Viruses are among the proven causes of several cancers and suspected in the etiology of others. Highly sensitive molecular techniques (e.g., polymerase chain reaction [PCR]), which can detect minute amounts of viral nucleic acids in human tumors, are useful in suggesting possible causal associations. Detection of viral DNA or RNA in human tumors, on its face, supports the conclusion that the virus is present in the tumor and may therefore have been instrumental in its development. However, reliance on these methods in recent years has generated controversial results in several instances, and some proposed associations have not withstood the test of time.
There are several reasons why use of these sensitive molecular techniques is insufficient in establishing causal associations between viruses and cancer. First, PCR is widely available but poorly standardized, and the methods sometimes yield false-positive results when applied to tumor tissues. Along these lines, quantitative PCR methods are helpful in distinguishing low-level PCR contamination from biologically relevant infection; in the latter case, one would predict at least one PCR-quantifiable copy of virus per tumor cell. Second, additional laboratory techniques are required to localize viruses within tumor cells, demonstrate expression of key viral oncoproteins, and establish integration into host DNA and clonality of the virus_all of which add important supportive evidence that the virus is present in the tumor and driving proliferation. Third, epidemiologic evidence from population-based studies in humans is required to demonstrate that host infection precedes development of the cancer and that infection is associated with an increased risk of cancer during follow-up. A strong magnitude of association (i.e., a large relative risk relating infection to a sizably increased cancer risk) provides important supportive evidence. These epidemiologic studies require rigorous design and validated measurements of host infection (e.g., assays for systemic antibody production or viremia).
Controversy typically arises when evidence from molecular studies of tumor tissues is not supported by, or actually conflicts, with other lines of evidence. Further controversy can arise from inconsistencies in molecular results within or across laboratories. Several examples will be reviewed, including a recent controversy surrounding the possible role of simian virus 40 in development of several cancers in humans. Investigators must treat initial reports of possible associations between viruses and cancer with caution. Replication of initial findings by independent research groups, and evaluation of the proposed association using several complementary laboratory and epidemiologic approaches, is required to establish a causal relationship.
Citation Information: Cancer Prev Res 2008;1(7 Suppl):CN09-03.
- American Association for Cancer Research