Microsatellite-unstable cancers are characterised by a large number of mutations within short repetitive DNA sequence regions, and can form novel peptides that the human immune system recognises as neoantigens. These cancers represent a starting point for the development of vaccines to prevent them appearing at an early stage of development. Microsatellite instability is particularly frequent in colon and cervical cancers.
At the 4th interdisciplinary symposium “Innovations in Oncology”, held in Berlin on 22nd June 2018 and ambitiously titled “Die Neuvermessung der Onkologie” (Remeasuring oncology), Magnus von Knebel Doeberitz gave a programmatic presentation about ways to effectively combat tumours with vaccines in the future. In the early 1990s, von Knebel Doeberitz, professor of molecular oncology at the Heidelberg University Hospital, worked at the German Cancer Research Center (DKFZ) in Heidelberg under Harald zur Hausen, the man who was awarded the Nobel Prize in Physiology or Medicine in 2008 for demonstrating that cervical cancer in humans is caused by certain types of papilloma viruses (HPV). Zur Hausen’s discovery led to the first effective cancer prevention vaccine. Von Knebel Doeberitz comments: “HPV vaccines have the potential to wipe all HPV-associated cancers from the face of the earth” - if the vaccine is used consistently (which is not yet the case, especially in Germany).
For a normal body cell to develop into a cancer cell it has to become "genomically unstable". This involves the reprogramming of genetic information that causes the cell to divide and acquire countless new mutations, eventually leading to the selection of a cell clone with tumour-specific properties. The development of malignant epithelial tumours can be divided into three stages: the first stage leads to genomic instability and is called a “breakthrough event”. It usually results from mutations in genes that are characteristic of certain cancers - mutations in the APC gene in the case of colon cancer, and a point mutation in the IDH1 gene in the case of certain brain tumours. The activation of certain viral oncogenes is the initial cancer-causing event in HPV-associated tumours.
As these initiated cells continue to divide, their randomly acquired growth-stimulating mutations give them a selective advantage over other cells. They can induce the second stage of carcinogenesis known as the premalignant "expansion phase." If a cell clone acquires further mutations that enable it to penetrate neighbouring tissue and metastasise, the cancer enters the malignant "invasion phase", the third stage of carcinogenesis. Carcinogenesis - from breakthrough event to malignancy - is thus driven by a sequence of mutations that are characteristic of many cancers. These mutations are called driver mutations. They are currently a main focal point of cancer research, notably in terms of the development of targeted therapies.
As von Knebel Doeberitz noted, not all mutations that drive carcinogenesis lead to antigenic structures that can be used to develop cancer vaccines. The structural changes associated with oncogenic point mutations (as in the IDH1 gene) are only weakly antigenic, and barely recognisable by the immune system. This is different from insertion and deletion mutations (so-called "in/dels"), which lead to a frame shift. When these changes occur in coding regions of the genome, frame-shift mutations can lead to completely new peptide sequences (“frame shift peptides”, FSP), which the immune system recognises as foreign and forms antibodies against them.
Such in/dels tend to occur mostly in tumours with a defective DNA repair system, especially in the area of short repetitive DNA sequences, the so-called microsatellites. "Microsatellite-unstable cancers" (MSI cancers) are particularly frequent in colon and uterine cancers. A high proportion (15-20 percent) of MSIs are hereditary; this inherited predisposition to MSI cancer is called Lynch syndrome. It is by far the most common hereditary predisposition to the development of cancer; however, only around half of Lynch syndrome carriers actually develop the cancer during their lifetime.
“The vast majority of MSI cancers share certain mutations in the coding microsatellites, including mutations that drive the development of MSI cancers”, said von Knebel Doeberitz. These shared mutations are expected to lead to FSPs that are common to all MSI cancers, including neoantigens that represent targets for the immune system. In a comprehensive bioinformatics-based approach, the scientists from Heidelberg looked at all coding microsatellites of the human genome and selected those that most frequently mutate in MSI cancers. They then predicted the microsatellites’ corresponding gene products. After evaluating gene expression and immunogenicity of these FSP candidates in vitro, 13 candidates were chosen and used to study whether patients with an MSI cancer produce an immune response against these theoretically predicted peptides.
The researchers found that all patients with MSI cancers had developed pronounced immune responses to these FSP neoantigens. Interestingly, a similarly strong immune response was observed in healthy Lynch syndrome carriers. It can be assumed that these individuals had already formed cells that produced the mutated peptides, but that their immune system was able to prevent the cells from developing into full-grown tumours in a process of "auto-vaccination." Based on these observations, the researchers concluded that it should be possible to identify driver mutations for the tumours’ breakthrough event and the expansion phase, and subsequently develop vaccines against these cancers. The principal feasibility of this approach has been successfully tested on a Lynch syndrome mouse model in collaboration with US scientists. Together, the researchers are working on a study protocol for clinical testing of vaccines against Lynch syndrome antigens.
The initial successes with vaccines for cancer prevention were achieved with tumours that are caused by exogenous factors, including cancer-causing HPV and the hepatitis B virus. In the case of MSI cancers and Lynch syndrome, the possibility of vaccinating cancers caused by endogenous driver mutations is now emerging for the first time. As Magnus von Knebel Doeberitz and his colleague PD Dr. Matthias Kloor write in a review published in the journal “Trends in Cancer”, it should now possible to shift cancer immune intervention strategies from end-stage disease towards early-stage adjuvant therapy and cancer prevention: "This transition," say the authors, "represents a significant step towards a more general assessment of cancer vaccination approaches. If successful, it could pave the way for the development of vaccines against other malignancies."
Kloor M, von Knebel Doeberitz M: The immune biology of microsatellite-unstable cancer. Trends in Cancer 2016 Mar, 2(3): 121-133. http://dx.org/10.1016/j.trecan.2016.02.004