Three health systems—Stanford Cancer Institute, Intermountain Healthcare and Providence Health and Services—have agreed to eliminate the electronic barriers between their medical records, tumor registries and genomics databases.
The three entities said they have started to use a common IT platform to achieve interoperability and guide clinical decision-making.
That platform is Syapse, a startup that is emerging as an important player in the ongoing conversation on bioinformatics and data sharing in oncology, led by Vice President Joe Biden and the National Cancer Moonshot Initiative.
On June 6, Biden announced the NCI Genomic Data Commons as part of the moonshot. The publicly accessible $20-million database consolidates NCI’s diverse datasets and contains raw genomic data and analyses of tumors, as well as clinical data on enrollment and treatment (The Cancer Letter, June 6).
Biden is expected to discuss other data-sharing initiatives at the National Cancer Moonshot Summit June 29 at Howard University in Washington, D.C.
Syapse, an informatics software program that integrates oncology data from electronic health records with genomic data, is making inroads into the U.S. health sector.
On June 2, Syapse launched the Oncology Precision Network, or OPeN, which enables interoperability between 79 hospitals and 800 clinics across 11 states. Stanford Cancer Institute, Intermountain Healthcare and Providence Health and Services jointly announced the network, which currently has 200,000 active patients and accrues about 50,000 new cancer cases each year.
Through OPeN, Syapse enables physicians in the network to search for specific gene mutations, synthesize treatment plans and compare patient outcomes from the merged multi-institutional database—and find matching clinical trials.
Experts in oncology bioinformatics say that Syapse is a pioneer because its software is licensed and can be adapted to individual health systems and institutions to achieve interoperability.
By contrast, most research consortia and data vendors generate revenue through collaborations with industry or by selling patient data.
“You’d be amazed how few are really doing this,” said James Ford, associate professor of medicine and genetics in the division of oncology at Stanford University. “A few years ago, there was almost none. So really the answer is, they were quite visionary in seeing that this is a need coming down the road with genomics and things like that.
“I would say they were really the first to set up—in kind of a small startup manner as opposed to a big lumbering company—so they were nimble and able to work with individual partners in terms of building a software system that works,” Ford said to The Cancer Letter. “It was pretty obvious upfront that they were capable of doing that and tweaking the software for each system’s particular needs.
“How you do that at a big academic center like Stanford is completely different from a large Utah-based network of hospital sites. They can do those things in real time.”
Syapse’s annual subscription fee starts at a little under $500,000 a year and depends on the number of users and the amount of tech support.
Besides OPeN, leading oncology data consortia that are comparable in terms of patient volume and accrual include:
• The American Society of Clinical Oncology’s CancerLinQ. Launched in 2010, CancerLinQ is expected to use patient care data from millions of physician and patient records from practices and hospitals to provide feedback and clinical decision support to care providers. When the system is completed, doctors will be able to receive personalized insights based on up-to-date findings (The Cancer Letter, Feb. 20, 2015).
• The American Association for Cancer Research’s Project GENIE, for Genomics, Evidence, Neoplasia, Information, Exchange. The initiative, a multi-phase data-sharing project designed to improve clinical decision making, includes AACR and seven institutions in genomic sequencing.
• ORIEN, the Oncology Research Information Exchange Network, founded by Moffitt Cancer Center and The Ohio State Comprehensive Cancer Center. ORIEN is a self-governed alliance of NCI-designated cancer centers built around a standard consenting and processing protocol called Total Cancer Care (The Cancer Letter, March 13, 2015).
Speaking at the annual meeting of the American Society of Clinical Oncology in Chicago June 6, Biden challenged these initiatives to interoperate and share data with NCI’s Genomic Data Commons (The Cancer Letter, June 10).
Ford: How Data Should Be Organized
With the advent of precision medicine, academic cancer centers have invested millions to build tumor registries and sequence patient genomic data.
These data troves now exist at many institutions, but experts say researchers have difficulty figuring out how the data can be used to guide clinical decisions, especially since individual institutions are electronically isolated and limited to their own patient pool.
Moreover, the absence of data standards and the lack of interoperability—even between hospitals that use the same EMR vendors—pose additional barriers. Physicians are unable to link genomics with conventional EMR systems, including Epic and Cerner, which have yet to provide a comprehensive solution for accommodating the data.
“It’s so hard, particularly for academic centers,” Ford said. “I mean, big networks like Kaiser or Intermountain Health have many sites that are already networked together and share a common platform.
“But academic centers tend to all be homegrown. Each one has its own system and its own health records and the IT part of this hooking them together—it’s easy to say, ‘Oh, we should all just network together, and it’s easy to do,’—but practically, it’s very hard to do, and to do that within the law and patient health information protection.
“One thing that traditional electronic health records are not good at is dealing with large genomic tests. They haven’t sorted out how to manage that in the system with these large file sizes and many genomic data points in a patient and how you link that to outcomes of the patient somehow. This in many ways is serving a need that we can’t really do from a research point of view, even from a clinical management point of view, in this age of personalized or precision genomics.
“It’s just a challenging thing, the whole problem, and that I think is the sweet spot that Syapse is trying to show expertise at.”
Syapse aggregates data in a way that gives a treating physician all the tools to guide clinical decision-making, Ford said.
“From a physician’s point of view, it’s been incredibly useful to me, because it’s a central place where I can see all the pieces of information about a patient that I want,” Ford said. “It has both the genomic test sequencing data and the tumor, the actionable variance that were found, the potential treatment options that those would suggest, the report and the comments from our molecular pathologist about that tumor, and then ultimately, if the patient had a targeted agent, how they responded.
“So, all those things in one place, which is how it should be, and it’s impossible based just on electronic health records.
“It’s helpful to other physicians because they can go in and look those things up, they can go see particular patients eligible for clinical trials, things like that. It’s just very useful to have many different kinds of information in one place that previously has been difficult to do in one place.”
Syapse’s mission is consistent with the goals of the moonshot, Ford said.
“Syapse connects multiple health care sites—academic, community, large systems—that have different health records and patient types, and networks all of them together for the benefit of learning more about precision medicine and individual cases that are often rare in any one place,” Ford said. “Putting many places together will gather more power on that information.”
Nadauld: It’s Been Working
Syapse is designed to work in both academic and community practice settings, said Jonathan Hirsch, founder and president of Syapse.
“Traditionally, it has been very difficult to coordinate the sharing of knowledge of best practices between the academic center and their community affiliates,” Hirsch said to The Cancer Letter. “One of the things our software does for an academic center is help them disseminate those best practices out from their experts to the community affiliates, and then receive back information about the care journey of those patients, compliance with those best practices, and outcomes.
“When it’s time to have the patient maybe referred to the academic center or to have the patient matched to a clinical trial at the academic center, our software can help automate that process rather than what occurs today, which is essentially the phone calls back and forth between different organizations and emails and disorganized mess.”
As a member of OPeN, Stanford can use Syapse to access de-identified patient information in large enterprise data warehouses, or EDWs, at Intermountain and Providence, which the health systems use for internal data reporting and analysis.
Although the EDWs at the two systems consolidate tumor registries, laboratory test results, and other health information, the warehouses don’t necessarily enable physicians to match patients with clinical trials or utilize genomics data in clinical decision-making.
Intermountain Healthcare chose Syapse because the platform aggregates and compiles genomic data as well as clinical outcomes data, said Lincoln Nadauld, director of precision oncology at Intermountain, a non-profit based in Salt Lake City that operates 22 hospitals and more than 185 physician clinics.
“Intermountain has an enterprise data warehouse already in place for many of those elements. What we didn’t have was a way to organize all of those standard data elements that we’ve been collecting for years, along with genomic data and clinical trials matching and targeted oncology treatments,” Nadauld said to The Cancer Letter. “The big thing for us is Syapse understood what we were trying to accomplish—and they spoke the language—Syapse understands that goal and vision and could help us achieve it.
“We have an EMR and that’s good for standard labs and vital signs and drugs, and stuff like that—we use Cerner—but it does not handle genomics data. That’s big and kind of difficult, and genomics data, for clinical purposes, is new enough that none of these EMRs have really been set up for that.
“When we adopted the Syapse solution to handle the genomics data and clinical trials matching, etc., we then had to start building this link between Syapse and our electronic data warehouse. It’s been working.
“For the first time, in a significant way, multiple health organizations can tear down the silos and begin sharing data, because that’s how we can improve patient outcomes.”
Brown: Never Too Soon for Genomics
For the Swedish Cancer Institute, Syapse is an unprecedented and efficient solution for the institute’s clinical trial operations, said Thomas Brown, executive director of SCI.
“For us at Swedish Cancer Institute, our ultimate goal was really focused on clinical trials matching,” Brown said to The Cancer Letter. “Syapse, a cloud-based system that can handle semi-structured data, has an emerging amount of experience effectively bulking on to the electronic medical record and specifically to Epic.”
According to Brown, Syapse specifically enables SCI to:
• Identify relevant on or off-label therapies that relate to the genetic alterations, mutations,
• Prioritize clinical trials for which specific mutations are relevant—these two purposes require identified data—and
• De-identify data for data-mining research.
SCI is a part of Providence, a non-profit that operates 34 hospitals and 600 physician clinics. SCI is one of the first sites to use Syapse, Providence is in the process of implementing Syapse throughout its network.
Before signing on in 2015, Brown said the institute looked at other platforms offered by companies such as Oracle and Flatiron, and ultimately decided to go with Syapse.
“We’re a non-university research practice with a focus on early-phase clinical trials,” Brown said. “What we’re really after was an IT platform that could help us collect, organize, and analyze the molecular phenotypic information, in this case, the genomic information, in the context of the clinical data that’s within Epic or EMR.”
Syapse is key to SCI’s personalized medicine program, Brown said.
“We have an enterprise data warehouse that is the common data to the different instances of Epic in Providence, and the scrubbed data from those different instances that’s used to interact with Syapse,” Brown said. “In addition, we have tie-ins to our formal tumor registry—that has very nicely scrubbed data—and also we’re tying in our clinical trial management system. That integrated IT platform is really one of the key aspects to our personalized medicine initiative.
“We also have a registration protocol, in other words, all of the patients for whom the panel is ordered, based on clinical, medical necessity, and then patients are asked to consider participation on an IRB-approved registration protocol. The purpose of the protocol is to observe how clinicians use this information. So every time a treatment decision is made, we want to make note of whether the molecular phenotypic data are factored in or not.
“At least, every year, we revisit the patient’s updated clinical status in the context of their molecular phenotypic information, in the context of any mutational data that derive from their tumor. Over time, we anticipate adding horizontal, serial genomic reassessments using liquid biopsy technology, or just repeating the molecular profiling of patients’ tumors on a subsequent biopsy over time.
“We’ve entered over 750 patients to this registration protocol, and it allows us, in a very disciplined way, to evaluate the impact of genomic profiling in the day-to-day clinical setting.
“We also have a molecular tumor board that meets every two weeks and, of course, the other piece has been the implementation of our IT platform. Lastly, we’ve just opened a state-of-the-art early-phase clinical trials unit that specifically focused on molecularly targeted therapies that derive from our personalized medicine initiative.”
Critics may contend that it’s too early to “routinely” rely on genomics data for patient treatment, but the science needs to be tested, Brown said.
“One can take the stance that things are too uncertain at this point, and one needs to wait before routinely using this technology—again, as a research practice, we felt it important to be involved, but to do so in a disciplined way in the context of this IRB-approved registration protocol,” Brown said. “We also have a molecular tumor board that meets every two weeks and, of course, the other piece has been the implementation of our IT platform.
“Lastly, we’ve just opened a state-of-the-art early-phase clinical trials unit that specifically focused on molecularly targeted therapies that derive from our personalized medicine initiative.”
Brown said he hopes the uptake of Syapse through OPeN will help expedite access to data and clinical trials for patients and providers.
“The point is we wanted to come together to share de-identified data, both for clinical purposes and to be available for data-mining research,” Brown said. “OPeN itself is not a clinical trials cooperative group, but we feel that OPeN will facilitate clinical trials by giving patients and providers access to molecular data that will help allow access to relevant clinical trials.”