Skip to main content
Premium Trial:

Request an Annual Quote

Fred Hutchinson Researchers Harness RNAi Screening to Create Cancer Drug ‘Oracles’


Researchers from the Fred Hutchinson Cancer Research Center are developing a new approach for identifying novel cancer drug targets, combining patient-derived stem cells that retain the specific characteristics of the original tumor with genome-wide RNAi screens to create so-called therapeutic prediction oracles.

According to Fred Hutchinson researcher Patrick Paddison, who described the approach last month at the GTC Nucleic Acid Summit in San Francisco, he and his colleagues adopted the term “oracle” in response to the common strategy of designing cancer drugs that inactivate the molecular engines of the disease.

“When people are doing cancer-lethal screens, the reflex reaction is to filter screen results … to identify activities that are inappropriately activated [and] are drivers of the cancer,” he told Gene Silencing News this week. “I use the term oracle to contrast that approach.

“We’re not trying to go out and rationally find these drivers,” he added. “What we’re trying to do is find the pressure points [in molecular processes] that arise uniquely in the cancer or transformed cells, and try to use those against the cancers. Whether they are drivers or not is somewhat irrelevant.”

Paddison said that the oracle approach stems from the work of other research groups that created glioblastoma multiforme, or GBM, stem cell models wherein stem cell-like cells are derived in culture from patient isolates, retaining tumor-initiating potential and tumor-specific genetic and epigenetic signatures more reliably than typical tissue culture cell lines.

Importantly, the technique also uses human neural progenitor cells, which are candidate cells of origin for brain tumors, as controls. These cells “have a lot of the same features that our brain tumor stem cells have … [expressing] many of the same genes,” he noted. They also are derived and grow in the same conditions as the patient-derived cells, improving their value as non-cancer controls.

Earlier this year, Paddison and collaborators reported in Genes & Development on genome-wide RNAi screens in the brain tumor isolates, which uncovered a novel viability requirement for GBM that does not appear to impact non-cancer cells.

The screens identified the plant homeodomain-finger domain protein PHF5A as “differentially required” for GBM stem cell expansion, as compared with untransformed neural stem cell controls and fibroblasts, according to their paper.

“Given PHF5A’s known involvement in facilitating interactions between the U2 snRNP complex and ATP-dependent helicases, we examined cancer-specific roles in RNA splicing,” the researchers wrote in Genes & Development. They found that in the GBM stem cells, but not in the controls, PHF5A facilitates recognition of exons with unusual C-rich 39 splice sites in thousands of essential genes.

Meanwhile, PHF5A knockdown in the stem cells inhibited splicing of these genes, leading to cell cycle arrest and loss of viability. This effect did not occur in the untransformed controls, astrocytes, or fibroblasts. Inhibition of PHF5A also compromised GBM stem cell formation in vivo and inhibited the growth of established GBM patient-derived xenograft tumors.

Taken together, the findings demonstrate that PHF5A is required to maintain proper exon recognition in brain tumor-initiating cells but, critically, not non-cancer cells.

Paddison said that his lab has now banked about 30 patient-derived brain tumor isolates, which can all be used for screening experiments to determine how widespread a particular genetic vulnerability is between cancer patients — something he hopes will aid in the creation of personalized treatments.

“What we’re very interested in is determining if there are biomarkers — whether the activation of a particular pathway or gene, or a difference in a particular cellular process — that tell us which patient subsets will be sensitive to a particular candidate therapeutic target,” he said.

He also said that he and his colleagues are exploring the development of oracles for other cancers, but that they are still lacking a way to propagate those cancers’ stem cells as easily as they can with GBM stem cells.