Headlines like “We’re fighting cancer with a universal vaccine” and statements such as “A vaccine could be developed which could be used against all cancers” are what’s wrong with science journalism, because they are far from the truth hidden in this study published in Nature.
When most people read “universal cancer vaccine”, they think about a prophylactic vaccine, much like the ones given to children to prevent diseases caused by bacterial or viral infections such as measles, diphtheria or the flu. This is not the case for this cancer vaccination strategy. The study in question reports a novel way for a therapeutic vaccine, given to patients to stop and kill existing cancer cells. And while this cancer vaccination strategy could be applied “universally” for most cancer types, there will be no universal vaccine that can be given to all patients! The vaccine must be specific for each cancer type – so it is a form of personalised therapy. Not universal at all.
Although the media have overstated and sensationalised the outcome of the study, the new approach is interesting and could still help a lot of cancer patients.
What does the science really show?
Cancer cells acquire mutations in their genetic code, allowing them to grow and divide in an uncontrolled manner. They stem from normal cells and still “look” mostly like normal cells to the immune system, which is why there is rarely a strong immune response mounted against cancer cells. The immune system struggles to distinguish normal from cancer cells.
The vaccine developed in the study at hand aims to activate the immune system of cancer patients and thereby enable them to fight cancer from within. To do so they target so called neo-antigens, molecules that are newly formed by mutations, are specific for the cancer cells and not expressed by other cells in the body. These neo-antigens are capable of inducing anti-tumour immune responses.
To prime the immune system for neo-antigens (it’s like given a hound the smell of its prey), RNA encoding for the neo-antigens was packaged into nanoparticles, small fatty acids (lipid)-coated particles and injected into tumour-bearing mice or patients. An innate immune cell type called dendritic cell takes up the RNA-nanoparticles and activates a large-scale immune response including neo-antigen-specific CD8+ T cells. These T cells are capable of finding and destroying neo-antigen expressing cells and therefore fight the tumour.
In mouse models, both prophylactic and therapeutic administrations of the RNA-nanoparticles led to the prevention or regression of several well-known cancer types, respectively. These promising results and good safety records in primates promoted the translation of the approach into patients. Of the three melanoma patients given nanoparticle treatment so far, all mounted a strong immune response against one or two of the four melanoma-specific antigens they were vaccinated against. One patient showed shrinking of their metastasis, while cancer growth was stunned in another. The final patient showed no new cancer growth after surgical removal of their metastasis.
These results are striking, especially given the mild side effects of flu-like symptoms in the patients. Nonetheless, we need to await the results from larger trials to see if this effect holds true before talking about “breakthroughs”.
What did the media not say?
While the news were quick to point out the great results in the early clinical trials and talked about the method being universally applicable in all cancer types, most failed to report the caveats of this study.
This “universally applicable” approach won’t actually work in all cancer types. In order to be able to vaccinate we need to know the specific neo-antigens produced by each cancer type and these are different between cancer types as well as between patients suffering from the same cancer type. Take this study for example, the three melanoma patients did not react to all four neo-antigens they were vaccinate against. They selectively mounted immune responses against one or two neo-antigens. Every person reacted differently based on the mutations their cancer had. Knowing the mutations in each tumour is essential for the success of this treatment. To achieve this, each tumour needs to be checked for mutations (by genome sequencing) and the vaccination needs to be adjusted accordingly.
Another issue is the fact, that cancer cells are not all alike. They are heterogeneous within a tumour. You may have some cancer cells expressing neo-antigens while others do not. So even if you successfully target one neo-antigen you may not be able to attack the whole tumour.
Lastly, not all tumours express neo-antigens. This study vaccinated three patients against four neo-antigens known for melanoma. Melanoma is the cancer type with highest mutation rate of all cancers, which is why a high number of neo-antigens are known for melanomas. A lot of these have been excessively studied regarding their ability to effectively activate a tumour-specific immune response and this study picked well characterised neo-antigens to proof the efficacy of their vaccine. While this is good news for melanoma patients, this means developing a vaccine for other cancer types will be trickier, if not impossible, if no unique neo-antigen is found. It is important to note that not every mutation is immunogenic – meaning that not all mutations found in cancer cells will lead to the production of a neo-antigen that can be recognised by the immune system. Thus, while all cancer cells carry mutations, we won’t be able to use all of them for vaccination strategies.
To conclude, this vaccination approach has valid success chances in some cancer types, however, it won’t be a universal cure for cancer. Unfortunately the study has been blown out of proportion in the headlines to get more clicks, in complete disregard of the false hope this may create. John Oliver has done a great piece on science journalism recently- so if you are interested in why there is a problem in reporting – check it out.
Featured Image: Study participant receives vaccine. Flickr. License: CCBY 2.0
Are you affiliated to DKFZ and motivated to blog about science, DKFZ, research life (or all of these)? To be a blogger, send us an email on firstname.lastname@example.org!