Making Personalized Cancer Vaccines

Filmed on March 16, 2017 in San Francisco
Part 2 of 3 from our Health x Design series.

Sign up for the newsletter

Amy Rappaport
Senior Scientist at Gritstone Oncology

Amy Rappaport is a Senior Scientist at Gritstone Oncology in Emeryville, CA, where she is working to advance innovative technology for personalized tumor vaccines. Prior to Gritstone, Amy developed diagnostic tests at Theranos, and as a postdoctoral researcher at Genentech, utilized mouse models to understand the progression of lung cancer. Amy earned her PhD in Biological Sciences at Cold Spring Harbor Laboratory in NY, where her research focused on identifying ways to target leukemia.


It's great to be here and have the opportunity to talk to you about the work I've been doing at Gritstone. Before I do, I just want to give you a little bit of background about me. I got my degree in biology at Cornell, and that was my first taste of research, where I was studying an invasive species of parakeet called the Monk Parakeet.

That was a lot of fun, but I wanted to do something more geared towards human health and disease. I decided to get my PhD at Cold Spring Harbor Laboratory in New York, in molecular biology and genetics. While I was there, I joined a lab focused in cancer biology, and did my thesis on understanding the genetics of leukemia, using mouse models.

Then I decided to move across the country to Genentech for my post-doctoral fellowship, where I continued using mouse models to understand the genetics, this time of lung cancer.

It was there that I learned two really important things. One, that I really like technology development, and two, that I wanted to do something that was actually making a product, and that would go then to patients.

I became a scientist at Theranos, where I got really valuable experience developing tests, and I was able to then take that test development experience, combine it with my background in cancer biology, and bring that to Gritstone Oncology, where I am now a scientist in vaccine technology. I've been here for about three months. So, at Gritsone, we are focused on turning tumor mutations into personalized cancer vaccines. Before I tell you what that means, I'll just put it into context for you.

Currently, a treatment for lung cancer, as an example, involves surgery to remove the tumor as well as chemotherapy. But the lung cancer five-year survival rate is 18 percent, and more than half of people die within one year.

There's another class of drugs called targeted therapeutics, and these can prolong survival, but they're only effective in a small subset of patients, and resistance is really common.

So there's a need for more effective therapies, and more options for cancer patients.

So what is cancer? Cancer is when mutations occur in normal cells that cause those cells to become abnormal and multiply. And as those cells multiply and grow, they gain more mutations, which make them more abnormal as they become malignant and invasive. No two people have identical tumors. So, what is a mutation? Your genes are made up of DNA, which is a code. That code gets turned into RNA molecules, which then become proteins. These proteins do all the functions within your cells. In cancer, there's a change in the code which leads to a mutant protein. The mutant proteins are on cancer cells, but they're not on normal cells. There's a huge number of mutations, and each individual cancer is different. That is why it's very difficult to develop a drug that can treat all of them.

At Gritstone, we're taking a different approach.

What we want to do is harness the body's own immune system to attack cancer cells.

Before I explain that, I'm first going to explain how the immune system works.

Imagine you have an infection with a flu virus. There is a certain type of immune cell called an antigen presenting cell, and it goes around your body, picking up little bits of foreign protein, like one that would come from a flu virus, and then uses that to teach a different type of immune cell called the T-cell. You can imagine the T-cell is like a drug-sniffing dog, and the antigen presenting cell is like a dog trainer, using this little bit of foreign protein to train the T-cell what to recognize. The T-cell then becomes activated, and goes around your body, finding the cells that have been infected with the flu, and kills them.

What we want to do is use the same system to teach the T-cells to recognize cancer mutant proteins, so that they can then go around your body, find the cancer cells, and kill them. Importantly, these will only kill cancer cells, and not normal cells. Unlike chemotherapy, for example, which is toxic to both, explaining why it has this devastating side effect. In order to do this, we're designing therapeutic vaccines that stimulate the immune system to attack the individual's tumor, based on their cancer mutant proteins.

They say therapeutic vaccines, to differentiate it from a prophylactic vaccine like measles, mumps, and flu, which you're familiar with, are preventative, but they work in a very similar manner to teach the T-cells to recognize foreign little bits of protein in the case of the prophylactic vaccines, or a little bit of mutant protein in the case of the cancer vaccines. These cancer vaccines are also personalized. For each patient, when they come in with a diagnosis, they'll have a biopsy, and a little bit of the tumor will be removed.

We will then sequence those tumors to identify cancer mutant proteins, and then synthesize a vaccine specific for that patient.

This has some clear challenges, the main one being identifying all the cancer mutant proteins in every single patient, and then out of the hundreds of mutations, identifying those most likely to induce an immune response, and then delivering those mutant proteins to immune cells.

The first line has actually become really easy in recent years, but only in the last 10 years or so, with improvements in DNA sequencing technology. The cost of DNA sequencing has really plummeted since about 2007.

It's very feasible to sequence every single patient and identify those mutations. Second problem, identifying those most likely to induce immune response, is still very challenging. At Gritstone we've put a lot of effort into developing a prediction algorithm based on functional data.

Then, the next problem, delivering those mutant proteins to immune cells, is what I've been working on, and my main focus at Gritstone. You could imagine, we could take these little pieces of foreign protein, and stick them straight into people. Some have done that, and it does work in a way. However, we think we can do it better. Again, harnessing the body's own machinery to our advantage. We're going to go one step backwards, and instead of injecting the proteins themselves, we're going to inject the RNA that encodes those proteins. We're going to deliver those in the cell's package into something called nanoparticles, which are simple little cells that we synthesized in a lab. These act as a mailman to deliver that package into the antigen presenting cells, which they can then teach the T-cells.

So, what do I actually do? I'm a scientist, I work in a lab, I wear a white lab coat, and I do experiments to test these vaccines. Here are some pictures of me in the lab with a colleague, doing an experiment to test these vaccines, and I will give you an overview of a recent experiment that we did, and how it worked to test this vaccine technology.

They start first by designing and synthesizing DNA that encodes our mutant proteins. We then can combine that with an enzyme called RNA polymerase, in a tube. And a couple hours later, we generate RNA that is an exact copy of that DNA. We can then package that into nanoparticles, and inject that into mice that have tumors in order to immunize them, and then measure their survival and their immune response. We measure their immune response by counting the number of those activated T-cells that we find in the mouse's body after vaccine vaccination.

Cancer vaccines are not a new idea. They've been around a long time, but it's only recently become feasible at all, and that's due to some really important breakthroughs in the last couple of years in understanding how cancer cells interact with immune cells. This has led to a huge new field called Cancer Immunotherapy. We're certainly not the only company who is trying to harness the immune system as treatment in various ways. In fact, there's at least two other small biotech companies in our building, that are also using the immune system to attack cancer, although with completely different approaches. It's a really exciting time to be in the field, and there's a real possibility that these different approaches can synergize with each other, as well as with other existing cancer therapies, to really boost the efficacy of cancer therapy. This could not just prolong survival a little, but have a big impact on patients.

In addition, we're also personalizing this cancer therapy, which in the past has been an extremely challenging thing to do.

However, because of many recent advances, in a large variety of different fields like amino therapy, but as well as DNA sequencing, bioinformatics, and also advances in manufacturing, this can be easier now.

What we need to do is rapidly manufacture vaccines for individual patients. That's really enabled us to move forward with this. Vice versa, some of the advances that we're making in bioinformatics and manufacturing could help lead to personalization in other types of cancer therapies, as well as other types of disease.

Hopefully this personalization will give us the ability to treat many more patients who would not respond to current cancer therapies. That's it. Thank you very much for your attention.