A new study suggests that patients whose melanoma tumors have a high number of genetic mutations might respond best to drugs that block CTLA-4.
Maui – The more mutations in the tumors of a patient with melanoma, the more likely the patient may be to respond to drugs that block cytotoxic T-lymphocyte antigen (CTLA)-4, according to a recent study. These findings may provide a rationale for examining exomes of patients for whom these treatments are under consideration, study authors conclude.
Ipilimumab earned Food and Drug Administration approval for use as an anti-CTLA-4 antibody in melanoma because it provides statistically significant benefits, said Hensin Tsao, M.D., Ph.D., at MauiDerm 2015. He is clinical director of the Massachusetts General Hospital (MGH) Melanoma & Pigmented Lesion Center, director of the MGH Melanoma Genetics Program and professor of dermatology at Harvard Medical School.
However, he said, "The real question is whether there is any way to predict who potentially could respond to CTLA-4 blockade and who wouldn't."
To answer this question, researchers at New York's Memorial Sloan-Kettering Cancer Center used whole-exome sequencing on tumor tissues from patients with melanoma who were treated with ipilimumab or tremelimumab (a CTLA-4 blocker under development by MedImmune under license from Pfizer, Inc.). This allowed investigators to analyze all protein-expressing parts of the genome, said Dr. Tsao.
"As far as we can tell, the exome examines about one percent of the genome." The rest consists of regulatory elements and non-coding RNAs that might be biologically important but do not necessarily code for protein, he said. Although this approach provided a targeted analysis of the genome, he added, it focused on some of the most important elements – those used to make tumor-associated protein products.
NEXT: Long-term benefits
This analysis showed that the tumors of people who enjoyed long-term benefits – defined as complete response for at least six months – had more missense mutations than did patients who experienced minimal benefit – defined as posttreatment progression.1 In one analysis, 17 patients who had more than 100 such mutations survived an average of around 85 months, versus 50 months for eight patients who had less than 100 mutations (p = 0.04). "Patients with more mutations lived longer," said Dr. Tsao.
Further analysis showed that these patients had many more epitopes, or proteins to which the immune system responds. As such, said Dr. Tsao, researchers hypothesized that every time a DNA mutation (resulting from ultraviolet exposure to the skin) produces a novel protein sequence, "The body no longer recognizes it as 'self,' because it's altered. And that becomes a source of immune response. In theory, at least, the idea is that all these mutations stimulate the immune system."
Therefore, said Dr. Tsao, people with high numbers of mutations responded better to immune treatment because "fundamentally they have more antigens they can present to the immune system. The idea is that the mutations themselves, beyond having an effect on the function of the cell, influence the immune response. It's something of a new look at what mutations do" in melanoma.
Traditionally, he explained, researchers believed that functional mutations simply boosted expression of or killed off certain genes, leading to the development of cancer. "Now we have a third element – the host immune element. When you generate new protein sequences, peptides and epitopes, you're exposing the immune system to new sequences that it can recognize." And because these mutations only happen in the cancer, Dr. Tsao said, the immune response generated is relatively tumor-specific. "Other cells such as blood, skin, and brain cells don't generate these mutations."
As part of the study, researchers used genomewide somatic neo-epitope analysis and patient-specific HLA typing to identify a neoantigen profile that was specifically present in tumors with a strong response to CTLA-4 blockade. They then validated this antigen profile in a second set of 39 patients with melanoma who were treated with anti-CTLA-4 antibodies. The predicted neoantigens indeed activated T cells from the patients treated with ipilimumab.
Furthermore, among the 15 patients from the original analysis who lacked the epitope signature, less than 10% survived beyond 30 months. Conversely, most of the 10 patients whose tumors had the signature survived beyond 80 months (p <0.001).
Altogether, said Dr. Tsao, the study's findings "may be just one part of the entire equation of why people respond to immune treatments. But it's an important new one." He expects that other parts of the picture will come into focus as analysis continues. In particular, "These findings could be generalized in the future to understand phenomena such as tumor regression and halo nevi." As exome sequencing becomes more rapid and routine, he foresees a day when "whole-exome analysis becomes part of the tumor analysis. Both the number and site of mutations could play a role in predicting prognosis and response to treatment."
Disclosures: Dr. Tsao serves on the editorial boards of several dermatology journals, has received research funding from the National Institutes of Health and the American Skin Association, and received consulting fees from Lubax Inc.
1. Snyder A, Makarov V, Merghoub T, et al. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N Engl J Med. 2014;371(23):2189-99.