NEW YORK - Dell has announced it will donate its cloud-computing architecture to speed up the search for successful pediatric cancer treatments, the company announced at a press conference here on Nov. 10.
The project is part of Dell's "Powering the Possible" initiative, which commits millions of dollars over many years to supporting pediatric treatment in underserved communities.
Pediatric cancer is a leading cause of death in children ages 1 to 14, according to Dell.
The Translational Genomics Research Institute (TGen) will use Dell's cloud infrastructure to store data for the world's first personalized medicine trial for pediatric cancer. TGen is a nonprofit organization that performs genomics research to develop quicker diagnoses and treatment of diseases.
Cloud infrastructures can speed computational processes for medical data and allow researchers to manage and store the information as well as collaborate on the findings.
"We believe the cloud, combined with genetic engineering, can be a game-changer," said Karen Quintos, senior vice president and chief marketing officer for Dell.
Dell is contributing the high-performance computing architecture as well as data-exchange and collaboration functionality.
"It's the next-generation high-performance computing platforms that are taking the number of calculations from billions to trillions," Dr. Jamie Coffin, vice president and general manager of Dell Healthcare and Life Sciences, told eWEEK.
The TGen Cloud architecture runs at up to 8.2 teraflops and is based on Dell Precision Workstations, which researchers will use for analysis and review.
Dell's high-performance computing will provide a dedicated clinical cluster to perform 8 trillion operations per second, according to Dr. Jeffrey M. Trent, president and research director of TGen and VARI.
The cloud also incorporates Dell's PowerEdge Blade Servers, Compellent Storage Center Arrays and Force10 network infrastructure.
Personalized medicine trials will focus on neuroblastoma, a cancer that attacks the sympathetic nervous system and affects heart rate, blood pressure and digestion. With tumors unique to each child, researchers need data that can provide information on how to attack individual tumors.
Cloud data could help reduce the trial and error needed to find treatment, according to Dell.
"You don't want to put patients on this drug if they're going to respond 5 percent of the time; you want to put them on it if they'll respond 90 percent of the time, and that's what targeted treatment is about," Coffin said.
"You want to get to the point where you can say I know this tumor is expressing these four pathways to grow," he added.
Finding treatment for a specific patient's tumor is called "precision medicine," noted Dr. Lee Helman, head of molecular oncology at the National Cancer Institute.
Data from the cloud project will be also transferrable to other types of chronic diseases, Helman added.
In addition, storing genomic data in a cloud repository can reduce the time required for tumor mapping and analysis from months to days, according to Dell.
"It allows us to take this down to weeks instead of years in the calculations, so that's probably the most significant part of what we're doing," Coffin said.
"This allows us to not wait to do the analysis for an individual patient," Coffin said. "And as you know with these diseases, they grow very fast-there isn't a lot of time to wait."
During the trials, scientists from the Neuroblastoma and Medulloblastoma Translational Research Consortium (NMTRC) and the Van Andel Research Institute (VARI) as well as parents and physicians aim to process data on patient tumors in real time to find drugs for specific patients.
In the first stages of the trial, 12 medical centers will enroll patients. They include the National Cancer Institute in Bethesda, Md.; Children's Mercy Hospital and Clinics in Kansas City, Mo.; and Connecticut Children's Medical Center in Hartford, Conn.
Later stages could expand the number of hospitals participating to around 20, according to Dr. Giselle Sholler, NMTRC chair and co-director of VARI's Pediatric Cancer Translational Research Program.
More than 200 billion measurements per patient will be analyzed, shared and stored in the cloud. The technology will enable scientists to move genomic data more effectively.
The trials will be exclusively in the United States the first two years, according to Coffin.
The shared environment of the cloud will eventually enable collaboration with international scientists for additional types of childhood cancer, he said.
"Having the shared collaborative infrastructure allows [oncologists] to be able to use 50 brains instead of one brain to make a treatment decision," Coffin explained. "Most times when oncologists are making treatment decisions they're doing it by themselves."
The cloud will allow TGen's clinical cluster to increase computation and collaboration capacity by 1,200 percent, Coffin said.
Cancer will be considered a treatable chronic disease in the future, he predicted.