The fact that 20-25% of patients with schizophrenia are treated with more than one antipsychotic1 suggests polypharmacy meets a need. Meta-analysis of blinded studies suggests no benefit from giving two drugs, but real world evidence indicates the opposite. Patient subgroups and specific drug combinations should be considered in a nuanced way.
Specific combinations may be particularly effective when compared with monotherapy
Polypharmacy is not superior to monotherapy when meta-analysis includes only randomized trials in which both patients and investigators are blinded to the treatment arm.2 When analysis is confined to non-blinded trials, polypharmacy is superior to single-agent treatment. So expectation clearly seems affect outcome, Christoph Correll (Charité Hospital, Berlin , Germany; and Zucker School of Medicine, New York, USA) told the EPA 2020 symposium which considered the issue.
That said, Professor Correll drew attention to one randomized discontinuation study suggesting that certain patients did benefit from the combined use of antipsychotics: many of those randomized to discontinue one of them chose – after a while -- to return to taking two drugs.3
Cohort studies avoid selection bias and are long term
Professor Correll also recognized that clinical trials of polypharmacy represent highly selected patients. Among those excluded are many of the most severely ill. This point was taken up by Jari Tiihonen (Karolinska Institute, Stockholm, Sweden; and University of Eastern Finland) who has co-authored several recent studies using Nordic registry data on rehospitalization.4,5
In the observational study of more than sixty thousand patients in Finland, antipsychotic polypharmacy was associated with a 7%-13% lower risk of psychiatric rehospitalization when compared with any monotherapy.4 Polytherapy also compared well against most monotherapies in a previous long-term observational study of almost thirty thousand patients in Sweden.
As was pointed out by Professor Correll, both studies showed particular benefit from combinations involving a long-acting injectable agent.
Nuanced view of a complex issue
The question of whether we should move towards a nuanced view of the merits or otherwise of polypharmacy -- one which considers the nature of symptoms, specific patient subgroups, and particular drug combinations -- was discussed in the symposium.
Professor TIIhonen drew attention to one high quality randomized trial that showed a benefit from adding a drug to existing antipsychotic therapy, albeit only on a secondary analysis of negative symptoms.6
And Professor Correll argued that there could be a pharmacological rationale for certain combinations. Although giving two antipsychotics risks drug interactions and enhanced adverse events, combining a partial dopamine D2 agonist with a full D2 antagonist may actually improve side effects, such as insomnia and prolactin elevation.
But there is less rationale for combining two agents with very similar receptor profiles.
And polypharmacy should be used only when adequate monotherapy options have been exhausted, he concluded.
Jari Tiihonen also came to a considered verdict. By analogy with antihypertensive therapy, there may be benefit in schizophrenia from combining agents with different mechanisms of action. Data do not suggest that all kinds of combination therapy are beneficial, but guidelines should modify their recommendations against any antipsychotic polypharmacy.
He also takes encouragement from another recent study suggesting that in the Finnish cohort polypharmacy is well tolerated. Risks of hospitalization for somatic reasons and all-cause mortality among people receiving drug combinations compare well with those among people on antipsychotic monotherapy.7
For the latest updates on sea.progress.im, subscribe to our Telegram Channel https://bit.ly/telePiM
Our correspondent’s highlights from the symposium are meant as a fair representation of the scientific content presented. The views and opinions expressed on this page do not necessarily reflect those of Lundbeck.
