Bifeprunox versus placebo for schizophrenia

Arka Chattopadhyay, Stephen Frey, Ghiselle Green

Bifeprunox is a novel antipsychotic drug designed to treat schizophrenia. However, research into the drug was ceased in 2009 due to rejection of licence to go to market by the US Food and Drug Administration (FDA), who could not approve the drug for acute or long-term symptoms of schizophrenia because more research was required to demonstrate convincing effects "beyond those already achieved" with currently licenced drugs. There were also concerns expressed over one death of a person whilst on the drug. To investigate the clinical and adverse effects of bifeprunox for people with schizophrenia. We searched the Cochrane Schizophrenia Group's Trials Register on 23 October 2015, which is based on regular searches of MEDLINE, EMBASE, CINAHL, BIOSIS, AMED, PubMed, PsycINFO, and clinical trials registries. There are no language, date, document type, or publication status limitations for inclusion of records in the register. All randomised clinical trials focusing on bifeprunox versus placebo for schizophrenia. We extracted data independently. For binary outcomes, we calculated risk ratio (RR) and its 95% confidence interval (CI), on an intention-to-treat basis. For continuous data, we estimated the mean difference (MD) between groups and its 95% CI. We employed a random-effects model for analyses. We assessed risk of bias for included studies and created 'Summary of findings' tables using GRADE. We included four randomised controlled trials (RCTs). We found evidence of missing data and poor reporting. When bifeprunox 20 mg was compared with placebo for schizophrenia, the drug resulted in a reduction of the Positive and Negative Syndrome Scale (PANSS) positive subscale score regarding positive symptoms (n = 549, 2 RCTs, MD -1.89, 95% CI -2.85 to -0.92, low-quality evidence) and the PANSS negative subscale regarding negative symptoms (n = 549, 2 RCTs, MD -1.53, 95% CI -2.37 to -0.69, low-quality evidence). There was a clear improvement regarding deterioration in the bifeprunox 20 mg group (n = 231, 1 RCT, RR 0.71 95% CI, 0.54 to 0.93, very low-quality evidence). The total number of participants with equal to or greater than 7% weight increase was similar between bifeprunox and placebo (n = 483, 1 RCT, RR 1.02 95% CI 0.31 to 3.33 moderate-quality evidence). There were no useable data for quality of life, economic outcomes, and service use. Our results showed some positive effects and a favourable adverse effect profile for bifeprunox, although there were few data overall and none were of high quality. It would seem that these data alone would not have been enough for the FDA to decide to halt progress of the drug to market. We can only assume that we are missing important data. Both the FDA and the relevant pharmaceutical companies have not made all relevant data accessible. As some of these trials also involved an additional haloperidol, olanzapine, quetiapine, or risperidone arm, these data are not only relevant to evaluation of bifeprunox. In not making all data accessible, it is hard to see how the FDA and the drug companies have fulfilled their full obligations to people with schizophrenia or their clinicians.