Treating schizophrenia – Thinking beyond the choice of antipsychotic

English and Castle suggest an updated Royal Australian and New Zealand College of Psychiatrists (RANZCP) Clinical Practice Guidelines (CPG) for schizophrenia and related disorders should be informed ‘most explicitly’ by the side-effect profile and risk amelioration potential of available antipsychotics. Given the (relative) comparable efficacy of most available antipsychotics, this approach is not unreasonable and would be in line with several other contemporary international guidelines. In addition, it would be prudent to incorporate the (appropriately informed) views and preferences of the patient and their carer(s) in a shared decision-making exercise. The authors highlight three key side effects of antipsychotics deemed most physically significant:

When it comes to psychotropics, there are arguably no such thing as ‘side’ effects – there are just ‘effects’. Effects may be interpreted as beneficial or adverse depending on the person, time and circumstances. For example, sedation from olanzapine may be considered beneficial in a person tormented by psychosis and/or insomnia, but adverse in a person attending a 9 am job interview. Suffice to say, the three listed ‘side’ effects are universally considered adverse (perhaps with the rare exception of a post-partum woman with inadequate milk supply prescribed a ‘galactagogue’ to induce hyperprolactinaemia). The word ‘side’ implies there is a ‘central’ effect, but psychotropic drugs have wide-ranging effects on multiple receptors across multiple brain regions – beneficial effects are tied together with other, more global effects. One such example is the indiscriminate blockade of D2 across multiple brain pathways by antipsychotics. In the mesolimbic pathway, acutely, it is hypothesised this results in diminution of positive psychotic symptoms. In the nigrostriatal pathway, chronically, it is hypothesised this may be a cause of TD.

TD is a chronic, potentially disabling motor syndrome characterised by persistent, repetitive abnormal involuntary movements. As English and Castle point out, upregulation of the D2-receptor in response to chronic D2 blockade is implicated in the evolution of TD. This, along with an increase in the proportion of D2-receptors in the ‘high affinity’ state for dopamine, is considered a neuroadaptive homeostatic response to an altered neurophysiological environment – a process described elsewhere as ‘dopamine supersensitisation’ (Chouinard et al., 2017). Some evidence, but importantly not all, suggests that the D2-partial agonist aripiprazole may not induce the same D2-receptor upregulation seen with chronic exposure to D2-antagonists (at least in animal models). In line with this, the authors reference a recent meta-analysis where aripiprazole carried the lowest risk of TD and they reasonably speculated about the newer D2/D3 partial agonists (brexpiprazole and caripirazine) carrying a similarly lower risk. In sum, the authors endorse an aetiological model for TD that involves upregulation of D2-receptors consequent to chronic D2-receptor antagonism and imply the problem is significant enough to warrant specifically listing agents that may reduce the risk of this happening. However, TD is considered (primarily) the manifestation of nigrostriatal dopamine supersensitisation – given antipsychotics indiscriminately block D2-receptors across the brain, what about the consequences of supersensitisation elsewhere?

Supersensitivity psychosis (SP) refers to psychosis hypothetically precipitated by stimulation of a ‘supersensitised’ striatal (particularly mesolimbic) dopaminergic post-synaptic membrane. SP may be considered aetiologically analogous to TD, albeit in a different brain pathway. The clinical paradox of SP is an increased risk of psychotic relapse in those taking long-term (>3 months) antipsychotics following dose reduction, switch or cessation; or ‘breakthrough psychosis’ in those on continuous treatment. From a neurophysiological standpoint, efforts to minimise the risk of TD should therefore also minimise the risk of SP – English and Castle’s nod to the partial agonists may be apt here – choice of antipsychotic based on pharmacodynamic profile is important. Dose, frequency and duration of exposure, however, are all variables that also affect the risk of developing TD and/or SP – the authors mention none of these. If we accept the putative pathophysiology of TD then we should seriously consider the same for SP and therefore reappraise those factors deemed material to the risk of developing both.

Chouinard recently proposed guidelines for the prevention of both TD and SP – the choice of antipsychotic formed but one of several recommendations (Chouinard et al., 2017). First, the guidelines highlight the role of prevention and detection (e.g. monitoring with validated scales, using lower risk agents at lower doses and, where clinically appropriate, increasing the interval between doses and/or discontinuing after a several-month taper). Second, they discuss the choice of antipsychotic (e.g. using second generation antipsychotics (SGAs) associated with few/no movement disorders including low-dose partial D2-agonists before the development of TD/SP [using after may worsen TD/SP], and avoiding higher risk drugs including most first generation antipsychotics (FGAs)). Finally, they suggest adding an adjunctive low-dose anticonvulsant (rationale provided). Though theoretically sensible, such recommendations have gained little traction in mainstream practice. However, emerging evidence supporting such “minimal medication” approaches to psychosis may just change this.

Expressing his concerns about the development of D2 supersensitivity consequent to continuous antipsychotic treatment, eminent schizophrenia researcher Robin Murray recently stated that 40% of those with FEP-in-remission should achieve good long term outcomes with either no antipsychotic medication, or very low doses (Murray et al., 2016). The current RANZCP recommendations for continuous antipsychotic treatment of all cases of first episode psychosis (FEP) for a minimum of 2–5 years at ‘target doses’ are thus at odds with this, and may actually be harmful to a significant proportion of people.

A recent Australian study examined 90 patients with FEP deemed ‘low risk’, randomised to antipsychotic or placebo (with intensive psychosocial intervention). At 6 months there was no difference in symptomatic or functional outcomes (Francey et al., 2020). In the United Kingdom, Morrison et al randomised 61 patients with FEP to either antipsychotic, psychological intervention or a combination. At 6 months, all three groups had improved significantly on the PANSS, with the highest proportion of ‘responders’ in the psychological intervention-only group (Morrison et al., 2020). Two systematic reviews, involving 2,250 people with psychosis or schizophrenia, have compared ‘minimal medication’ approaches (including psychosocial treatments) with antipsychotic treatment-as-usual (TAU). Compared to TAU, ‘minimal medication’ approaches were at least as effective in attenuating symptoms and improving functioning with no increased risk of harms (Cooper et al., 2021). In response to pressure from consumer groups, some countries have created alternatives to antipsychotic-TAU-based services for psychosis. Since 2015, Norway has established 14 hospital units with ‘drug-free’ treatment programmes that include outpatient services. In the United States, a ‘Soteria House’ operates in the state of Vermont (Cooper et al., 2021). Formal evaluations of these programmes are pending, but early reports are favourable.