What Does Schizophrenia Teach Us About Antipsychotics?

Abstract

Objective:

To examine how advances in our understanding of schizophrenia have shaped thinking about antipsychotics (APs) and their role in treatment.

Method:

Three specific developments in the field of schizophrenia are highlighted: advances in knowledge related to the earliest stages of schizophrenia, specifically the prodrome; reconceptualization of schizophrenia as an illness of multiple symptom domains; and greater clarification regarding the efficacy of clozapine and a new generation of APs.

Results:

Evidence indicating that negative and cognitive symptoms are present during the prodrome suggests that intervention at the time of first-episode psychosis constitutes late intervention. The limited efficacy of APs beyond psychosis argues against a magic bullet approach to schizophrenia and for polypharmacy that is symptom domain–specific. Clozapine’s unique, but limited, efficacy in treatment resistance supports subtyping schizophrenia based on treatment response.

Conclusions:

Advances in our understanding of schizophrenia have important implications regarding the current use of APs, expectations regarding response, and future drug development.

Drug development in the field of schizophrenia has flourished since the 1990s, beginning with clozapine’s reintroduction and paralleled by technological advances in areas (for example, neuroimaging and electrophysiology) that have markedly impacted thinking regarding the illness’ underlying neurobiology.

Shifts in how we conceptualize the illness clinically, combined with a greater understanding regarding how medications work within these newer frameworks, offer yet another opportunity to advance pharmacotherapy and form the basis of the present discussion. Here we comment on 3 developments that have evolved out of these advances, developments we argue hold important ramifications regarding expectations about APs, and the search for new agents.

Antipsychotic Treatment, by Definition, Constitutes Late Intervention in First-Episode Schizophrenia

The 1990s also witnessed a growing interest in the early stages of schizophrenia, specifically FEP. Driving this line of thinking was evidence that considerable delays in implementing treatment often occur, and, further, delayed treatment (that is, DUP) compromises outcome. While a body of evidence has supported this hypothesis, findings have been inconsistent, and debate regarding the benefits of early intervention continues.^1–5 Notably, there appears to be a disparity between clinical and functional outcome in as much as remission in psychosis does not guarantee functional recovery.^6,7

As this work unfolded, schizophrenia was being reconceptualized, no longer positioned as an illness of only positive symptoms. Through the 1980s, negative or deficit symptoms garnered attention,^8–11 and their importance was reflected in the development of clinical scales that clearly differentiated, while capturing, both positive and negative symptoms.^12–14 Through the 1990s, increased focus was given to neurocognition, and by the turn of the century, the notion of schizophrenia as an illness of multiple symptom domains was firmly established.^15–17

Shedding further light on these other symptoms was an evolving line of investigation also focused on the early stages of schizophrenia, but in this case the illness’ prodrome. This stage precedes the first psychotic break, can span a period of years, and is characterized by various nonpsychotic symptoms and behavioural changes often associated with functional deterioration. Evidence gathered suggests that cognitive deficits (both neurocognition and social cognition), as well as negative symptoms, are in place during the prodrome.^18–21

Clinical Implications

• Positive symptoms represent the end stage in schizophrenia, with evidence that other key symptom domains are in place by the time of FEP.

• The idea that a single agent will prove useful across these different symptom domains (the magic bullet approach) is naïve and has not been substantiated, endorsing a new form of polypharmacy that is presently being pursued.

• From the standpoint of psychosis, evidence based on treatment response defines 3 subgroups that should be treated as such as we seek to understand the underlying neurobiology.

Limitations

• Efforts to intervene earlier than FEP remain hampered by the absence of clearly established clinical and (or) biomarkers that accurately delineate those who will convert.

• Acknowledging schizophrenia’s heterogeneity represents an important advance that also highlights major gaps in treatment (for example, negative and cognitive symptoms; clozapine-resistant schizophrenia).

Why are these details important? A closer examination of functional outcome and its determinants has established that domains other than psychotic or positive symptoms may be more relevant. While such a statement may be skewed, in that people are generally being treated for their positive symptoms, it remains that both cognitive and negative symptoms have been posited to play a more important role in functional recovery.^22–24

The implications of these different lines of investigation are considerable. First, from a clinical standpoint, the onset of positive symptoms really represents a late stage of schizophrenia regarding its evolution; by the time a first psychotic break occurs, the other key symptom domains are already established. Second, it is these domains that seem to play a critical role in functional recovery. Finally, evidence indicates that APs are not particularly effective in treating these other symptoms.^25–27 That early intervention with APs has not dramatically impacted measures of functional outcome is consistent with this, and highlights the current limitations facing FEP programs and the concept of early intervention, as well as the potential impact of APs, even used in the earliest stages of psychosis, on functional outcome.

Embracing Polypharmacy: the Myth of Magic Bullets

The magic bullet era surrounding APs may be drawing to a close. In retrospect, it was relatively short lived, and our embracing of this model seems a complex interplay of science, hope, and marketing. The era was short lived in that APs were seen as just that (that is, anti-psychosis drugs) almost exclusively prior to clozapine’s reintroduction in the early 1990s. As noted, interest in the negative symptoms really only took a foothold in the 1980s (and largely argued against the potential use of APs, or pharmacotherapy generally, in effectively treating primary negative-deficit symptoms).^9–11 Similarly, interest in neurocognition did not gain momentum until the 1990s.^15–17

The finding that clozapine also appeared to impact negative symptoms in the work involving TRS²⁸ seems an important catalyst in setting the stage for expectations that APs could be more than that. It is likely that time further contributed to this shift; almost immediately on the heels of clozapine came the advent of the newer atypical APs, drugs that were to mirror clozapine and, ideally, circumvent its burdensome side effect profile. The next decade saw claims regarding the benefits of these new medications expand considerably,²⁹ perhaps fuelled by aggressive marketing. In fairness, it was likely driven as well by the hope that we could set behind us the therapeutic nihilism accompanying the evolution of the conventional APs and the growing recognition that even highly selective DA D₂ antagonists (for example, haloperidol and pimozide), which theoretically should represent the ideal AP, fell woefully short.^30–32 Be that as it may, claims of broader, as well as greater, efficacy grew, but the science to firmly establish these could simply not keep pace. As a result, we have, more recently, witnessed a series of investigations that temper, if not dismiss, these claims.^33–35

As the dust settles, we face a very different landscape. For example, evidence that these drugs meaningfully impact cognitive deficits and negative symptoms has not been forthcoming.^25–27 Remembering what resurrected clozapine (that is, efficacy in TRS), it is now also clear that these other new drugs are not comparable.^36,37 That said, enthusiasm generated by the introduction of a new class of APs generated renewed interest in research that fundamentally changed how we conceptualize the illness. It is now patently evident that schizophrenia represents much more than psychosis.³⁸ Further, the apparent disconnect between clinical and functional outcome might better be understood in the context of these other symptom domains.^22–24

It is here we pick up the polypharmacy story, which had, to a considerable extent, been built around the practice of using multiple APs to manage TRS. Evidence has been equivocal, at best, particularly when other factors, such as cost and side effect burden, are taken into consideration.^39–41 However, the field of schizophrenia is now courting polypharmacy through another door. It has become increasingly apparent that our expectations of a magic bullet approach to schizophrenia were overly optimistic at best, or simply misguided at worst. In abandoning such an approach, we are still left to deal with an illness that has multiple domains, and the more recent shift in focus from clinical to functional recovery only reinforces this.

Our new definition of polypharmacy is no longer about combining APs. It is now about the possibility of accessing other drugs that can be used in conjunction with anti-psychosis agents to improve these other symptoms. Indeed, it may well provide indirect support for some of the other forms of polypharmacy identified as prevalent (that is, use of multiple psychotropics for treatment of anxiety or depression).⁴⁰ Only now we will be looking to add anti-negative symptom drugs, anti-cognitive deficit drugs, and perhaps others as our understanding of schizophrenia’s complexity evolves.

Interestingly, this approach very much aligns with the growing interest in personalized or individualized medicine. Clinical practice tells us that people with schizophrenia differ markedly across these different symptom clusters, a point captured in the multi-domains now detailed for clinical assessment in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition.⁴² By having different drugs for each, we can individualize treatment in a fashion that best suits a particular individual. What is less clear at present, though, is whether the foreseeable future will provide us agents that can effectively treat these other symptoms in a clinically meaningful way.

Subtyping Schizophrenia by Antipsychotic Response

With the exception of clozapine in TRS,^36,37 APs are generally considered to be similar in terms of clinical response. Moreover, all APs share in common DA D₂ antagonism, considered critical to their AP effects.⁴³ Clinicians routinely initiate AP treatment with the goal of symptom remission, switching agents in the case of suboptimal response to achieve this. Evidence suggests that, as a group, people with first-episode schizophrenia demonstrate a high response rate to the first AP, in the range of 70%, regardless of drug choice. Thereafter, the response rate drops precipitously, findings indicating that subsequent trials are associated with a response in the range of 20% or less.⁴⁴ Operational criteria have been established to define TRS, and for this group about 30% to 60% of people will demonstrate a favourable response.^28,45,46 For those who prove ultra-resistant (that is, suboptimal response to clozapine), no treatment or combination of treatments has proven to demonstrably capture a substantial number of this population.⁴⁷

This implies at least 3 types of schizophrenia based on treatment response.⁴⁸ The first group, the largest, responds to one of the currently available APs (other than clozapine as it cannot be used at this point), and can be designated AP responsive. There is no compelling evidence that atypical agents are superior in this group, although findings do suggest there is a modest chance (<20%) that one agent may work when another has failed. The second group represents people who respond to clozapine (that is, TRS) and falls under the category of clozapine responsive, while the third group constitutes a clozapine nonresponsive sample (that is, URS).

These findings have important implications, both clinically and theoretically. Clinically, undertaking multiple trials of APs in nonresponsive people before moving to clozapine is simply not supported by the evidence, which highlights clozapine as the single agent with proven efficacy, at least for a subgroup of those designated TRS.^36,37 Unfortunately, for the URS population, there is no empiric evidence presently that can be used to guide clinical decision-making in prioritizing interventions.⁴⁷

Theoretically, this same information can be used to guide research in drug development. For one subgroup, again the largest it seems, response can be achieved with DA D₂ antagonism, common to all APs. That it is purely DA D₂-related may be an overstatement, given evidence that a small number of people will respond to another non-clozapine AP⁴⁴; however, DA D₂ antagonism appears central.⁴³ For a second subgroup, TRS, response is tied to clozapine, which has in common with other APs DA D₂ binding.⁴⁹ However, its unique response in this population implicates other mechanisms are critical as well, although, to date, these have not been clearly delineated. Finally, there is this third subgroup of people who fail to respond to all APs currently available, including clozapine. Other mechanisms must define this group, and, in fact, it may be that more than one subsample constitutes this URS population.

Summarizing, treatment response defines at least 3 forms of schizophrenia. Such an approach aligns with the current notion that schizophrenia is heterogeneous in nature, and may better serve efforts aimed at improving treatment. As an example, mixing TRS and URS people clearly obfuscates efforts to find clozapine-like agents that are superior from a side effect profile, as a disproportionate number of URS subjects in a study sample would leave a putative agent appearing ineffective.

Conclusions

Looking back, it now seems naive that APs were once viewed as a panacea for the treatment of schizophrenia. Even as anti-psychosis drugs, these agents have marked limitations. The importance of DA D₂ occupancy remains, but from an efficacy standpoint captures only a subgroup of people. Better delineating the poorly responsive population (for example, clozapine responders, compared with nonresponders) seems a useful starting point to advance future treatments, while highlighting that we still do not understand the underlying pharmacological mechanisms that make clozapine unique. The notion that early treatment with APs is going to substantially improve measures of functional outcome seems misguided, at least based on current knowledge regarding how the illness unfolds, the importance of other symptom domains, and the limited impact of current drugs. Polypharmacy represents a next logical step, combining and individualizing symptom domain–specific agents, although, at present, the evidence involving investigational compounds has yet to match the proven efficacy of APs for psychosis.

Abbreviations

AP

antipsychotic

DA

dopamine

DUP

duration of untreated psychosis

FEP

first-episode psychosis

TRS

treatment-resistant schizophrenia

URS

ultra-resistant schizophrenia