Treatment resistance in psychiatry: state of the art and new directions

Abstract

Treatment resistance affects 20-60% of patients with psychiatric disorders; and is associated with increased healthcare burden and costs up to ten-fold higher relative to patients in general. Whilst there has been a recent increase in the proportion of psychiatric research focussing on treatment resistance (R² = 0.71, p < 0.0001), in absolute terms this is less than 1% of the total output and grossly out of proportion to its prevalence and impact. Here, we provide an overview of treatment resistance, considering its conceptualisation, assessment, epidemiology, impact, and common neurobiological models. We also review new treatments in development and future directions. We identify 23 consensus guidelines on its definition, covering schizophrenia, major depressive disorder, bipolar affective disorder, and obsessive compulsive disorder (OCD). This shows three core components to its definition, but also identifies heterogeneity and lack of criteria for a number of disorders, including panic disorder, post-traumatic stress disorder, and substance dependence. We provide a reporting check-list to aid comparisons across studies. We consider the concept of pseudo-resistance, linked to poor adherence or other factors, and provide an algorithm for the clinical assessment of treatment resistance. We identify nine drugs and a number of non-pharmacological approaches being developed for treatment resistance across schizophrenia, major depressive disorder, bipolar affective disorder, and OCD. Key outstanding issues for treatment resistance include heterogeneity and absence of consensus criteria, poor understanding of neurobiology, under-investment, and lack of treatments. We make recommendations to address these issues, including harmonisation of definitions, and research into the mechanisms and novel interventions to enable targeted and personalised therapeutic approaches.

Fig. 1. Scatterplot showing the relationship between the number of papers published in the field of treatment resistance in psychiatry and time (2000–2019).

The number of papers published on treatment resistance in psychiatry are presented as a percentage of the total number of publications in the field of psychiatry overall. The solid blue line corresponds to the regression estimate with the corresponding 95% confidence interval, indicated by grey shading, showing a significant increase in the percentage of psychiatric research focusing on treatment resistance over the last two decades.

Fig. 2. Treatment resistance consists of three core components.

Establishing treatment resistance requires concurrent confirmation of the following: 1) that the correct psychiatric diagnosis has been made; 2) that a patient has received adequate treatment; 3) that symptoms have not adequately responded despite treatment.

Fig. 3. Pseudo-resistance to treatment in psychiatry: treatment related factors.

A Poor concordance with medication or forgetfulness may result in insufficient drug being taken to achieve a therapeutic response, B Polymorphisms in P-glycoproteins in the gut endothelia may result in poor absorption of drugs and insufficient drug exposure. C Smoking tobacco induces expression of CYP450 enzymes, particularly CYP1A2, in the liver (D) resulting in enhanced break down of psychiatric medication metabolised by these enzymes. Polymorphisms in CYP450 enzymes that enhance their activity or co-administration of other psychiatric/non-psychiatric medications that act as enzyme inducers will have a similar effect. E Poor brain accumulation of drug owing to poor blood brain barrier permeability and/or polymorphisms in P-glycoprotein may result in insufficient central nervous system drug levels to achieve a therapeutic response.

Fig. 4. Algorithm for approaching non-response to treatment in psychiatric illness.

Persistent symptoms despite treatment could be due to treatment resistance or due to other factors that give the impression of treatment resistance when in fact adequate treatment has not been received (pseudo-resistance). Pseudo-resistance may be secondary to an incorrect primary diagnosis/psychiatric comorbidity/substance abuse, or be treatment related, including poor treatment adherence, malabsorption of drug, poor blood brain barrier penetrance of drug or fast metabolism of drug (see Table 2 and Fig. 3).

Fig. 5. Main putative disease models to explain treatment resistance in psychiatry.

A Treatment responsive and resistant illnesses are defined by the same neurobiological alterations (neurotransmitter 1), however the alterations are more marked in patients with treatment resistance. As such, higher doses of drugs targeting neurotransmitter system 1 (orange triangles) are required for therapeutic benefit. B Treatment responsive and resistant illnesses are defined by different underlying neurobiological mechanisms, for example two different neurotransmitter systems (neurotransmitters 1 and 2). As such, a drug targeting only neurotransmitter system 1 will be ineffective in patients with dysregulation in neurotransmitter system 2. C Treatment resistance arises from a combination of neurobiological alterations seen in responsive illness (neurotransmitter 1) in combination with a different neurobiological process (neurotransmitter 2). As such, treatments that act on both targets are likely to be needed.