< Back to Articles

Treating Cancer With Your Own Personal Vaccine


View Web Version


Tumor mutations are potential vaccine targets because they are not found in healthy tissue. But every person’s cancer holds its own unique set of mutations, making a generalized vaccine impossible. Now a group at Johannes Gutenberg University in Germany have created an algorithm to prioritize the mutations that trigger the strongest immune response in mice. Based on this information, they made individual vaccines that caused the immune system to destroy even the most aggressive tumors.

“It’s exciting from the pharmaceutical perspective since this would be the first type of treatment in which a patient’s genome is analyzed and a drug is tailored exactly for use by an individual patient,” said Ugur Sahin, lead researcher on the work.

Previous work showed that the immune system can target cancer mutations, and such mutations can be used to generate vaccines, but it wasn’t clear how many mutations are suitable for immunotherapy or which ones worked the best.

Since cells present foreign peptides such as those from tumor mutations on their surface using MHC I, which is targeted and consumed by CD8+ cytotoxic T cells, the general assumption of tumor immunologists was that CD8+ T cells were most important for tumor removal. Sahin and his group sequenced the DNA of tumors in mouse models of melanoma, lung, and colon cancer and identified more than 500 mutations in each tumor. They chose to focus on 50 random mutations and 50 CD8+-focused T cell mutations and found that about 20% of the mutations triggered an immune response. They then evaluated whether these mutations were recognized by CD8+ or CD4+ T cells. “The surprising finding was, the most frequent immune response is mediated by CD4+ T cells,” said Sahin. MHCII on antigen-presenting cells present these mutations as neoepitopes, and CD4+ T cells recognize them and get rid of the tumor.

“The challenge for exploiting this discovery is that we would have to sequence the mutations of the patients, and we would have to have the tool to predict the right mutations for cancer immunotherapy,” said Sahin. The team developed a prioritization algorithm to choose which tumor mutations (based on good MHC class II-binding capacity and abundant expression of the mutation) would evoke the strongest response. They were able to create mRNA vaccines first using just 1 mutation, then at least 10, and showed that the T cells eliminated even established, growing tumors, causing the mice to live longer. This occurred despite the fact that the tumor microenvironment usually prevents the immune system from targeting tumors. This helps in the fight against resistance, since even if a cancer loses 1 or 2 mutations and metastasizes, the vaccine should still work since it typically targets about 10 mutations.

Sahin thinks that this approach could work in up to 75% of human cancers. The team has launched a clinical trial of 15–30 patients with melanoma and plans to begin trials for breast cancer as well. He hopes that the results will be published later this year. “We are convinced that this may become the future of cancer treatment,” said Sahin. ”We are not there yet, but at least in my eyes it’s only a technological challenge; it’s a challenge of doing and not a challenge of too many unknowns.”


1. Kreiter, S., et al. 2015. Mutant MHC class II epitopes drive therapeutic immune responses to cancer. Nature. doi:10.1038/nature1442. [Epub ahead of print]

< Back to Articles