If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Will cancer vaccines change the way we treat cancer?

Experts weigh in on whether candidates in the clinic will live up to mounting hype

by Gina Vitale
June 25, 2023 | A version of this story appeared in Volume 101, Issue 20
An illustration showing a syringe exit a cancer cell as if a missile went through the cancer cell making it explode.

Credit: Mark Smith


In brief

Imagine a world where a vaccine could treat cancer. The concept isn’t new—researchers have been working for decades to formulate vaccines that will spur a person’s immune system to fight their cancers. It’s been a long and largely unsuccessful road. But recent advances in technology have facilitated a new wave of vaccine candidates, and some are from newly-minted household names like Moderna and BioNTech. Clinical trials, though still early, are yielding promising results. Will this new wave of vaccines change how clinicians treat cancer?

Humans have been trying to outsmart cancer for eons. Scientists have tried to kill it head-on with radiation or chemotherapy. They’ve engineered treatments like checkpoint inhibitors to give the body’s own defenses a fighting chance, carved out cancer cells with precision tools, and much more.

But none of these treatments are perfect. Chemotherapy and radiation can kill more than just cancer. Checkpoint inhibitors aren’t effective in certain tumors. Cancer that’s removed comes back.

What if there were another way? Every year, doctors deliver vaccines to ward off disease. A small syringe introduces a harmless segment of a pathogen to the body, and the body learns to fight that pathogen. What if, through a poke of the needle, someone could similarly spur their immune system to fight cancer they already have?

A very effective cancer-treating vaccine that targets a specific antigen—a substance that provokes the body’s immune response—doesn’t exist yet. But with versatile tools like messenger RNA (mRNA), it’s not the moon shot it once was. Vaccine candidates are speeding through the clinic. They aren’t perfect—they need to be combined with other agents to be effective, and big questions remain about whether they’ll work well for multiple types of cancer. But as positive early results roll out from clinical trials by COVID-19-minted household names like Moderna and BioNTech, excitement is mounting. If the therapies succeed in the proving ground of the clinic, vaccines may be a paradigm shift in the way we treat cancer.

What is a cancer vaccine?

A vaccine, defined broadly, is a substance that spurs the body’s immune system against a particular pathogen or disease. When most people think of vaccines, they probably picture something like the COVID-19 shot—a jab intended to prevent illness rather than treat it. There are a few of those for cancer already.

Two- or three-dose injections of Gardasil 9 prevent human papillomavirus (HPV), a virus that causes most cervical cancer cases and can also cause other types, like anal and penile cancer. The hepatitis B virus (HBV) vaccine works similarly, warding off hepatitis B, which can lead to liver cancer. Rather than fend off cancer directly, these vaccines thwart viruses that may cause cancer.

But generally, the term cancer vaccines refers to vaccines that treat cancer. A big use case is adjuvant therapies—doctors would give them to someone after they have already received their main treatment, most likely surgery. The vaccines would then help prevent the cancer from returning.

Several connected, fuzzy-looking blobs colored various shades of orange.
Credit: Anne Weston/EM STP/Francis Crick Institute/Science Source
Pancreatic cancer cells

“The concept is to get the immune system to reject the cancer, the way it could reject a kidney transplant or a lung transplant,” says Jay A. Berzofsky, chief of the Vaccine Branch of the US National Cancer Institute (NCI).

Only one therapeutic cancer vaccine targeting a specific antigen has been approved by the US Food and Drug Administration. Provenge treats prostate cancer, but it doesn’t have a clear mechanism of action, and while it boosts overall survival, it doesn’t improve progression-free survival, says Don J. Diamond, a professor of hematology and hematopoietic cell transplantation at the City of Hope National Medical Center and Beckman Research Institute. The lack of approvals reflects the difficulties that have dogged the field.

“I think it’s just because it’s been quite a tricky area to research,” says Anna Osborne, a senior manager at Citeline, a pharmaceutical information services firm. There’s been a lot of activity and a lot of clinical trials, many of which have failed, she says. “So it’s just been quite a slow area to grow.”

That’s starting to change with some positive trial data. In April, Moderna and Merck & Co. released the first detailed data from a Phase 2b clinical trial for a cancer vaccine in people with high-risk stage 3/4 melanoma who had their tumors removed. And in May, BioNTech published results from a small Phase 1 trial of a vaccine collaboration with Roche’s Genentech that targets pancreatic ductal adenocarcinoma, the most common type of pancreatic cancer (Nature 2023, DOI: 10.1038/s41586-023-06063-y). Participants in that trial had also had their cancers removed.

Unsurprisingly for the mRNA leaders Moderna and BioNTech, their cancer vaccines are based on mRNA. In some ways, the vaccines are quite similar to the COVID-19 jab. But instead of a string of mRNA encoding for the SARS-CoV-2 spike protein, these cancer vaccines hold mRNA that encodes for proteins found on the vaccine recipient’s cancer cells. Specifically, they’re neoantigens—proteins that are made because of mutations in the cancer cells’ DNA. The vaccine stimulates immune cells like T cells to attack the neoantigens and therefore the cancer.

These neoantigens are a logical target because they’re unique to tumor cells. T cells don’t instinctively attack proteins that are shared between normal and cancer cells because T cells don’t see those proteins as foreign. Neoantigens are caused by tumor mutations, so they are blatantly non-self.

“Mutations are the grist by which your T-cell repertoire can do an effective job,” Diamond says.

Moreover, these two mRNA vaccines are personalized to each patient. Formulators aren’t just choosing neoantigens that they expect to be present on many people’s cancers. Rather, they sequence a person’s own tumor cells and determine which neoantigens are likely to be present. They also predict which ones are most likely to induce an immune response.

The BioNTech-Roche vaccine contains two strands of mRNA, each encoding up to 10 neoantigens, for a maximum 20. The Moderna-Merck vaccine encodes up to 34 neoantigens in a single strand of mRNA.

“We think having multiple shots on goal, if you will, is important,” says Eric Rubin, senior vice president of global clinical development at Merck. “Because despite our improvements in the field in predicting neoantigens, it’s not perfect. And so we think this enables the likelihood that at least one or possibly more of those [neoantigens] will be effective in enabling the T cells to destroy the cancer.”

In the Phase 2b trial, Moderna and Merck’s jab, used in combination with the checkpoint inhibitor Keytruda, reduced the risk of cancer recurrence or death by 44% over a year compared with using Keytruda on its own. It also cut the risk of death or spread of the cancer from its original location to distant parts of the body by 65%, the companies announced in June. The trial included 157 people, 107 of whom received the vaccine.

In BioNTech and Roche’s trial, 8 of the 16 vaccine recipients developed a substantial T-cell response specific to a neoantigen or multiple neoantigens in the vaccine. And at an 18-month median follow-up, people that had the boosted T-cell response had a longer median recurrence-free survival than the people that didn’t have that boosted response.

It’s not personal

Personalized cancer vaccines, as evidenced by the early positive results of the Moderna-Merck and BioNTech-Roche candidates, are on the rise. But vaccines tailored to individuals are not the only options.

“We can make a business out of personalized cancer vaccines—a very successful business, assuming everything works,” says Andrew Allen, cofounder, president, and CEO of Gritstone Bio, which develops immunotherapies for cancer and infectious diseases. “However, it’s always going to take you a bunch of time, and it’s always going to be a bit more expensive. So off the shelf is desirable.”

An off-the-shelf vaccine is what it sounds like—something that can be taken off the shelf and administered with no personalization required. To be effective, an off-the-shelf vaccine must have a target that is common to many recipients but still distinct from self. That’s a challenge in oncology, as everyone’s cancers are different.

Two electric-blue, fuzzy balls in the act of separating from each other. Long, thin tendrils extend out from them.
Credit: Anne Weston/Science Source
Cancer cells dividing

Nora Disis, director of the Cancer Vaccine Institute at the University of Washington School of Medicine, is focused on treating cancers that don’t typically have a lot of mutations, such as breast cancer. Such cancers are less suited to treatment with the neoantigen approach. So for her cancer vaccine efforts, Disis is targeting proteins that drive the cancer to grow rather than neoantigens specific to each person.

The advantage to targeting cancer-driving proteins, the NCI’s Berzofsky says, is that they are essential to the tumor’s survival. “If you can target something that tumor cells can’t do without, then it’s much more effective than if you target something that the cancer cells can escape from by just getting rid of that antigen,” he says.

Many shared tumor antigens, including cancer-driving proteins, are present in much higher amounts in tumor cells than in normal cells, but since they are also expressed in normal cells, it’s tricky to get the body to attack them. “They’re not seen as foreign by the immune system, whereas a mutation is seen as foreign. It’s never been shown to the immune system before,” Disis says.

Disis and her collaborators have tested a DNA-based shot that encodes a single antigen based on part of a driver protein that is upregulated in breast cancer cells. They have seen good results in people with advanced-stage breast cancer, she says. Her team is also working on versions with multiple antigens. “The vaccines generate similar magnitude of immunity as the mutated antigen vaccine,” she says. “I think it’s a promising approach for patients with low mutational tumors.”

Haval Shirwan, an immunologist at the University of Missouri School of Medicine, doesn’t think vaccines based on nonmutated proteins are promising on their own. But they could be when used in combination with other approaches, like including neoantigens specific to an individual’s tumor in the vaccine formulation.

Another nonpersonalized vaccine strategy is to target genes that, when mutated, cause or contribute to many cancers, Shirwan says. Different people with the same tumor types can have these mutations in common. Since these genes are well studied, researchers could try to use the genes’ protein products as antigens and immunize the body against them, Shirwan says.

Mutations are the grist by which your T-cell repertoire can do an effective job.
Don J. Diamond, professor of hematology and hematopoietic cell transplantation, City of Hope National Medical Center and Beckman Research Institute

Gritstone is pursuing both individualized and off-the-shelf vaccine programs. The firm’s personalized vaccine, Granite, relies on individual neoantigens. For its off-the-shelf vaccine program, called Slate, the firm is targeting neoantigens that are shared across some people.

Allen predicts that in 20 years, or maybe even 10, people in the field will be able to develop off-the-shelf products that can produce as robust an immune response as the personalized ones.

RNA renaissance

RNA is a popular medium for cancer vaccines, as demonstrated by companies like Moderna, Merck, BioNTech, and Roche. That’s not a coincidence. RNA offers several advantages—a significant one is speed, according to Merck’s Rubin.

“You can go from biopsying the patient’s tumor to developing an individualized neoantigen therapy within a few weeks,” Rubin says. “And so compared to other ways to generate vaccines, this is much faster.”

A blue strand with little spikes protruding from it, looping around itself.
Credit: Shutterstock
Messenger RNA is used to encode for neoantigens in both Moderna and Merck's and BioNTech and Roche's cancer vaccine candidates.

RNA is also cost effective to produce. And it’s benefiting from the halo of two successful COVID-19 vaccines, which skyrocketed mRNA vaccines into the mainstream. Michael Super, a senior staff scientist at the Wyss Institute for Biologically Inspired Engineering, recalls that when the pandemic was starting, the fastest vaccine development he could think of took 5 years.

“I was absolutely gobsmacked when they were approved within a year,” he says of the mRNA vaccines for COVID-19. “I suppose it took a pandemic to kick the status quo into a different gear. And I think the silver lining in this horrendous pandemic has been the fact that mRNA vaccines and others now are fast-tracking in cancer and other diseases.”


It’s not just one type of RNA that’s getting the attention. Gritstone, as part of its personalized vaccine program, uses self-amplifying messenger RNA (saRNA), or RNA that makes copies of itself. One benefit, Allen says, is that when RNA replicates itself in the cell, the result is double-stranded RNA.

“Your mammalian cells typically think, ‘Aha, there is a virus in the vicinity,’ because it’s viruses that usually create double-stranded RNA,” he says. “And so in principle, it’s that extra degree of foreignness.”

Also, because it xeroxes itself, saRNA can be delivered in lower doses than mRNA, says Steve Reed, CEO and cofounder of HDT Bio, which is also developing cancer vaccines using saRNA. HDT’s vaccine technology, authorized for emergency use in India for COVID-19, is the first saRNA-based vaccine platform in the world authorized for humans, according to the company. The vaccine platform uses 10 µg of saRNA, Reed says; that’s less than the current mRNA COVID-19 vaccines, which use 30–100 µg.

I think we will see vaccines approved for use in the treatment of cancer within the next 5 years. No question.
Nora Disis, director of the Cancer Vaccine Institute, University of Washington School of Medicine

In Gritstone’s program, neoantigens are initially delivered using an adenoviral vector. Then, patients get a booster, this time using saRNA to encode for those neoantigens. “Same antigens, different vector, which has been shown to give you an even stronger T-cell response,” Allen says.

“At some level, of course, it doesn’t really matter which is better or worse,” Allen says. “All that matters for us is, does it work?”

Checkpoint cleared

Even before the data drops of the past few months, there was a lot of hype surrounding cancer vaccines. But these new treatments generally don’t work in a vacuum. The way Disis defines them, they don’t count as a cancer vaccine unless they’re combined with, or include, an agent that will help provoke an immune response against the antigen or antigens.

One such agent is a checkpoint inhibitor. Some cells in the body contain checkpoints that turn T cells off to prevent the immune system from going rogue and attacking healthy cells. But cancer learns to exploit those checks and turn off T cells before they can attack cancer cells. A checkpoint inhibitor is a drug, usually a monoclonal antibody, that modifies the behavior of an immune system’s checks and balances. Inhibiting the checkpoint prevents cancer from putting the brakes on T cells.

On the top half, a light-blue antibody binds to a dark-blue protein on the surface of a cell. On the bottom half, a yellow antibody binds to a red protein on the surface of a cell.
Credit: Juan Gaertner/Science Source
Checkpoint inhibitors (yellow, light blue) block an interaction in which cancer cells shut down T cells. Checkpoint inhibitors bind to either a protein (red) on the cancer cell or a protein (dark blue) on the T cell to prevent the two proteins from connecting.

Decades ago, before these immune suppression tactics were discovered and checkpoint inhibitors were developed, vaccine attempts were hampered, according to Berzofsky. “There was some discouragement that you could make a vaccine but it couldn’t kill the cancer,” he says. “The big breakthrough was the discovery of these checkpoint inhibitors.” And, he notes, checkpoint inhibitors are “just the tip of the iceberg”—there are many other immunosuppressive mechanisms that are still ripe to be blocked.

Importantly, checkpoint inhibitors and cancer vaccines have somewhat of a symbiotic relationship, Berzofsky says. “The vaccines alone won’t work well if you don’t block the immunosuppression from the tumor,” he says. And in tumors that don’t stimulate a strong immune response to begin with, called cold tumors, agents such as checkpoint inhibitors aren’t effective unless you can spark an immune response. “And you do that with the vaccine.”

Both the BioNTech–Roche and Moderna–Merck teams have taken advantage of the synergy between checkpoint inhibitors and cancer vaccines. BioNTech and Roche’s vaccine is used in combination with the checkpoint inhibitor Tecentriq, made by Roche’s Genentech (15 out of 16 vaccine recipients in the recent pancreatic cancer trial also received chemotherapy). Moderna and Merck’s vaccine is used in concert with Merck’s Keytruda.

“In cancer patients, you do have T cells which can recognize the cancer cells, but they’re inactive until you give Keytruda,” Merck’s Rubin says. “Keytruda takes the brakes off, and now you can have a very robust anticancer response.”

Keytruda’s been a blockbuster drug for Merck, pulling in almost $21 billion in 2022 alone. Rubin says he was lucky enough to have been at Merck in Keytruda’s early days: he led the Phase 1 trial that led to the early initial approvals in melanoma and lung cancer. The cancer vaccine being developed in collaboration with Moderna reminds him of that time. “This has that same feeling of being really a paradigm shift that has the potential to help patients with multiple different cancer types.”

Hot and cold

Along with needing to be used with an immune-stimulating agent, cancer vaccines come with another caveat. The success of a cancer vaccine will depend on the immunogenicity, or ability to stimulate an immune response, of the cancer it’s treating.

Moderna and Merck’s clinical trial results were from tests in melanoma. Melanoma is generally a hot tumor—meaning it spurs a strong response from immune cells to come fight it. It makes sense that cancer vaccines would work better with hot tumors because they help stimulate and strengthen an already-robust T-cell response.

“Melanoma . . . is a rare example of the hottest of hot tumors that responds very well to a whole plethora of immunotherapies,” the City of Hope’s Diamond says. “It was a natural place to start, but it is somewhat of an outlier.”

Tendrils extend from pink spiky balls (T cells) onto larger, blue spiky balls (cancer cells).
Credit: Maurizio De Angelis/Science Source
Cancer vaccines spur T cells (white) to attack cancer cells, like the one shown here (orange).

Still, Merck’s Rubin thinks the Phase 2b trial data show that the vaccine-Keytruda combo will have potential beyond melanoma. He notes that Merck has moved Keytruda into earlier lines of therapy for several types of cancer. “The underlying biology of why we would combine a neoantigen vaccine therapy with Keytruda, I think likely holds for multiple cancer types,” he says.

A cold tumor is one that does not induce a strong T-cell swarm. Even checkpoint inhibitors aren’t as effective with cold tumors, because while the inhibitors enable T cells to keep fighting, there aren’t as many T cells present for them to help. Whether cancer vaccines are able to stimulate a sufficient T-cell response when there isn’t much of one to start is yet to be proved, although Diamond says the neoantigen vaccines are working at it.

BioNTech’s effort is offering some hope for treating pancreatic cancer, which is typically cold. Its vaccine saw a 50% response rate—the numbers are striking, Diamond says, but with caveats.

“I think that as opposed to anything else that has been proffered [for pancreatic cancer] and then hopes dashed, there’s greater possibilities here. The science behind it is stronger,” he says. “It’s just a question of whether the durability will be enough that it actually will prove to have lasting benefit. But there is something in between now and lasting benefit that would still be a benefit.”

The NCI’s Berzofsky thinks that cancer vaccines will show positive results against cold tumors when used in combination with checkpoint inhibitors and other agents. “That’s the whole point of the vaccine,” he says. “In principle, a vaccine might be more important in a cold tumor than a hot tumor, because the hot tumor already has the immune response.”

Why now?

It’s been a long road for cancer vaccines. Decades were largely marked by failure. But the science has progressed—enough, it seems, to facilitate some measure of success.

According to Gritstone’s Allen, two things are enabling this new and rapid wave of development: researchers have figured out which antigens to use and have better vectors for getting them into the body. “I think it’s the combination that’s the magic,” he says.

In my opinion, yes, it’s hype. And yet, it’s the best hope we’ve got.
Michael Super, senior staff scientist, Wyss Institute for Biologically Inspired Engineering

The technology, too, has changed since scientists first started attempting cancer vaccines. Berzofsky describes having made a promising cancer vaccine in the 1990s but says it was too cumbersome and expensive to bring to market. Sequencing the target genes and synthesizing a vaccine were slow processes. “By the time you did all that, a month or two would go by and then you’d have to hope that the patient was still alive, and well enough to get vaccinated,” he says.

COVID-19 vaccines also paved the way. Moderna and BioNTech’s mRNA-based COVID-19 vaccines showed that RNA could make for a safe and effective modality. “I really think the field has changed dramatically because of COVID, which will then help the cancer vaccines tremendously,” the Wyss Institute’s Super says.

Even with the new knowledge and advancements in technology, cancer vaccines have a lot left to prove. They’ll need to make it through Phase 3 trials, and studies will have to show how they can square up against cold tumors.


The vaccines will also need to be able to adapt. Cancers change themselves to evade the immune system, and cancer treatments need to keep up—just like vaccines against seasonal viral infections. Fortunately, once the cancer vaccine platforms are established, they may be well suited to evolve in such a way. “The gorgeous thing about the mRNA vaccines is you just change the mRNA,” Super says. “And you can keep the rest of it the same.”

Ultimately, cancer vaccines could be a big tool in the oncology toolbox. More data are needed to see exactly how useful they could be, and for how many people. But the University of Washington’s Disis is now optimistic that it’s only a matter of time before the vaccines make their way to patients.

“I think we will see vaccines approved for use in the treatment of cancer within the next 5 years,” Disis says. “No question.” She also anticipates that eventually, “there will be a vaccine for almost any type of patient.”

“In my opinion, yes, it’s hype,” Super says. “And yet, it’s the best hope we’ve got.”


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.