Early intervention in psychosis might be associated with better outcomes. However, intervention in the pre-psychotic phase has been questioned as, using current criteria, only 20-50% of individuals classified as prodromal develop a psychotic disorder within a 1-2 years period. Treatment agents investigated in the pre-psychotic phase of schizophrenia and other psychotic disorders should, therefore, not have major side effects. This proposal investigates omega-3 fatty acids (1.2 gramm per day eicosapentaenoic acid/docosahexaenoic acid;EPA/DHA) as a beneficial and possible preventive therapeutic agent in young people at ultra high-risk for developing a psychotic disorder.

1. Aims of the study The principal aim is to test if 1.2 g/day EPA/DHA can prevent transition to first-episode psychosis in 13-25 year old ultra-high risk individuals.

   Specifically we propose to investigate:

   • The clinical effects of EPA/DHA supplementation as an adjunct to standard therapy in individuals with 'At-Risk Mental State' (ARMS) for psychosis as defined by the PACE criteria (Yung et al., 1998).
   • Lipid metabolism in peripheral tissue pre/post treatment by 1./analyzing bioactive lipid composition of red-blood cell membranes, 2./measuring phospholipase A2 (cPLA2) activity in serum (the enzyme responsible for the cleavage of arachidonic acid (AA) and other precursors of bioactive lipids from glycerophospholipids (GPL) and 3./the topical niacin flush test (a clinical test of the AA-prostaglandin D2 cascade).

2. Background and evidence that bioactive lipids are altered in schizophrenia and can be influenced by EPA/DHA supplementation

   There is suggestion that early intervention in psychosis might be associated with better outcomes (Norman & Malla, 2001). However, intervention in the pre-psychotic phase has been questioned as, using current criteria, only 20-50% of individuals classified as prodromal develop a psychotic disorder within a 1-2 years period (McGlashan et al., 2001). Treatment agents investigated in the pre-psychotic phase of schizophrenia and other psychotic disorders should, therefore, not have major side effects. This proposal introduces EPA/DHA, two omega-3 essential fatty acids (EFA), as a beneficial and possible preventative therapeutic agent in young people at ultra high-risk for developing a psychotic disorder.

   Bioactive lipids and their role in the brain Bioactive lipids are molecules that have both intra- and intercellular roles, including mediation, modulation and control of neurobiological processes, such as ion channel and receptor activity, neurotransmitter release, synaptic plasticity, second messenger pathways and neuronal gene expression (Agranoff et al., 1998). Emphasis has been placed on AA and its metabolites, known collectively as eicosanoids. A major proportion of lipids in the brain consist of bioactive lipids such as AA and its metabolites, also referred to as EFA, which are mainly bound to GPL. Bioactive lipids are released through direct and indirect enzymatic pathways (e.g., phospholipases) from membrane GPL. AA is a precursor of prostaglandins, thromboxanes, leukotriens (5-HpETE) and prostacyclins. Animal studies and preliminary studies in humans have shown an association between bioactive lipid metabolism, behaviour and cognition (Zimmer et al., 2000).

   Reduced membrane EFA in schizophrenia Abnormal membrane GPL EFA metabolism has been suggested to contribute to the aetiopathophysiology of schizophrenia. A recent review of 15 published studies confirmed a depletion of bioactive lipids in cell membranes of patients with schizophrenia (Fenton et al., 2000). The most consistent findings were reductions in AA and its precursors, and these were independent of drug treatment (Yao et al., 1996). Reductions in AA and its precursors have also been found in post mortem brains of patients with schizophrenia, relative to normal control brains [Yao et al., 2000]. Yao and van Kammen (1996) suggested that defective uptake of AA into membrane GPL was a possible aetiopathological mechanism in schizophrenia, whereas Peet et al. (1996), who reported an additional increase of EFA peroxidation products, suggested there was increased breakdown of membrane GPL.

   Khan et al. (2002) reported on erythrocyte membrane EFA levels and levels of plasma lipid peroxides, products of damaged EFAs, in drug-naive patients within +/-4.5 days of onset of psychosis. The levels of EFAs, particularly AA and docosahexaenoic acid (DHA) were significantly lower in drug-naive patients at the onset of psychosis compared to matched normal controls. These lower EFA levels were associated with significantly higher levels of lipid peroxides in patients. The levels of AA and DHA were also lower and lipid peroxides higher in chronic medicated patients than normal controls. Interestingly in context with this proposal, EFA levels were higher in chronic medicated patients than drug-naive first-episode patients. Khan et al. concluded that these findings could indicate that lower membrane AA and DHA most likely predate the illness and probably contribute to the onset of illness. The lipid peroxidation data suggest that possible increased oxidative stress may be one of the mechanisms of reduced membrane EFAs. The findings also imply that supplementation of EFAs and/or antioxidants might provide effective treatments for early psychosis. This view is supported by Horrobin et al. (2002) who showed that increase in red cell AA levels resulted from treatment with the optimal levels of EPA and that, clinical improvement was highly significantly positively correlated with rises in red cell membrane AA in individuals with schizophrenia.

   Treatment studies in schizophrenia Three randomized controlled treatment studies conducted over 12 weeks found 2g/day EPA significantly more effective than placebo in reducing psychopathological symptoms in individuals with schizophrenia (Peet et al., 2001; Emsley et al., 2002). Symptom improvements in those studies were both, clinically relevant and statistically significant. A dose-ranging exploratory study of the effects of EPA in individuals with schizophrenia who experienced persistent symptoms found 2 g EPA/day significantly more effective in reducing symptom scores on psychiatric rating scales than 1g and 4g EPA/day (Peet et al., 2002).

   On the other hand, Fenton et al. (2002) investigated augmentation of neuroleptics with 3 g/day of EPA on symptoms and cognition in patients with schizophrenia or schizoaffective disorder and reported a negative finding. The patients in Fenton et al.s' study had, however, been ill for two decades and had substantial symptoms, despite treatment with newer neuroleptics, including clozapine. The patients described as benefiting from EPA in the other studies were younger and had a shorter duration of illness.

   It must be emphasized that in all EPA treatment studies, no treatment-related side effects or adverse biochemical or haematological effects have been observed. EPA proved safe to administer to schizophrenic patients as an adjunct therapy. EPA did not cause side effects other than mild gastrointestinal symptoms by itself, nor did it enhance the side effects of existing drugs. Patients found EPA highly tolerable. The proportion of patients who completed 12 weeks (89%) compares favourably with mean withdrawal rates of 54% in the novel neuroleptic groups and 67% in the placebo groups in trials in the FDA database (Peet et al., 2002). Acceptance of a substance which is normally found in the human body without significant side effects, with a potency potentially similar to the antipsychotic drugs in the early phase of psychotic disorders could contribute to reduce the duration of untreated psychosis and to increase compliance.

3. Study design We will use a prospective, randomized, double-blind, placebo-controlled, single-centre study design. Eighty one individuals aged 13-25 will be randomly assigned in two treatment conditions at the University Clinic for Child and Adolescent Neuropsychiatry, Vienna, Austria. Randomization codes will be generated and stored off site. The treatment groups will receive 1.2 gramm per day EPA/DHA or placebo for 12 weeks. Follow-up assessments will be conducted at 1,2,3,4,8,12,26,and 52 weeks. All patients will receive standard treatment, which includes management by a psychiatrist or resident psychiatrist and non-neuroleptic pharmacotherapy as clinically indicated.

Inclusion Criteria:

1. /written informed consent (for individuals under 18 written informed consent of at least one of the parents is required),
2. /age between 13 and 25 years,
3. /ARMS as classified by the PACE criteria (Yung et al., 1998)

PACE criteria for ARMS include one or more of following characteristics which must have occurred within the last 12 months:

• Frank psychotic symptoms < 1 week (Transient psychosis group)
• Attenuated psychotic symptoms > 1 week, > 2 times per week
• Decline in global function (drop in GAF of > 30%) plus family history of psychosis or individual has schizotypal personality disorder To operationalize PACE criteria duration and severity ratings of psychotic symptoms will be performed using the Positive and Negative Syndromes of Schizophrenia Scale (PANSS) (Kay et al., 1987) applying following cut-off scores, following Morrison et al (2002): Ad 1) Transient psychosis is defined with the presence of symptoms that score 4 or more on hallucinations, 4 or more on delusions, or 5 or more on conceptual disorganizations, last less than one week and resolve without antipsychotic medication. Ad 2) Attenuated psychotic symptoms are defined by the presence of symptoms that score 3 on delusions, 2-3 on hallucinations, 3-4 on suspiciousness or 3-4 on conceptual disorganization.

Exclusion Criteria:

1. /Acute suicidal behaviour, aggressive behaviour (PANSS hostility, suicidality = 7),
2. /Drug abuse that contributed decisively to the presentation of the index episode, (dependency on morphine, cocaine, amphetamine, but not THC),
3. /Alcohol abuse if considered as major problem,
4. /Epilepsy,
5. /Mental Retardation (IQ<80),
6. /Pregnancy and lactation,
7. /Structural changes in MRI or CT scan (e.g., tumours), expect for enlargement of ventricles or sulci,
8. /Previous history of antipsychotic drug (>1 week) or mood stabilizer treatment,
9. /Laboratory values more than 10% outside the normal range for transaminases, CRP or bleeding parameters,
10. /Individuals with organic brain syndrome,
11. /Individuals who are taking anticoagulants,
12. /Individuals who are taking omega 3 supplements, currently or within 8 weeks of being included in the trial,
13. /Individuals who have other, severe, intercurrent illness which in the opinion of the investigator may put them at risk or influence the results of the trial or affect ability to take part in the trial.