“Dependency Map” Bears Fruit for Childhood Cancers

The first comprehensive catalog of genetic vulnerabilities in pediatric tumors has identified dozens of new therapeutic targets, mostly in genes not known to harbor activating mutations.

The findings—part of the Cancer Dependency Map (DepMap) Project at the Broad Institute in Cambridge, MA—reveal that pediatric cancers, despite their lower mutational burden relative to adult tumors, harbor just as many genes that cancer cells rely upon for growth and survival (https://depmap.org/portal/; Nat Genet 2021;53:529–38). Many of those cancer-essential genes are also specific to pediatric tumors, highlighting the need for focused drug development on childhood cancers.

DepMap builds on prior Broad initiatives, including the Cancer Cell Line Encyclopedia and Project Achilles, but adds CRISPR-based loss-of-function screens. Since the project's first public data release in 2018, researchers have used DepMap to characterize and validate druggable dependencies in neuroblastomas, Ewing sarcoma, and rhabdoid tumors, and to determine how promising drugs inhibit pediatric cancer cell lines. But until now, there hadn't been an analysis of all the pediatric samples in the DepMap library of 800-plus cell lines, mostly of adult cancer origin.

Francisca Vazquez, PhD, and her Broad colleagues performed that analysis on 82 cell lines from 13 types of childhood cancer. Working with study co-lead Kimberly Stegmaier, MD, of Dana-Farber Cancer Institute (DFCI) in Boston, MA, Vazquez and her team identified 235 genes required for the growth of certain cancers. About two thirds of these dependencies were shared between adult and pediatric tumors, but 14% were unique to childhood cancers.

Surprisingly, the researchers found little correlation between the number of genetic alterations in the cancer genome and the number of genes needed for a tumor cell to persist. Despite their relatively simple genomes, the pediatric tumors had just as many selective dependencies as their mutation-laden adult counterparts—with changes in gene expression, not sequence, largely behind their vulnerabilities.

“That's an unexpected finding with profound consequences,” says Paul Geeleher, PhD, of St. Jude Children's Research Hospital in Memphis, TN, who was not involved in the research. “It gives you potential therapeutic targets that you wouldn't have identified using mutation profiles—or, in all honesty, potentially any other data source.”

One novel vulnerability discovered was TRIM8, a gene found to be essential only in Ewing sarcoma, a pediatric solid tumor involving bone. Since identifying the target through the Pediatric Cancer DepMap effort, Stegmaier's lab has been studying the role of TRIM8 — which encodes an enzyme that regulates cellular turnover of proteins — in the disease and working on pharmacological strategies for disrupting TRIM8 function.

In another study, a team led by St. Jude's Charles Roberts, MD, PhD, and DFCI's William Hahn, MD, PhD—both of whom were involved in DepMap—built on the observation that rhabdoid tumors were highly dependent on MDM2 and MDM4, two key suppressors of p53 activity. They showed that common mutations in SMARCB1 caused these tumors to rely on MDM2 and MDM4 for survival. Inhibition of one or both proteins using clinical-stage drugs such as idasanutlin (Roche) led to marked tumor regression in mouse models (Cancer Res 2019;79:2404–14). The dual MDM2/MDM4 inhibitor ALRN-6924 (Aileron) is now being tested in children with intact p53.

The pediatric DepMap is “the real deal,” says Roberts. “It really will be a key way that we're going to get new insights into vulnerabilities that will lead to translation.” –Elie Dolgin