Pooled Antipsychotic Data Will Be Mined For Early Response Signals
|
A consortium of international drug companies hopes to leverage their pooled schizophrenia data into a means of determining which patient subgroups are most likely to respond to treatment during clinical trials. The road to identifying responders runs through a database of more than 23,000 patients from 67 industry-sponsored clinical trials of atypical antipsychotic medications. Using alternate analytical strategies, researchers at Bar Ilan University in Israel and the Institute of Psychiatry of Kings College in London will search the database for strong, early signals that a patient will respond to a test drug. The large database also provides the first opportunity for researchers to examine patients who get better on placebo, notes Jonathan Rabinowitz of Bar Ilan, who is leading the analytical effort. There are 2,000 such patients and their information will enable researchers to assess "the nature of the placebo and active treatment signals to see how they might differ," he says. The researchers will examine the sensitivity of the various measurement tools to detect differences between placebo and active treatment signals and the impact of variations in study designs. Smaller, Faster Trials For drug developers, there is a clear benefit to identifying markers of who is more likely to respond: that could allow trials to be smaller and faster. They also could limit exposure of patients to experimental medications, according to Rabinowitz. The search for early responder identifiers is one of several projects being conducted by the Novel Methods leading to New Medications in Depression and Schizophrenia (NEWMEDS) consortium, which is sponsored by the Innovative Medicines Initiative (IMI), a cooperative effort of the European Union and the European Federation of Pharmaceutical Industries and Associations. Like FDA's Critical Path initiative, IMI fosters collaborative research by industry and academia to find ways to develop drugs and bring them to market more quickly (Also see "European Union And Industry Launch Massive Pharmaceutical R&D Program" - Pink Sheet, 14 Jan, 2008.). The schizophrenia database contains information on efficacy, demographics, background variables, treatment and study methodology for 23,401 participants in 29 placebo-controlled trials and 38 studies with active comparators. Information on 1,493 participants from the NIH-sponsored CATIE study will be included as well, and NEWMEDS has a commitment for data from another large, publicly funded European trial. The database covers compounds studied in industry trials, either as the investigational agent or the active comparator: Janssen's Risperdal (risperidone), Lilly's Zyprexa (olanzapine), Lundbeck's Serlect/Serdolect (sertindole), which received a complete response letter from FDA in 2009, Janssen's Invega Sustenna (paliperidone), AstraZeneca's Seroquel (quetiapine), Pfizer's Geodon (ziprasidone), Bristol-Myers Squibb's Abilify (aripiprazole), clozapine (Novartis' Clozaril and generics), haloperidol (Ortho McNeil's Haldol and generics), chlorpromazine (GlaxoSmithKline's Thorizine and generics) and bifeprunox, on which Lundbeck and Solvay discontinued development in 2009. The database does not include information on the newer atypical antipsychotics – Sunovion's Latuda (lurasidone), which was approved in October, Vanda's Fanapt (iloperidone) or Organon USA's Saphris (asenapine). Depression And Genetics The schizophrenia database does not include genetic information or samples to search for biomarkers. A second NEWMEDS project, however, is taking that approach in depression, using a dataset containing data on the genetics and clinical response from more than 1,800 patients. GSK and Pfizer have contributed samples from their clinical trials and AstraZeneca will make additional samples available for replication genotyping. The samples are from patients taking selective serotonin reuptake inhibitors – citalopram (Forest's Celexa and generics), Forest's Lexapro (escitalopram), GSK's Paxil (paroxetine) and Pfizer's Zoloft (sertraline). Other submissions are coming from two publicly funded projects in Europe – GENDEP and GenPod. The database "finally gives us a big enough, powerful enough sample to address how genetics influences antidepressant response," says Peter McGuffin of the Institute of Psychiatry at King's College, where the research will be conducted. "This could lead to changes in the way we select patients for trials, and in the long run, how we select treatments for individuals," he adds. Researchers will utilize the Illumina bead array platform to look at the samples to detect more than one million common variants across the genome – including human genes and the stretches of DNA between genes. Common variants are those carried by at least several percent of individuals. The study will allow inferences to be made about rarer variants that may have a large effect in a small subset of subjects, McGuffin says. Search For Non-Responders The long-term goal is to develop a comprehensive list of predictors to define the group of patients who do not respond well to currently available antidepressant drugs. "If the development of new antidepressants can focus on this subgroup, then we may get genuinely novel treatments rather than more of the same," he contends. Genetic analysis is a mainstay of personalized medicine as both payers and drug companies look for guidance on how to match patients with therapies. Advances have already been made with genetic predictors for response to antipsychotics. With an eye toward improving antipsychotic usage, Medco has joined with genetic test developer SureGene to study whether a brain-specific enzyme that interacts with certain neurotransmitters can identify which patients will respond to olanzapine, risperidone, quetiapine and ziprasidone (Also see "Antipsychotic Drugs Are Medco's New Personalized Medicine Target" - Pink Sheet, 6 Sep, 2010.). Vanda is looking at developing a pharmacogenetic test based on a polymorphism in the CNTF gene to screen patients for Fanapt. Patients in the pivotal trials who had a "good" copy of the gene had the most notable benefit from the drug (Also see "Fanapt Pharmacogenetic Data Suggestive But Not Conclusive, FDA Finds" - Pink Sheet, 1 Aug, 2009.). IMI Sets IP Policy Retention of rights to intellectual property has been an issue for pharmaceutical companies wishing to participate in such research consortia. Several years ago the Predictive Safety Testing Consortium developed a scheme in which companies can patent information they contribute to the consortium, but must allow other members access to it (Also see "Critical Path Consortium To Allow Patents But Preserve Access For Partners" - Pink Sheet, 20 Mar, 2006.). NEWMEDS is governed by IMI's intellectual property policy, which provides that each participant retains ownership of property it brings to the collaboration. Results generated under a project or stemming from a project but outside the project objectives belong to the participants who generate it, although participants may agree on a different allocation of ownership. Eight other projects are planned by NEWMEDS to look for animal models to guide drug discovery and to develop tools and tests for finding early indications of efficacy in healthy volunteers. Drug companies participating in the consortium are Lundbeck, AstraZeneca, Lilly, Pfizer, Janssen, Novartis, Orion, Roche and Servier. In addition to Bar Ilan and King's College, other researchers are from the Karolinska Institute in Stockholm, the University of Cambridge, the Central Institute of Mental Health in Manheim, the Instituto de Investigaciones Biomedicas de Barcelona and the University of Manchester. Although GSK is providing samples and data from its clinical trials, the company is not actively participating in NEWMEDS because of its decision to exit the depression space ( (Also see "GSK Spins Out Pain Assets, Raises $35.4m" - Pink Sheet, 5 Oct, 2010.) Also participating are Psynova Neurotech, Reykjavik, which studies biomarkers for neuropsychiatric illnesses, and deCODE Genetics, Cambridge, which analyzes the human genome. GABO:mi, Munich, is the project manager. Funding for NEWMEDS is €24 million ($32.1 million), with €8 million ($10.7 million) provided by the IMI Joint Undertaking, about €13 million ($17.4 million) from EFPIA and about €2.5 million ($3.4 million) from other sources. NEWMEDS is one of 15 projects selected in May 2009 for the first round of IMI-funded industry/government collaborative research. |