Exploring the Challenges and Promise of Personalized Cancer Vaccines
June 14, 2024
By Beth Dougherty
In a small, nondescript conference room in Maryland, seats were hard to come by. Representatives from the US Food and Drug Administration (FDA) — experts in pharmacology, pharmaceutical manufacturing, biologics, you name it — filed in until the place was packed.
They were all waiting for Catherine Wu, MD, Dana-Farber chief of Stem Cell Transplantation and Cellular Therapies, and her small team. It was 2013 and the group would be proposing the first test of their personal NeoVax cancer vaccine in human patients.
To proceed, they needed an FDA green light.
"It was a completely new individualized approach, something the FDA had never seen before," says Wu. "I wish I had a picture to record us going in. It was a big moment. It was go time."
Wu, who earlier this year won the Sjöberg Prize for her pioneering research in the field of personalized vaccines to treat cancer, credits the vision for personal cancer vaccines to "a prepared mind." Her thoughts about making them stretches back to the beginnings of her medical and scientific career.
Neoantigens Spark the Personalized Vaccine Journey
Wu has a clinical background in stem cell transplantation. A challenge in that field is the stem cell donor's immune system's strong reaction to seeing unfamiliar flags, called antigens, on patient cells. The donor immune system is detecting molecular differences between the patient and the donor cells that make the patient's cells look like invaders. Wu's insight was that a patient's immune system can make that same kind of molecular differentiation, only between the patient's normal cells and cancer cells.
"It's not a big conceptual leap to recognize that mutations caused by cancer also have the potential to create new antigens that have never been seen before by a patient's immune system," says Wu.
Personal neoantigen cancer vaccines unleash T cells with superpowers: The ability to detect mutations specific to a patient’s cancer and to eliminate cancer cells with those mutations. Before vaccination, there are not a lot of these special neoantigen specific T cells and other T cells do not specifically target the cancer. After vaccination, T cells (blue) are able to recognize multiple cancer-specific mutations, in the form of neoantigens, enabling the immune system to more effectively attack the cancer.
For those who aren't steeped in immunology and cancer genetics, a mutated gene creates proteins that differ from those created by the unmutated gene. When cells break proteins down, they present pieces of them — antigens — on the cell surface for the immune system to inspect. A normal protein will present a familiar antigen that the patient's immune system will ignore. But a mutated protein will produce a new, unfamiliar antigen.
Those get the immune system's attention. In the world of cancer vaccines, these unfamiliar antigens are called neoantigens. They appear only on cancer cells, and not on normal cells, because the mutations are specific to the patient's cancer. This specificity makes neoantigens a tantalizing target.
Early on in Wu's career, understanding the power of cancer-specific antigens might have been easy, but predicting what those neoantigens would look like — and doing so for each individual person's cancer's unique set of mutations — would have been impossible. It wasn't until the advent of next-generation genomic sequencing, around 2008, when cancer-related gene mutations could even be identified rapidly and precisely.
Wu was there during the explosion of genomic and immunoproteomic data that happened at that time and the emergence of a new field of computational genomics that followed. An abundance of data made it possible to see patterns, and to use this data for generation of prediction tools for determining which candidate neoantigens would be optimal to target. Wu likens these data availability changes to the difference between using a crowdsourced GPS application to navigate.
"If you're in a signal poor area, the prediction is not very good," Wu says. "But in Boston and other urban areas, where there is a lot of data, predictions are very reliable."
With data in hand, computational biologists around the world started to create algorithms and use machine learning to make incredible predictions — such as what an antigen produced by a cancer-related mutation might look like and which neoantigens might trigger a strong response from the immune system. With the ability to make reliable predictions, it was possible for Wu to envision NeoVax, a personal neoantigen-based cancer vaccine — and to bring the idea to the FDA for approval.
It's an exciting time, especially in the context of other cancer immunotherapies. Right now, we're out of the gate with the early phase one studies of vaccines and we're making investments in future trials, armed with everything we've learned so far. We're making progress on so many fronts.
NeoVax Takes Flight: Early Trials Show Promise
Wu and her team presented their proposal to the packed FDA conference room in Maryland back in 2013. Their plan was to treat a small cohort of patients with advanced melanoma with NeoVax. At around the same time, in Europe, a team from a pharmaceutical company, BioNTech, was making a similar proposal for a personal neoantigen vaccine in melanoma, which was already known to respond well to immunotherapies.
"The good news is that they said 'yes,'" Wu says.
With a green light from the FDA, she and her colleague Patrick Ott, MD, PhD, clinical director of Melanoma Treatment Center, whom she had teamed up with early on, carried out that first trial and reported encouraging results in 2017.
"We had these really nice responses," says Ott. "But we needed to find an effective way to boost the immune system to produce even stronger ones."
One possibility was to combine NeoVax with other immunotherapies, such as immune checkpoint inhibitors. The rationale for that approach is that the immune checkpoint inhibitor will unmask the cancer, making it easier for the T cells primed by the vaccine to enter the tumor and attack.
"I think of it like in 'Star Wars,' when they bring down the shields of the Death Star, opening the opportunity to target it," Wu says. "Following this analogy, if you bring down the shield, you can then get the fighters — the T cells — in."
To test this idea, Wu, Ott, and clinical investigators at Dana-Farber have opened trials of NeoVax in combination with various immune checkpoint inhibitors in melanoma, glioblastoma, chronic lymphocytic leukemia (CLL), and kidney cancer.
The trial in kidney cancer combines NeoVax with a low dose of an immune checkpoint inhibitor after surgery in patients who are at a high risk of the cancer coming back. The team of investigators, led by Toni Choueiri, MD, director of the Lank Center for Genitourinary Oncology, will be reporting data from that study in the coming year.
Inhye Ahn, MD, a physician in the lymphoma program at Dana-Farber is leading another combination strategy in a trial of the vaccine for CLL. Ahn is testing the vaccine alone as well as in combination with two different immunotherapy combinations.
One combination adds an immune checkpoint inhibitor to the vaccine. The other adds an immunotherapy that dampens the portion of the immune system that prevents runaway immune reactions. "By removing the inhibition, it sort of creates an immunological space for the newly trained T cells to grow," says Ahn.
"It's an exciting time, especially in the context of other cancer immunotherapies," says Wu. "Right now, we're out of the gate with the early phase one studies of vaccines and we're making investments in future trials, armed with everything we've learned so far. We're making progress on so many fronts."
We're literally in a race against the tumor. We want to make the best product we can, which takes time, to optimize the ability of the patient's immune system to activate and respond as well as possible.
Race Against Time for Glioblastoma
In glioblastoma, the focus is not only on combinations, but also on timing, says David Reardon, MD, clinical director in the Center for Neuro-Oncology at Dana-Farber. Along with Wu and Ott, Reardon began testing NeoVax in glioblastoma in 2014.
An initial group of patients received the vaccine after surgery to remove the tumor. Based on early results, Reardon expanded the trial. New patients would receive an immune checkpoint inhibitor with the vaccine and be randomized to receive different therapy combinations on different schedules.
"It's not simply putting them together," says Reardon. "Our immune system responds very elegantly in an orchestrated series of events. The timing of the immune checkpoint inhibitor relative to the start of vaccination could make a big difference."
Reardon, with Wu and Ott, is doing a deep dive into the immune responses of patients who are receiving the medicines on different schedules. They hope to determine how the schedule affects not only the initial immune response but also the durability of that response.
For patients with glioblastoma, timing is everything. The average time to recurrence after surgery is six or seven months. It takes several months to create a personalized vaccine. For each patient, the tumor tissue must be sequenced and analyzed to identify mutations and make predictions. After that, a team of Dana-Farber experts gathers to hand-select the best neoantigens to include in that patient's vaccine. Then the vaccine must be manufactured and tested for quality. Once the vaccine is given to the patient, it takes at least one to two months for the immune system to begin mounting an immune response.
"We're literally in a race against the tumor," says Reardon. "We want to make the best product we can, which takes time, to optimize the ability of the patient's immune system to activate and respond as well as possible."
Visualizing Neoantigens
A mutated gene creates mutant proteins with portions that differ from the normal protein. When proteins are degraded, fragments are presented on the cell surface for the immune system to inspect. A normal protein will present a familiar antigen that the immune system will ignore. But a mutant protein will produce a new, unfamiliar antigen called a neoantigen (orange). Cancer vaccines help T cells target and destroy cells presenting neoantigens.
NeoVax and Ovarian Cancer: A New Frontier
So far, most of the personal neoantigen vaccine efforts have focused on cancers that respond well to immunotherapy and have high numbers of cancer-related mutations — specifically mutations caused by misspellings of genes.
Ovarian cancer is an outlier. It doesn't respond well to immunotherapy in general and it is driven by abnormalities in the genome that are more structural — two genes fused together or rearranged — rather than by misspellings.
To make the cancers more responsive to immunotherapy, Panos Konstantinopoulos, MD, PhD, director of translational research in Gynecologic Oncology at Dana-Farber, combined NeoVax with nivolumab in a trial for high-grade serous ovarian cancer, a form of cancer with a strong need for medicines to stave off recurrence. Ten patients have enrolled in the trial and early unpublished data hints that the vaccine is triggering immune activity.
"We want to make improvements, so we're working on the next generation vaccine," says Konstantinopoulos.
One advancement is related to the selection of neoantigens. Wu's lab is working on expanding the capabilities of the neoantigen prediction algorithms to recognize more antigens, such as those produced by structural changes in the genome.
"Our projects are hitting on the biggest challenges in vaccinology," says Wu. "What is the antigen? What are you going to combine it with? How are you going to deliver it?"
Exploring New Delivery Systems
Like combination strategy and timing, vaccine delivery presents another avenue to improve the potency of the vaccines. The NeoVax vaccine is made of peptides, tiny protein fragments that mimic antigens and train the immune system to detect them. That training process is spurred on by the inclusion of another chemical called poly-ICLC in the vaccine formulation.
The addition of this chemical, called an adjuvant, is like tossing back a shot of espresso before cramming for an exam. It primes the immune system for learning. Dana-Farber researchers have several projects underway to learn more about how different adjuvants and delivery mechanisms improve immune system training and help it remember what it learned much longer.
For instance, Ott, in a trial combining NeoVax with two immune checkpoint inhibitors, ipilimumab and nivolumab, for melanoma, is also adding a drug called Montanide as a potential next-generation adjuvant.
Yet another mechanism, devised by Dana-Farber researcher William Shih, PhD, is using a nanoparticle called DoriVac to deliver vaccines. The nanoparticle is made of "DNA origami," so called because the DNA has been folded into a rectangular shape. Shih has found that the number and placement of adjuvant chemicals on one end of the DNA origami block influences the immune system's response to the antigens attached at the other end.
Other delivery mechanisms are also in the works. Moderna and Merck, and BioNTech and Genentech, for instance, have created personal neoantigen vaccines that use mRNA, rather than peptides, to introduce the neoantigens to the immune system. These companies are testing their vaccines in several cancers and in large registration trials.
"We are making so many investments in the future, so many at one time, that it can be hard sometimes to measure what's going on," Wu says. "But we as a field are learning so much and there are many more studies along the way. My hope is that we have the tools we need — the combinations, the timing, the algorithmic predictions, and the delivery mechanisms — to find the way forward."
