by Cathy Gutierrez
figures by Lillian Horin

“The history of cancer vaccines is a history of failure.”

This is the leading sentence of a 2005 article that summarized the history of cancer vaccines. Cancer vaccines have long been the Holy Grail of cancer research. For centuries, scientists have been devising ways to train the body to destroy tumors. Despite the success of early preventive cancer vaccines (such as Gardasil for prevention of cervical cancer), vaccines to treat cancer had largely been unsuccessful. However, two recent clinical trials generated promising results using new vaccines that are tailored to a person’s particular combination of mutations, or acquired changes in their DNA. Although these results are preliminary, they may be a game-changer in the way we treat cancer—they represent a divergence from the one-size-fits-all method that has dominated the cancer treatment field for centuries.

How Do Cancer Vaccines Work?

Cancer vaccines use the body’s immune system to fight cancer. Our immune systems are formidable armies who are experts at detecting and destroying foreign agents, such as bacteria and viruses. They perform their duties using an intricate surveillance system, which recognizes antigens, or the proteins in germs that are not normally found in the human body. Cancer cells can trigger this surveillance system because they often contain mutations that correspond to antigens that immune cells can spot just as they detect foreign proteins on bacteria.

So if our immune systems can recognize cancer cells, why do we still get cancer? Cancer cells are harder to target than bacteria for several reasons. First, they are still human cells and may not look so different from normal cells while in the early stages of cancer. And once our immune cells are finally able to distinguish them from healthy cells, the immune response might not be strong enough to destroy the entire tumor. Second, cancer cells are highly adaptable. Tumors often develop ways to avoid detection and destruction by releasing substances that make immune cells less active or completely nonfunctional. Third, people with cancer tend to have weaker immune systems due to age, prolonged illness, or side effects of cancer treatments.

Cancer vaccines help to boost our immune response against cancer in the same way that vaccines against infectious diseases protect against the flu and other infections. In a nutshell, vaccines mimic disease—they deliver small amounts of antigens into the body that come from bacteria, viruses, or cancer cells but cannot cause disease. Antigens can then be collected by certain types of immune cells that are broadly called antigen presenting cells. These cells will ingest the antigens, process them, and then present the fragments on their surface using molecules called MHC class II proteins. MHC class II proteins can attach to and activate CD4+ T cells, also known as helper T cells. Helper T cells can kill cells carrying the antigens directly, but they can also activate other killing immune cells, such as CD8+ killer T cells (Figure 1). Helper T cells can even generate CD8+ T cells with “memory.” These “memory” T cells will respond strongly if they ever come across the offending antigen in the future.

Non-immune cells, including tumor cells, can also present their own internal antigens directly to T cells. They do so by presenting their antigens on MHC class I proteins to CD8+ killer T cells. If CD8+ killer T cells view these internal antigens as foreign, they can directly attack and destroy the offending cell.

Both of these systems can be harnessed in cancer treatment vaccination. In preliminary studies using cancer vaccines in the lab and in mice, scientists have shown that CD4+ and CD8+ T cells are activated, and that both can generate strong immune responses.

Promising Clinical Trials

After decades of hard work and little to show for it, two clinical trials published in July 2017 in Nature showcased two new cancer vaccines that appear to prevent cancer from returning in patients with late-stage skin cancer. In both trials, patients had their initial tumor removed and were treated with the new vaccines in an adjuvant setting, which includes additional therapies designed to increase the immune response to vaccination.

The first trial treated 13 melanoma patients, eight of whom remained tumor-free two years after tumor removal and vaccination. The remaining five patients’ tumors had already spread to other parts of their bodies by the time they received the vaccine, yet they still saw some benefit. Of these five patients, two of them saw their tumors shrink, and a third was treated successfully with a different type of immunotherapy called a PD-1 inhibitor that essentially turns off the “off switch” on T cells, enabling them to attack cancer more effectively.

The second trial treated six patients, four of whom were cancer-free two years after tumor removal and vaccination. The two patients who saw their tumors return experienced a complete response to PD-1 inhibitors. For perspective, only 6% of patients receiving PD-1 inhibitors alone have experienced a complete response to treatment.

Caveats to these studies include their small sample sizes and lack of a control group of patients who received just the standard treatment and not the vaccine. More studies will be necessary to appreciate the benefit of cancer vaccines, but some of these experiments are already underway.

What makes these cancer vaccines so different from the many others that have failed? Quite simply, technology has changed. For some time now, researchers have recognized that cancer can be vastly different from patient to patient—in fact, cancer can be different from cell to cell, within the same patient. Until recently, however, it was difficult, costly, and time-consuming to figure out what made each cancer cell different. Now, developments in the field of DNA sequencing—the reading of our genetic code, the blueprint from which our cells are made—have allowed us to become faster at detecting mutations.

Many of these mutations are harmless, but some encourage cancer cells to grow out of control and others code for new types of cancer proteins called “neo-antigens,” or new antigens. Since neo-antigens are not found in healthy cells, T cells can recognize and attack the cells that present them (Figure 1). New machine learning computer algorithms, which provide computer programs with the ability to access data, learn, and improve from experiences without being specifically programmed, have been trained to recognize mutations in DNA that can produce neo-antigens. Both of the clinical trials mentioned above used cancer vaccines that help T cells target neo-antigens.

These new cancer vaccines are impressive because they’re entirely unique to each patient (Figure 2). Each patient that enrolled in these trials had her tumor sequenced to identify the mutations in her cancer cells. A personalized vaccine was then generated with a concoction of 20 neo-antigens that were predicted based on the DNA of each patient’s cancer cells. Although 20 neo-antigens may seem like a big ask given the amount of work that goes into making each of these vaccines, researchers believe that targeting more neo-antigens means fewer cancer cells can squeak by unnoticed simply because they happen to lack the neo-antigens in that particular vaccine.

What’s Next for Cancer Vaccines?

Although the future looks promising for cancer vaccines, there are some significant wrinkles to iron out. First, producing the vaccines for the successful Nature studies took months. However, researchers behind both studies seem confident that they can shorten future production time to 6 weeks or less. Additionally, cost remains a factor—the production of a single neo-antigen vaccine has been estimated to cost around $60,000.

Our growing understanding of cancer and its complexity has made it clear that there is no single cure for cancer. Instead, personalized cancer vaccines leverage our new ability to read and understand our individual DNA blueprints to design therapies that are custom-made for each patient.

In the process of producing these personalized vaccines, we are gaining more insight into what makes cancer “tick,” which can be used to make better cancer vaccines in the future. If certain neo-antigens are common across patients, panels with different combinations of neo-antigens could lower production time and costs. We are also realizing that combining cancer vaccines with other treatments, like PD-1 inhibitors, can help our immune systems find and destroy cancer cells.

We still have much to learn, but the last few months have started a firm, hopeful shift away from the 2005 headline, “The history of cancer vaccines is a history of failure.” Cancer treatment vaccines were, for a long time, the stuff of science fiction—but today, finally, they are becoming reality.

Cathy Gutierrez is an M.D./Ph.D. student at Harvard Medical School.

For more information:

• A History of Cancer Vaccines
• Cann SH, Van Netten JP, Van Netten C. Dr William Coley and tumour regression: a place in history or in the future.
• Kucerova P, Cervinkova M (April 2016). “Spontaneous regression of tumour and the role of microbial infection–possibilities for cancer treatment”.
• Cover image: Cancer cell by Destroyer of furries