CEO SUMMARY: In a pioneering collaboration, the pathology department at Beth Israel Deaconess Medical Center in Boston, Massachusetts, will work with GenomeQuest, Inc., to perform whole genome sequencing of tumor specimens. GenomeQuest will handle sequencing, assembly, and annotation of the genetic data. BIDMC will analyze these whole human genome sequences to develop companion diagnostic tests and to find new ways to advance personalized medicine.
IN WHAT SEEMS TO BE THE FIRST COLLABORATION of its kind in anatomic pathology, pathologists at Beth Israel Deaconess Medical Center (BIDMC) in Boston, Massachusetts, are partnering with a genetic data management company to use whole human genome sequences for diagnostic purposes.
What adds excitement to this innovative collaboration is that the participating pathologists believe their pioneering work may lead to what they call the “primary-care pathologist.” From the birth of an individual, pathologists would use their analysis of that individual’s whole genome sequence to prevent disease, optimize health, and guide the care team.
Earlier this month, BIDMC pathologists and GenomeQuest, Inc., of Westborough, Massachusetts, announced an innovative two-year agreement. Essentially, as the pathology department at BIDMC does whole human genome sequencing and multi-gene analysis, GenomeQuest will provide sophisticated support in whole-genome data management and analysis.
In recognition of the importance of informatics support for the whole genome sequencing effort, GenomeQuest “will also provide API [Application Programming Interface] access and training to the [BIDMC] department’s scientific investiga- tors and applied mathematicians.”
Pathologists at BIDMC want to use whole genome sequencing of tumors as a way to develop useful diagnostic information. GenomeQuest will perform the sequencing, assembly, and annotation of entire genomes from the tumor specimens collected at BIDMC. Beth Israel pathologists will then take that information and interpret it.
“We think that the landscape of molecular diagnostics is going to be revolutionized by genomic approaches— rather than by analyzing one or two genes at a time,” observed Mark Boguski, M.D., Ph.D., and Associate Professor of Pathology at BIDMC. Boguski also serves in the Center for Biomedical Informatics at Harvard Medical School (HMS) and has been described as “one of the fathers of computational biology.”
“As the diagnosticians of medicine, pathologists are responsible for all the other laboratory tests,” added Boguski. “We consider a genotype to be no different than any other lab specimen that we get.”
Boguski points out that, since 1990, when the whole human genome sequencing project commenced, the popular wisdom was that the medical benefits from the human genome project “were largely couched in terms of new therapeutics.”
Boguski and his pathologist colleagues at BIDMC believe that the major first clinical fruits from whole human genome sequencing will actually come from diagnostics, not therapeutics. “In the last couple of years, we’ve come to the realization that whole genome sequencing is going to make its biggest, earliest impact in precision diagnostics for personalized medicine.”
“Further, we think that the landscape of molecular diagnostics is going to be revolutionized by genomic approaches rather than by analyzing one or two genes at a time,” he declared. “This has implications for personalized medicine because you can’t personalize medicine until you get a very precise diagnosis—not only for that patient, but for that patient’s disease at some point in time.
“Keep in mind,” said Boguski. “A lot of diseases change over time, and so precision diagnosis happens multiple times within the course of a patient’s disease. It’s going to be absolutely essential.”
Boguski had another observation about how whole genome sequencing can be transformational to laboratory medicine as it has been practiced for the last 100 years, stating, “I recently read a statistic that 70% of clinical decisions are based on laboratory values. Now we have a whole human genome sequence to use as the new laboratory value and it has more information in it than the combination of many existing laboratory assays!
“You can’t have personalized medicine without precision diagnosis, and increasingly that diagnosis is going to be multifactorial, based on whole-genome data sets,” he emphasized.
“There’s another aspect to this,” continued Boguski. “Traditionally as pathologists, we get involved when somebody becomes sick and a clinician sends us a specimen to diagnose a disease.
“But increasingly, well patients—and very likely newborns—will have their genomes analyzed,” he noted. “The role of pathologists will not simply be to diagnose disease, but also to help develop health maintenance plans, to prevent disease, and to participate in the lifelong management of individual healthcare in an effort to preserve health.
“Here at BIDMC, we discuss the idea of the “primary-care pathologist” and we see this as a model for the future,” Boguski added. “This is a new role for us and I think it’s one that the pathology profession must aggressively embrace. In this way, whole human genome sequencing redefines pathology’s future mission.”
Pathology leaders at Beth Israel Deaconess have a razor-sharp focus on the goal of this new genomic research initiative. They are concentrating solely on the parts that will result in putting usable gene data in front of physicians at the point of care. They expect to deliver a highly-advanced, evidence-based diagnosis.
“This is a stepwise process,” explained Jeffrey Saffitz, M.D., Ph.D., and Chairman of the Department of Pathology. “In order to bring this into medical practice we’ve developed a model that involves using whole genome sequence data in clinical practice.
“To get there, this model involves a considerable amount of outsourcing,” Saffitz noted. “We don’t want to be the ones who do the sequencing, and we don’t want to be the ones who do the initial data annotation. Rather, we want to interpret the data and help clinical physicians use this genetic information in the context of all the other things going on in diagnosis and treatment of the patient.
“We have partnerships with members of the technology community who are leading next-generation sequencing companies,” he continued. “These companies can do the sequencing far better, much more rapidly, and much more economically than any research institute or hospital laboratory.
“Our new partnership with Genome-Quest is the next step and gives us access to the data in a form that we can analyze,” explained Saffitz. “Our view is that pathologists in the future won’t be inputting gene sequence data. We will be dealing, first, with specimen control and, second, with data interpretation. Both of these activities are primary to the role of the pathologist.”
Boguski observed that demand for whole human genome sequence data is growing from several sources. “Of course, pharmaceutical companies are already looking at ways to use genome-wide, genome-scale information to manage prescriptions to produce real cost-benefit value propositions,” he stated.
“As well, people from the benefits management industry are saying that they’re ready to use some of this information right now,” continued Boguski. “Health insurers tell us ‘we will pay for something where you can demonstrate value.’
“This will be one challenge in our effort,” he added. “Our goal is to demonstrate how this new diagnostic approach will create value in healthcare.
“This value can come in several ways: 1) by directing the rational allocation of extensive resources; 2) by using the most effective therapy; 3) by getting away from this trial-and-error approach; and, 4) by minimizing side effects and adverse outcomes,” stated Boguski. “All of these things are ultimately going to save money for the healthcare system. So again, we have to begin with first steps. We can’t do all this yet, but we see this coming.”
One obvious target for the pathologists at BIDMC is to use the whole human genome sequence data to develop companion diagnostic assays. “Currently, companion diagnostics are characterized by individual, narrowly-focused technology platforms with a proprietary set of reagents,” observed Boguski.
“However, in the future, molecular diagnostics will simply be software filters and algorithms applied to whole genome or transcriptome data,” he said. “This work will fundamentally change the nature of what a companion diagnostic means for drug development.”
Boguski recommends that clinical laboratories should take steps to prepare for these developments. “Every clinical lab should be bringing its staff up to speed on genomics technology and its applications in clinical diagnostics,” he advised. “We are developing residency training programs, and there is early talk about CME programs for people who are already practicing out in the community.
“At this point, pathologists and laboratory professionals should keep an eye on the technology providers because there will be a time when it will make sense for them to start to gain access to this technology,” he continued. “And it probably will be an outsourcing model at first, because that’s the most cost-effective way.”
Creating a Clinical and Business Model to Support Diagnostic Use of Whole Genome Screening
“WE ARE AN ACADEMIC MEDICAL CENTER and a tertiary-care facility,” said Jeffrey Saffitz, M.D., Ph.D., and Chairman of the Department of Pathology at Beth Israel Deaconess Medical Center (BIDMC). “But when it comes to whole human genome sequencing, we also are an institution with limited capital resources.
“Our clinical laboratory lacks access to the resources required to purchase multi-million- dollar sequencing machines and to develop a super-computer center for doing whole human genome sequencing,” he explained. “In order for us to provide what we anticipate will be essential clinical services to our doctors, we have come up with a different approach.
“Using collaborations like the one with GenomeQuest, we can access the latest technology and tap into the most powerful sequencing and analytical resources available,” noted Saffitz. “We can then tailor the resulting flow of diagnostic information to the needs of our physicians and their patients.
“This is not necessarily the model that will prevail at every institution in the future,” he continued. “But we envision a future in which every community hospital clinical lab will have the capacity to do this type of testing and analysis.
“At this early stage in genome sequencing, we are using this outsourcing model to leverage our existing resources,” Saffitz noted. “Frankly, we also question whether we need to do all this sequencing in-house.”
“What we are doing here at BIDMC is meant to be a portable and sustainable model,” added Mark Boguski, M.D., Ph.D., and Associate Professor of Pathology at BIDMC. “If this is really going to be part of standard-of-care medical practice, basing it on a big-box genome center right next door is just not viable economically.
“So, as we move forward, we are taking a very business-process-oriented approach to this area of lab science, while working to answer this question: ‘what would it really take to pull this off in 5,000 hospitals across the country if it becomes the standard of care?’” declared Boguski. “Developing a viable model that can be adopted in other clinical environments is part of the experiment here.”
Training New Pathologists With Genomics Curriculum
PATHOLOGISTS AT BETH ISRAEL DEACONESS MEDICAL CENTER took the lead in developing a genomics training curriculum for residents in pathology programs. It has gained significant national traction.
Several meetings took place at BIDMC during which the key players in pathology— residency, education programs and other participants—came onboard to this goal. According to Mark Boguski, M.D., Ph.D., he and Jeffrey Saffitz, M.D., Ph.D., Chairman of the pathology department “have thrown down a challenge” to the pathology community to have, by 2012, a genomics and personalized medicine curriculum as part of every pathology training program in North America and in all ACGME-approved pathology residencies.
“We think this is an achievable goal,” he said. “And we’ve already made considerable progress. It’s very important that, as a discipline, pathology stands up and says, ‘we will prepare the future practitioners to function in this new world.’
“This will be key for helping the pathology profession maintain a leadership position in this new age of personalized medicine,” concluded Boguski. “For that reason, the genomics education component is absolutely crucial.”