Lack of improvement at week 2 predicts later antipsychotic non-response in people with acute exacerbations of schizophrenia or schizophrenia-like psychosis

What is already known on this topic?

• Because there is no conclusive evidence on whether early signs of improvement predict later non-response to antipsychotics in patients with schizophrenia spectrum disorders, the treatment guidelines for schizophrenia, such as those from National Institute for Health and Care Excellence,1 remain unclear as to when a treatment should be changed if patients are unresponsive to a recently prescribed antipsychotic.

• Therefore, Samara and colleagues performed this diagnostic meta-analysis to examine whether a lack of significant improvement at week 2 could predict ultimate treatment failure.

Methods of the study

The diagnostic test meta-analysis using mostly individual patient data was conducted to assess whether lack of improvement at week 2 predicts later non-response to acute treatment (4–12 weeks). Diagnostic test meta-analysis is a novel technique that allows researchers to synthesise the results of studies on diagnostic tests and obtain their overall test parameters.2 The authors conducted a comprehensive literature search of electronic databases until February 2014, with terms combining antipsychotic drugs, schizophrenia and prediction of response. They also contacted study authors and pharmaceutical companies for additional trial data. The lack of improvement after treatment at week 2 was defined as ≤20% reduction in total score on either the Positive and Negative Syndrome Scale (PANSS) or the Brief Psychiatric Rating Scale (BPRS) from baseline to end point (corresponding to less than ‘minimally improved’). Later antipsychotic non-response was defined as ≤50% reduction in total score on either the PANSS or the BPRS from baseline to end point at 4–12 weeks (the cut-off for ‘much improved’). Secondary outcomes were absent cross-sectional symptomatic remission and ≤20% PANSS or BPRS reduction at end point. Potential moderator variables were examined by meta-regression.

What does this paper add?

• Thirty-four studies (9460 patients) were included in this diagnostic meta-analysis, using individual patient data for almost all studies (94.1%).

• A ≤20% PANSS or BPRS reduction at week 2 predicted non-response at end point with a specificity of 86% and a positive predictive value (PPV) of 90%. The sensitivity analysis using data for observed cases did not substantially change the results (specificity was 86% and PPV was 85%).

• The test specificity was significantly moderated by a trial duration of <6 weeks, higher baseline illness severity and shorter illness duration.

Limitations

• High discontinuation rates were observed in most of these trials (mean=29.2%). However, there was no significant difference in sensitivity between intention-to-treat analysis and observed-case analysis in the sensitivity analysis.

• The meta-analysis mainly included studies of chronically ill patients. Several studies have shown that response patterns in first-episode3 or treatment-resistant4 patients may differ from those in chronically ill patients. However, there were insufficient numbers of first-episode studies (N=6) and treatment-resistant studies (N=1) to detect a difference between the two patient populations.

What next in research?

• Because several studies have shown that a partial population of first-episode3 or treatment-resistant4 patients have a later response to antipsychotics, a longer term study including first-episode or treatment-resistant schizophrenia patients is required.

• Findings from the Psychiatric Genomics Consortium genome-wide association study showed that allelic variation at the dopamine receptor D2 gene (DRD2) locus is associated with schizophrenia risk.5 Dopamine D2 receptor binding is the main mechanism of antipsychotic action and DRD2 variant was related to antipsychotic response.6 Genotype might also be one of the potential moderators of the test performance.

Do these results change your practices, and why?

Yes. The study reported that patients not showing minimal improvement by week 2 of antipsychotic treatment are unlikely to respond later (after 4−12 weeks). The evidence suggests that if a patient's psychopathology does not improve minimally after treatment with antipsychotics for 2 weeks, the treating physician is able to switch from the antipsychotic to another antipsychotic. However, the application of these results is more appropriate for patients who are neither in their first episode of schizophrenia nor exhibiting treatment resistance.