Infections and cancer - an overview
This page contains information on
- the burden of cancers caused by infection,
- viruses in general,
- viruses and cancer,
- causal association,
- basic mechanisms of cell growth transformation and
- how we can reduce the burden of infection-associated cancer.
The burden of cancers caused by infection
Table 1.11 presents the the estimated number of cancers worldwide linked to infectious agents. In some instances, including Epstein-Barr virus (EBV) and Burkitt/Hodgkin lymphoma, and human herpes virus type 8 (or Kaposi sarcoma-associated herpesvirus, HHV8) and Multicentric Castleman's Disease, not all cases of the related tumours carry the infectious agent, but the virus clearly accelerates a process that otherwise happens rarely. In other instances, such as human papillomavirus (HPV) and cervical carcinoma; EBV and nasopharyngeal carcinoma; and HHV8 and Kaposi sarcoma, every case of the disease involves the infectious agent.
The process of cancer development is complex, requiring multiple steps in addition to virus infection, and so the latency period (from viral infection to appearance of the virus-positive tumour) can be many years. Although only a small proportion of virus-infected people develop these cancers, the total burden of infection-associated cancer is very large, with 2 million new cases of cancer worldwide in 2008 attributed to infectious, around 16% of all new cancer cases (Table 1.1), although the proportion is lower in developed countries.2 A study published in December 2011 estimated that around 3% of cancer cases in the UK are linked to infections.6
section updated 22/05/12
Viruses in general
Viruses are very simple forms of life, consisting of genetic information (as a DNA or RNA genome) wrapped in a protective coat of proteins. When viruses infect an animal or human being (the host), they invade living cells, and usually turning them into 'factories' that replicate the virus and produce new virus particles. These particles can either invade more cells in the host or spread the infection to a new individual. Infection need not always lead immediately to full virus replication. Some viruses are able to persist in the body for many years by hiding inside cells in 'latent' form; the virus genome is present within the cell but only a few of the virus' genes (sometimes called the 'latent' or 'early' genes) are expressed and no new virus particles are produced. With certain viruses, this type of persistence can predispose the latently-infected cells to cancerous change. The general rules of virus infection are outlined in Table 1.2.
Viruses and cancer
Work on cancer in animals first revealed the link with viruses and provided the foundation upon which all present work on virus-associated human cancers is based. The types of viruses that are linked to human cancer can often be found in many people in the normal healthy population, not just in the few who develop the malignancy. Cancer therefore represents a very rare accident of long-term infection with such a virus. Almost all forms of cancer arise through a multi-step process; a series of genetic accidents must accumulate in a cell before that cell becomes malignant and multiplies to form a tumour. In the case of virus-associated tumours, one of these 'genetic accidents' is viral infection (Table 1.3).
Causal association
What kinds of evidence are needed to link a virus with a particular type of cancer? It is not enough just to show that all patients with the cancer in question have a history of infection with that particular virus, since many healthy individuals have likewise been infected. The evidence needed for a direct causal association is summarised in Table 1.4.
For most virus types linked to cancer, the crucial evidence comes from the cancer tissue itself; often every cancer of a certain type is virus-positive, and every malignant cell within the cancer, carries the same virus genetic information. In such cases, each cancer is made up of a clonal population of cells descended from a single progenitor cell infected with the virus before clonal growth began. For most, but not all, cancer-associated viruses, the virus genome is present in the tumour cells and, in addition,, certain virus genes continue to be expressed. This is strong evidence that the virus is acting directly to promote tumour growth. In most cases of this kind, infecting normal cells with the virus in the laboratory, or introducing individual virus genes into such cells experimentally, can be shown to alter ('transform') cell growth.
Not all infectious agents linked to cancer act in this way. There are a few agents that appear to promote cancer development indirectly. They do so by establishing chronic infections at certain sites in the body and creating a local environment where cells, even uninfected cells, are at greater risk of becoming cancerous.
Basic mechanisms of cell growth transformation
Considering the whole range of viruses known in animals as well as man, only a small number of agents within particular virus families have direct growth-transforming capacity. What are these viruses and how do they work?
(i) Retroviruses are unusual RNA viruses which replicate by converting their genetic information into DNA form (the provirus), integrating this into the DNA of the host cell and producing new copies of the virus' RNA genome using this provirus as a master template. Very occasionally, the DNA provirus may integrate near a 'cellular oncogene' (a growth-promoting gene in the cell's own genome), liberating that gene from its usually tight control and causing it to drive the cell into growth (Figure 1.1 (a)). Such 'chronically oncogenic' viruses are found naturally in some animal species and produce tumours late in life. Very rarely, such viruses develop into 'acutely oncogenic' variants by capturing cellular oncogene sequences into the viral genome itself. These variants, so far only seen in the laboratory, produce tumours much more efficiently because they carry their own oncogene and can express it wherever they integrate in the cell genome ( Figure 1.1 (b)). Yet a third mechanism of retrovirus-induced cell transformation exists (Figure 1.1 (c)) and is described in the context of a human retrovirus human T-cell lymphotropic virus type 1 (HTLV1).
(ii) Certain DNA viruses of the polyoma-, papilloma-, adeno- and herpesvirus families are linked with cancer either in animals or in man. The viruses all possess one or more genes which are used early in the normal infectious cycle and transiently activate cell growth; this is important to these viruses because a transiently 'activated' cell becomes a much better factory for virus replication. Very occasionally, and with the exception of the herpesviruses, the viral genome accidentally integrates into the cell DNA and may do so in such a way that the early, growth-activating genes of the virus are permanently expressed. The normal infectious cycle is interrupted and the cell permanently activated into growth (Figure 1.2 (a)). The viral genome of the cancer-associated herpesviruses is much larger. It can be stably maintained in the cell and express growth-activating latent genes without integration.
At least one other family of DNA viruses linked to cancer, the hepadnaviruses , act through accidental integration of viral DNA fragments into the cell genome. In this case, the integrated viral genes do not necessarily need to be expressed. In some circumstances (the best known example is the woodchuck hepadnavirus), integration can occur close to a cellular oncogene, c-myc; the resulting activation of the gene promotes tumour development. In other cases (for example, with the human hepadnavirus hepatitis B) integration is more random but the acquisition of foreign viral DNA appears to make the cellular genome more susceptible to further genetic accidents, and cancer (Figure 1.2 (b)).
How can we reduce this burden?
The ultimate aim is to develop vaccines against these cancer-causing viruses, and the ideal vaccine is one that will provide sterile immunity by preventing the vaccinated person ever becoming infected with the virus. Such vaccines have already been developed successfully for the hepatitis B virus,3 and two of the high risk HPV types, HPV 16 and HPV 18, for which there is a vaccination programme in the UK.4, 5. In the long run such a vaccine, extended to include other high risk HPV types, has enormous potential to prevent HPV-associated cervical carcinoma. While anti-viral drugs are in short supply, there are highly effective antibiotics against H. pylori and evidence from randomised trials suggests that antibiotic treatment can bring a reduction in gastric cancer risk as well as preventing MALT lymphomas.
These examples are spurring on efforts towards the long-term goal of developing vaccines or anti-viral drugs that can target other cancer-associated viruses, HTLV1, EBV, HHV8 and the hepatitis C virus. Each of these presents their own special challenges, but research continues in an effort to meet these challenges and to further reduce the burden of virus-associated cancers worldwide.
Increase text
Decrease text
Print this page