Probing T-cell Infiltration in Tumors

The lack of infiltrating T cells in some tumors is not because neoantigens necessary for T-cell recognition are in short supply, according to recent research from the University of Chicago in Illinois. Instead, the study results suggest that these tumors, which respond poorly to immunotherapy, are unable to recruit a specific subset of dendritic cells that spurs T-cell trafficking, thereby helping initiate antitumor immunity (Proc Natl Acad Sci USA 2016;113:E7759–68).

Led by immunologist Thomas Gajewski, MD, PhD, the researchers analyzed two groups of metastatic melanoma samples from The Cancer Genome Atlas (TCGA): 106 had beneficial T cell–led inflammation and another 91 did not. Both cohorts had comparable mutational loads and immunogenic antigens, including mutation-derived neoantigens, they reported. Similar expression levels of HLA-A*0201, a cell-surface protein responsible for presenting antigens to the immune system, were also observed.

Next, the team synthesized 707 potential neoantigen peptides and evaluated them in both groups; no significant differences in peptide binding to HLA-A*0201 were found. They randomly assessed 40 peptides for their ability to elicit cytotoxic T-cell responses in vitro, and confirmed that approximately a quarter were immunogenic—five from the inflamed cohort, and six from the noninflamed cohort.

“The main question we addressed in our study was, ‘Does T cell–driven inflammation in tumors depend on mutational load and neoantigen density?’” says co–lead author Stefani Spranger, PhD. “It appears the answer is no.” The findings substantiate clinical observations that immunotherapy can be effective even in tumor types with a low number of neoantigens, such as renal cell carcinoma, adds Jason Luke, MD, the study's other co–lead author.

One reason why some tumors are devoid of infiltrating T cells and are noninflamed may be because they also lack Batf3-lineage CD103+ dendritic cells, Spranger notes. She and Gajewski have shown that tumor-intrinsicβ-catenin signaling hinders the recruitment and activation of this dendritic cell subset; this in turn keeps T cells from being activated and entering the tumor microenvironment (Nature 2015;523:231–5).

Another culprit that suppresses T-cell trafficking to tumors is PTEN loss and concomitant activation of PI3K signaling, reported by Patrick Hwu, MD, chair of the department of melanoma medical oncology at The University of Texas MD Anderson Cancer Center in Houston, and his colleagues (Cancer Discov 2016;6:202–16). This, together with the findings from Gajewski's group, demonstrates that “mutational load may be an important factor for immunogenicity in some cases, but there are many other elements that are underappreciated by our field,” Hwu says.

Luke and Spranger also examined the correlation between mutational load and T cell–driven inflammation in cancers besides melanoma. The team developed a 160-gene signature of T-cell inflammation and applied it to 30 different solid tumor samples from TCGA, including prostate cancer and lung adenocarcinoma. “The number of mutations present in a given tumor type had little bearing on whether or not inflammation occurred,” Luke says.

“It's a relief to know that mutational load is not the be all and end all of immunotherapy responsiveness,” he adds. “We're now exploring therapeutic interventions, including β-catenin inhibitors, that may improve T-cell trafficking into noninflamed tumors and ultimately increase the number of patients who benefit from immune checkpoint blockade.” –Alissa Poh

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