Simian virus 40 and human malignancy: contamination of early polio vaccine may be linked to rare tumours

British Medical Journal, March 21, 1998 by S.C. Stenton

The introduction of the Salk parenteral vaccine in the mid-1950s led to a dramatic decline in the incidence of poliomyelitis. By 1961, the majority of young adults in Britain and America had been immunised and the numbers of reported cases of poliomyelitis had fallen from 8000 a year to 100 a year.[1] At that point in the mass immunisation programme, a contaminating virus was identified in the rhesus monkey kidney cells that were used to culture the poliovirus. It was named simian virus 40 (SV40). It was more resistant than poliovirus to chemical denaturation and survived into some vaccine samples. There are no reliable data about the proportion of batches that were contaminated with live SV40, and estimates range up to 30%.[2] Early worries that the contaminant might be implicated in the development of human cancers have recently resurfaced.

SV40 was characterised as a double stranded DNA virus belonging to the group of papovaviruses. They share with adenoviruses (another DNA virus) a potent ability to induce tumours in species that are not their natural hosts. SV40 itself was found to be highly oncogenic in hamsters shortly after it was identified, and epidemiological surveillance of immunised cohorts was begun.[3] Except for one study, which reported an increased incidence of neural tumours in children of mothers vaccinated during pregnancy, all studies were essentially negative.[4] Occasional cases were reported of SV40 infection in association with tumours, but until recently the view was that SV40 has no role in the pathogenesis of human malignancy.

SV40 has now re-emerged as a potentially oncogenic virus. In 1992 Bergsagel et al used polymerase chain reaction techniques to search for DNA from human polyomaviruses, which are usually asymptomatic, in childhood ependymomas and choroid plexus tumours. They identified DNA which more closely matched that of SV40.[5] Since then SV40-like DNA has been identified in other human tumours, particularly osteosarcomas and malignant mesotheliomas though not in adenocarcinomas.[6,7] These findings mirror the range of tumours induced by SV40 in animals: injection of SV40 into hamsters results in lymphoid tumours and osteosarcomas, SV40 transgenic mice develop choroid plexus tumours, and intrapleural SV40 seems more potent than asbestos in inducing mesotheliomas.

DNA viruses such as SV40 carry only a limited amount of genetic information, and in order to reproduce they must subvert normal cellular DNA replication. This process is facilitated by viral proteins that inactivate products from cellular tumour suppressor genes. These products normally have inhibitory effects on DNA replication, and if their function is impaired this can contribute to the escape from replicative control that is an important step in the development of malignancy. When viruses enter cells which do not support their replication their DNA can become incorporated into the host genome, allowing inhibitors of tumour suppressor genes to be produced. The SV40-like DNA found in human tumours codes for the large T antigen, which inactivates the products of tumour suppressor genes.[8] The T antigen is structurally similar to the e7 and e8 antigens of the papillomaviruses, which are now recognised as important in the aetiology of cervical cancer.[9]

The identification of virus-like DNA in tumours, the studies in animals, and the molecular actions of SV40 all suggest that it might have a role in some human malignancies. Epidemiological studies make it unlikely that the virus plays an important part in the aetiology of common cancers, but there are few other examples of known human oncogenic viruses and if the findings are confirmed they would be of considerable importance. For the present, however, we must remain cautious. The polymerase chain reaction techniques used to identify the viral DNA from fixed specimens are poorly standardised, and SV40 is a commonly used laboratory virus which might contaminate assay systems. No large scale studies have been undertaken, control tissue has often been inadequate, and the findings have not been replicated in all laboratories.[10]

Even if the identity of the DNA is confirmed as viral in origin, its source would remain unclear as SV40-like DNA has been identified in tumours from those who are far too young to have been immunised with contaminated vaccines. If this cannot be explained by artefact or misidentification then it implies either some other source of human SV40 infection or vertical transmission from immunised subjects. It thus remains possible that a late adverse effect of the polio vaccination programme is emerging, although any risk of cancer is likely to be more than outweighed by the benefit of vaccination to the postwar generation. Indeed, if it leads to an improved understanding of tumour biology it might even result in a treatment for tumours such as mesotheliomas, which to date have proved depressingly resistant to treatment.

[1] Salk J. Salk D. Vaccination against poliomyelitis. In: Voller A, Freedman H, eds. New trends and developments in vaccines. Lancaster: MTP Press 1978:138-9