This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American
When it comes to treating pediatric cancer a group of academic researchers, oncologists and pathologists believes that a more personal approach isn't just more humane, it's the key to survival. For members of the Neuroblastoma and Medulloblastoma Translational Research Consortium (NMTRC) this personal touch means using genomic analysis to develop highly targeted therapies to treat each individual tumor. Such analysis produces a lot of data, which means it holds a lot of potential for helping kids with neuroblastoma, an often aggressive malignant tumor that develops from nerve tissue. It also means the researchers need access to some serious computer power to make sense of that data.
NMTRC announced a big boost to their work Thursday—they will soon be doing their research on a new high-performance cluster of computers donated by Dell and dedicated exclusively to their work. The timing couldn't be better as NMTRC in May launched a clinical trial studying pediatric neuroblastoma patients to find ways of more quickly and effectively treating their cancer. Neuroblastoma is most commonly diagnosed before age five and occurs in about one out of 100,000 children (it is slightly more common in boys). While the prognosis for infants with neuroblastoma is good, only 30 percent of children diagnosed after the age of 12 to 15 months survive (pdf).
Doctors have had few treatment options to choose from. Over the past 25 years the FDA has approved only one treatment for pediatric cancer, as compared to 50 for adult cancer patients. The NMTRC trial aims to match kids with effective drugs more quickly and to potentially identify treatments physicians hadn't previously thought of. It will enroll 14 kids with late-stage neuroblastoma who haven't responded well to conventional therapy. Physicians will take a biopsy from each patient and submit it for genomic analysis.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
The results of that analysis will be fed into the computer cluster, which will analyze it using software developed by the Van Andel Research Institute's Pediatric Cancer Translational Research Program in Grand Rapids, Mich. Doctors will then create a list of FDA-approved drugs to consider for treatment, Giselle Sholler, NMTRC chair and co-director of Van Andel's Pediatric Cancer Translational Research Program, said Thursday at a press conference hosted by Dell.
It's easier to develop treatments for adult patients because there is typically a large sample of people to study and there's a wider variety of FDA-approved drugs and dosages to choose from, Jeffrey Trent, president and research director of the Translational Genomics Research Institute (TGen), said at Thursday's press conference. TGen will operate the cluster at its facility in Phoenix.
The Dell compute cluster, will initially consist of 148 CPUs capable of 8.2 teraflops (or 8.2 trillion floating point operations per second) and can expand to 13 teraflops . (That's quite a bit of computing power although the world's fastest supercomputers operate at thousands of teraflops.). Eventually, 11 NMTRC sites will be tied into the compute cluster using the cloud-computing model—they'll communicate over a single network, be able to use the same software and all have access to a centralized database. This infrastructure will also make it easier to admit additional research facilities to the project, Trent said, adding that 17 other research sites have expressed interest in joining the pediatric cancer trial.
One model for the pediatric cancer clinical trial is the Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination (BATTLE) study, whose results thus far indicate that patients prescribed treatment with existing drugs based on their tumor biomarkers benefit more than patients whose treatment is not based on their tumor biomarkers. The study, conducted by University of Texas M. D. Anderson Cancer Center researchers and funded by the U.S. Department of Defense, successfully completed its Phase II clinical trial program last year. Another model, the I SPY 2 TRIAL (Investigation of Serial Studies to Predict Your Therapeutic Response with Imaging And moLecular Analysis 2), is designed to determine whether women with newly diagnosed locally advanced breast cancer benefit from investigational drugs in addition to standard chemotherapy rather than standard chemotherapy alone before having surgery.
Once a personalized neuroblastoma treatment model is established, the same high-performance computing and analysis approach can be expanded to a variety of other diseases, Sholler said.
Image portrays an MRI of an adrenal mass diagnosed as neuroblastoma. Courtesy of WebMD