A new approach to treating cancer is to use each patient’s natural immune system to attack their tumor. This approach takes advantage of the fact that cancer is caused by mutations that occur only in tumor cells. We now know that these mutations allow the immune system to see a tumor as a “foreign” invader, almost like a viral infection. This knowledge has led to the idea that we could design cancer vaccines. Each vaccine would be unique to each patient and train their immune system to attack their unique tumor. The vaccine treatment involves injecting small amounts of harmless pieces of tumor protein into the arm of each patient. Cancer vaccines are promising because they have few side effects compared to other cancer treatments. If cancer vaccines are to be a success, we need to become good at finding the tumor mutations that are best for training each patient’s immune system. So far, early attempts to find good mutations have focused on the simplest and smallest forms of mutations. In some patients, we do not find the right mutations to create a vaccine. In our study we will explore a type of larger and more complex mutation that causes incorrect assembly of proteins in tumor cells. These provide more options for vaccine design. Finding such mutations should lead to better cancer vaccines. Our study should also allow us to design vaccines for more patients and help us to understand what makes a good cancer vaccine.