The Vision of G-DOC
October 26, 2010
Louis Weiner says medicine today reminds him of the old Hindu proverb concerning a blind man and an elephant: without vision, you can only feel one thing – the trunk, say, or the tail – and so have an incomplete picture of the entire animal. Weiner wants physicians – and especially oncologists – to open their eyes and minds and see the whole elephant.
Louis Weiner, MD, has long been frustrated. Despite the molecular revolution that has been taking place in medicine for decades, the way many physicians evaluate their patients dates back to the early part of the last century.
First, physicians take a medical history asking questions patients they think are important. Then, the doctors perform a physical exam, and may order additional laboratory or radiological tests. Using all this information, the physician comes up with a treatment plan for the patient.
So far, so good? Not according to Weiner, who says that all the information collected by the typical physician is filtered through the doctor’s personal – and thus, subjective - experience. “It is idiosyncratic. The picture of health and disease is highly edited through the perspective of the doctor – just like feeling only the elephant’s trunk or tail,” he says.
“Physicians don’t take full advantage of knowledge gathered over the last 30-40 years on the molecular features of the diseases we are evaluating,” he says. “But it isn’t their fault, because there has been no way to effectively connect clinical and nonclinical information in order to form an accurate picture. There are few analytical capabilities that permit us to make sense of the vast amount of information that has been collected.
“We are using 19th and 20th century tools even though we have 21st century information, including all the output from clinical trials, and the capacity to molecularly analyze people’s disease process,” he laments.
So, Weiner decided to do something about that in the specialty that he practices – oncology.
And the solution he created and helped to shape from the moment he came to head the Georgetown Lombardi Comprehensive Cancer Center three years ago is now up and running.
Called the Georgetown Database of Cancer (G-DOC), the purpose of G-DOC “is to demonstrate that collecting and then analyzing molecular and clinical data in tandem will permit us to make more informed decisions about patient management,” Weiner says.
G-DOC is one part of the systems medicine paradigm that Georgetown University Medical Center (GUMC) has launched. Systems medicine is an approach that will allow health care providers to understand the interplay between genetics and the environment in such a way that one day, they will be able to predict who is at greatest risk, rather than simply react to symptoms,” said Howard J. Federoff, MD, PhD, executive vice president for health sciences at GUMC and executive dean of its School of Medicine, who has championed systems medicine at the University. “It holds promise to enable broad applications on individualized primary prevention.”
For Weiner, who is a specialist in the treatment of gastrointestinal cancers, G-DOC can help figure out how to treat each individual who comes to Lombardi with a newly diagnosed stage II colon cancer. He knows patients with this diagnosis have an 80 percent chance of being cured by surgery, but also realizes that in 20 percent, the cancer will return. “We don’t have effective tools now to help us discriminate who is destined to be cured and who is not,” he says. So he and his team have been conducting a pilot study that is collecting all kinds of information on patients with this diagnosis, including analysis of genes, proteins and other molecules from tumor tissue, as well as detailed outcome data collected over a five-year period, to help them predict who is destined to be cured with surgery alone, and who may need more aggressive treatment.
That outcome profile will then be applied to new patients in the future, thus ensuring they will receive just the right treatment.
The vision is that every patient treated by Lombardi who signs an informed consent will have information on their particular cancer and treatment outcome entered into G-DOC. More or less data will be collected, depending on funded, ongoing studies, but enough will be available, in general, to inform oncologists who are designing treatment plans for their patients.
Eventually, through Lombardi’s affiliation with MedStar and the Washington Hospital Center, G-DOC will have access to one of the largest patient populations available in a healthcare system.
While many medical centers collect detailed information on their patients and try to connect that data to clinical outcomes, the G-DOC approach is unique in two ways, Weiner says: “First of all, we are not just looking at one or two different types of molecular analysis, but at a whole battery of them to offer a true picture of the whole cancer – the whole elephant,” he says. “That means the genes that are expressed and mutated, the gene methylation profile, what metabolites are produced in urine and tumors, what mRNA is expressed, etc.
“The second piece is that we are tying all of our work to clinical important questions, such as prediction of treatment benefit,” Weiner says. “This is an initiative that is directly applicable to notion of personalized medicine. We are taking a catch phase and beginning to reduce it to practice in a very real way.”
But as straightforward as G-DOC sounds, it has taken years of work to fulfill Weiner’s vision, and Subha Madhavan, MS, PhD, has led its development.
Madhavan, who was recruited by Weiner from the National Cancer Institute, says G-DOC is unique because it brings a lot of datasets together and connects them via one portal – available to any GUMC researcher on a computer through their web browser. Eventually, some aspects of G-DOC will be made available to researchers outside of GUMC.
G-DOC also includes all the tools now available on numerous websites to analyze these datasets plus some new ones that Madhavan and others have had to create. “G-DOC has methods of analyzing data that are commonly used by biostatisticians and new ones are added on a regular basis,” she says. “And while the overall idea of G-DOC is not exactly novel, we have unique ways of approaching its creation and in tailoring it to the disease areas of interest at Georgetown.”
The G-DOC is now “live” with detailed data on 2,593 breast, colorectal, and liver cancer patients – with more information being entered everyday.
G-DOC is already being utilized by researchers. As part of a $1 million gift from the Robert M. Fisher Memorial Foundation, information from 200 breast cancer patients treated by Georgetown oncologists has been entered into the G-DOC and the funds will also be used to conduct a prospective trial of the G-DOC approach in women with early stage breast cancer.
The data collected includes all the “omics” information – molecular analysis of genomics, proteomics, metabolomics, methylomics, transcriptomics from tumors, as well as detailed clinical treatment and outcome information and patient questionnaires about their history, lifestyle, and potential risk factors.
G-DOC’s data analysis template may lead to other similar databases as GUMC expands it systems medicine approach, says Madhavan. For example, a similar database is being considered to analyze information collected from the new $38 million Georgetown-Howard Universities Center for Clinical and Translational Sciences.
“Helping to build G-DOC has been very intellectually stimulating,” Madhavan says. “It is fascinating to see the challenges oncologists face, and to find ways to make their work flow more efficient and effective.”
“We believe G-DOC will really help train the next generation of physicians, empowering them to make better decisions, based on evidence, in the care of their patients,” Weiner says. “That is a vision we can all agree is important.”