Chromatin-Remodeling Genes Promote Immunotherapy Resistance

A pair of studies has uncovered a new mechanism of resistance to checkpoint inhibitors that involves proteins that remodel chromatin to regulate gene expression.

Previous studies have identified some mechanisms that enable tumors to withstand immunotherapy. For instance, recent studies point to inhibition of IFNγ signaling, which is crucial for antitumor responses. In most cases, however, “the causes for resistance are largely unknown,” says Kai Wucherpfennig, MD, PhD, of Dana-Farber Cancer Institute in Boston, MA.

To search for genes that might be responsible, Wucherpfennig and his colleagues performed a genome-wide CRISPR/Cas9 screen to look for genes that would sensitize a mouse melanoma cell line that is resistant to PD-1 and CTLA4 inhibitors. Inactivation of any of more than 100 genes made the cancer cells susceptible to T-cell attack, the scientists found.

Three of the genes that hadn't been previously linked to immunotherapy resistance—Pbrm1, Arid2, and Brd7—encode proteins of the PBAF form of the SWI/SNF complex, which remodels chromatin and modifies gene expression. To investigate how these proteins foster resistance, the researchers analyzed gene expression in mouse melanoma cells lacking each of the three genes. Loss of PBRM1 and ARID2 increased expression of genes involved in IFNγ signaling. For example, two of the genes upregulated by IFNγ encode the chemokines CXCL9 and CXCL10, which draw CD8 T cells into tumors. The researchers found that cells lacking PBRM1 increased production of these chemokines.

Working separately, a team led by Eli Van Allen, MD, and Toni Choueiri, MD, also of Dana-Farber, zeroed in on PBRM1, too. The scientists wanted to track down genes that promote resistance to checkpoint inhibitors in clear cell renal cell carcinoma (ccRCC) and performed exome sequencing on tumor samples from 35 patients with metastatic ccRCC who had received anti–PD-1 therapy. They discovered that nine of the 11 patients who showed complete or partial responses carried PBRM1 mutations, whereas only three of 13 nonresponders did.

The team confirmed the result in a second group of 63 patients with ccRCC who had received anti–PD-1 treatments, sometimes in combination with anti-CTLA4 therapy. In this group, 17 of 27 responders harbored PBRM1 mutations, compared with only four of 19 nonresponders.

When the researchers measured PBRM1 gene expression in ccRCC cell lines that lack PBAF and in ccRCC tumors from patients with PBRM1 mutations, they also uncovered a link to IFNγ. The cells showed increased expression of genes in the JAK–STAT3 pathway, which promotes IFNγ signaling.

The results of the studies “implicate chromatin remodeling biology in the selective response to immunotherapy,” says Van Allen. Given that the same mechanism promotes resistance in two different types of tumor cells, “it could be relevant in a lot of human cancers,” adds Wucherpfennig. Whether the discovery will provide clinical benefits remains to be seen. For example, it's not clear if this resistance mechanism is a feasible drug target, says Van Allen, because the SWI/SNF proteins have numerous functions and interactions that require additional study. Choueiri adds that “further larger prospective studies are needed to investigate PBRM1 as a biomarker in the setting of checkpoint blockers.”

“They [the two groups] add pieces to the puzzle” of how resistance to checkpoint inhibitors occurs, says Antoni Ribas, MD, PhD, of the University of California, Los Angeles, who wasn't connected to either study. “But the puzzle is still very complicated,” he cautions. Brent Hanks, MD, PhD, of Duke University School of Medicine in Durham, NC, agrees that the studies have uncovered one of many factors that likely contribute to resistance. PBRM1 “is another marker that will help us determine whether a tumor is vulnerable to T cell–mediated killing.” –Mitch Leslie