Personalized Cancer Vaccine: New Feat Played on Old Ideas Brings New Hope for Cancer Patients

The concept of immunosurveillance was first proposed in 1957 by two scientists Burnet and Thomas: immune system is at constant looking out for the “transformed” cells and eliminates them before the situation is getting of out control. It is supported by the observation that patients with compromised immunity, congenital or acquired (as such HIV infection) tend to have much higher risk for developing cancers. In theory, by boosting the cancer patients' immune response, the immune system might be able to fight off cancer. However, killing cancer cells by enhancing the immune system proved to be elusive. The problem lies in a well know phenomenon that cancer cells can play tricks on the immune system to evade immune surveillance. First, they can hide themselves from being discovered by covering up their neoantigens, the landmarks used by immune cells to tell apart cancer cells from normal cells. Second, they produce several substances to suppress or even kill the immune cells when they are closing in.

With the new advent of genomics and personalized medicine, scientists start to break the ice. For example, new therapy using anti-PD1 monoclonal antibody (Keytruda or pembrolizumab) had shown promises in improving survival for advanced stage lung cancer patients. Anti-PD1 has now also been approved by FDA for treating other type of cancers. How does anti-PD1 work? It works by blocking the cancer cells's ability to "switch off" the attacking T lymphocytes, a key player in cancer immune response. PD-L1 is a protein produced by cancer cells. PD1-L1 binds to PD1 on T lymphocytes and "turn off" the T cells. Anti-PD1 blocks PD1/PD1L-1 engagement, thus allows for resumption of the T cells attacks against cancer cells. Another example is chimeric antigen receptor T cells (CAR-T), the modified patients' own T cells, which has demonstrated long lasting remissions for therapy-refractory leukemia and lymphoma. The production of CAR-T involves genetic re-engineering the patient’s own T-cell receptor to produce "super” T cells for cancer killing. The re-engineered "super" T cells are grown in large number in Petri dish and infused back to the patients. It is like to re-arm soldiers with better weapons and sent them back to the battle field to win the battle.

In the recent two papers published in the July 5 issue of the scientific journal “Nature”, two teams published their “game changing” studies on personalized cancer vaccines, both were able to elicite strong immune response towards the patients' cancer cells. One is Catherine Wu’s team at the Dana-Farber Cancer Institute in Boston. They have treated 6 melanoma patients using their formulated vaccines, 4 of them are disease free two years later. Two showed recurrent disease but the tumor melts away when subsequently treated with anti-PD1. The second team is led by Ugur Sahin at the University of Mainz in Germany. They treated 13 melanoma patients with specially designed vaccines. Eight patients with no visible tumor at the time of vaccination remained tumor free a year later. For the five patients with their tumor already spread at the time of vaccination, the tumors shrank in two. Another patient was tumor free after anti-PD1 therapy.

Cancer vaccine is not a new idea. Many have tried and failed. The idea is based the differences in surface landmarks between cancer cells and normal cells and are produced by DNA mutations. Cancer cells are known to carry abundant mutations in their DNAs. DNA mutations are why a normal cell is transformed into a cancer cell in the first place. Even for normal cells, DNA mutations occur all the time. Most of the mutations are corrected by the DNA repair mechanism. For the ones missed by the DNA repair machinery, majority of them are harmless. For the few detrimental ones, most of them will trigger a process called programed cell death or apoptosis: the cell dies and the story ends. Thus, only a small bunch of mutations may eventually cause cancer. The cancer causing mutations unleash the power for the cells to proliferate uncontrollably. It is the uncontrolled proliferation that makes cancer cells such bad players. Proliferation by itself isn't a bad thing. In fact, most of our cells proliferate, in a controlled fashion. Our skin and bowel cells turns over rather rapidly to maintain our health status - the processes are well programed and controlled. In a sense, our cells work as supercomputers. The programs to run the computers are coded in the DNAs. Cancer cells started with acquiring bad mutations. And when cancers grow they tend to accumulate more mutations. Some mutations are in the coding regions - DNA “codes" for programing the proteins. While ptoteins are antigens, landmarks recognized by the immune system. In theory, If one can produce enough amount of cancer specific neoantigens, one can use them to vaccinate the cancer patients to set off immune response to kill cancer cells, just like the vaccines used for infectious diseases such as flu and hepatitis. But there is one problem: the mutations in cancers are somewhat random. As a results, even for the same type of tumors different patient may bear different mutations, thus express different neoantigens; even the same patient at different cancer stage may bear different sets of neoantigens.

How did the above mentioned two teams do differently to solve the problem? They turned to the newly developed genomic tools. First, they collected each of the patient’s tumor sample. They then did in-depth DNA sequencing using the high-through-put next generation DNA sequencing. Using bioinformatic algorithms, mutations producing neoantigens are identified. Neoantigens best suitable for inducing effective immune responses are selected. 10-20 of the selected neoantigens were then concocted into injectable vaccines, each Conststing a mixture of 10-20 neoantigens. The results are potent immune response towards the melanoma cells, one of the most malignant cancers known.

You may ask why the cancer cells did not trigger effective immune responses before the vaccination. Well the answers lie in the cancer cells ability to hide (cover up) their neoantigens, thus preventing them from being "seen" by the immune cells. Vaccination essentially feed neoantigens directly to the immune cells so that effective immune response can be triggered.

Why the results from these two studies are so exciting? Because it answered some of the most fundamental questions in cancer immunology. For a long time, there are doubts on if the immune system is powerful enough to combat cancer at all, especially when the cancer is at advanced (late) stage. Some even propose that due to the cancer cells high proliferation rate, immune system will never be able to “keep up with” cancer growth, especially for the fast-growing cancers. The antigenic properties of the neoantigens, namely, how effective the neoantigens may serve as antigens, were also unknown. Another concern is cancer cells can hide their neoantigen so well that even the specially "trained" immune cells would not be able to find them. There are also evidence that for a given tumor the neoantigens may change to fast; they are moving targets for the immune system. While, most of the questions seem settled with the two new studies: when the right sets of neoantigen are given in the right way, effective cancer killing immune response can be induced. If the approach proves to be effective in the fight against other type of cancers, a new era of cancer treatment will arrive. If so, winning one of the most challenging battles which has ever seemed elusive to us may indeed be within the grab.

Nonetheless, there are still obstacles ahead. The top issues right now are the costs, technologic difficulties and the time needed to produce such vaccines. If it can only be done in high profile research institutes to serve only a few, it may potentially become a game only for the riches. I am rather optimistic on this front. Looking at the history of technological advances and their adaption as seen for cell phone, high-def TVs and personal computers, it seems commercialization will lead to the reduction of costs and mass adaption. With time, better algorithms for generating vaccines may also be developed. When the the number of cases increases, it may even be possible to establish neoantigen banks, thus making pre-made vaccines possible. It seems we may indeed be at the dawn of winning the war against cancer.

Reference
Dunn GP, Bruce AT, Ikeda H, Old LJ, Schreiber RD (Nov 2002). "Cancer immunoediting: from immunosurveillance to tumor escape". Nature Immunology. 3 (11): 991–8.
Ott, P. A. et al. Nature http://dx.doi.org/10.1038/nature22991 (2017).
Sahin, U. et al. Nature http://dx.doi.org/10.1038/nature23003