Altered brainstem responses to modafinil in schizophrenia: implications for adjunctive treatment of cognition.

Candidate pro-cognitive drugs for schizophrenia targeting several neurochemical systems have consistently failed to demonstrate robust efficacy. It remains untested whether concurrent antipsychotic medications exert pharmacodynamic interactions that mitigate pro-cognitive action in patients. We used functional MRI (fMRI) in a randomized, double-blind, placebo-controlled within-subject crossover test of single-dose modafinil effects in 27 medicated schizophrenia patients, interrogating brainstem regions where catecholamine systems arise to innervate the cortex, to link cellular and systems-level models of cognitive control. Modafinil effects were evaluated both within this patient group and compared to a healthy subject group. Modafinil modulated activity in the locus coeruleus (LC) and ventral tegmental area (VTA) in the patient group. However, compared to the healthy comparison group, these effects were altered as a function of task demands: the control-independent drug effect on deactivation was relatively attenuated (shallower) in the LC and exaggerated (deeper) in the VTA; in contrast, again compared to the comparison group, the control-related drug effects on positive activation were attenuated in LC, VTA and the cortical cognitive control network. These altered effects in the LC and VTA were significantly and specifically associated with the degree of antagonism of alpha-2 adrenergic and dopamine-2 receptors, respectively, by concurrently prescribed antipsychotics. These sources of evidence suggest interacting effects on catecholamine neurons of chronic antipsychotic treatment, which respectively increase and decrease sustained neuronal activity in LC and VTA. This is the first direct evidence in a clinical population to suggest that antipsychotic medications alter catecholamine neuronal activity to mitigate pro-cognitive drug action on cortical circuits.