Genetic clues to prostate cancer

The Daily Mail has reported that “nine further genes associated with a higher risk of prostate cancer have been discovered”, and this raises hopes of developing new drugs to treat the condition. The research reportedly showed that men who carry the faulty versions of these genes are twice as likely to develop prostate cancer as men who do not.

The news story is based on four studies published in the Nature Genetics journal that used genetic analysis to find variations that are more common in men with prostate cancer. The variants identified in these studies may not themselves be causing the increase in prostate cancer risk, but may instead lie close to genes that are having this effect. The authors of some of the research suggest further research will be needed to confirm the genes responsible.

Many regions within our DNA appear to contribute to prostate cancer risk, and more are likely to be discovered. As yet, it is not clear whether a screening programme based on what we know about the genetics of prostate cancer would be helpful in identifying men at risk of the disease.

Deciding on whether to start a screening programme is a complex process. It would need to take into account a number of issues such as testing methods, what constitutes being at risk, what treatment should follow and how reliable the test will be. The consideration of these issues will need to be decided using the evidence provided by further research.

Where did the story come from?

The research discussed in the news had been published as four papers in the peer-reviewed journal Nature Genetics . All of the studies were carried out by large international collaborations of scientists.

The first authors on the individual papers were:

• Dr Meredith Yeager from the National Cancer Institute in the US.
• Dr Julius Gudmundsson from deCODE genetics in Iceland.
• Dr Rosalind A Eeles from The Institute of Cancer Research in the UK.
• Dr Ali Amin Al Olama from the Cancer Research UK Genetic Epidemiology Unit.

What kind of scientific studies are these?

All of the research was based on genome-wide association studies (a type of case-control study). These studies look at a large number of specific sites across people’s DNA to identify variations that are more common among people who have a specific condition (cases) than those who do not (controls).

Study one
This genome-wide association study by Dr Yeager and colleagues compared the genetic make-up of 10,286 men with prostate cancer (cases) and 9,135 men without prostate cancer (controls). All participants were of European ancestry.

Study two
This study by Dr Gudmundsson and colleagues took data from previous genome-wide association studies from Iceland and other locations including the US and Europe. The researchers pooled this data together in order to try to identify new variations that were associated with prostate cancer, and to look closely at two regions in the DNA that previous research had associated with prostate cancer. These regions were on the long arms of chromosomes 8 and 11.

Once they had carried out this analysis, they looked at all of the variants reported to be associated with prostate cancer in Icelandic samples. They used a statistical model to estimate what proportion of the population carried the highest risk variants and what their risk of prostate cancer would be in relation to the population as a whole.

*Study three
*The study by Dr Eeles and colleagues was an extension of an earlier genome-wide association study. This extension study had two stages. In the first stage, the researchers looked at a further 43,671 genetic sites in 3,650 men with prostate cancer (cases) and 3,940 controls. In the second stage, they looked at these sites in an additional 16,229 cases and 14,821 controls from 21 studies.

The researchers used a statistical model to determine how much of the excess risk that runs in families could be explained by the variants they identified. They then calculated the effect these variations have on risk of prostate cancer in men with the highest and lowest levels of genetic risk, based on these variations compared to the population as a whole.

Study four
The study by Dr Al Olama used the same methods as used in a genome-wide association study but concentrated on a smaller stretch of DNA. The researchers took a closer look at a region on the long arm of chromosome 8, which was previously found to contain variations associated with an increased risk of prostate cancer. They looked at genetic variations in 5,504 men with prostate cancer and 5,834 controls.

What were the results of the study?

Study one
A single-letter variation on the long arm of chromosome 8 was found to be associated with prostate cancer susceptibility. Men who carried two copies of the high risk form of the variation were 33% more likely to have prostate cancer than men who carried none. Men who carried one copy of the high-risk form were 17% more likely to have prostate cancer.

Study two
The second study identified five variations in the DNA that were associated with prostate cancer susceptibility, on the long arms of chromosomes 3, 8, 19, and 11. Individually these variations increased the risk of prostate cancer by between 12% and 23%.

When this research was combined with previous studies, 22 genetic risks were reported to be associated with prostate cancer in the Icelandic men sampled. The researchers estimated that about 1.3% of the Icelandic population carried the greatest number of high-risk variants, and these men would have over 2.5 times the odds of having prostate cancer compared to the general population. They estimated that 9.5% of the Icelandic population were estimated to carry the lowest proportion of high risk variants, and these men would be at less than half as likely to have prostate cancer compared to the general population.

Study three
The researchers identified seven new variations in the DNA that were associated with prostate cancer susceptibility, found on chromosomes 2, 4, 8, 11, and 22. These variations were near various genes that could potentially be playing a role in prostate cancer, including the gene NKX3.1 on chromosome 8. The researchers estimated that the new variations that they identified would explain about 4.3% of the excess risk of prostate cancer seen in first degree relatives of men who have prostate cancer. Combined with other previously identified variations, they would explain about 21.5% of the excess familial risk.

Based on this model, the 10% of men with the greatest genetic risk would be at about 2.3 times the risk for prostate cancer in the general population, and the top 1% would be at about three times the risk. Men in the bottom 1% of genetic risk were estimated to have around one-fifth of the risk for prostate cancer compared to the populations’ average risk.

Study four
The fourth study confirmed that three areas reported to be associated with prostate cancer in previous studies were associated with the disease. They also identified two new genetic variations that were associated with prostate cancer risk. Carrying the low risk forms of these two variations reduced the risk of prostate cancer by 13% and 10%.

What interpretations did the researchers draw from these results?

The study by Dr Eeles and colleagues concludes that prediction of a man’s risk of prostate cancer based on the variants identified “may have implications for targeted screening and prevention”. Dr Yeager’s group says, “further investigation is warranted to determine the molecular basis” of each of the prostate cancer associated variations found. Dr Gudmundsson’s group, who combined the genetic markers into a model for predicting risk, say that given the pace of new discoveries, the model will need constant updating.

They all say that further work will be needed to identify the genes that are actually causing these increases in risk.

What does the NHS Knowledge Service make of this study?

These four studies increase the information we have about sites in our DNA that are associated with prostate cancer risk. These variants may not themselves be the cause of an increased prostate cancer risk, but may instead lie close to genes that are having this effect. As the authors of some of the research suggest, further research is needed to identify these genes.

Many regions within our DNA appear to contribute to prostate cancer risk and further regions are likely to be discovered. At present, it is not clear whether a screening programme based on what we know about the genetics of prostate cancer would be helpful in identifying men at risk of the disease.

Deciding whether to start a screening programme is a complex process, and there would be several considerations that would need to be assessed. These considerations could include:

• The ability of the potential testing methods used to differentiate between people at greater and lesser risk.
• What threshold is appropriate for deciding that a person is at risk.
• What the risks of the test are.
• Whether effective measures are available to reduce risk in those identified.

These types of issues need to be considered and researched in the future, and used to inform these decisions on screening.