fb-pixel Skip to main content

Targeting the ‘gene next door’ to fight hard-to-treat tumors

Konstantinos Mavrakis works in the lab at Novartis Institutes in Cambridge. “If we can make the right therapy, it would selectively kill tumors.” said a Novartis oncologist.David L. Ryan/Globe staff

CAMBRIDGE -- Cancer scientists toiling here in one of the world’s largest research clusters are setting their sights on what could become a game-changing approach to battling hard-to-treat tumors.

Experiments at Novartis Institutes for BioMedical Research and the Broad Institute of MIT and Harvard, described in a pair of papers published last week in the medical journal Science, identified the gene PRMT5 as a promising target that drug developers can use in knocking out cancer cells.

While the full impact of their findings isn’t yet clear, Novartis has deployed between 30 and 40 researchers here to screen drug compounds aimed at PRMT5, which they have linked directly to some of the most intractable blood, brain, and pancreatic cancers.


Two research teams across town at the Broad, meanwhile, are engaged in follow-up studies to better understand the mechanisms of PRMT5 and determine if there might be other drug targets that function in similar ways.

Researchers caution they are a long way from fielding effective cancer medicines based on discoveries from their experiments, code-named Project Drive at Novartis and Project Achilles at the Broad. But if their initial hypotheses pan out, the work could open a new cancer research field and give industry and academic drug discovery teams an important new tool.

“This is an exciting new research target,” said Bill Sellers, global head of oncology at Novartis Institutes. “If we can make the right therapy, it would selectively kill tumors.”

The PRMT5 gene wasn’t the initial focus of the two research programs. Working independently, the Broad and Novartis groups sought to zero in on a cancer research mystery that has long bedeviled scientists: the nature of “tumor suppressor” genes that mutate or disappear from cells that become cancerous. In healthy cells, these genes act to fend off the growth of tumors.


Their experiments, which at first examined a key tumor suppressor gene called CDKN2A, suggested that many cancers with the missing tumor suppressors depend on PRMT5 for survival. They concluded that eliminating that bystander gene could kill cancer cells with the missing suppressors.

This is important because cancers that lose their tumor suppressor genes are often the most difficult to treat. While scientists have designed molecules that can switch off other cancer-causing genes known as oncogenes, tumor suppressor genes have historically proved resistant to such efforts.

“Finding therapeutic targets that are going to be active when tumor suppressors are deleted is very challenging, and we think we’ve found one,” Sellers said.

Levi Garraway, an institute member at the Broad who oversaw its research, calls PRMT5 “the gene next door” because it’s essential to one of the most common sections of the genome to get deleted in cancer cells.

“Our data and the data from Novartis are telling us there might be something you can exploit to make a drug against PRMT5,” said Garraway, who also directs the Joint Center for Cancer Precision Medicine at the Broad and partners Dana-Farber Cancer Institute and Brigham and Women’s Hospital in Boston. “This gives us a window of opportunity.”

Novartis scientists, who began their work in mid-2013, screened thousands of genes contained in 390 cancer cell lines whose genomes had previously been mapped through a Novartis-Broad project called the Cancer Cell Line Encyclopedia.

When tested, the PRMT5 “jumped right out” for its potential to knock out cancers with deleted tumor suppressors. Sellers said.


PRMT5 is a gene well known in the cancer research community, and some other drug programs are also targeting it. Cambridge-based biotech startup Epizyme Inc., working in partnership with British drug giant GlaxoSmithKline PLC, is seeking to inhibit the gene to treat an aggressive form of non-Hodgkin lymphoma.

But the Novartis and Broad research points to a broader drug discovery approach based on their findings of how the target interacts with cancers missing tumor suppressors.

“This data really clarifies the path for going after PRMT5 as we move forward,” said Garraway at the Broad. “We have a target that’s relevant to a large number of hard-to-treat cancers that we now may be able to deliver drugs against.”

Sellers said cancer research at Novartis has flowed from the mapping of the human genome more than a decade ago to a current effort to sequence the genomes of cancer cells themselves. (The genetic makeup of between 10,000 and 20,000 types of tumors is already known.)

“This is helping us find all the mutations that cause all the cancers,” he said. “The next step is how do we make drugs that take advantage of these mutations.”

Robert Weisman can be reached at robert.weisman@globe.com. Follow him on Twitter @GlobeRobW.