Hostname: page-component-7bb8b95d7b-s9k8s Total loading time: 0 Render date: 2024-09-27T02:25:44.694Z Has data issue: false hasContentIssue false

How to classify antipsychotics: time to ditch dichotomies?

Published online by Cambridge University Press:  14 November 2023

Robert A. McCutcheon*
Affiliation:
Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford, UK; Oxford Health NHS Foundation Trust, Oxford, UK; and Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King's College London, London, UK
Alistair Cannon
Affiliation:
South London and Maudsley NHS Foundation Trust, London, UK
Sita Parmer
Affiliation:
South London and Maudsley NHS Foundation Trust, London, UK
Oliver D. Howes
Affiliation:
Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King's College London, London, UK; South London and Maudsley NHS Foundation Trust, London, UK; Institute of Clinical Sciences, Faculty of Medicine, Imperial College London, London, UK; and H. Lundbeck A/S, Copenhagen, Denmark
*
Correspondence: Robert A. McCutcheon. Email: robert.mccutcheon@psych.ox.ac.uk
Rights & Permissions [Opens in a new window]

Summary

The dichotomies of ‘typical/atypical’ or ‘first/second generation’ have been employed for several decades to classify antipsychotics, but justification for their use is not clear. In the current analysis we argue that this classification is flawed from both clinical and pharmacological perspectives. We then consider what approach should ideally be employed in both clinical and research settings.

Type
Discussion
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press on behalf of the Royal College of Psychiatrists

Over 20 antipsychotics are licensed for the treatment of schizophrenia. Given this number, a classification system is a potentially useful heuristic for both clinician and researcher. In the past three decades the predominant classification of antipsychotic drugs has been into ‘typical’ and ‘atypical’ groupings. More recently the terms ‘first’ and ‘second generation’ have been used (Fig. 1(a)), but in practice this is used as a synonym for the typical/atypical classification. An ideal classification system should map to the pharmacological and/or clinical effects of the drugs and it is not clear that this approach achieves this. More recently a more pharmacologically precise approach, ‘neuroscience-based nomenclature’, has been proposed but it is yet to be widely adopted (Fig. 1(a)). In the current paper we discuss the typical/atypical classification criteria, the evidence supporting their use and drawbacks of the classification, before discussing alternatives.

Fig. 1 Quantifying atypicality.

(a) Trends in nomenclature: to quantify the use of the ‘atypical’ terminology we searched PubMed using the search term ‘atypical antipsychotic’ to demonstrate what percentage of publications using the word ‘antipsychotic’ have employed this method of classification. We did the same for ‘generation antipsychotic’. We then searched for citations of the first paper describing and recommending the pharmacology-based neuroscience-based nomenclature (NbN).Reference Zohar, Nutt, Kupfer, Moller, Yamawaki and Spedding1 The figure shows that the use of the typical/atypical classification remains frequent and, although it has declined, it has been replaced by ‘first/second generation’ terminology, which essentially duplicates it. FDA, US Food and Drug Administration. (b) Atypicality, efficacy and side-effects: the hatched bars show antipsychotics grouped into typical (vertical hatching) and atypical (diagonal hatching), as defined in clinical guidelines, or on the basis of receptor profile when this was not available (e.g. molindone).Reference Taylor2,Reference Seeman3 The next three columns show the relative side-effect burden and efficacy according to a recent network meta-analysis,Reference Huhn, Nikolakopoulou, Schneider-Thoma, Krause, Samara and Peter4 whereby a lighter colour indicates a lower ranking for side-effect burden or higher ranking for efficacy. ‘n.a.’ indicates that data are not available; EPSEs, extrapyramidal side-effects. Atypical drugs should have lighter colours across all three domains than typical drugs. However, the figure illustrates that neither side-effect burden or efficacy neatly maps to this classification scheme. (c) Pharmacological differences between typical and atypical drugs: the hatched bars show antipsychotics grouped into typical and atypical, as in part (b). The next three columns show the relative affinity for the dopamine D2 receptor, the serotonin (5-HT) 2A receptor and the ratio between the two. Affinities obtained from McCutcheon et al.Reference McCutcheon, Harrison, Howes, McGuire, Taylor and Pillinger5

What is atypicality?

The term ‘atypical’ was first used in 1975 to describe antipsychotic medications such as clozapine, thioridazine, and sulpiride, which were observed to induce catalepsy in rats to a lesser degree than ‘typical’ antipsychotics such as haloperidol and chlorpromazine.Reference Costall and Naylor6 A formal definition, however, was not elaborated until the 1990s with a review by Kinon & Lieberman in which three criteria were specified: (a) a lack of extrapyramidal side-effects (EPSEs) and tardive dyskinesia; (b) increased therapeutic efficacy; and (c) minimal elevation of prolactin levels.Reference Kinon and Lieberman7

There was not, however, an attempt by the field to systematically categorise antipsychotic drugs according to these criteria. Apart from clozapine, drugs developed prior to the approval of risperidone in 1993 were generally understood to show ‘typical’ properties, whereas those developed subsequently came under the atypical umbrella. Figure 1(b) shows this classification against meta-analytical estimates of efficacy and side-effect burden for current antipsychotics. This illustrates that atypical drugs are somewhat more likely to have a lower propensity for inducing EPSEs and hyperprolactinaemia than the typical counterparts. However, the boundary is not clear-cut, with considerable overlap between groups across criteria. For example, some atypical drugs, such as risperidone and paliperidone, appear more likely to induce hyperprolactinaemia than several typical drugs, such as pimozide or haloperidol. Similarly, several atypical drugs, such as cariprazine and molindone, are more likely to cause EPSEs than typical drugs such as chlorpromazine and thioridazine.Reference Huhn, Nikolakopoulou, Schneider-Thoma, Krause, Samara and Peter4 Moreover, there is no distinction in efficacy between the two categories. Even the archetypal antipsychotic, clozapine, although on average more effective than other antipsychotics, does not clearly separate in terms of efficacy from all typical drugs.Reference Huhn, Nikolakopoulou, Schneider-Thoma, Krause, Samara and Peter4,Reference Mizuno, McCutcheon, Brugger and Howes8

Is the situation even less clear-cut? The limitations of side-effect comparisons

Although the above demonstrates some shortcomings of the typical/atypical classification there does still appear to be, on average, a greater propensity for EPSEs and hyperprolactinaemia to occur following treatment with typical compared with atypical antipsychotics. However, even this difference is probably exaggerated owing to the nature of the trials that the meta-analytical estimates of side-effect burden are based on.

Antipsychotics antagonise dopamine D2 receptors (D2R) across the striatum,Reference Kaar, Natesan, McCutcheon and Howes9 including regions critical to normal movement, and it is therefore understandable that D2R blockade can also lead to EPSEs. Positron emission tomography (PET) studies have shown that EPSEs are related to D2R occupancy, and the risk is greatest when occupancy of dopamine receptors by dopamine antagonists exceeds ~85%.Reference Kapur, Zipursky, Jones, Remington and Houle10,Reference Farde, Nordström, Wiesel, Pauli, Halldin and Sedvall11

Receptor occupancy is related to dose and therefore, as expected, higher doses are associated with a greater risk of EPSEs.Reference Seeman12 PET studies indicated that the doses of typical antipsychotics used in many clinical trials, particularly the older ones, would be expected to result in D2R occupancy >85%, whereas the doses of atypical antipsychotics used in clinical trials tend to be associated with D2R occupancy <85%Reference Leucht, Leucht, Huhn, Chaimani, Mavridis and Helfer13 This difference in D2R occupancy is likely to account for some of the higher rates of EPSEs seen in older trials of typical agents. Even in head-to-head trials between atypical and typical drugs the doses of the typical agents have frequently been associated with markedly higher D2R occupancy. For example, an important early trial of olanzapine used olanzapine doses in the range of 5–15 mg daily, compared with haloperidol doses of 15 mg daily in the comparator arm.Reference Beasley, Tollefson, Tran, Satterlee, Sanger and Hamilton14 A dose of 15 mg olanzapine has been shown to be associated with around 70% occupancy of striatal D2 receptors.Reference Kapur, Zipursky, Remington, Jones, DaSilva and Wilson15 The dose of haloperidol required for similar occupancy is around 2.5 mg, with doses above 5 mg approaching 90% occupancy.Reference Kapur, Zipursky, Jones, Remington and Houle10,Reference Kapur, Zipursky, Roy, Jones, Remington and Reed16 It is therefore unsurprising that EPSEs would occur with greater frequency in the haloperidol arm as receptor occupancies would be expected to be markedly higher. When between-class comparisons have been restricted to trials that have used doses of typical antipsychotics expected to give similar rates of D2R occupancy to the atypical dose, rates of EPSEs between classes are similar.Reference Leucht, Wahlbeck, Hamann and Kissling17 Moreover, doses of atypical antipsychotics that would be expected to results in D2R occupancy >85% are associated with higher rates of EPSEs.Reference Rochon, Stukel, Sykora, Gill, Garfinkel and Anderson18 Therefore, much of the difference observed in EPSEs between typical and atypical drugs may be an artefact of dosing differences leading to differences in receptor occupancy.

Hyperprolactinaemia also results from dopaminergic antagonism, and therefore the arguments made above for EPSEs also apply to prolactin effects.Reference Lobo, Whitehurst, Kaar and Howes19 A mechanistic distinction between EPSEs and hyperprolactinaemia is that in the latter the side-effect arises from antagonism at the pituitary, which unlike the striatum, is located outside of the blood–brain barrier. This means that drugs with poor penetrance of the barrier are more likely to induce hyperprolactinaemia.Reference Kapur, Langlois, Vinken, Megens, De Coster and Andrews20 There is no evidence, however, for a distinction in blood–brain barrier penetration along typical/atypical lines.Reference Loryan, Melander, Svensson, Payan, König and Jansson21

It is more difficult to obtain drug-specific risks of tardive dyskinesia as clinical trials are often of insufficient duration to observe its emergence. Meta-analysis of relevant clinical trial data does, however, suggest that that the risk may be higher following long-term treatment with typical as opposed to atypical antipsychotics.Reference Carbon, Kane, Leucht and Correll22 In this case the differences do not appear to be driven by dosing differences.Reference Carbon, Kane, Leucht and Correll22 The limited number of studies that are available, however, make it difficult to determine whether this is truly a class effect, for example when individual compounds were examined there was no evidence that quetiapine, paliperidone or ziprasidone had a reduced propensity for inducing tardive dyskinesia. When specific pharmacodynamic factors are considered it appears that D2R affinity rather than ‘atypicality’ may be the factor of interest.Reference Tenback and van Harten23

Although not a component of the original criteria for atypicality, metabolic side-effects have increasingly been associated with atypical antipsychotics. Again, however, when the evidence is examined, it does not divide neatly along class lines. Several atypical antipsychotics, such as lurasidone, ziprasidone and molindone, show less propensity to induce weight gain than typical antipsychotics such as chlorpromazine and thioridazine (Fig. 1(c)).

Efficacy

Atypical antipsychotics were proposed not only to possess a more benign side-effect profile, but also to display greater efficacy. Initial trials supported this stance, but as evidence accumulated the proposed benefit appeared to be less clear.Reference Huhn, Nikolakopoulou, Schneider-Thoma, Krause, Samara and Peter4,Reference Leucht, Corves, Arbter, Engel, Li and Davis24,Reference Tollefson, Beasley, Tran, Street, Krueger and Tamura25 A major blow to the hypothesis were the findings of the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE), funded by National Institute of Mental Health (NIMH).Reference Lieberman, Stroup, McEvoy, Swartz, Rosenheck and Perkins26 In its first phase CATIE randomised over 1000 patients to either a typical antipsychotic, perphenazine, or one of four atypical drugs (olanzapine, risperidone, quetiapine or ziprasidone). Participants randomised to the typical treatment were no more likely to discontinue their medication owing to a lack of effectiveness than those randomised to one of the atypicals. Two European studies had a similar rationale, the Cost Utility of the Latest Antipsychotic Drugs in Schizophrenia Study (CUtLASS 1, n = 227) trial provided similar findings to CATIE,Reference Jones, Barnes, Davies, Dunn, Lloyd and Hayhurst27 but the European First Episode Schizophrenia Trial (EUFEST, n = 500) did find that haloperidol was associated with a greater risk of all-cause discontinuation than several atypical antipsychotics.Reference Kahn, Fleischhacker, Boter, Davidson, Vergouwe and Keet28 Later meta-analyses have confirmed that there is no clear distinction in efficacy along typical/atypical group lines.Reference Huhn, Nikolakopoulou, Schneider-Thoma, Krause, Samara and Peter4

Similarly, early trials suggested that the atypical compounds were not only advantageous in terms of psychotic symptoms, but also that they showed a benefit in treating cognitive symptoms, a crucial domain given that no existing treatments appeared to show significant benefits here.Reference McCutcheon, Keefe and McGuire29 Again CATIE produced findings in contradiction of this hypothesis, with individuals on the typical treatment showing the greatest improvement in neurocognitive outcomes,Reference Keefe30 and network meta-analyses do not show any clear pattern of superiority for atypical over typical.Reference Baldez, Biazus, Rabelo-da-Ponte, Nogaro, Martins and Kunz31

Pharmacology

The criteria proposed to distinguish typical and atypical drugs solely reflect clinical considerations, but a distinction in pharmacodynamic mechanisms is implicit and led to considerable efforts to investigate proposed underlying mechanisms.Reference Duncan, Zorn and Lieberman32 We have demonstrated that the typical/atypical divide does not accurately separate drugs in terms of clinical effects but there may be a value to its continued employment if it summarises fundamental pharmacological difference between the two groups.

Trials of clozapine in the 1980s demonstrated the drug's properties of both improving symptoms in patients where other drugs had failed and having a low risk of hyperprolactinaemia and movement side-effects. This motivated efforts to develop compounds that shared clozapine's pharmacological features, in the hope they would also share its clinical profile. Although antagonism of the dopamine D2 receptor had already been established as central to antipsychotic efficacy, the effects of clozapine implied the existence of additional mechanisms suitable for therapeutic exploitation.

High affinity for the serotonin (5-HT) 2A receptor relative to the affinity for the D2 receptor was proposed as a key factor underlying atypicality.Reference Meltzer33 Fig. 1(c) summarises the ratio of D2 to 5-HT2A affinities across the atypical/typical divide. This shows that, although the D2 ratio separates a number of atypical and typical drugs, there are some notable exceptions. In particular, the D2/5-HT2A ratios of amisulpride, lurasidone and molindone overlap with those seen among typicals, whereas they would put thioridazine and chlorpromazine among the atypicals. Similarly, brexpiprazole fits with the atypical pattern of low D2/5-HT2A ratio, whereas cariprazine fits the typical pattern. Thus, the groupings do not reflect D2/5-HT2A ratios.

Although some separation exists between typical and atypical compounds based on D2/5-HT2A ratio, it is unclear as to why this mechanism should be afforded priority over others, given that its clinical relevance is unclear. That the ratio is unlikely to be central to efficacy is indicated by the fact that the most efficacious non-clozapine antipsychotic is amisulpride,Reference Huhn, Nikolakopoulou, Schneider-Thoma, Krause, Samara and Peter4 a drug that possesses negligible affinity for the 5-HT2A receptor. It is also clear that other receptor systems play more important roles in determining side-effect burden, such as the histamine H1 receptor for weight gain.Reference Kroeze, Hufeisen, Popadak, Renock, Steinberg and Ernsberger34 An alternative approach to selectively focusing on specific receptors is to examine the full receptor profile of each drug in an unbiased data-driven fashion. Using this method it is apparent that the typical/typical divide captures only a minimum of pharmacological differences at best.Reference McCutcheon, Harrison, Howes, McGuire, Taylor and Pillinger5

Other mechanisms proposed to underlie atypicality include ‘fast dissociation’ of drugs from the D2 receptor. The affinity of an antipsychotic (i.e. the K i) is determined by the rate at which the drug binds to (k on) and the rate at which it dissociates from (k off) the receptor. In practice, however, k on hardly varies between antipsychotics, which means the dissociation rate is a proxy for affinity, with compounds displaying fast dissociation possessing a low affinity.Reference Kapur and Seeman35 From Fig. 1(c) we can see how an archetypal atypical compound, risperidone, shows greater affinity (and thereby slower dissociation) for the D2 receptor than typical compounds such as sulpiride and thioridazine. Although koff may well be an important mediator of clinical effects, it varies gradually across compounds (Fig. 1(c)) and it is therefore hard to see how it could be used to delineate a dichotomy.

In summary, the typical/atypical dichotomy was built on a presupposition that the antipsychotics that came to market in the years following the Food and Drug Administration (FDA) approval of risperidone differed from earlier medications in terms of side-effects and clinical efficacy. However, it has subsequently become clear that observed differences in side-effect profile primarily reflected differences in dosing and that efficacy differences do not separate with categorical boundaries. This is neatly illustrated by the fact that the original antipsychotic, chlorpromazine, an archetypal typical antipsychotic, is highly similar to one of the most recently approved antipsychotics, lurasidone, on all the Kinon & Lieberman criteriaReference Kinon and Lieberman7 and as regards D2/5-HT2A ratio (Fig. 1c). The fact that these two compounds are more similar to each other than to other compounds within their ‘class’ shows how the classification could lead to false distinctions in a research setting. It also shows that it is not helpful in a clinical setting, as a clinician may consider they are making a marked switch in treatment strategy when in fact they are changing to a drug with similar side-effect and efficacy profiles.

Alternative classification schemes

Broad classification schemes such as the World Health Organization's Anatomical Therapeutic Chemical (ATC) system primarily classify medications on the basis of clinical indication, with more fine-grained categorisation based on chemical structure. Although broad categories are useful for facilitating epidemiological monitoring of drug use, the system is not suitable for clinical use, given that the chemically based subgroupings are distinct from clinical effects and unfamiliar to clinicians.

An extension of the typical/atypical classification that has seen widespread adoption in both research and clinical settings is the addition of an extra grouping of ‘third-generation’ drugs, namely aripiprazole, cariprazine and brexpiprazole. This grouping appears justified in that the compounds share a common pharmacological mechanism (partial agonism of the dopamine D2 receptor) and a similar clinical profile.Reference Huhn, Nikolakopoulou, Schneider-Thoma, Krause, Samara and Peter4 This common property accounts for the fact that these drugs are not associated with raised prolactin concentrations and can even be used as augmentation agents to reduce prolactin levels in cases of dopamine antagonism-associated hyperprolactinaemia, presumably because the D2 partial agonism counters the D2 antagonist's effects on D2 receptors in the pituitary.Reference Zhu, Zhang, Siafis, Zhuo, Zhu and Wu36 D2 partial agonism is also thought to account for the fact that rates of extrapyramidal side-effects are much lower than would be expected given the high striatal D2 receptor occupancy levels (generally above 80%) seen at clinical doses with these drugs.Reference Sparshatt, Taylor, Patel and Kapur37 This does not, however, address the issues outlined above that still pertain to most antipsychotics in the typical/atypical groupings. Moreover, inventing the term ‘third generation’ to categorise them rarefies the typical/atypical categorisation and also suggests a linear evolution in the development of antipsychotics, whereas, in fact, aripiprazole was developed before a number of drugs usually included in the second-generation category.

The neuroscience-based nomenclature (NbN) was developed to address the fact that indication-based classification systems do not reflect the underlying pharmacology, often have little bearing on clinical effects, and that existing schemes such as typical/atypical have the flaws outlined above.Reference Zohar, Nutt, Kupfer, Moller, Yamawaki and Spedding1 In many respects this is an advance on the typical/atypical scheme in that there is an attempt made to reflect pharmacology in the scheme, although it has not seen widespread uptake yet (Fig. 1(a)). A potential drawback of the NbN scheme is, however, that it selects certain aspects of the pharmacology over others, based on expert consensus that these aspects are central to the actions of the drugs, and uses these to the make categories. For example although dopaminergic, serotonergic and adrenergic mechanisms are used in the scheme, histaminergic affinities do not play a role. This is despite the fact that antagonism of the histamine H1 receptor is central to the sedative and weight gain properties of a number of psychotropics.Reference Kroeze, Hufeisen, Popadak, Renock, Steinberg and Ernsberger34

An alternative to an expert consensus approach is to use a data-driven approach. This was recently applied to classify antipsychotics on the basis of their receptor affinity profile,Reference McCutcheon, Harrison, Howes, McGuire, Taylor and Pillinger5 using a multivariate approach to identify clusters of drugs with similar receptor profiles. This identified four clusters, one with high affinity for muscarinic receptors (e.g. olanzapine and quetiapine), one with relatively low antagonism of the dopamine D2 receptor (e.g. the partial agonists and lurasidone), one with serotonergic antagonism (e.g. risperidone) and one with relatively pure dopaminergic antagonism (e.g. amisulpride). These clusters also mapped to side-effect profiles with greater accuracy than the approaches we have discussed above. A drawback of a data-driven approach, however, is that all receptors are assigned an equal level of importance regardless of their magnitude of impact in mediating clinically relevant effects.

Drugs that are primarily muscarinic receptor agonists or trace amine-associated receptor 1 (TAAR1) agonists have recently shown efficacy in large clinical studies, and these appear distinct from existing antipsychotics because they do not block D2 receptors and show different side-effect profiles.Reference Brannan, Sawchak, Miller, Lieberman, Paul and Breier38,Reference Koblan, Kent, Hopkins, Krystal, Cheng and Goldman39 Although novel mechanisms of action have the potential to advance the treatment of psychotic disorders, care must be taken when considering how to categorise these compounds. The role of market incentives in shaping language should not be underestimated. It is likely that this played a significant role in cementing the current typical/atypical dichotomy, and any novel categorisation should not be guided by a desire to promote novel compounds over their off-patent competitors. If these new agents become approved, it could be that a single category of ‘dopamine receptor blocker’ subsumes the typical/atypical dichotomy to distinguish current drugs from new entrants. However, although pharmacologically accurate, this would obscure important pharmacological and clinical differences between existing compounds, which could be detrimental to patient care. Moreover, the mechanism underlying the action of any new drug would need to be established in clinical studies before a new classification could be justified. For these reasons, we caution against a rush to new categorisations if novel drugs are approved and suggest that it is preferable to keep the pharmacologically based categories described above until there is sufficient understanding of the clinical pharmacology of new drugs.

Fundamentally, any form of classification is a form of dimensionality reduction and so entails a loss of information. The loss of precision inherent when using groupings must be compensated for adequately in terms of any gains obtained in terms of heuristic value. An alternative to groupings is to treat each compound individually. This approach means each drug would be considered in terms of its unique pharmacology. We argue that this is preferable to the typical/atypical classification because of the flaws in both the principles and practical application of the latter. However, considering each drug separately has limitations. For example, if researchers wish to investigate common underlying mechanisms they need groupings, and busy clinicians may find it challenging when faced with making rapid treatment recommendations with over 20 drugs and no schema to help guide the process. Fortunately, new digitally aided approaches can facilitate what would otherwise be an infeasible task in clinical practice. For example, a tool has been developed without the need for a classification scheme that allows antipsychotics to be ranked by patients and clinicians on the basis of multiple side-effect preferences to aid decision-making.Reference Pillinger, Howes, Correll, Leucht, Huhn and Schneider-Thoma40

In terms of clinical guidelines, our review of the efficacy and side-effect data makes it clear that there is minimal benefit to using the typical/atypical groupings and, if compounds are to be specified, they should be individually described. When it comes to research it is possible to use bespoke groupings that better address the research question. For example, if the hypothesis is that the D2/5-HT2A ratio is critical for clinical efficacy, then it is most logical to make groupings explicitly along these lines. Likewise, if the question is whether affinity for histamine 1 receptors underlies weight gain, then grouping based on H1 affinity is a better way to test this.

Conclusions

The classification of antipsychotics into two categories of typical and atypical has been the dominant taxonomic approach for over 30 years. Over this period, increasing evidence has accumulated that this category is fundamentally flawed in conception and application. As a result, the dichotomy now serves more to obscure than illuminate differences between compounds and we recommend that it is no longer used. Alternatives include NbN or a data-driven approach. These have the advantage over the typical/atypical classification of not being based on flawed criteria that are not applied consistently in practice. Nevertheless, classification inevitably involves some loss of information that may in some circumstances outweigh its benefits, and different classifications may be more or less appropriate depending on the issue at hand. We recommend that researchers and clinicians consider whether a given system is fit for their specific purpose and whether to use one at all.

Data availability

Data availability is not applicable to this article as no new data were created or analysed in this work.

Author contributions

R.A.M., A.C., S.P. and O.D.H. together wrote the manuscript and all approved the final version.

Funding

R.A.M.'s work is funded by a Wellcome Trust Clinical Research Career Development (224625/Z/21/Z). This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Declaration of interest

R.A.M. has received speaker/consultancy fees from Karuna, Janssen, Boehringer Ingelheim and Otsuka and is director of a company that hosts psychotropic prescribing decision tools. O.D.H. is a part-time employee of H. Lundbeck A/S and has received investigator-initiated research funding from and/or participated in advisory/speaker meetings organised by Angellini, Autifony, Biogen, Boehringer-Ingelheim, Eli Lilly, Heptares, Global Medical Education, Invicro, Jansenn, Lundbeck, Neurocrine, Otsuka, Sunovion, Rand, Recordati, Roche and Viatris/ Mylan; he has a patent for the use of dopaminergic imaging.

References

Zohar, J, Nutt, DJ, Kupfer, DJ, Moller, H-J, Yamawaki, S, Spedding, M, et al. A proposal for an updated neuropsychopharmacological nomenclature. Eur Neuropsychopharmacol 2014; 24: 1005–14.CrossRefGoogle ScholarPubMed
Taylor, D. The Maudsley Prescribing Guidelines in Psychiatry (14th edn). Wiley-Blackwell, 2021.CrossRefGoogle Scholar
Seeman, P. Antipsychotic drugs, dopamine receptors, and schizophrenia. Clin Neurosci Res 2001; 1: 5360.Google Scholar
Huhn, M, Nikolakopoulou, A, Schneider-Thoma, J, Krause, M, Samara, M, Peter, N, et al. Comparative efficacy and tolerability of 32 oral antipsychotics for the acute treatment of adults with multi-episode schizophrenia: a systematic review and network meta-analysis. Lancet 2019; 394: 939–51.Google Scholar
McCutcheon, RA, Harrison, PJ, Howes, OD, McGuire, PK, Taylor, D, Pillinger, T. Data driven taxonomy for antipsychotic medication: a New classification system. Biol Psychiatry 2023; 94: 561–8.Google Scholar
Costall, B, Naylor, RJ. Detection of the neuroleptic properties of clozapine, sulpiride and thioridazine. Psychopharmacologia 1975; 43: 6974.Google Scholar
Kinon, BJ, Lieberman, JA. Mechanisms of action of atypical antipsychotic drugs: a critical analysis. Psychopharmacology (Berl) 1996; 124: 234.CrossRefGoogle ScholarPubMed
Mizuno, Y, McCutcheon, RA, Brugger, SP, Howes, OD. Heterogeneity and efficacy of antipsychotic treatment for schizophrenia with or without treatment resistance: a meta-analysis. Neuropsychopharmacology 2020; 45: 622–31.CrossRefGoogle ScholarPubMed
Kaar, SJ, Natesan, S, McCutcheon, R, Howes, OD. Antipsychotics: mechanisms underlying clinical response and side-effects and novel treatment approaches based on pathophysiology. Neuropharmacology 2020; 172: 107704.CrossRefGoogle ScholarPubMed
Kapur, S, Zipursky, R, Jones, C, Remington, G, Houle, S. Relationship between dopamine D(2) occupancy, clinical response, and side effects: a double-blind PET study of first-episode schizophrenia. Am J Psychiatry 2000; 157: 514–20.CrossRefGoogle ScholarPubMed
Farde, L, Nordström, AL, Wiesel, FA, Pauli, S, Halldin, C, Sedvall, G. Positron emission tomographic analysis of central D1 and D2 dopamine receptor occupancy in patients treated with classical neuroleptics and clozapine. Relation to extrapyramidal side effects. Arch Gen Psychiatry 1992; 49: 538–44.CrossRefGoogle ScholarPubMed
Seeman, P. Atypical antipsychotics: mechanism of action. Can J Psychiatry 2002; 47: 2738.Google Scholar
Leucht, S, Leucht, C, Huhn, M, Chaimani, A, Mavridis, D, Helfer, B, et al. Sixty years of placebo-controlled antipsychotic drug trials in acute schizophrenia: systematic review, Bayesian meta-analysis, and meta-regression of efficacy predictors. Am J Psychiatry 2017; 174: 927–42.Google Scholar
Beasley, CM Jr, Tollefson, G, Tran, P, Satterlee, W, Sanger, T, Hamilton, S. Olanzapine versus placebo and haloperidol: acute phase results of the north American double-blind olanzapine trial. Neuropsychopharmacology 1996; 14: 111–23.CrossRefGoogle ScholarPubMed
Kapur, S, Zipursky, RB, Remington, G, Jones, C, DaSilva, J, Wilson, AA, et al. 5-HT2 and D2 receptor occupancy of olanzapine in schizophrenia: a PET investigation. Am J Psychiatry 1998; 155: 921–8.Google Scholar
Kapur, S, Zipursky, R, Roy, P, Jones, C, Remington, G, Reed, K, et al. The relationship between D2 receptor occupancy and plasma levels on low dose oral haloperidol: a PET study. Psychopharmacology 1997; 131: 148–52.CrossRefGoogle ScholarPubMed
Leucht, S, Wahlbeck, K, Hamann, J, Kissling, W. New generation antipsychotics versus low-potency conventional antipsychotics: a systematic review and meta-analysis. Lancet 2003; 361: 1581–9.Google Scholar
Rochon, PA, Stukel, TA, Sykora, K, Gill, S, Garfinkel, S, Anderson, GM, et al. Atypical antipsychotics and parkinsonism. Arch Intern Med 2005; 165: 1882–8.Google Scholar
Lobo, MC, Whitehurst, TS, Kaar, SJ, Howes, OD. New and emerging treatments for schizophrenia: a narrative review of their pharmacology, efficacy and side effect profile relative to established antipsychotics. Neurosci Biobehav Rev 2022; 132: 324–61.Google Scholar
Kapur, S, Langlois, X, Vinken, P, Megens, AAHP, De Coster, R, Andrews, JS. The differential effects of atypical antipsychotics on prolactin elevation are explained by their differential blood-brain disposition: a pharmacological analysis in rats. J Pharmacol Exp Ther 2002; 302: 1129–34.Google Scholar
Loryan, I, Melander, E, Svensson, M, Payan, M, König, F, Jansson, B, et al. In-depth neuropharmacokinetic analysis of antipsychotics based on a novel approach to estimate unbound target-site concentration in CNS regions: link to spatial receptor occupancy. Mol Psychiatry 2016; 21: 1527–36.Google Scholar
Carbon, M, Kane, JM, Leucht, S, Correll, CU. Tardive dyskinesia risk with first- and second-generation antipsychotics in comparative randomized controlled trials: a meta-analysis. World Psychiatry 2018; 17: 330–40.CrossRefGoogle ScholarPubMed
Tenback, DE, van Harten, PN. Epidemiology and risk factors for (tardive) dyskinesia. Int Rev Neurobiol 2011; 98: 211–30.CrossRefGoogle ScholarPubMed
Leucht, S, Corves, C, Arbter, D, Engel, RR, Li, C, Davis, JM. Second-generation versus first-generation antipsychotic drugs for schizophrenia: a meta-analysis. Lancet 2009; 373: 3141.Google Scholar
Tollefson, GD, Beasley, CM Jr, Tran, PV, Street, JS, Krueger, JA, Tamura, RN, et al. Olanzapine versus haloperidol in the treatment of schizophrenia and schizoaffective and schizophreniform disorders: results of an international collaborative trial. Am J Psychiatry 1997; 154: 457–65.Google Scholar
Lieberman, JA, Stroup, TS, McEvoy, JP, Swartz, MS, Rosenheck, RA, Perkins, DO, et al. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med 2005; 353: 1209–23.Google Scholar
Jones, PB, Barnes, TRE, Davies, L, Dunn, G, Lloyd, H, Hayhurst, KP, et al. Randomized controlled trial of the effect on quality of life of second- vs first-generation antipsychotic drugs in schizophrenia: cost utility of the latest antipsychotic drugs in schizophrenia study (CUtLASS 1). Arch Gen Psychiatry 2006; 63: 1079–87.Google Scholar
Kahn, RS, Fleischhacker, WW, Boter, H, Davidson, M, Vergouwe, Y, Keet, IPM, et al. Effectiveness of antipsychotic drugs in first-episode schizophrenia and schizophreniform disorder: an open randomised clinical trial. Lancet 2008; 371: 1085–97.Google Scholar
McCutcheon, RA, Keefe, RSE, McGuire, PK. Cognitive impairment in schizophrenia: aetiology, pathophysiology, and treatment. Mol Psychiatry 2023: 117.Google Scholar
Keefe, RSE. Neurocognitive effects of antipsychotic medications in patients with chronic schizophrenia in the CATIE trial. Arch Gen Psychiatry 2007; 64: 633.Google Scholar
Baldez, DP, Biazus, TB, Rabelo-da-Ponte, FD, Nogaro, GP, Martins, DS, Kunz, M, et al. The effect of antipsychotics on the cognitive performance of individuals with psychotic disorders: network meta-analyses of randomized controlled trials. Neurosci Biobehav Rev 2021; 126: 265–75.Google Scholar
Duncan, GE, Zorn, S, Lieberman, JA. Mechanisms of typical and atypical antipsychotic drug action in relation to dopamine and NMDA receptor hypofunction hypotheses of schizophrenia. Mol Psychiatry 1999; 4: 418–28.CrossRefGoogle ScholarPubMed
Meltzer, HY. What's atypical about atypical antipsychotic drugs? Curr Opin Pharmacol 2004; 4: 53–7.CrossRefGoogle ScholarPubMed
Kroeze, WK, Hufeisen, SJ, Popadak, BA, Renock, SM, Steinberg, S, Ernsberger, P, et al. H1-histamine receptor affinity predicts short-term weight gain for typical and atypical antipsychotic drugs. Neuropsychopharmacology 2003; 28: 519–26.Google Scholar
Kapur, S, Seeman, P. Antipsychotic agents differ in how fast they come off the dopamine D2 receptors. Implications for atypical antipsychotic action. J Psychiatry Neurosci 2000; 25: 161–6.Google ScholarPubMed
Zhu, Y, Zhang, C, Siafis, S, Zhuo, K, Zhu, D, Wu, H, et al. Prolactin levels influenced by antipsychotic drugs in schizophrenia: a systematic review and network meta-analysis. Schizophr Res 2021; 237: 20–5.Google Scholar
Sparshatt, A, Taylor, D, Patel, MX, Kapur, S. A systematic review of aripiprazole – dose, plasma concentration, receptor occupancy, and response: implications for therapeutic drug monitoring. J Clin Psychiatry 2010; 71: 1447–56.CrossRefGoogle ScholarPubMed
Brannan, SK, Sawchak, S, Miller, AC, Lieberman, JA, Paul, SM, Breier, A. Muscarinic cholinergic receptor agonist and peripheral antagonist for schizophrenia. N Engl J Med 2021; 384: 717–26.CrossRefGoogle ScholarPubMed
Koblan, KS, Kent, J, Hopkins, SC, Krystal, JH, Cheng, H, Goldman, R, et al. A Non-D2-receptor-binding drug for the treatment of schizophrenia. N Engl J Med 2020; 382: 1497–506.Google Scholar
Pillinger, T, Howes, OD, Correll, CU, Leucht, S, Huhn, M, Schneider-Thoma, J, et al. Antidepressant and antipsychotic side-effects and personalised prescribing: a systematic review and digital tool development. The Lancet Psychiatry 2023; 10: 860–76.Google Scholar
Figure 0

Fig. 1 Quantifying atypicality.(a) Trends in nomenclature: to quantify the use of the ‘atypical’ terminology we searched PubMed using the search term ‘atypical antipsychotic’ to demonstrate what percentage of publications using the word ‘antipsychotic’ have employed this method of classification. We did the same for ‘generation antipsychotic’. We then searched for citations of the first paper describing and recommending the pharmacology-based neuroscience-based nomenclature (NbN).1 The figure shows that the use of the typical/atypical classification remains frequent and, although it has declined, it has been replaced by ‘first/second generation’ terminology, which essentially duplicates it. FDA, US Food and Drug Administration. (b) Atypicality, efficacy and side-effects: the hatched bars show antipsychotics grouped into typical (vertical hatching) and atypical (diagonal hatching), as defined in clinical guidelines, or on the basis of receptor profile when this was not available (e.g. molindone).2,3 The next three columns show the relative side-effect burden and efficacy according to a recent network meta-analysis,4 whereby a lighter colour indicates a lower ranking for side-effect burden or higher ranking for efficacy. ‘n.a.’ indicates that data are not available; EPSEs, extrapyramidal side-effects. Atypical drugs should have lighter colours across all three domains than typical drugs. However, the figure illustrates that neither side-effect burden or efficacy neatly maps to this classification scheme. (c) Pharmacological differences between typical and atypical drugs: the hatched bars show antipsychotics grouped into typical and atypical, as in part (b). The next three columns show the relative affinity for the dopamine D2 receptor, the serotonin (5-HT) 2A receptor and the ratio between the two. Affinities obtained from McCutcheon et al.5

Submit a response

eLetters

No eLetters have been published for this article.