This week, more than 1,100 bioinformaticians, patient advocates, computer scientists, clinicians, and others attended the sixth annual meeting of the National Cancer Institute Cancer Biomedical Informatics Grid Initiative held in Washington, DC.
Over the course of the three-day conference, participants were brought up to speed on the latest uses of the caBIG suite of tools in several cancer research efforts covering a range of topics such as biospecimen collection and management and clinical trial management, all of which were in keeping with the tagline for this year’s meeting — building a collaborative biomedical network — according to Kenneth Buetow, director of NCI's Center for Biomedical Informatics and Information Technology.
In his opening remarks, Buetow said that the caBIG community has been "working aggressively" to solve the problems of processing, analyzing, and linking large quantities of genomic and other types of data stored on "islands" in institutions and laboratories across the nation. So far, he said, these efforts have made a "dent" in the problem with the development of tools that support clinical research, track clinical trials, and facilitate data capture.
He noted that currently, caBIG has more than 2,300 active participants from over 700 organizations in 15 countries. In addition, the community has more than 19 licensed support service providers and over 70 software tools such as caTissue, caIntegrator, and the Cancer Genome Workbench, among others. Buetow also noted that caGrid, the underlying infrastructure for caBIG, contains more than 2.1 million biospecimens and that there are 4.7 million images stored in the National Biomedical Imaging Archive — a repository of medical images.
But it’s more than simply having the tools, Buetow noted. "We think the true power of caBIG is that it's not just molecular analysis, or imaging, or biospecimens, or clinical trials, but the continuum — the integration and the interconnection of those into a continuum, into a whole that is more than the sum of the parts," he told BioInform in an interview after his talk. "The true strength of the biomedical community is its diversity, but that diversity needs to be connected in collaborative frameworks."
For the meeting, the caBIG team selected presentations that covered a variety of tools and use cases in order to cover the entire "multidimensional space" of translational research and, as such, there wasn’t "a deep dive" into one particular area, he said.
For example, during one of the plenaries, JingHui Zhang, a researcher at St. Jude’s Children’s Hospital, showed how her research team used the Cancer Genome Workbench to analyze, visualize, and manage genomic data as part of a study that identified mutations in childhood cancers for the Therapeutically Applicable Research to Generate Effective Treatment, or TARGET, study, which Zhang described as the childhood cancer version of The Cancer Genome Atlas.
Following her presentation, William Carson, a professor of surgery at Ohio State University, presented the Biorepository and Biospecimen Resource, or BBR, a centralized IRB-approved repository developed by OSU to collect, store, and disseminate biospecimens. BBR uses tools like caGrid to “enable real-time querying and integration of phenotypic data” to discover cohorts, for example. Carson noted that prior to the development of the resource; OSU had as many as two dozen separate tissue banks containing things like tumor biopsies, serum samples, and so on.
Translational research in the current scientific and clinical climate is a "complex ecosystem," Buetow remarked during a plenary session.
He added that translational informatics requires scientists and clinicians to "seamlessly join multiple components," including information and tools for clinical research, a "rich" collection of biosamples and reagents for next-generation molecular analysis, as well as tools and infrastructure to capture and interpret terabytes and petabytes of genomic data.
What caBIG does to support translational research "is permit this capacity to have all this information be integrated and integratable across the entire spectrum of the use of the information," Buetow said.
Cancer isn’t restricted to a particular population or to one corner of the world, however, and now, as "the caBIG framework has more maturity, more deliverables, more capabilities," Buetow said that the community can "interact with the international community in a value proposition."
Other countries "see the value of the caBIG infrastructure and are interested in adding to it and extending it and so we get the benefits of sort of globalization with just being an open community," he said.
Indeed, Rozmin Jamal, director of clinical research at Aga Khan University in Pakistan, told BioInform that researchers at her university are considering using some of the caBIG tools to capture some of the "early clinical data that are mandatory such as informed consents" and then to incorporate other tools over time depending on how successful the researchers are in using the tools.
"We have been looking at electronic data management and data capture in clinical research," she said, noting that prior to considering caBIG, she had spent a significant amount of time evaluating both commercial and open source software and found that these tools either would not support the kind of research the university plans to do or were "prohibitively expensive."
She said that the university plans to open a clinical research center in the next few months to conduct research in several therapeutic areas, including oncology, cardiology, and neuroscience, and hopes to use several caBIG components to track and manage patient data in clinical trials.
Growing Pains and Next Steps
As the caBIG community grows, Buetow said, more members are using tools for applications well beyond their intended use, which creates some difficulties.
In addition, some researchers are finding that caBIG tools are not always easy to adopt. In particular, several speakers at the conference noted that the caTissue suite, caBIG's biorepository tool for managing, tracking, and annotating biospecimens, poses deployment challenges.
Buetow said that the caBIG development team plans to address concerns with caTissue by "refactoring some of our traditional applications into services frameworks." This way they can cater to users from institutions that require tissue bank biospecimen repository services as well as users who want a specific application with user interfaces and other components.
A larger community means more voices, Buetow noted, adding that one of the ongoing challenges for caBIG is making sure that everyone's voice is heard and their complaints addressed.
In a keynote presentation, Patrick Soon-Shiong, executive chairman of Abraxis Bioscience, noted that it takes 17 years for research findings to make their way into clinical care, a gap that Buetow called "intolerable." To shorten the time from bench to bedside, he said, the caBIG community will have to "embrace new paradigms."
In line with this, the caBIG organizers have begun taking steps toward the development of the next generation of the tool suite, which it calls the caBIG 2.0 roadmap. For example, the caBIG team last month announced that it was seeking input from the informatics community on caGrid 2.0 (BI 08/06/2010).
"Part of our goal is to make sure that we keep pace with the evolving technological landscape," Beutow said this week, adding that the caBIG 2.0 roadmap activities are an attempt to acknowledge the fact that information technology is evolving rapidly and that caBIG "needs to have a dynamic framework" in order to make use of emerging capabilities, such as cloud computing and hadoop file systems.
Buetow also said that the caBIG effort plans to continue to "blur and blend" the line between research and care.
"We need to have ways that we can make sure that every observation that’s occurring in the clinic can inform our next research investigation," he said. "But we also need to be sure that our unique findings that are coming out of our research investigation can flow into clinical and care settings."