Effectiveness of antipsychotics in first-episode schizophrenia and schizophreniform disorder on response and remission: An open randomized clinical trial (EUFEST)

Abstract

Background

Predefined response and remission criteria may hold more clinical relevance than mean scores on rating scales. We compared the effectiveness of low doses of haloperidol and regular doses of second generation antipsychotics (SGAs) on ≥ 50% response and remission.

Methods

In an open randomized clinical trial in 14 countries, 498 unselected first-episode patients with schizophrenia were assigned to haloperidol (1–4 mg/d; n = 103), amisulpride (200–800 mg/d; n = 104), olanzapine (5–20 mg/d; n = 105), quetiapine (200–750 mg/d; n = 104), or ziprasidone (40–160 mg/d; n = 82). Primary outcomes were ≥ 50% response and remission within 12 months, as measured with the Positive and Negative Syndrome Scale. Analysis was by intention-to-treat.

Results

Within 12 months, the proportions of patients with ≥ 50% response were 37% for haloperidol, 67% for amisulpride, 67% for olanzapine, 46% for quetiapine, and 56% for ziprasidone. Comparisons with haloperidol showed a higher likelihood for ≥ 50% response with amisulpride (hazard ratio [HR] 2.27, [95% CI 1.51–3.42]), olanzapine (HR 2.07 [1.38–3.10]), and ziprasidone (HR 1.62 [1.02–2.56]). Within 12 months, the proportions of patients in remission were 17% for haloperidol, 40% for amisulpride, 41% for olanzapine, 24% for quetiapine, and 28% for ziprasidone. Comparisons with haloperidol showed a better chance for remission on amisulpride (HR 2.49, [95% CI 1.43–4.35]), olanzapine (HR 2.58 [1.48–4.48]), quetiapine (HR 1.96 [1.06–3.64]), and ziprasidone (HR 2.03 [1.07–3.87]).

Conclusions

Substantial proportions of first-episode patients with schizophrenia showed clinically meaningful response and remission rates within 12 months. The proportions of response and remission were higher for most SGAs as compared to haloperidol.

Introduction

The Positive and Negative Syndrome Scale (PANSS) is often used to assess the severity of psychopathology in psychotic disorders (Andreasen et al., 2005, Leucht et al., 2008). Although changes in mean values of the PANSS may be significant, the clinical relevance of such changes is often less clear. Dichotomized outcomes like response and remission may be more useful in this regard. Previously, investigators evaluating dichotomized outcomes have used a wide range of cutoff scores. Fortunately, criteria for response and remission have been introduced recently. For acutely ill, nonrefractory patients with schizophrenia – such as most first-episode patients who generally respond well to antipsychotics – a cutoff of at least 50% PANSS reduction may be informative, corresponding to ‘much improvement’ on the Clinical Global Impression scale (CGI; Leucht et al., 2008). Other cutoffs may provide meaningful additional information, i.e. symptom reductions of ≤ 0%, 0–24%, 25–49%, 50–74%, and 75–100% (Leucht et al., 2008). Since ‘responders’ can still experience significant symptoms at endpoint (van Os et al., 2006), criteria for remission have been introduced that reflect a sustained relative absence of core symptoms of schizophrenia (Andreasen et al., 2005). According to these criteria patients are in remission if eight specific PANSS items are only mildly present for at least 6 months.

In the European First-Episode Schizophrenia Trial (EUFEST), an open-treatment randomized clinical trial, we recruited first-episode schizophrenia patients with broad inclusion criteria to compare the effectiveness of low doses of haloperidol versus regular doses of the second generation antipsychotics (SGAs): amisulpride, quetiapine, olanzapine, and ziprasidone on treatment discontinuation during 12 months of follow-up (Kahn et al., 2008). Though we found higher discontinuation rates on haloperidol (72%) as compared with the SGAs (33–53%), mean PANSS total scores did not differ between treatment arms. For the present study we have analyzed the effect of these study drugs on dichotomized PANSS total scores: ≥ 50% response within 12 months and remission within 12 months.