Many patients do not respond to checkpoint inhibitors given alone. We think an effective cancer vaccine is a promising way to drive T cells into the tumor and also enhance T-cell expansion, broadening their repertoire.

— Patrick A. Ott, MD, PhD

Tweet this quote

The field of cancer vaccines may be reinvigorated by a new understanding, and the therapeutic leveraging, of neoantigens. Researchers from Dana-Farber Cancer Institute in Boston are exploring this novel approach as a means of protecting patients with high-risk melanoma from recurrence. Early results from a phase I study were reported at the 2nd International Cancer Immunotherapy Conference by Patrick A. Ott, MD, PhD, Clinical Director of the Melanoma Center and the Center for Immuno-Oncology.¹

The senior author of the abstract is Catherine J. Wu, MD, also of Dana-Farber and the Broad Institute in Cambridge, Massachusetts.

Neoantigens arise from somatic tumor mutations and can encode changes in amino acids; these changes can generate novel peptides with the potential to stimulate T-cell responses against the tumor. Since these neoepitopes provide exquisite tumor specificity and are expected to be highly immunogenic, they have long been envisioned as highly valuable tumor antigens. Growing evidence has supported tumor neoepi­topes as key antigens that can trigger immune-mediated tumor killing in vivo across various cancer immunotherapy approaches, including checkpoint blockade and adoptive transfer with tumor-infiltrating lymphocytes, according to Dr. Ott. 

Catherine J. Wu, MD

“Neoantigens are a very attractive target as a vaccine because they are novel antigens similar to a virus that are potentially much more visible to the immune system than antigens previously used in cancer vaccines. The mutated antigen is present only in the tumor, not in normal cells,” he explained. “Neoantigens have been challenging to use in vaccines, however, because the technology wasn’t there. Now, with new sequencing technologies and better ways to predict the mutations that will actually generate peptides the immune system will be able to see, it is easier to identify the whole spectrum of neoantigens from a given tumor. With this approach, what the individual mutation actually does to the tumor is not as relevant as the fact that it makes the tumor different and recognizable for the immune system.”

This approach is personalized because each patient’s tumor is unique. The hypothesis of the study is that a personalized multiepitope neoantigen vaccine can stimulate naive and preexisting tumor-reactive T cells to broaden and enhance the tumor-specific T-cell repertoire in cancer patients, Dr. Ott said

Study in High-Risk Patients

By leveraging next-generation sequencing technologies and new analytic pipelines to systematically identify personal tumor neoantigens, Dr. Ott and his team tested the feasibility, safety, and immunogenicity of a multiepitope cancer vaccine targeting personal tumor neoantigens in patients with stage IIIB–IV, M1b surgically resected melanoma. This subgroup of patients historically has a 40% to 80% risk of recurrence.

Patients’ tumors were sequenced and peptides were generated based on the mutations. Since not all epitopes are compatible with the immune system, the researchers developed a means of selecting those that will bind to human leukocyte antigen and thus be presented to the immune cells. The peptides were synthesized to create the vaccine, which was composed of four different peptide pools. Patients received all four peptide pools as injections in four different locations on the body. They received 5 priming doses within the first 3 weeks and 2 boosts within 24 weeks to these nonrotating sites.

Of 11 patients enrolled in the study, vaccine was prepared for 8, and 6 patients were ultimately vaccinated. These six patients experienced only minimal adverse events.

Immunogenicity Demonstrated

The analysis identified ex vivo interferon-gamma responses against the peptide pools, showing that T cells were being stimulated, and strong, polyfunctional ex vivo CD4-positive and CD8-positive T-cell responses against multiple neoepitope pools in all dosed patients. The neoepitope-specific CD4-positive and CD8-positive T-cell responses were largely restricted to mutated epitopes, confirming that the immunologic response was specific to the mutant and not the wild-type counterpart peptide. These steps showed that the vaccine strategy was, indeed, immunogenic, Dr. Ott explained. 

“What is exciting about this is that we see robust responses in all six patients,” he reported. “Typically most cancer vaccines induce responses only in subsets of patients and are not seen without stimulation in vitro. Some patients responded at lower levels but were still clearly responders.”

The researchers also confirmed endogenous recognition of the neoepitopes. “We wanted to know if the peptides can be processed and presented by T cells, which is critical for their function in patients, and we found that they could. We tested this with three independent approaches,” he said. “The T cells stimulated with the neoantigen vaccine could see the patients’ own melanoma cells, and a subset of the peptides was identified by high-resolution mass spectrometry on the patients’ own melanoma cells.”

The study shows, he concluded, that a personalized neoantigen vaccine is safe, feasible, and strongly immunogenic in cancer patients. Dr. Ott believes the data can reinvigorate the field of cancer vaccines and provide a rational path for building combinatorial approaches to extend their efficacy.

“It’s a very attractive vaccine approach. Many patients do not respond to checkpoint inhibitors given alone. We think an effective cancer vaccine is a promising way to drive T cells into the tumor and also enhance T-cell expansion, broadening their repertoire,” he said.

It is hoped that the T-cell response to any remaining malignant cells will prevent recurrences, but it is too early to determine that, he added. The next step is to combine this approach with a programmed cell death protein 1 (PD‑1) inhibitor, to try to optimize the effect of each strategy. What has been proven so far is immune reactivity. “We haven’t seen tumor shrinkage, because these patients don’t have tumors. Time will tell,” Dr. Ott commented. ■

Disclosure: Dr. Ott reported no potential conflicts of interest.

Reference

1. Ott PA, Hu Z, Keskin DB, et al: A personalized neoantigen vaccine in patients with high-risk melanoma. 2016 International Cancer Immunotherapy Conference. Abstract PR02. Presented September 25, 2016.