EACR 21, 26—29 June 2010, Oslo

Interview with Professor Arnold Levine (Cancer Institute of New Jersey, USA)

The role of tumour suppressor cells in the prevention of cancer

My field of study for literally all my career, 40 years, 41 years in my career, has been cancer biology. I started off by studying viruses that cause cancer and that led me to study proteins that interact with viral proteins that cause cancer. Some of those proteins turned out to be important in the origins of human cancer; in particular the p53 protein and the p53 gene was discovered first by its interaction with an oncogene from a virus that causes cancer in animals. That led us back to the gene and the gene led us back to mutations in humans and today we now understand that about 50% of human beings that have cancer have mutations in the p53 gene.

In the United States there are about 250 families that inherit p53 mutations and anyone who inherits a p53 mutation will, with 100% probability, develop cancer, up to five independent cancers over a lifetime. So that’s called Li-Fraumeni syndrome. The great majority of p53 mutations in cancer are spontaneous somatic mutations and these inactivate the function of the gene and it’s these inactivations that lead to cancer. So that means that the protein that the gene p53 encodes prevents cancer, and that’s why it’s called a tumour suppressor gene.

What were you talking about at EACR 2010?

p53 is usually one of the mutations that will give rise to cancer and so it would be common to find five or six different mutations in critical genes, some of which are oncogenes, others of which are tumour suppressor genes that contribute to cancer. What we’ve begun to learn is that if you treat the cancer with a drug that will inactivate one of the mutated genes, like an oncogene like ABL for example, if you treat with the drug the ABL protein kinase in chronic myelogenous leukaemia then you’ll stop the cancer. So all five of the mutations appear to be important and central in the growth and the spreading of the cancer and attacking one will be sufficient to stop it. Unfortunately there are then mutations that give rise to resistance to these drugs and so, probably in the long run, the best strategy will be to attack two or three of these mutations.

How is p53’s role in the fidelity of gene transmission relevant to cancer?

With p53, it is involved in fidelity of somatic cell transfer and what happens there is if there is DNA damage and the cell is replicating itself, it duplicates the DNA. The damage site is a site for increased mistakes and that means increased mutation rate. That inactivates tumour suppressor genes, it activates oncogenes and so it raises the probability of cancer. p53 sees this damage in the DNA and it kills the cell, preventing the cell from propagating and restricting the cancer that arises. Its absence, therefore, would give rise to cancer.

The sisters of p53 in the human genome called p63 and p73, they’re related transcription factors, play a very similar role in the germline, especially the female germline. The fidelity of information transfer in the germline, in the egg, to the offspring is very important and if mistakes are made in the germline then you will spontaneously have a mutation that will give rise to an inherited alteration and some subsets, not all of them but some subsets, of these mutations that are passed through the germline could inactivate tumour suppressor genes and begin cancer at a young age in the offspring.

The offspring are at increased cancer risk even if the parents don’t have cancer?

The parents have the propensity to make mistakes in the eggs so the offspring is at risk but the parents don’t have to have that mutation that gave rise to cancer. This is a new concept and the concept would be that there would be inherited mutations that give rise to cancer but it wouldn’t be familial, it wouldn’t be found in the families, it would only be found in the offspring. The only way we’re going to know these are important contributors to cancer is if we see this in five or ten or twenty independent individuals that come from twenty different families. If they come from twenty different families and they have the same mutation which was spontaneous because the egg had that mutation, then we have uncovered a new tumour suppressor gene.

Are you screening for this?

Sure, as soon as the concept became clear we began a study in a large dataset that’s called the Framingham dataset. It’s three generations of individuals who have offspring that have developed cancer in families that don’t seem to have a large rate of cancer, for example. In the Framingham study we can identify about 630 women who have developed early onset breast cancer. It’s those families, the 600 or so families, that would be at potential risk for a spontaneous mutation and we will have the data shortly to test this hypothesis but it’s just a hypothesis at this time.

If this proves to be a correct hypothesis, the long-term goal will be to identify women at risk and then to follow them more closely. Because if the cancer can be found at early times then there’s a higher probability it will be cured.