study of patients diagnosed between 1920 and 1961, Remschmidt et al., 2007, as well as SRF Q&A with Gogtay below).

A progressive loss of GM occurs in adult-onset schizophrenia, but it is particularly pronounced in COS during adolescence. By early adulthood, overall GM loss in COS patients becomes more restricted, mostly affecting the prefrontal and temporal cortices bilaterally. Because the healthy COS siblings in the study were aged 8 to 28 years, a period of intense brain plasticity, Gogtay and colleagues predicted that repeated MRI scans over time would reveal whether these subjects followed a similar developmental trajectory.

Reappearing Act
Gogtay and colleagues performed 113 scans on the 52 siblings of patients with COS and 108 scans on 52 age-, sex-, and scan-interval-matched healthy controls. Using an automated measurement procedure on the acquired images, the researchers found significant GM deficits in the COS siblings in the left prefrontal cortex and bilaterally in the temporal cortex at 8 years of age compared with healthy controls. These deficits in frontal and temporal regions had disappeared in the COS siblings by age 20, and this diminution of cortical thinning was positively correlated with improved scores on the Global Assessment Scale (GAS).

“The direct relationship between normalization of cortical thickness and GAS scores would suggest that in the absence of core schizophrenia spectrum pathologic features or other confounding factors such as medication exposure, restitutive ‘normalization’ of GM deficits...is closely related to their overall functioning,” the authors write. “Whether this relationship between GM thickness and GAS score reflects cause or effect, or influence of yet another factor, remains unknown.”

Because the pattern of early frontal and temporal GM deficits seen in the asymptomatic COS siblings in the Gogtay team’s study was strikingly similar to that which persists into adulthood in COS, the authors conclude that these deficits are likely to be a familial/trait marker. On the other hand, a pattern of significant parietal GM deficits that is common in COS was not observed in the healthy siblings, leading the researchers to propose that these abnormalities may be under environmental control or secondary to illness.

Gogtay and colleagues caution that their study concentrated on the cortical surface, so the results do not address the possibility of persistent GM deficits in COS siblings in deeper brain structures such as the hippocampus and amygdala. Also, ...while all subjects had 2 scans, relatively few had 3 or more scans. Thus, although partially longitudinal, the group of siblings did not overlap with those at younger ages and the study remains susceptible to unknown cohort effects,” the authors write.

A Slowing Rate of Decline
The recent longitudinal study of brain-volume loss in adult patients with schizophrenia by van Haren and colleagues in the Netherlands, published online June 26 in Biological Psychiatry, also revealed a normalization of gray matter change over time. Patients in the study exhibited a linear trajectory of total cerebral GM loss, from 6.3 ml/year at age 20 to 5.7 ml/year at age 50. By contrast, in healthy controls, volume loss significantly decreased (from 4 ml/year to 2 ml/year) between early adulthood and 29 years of age. By age 50, controls showed a loss of 5 ml/year, comparable to the patients. “These findings suggest the excessive brain tissue loss in schizophrenia to occur during the first 10 to 20 years of the illness. Later in life, the degree of cerebral and gray matter volume loss in patients is similar to that observed with normal aging…. In addition, patients with a lower level of functioning showed more brain tissue loss during the interval relative to good functioning patients,” the authors write.—Peter Farley.

References:
Gogtay N, Greenstein D, Lenane M, Clasen L, Sharp W, Gochman P, Butler P, Evans A, Rapoport J. Cortical brain development in nonpsychotic siblings of patients with childhood-onset schizophrenia. Arch Gen Psychiatry. 2007 Jul;64(7):772-80. Abstract

van Haren NEM, Pol HEH, Schnack HG, Cahn W, Brans R, Carati I, Rais M, Kahn RS. Progressive brain volume loss in schizophrenia over the course of the illness: evidence of maturational abnormalities in early adulthood. Biol Psychiatry. 2007 Jun 26; [Epub ahead of print] Abstract

Q&A with Nitin Gogtay. Questions by Peter Farley.

Q: Childhood-onset schizophrenia (COS) is so rare that most clinicians have never encountered a case. What is its incidence?
A: We don’t know the exact prevalence of the illness. If adult-onset schizophrenia happens in 1 percent of the population worldwide, this happens in about 1 in 30,000, and is probably more rare than that.

Here [at the Childhood-Onset Schizophrenia Study in the Child Psychiatry Branch of the NIMH] we go through an enormous screening and evaluation process, and practically every child is only diagnosed after full medication washout. When these kids are admitted, they’re often on six to nine medications, two to three neuroleptics, and two to three mood stabilizers, all at adult doses. We don’t formally diagnose them until they’ve been without medications for as long as 3 weeks. This is very important to do, because medications can confound and confuse the picture in children—as many as 30 percent of the children we admit leave the unit with a diagnosis other than schizophrenia.

Over 14 years we have assembled about 96 children after screening over 2,000 charts. It’s a small cohort of well-identified cases, but ours may be one of the largest samples, if not the largest, in the world of this phenotype.

Q: What is the average age of onset?
A: The statistical average is 10 years old. The youngest we’ve seen so far is 6.5 years old. But it’s difficult to say whether there is a clustering at age 10. Children are often misdiagnosed because it’s hard to interpret whether there are real hallucinations or delusions, for example.

Q: Do current theories see COS as an accelerated form of the more typical illness, or is it thought to emerge from some different mechanism?
A: It’s probably more genetic salience, because we have not been able to identify any factors specific to the illness. On the neurobiological and neurocognitive measures, the illness seems continuous with adult-onset illness—it’s the same illness, but it happens very early on. This follows a pattern seen elsewhere in medicine: when adult-onset illnesses like diabetes or cancer happen early on in life, they tend to be much more severe. The same thing seems to be happening here in schizophrenia.

While we haven’t been able to identify any specific risk factors that could act as stronger triggers, we do find a higher incidence of cytogenetic abnormalities and chromosomal abnormalities, and we have seven or eight cases that are positive for the risk genes that have been identified in adult-onset cases, which is a very high incidence for such a small cohort.

Q: What is known about the long-term outcome of COS patients?
A: COS has a much poorer outcome. Our kids are very sick, and there is no episodicity—they get sick and they stay sick. They are very treatment-refractory. Those who respond well to medications, usually clozapine, do better than those who do not, but the majority remain symptomatic. It’s more like a chronic adult schizophrenia phenotype seen very early on in life.

Q: What piqued your interest as a researcher in this form of the disease?
A: I’m a neuropathologist by training and did research in that area for many years. I eventually decided to do a psychiatry residency with the interest of pursuing research on schizophrenia. When I heard about this long-term study I came to speak to Judy Rapoport, who began the project about 14 years ago. I was fascinated by the nature of the cohort, because this is a very unusual phenotype.

These kids are psychotic, by definition, before they are 13 years old, and this is a very severe form of the illness, amplified many-fold in terms of severity compared to adult-onset schizophrenia. Because it happens so early on in life, I think one also gets to see schizophrenia in a really pure phenotype, because it’s not “contaminated” by multiple bouts of substance abuse, multiple treatments, and so on. There’s also a feeling that, because it is so severe and happens so early in life, maybe there is more genetic salience.

We study a very homogeneous, severe phenotype of schizophrenia, and we feel that the biological factors are likely to be more severely manifest, so we’re likely to get more insights into the neurobiology of the illness.