NEW YORK (GenomeWeb) – A team led by researchers from the University of California, San Diego's School of Medicine and Moores Cancer Center has uncovered 172 new combinations of mutated genes and drugs that could form the basis for future cancer therapies, work that was published today in the journal Molecular Cell.
"Oncologists here at Moores Cancer Center at UC San Diego Health and elsewhere can often personalize cancer therapy based on an individual patient's unique cancer mutations," Trey Ideker, senior author and professor of genetics at UC San Diego School of Medicine, said in a statement.
However, "the vast majority of mutations are not actionable — that is, knowing a patient has a particular mutation doesn't mean there's an available therapy that targets it," Ideker added.
Gene mutations in cancers usually do one of two things, promote cell growth or prevent cell death. Many therapies inhibit cell growth by targeting the mutations that promote it, since it's much harder to develop therapies that restore tumor-suppressor genes.
An emerging strategy to develop therapies identifies "synthetic lethal" genetic interactions between tumor suppressor genes (TSGs) and other genes so that simultaneous disruption of both gene functions causes rapid and selective cell death, the researchers said in the study. However, there is currently no large cross-species data set available to help pinpoint these potential therapy targets.
"The goal of this study was to expand the number of mutations we can pair with a precision therapy," Ideker said.
To create a resource that could provide researchers with a broad network of synthetic lethal interactions connecting TSGs to druggable targets (DTs), Ideker and his colleagues began by screening 169,000 interactions between yeast orthologs of human TSGs and DT genes.
To test for genetic interactions among all possible TSG and DT orthologs, the researchers used synthetic genetic array technology, which uses high-throughput robotic colony pinning on agar to create and score the growth of large numbers of double gene deletion strains in parallel, in the yeast Saccharomyces cerevisiae.
In total, 1,420 synthetic lethal interactions and 996 epistatic interactions were identified under untreated conditions, with an average of 14 and 11 synthetic lethal interactions per TSG and DT respectively, the researchers stated in the paper. They also found that a pan-cancer analysis of The Cancer Genome Atlas identified 16 TSGs that are associated with coordinated upregulation in genes for which a negative TSG-gene interaction is found in yeast.
Using the yeast network as a guide, the researchers then performed a tumor suppressor-drug interaction screen in human cancer cells. They prioritized 21 drugs for which the yeast DTs were involved in the greatest numbers of synthetic lethal interactions. The research team tested these drugs one at a time for lethal interaction with 112 different tumor-suppressor gene mutations in human cancer cells growing in the lab.
The researchers used these first two genetic maps to create two conserved cancer networks (CoCaNets). Then they continued to look at the intersection between the first two genetic maps to produce three more CoCaNets to help pinpoint viable therapy pathways.
Ultimately, the researchers ended up with 172 drug-gene mutation combinations, 158 of which had not been previously discovered, that successfully killed both yeast and human cancer cells.
While this offers many new treatment avenues to explore, the researchers noted that their results indicated that Irinotecan, a drug currently only indicated for use in colon cancer by the US Food and Drug Administration, should be evaluated for efficacy in any tumor with a mutation that inhibits RAD17 — a tumor-suppressor gene that normally helps cells fix damaged DNA.
"We've created an important translational research resource for other scientists and oncologists. And since many of the cancer-killing interactions we discovered involve already FDA-approved drugs, it may mean they could reach clinical translation rapidly," John Paul Shen, co-first author, clinical instructor, and postdoctoral fellow at UC San Diego School of Medicine and Moores Cancer Center, said in a statement. "If these results are validated in subsequent testing, in the future an oncologist will have many more options for precision cancer therapy."
These results are preliminary, and Shen and his colleagues noted that the next step will be to test these drug-gene mutation combinations in more human cancer cell types. They hope this will eventually transition into mouse models.
The research team has made the data from all five networks available on the Network Data Exchange. With so many new combinations to explore, the researchers hope that other research teams will make use of their data to further test each combination in a variety of conditions.