Redefining Clinical Research In The Age Of Genomic Science

By Tom Martorelli, historian, nonprofit strategist, and writer

In 2013, TransPoC, Inc. (Translational Proof of Concept) was formed to solve a serious problem: the diminishing number of new drug therapies resulting from oncology research. This problem has become more challenging given the overwhelming volume of genomic data available since the completion of the Human Genome Project in 2003.

Like others, TransPoC’s founders believe the way to accelerate discovering new therapies is through collaboration and data sharing on a large scale. They see it as a classic collective action problem. All “discovery-enabling organizations” know there are inefficiencies and duplication of effort in oncology research. They also see the potential in pooling resources to work from a shared platform.

TransPoC’s success hinged on developing incentives to overcome two key hurdles. First, the participating organizations must share research data, which many guard as trade secrets. And second, they need to commit start-up funding to a collective, instead of spending it themselves. Since its founding in 2013, TransPoC has made a lot of progress toward overcoming these barriers, but its success remains incomplete.

The challenge is more complicated than a scientific problem for researchers needing to share data; it’s also larger than the business problem of independent companies being unwilling to fund a collective effort. TransPoC believes there may also be a language barrier preventing researchers from fully understanding the data from each other’s studies. Lihua Yu, VP of bioinformatics at H3 Biomedicine, one of TransPoC’s founding organizations, explains how access to the data itself is a major hurdle.

“TransPoC wants to build a BioIT platform to answer drug discovery questions in the way oncology researchers ask them,” explains Yu. “Today, even a single study produces terabytes of data. And that data is complex, multidimensional, and math-oriented. The chemists and biologists who need it most can’t gather the data they need or format it in a way that is useful for designing new studies. They have to wait for mathematicians to go into these huge files, pull data, and organize it. TransPoC’s goal is to remove this time-consuming bottleneck and create an informatics system that is more user-friendly, allowing scientists to access the platform on their own, giving them the opportunity to compose and evaluate many more hypotheses.”

Other organizations before TransPoC have attempted similar, if limited, collaborative research efforts, with varying degrees of success. Examples include precompetitive testing of compounds for safety, searches for biomarkers to identify patients likely to respond to targeted therapies, pooled research efforts aimed at less well-studied areas of the human genome, and attempts to set standards for industrywide data analysis and reporting.

But each of these collaborations limits itself through a narrow definition of what is “precompetitive.” The fact that it found precompetitive information at different stages of the research process led to TransPoC’s big idea: Why not create a research organization dedicated to precompetitive data sharing for its own sake, at multiple stages of research?

SHARING DATA WHILE STILL PROTECTING TRADE SECRETS
TransPoC’s solution is designed to address the twin problems of scale and sharing intellectual property. The platforms TransPoC envisions require collective funding; they are so enormous that even the largest pharmaceutical firms would struggle to build them. TransPoC’s sponsorship model bases fees on the degree to which sponsors use its platforms and encourages them to share some data while keeping valuable trade secrets proprietary.

TransPoC’s business plan includes four basic elements:

• Cancer Cell Proof-of-Concept Network, (CPN or Cell PoC): A research platform enabling screening of compounds against more than 1,000 genetically characterized cancer cell lines
• Mouse Clinical Trials Proof-of-Concept Network (MPN or Mouse PoC): A global platform for performing multisite clinical trials using genetically characterized PDx (patient-derived xenograft) models
• Integrated Genomics, Bioinformatics, and Scientific IT (BioIT): Synchronization, storage, collaboration, and access to integrated molecular and pharmacological profiling data, to overcome the “language barrier” among researchers
• Physician/Patient Proof-of-Concept Network (Patient Portal or PPN): A network to bring patients and physicians closer to research with the intent of further catalyzing drug discovery and improving outcomes

Having worked for biotechnology companies, pharmaceutical corporations, and academic institutions themselves, TransPoC’s founders knew the incentives that would encourage discovery-enabling organizations to join their efforts would not be easy to design. Researchers might be willing to share research data, but certainly not all of their trade secrets. Start-up costs needed to be funded, but they might not reach their goals through donations alone. Dr. Markus Warmuth, CEO of H3 Biomedicine and president of TransPoC, remembers these deliberations very well.

“First, we focused on the value for scientists,” Warmuth says. “TransPoC’s translational approach allows researchers to expose their original hypotheses to as many data points as possible to ‘pressure test’ them before going to clinical trials. One molecule can be tested against thousands of cell lines, not just one. And since it is an integrated platform including xenographic data, patient-derived cell lines, and its own informatics system, TransPoC can rightfully claim a unique space having all three of these. Further, TransPoC can facilitate new relationships between research practitioners and drug discovery-enabling organizations.”

A physician and researcher himself, Dr. Warmuth recalls this as the easy part. The tougher challenge would be convincing the CFOs of potential sponsors. “We had to offer a precompetitive, capital-efficient answer to the ‘build or buy’ question each of these organizations face. And we wanted the answer to be ‘buy TransPoC.’”

To create their best answer, TransPoC’s staff and board hammered out the details of a multitiered sponsorship structure with annual fees based on company size. There would be separate fees for each of TransPoC’s CPN and MPN platforms. CPN fees would range from $200,000 to $2 million per year and would buy access for testing 10 to 100 compounds. MPN fees would range from $850,000 to $2 million per year buying access for 40 to 100 PDx (patient-derived xenograft) models. Sponsors would be required to match each proprietary compound submitted to TransPoC’s CPN and MPN PDx platforms with another nonproprietary compound. The data resulting from these nonproprietary compounds would be made available to the public, while the data from proprietary compounds would remain the intellectual property of the sponsor. Dr. Warmuth recalls the devil in these details.

“We were looking for a very specific balance, because we needed to articulate our mission and value proposition at the same time. We had to appeal to scientists seeking the benefits of data sharing while also convincing corporate managers we would protect their intellectual property. Actually, we came up with two sports analogies. First, soccer. TransPoC’s collaborative model reduces the pressure a company faces in focusing expensive clinical trials on compounds with the highest chance of success. We used the example of a research budget that allocates $20 million for each trial. Through TransPoC’s efficiencies and data sharing, TransPoC could allow this company to take four $5 million ‘shots on goal’ instead of a single one for this same $20 million research allocation.

“And second, baseball. We think TransPoC is much like the Major Leagues’ farm system, particularly the model used in the Dominican Republic. Every team wants a large farm system with a large number of prospects, so they build a common infrastructure, including multiple stadiums and other capital assets. But each team still competes with the others by developing its own players, even though they all play in these shared stadiums. Similarly, discovery-enabling organizations can develop their own compounds and PDx models in a shared research platform, concentrating efforts on projects with the best potential for success.”

FACING THE COLLECTIVE ACTION PROBLEM
In 2015, TransPoC has made progress toward implementing its business plan, launching its nonprofit organizational structure, and designing pilot studies to demonstrate its value proposition to potential sponsors. TransPoC is still seeking its start-up funding, but its founders remain optimistic. Josh Sommer, executive director of the Chordoma Foundation, a founding partner, thinks TransPoC is an idea whose time has come.

“I believe strongly in the value of collaboration and data sharing, though I experience the collective-action problem firsthand,” Sommer says. “Many times, we wrestle with the choice of outsourcing our foundation’s research budget to a larger organization that might make progress toward a cure more quickly than we could on our own with our limited resources. Yet my board rightfully wants our research to focus on our core mission of finding a cure for very specific cancers — those of the skull and spinal column — and that argues against taking this risk.

“TransPoC, or an organization very much like it, will succeed one day. But it may need a ‘Switzerland’ — a large, neutral funding source to contribute the funding that my organization and others are hesitant to invest on our own. One day, the right scheme of incentives will encourage enough organizations to take the risk of collaboration and data sharing that will lead to greater success in the new world of genomic research.”