By Turna Ray
The Van Andel Research Institute has received an investigational device exemption from the US Food and Drug Administration to test the feasibility of using gene expression profiling to direct therapy for children with neuroblastoma.
The IDE will enable scientists in the Neuroblastoma and Medulloblastoma Translational Research Consortium, a nationwide network of 11 pediatric cancer clinical trial sites hosted at Van Andel, to conduct a 14-patient trial in which they will try to pinpoint treatments that best suit individual participants based on their gene expression profiles.
The NMTRC has formed an expert committee to review each patient's profile against gene-drug associations that Van Andel has been gathering from various databases and through collaborations.
"We have a knowledgebase of different signatures that might predict response to different therapies," Craig Webb, co-director of the Pediatric Cancer Translational Research Program at the Grand Rapids, Mich.-based research institute, told PGx Reporter this week. This databank contains information on the mechanisms of action for around 180 drugs, some of them indicated for the treatment of cancer and others indicated for other diseases.
The researchers have collated information on gene-drug associations from public databases, such as the Pharmacogenomics Knowledge Base and the DrugBank resource. NMTRC has been working with pathway informatics firms GeneGo and Ingenuity to analyze the data and has also been using the Broad Institute's Connectivity Map to match drugs to genome-wide disease signatures.
"We're working with an accumulation of all that [data] in our knowledgebase. There is no one particular signature in mind and no one particular drug in mind," Webb said. "We are scraping the world of publicly available data and putting it together in one system."
The feasibility trial will take between six to 12 months and through it researchers will attempt to show through correlative retrospective analysis that they can use gene expression profiling "to support real-time decision-making in pediatric oncology" and improve outcomes. According to Webb, it typically takes five days after a sample has been taken from a patient to receive a gene expression profiling report. "For this study we will have … discussions across the consortium, so we have another five business days on top of that," he explained.
The genomic analysis will be conducted using Affymetrix's FDA-approved GeneChip platform. Intervention Insights, an informatics firm that holds the commercialization rights to Van Andel's bioinformatics platform, will provide sample logistics, work with the CLIA lab, provide informatics services, and yield the final report for each patient.
NMTRC centers participating in the trial include Children’s Mercy Hospital and Clinics in Kansas City, Mo.; Connecticut Children’s Medical Center in Hartford, Conn.; Oregon Health & Science University in Portland, Ore.; Rady Children’s Hospital in San Diego, Calif.; and the Medical University of South Carolina in Charleston, SC.
Following the retrospective feasibility study, NMTRC will validate gene-drug association leads through prospective randomized-controlled trials at multiple sites. In these follow-on studies, NMTRC expects to randomize between 50 and 150 pediatric patients who will either receive treatment based on their gene-expression profile, or treatment that the physician chooses. Then, researchers will compare the outcomes head to head.
Off-Label Risks
In the feasibility study, because researchers aren't restricting themselves to a particular class of drugs, many of the gene-association based treatment recommendations may be for drugs that aren't indicated by the FDA as cancer therapies. Also, because the pediatric patients coming into the study will have exhausted all other treatment options, there are often no established therapies for these children.
For example, neuroblastoma in the pediatric population usually affects children six and younger, and kids with Stage 4 disease have less than 40 percent chance of survival for five years. These patients are usually treated with chemotherapy, surgery, bone marrow transplants, radiation, and antibody therapy.
"Most pediatric patients go off label after the first round. Remember, these patients are late-stage, and have gone through multiple lines of prior therapy, so they have really exhausted all the standard of care," Webb noted. "So, the scientists and informaticians running the [feasibility] study will not restrict the class of compounds we look at … We go across all indications and we often see the aspirins come up. So, we'll leave it up to the discretion of the physician and the tumor board to make that decision."
[ pagebreak ]
Due to the fact that NMTRC's research might lead the expert committee to recommend off-label uses of many drugs, the NMTRC hasn't enlisted the help of any pharma partners yet. Drug firms have been skittish to join genome analysis studies that seek to treat cancer patients with off-label drugs (PGx Reporter 03/30/2011).
NMTRC is "developing a tool that can lead to off-label uses of drugs." Webb said. "So, we're not really sure how to deal with pharma at this point."
Directing off-label drugs to pediatric patients using genomic data "has an element of risk," Webb acknowledged. But in this case, "it's [a risk] that the FDA, pediatric oncologists, and the patients have agreed to."
Next-Gen Side Project
NMTRC's work to use genomics to develop a personalized treatment strategy for pediatric cancer patients began with a collaboration with the Helen De Vos Children's Hospital in 2006, where researchers conducted a proof-of-concept trial involving 14 patients.
"We initially used gene expression profiling since our specific methods of analysis utilize gene expression signatures to identify sensitivity/resistance to multiple agents, network analysis of transcriptional profiles to identify likely driver nodes/hubs, and/or simple drug target rules to identify drugs," Webb said.
Because the group's discussions with the FDA regarding the feasibility study relied on data from the proof-of-concept trial, the researchers decided to stick with gene expression profiling rather than move to another method such as genotyping or next-generation sequencing. In addition, the fact that Affy's GeneChip platform is FDA-approved is a positive from a regulatory standpoint, Webb noted.
However, in addition to the feasibility study and the follow-on randomized-controlled trial, which will also use gene-expression data, NMTRC will supplement its findings with genotyping data using Sequenom's OncoCarta panel, and by next-generation sequencing conducted by the Pediatric Oncology Branch of the National Cancer Institute.
Webb explained in an e-mail that it will be "relatively easy to plug RNA-seq into our existing methods, and simple rules-based approaches can be used to associate genotype with drug" via resources such as PharmGKB. NMTRC hasn't settled on the next-gen platform it will use for this aspect of the study, but it is considering capabilities housed at TGen and at NCI.
NMTRC Chair Giselle Sholler has been working for the past five years to garner funding for this study from a variety of organizations and has received donations and grants from the Witmer Foundation, Owen Moscone Foundation, Friends of Will, Max’s Ring of Fire, the Ishan Gala Foundation, Lillie's Friends Foundation, NB Alliance, Melina’s White Light, Ethan’s Rodeo, Solving Kids Cancer, Hyundai Hope on Wheels, St. Baldrick's Foundation, and the Pappas Foundation. Additionally, many parents of pediatric cancer patients have contributed money to NMTRC's efforts, Sholler noted in a statement.
Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at tray [at] genomeweb [.] com.