By Allison Proffitt 

 

December 19, 2013 | When Kathy Giusti was diagnosed with multiple myeloma in 1996, she was offered drugs that had been used in 1956. She found it unacceptable, so she started the Multiple Myeloma Research Foundation (MMRF) to accelerate cures for the rare disease. The Foundation (of which Kathy, now in remission, serves as head) is now launching a pair of gateways, built by John Quackenbush’s GenoSpace, to link research, researchers and patients that they believe could serve as a model for research for other diseases. 

 

MMRF’s vision has always been to get many stakeholders involved to accelerate multiple myeloma research: researchers, drug companies, hospitals, patients and more, said Walter Capone, MMRF’s Chief Operating Officer. From 1998 to 2004, Giusti and the MMRF built the Multiple Myeloma Research Consortium consisting of 18 academic centers, a multiple myeloma tissue bank, 47 different trials, and 26 different drugs. Six of those drugs have been approved. 

 

With a stocked tissue bank, MMRF launched a $15 million genomics mapping initiative. In 2011, the Foundation published the characterization of multiple myeloma—at least ten subtypes of cancer involving 50 primary mutations and up to 400 secondary mutations—and published the data at myelomagenomics.org. 

 

“It was very clear that only by driving forward an initiative focused on personalized medicine, specifically to identify targets and drive drugs forward for those targets in myeloma, would we really be able to make the next significant advancement in trying to find a cure for the disease,” Capone said. 

 

Now, in what Capone calls the “outgrowth of almost a decade of research,” MMRF has launched two gateways—one for researchers and one for patients—giving access to MMRF’s CoMMpass Study: the clinical information from multiple myeloma patients collected over ten years. 

 

But the researcher gateway won’t be home to just one study. “We’ll be able to house all the significant studies, all the significant data that’s available with this particular focus—to accelerate our understanding and to share the insights with drive forward cures more quickly,” Capone said. 

 

An Atmosphere of Collaboration 

 

When the gateways were first conceived, MMRF wanted to build a place where academic medical center and drug companies would want to come together. “We kind of had to create a space where people would be, on the one side giving up certain aspects or certain things they’re used to having, while at the same time benefiting from collaborating but not exposing themselves to undue risk from collaborating.” 

Academic centers, Capone said, weren’t used to sharing data or samples from patients that they believed to be “their patients.” The idea of conveying clinical data and bone marrow samples to a third party to manage was completely foreign, he recalled. 

“We got them over the hump, by indicating to them that it was not our intent to house and to horde that data within our own walls. Our only intent was to take this data and create a public access portal that all researchers could take advantage of. And—by the way—the institutions would get back the most extensively profiled full genome, whole exome, and RNA sequence data on the patients they submit to us. They’d have access to the full spectrum of that data.” 

 

They key to getting buy in, Capone said, was to understand their own research programs and to offer data that they wanted. “Based on the quality of the scientific work that we were doing and the longitudinal nature of the study itself—at least five years with follow up if not 10 years—it would take any one of them tens of millions of dollars to do. But they could buy in for a percentage of that if they only collaborated.” 

 

The proposal won over more than 50 institutions including Millennium Pharmaceuticals, Mt. Sinai hospital, University of California San Francisco Medical Center, University of Chicago, and Bristol-Myers Squibb. 

 

Space Planning 

 

The next challenge might have been even larger. MMRF needed a platform to connect these groups and their data. The platform needed to enable researchers around the world to access on their desktops exquisitely-detailed genomic data correlated with clinical data and demographics, treatments, and laboratory values. It’s, “the full gamut of clinical annotation that provides context for the underlying biological changes that are seen,” Capone said. Researchers needed everything. 

MMRF was intent on collecting multiple time points from each patient: samples at diagnosis, after treatment or transplant, and best response. “[We wanted] to look at how myeloma is changing and evolving based on its actual biology and the evolution of those tumor cell sub-populations as well as the pressure of different drugs and how they move and shift the clonal populations of the tumor cells in these patients,” Capone explained. And they wanted all of it as easily accessible as an Amazon listing. 

 

MMRF spent a year developing the requirements and invited eight companies to submit proposals including Oracle, SAS and IBM. 

 

For most of the companies, Capone says, “Their orientation was very much focused on clinical trial conduct. Basically, a single study that would look at two drugs in comparison and render that information… to the sponsor. They had exquisite data management and structure and execution in that front.” 

 

Their weaknesses came with integration though. “When it came to sharing that information across institutions or enabling them to basically do social networking in a research setting, we came to find that didn’t exist in the platform they were offering. At the same time, they hadn’t integrated the next generation sequencing technology into the platform to be able to integrate genomics and clinical data together. That was pretty much a common theme across all the vendors, with the exception of GenoSpace.”

 

Selling the Startup 

 

To hear John Quackenbush tell it, GenoSpace got the gig at the last minute. 

 

When Quackenbush first heard of the MMRF proposal, he said, “They’d settled on a vendor that we knew probably wouldn’t be able to capture this data in an effective way.” Quackenbush, GenoSpace’s CEO and a professor at Dana-Farber, decided to change their minds. 

 

“We made a journey down to Norwalk, Connecticut. We met with Kathy and Walter and others from MMRF and we laid out why we thought we were going to be the best solution to the problem. Being a small startup company, not having a long history of doing this, I think they were justifiably skeptical, but we have strong reputations in the field as individuals so MMRF was willing to take a chance.” 

 

At the helm of GenoSpace, Quackenbush is joined by Mick Correll as COO (also Associate Director of the Center for Cancer Computational Biology, Dana-Farber Cancer Institute, with a background from InforSense and LION Bioscience); Daniel Meyer as CFO (with Arboretum Ventures and Point Judith Capital); and Niall O’Connor as Director of Engineering (with experience at Dana-Farber Cancer Institute and Fidelity Data Security). 

 

It’s a strong team, and their tenacity paid off. MMRF set up a head to head competition between GenoSpace and another vendor. “Within about six weeks, we’d really taken our core infrastructure and around it wrapped a prototype portal that met almost all their requirements,” Quackenbush said. “They were convinced we could really deliver.”

 

What GenoSpace creates, Quackenbush explains, is an operating system to enable genomic medicine, broadly-defined. Genomic data is fundamentally identifiable, Quackenbush explained. The idea of de-identification “goes out the window,” when you work with genomic data. There’s no point in pretending otherwise. 

 

So GenoSpace chose to focus on security. The core infrastructure is based on data security with a data store that can handle genomic data and data security protocols with encryption. Clinical data is also identifiable, so again GenoSpace chose to link that to genomic data with strong protection. 

 

The resulting products—GenoSpace for Research, GenoSpace for Communities, and an offering for clinical care (PathGroup is using it)—allow for transition from purely a research space to a very personal clinical space. “By focusing on security, what we could do was open up this broader universe of applications.” 

 

Quackenbush calls the applications a “LEGO toolkit of different analytical pieces that we can kind of snap into place.” MMRF is using GenoSpace for Research to power the researcher gateway, and other groups—Quackenbush mentions a large sequencing project in the UK—are taking it on as well. 

 

GenoSpace and MMRF have signed a six year contract that can be renewed, and Quackenbush says GenoSpace is now actively supporting the gateways, updating software, and refining tools. 

 

Building Community 

 

But the vision wasn’t just a deployment of infrastructure for researchers. Early on, both GenoSpace and MMRF independently identified a community gateway as a priority. 

 

“One of the things we recognized going to MMRF was that CoMMpass was a start,” Quackenbush said. “Those 1,000 patients are going to provide incredibly valuable data, but there are 10,000 patients a year diagnosed with multiple myeloma in the country. They represented a tremendous opportunity to get information that would drive research.”

 

The real intent of the community gateway is for people with multiple myeloma to have all of their data and community in one place, and to connect patients to the clinical trials born of the research efforts, Capone said. 

 

Quackenbush believes it to be a rare opportunity. “Many of these patients would have some kind of genomic data. Many of them would be willing to share clinical information. They’d be willing to share genomic data. They’d be willing to share their personal experiences, anything that they saw as being useful for driving forward the search for new therapies and better ways to treat.” 

 

Capone believes the MMRF community is different from other patient communities (PatientsLikeMe, for example) because of the focus on multiple myeloma. The gateway can host records, so that patients can easily access it when it’s time for follow up appointments with their doctors. It’s also an “intelligent” community, he says. “The more patients provide information about their background, the more we can tailor content, assistance, or support as well as research and trials that might be appropriate for them.” The data coming into the gateway can help “connect and guide” patients to the resources that would be most helpful to them. 

 

“By nature the disease itself is complex with many subtypes, and much diversity or heterogeneity. It is a very good model for how other cancer types can be addressed or attacked in a similar way. All the tools that are being applied right now are completely transferable and applicable to leukemia, lymphoma, other solid tumors, particularly where there’s a molecular competent to the conditions as well as the application or investigation of immunological agents,” Capone says. 

 

Editor's Note: John Quackenbush will keynote the 2014 Bio-IT World Conference & Expo in Boston, April 29-May 1. For a full agenda and to register, see www.bio-itworldexpo.com