When asked about pancreatic cancer, physicians roll out a grim parade of statistics. Only around one in ten people survive for more than five years after diagnosis. Average survival is just four months. “It is amongst the deadliest of cancers,” says Marc Besselink, a pancreatic surgeon at the Amsterdam University Medical Center.

In the Netherlands, he says, 65% of people with pancreatic cancer do not receive treatment to target the tumour. This is not through lack of action, but because the disease is usually too advanced by the time people are diagnosed. Vague symptoms, such as back pain, unexplained weight loss and loss of appetite, mean that more than half of people will already have metastatic disease — in which the cancer has spread to other areas of the body, such as the liver and lungs — when diagnosed. Surgery is an option only if the cancer has not spread. People without metastatic disease can also receive a potent cocktail of chemotherapies. But these interventions are life-saving for only a very small minority of people.

Nature Outlook: Pancreatic cancer

Scientists are looking for better options, including cancer vaccines. Advances over the past few years have buoyed cancer biologists with enough hope that some suggest that a cure could be possible. The objective with therapeutic vaccines — which have shown promise for prostate and skin cancer — is to train a person’s immune system, after surgery and chemotherapy, to seek and destroy pancreatic cancer cells, no matter where they are. Immune memory then guards against a tumour returning.

Two strategies are at the vanguard. Both tutor the immune system to attack proteins that are more plentiful on the surface of pancreatic cancer cells than they are on other cell types. One targets changes to proteins that are specific to a person’s cancer to create a tailored vaccine, the other takes aim at subtle alterations in a protein that are hallmarks of pancreatic cancers.

People are already receiving these vaccines in early clinical trials, and companies hope to eventually roll out vaccines to hospitals. Pancreatic cancer has a reputation as a bruising opponent for new treatments, and there are many hurdles to overcome. But therapeutic vaccines would be “a game changer if they work for pancreatic cancer”, says Shivan Sivakumar, an oncologist at the University of Birmingham, UK.

A whole-body approach

The low rate of success with standard forms of treatment is only part of the reason why scientists are developing therapies. Many routine procedures have serious downsides.

Surgery, for instance, is tricky, given the pancreas’ location deep in the body and delicate nature. “You think you have this curative surgery but then one year later, half the patients are dead,” says immunologist Stephanie Dougan at Dana-Farber Cancer Institute in Boston, Massachusetts. And chemotherapy can only be used for so long — the toxicity of these drugs is not limited to cancerous cells and the resulting side effects can be difficult for people to tolerate.

The disease’s aggressive nature and tendency to spread early means that a whole-body approach that harnesses the immune system might make the most sense. Immunotherapy has already been used to successfully treat other cancers — greatly improving survival in metastatic melanoma, for example. But pancreatic cancer has proved less amenable to immune-targeted therapies. Some cancers such as melanoma have a large number of DNA mutations and therefore already attract immune cells to the tumour. Pancreatic cancers have fewer mutations and are often overlooked by patrolling immune cells. A moat of non-cancerous cells, scar tissue and blood vessels around the tumour can also impede drugs and radiotherapy.

Support for new treatments has, therefore, been in short supply. Besselink experienced this in his pleas to pharmaceutical firms to test their cancer drugs in individuals with pancreatic cancer. “We said we would pay for everything. Just give us the drug for free. They still wouldn’t do it,” he says, possibly because treatments for other cancers are more lucrative.

Vinod Balachandran at Memorial Sloan Kettering Cancer Center in New York did successfully pitch an idea to industry. His group had been grappling with the puzzle of why some people survive five or more years after a pancreatic cancer diagnosis. The researchers found that, in this small group of people, immune cells seem to spontaneously recognize the cancer. There were more T cells in the tumours of these people compared with in people who did not survive the cancer, and clones in the blood could be detected up to a decade later1. They also found that the T cells recognized proteins that were specific to each person’s tumour and had only a few alterations.

In a clinical trial2, the researchers applied these findings to people with a new diagnosis of pancreatic cancer and created personalized vaccines. Working with biotechnology firms Genentech and BioNTech, the team first sequenced both healthy and cancerous tissue that was removed from participants during surgery. By comparing the two, they were able to identify mutant proteins that were unique to each person’s cancer. Up to 20 of these neoantigens were then combined and delivered to people in the form of a messenger RNA vaccine.

The first person was given their personalized vaccine in December 2019, roughly 9 weeks after surgery to remove the tumour. They were also given chemotherapy to target surviving cancer cells, as well as a therapy known as a checkpoint inhibitor, which helps to release the ‘brakes’ that some tumours place on T cells. “They get the works,” says Drew Pardoll, an oncologist at Johns Hopkins Medicine in Baltimore, Maryland.

As predicted by lab studies, much of the vaccine’s neoantigen payload ended up in lymph nodes. This is an ideal place for them to be found by the immune system’s dendritic cells, which pick up and display suspicious proteins to T cells to train them to attack the cells that bear them. Initial results showed that the vaccine was safe and triggered an immune response in half of the 16 people vaccinated. Six of the people who responded to the vaccine did not have any disease relapse after three years. “That might not sound like a lot, but in my practice, I can assure you that after three years around 90–95% of my patients will have relapsed,” says Sivakumar.

When the researchers looked at blood samples from the eight participants who responded, they found T cells specific to the person’s tumour circulating in substantial quantities up to four years later3. Much of the immune response consisted of CD8+ cells, also known as killer T cells, which can destroy cancer cells. This was encouraging, especially because participants had received chemotherapy after the vaccine, which can sometimes cool immune responses.

Some researchers strike a note of caution, however. “Roughly half of the patients didn’t mount a response to any of the antigens” that the researchers could measure, says Pardoll. And even for the eight who did respond to the treatment, the study “doesn’t prove that the vaccine was responsible for the long relapse-free survival”. Proving cause and effect will require a larger trial in which individuals undergo surgery and then receive either standard care, or standard care and the vaccine.

Sivakumar is now running an arm of the mRNA vaccine trial in the United Kingdom sponsored by Genentech and in collaboration with BioNTech. If such a vaccine proves protective, the plan would be to deploy it after surgery to stop the return of the cancer. “This could be paradigm shifting,” says Sivakumar.

Hitting the driver

Bespoke cocktails of neoantigens are not the only option for creating pancreatic cancer vaccines. The other promising approach involves targeting changes to a specific protein called KRAS. Mutations to the gene that encodes it affect almost everyone with the disease.

Usually, KRAS acts like a switch to control cell growth and proliferation. But small changes to the protein can lock it into a permanently activated state that leaves cells free to divide unchecked and develop into tumours. Such ‘driver’ mutations occur in up to one-quarter of all human cancers, but in pancreatic cancers that figure rises to a staggering 93%. “Cancer cells can’t survive without it,” explains Neeha Zaidi, an oncologist at Johns Hopkins Medicine. She is developing vaccines that contain key peptides of mutant KRAS proteins that are common to most pancreatic cancer cells. In theory, KRAS-based vaccines could be mass produced — there should be no need for individualization.

Hopes of zeroing in on this protein were buoyed a decade ago by the case of a 50-year-old woman with metastatic colorectal cancer. She had received surgery, radiation and chemotherapy, but seven tumour deposits in her lungs resisted all attempts at treatment. Tissue removed from theses tumours contained T cells that specifically recognized a mutant form of KRAS. This protein was altered at a single amino acid — a seemingly minor change, but one that is seen in 13% of people with colorectal cancer and 45% of people with pancreatic cancer. When the woman’s T cells were grown in a lab and reinfused in greater quantities, six of the tumour deposits disappeared — the final one was removed surgically4. Almost everybody in her situation not receiving this treatment would have been dead in five years, says Steven Rosenberg, a surgeon at the National Cancer Institute in Bethesda, Maryland, who was in charge of her care. “She remains continuously disease- free now ten years later.”

A few companies are actively pursuing therapeutics that target driver mutations. Elicio Therapeutics in Boston, Massachusetts, has two vaccines in development that target KRAS. The first contains two mutant KRAS peptides. A key part of this peptide vaccine is that the drug has a tail containing a water-loving and a fat-loving part. The water-loving portion allows it to move easily through tissue, and the fat-loving part allows it to latch onto the protein albumin at the injection site and hitch a ride to lymph nodes.

Elicio has completed a phase I safety trial in people with pancreatic or colorectal cancer who had already undergone surgery and chemotherapy. The vaccine candidate seemed to trigger T cells against KRAS in 21 of 25 individuals5. “That you get T-cell responses that are good enough against these driver mutations in the vast majority of vaccinated patients is really exciting,” says Dougan, who is involved in a separate collaboration with Elicio on T cells. Earlier this year, Elicio received the support of the US Food and Drug Administration to begin a larger trial to test the potential vaccine’s efficacy. Results of a phase II trial are expected by early 2026.

Elicio’s second KRAS-targeting vaccine formulation contains seven of the most common KRAS peptides. These are present in 88% of people with pancreatic ductal adenocarcinoma, the most common type of pancreatic cancer. The aim is to produce an off-the-shelf vaccine that will target cancer in as many people as possible.

Zaidi and her colleagues are also testing a KRAS-based vaccine targeting six peptides6. Like Elicio’s vaccine, Zaidi’s candidate provokes a response not only from killer T cells, but also from CD4+ ‘helper’ T cells. Although killer T cells are crucial, Zaidi and her colleagues found that helper T cells boost the effectiveness of personalized vaccines targeting neoantigens. This is also supported by reports of responses to tumour targets. “We need CD4+ T cells to help get better CD8+ T cells,” says Zaidi.

Although KRAS is ubiquitous and has an early role in pancreatic cancer, “we haven’t been able to target KRAS particularly well, despite having known about it for 40 years”, says Dougan. But recent progress is reason for hope, she says.

Packing a protective punch

Most research has focused on administering vaccines after surgery, but some scientists are setting their sights on delivering vaccines before a tumour establishes itself in the pancreas and builds a protective barrier. Physicians regularly perform ultrasounds on people with symptoms such as stomach pain, weight loss or nausea and find jelly-like cysts in the pancreas. These cysts are not cancerous, so surgery to remove them is not warranted. But they can develop into a tumour, says Besselink. A vaccine might prepare a person’s immune system to ward it off.

Preventive vaccines have been developed against viruses, such as human papillomavirus and hepatitis B, that can cause some cancers. Currently, there is no vaccine that targets precancerous cells to prevent cancer, but some researchers see KRAS-targeting vaccines as an opportunity to change that. “Mutated KRAS is present in 80–90% of precancers,” says Zaidi. “So it is not only an ideal target for cancer, but for precancer.” Vaccines could also be given to the roughly 10% of people with pancreatic cancer who are genetically predisposed to getting the disease. Zaidi and her colleagues are currently running a trial in which people at risk of developing pancreatic cancer and those with cysts receive the group’s KRAS-targeted peptide vaccine. “This is the first ever vaccine being tested in people who are at high risk,” says Zaidi. The first 20 people have been vaccinated safely, with promising signs of a T-cell response in their blood7. A second group with pancreatic cysts is now receiving the vaccine.

Even with the many hurdles, early results from vaccine work — both prophylactic and therapeutic — are serving as a bright spot in a field with dim prospects. “This is super exciting,” says Besselink. The potential for a vaccine to train the immune system to keep pancreatic cancer in check for a lifetime has reset how physicians and researchers think about future treatments. “The goal is not to extend survival by a few months,” says Dougan. “The goal is to take people who would have died and cure them of their cancer.”