NEW YORK CITY—One of the holy grails in schizophrenia research has been the possibility of preventing the disease, perhaps by intervening during the long prodromal phase. Although it remains unclear whether such interventions are effective, several major research programs are currently under way in this area, and early findings are yielding intriguing results. For example, presenters at the annual meeting of the American Academy of Child and Adolescent Psychiatry (AACAP) described studies suggesting that clinicians can often successfully identify which persons are likely to develop schizophrenia in the near future. Moreover, data from three recent or ongoing trials provide some provocative hints that treating patients during the prodromal phase may delay the onset of full-blown schizophrenia and improve outcome, at least in the short term.

Prevention encompasses at least two steps—successful identification of who is at high risk and successful treatment. Both steps have their pitfalls, including the possibility that patients not fated to develop schizophrenia will be unnecessarily exposed to the adverse effects of long-term treatment. On the other hand, "there may be a measurable and significant risk—in the form of turning psychotic—to doing nothing," noted Thomas H. McGlashan, MD, Professor of Psychiatry at Yale University School of Medicine.

The danger of the "wait-and-see" approach is illustrated by a groundbreaking randomized trial led by Patrick D. McGorry, MBBS, PhD, MRCP (UK), Professor of Psychiatry and Director of the Centre for Young People's Mental Health at the University of Melbourne, Australia. The study, the only schizophrenia prevention trial completed thus far, involved 59 persons with prodromal symptoms who received either "supportive following" or a multimodal treatment regimen consisting of low-dose risperidone, cognitive behavioral therapy, and (if necessary) antianxiety or antidepressant medications. After six months in the study, schizophrenia was diagnosed in 10 of the 28 control participants (36%) but in only four of the 31 of treated subjects (13%).

THE NEXT STEP: MEDS VERSUS PLACEBO

Dr. McGorry has often emphasized the potential importance of psychosocial interventions in schizophrenia prevention. However, because the members of the treatment group in his trial received both psychotherapy and pharmacotherapy that the control group did not, it was impossible to determine the relative contribution of the two components. Dr. McGlashan and colleagues are currently conducting what he calls "the next-step study"—a trial in which all patients are given psychosocial treatment and are randomized to receive either placebo or an antipsychotic as well. The study, which is part of Yale's PRIME program (Prevention through Risk Identification, Management, and Education), will thus provide evidence as to whether medication has preventive properties for prodromal patients.

Subjects take part in the study for two years. During the first year, patients receive psychotherapy, a novel neuroleptic (or placebo), and family intervention as needed. Pharmacotherapy is stopped after 12 months, and patients are followed. If a subject's diagnosis changes to schizophrenia at any point during the trial, the medication blind is broken and the subject receives six months of open-label treatment.

Patients are assessed using the Structured Interview for Prodromal Symptoms (SIPS), an instrument developed at Yale. About 80% of participants have attenuated psychotic symptoms—unusual thought content, grandiosity, perceptual abnormalities—but do not meet criteria for schizophrenia. They may, for example, feel that they are being persecuted but in discussion can bring perspective to their plight. The remaining 20% of subjects are considered at risk for developing schizophrenia because they have had brief intermittent psychotic episodes (lasting no more than a few minutes) or because they have a family history of schizophrenia and have had a recent, significant loss of functioning.

A representative example is Jack, a 17-year-old who had become increasingly withdrawn, irritable, and socially awkward. His grades had begun to slip, and his periodic suspiciousness often seemed to border on paranoia; at times, he said, he felt that the entire sophomore class was watching him.

Preliminary data from the trial suggest that the SIPS is not only reliable but has value for predicting the likelihood that a person will develop schizophrenia. For example, six of the first 13 subjects (46%) whose SIPS scores indicated prodromal status developed full schizophrenia within six months of enrolling in the trial, whereas none of the nine subjects who didn't meet prodromal criteria converted during that time frame. Similarly, after a year, seven of 12 SIPS-positive subjects (58%) had schizophrenia, but none of seven SIPS-negative persons did.

Although the study is ongoing and the researchers are still blind to treatment assignments, some interesting trends are already apparent, Dr. McGlashan noted. Of the 32 patients randomized thus far, three were still prodromal at the end of the medication phase, 11 had remitted, and eight had dropped out. The remaining subjects had converted to schizophreniform disorder (n = 8) or psychotic mania (n = 1), or had developed nonpsychotic affective disorder (n = 1). Encouragingly, none of the eight converters to schizophreniform disorder have been hospitalized, and most have lost only minimal time from work or school. Moreover, all have had a robust response to treatment, a finding consistent with recent evidence that short durations of untreated psychosis may be associated with better outcomes. "This is a group with a duration of untreated psychosis of zero, and there does appear to be an advantage," Dr. McGlashan said.

As to whether neuroleptics substantially reduce the rate of conversion to schizophrenia, Dr. McGlashan acknowledged that "I don't think we're going to have an answer for awhile." But the SIPS, he believes, represents a crucial first step: Although not 100% sensitive and specific, it does have good reliability and predictive value.

IS COMMUNITY CARE HELPFUL?

The other major prevention effort currently in progress, the Recognition and Prevention (RAP) program at Hillside Hospital in Glen Oaks, New York, and Schneider's Children's Hospital in New Hyde Park, New York, is less concerned with testing the prophylactic effects of a specific medication than with examining whether an intervention resembling typical community care has preventive benefits. Thus far, 55 adolescents have completed the full research battery, which includes at least six months of treatment, reported Barbara Cornblatt, PhD, Director of RAP, at the AACAP meeting.

The course of treatment is based on the manifestation and severity of the subject's symptoms and may include individual, group, and/or family therapy, medication management, and psychoeducation. For subjects with mild to moderate symptoms, such as social withdrawal and perhaps unusual perceptions, the goal is to prevent psychosis; for those who have already received a DSM-IVdiagnosis of psychosis–not otherwise specified (PNOS), the goal is to prevent chronic schizophrenia.

One focus of the study is to identify which symptoms might best predict conversion to psychosis and schizophrenia. In the initial sample, the most commonly observed symptoms of prodromal adolescents without PNOS have been social withdrawal/isolation (68%), decline in school functioning (39%), odd behavior (39%), and magical beliefs (26%). PNOS patients have displayed these symptoms as well, in the same order of frequency, plus hallucinations (54%).

In addition to psychotherapy, most subjects (80%), including nearly all PNOS patients, have received pharmacotherapy. "We're getting definite indications that we can either improve or stabilize these kids using medications," Dr. Cornblatt said in an interview. Although more than half of patients have been treated with an atypical antipsychotic, other medications have played a sizeable role in the study as well: about a third of the patients have received an antidepressant (especially those with attenuated negative and disorganized symptoms), and 19% have been given another class of medication (typically a mood stabilizer). "We're getting just as good results using selective serotonin reuptake inhibitors and other medications" as with antipsychotics, according to Dr. Cornblatt.

Not surprisingly, outcome has varied according to the nature and severity of patients' initial symptoms. Although 45% of PNOS patients have developed schizophrenia after six or more months in the program, a fifth have remained stable and 35% have actually improved. Treatment has been even more successful in the two less severely affected subgroups—patients with attenuated negative or attenuated disorganized symptoms and those with attenuated negative symptoms and at least one positive symptom. "In these two nonpsychotic groups, we've either improved or stabilized the vast majority of kids," Dr. Cornblatt said.

Interestingly, of the first 27 subjects without PNOS at enrollment, none of the 20 drug-treated patients—but two of the seven patients without medication (29%)—converted to schizophrenia after at least six months in the study. The medication group included 10 subjects treated solely with antidepressants, suggesting that these drugs "may be an effective firstline treatment of prodromal symptoms," noted Dr. Cornblatt and colleagues. They further suggested that "in the absence of base rates, treatment should be symptom based, not prevention oriented."

The Hillside team has just embarked on a new, five-year naturalistic study funded by the National Institute of Mental Health. The expanded project, they hope, will yield further insights into the range of medical and psychosocial therapies that may help prevent conversion to schizophrenia.

SCHIZOTAXIA REVISITED

Work by Ming Tsuang, MD, PhD, DPhil, and colleagues at Massachusetts Mental Health Center is also examining the feasibility of preventing schizophrenia. Schizotaxia is a concept that may be useful in these efforts, suggested William Stone, PhD, Assistant Professor of Psychology at Massachusetts Mental Health Center, at the AACAP meeting.

Dr. Stone and his colleagues have expanded the schizotaxia concept—developed in the 1960s by Paul Meehl, MD—to refer to the "premorbid substrate" of schizophrenia, including both genetic predisposition and adverse environmental factors (such as birth complications). According to this model, schizotaxia encompasses schizophrenia but does not lead to it unless the combination of genetic factors and environmental insults are sufficiently robust.

As part of an initial attempt to validate the schizotaxia model, Dr. Stone and colleagues studied six persons who had moderate or greater levels of negative symptoms, had first-degree relatives with schizophrenia, and had scored two standard deviations below the norm in at least two of three cognitive domains: attention, verbal memory, and executive function. None of the subjects had a history of psychosis.

To see whether such individuals might benefit from pharmacotherapy, the researchers treated the six subjects with risperidone, starting at 0.25 mg/d and increasing the dosage to a maximum of 2 mg/d over six weeks. By the end of the trial, five of the six had improved scores on the Scale for the Assessment of Negative Symptoms (SANS); in three cases, the score was cut in half. All but one of the subjects reported subjective cognitive improvements, especially in attention. Interestingly, only one of three subjects who reported improvement in social functioning attributed the changes to the medication. Adverse effects were mild and transient.

A typical case was a 38-year-old woman whose symptoms included anhedonia, flat affect, attention problems, and executive deficits. During the trial, her SANS score decreased from 23 to 12 and her performance on a test of attention improved to normal levels; she also showed moderate improvement in executive function. Other people said she was more pleasant to be around. The woman also reported subjective improvements that reversed after treatment ended.

"If we can successfully treat these kinds of symptoms in adults and have few negative side effects, could we begin to think about the same kind of treatment at earlier and earlier ages?" Dr. Stone asked. Validation of the schizotaxia model, he said, would be an important step toward the implementation of prevention programs. It may be time, he continued, to initiate field trials to develop a biologically informed approach to classification that considers schizophrenia and schizotaxia as distinct diagnoses. Such a system might allow clinicians to better identify which patients are at risk for schizophrenia and would help target interventions to the most appropriate patients.

—Peter Doskoch

Suggested Reading
1. McGlashan TH. Early detection and intervention of schizophrenia: rationale and research. Br J Psychiatry Suppl.1998;172:3-6.
2. McGorry PD, Jackson HJ, eds. The Recognition and Management of Early Psychosis: A Preventive Approach.New York, NY: Cambridge University Press; 1999.
3. Tsuang MT, Stone WS, Faraone SV. Towards reformulating the diagnosis of schizophrenia. Am J Psychiatry.2000;157:1041-1050.
4. Tsuang MT, Stone WS, Faraone SV. Towards the prevention of schizophrenia. Biol Psychiatry.2000;48:349-356.
5. Yung AR, Phillips LJ, McGorry PD, et al. Prediction of psychosis. A step towards indicated prevention of schizophrenia. Br J Psychiatry Suppl.1998;172:14-20.

Return to table of contents