New Data Mining Platform Accurately Predicts Effectiveness of Metastatic Prostate Cancer Treatments

Scientists Use Project Data Sphere® Research Platform to Identify New Surrogate Survival Endpoint that Can Dramatically Reduce Clinical Trial Size and Transform Phase III Trial Design

Shared data can accelerate cancer research, unleashing the power of data to achieve the National Cancer Moonshot’s goals

CARY, N.C.--()--A report published in today’s issue of the international journal The Lancet Oncology 1 demonstrated the power of data mining to enhance our understanding of cancer, improve its treatment and accelerate clinical research.

“Our research shows that tools created with shared data can detect whether or not a new therapy is working with significantly fewer patients than are currently required in standard clinical trials,” said Tito Fojo, MD, PhD. “This is encouraging and makes a strong case for the expansion of broad-access data sharing and collaborative data-mining initiatives, especially in oncology.”

Using completed Phase III clinical trials collected by Project Data Sphere, LLC (PDS), researchers from around the globe created tumor growth models that accurately predicted patients’ Overall Survival (OS) rate – a clinical endpoint that usually takes months or even years to ascertain. By creating a useful new surrogate endpoint, researchers found they needed only 16 to 47 patients to establish the superiority of one treatment over another – far fewer than the roughly 1,000 patients needed to detect differences between treatments in present-day metastatic prostate cancer trials.

“The goal of PDS is to enable physicians to improve care for cancer patients through new insights derived from shared clinical trial datasets,” said Dr. Martin J. Murphy, PDS Chief Executive Officer and study co-author. “The research performed by this multinational group of investigators is a fantastic example of how this shared data can validate a new endpoint and eventually reduce the time needed to prove a cancer treatment’s effectiveness.”

The study results highlight how researchers can use shared clinical data to detect patterns and ultimately accelerate cancer research, fulfilling Vice President Joe Biden’s call to unleash the power of data to achieve the National Cancer Moonshot’s2 goals.

“The value of the Project Data Sphere research platform is the ease with which you can get access to multiple large datasets with incredibly rich control arm data,” said Dr. Fojo. “The data providers do tremendous work up front to clean and anonymize the data prior to uploading it to the platform, which enhances the experience for researchers. We used only a small fraction of PDS’s data in our study, so we believe that applying our method to even larger datasets will yield even greater insights and will alert others to the untapped potential of this database.”

Accessing de-identified patient data aggregated within the PDS research platform from the control arms of eight metastatic castration-resistant prostate cancer (mCRPC) trials, the researchers used previously derived mathematical models to discern how mCRPCs typically grow. They then evaluated the effects of different treatments on mCRPC growth rates to arrive at an algorithm that predicts survival based on tumor growth rate alone.

The study involved research collaborators from 12 organizations, from across industry, government and academia, including AstraZeneca; Sanofi US; Pfizer, Inc.; Celgene Corporation; Mayo Medical Center; Memorial Sloan Kettering Cancer Center; Janssen, the Pharmaceutical Companies of Johnson & Johnson; The National Cancer Institute Biometric Research Program; The Silberman Institute of Life Sciences at the Hebrew University (Israel); the Division of Medical Oncology, Department of Medicine, Columbia University/New York Presbyterian Hospital; the James J. Peters VA Medical Center; and Project Data Sphere, LLC.

About Project Data Sphere, LLC

Project Data Sphere, LLC, is an independent, not-for-profit initiative of the CEO Roundtable on Cancer Inc.’s Life Sciences Consortium. The CEO Roundtable on Cancer is a 501(c)(3) nonprofit organization founded by President George H.W. Bush to develop and implement initiatives that reduce the risk of cancer, enable early diagnosis, facilitate access to the best available treatments and hasten the discovery of novel and more effective anti-cancer therapies.

The Project Data Sphere research platform provides data scientists with no-cost access to cancer trial data provided by leading research organizations, analytics software contributed by SAS, and a collaborative online research environment to facilitate analyses of tens of thousands of patients’ records in the database.

For more information, visit www.ProjectDataSphere.org, or follow PDS on Twitter at https://twitter.com/ProjectDataSphere.

1 Wilkerson, Julia et al., “Estimation of tumour regression and growth rates during treatment in patients with advanced prostate cancer: a retrospective analysis.” The Lancet Oncology, 2016, doi: 10.1016/S1470-2045(16)30633-7. Available at http://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(16)30633-7/fulltext
2 Report of the Cancer Moonshot Task Force. October 17, 2016. Available at www.whitehouse.gov/sites/default/files/docs/final_cancer_moonshot_task_force_report_1.pdf

Contacts

Deborah Dion
Project Data Sphere, LLC
An independent initiative of the Life Sciences Consortium of the CEO Roundtable on Cancer
919-531-0962
Deborah.Dion@ProjectDataSphere.org

Contacts

Deborah Dion
Project Data Sphere, LLC
An independent initiative of the Life Sciences Consortium of the CEO Roundtable on Cancer
919-531-0962
Deborah.Dion@ProjectDataSphere.org