September 23, 2005

Early Schizophrenia Treatment - Science Meeting

International Prodromal (Early Phase of Schizophrenia) Research Network Meeting Summary

The International Prodromal Research Network (IPRN) is a collection of leading schizophrenia researchers from Europe, United States and Australia.

Last week the group had a two-day meeting in Napa, California.

The meeting is sponsored by the Staglin Foundation and by an educational grant from Janssen Research Foundation. Organizers Tyrone Cannon, Ph.D from UCLA and Barbara Cornblatt, Ph.D, from Hillside Hospital and Albert Einsten College of Medicine, planned 4 sessions of debate and new data presentations. Over the course of two days, the group met to discuss topics including:

1. Treatment and Prevention,
2. Predicting Clinical Outcomes,
3. Characterizing Neurobiological changes,
4. New Directions for research.

Following is a brief summary of the topics presented at the meeting.

Session 1:
The first session was about Treatment and Prevention. The keynote speaker was Gahan J. Pandina, Ph.D of Janssen Pharmaceutica. His lecture focused on issues related to treatment adherence. Treatment adherence refers to the way that patients and physicians are able to work together such that the patient follows the advice of their doctor. He reported that for ever 10 days off of medication per year, there is an additional 6% chance of resulting hospitalization such that for those with 30 days or more off medications, there is over 20% chance of hospitalization. Treatment adherence also affects the veracity of research findings because if people are not known for sure to be taking medications in the study, it makes it more difficult to attribute the results found to the mechanisms in the medications. He presented many factors that contribute to difficulty with adherence including lower socioeconomic status, education and cognitive status. There were differences between men and women with men more likely to have a benefit in their adherence based on their intellectual abilities while women’s adherence rates were more related to social functioning measures and organizational abilities. Ultimately, there may be many explanations for adherence rates and factors such as co-morbid mood disorders like depression and anxiety need to be factored in as well as substance abuse and other important issues.

Tom McGlashan of Yale and Patrick McGorry of the Orygen Research Center in Australia lead a discussion of research strategies. The discussion focused on whether studies should be done as randomized, controlled trials that involved medications or if there should be an aspect of study design that incorporates naturalistic studies. Naturalistic studies are ones in which there is less intervention so that researchers can follow subjects to gather information and understand what factors might lead to a conversion into psychosis without having interference from medication related factors. Issues that complicate the discussion regard how to accurately describe someone in the prodromal period. Without strict diagnostic criteria, it is hard to determine the factors that best describe subjects’ symptoms and how to systematically determine how they change over time. Since randomized controlled trials rely on a systematic diagnosis on order o put people into the correct group, there are limitations on that type of research for this reason. However, the gold standard method of doing research is a randomized and controlled trial because it is the best way to isolate variables for accurate testing against comparator groups.

On the other hand, using naturalistic studies, either of the type that allows or does not allow medications, can help provide interesting data on the phenomenology of the disorder. In some naturalistic studies the personnel are allowed to use medications however they see fit and in other studies, medications are not allowed. It is often very difficult to recruit patients for purely observational studies as most potential subjects want to be in a study where they will be offered some sort of treatment. Also, most patients who arrive at an early psychosis clinic have been exposed to antipsychotic medication prior to their arrival and screening. This may lead to changes in the brain that would not have happened in the absence of medication and while the treatment may be necessary, it may impact research findings. The pressures to use medications may be either based on physician compassion or economic interest, but regardless, they have an impact on the future course of someone who is still not declared themselves to have a fully diagnosable syndrome such as schizophrenia or bipolar disorder.

Ultimately, the discussion focused on the need for both types of studies and for more data collection in general. It may be that more thought is needed on when to start medications and that starting too early may have lasting effects both psychiatric and not. It is not known if medications have protective or worsening effects long term in this population. Some believe that medications like lithium may protect the brain, but it is not proven. However, regardless of study type, it was agreed that there are many benefits for participating in a study on early psychosis because in all cases the amount of follow-up is increased such that if a person does develop a full-blown syndrome such as schizophrenia, it is more likely to be caught earlier and treatment that is needed can be started more promptly.

Session 2:
The focus of the second session of the meeting was on predicting outcomes both clinically and functionally. TK Larsen from Rogaland Psychiatric Hospital in Norway presented the keynote talk. He spoke about the meaning true vs. false positives in a prodromal research program. He proposed using the term, “hypopsychosis”, much like we use the term hypomania to describe someone with manic features that does not meet the full manic criteria, for someone who meets partial criteria for a psychotic disorder but not completely.

The discussion portion focused on risk factors for conversion to psychosis. Barbara Cornblatt and Alison Yung from the Orygen Research Center in Australia led the discussion. The discussion started with an acknowledgment that people who present to at-risk clinics already have demonstrated that they have symptoms and are already having difficulties. Whether they convert to schizophrenia or not, they are already people who need care and treatment in general; even if that is just close observation. The difficulty revolves in how to characterize the functional impairment, especially when there are significant differences between patients. The declines most notable in patients referred to these clinics are in the cognitive domain. Over 90% have some impairment in school and many have social isolation and other dysfunctional social skills. However, despite these common symptoms, which are found in many other situations and disorders, there is a much smaller rate of conversion to formal psychosis. However, the functional deficit is important and can have long lasting impacts on the patient.

There may be many potential underlying causes for these similar symptoms. The etiology however remains elusive. Is there a disorder of neurodevelopment? Neurotransmission (how the nerves and brain transmits messages)? Neuroplasticity (how the brain selectively prunes neurons in development)? Despite the outward appearance of similarity, there is a large variation in the background that leads to the condition genetically and biologically. If the outcomes are also different, it makes it difficult to characterize large numbers of subjects because one has to account for significant variation. The variation also makes it more difficult to ascertain the benefit of early treatment because it is hard to say whether the person will ultimately have schizophrenia and if not, it is difficult to characterize what disorder (or if there is a disorder) the patient has and what treatment would be most appropriate.

Session 3:

The theme fro this session was characterizing neurobiological changes in people considered at risk for psychosis. The keynote for this session was Ming Tsuang of UCSD and Harvard who gave a talk updating recent findings on genetic research. He detailed the results of recent work (which has been reviewed at In essence, he outlined a new technique in which red blood cells can be used to look at local genetic changes in the tissues in which they are harvested. This allows for researchers to study specific changes that may change in the brain based on which genes are turned on or off locally. This is exciting because while there may be a few genes related to schizophrenia, this allows to look at more genes at once and also to look for special markers that can alert us to more specific changes we might not otherwise have been able to see. He outlined future goals of using bioinformatics to help catalogue the data using more sophisticated methodologies that will allow us to look at more information at once than before. These techniques may help us to identify specific changes earlier and allow us to focus on prevention ultimately instead of treating problems that have already developed more fully.

The discussion section focused on the timing of neurobiological changes in people at risk for psychosis. The discussion was led by Daniel Mathalon of Yale and Greg Brown of UC San Diego. Are changes occurring during maturational processes or are they more or less just becoming unmasked at later ages, but already the result of a previous process? Evidence was reported that indicates that there are changes in the structures of the brain, namely the ventricles and some areas of gray matter that may progress with age. It may be that there is an insult to the brain in utero or at birth that may predispose a person to difficulty and a 2nd insult or also know as a 2nd “hit” may lead to the ultimate outcome. The first insult may be genetic or from the result of some kind of pregnancy related problem. Some other evidence of progression can be seen in specialized MRI scans that can look for specific molecules in the brain. These scans show evidence of increased brain cell breakdown. There are also functional tests that one can measure like mismatch negativity (see recent article review that may also shed light on brain mechanisms or early warning signs for impending psychosis.

The discussion continued with a presentation of 4 possible models including problems with programming and assembly during development that predispose one for a greater likelihood of schizophrenia. These problems may result from genetics or from prenatal traumas. Maturational changes may build on these poor foundations and lead to greater likelihood for psychosis as well.

Session 4:
The fourth session was more of a wrap up and look to the future. The keynote was given by Anil Malhotra of Zucker Hillside hospital in New York. He outlined several genes that are implicated in schizophrenia. The four he mentioned most specifically are the G72, DISC1, COMT and dysbindin genes. These are all related to various possible mechanisms though more work needs to be done to determine the mechanism for these changes. These genes seem to confer risk both to schizophrenia and to bipolar illness linking the two. Such linkages are becoming more frequent in research and there may be less difference between the mechanisms of these disorders than previously thought, though much more work remains on this topic as well.

The subsequent discussion included this topic of whether the schizophrenia prodrome should also include a bipolar prodrome. It was mentioned that there are similar percentages of people with bipolar as schizophrenia as well as similarity in heritability/genetics, relapse/remitting course, elevated risk of suicide, and that antipsychotic medications seem to help both syndromes. Additionally, within people who have bipolar disorder, nearly 60 percent have some form of psychosis at some point in their mania while many people with schizophrenia will ultimately have some elevation or depression of mood. Within at-risk psychosis clinics, there is much heterogeneity within the population such that some patients end up with schizophrenia, some with an anxiety disorder, some with a mood disorder like bipolar disorder and some with no disorder ultimately. At the initial presentation, it is difficult to impossible to fully predict clinical outcome. This interaction between bipolar disorder and schizophrenia is one of the very active areas in prodromal research.


Post a comment

Please enter this code to enable your comment -
Remember Me?
(you may use HTML tags for style)
* indicates required