AORTIC 2011, Cairo, Egypt 30 November–2 December 2011
Infectious diseases as a cause of cancer
Dr Robert Newton – University of York, UK
One of the critical things about infections as a cause of cancer is they are, at least theoretically, preventable because you may be able to stop people from getting infection or you may be able to treat that infection. If you do that at an appropriate stage, you may prevent them from going on to develop the cancer that’s caused by that infection. So it’s actually quite an important area of study.
In terms of infections as a cause of cancer, we know that about one in five cancers around the world are caused by an infection, that’s about two million cases of cancer a year, and the majority of those occur in developing countries rather than the developed world because that’s where the infections are most common. They’re second only to tobacco as a major preventable cause of cancer.
If we take two of the most important, or three of the most important, infections, hepatitis B is one of them, it causes about 50% of the world’s cases of liver cancer. We’ve had a safe and effective vaccine against hepatitis B available for several decades now. In fact, I’ve been vaccinated, most medical practitioners will have been vaccinated against hepatitis B. It’s very effective at preventing not only the cancer but also the hepatitis and the psorosis that’s associated with infection.
So that’s one. Another set of infections, human papilloma viruses, are sexually transmitted but very common in human populations and they cause cancer of the cervix, which is one of the most important causes of cancer in women worldwide. Again, we can now vaccinate against human papilloma viruses with a view to preventing cancer in the future. So in the UK, for example, we started vaccinating twelve year old girls against these important cancer causing viruses about three years ago. Now obviously it will take a couple of decades before we start seeing declines in cervical cancer because women don’t get it until they’re in their… generally until they’re over the age of thirty. So in the meantime we have to maintain our screening programmes, but that’s another tumour that we hope to be able to prevent.
The third one, I think, that’s worth highlighting is stomach cancer which is caused by a bacterium called Helicobacter pylori and the interesting thing about this bacterium is that you can treat it with antibiotics. We can make it go away, thereby reducing the risk of stomach cancer. And the incidence of stomach cancer is beginning to decline worldwide, probably because of the declines in the prevalence of Helicobacter. So these are important preventable causes of disease.
What are the costs involved with preventing these diseases?
Cost is a difficult thing, it’s always a balance. In reality, prevention of cancer, or of any disease, in the end is cheaper than treatment of that disease. Treatment of cancer can be inordinately expensive and it can involve chemotherapy, that’s drugs; it can involve a radiotherapy, radiation treatments; it can involve surgery. So you can have whole teams of individuals involved in the treatment of a single individual with cancer. Clearly it’s better to prevent that cancer from occurring in the first place.
So a vaccine like the hepatitis B vaccine is now dirt cheap, it costs really only a few pence to produce, it’s off-patent, it’s relatively cheap to buy. The newer vaccines, the papilloma virus vaccines are, at the moment, very expensive because the drug companies have to recoup the enormous costs associated with developing the vaccine but they’ll do that relatively quickly and in time the costs of those vaccines will go down very significantly. Again, the situation where prevention, in the end, ends up being a lot cheaper. The problem we always have, of course, is that for an initial big up-front cost a government has to provide that money and it won’t be that government that’s in power when the benefits of that decision are reaped in the future. So the problem with public health is it does require real long-term thinking. It takes decades, often, for the benefits of things that we do now to be seen.
What are the benefits to cancer registries in Africa?
Cancer registration, fundamentally it’s about counting cancers, it’s about measuring the burden of disease in populations. We’ve been doing this routinely in Western populations for decades, it was something that was really pushed by the late, great Sir Richard Doll, a very famous epidemiologist who died just a few years ago, although it wasn’t initiated by him. The importance of counting the burden of disease, of counting cancers is that it really underpins all of public health and all of public health decision making. If you don’t know what’s killing your population, if you don’t know what’s common, it’s very difficult to make decisions about where to invest and what to put your money in. So cancer registration is really the process of measuring the burden of cancer in different populations; we’re good at it in the West.
In Africa we’re not very good at it. There are five cancer registries in Africa that meet the standards that would be appropriate for a cancer registry in the West, that’s five out of 54 African countries. So most of what we know about the burden of cancer on the African continent comes from just those five centres in five different countries and everything else is an extrapolation from those data. So one of the key messages from this conference is that we need to expand the scope of cancer registration on Africa, we need to get much better at measuring the burden of disease on this continent in order to be able to make sensible decisions about how to prevent disease and where to put our resources.
Will they have to combine the cancer registry with other registries?
Usually what happens with a cancer registry is they focus just on cancers but on all cancers. So they measure the burden of all cancers across a population and you’re able to work out how those cancers vary by age and sex, for example, and then it allows you as well to make comparisons between different populations. I think one of the important things to mention about differences between populations is it’s very often those differences in the burden of disease across different populations that provides clues as to what the underlying cause of that cancer may be. It’s common in one place with one set of exposures, it’s far less common in another place where those exposures are less common. It’s that diversity in the distribution of cancer burden across different populations that very often points to what the underlying cause may be.
Are there infections we have not yet discovered?
Very definitely. There are a number of tumours that are suspected as having underlying infectious causes for which no infection has yet been identified. There are many laboratories across the world that are actively hunting for new viruses and other types of infection that may cause these cancers. The problem with identifying viruses as a cause of cancer is that they’re very, very small and it’s a very, very difficult job to identify those infections in tumour tissue. It may sound simple but it really isn’t. There’s also quite a lot of luck involved in identifying those infections. So it really is a very challenging job and it’s an area where quite a lot of investment is being made.
Is it difficult to link the initial infection with the end tumour?
It’s certainly true to say that for most of the cancers we know are caused by infections, the infection itself will pre-date the development of cancer by decades. That’s true of most cancers and most exposures. We know that tobacco causes more cancers than anything else but you have to smoke for decades before your risk of cancer really starts to go up and the same is true of infections. You can be infected with hepatitis B virus for thirty or forty years before the cancer develops, and the same for papilloma viruses in relation to cervix cancer. That’s one of the problems we face, that the period of exposure has to be so long before the cancer develops that sometimes it’s quite difficult to work out where the associations lie.
