A challenge to the dopamine orthodoxy in schizophrenia?

According to Kuhn’s argument science progresses by ‘paradigm shifts’ (Kuhn, 1962), an expression now so accepted that it has become entrenched in the lexicon of multiple disciplines. Kuhn’s classic examples included the shift from the Ptolemaic geocentric view of the planetary system to the Copernican heliocentric notion. On the horizon for the treatment of schizophrenia are non-dopaminergic acting agents which, if primary pharmacological actions mean anything at all, sit uncomfortably within the prevailing dopamine paradigm of antipsychotic drug action. Does the advent of such agents herald a Kuhnian ‘revolution’ in the understanding of antipsychotic drug action with an attendant ‘paradigm shift’ from the dopamine hypothesis of schizophrenia?

When introduced into clinical practice seven decades ago, chlorpromazine was rightly considered a breakthrough treatment for schizophrenia. It enabled many patients, who would otherwise have been confined long-term to mental hospitals, to be discharged to the community. While there were significant therapeutic benefits, these were offset by the associated neurological side effects (Stroup et al., 2000). The intervening years have resulted, arguably, in little in the way of improved efficacy with respect to the pharmacological treatment of the positive and negative symptoms, but current medications have a somewhat more benign neurological side effect profile (Stroup et al., 2000). Nevertheless, like the prototypical phenothiazine antipsychotic medications, current drugs are believed to derive their therapeutic benefits, at least in part, by antagonism/partial agonism at the dopamine D2 class of receptors (this includes D2-Short, D2-Long, D3 and D4 receptors) (Ginovart and Kapur, 2012). Recognition of the pharmacological actions of medications contributed significantly to an initial formulation of a dopamine hypothesis of schizophrenia. A more sophisticated iteration of the hypothesis focussed on the activity of dopamine within specific neuronal circuits. Overactivity of dopamine within the mesolimbic pathway is postulated to be responsible for positive symptoms of the disorder, while underactivity in meso-cortical pathways is posited to be responsible for negative and cognitive symptoms (Stahl, 2004). Within this framework of the modified hypothesis, alterations of the activity of other neurotransmitter systems (most notably a hypoactive glutamatergic system) have been accommodated. Although the hypothesis has been criticised as too reliant on the inference from drug effects to underlying disease process (Moncrieff, 2009), it nevertheless remains a plausible explanation of some symptoms of the disorder. The hypothesis has formed a useful drug discovery paradigm to the extent that all currently approved treatments can be considered to conform to a dopaminergic view of schizophrenia/antipsychotic drug action. Pre-clinical and early phase studies indicate that molecules without a primary dopamine D2-antagonist/partial agonist effect may be effective for the treatment of schizophrenia. At face value, the efficacy of such compounds questions the continuing utility of the dopamine hypothesis. A case in point is that of ulotaront (SEP-363856), which is in the preliminary stages of clinical evaluation.

Ulotaront is a trace amine-associated receptor (TAAR-1) and 5HT1A agonist. In pre-clinical studies the molecule blocked phencyclidine (PCP)-induced hyperactivity, pre-pulse inhibition and PCP-induced deficits in social interaction and cognition, widely used mouse phenotypes for antipsychotic drug discovery (Dedic et al., 2019). A short-term (4-week) double-blind, placebo-controlled trial showed superiority for the active compound in treating relapse of schizophrenia (Koblan et al., 2020). In the open label continuation phase of this study ulotaront maintained efficacy over 6 months (Correll et al., 2021). Clearly, further clinical evaluations are necessary to robustly establish clinical efficacy. Nevertheless, taken together the two parts of this study offer preliminary validation of non-dopaminergic mechanisms as treatments in schizophrenia. Closer examination of the putative mechanism of action of ulotaront (and chemical congeners) however posits that modulation of dopaminergic activity through effects on dopamine D2 receptors is important in understanding its effectiveness in treating schizophrenia. It has been demonstrated that activation of the TAAR1 receptor results in heterodimerisation with D2-like receptors preventing dopaminergic hyperactivity, in particular through a decrease in mesolimbic dopamine neuron firing (Berry et al., 2017). However, the administration of TAAR1 agonists is associated with enhanced prefrontal cortical dopamine levels and neuronal activation, suggesting an ability to enhance cognitive function and relieve negative symptoms (Revel et al., 2013).

While other actions of the drug are also likely to contribute to its efficacy, these too have been accommodated within the expanded dopaminergic hypothesis. Dopamine hyperactivity is posited to arise as a result of glutamatergic hypoactivity within an indirect circuit and hyperactivity within a direct circuit (Carlsson and Carlsson, 1990). TAAR1 receptor regulation of NMDA receptor subunits has been reported to increase excitatory postsynaptic potentials while activation of TAAR1 can overcome NMDA receptor blockade (Berry et al., 2017). TAAR1 activation can thus overcome hypoactivity in the direct glutamatergic circuit and prevent dopamine hyperactivity (Berry et al., 2017). Furthermore, the 5HT1A agonist properties of ulotaront, in common with other 5HT1A agonists, can stimulate the release of dopamine in the medial prefrontal cortex contributing to cognitive enhancing properties of the drug. The TAAR1 system has been described as an ‘endogenous rheostat’ which serves to modulate the activity of other neurotransmitter systems (Berry et al., 2017). Agonist actions affecting the dopaminergic system would appear to offer a novel approach to the treatment of schizophrenia without some of the drawbacks of existing agents. At a theoretical level ulotaront (and related TAAR1 agonists such as ralmitaront) hold promise as treatments devoid of neurological and possibly metabolic side effects. While not the only non-dopaminergic agent in clinical development for schizophrenia, the combination of xanomeline–trospium, a muscarinic-cholinergic agonist, is another medication in development (Weiden et al., 2022), exploitation of the TAAR1 class of agent offers the potential for superior outcomes in the disorder and a deeper understanding of neurobiological underpinnings.

Accommodation of the actions of TAAR1 agonists within the framework of the dopamine hypothesis represents a quintessential Kuhnian response: reconciliation of the observed data within an existing hypothesis, irrespective of the ‘goodness of fit’, what Kuhn termed the ‘resumption of normal science’. The interconnectedness of neural circuits and their dependence on the actions of multiple neurotransmitters offers a broad church within which explanations of symptoms and therapeutic modalities can be accommodated. Thus, dopamine as a common final pathway of psychosis and antipsychotic drug effects has proven adaptable, even if ultimately shown to be incorrect, and able to incorporate actions at glutamatergic, serotonergic, and now trace amine dependent receptors within its rubric. For the time being, it appears that the dopamine hypothesis will remain as the dominant paradigm of psychosis and antipsychotic drug action. A paradigm shift awaits further developments.