Imaging Brain Fatty Acid Amide Hydrolase in Untreated Patients With Psychosis

Abstract

Background: The brain's endocannabinoid system, the primary target of cannabis, has been implicated in psychosis. The endocannabinoid anandamide is elevated in cerebrospinal fluid of patients with schizophrenia. Fatty acid amide hydrolase (FAAH) controls brain anandamide levels; however, it is unknown if FAAH is altered in vivo in psychosis or related to positive psychotic symptoms.

Methods: Twenty-seven patients with schizophrenia spectrum disorders and 36 healthy control subjects completed high-resolution positron emission tomography scans with the novel FAAH radioligand [¹¹C]CURB and structural magnetic resonance imaging. Data were analyzed using the validated irreversible 2-tissue compartment model with a metabolite-corrected arterial input function.

Results: FAAH did not differ significantly between patients with psychotic disorders and healthy control subjects (F_1,62.85 = 0.48, p = .49). In contrast, lower FAAH predicted greater positive psychotic symptom severity, with the strongest effect observed for the positive symptom dimension, which includes suspiciousness, delusions, unusual thought content, and hallucinations (F_1,26.69 = 12.42, p = .002; Cohen's f = 0.42, large effect). Shorter duration of illness (F_1,26.95 = 13.78, p = .001; Cohen's f = 0.39, medium to large effect) and duration of untreated psychosis predicted lower FAAH (F_1,26.95 = 6.03, p = .021, Cohen's f = 0.27, medium effect). These results were not explained by past cannabis exposure or current intake of antipsychotic medications. FAAH exhibited marked differences across brain regions (F_7,112.62 = 175.85, p < 1 × 10^-56; Cohen's f > 1). Overall, FAAH was higher in female subjects than in male subjects (F_1,62.84 = 10.05, p = .002; Cohen's f = 0.37).

Conclusions: This first study of brain FAAH in psychosis indicates that FAAH may represent a biomarker of disease state of potential utility for clinical studies targeting psychotic symptoms or as a novel target for interventions to treat psychotic symptoms.

Keywords: Endocannabinoid; Fatty acid amide hydrolase; Imaging; Positron emission tomography; Psychosis; Schizophrenia.

Figure 1:

[¹¹C]CURB λk₃ in healthy controls and patients with psychotic disorders. Means are adjusted for sex and FAAH rs324420 genotype. Hipp: hippocampus; LST: limbic striatum; AST: associative striatum; SMST: sensorimotor striatum; ACC: anterior cingulate cortex; mPFC: medial prefrontal cortex; DLPFC: dorsolateral prefrontal cortex.

Figure 2:

Partial residuals plots of [¹¹C]CURB λk₃ and the positive symptom dimension in psychosis patients (n=27). Fit line and shaded areas reflect the model fit and 95% confidence interval of the overall random effects linear mixed model, controlling for FAAH rs324420 genotype, sex and including all brain regions (n=8). LST: limbic striatum; AST: associative striatum; SMST: sensorimotor striatum; ACC: anterior cingulate cortex; mPFC: medial prefrontal cortex; DLPFC: dorsolateral prefrontal cortex.

Figure 3:

[¹¹C]CURB λk₃ in males and females in the entire sample (n=63). Group means are adjusted for diagnostic group and FAAH rs324420 genotype. Hipp: hippocampus; LST: limbic striatum; AST: associative striatum; SMST: sensorimotor striatum; ACC: anterior cingulate cortex; mPFC: medial prefrontal cortex; DLPFC: dorsolateral prefrontal cortex.