Abstract
Researchers have developed a new software tool, called LOHHLA, which can pinpoint allele-specific HLA loss in tumors. When applied to a cohort of patients with non–small cell lung cancer, this genomic event was detected in 40% of cases, increasing in frequency as the disease spread to the brain.
Loss of heterozygosity (LOH) at the HLA locus on chromosome 6—a known occurrence in cancer—reduces the capacity for neoantigen presentation to cytotoxic T cells, thereby aiding immune evasion on the part of tumors. Until now, determining the extent and frequency of this genomic event has been challenging, but a new computational method developed by researchers at University College London and the Francis Crick Institute, also in London, UK, may prove useful.
“Because the HLA locus is highly polymorphic, very few sequencing reads align well with the human reference genome, making it quite difficult to assess whether LOH has occurred,” explains Rachel Rosenthal, one of the study authors (Cell 2017;171:1259–71). She also presented the researchers' work during the 2017 AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics, held October 26–30 in Philadelphia, PA.
The team created a software tool, dubbed LOHHLA, that aligns sequencing reads to known information about a patient's specific HLA type instead. As such, they could pinpoint “exactly which HLA alleles are lost at any particular time in the tumor,” says senior author Charles Swanton, MD, PhD. When LOHHLA was applied to a cohort of 90 patients enrolled in the TRACERx study, all with non–small cell lung cancer (NSCLC), the team detected loss of the paternal or maternal HLA allele in 40% of cases. The prevalence of HLA LOH was higher in lung squamous cell carcinomas than in lung adenocarcinomas, they found.
“This event is subclonal, occurring more commonly in the branches, not the trunk, of the tumor's evolutionary tree,” Swanton observes. “As well, we saw that the same HLA allele was lost multiple times, in separate cellular populations of the same tumor, indicating that there's profound selection going on in an evolving tumor mass.” HLA LOH increased in frequency as lung tumors spread to the brain, he notes, suggesting that this process could enable metastasis.
Besides HLA LOH, other genomic alterations being investigated as potential mechanisms of immune evasion include IFNγ pathway mutations and B2M deficiency. Led by Katerina Politi, PhD, a group of researchers at Yale Cancer Center in New Haven, CT, recently examined tissue samples from 14 patients with NSCLC whose tumors had become resistant to immune checkpoint inhibitors (Cancer Discov 2017;7:1420–35). They reported homozygous loss of B2M—which results in tumors lacking cell-surface HLA class I expression—in one case, as well as in a matched patient-derived xenograft (PDX). Low B2M protein levels were found in another two PDXs established from the researchers' collection.
Politi and her team also used CRISPR/Cas9 to knock out B2M in an immunocompetent mouse model of NSCLC and showed that this conferred resistance to PD-1 blockade. “We wanted to functionally test whether B2M loss would result in acquired resistance to immune checkpoint inhibitors,” she says. “Previously, this had been more of a correlation.”
Overall, Politi considers the work done by Swanton, Rosenthal, and their colleagues “really interesting. They've found that HLA LOH happens in a large fraction of lung cancers, which suggests a predisposition to immune escape. Just how this impacts therapeutic resistance is not really known, but I think it's important to figure out.”
Better characterizing the neoantigen landscape of tumors is equally essential, Swanton says—hence the need for tools such as LOHHLA. Otherwise, “one could potentially misinterpret the number of productive neoantigens present on a tumor cell, based on not knowing which HLA alleles have been lost.”
“Thinking about this brave new world, if you like, of personalized immunotherapy, it will be key to understand individual patients' HLA allele repertoire,” he adds. “You wouldn't want to develop vaccines or other cell therapies against neoantigens that simply can no longer be presented for immune surveillance.” –Alissa Poh
For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/content/early/by/section.
RSS Feeds