A CRISPR-based screen of all 19,050 genes in the genome has revealed around 100 genes that cancer cells must express in order for T cells—and, thus, immunotherapies—to effectively recognize and kill tumors.

A genome-wide screen has revealed around 100 key genes that cancer cells must express in order for the immune system to effectively recognize and kill them.

The new findings may help explain why checkpoint inhibitors, cancer vaccines, and other immunotherapies mediated by T cells are effective against only some tumors (Nature 2017;548:537–42). They also point toward candidate biomarkers for identifying patients most likely to benefit from immunotherapies, and reveal potential targets for drug combinations that might overcome resistance mechanisms. “We hope to use this information to make nonresponders be responders,” says lead investigator Nicholas Restifo, MD, of the NCI Center for Cancer Research in Bethesda, MD.

Restifo and his colleagues developed a CRISPR-based screen to systematically knock out genes in a human melanoma cell line. They delivered more than 123,000 guide RNAs corresponding to all 19,050 protein-coding genes plus another 1,864 that targeted miRNAs; pooled all the gene-edited cancer cells together; added CD8+ effector T cells engineered to recognize tumor antigens; and then sequenced the guide RNAs of all surviving melanoma cells—which told the researchers which genes, when knocked out, rendered the cells resistant to T-cell killing.

In total, Restifo's team found more than 500 candidate genes—around 100 of which they validated—that had to be present in the tumors for T cells and, thus, cancer immunotherapies to do their job. Consistent with the decades-old observation that defective antigen processing explains why some patients don't respond to cancer vaccines, the top two hits in Restifo's screen were involved in antigen presentation. Other top candidates played a role in the release of cytokines, which can drive the infiltration and activation of T cells.

There were also many hits that had no prior established link to T-cell function. One such gene, called APLNR, encodes a rhodopsin-like receptor that plays a role in regulating blood pressure but is often mutated in tumors. Restifo and his colleagues showed in patient data, mouse models, and cell-based experiments that loss-of-function mutations in APLNR reduce the efficacy of immunotherapies. Further mechanistic investigations showed that the APLNR gene product binds to JAK1 to modulate responses to IFNγ in tumor cells.

“It's certainly a comprehensive and heroic effort,” says Lisa Butterfield, PhD, of the University of Pittsburgh, PA. “I'm sure that this study is going to yield a number of interesting targets in the coming months and years.”

Restifo's team's findings, published in early August, complement those reported in late July by Nick Haining, MD, of Dana-Farber Cancer Institute in Boston, MA, who ran a smaller CRISPR-based screen in mouse models and also pinpointed genes involved in antigen presentation and IFNγ signaling as being essential for immunotherapies to work (Nature 2017;547:413–8).

“Combined,” says Antoni Ribas, MD, PhD, of the University of California, Los Angeles, “these studies are providing an increasing molecular understanding of how cancer cells react to T cells and how some may develop resistance.” –Elie Dolgin

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/content/early/by/section.

©2017 American Association for Cancer Research.
2017
American Association for Cancer Research.