Skip to main content
Premium Trial:

Request an Annual Quote

Dell Provides Funding, Cloud Infrastructure to Support Pediatric Cancer Personalized Medicine Effort


By Uduak Grace Thomas

Dell said this week that it is providing funding and a private cloud infrastructure to support pediatric cancer research using gene-expression profiling and, eventually, next-generation sequencing.

The first project in the initiative is an ongoing clinical trial that aims to select targeted treatments for children with neuroblastoma based on their gene- expression profiles.

Dell is committing $4 million to the initiative this year, with plans to increase that amount in the future as the trial grows and as the company expands the effort to include other studies focused on other pediatric cancers.

This trial, which began at the end summer, is being conducted by the Neuroblastoma and Medulloblastoma Translational Research Consortium, the Van Andel Research Institute, the Translational Genomics Institute, and a number of other research institutes and hospitals.

The trial has enrolled three patients to date but plans to enroll 14. As BioInform sister publication Pharmacogenomics Reporter reported in June, the researchers will first conduct a feasibility trial, which will take between six and 12 months, to demonstrate that they can use gene expression profiling on the Affymetrix GeneChip to guide therapy for pediatric cancer patients (PGx 6/1/2011).

Future stages of the trial will involve next-generation sequencing, organizers said this week.

In addition to the funding, Dell is providing a private cloud, which TGen will host, comprising PowerEdge blade servers, PowerVault storage arrays, Compellent Storage Center arrays, and Force10 network infrastructure. Dell also plans to install workstations at participating sites so that each center has dedicated hardware with which they can access the cloud.

Dell said the system will provide maximum performance of about 13 teraflops and is expected to provide a 1,200 percent increase in compute power over TGen's existing clinical compute cluster.

The platform will have 148 central processing units, 1,192 cores, 7.1 terabytes of RAM, and about a quarter of a petabyte of disk storage, Dell said.

The partners believe that the improved infrastructure will permit near-real-time processing of genomic information on patient tumors and prediction of the best drugs for specific patients.

A Collaborative Tool

In addition to speeding up the data analysis, the cloud infrastructure will also support collaboration among the physicians, genetic researchers, pharmacists, and computer scientists working in the trial, James Coffin, vice president and general manager of Dell's healthcare and life sciences division, said at an event announcing the investment that was held this week in New York.

Coffin said the trial infrastructure will support data from 12 US medical centers in its first year and 20 institutions the following year with plans to bring on additional centers at a later date.

The trial specifically targets relapsed patients who have exhausted all other therapy options without seeing an improvement in their condition, Giselle Sholler, the chair and co-director of VARI's pediatric cancer translational research program, explained during the event.

Sholler, who is leading the trial, explained that the process begins by collecting a small amount of the patient's tumor, which then undergoes RNA-expression analysis in a CLIA-certified laboratory.

TGen researchers then analyze the data using VARI's FDA-approved software, which processes the expression data from the Affy chips and matches this data to treatments, Spyro Mousses, vice president of TGen's office of innovation, told BioInform.

The software suggests a list of treatment options that target molecular pathways and halt the cancer's progress selected from a catalog of 150 drugs that have recommended pediatric dosing, Sholler said.

The report of the analysis findings and treatment options is reviewed by a molecular tumor board comprised of oncologists, medical geneticists, bioinformaticians, and pharmacists from 11 different institutions who use this information to design an individualized treatment plan for the patient.

As a final safety step, the treatment plan is reviewed by an independent medical monitor. Once approved, the patient will begin the therapy.

The entire process — from the time the tumor is biopsied to the onset of treatment — is expected to take less than two weeks, Sholler said.

Scaling Up

Sholler said that while treatment decisions are currently based on expression profiling alone, the partners plan to eventually include both DNA and RNA sequence information from patients' tumors.

Jeffrey Trent, president and research director of TGen and VARI, said that the consortium is currently exploring next-generation sequencing platforms from all the major providers, including Illumina, Life Technologies, and Pacific Biosciences, and aims to develop analysis infrastructure that works with data from all the systems.

He added that the NCI and TGen will share the sequencing responsibilities for the trial.

Additionally, TGen researchers have developed an analysis pipeline, comprised of commercial and in-house solutions, that will align sequence reads, perform base calling, and annotate the genomes, Mousses told BioInform.

A second pipeline will perform a series of steps to match genetic information to "higher-level concepts," such as protein interaction data or drug-gene interaction data contained in published literature and public databases, he said.

This way, if a new mutation is detected, the pipeline "will allow us to abstract the higher-level concepts that might also be linked to a drug, so that we could pull information that might not be in the database," Mousses explained.

Over the next six months, the partners plan to install the hardware for the cloud infrastructure at TGEN as well as to implement workstations at each study site, Trent said.

For now, however, since the trial is only using expression data, which isn't particularly "data heavy," that information can be transferred across standard network infrastructure, he said.

Have topics you'd like to see covered in BioInform? Contact the editor at uthomas [at] genomeweb [.] com.

The Scan

UCLA Team Reports Cost-Effective Liquid Biopsy Approach for Cancer Detection

The researchers report in Nature Communications that their liquid biopsy approach has high specificity in detecting all- and early-stage cancers.

Machine Learning Improves Diagnostic Accuracy of Breast Cancer MRI, Study Shows

Combining machine learning with radiologists' interpretations further increased the diagnostic accuracy of MRIs for breast cancer, a Science Translational Medicine paper finds.

Genome Damage in Neurons Triggers Alzheimer's-Linked Inflammation

Neurons harboring increased DNA double-strand breaks activate microglia to lead to neuroinflammation like that seen in Alzheimer's disease, a new Science Advances study finds.

Long COVID-19 Susceptibility Clues Contained in Blood Plasma Proteome

A longitudinal study in eBioMedicine found weeks-long blood plasma proteome shifts after SARS-CoV-2 infection, along with proteomic signatures that appeared to coincide with long Covid risk.