CDL Perspective
Dr. David Fenstermacher shares his views on data sharing, the need for collaborative research, his experience deploying caBIG®, and how he is leveraging caBIG® to support scientific inquiry at Moffitt Cancer Center.
Dr. David Fenstermacher, Chair and Executive Director, Department of Biomedical Informatics, Moffitt Cancer Center
David Fenstermacher is the Chair and Executive Director of the Department of Biomedical Informatics at the H. Lee Moffitt Cancer Center and Research Institute (HLMCC&RI). During his tenure in biomedical informatics, Dr. Fenstermacher has designed and directed the implementation of several bioinformatics distributed computing systems to support basic and clinical research. He has also designed data management systems for more specialized projects including integrating clinical (patient and lab test data), genomics (SNP and microarray) and proteomics (2D-gel electrophoresis and mass spectrometry data) data to support studies focused on cancer and other human diseases.
Q: Can you briefly describe what you're working on at Moffitt?
One of the driving forces at Moffitt has been our Total Cancer Care™ program—which is our vision for personalized cancer medicine. Total Cancer Care™ is a prospective study that allows access to a patient's total medical record; enables the use of residual tissue for research after clinical needs have been met; and allows us to reach out to the patient for follow-up. But, it is also a paradigm for the way we approach the patient experience and aim to provide truly personalized cancer care. We really try to partner with patients and their families from the point of diagnosis through treatment and survivorship.
Over 90% of the patients we have approached to participate have consented, and at this time no one has retracted their consent. We often hear from people who are excited to learn that their experience as a cancer patient might help others, and that the information gathered from their treatment will inform the care of other patients in the future. They feel it's their contribution to the progression of cancer research.
A protocol has been established to collect tissue samples and clinical data from patients, and there are 18 members in our consortium operating out of 19 separate facilities—mainly primary care hospitals, though we do have two academic partners—around the country.
Q: Why did you decide to use caBIG® as a part of your approach at Moffitt?
I came to the program pretty early and have believed in the value of data sharing for a long time.
Moffitt is a stand-alone cancer center and has invested heavily in best-in-class commercial software tools. But, when we wanted to set up a mechanism for our researchers to easily share data we quickly realized that trying to adapt our commercial tools was going to be problematic. As we discussed ways we could address this challenge we reviewed various resources that could help facilitate data sharing, and determined a layer of caBIG® tools that will give us the ability to share data.
No one cancer center can do everything alone; if we're really going to do personalized medicine, we need to have access to large patient populations and associated biospecimens. It wasn't hard to convince my team to put out tools for data sharing, since we know we have to be part of that game. Now, Moffitt researchers have access to all Total Cancer Care data, and the tools allow our researchers to get into caGrid to look for other things. Any Moffitt researcher that wants to use this data has access to it, which is one of the major ways caBIG® is helping to advance science here.
The catalyst for TOTAL CANCER CARE was to provide better patient care, but when we got into project we saw that some of the challenges are pretty big, and very widespread; for example, not everyone has a standard EMR. But, caBIG® helps provide some of the infrastructure that is needed to address that. You can use caBIG® to make things available that have been in silos—this is where things get exciting.
Q: Why is data sharing so important?
When you see the results from our CER project, it's easy to understand that this type of data sharing should be happening on a larger scale.
More self-interestedly, implementing these systems now is keeping us ahead of the HIT curve, so when meaningful use is defined, Moffitt is poised to be a leader.
Q: Can you explain your role as a Center Deployment Lead (CDL) at Moffitt Cancer Center?
My role is to lead the Moffitt Cancer Center's efforts to provide caBIG® connectivity for our researchers. Basically I lead our local team. From a broader perspective, being a CDL is really about community. For example, the Annual Meetings give us a chance to connect and learn about exciting things going on at other institutions. It gives us a forum to discuss ideas when it comes to sharing data and moving toward molecular based personalized medicine for cancer. It's really one of the first times informatics has been discussed across multiple institutions. We all do things differently. Some have adopted a lot (e.g. Arkansas) and other are trying to adapt what they are their commercial and open-source tools to connect to the grid. Even though the NCI has created a wealth of resources to assist with using caBIG® tools, the fact we know each other and can learn for personal experiences is why being a CDL is so valuable.
Q: What challenges have you had as a CDL?
It was a little harder to convince people of the value on the “semantic” part of things. When you're a bioinformatician, this [semantic interoperability] is common sense because we live and breathe it every day; we see how differences in the way data is collected can affect the validity of experiments, but changing from how you've always done something is difficult.
Q: What do you see as the future of research, and where does caBIG® fit in?
There are lots of areas where we [research community] need to get better at standardizing things but caBIG® has really been out there pushing and I think it's really starting to get some traction, particularly with the money that has been earmarked for HIT through ARRA. I hope caBIG® will provide the transparent data sharing framework that allows researchers to take advantage of cancer research going on throughout the country. I also believe that caBIG® will be one of the catalysts to establish semantic infrastructures that will lead to the next generation of data interoperability that is needed for healthcare reform.
caBIG® has also increased the emphasis and visibility of bioinformatics within cancer centers. Before very few people knew about bioinformatics at all and caBIG® has demonstrated it's not only about analysis of the data, it's also about making the data usable by different organizations. It's not just useful for one researcher; there is really widespread value to be had.
I think that health information exchanges will also be important, so that one day a patient can say, “I'm a patient here at Moffitt and I'm moving to California, and I'd like to continue my treatment at UCSF.” With the right infrastructure for information exchange, we should be able to shoot that data over with no interruption in treatment. That doesn't exist today. It's a huge endeavor, and we are trying to make inroads with projects like Total Cancer Care, which is a pilot in and of itself for the future of what healthcare can look like. There is certainly an opportunity for caBIG® and caGrid to play a huge role both in exposing the problems and challenges around data-sharing, but also in addressing those technical issues.
Q: What advice to you have for institutions or individuals who are considering caBIG®?
caBIG® is a very dynamic community made up of biomedical research and information technology professionals from academia, the private sector and government. The NCI Center for Biomedical Informatics and Information Technology has created this environment to accelerate scientific discoveries through the sharing and integration of data across the cancer research community and has created software tools, support structures and basic information infrastructure that can be utilized by individual laboratories and institutions. caBIG® can be quite overwhelming for the newcomer. Anyone considering becoming part of caBIG® or using the tools and infrastructure should take advantage of the resources to learn more about the program, even contact current users of caBIG® tools. There are numerous ways to take advantage of caBIG® tools and infrastructure and newcomers should be open to the vast possibilities of how caBIG® can enhance their local biomedical informatics infrastructures. One of the best ways to learn more is by attending the caBIG® annual meeting and by becoming a member of the community.
From an interview with David Fenstermacher:
As part of the Moffitt Total Cancer Care project we have an ARRA grant that is funding the creation of an infrastructure to support comparative effectiveness research (CER). We're creating a comprehensive data-mart that includes a variety of information such as biospecimen data, data derived from unstructured text using natural language processing, transcriptomics data, and others and we are using caBIG® terminology resources to allow us to link into the Enterprise Vocabulary Services. We chose caBIG® because we wanted to take advantage of the infrastructure to make this data shareable via caGrid.
