Reviewers: John Han-Chih Chang, MD and Kenneth Blank, MD
Source: Journal of the American Medical Association 1998; Volume 279 (4): pages 292 - 295.
Simian Virus 40 (SV40) has been implicated to have tumorgenic effects in rodent models. DNA sequences similar to SV40's have been extracted from neoplastic cells of ependymoma, osteosarcoma and mesothelioma patients. The relevance this poses stems from the fact earlier polio vaccines were contaminated with SV40. The earlier polio vaccines were created from simian (monkey) kidney cell cultures, and despite formalin inactivation there were various titers of live SV40. An earlier study from Germany revealed no significant increase in the rate of malignancies in the patient population exposed. The follow up was greater than 20 years and the number of patients greater than 800,000. This study attempted to make a statement regarding cancer incidence in those in the United States that were at risk for exposure to the SV40 contaminated vaccine.
The comparison groups were as follows: 1 - infants at high risk for exposure (born 1956 - 1962), 2 - children at high risk for exposure (born 1947 - 1952) and 3 - unexposed individuals (born 1964 - 1969). Data was obtained from the Surveillance, Epidemiology, and End Results (SEER) program of the National Cancer Institute (NCI), the Connecticut Tumor Registry, the National Center for Health Statistics and the US Bureau of the Census. The malignancies evaluated were ependymoma, osteosarcoma, mesothelioma and brain cancer.
In the article, figure 1 displayed the incidences of each malignancy. Essentially, there were no significant differences between the comparison groups evaluated in incidence rates. The differences were noted as relative risk (RR) compared to those unexposed (used as the standard or RR = 1) except for brain cancer, which was evaluated as mortality from disease per person-years of observation.
For ependymomas, the RR's were 1.06 and 0.98 in groups 1 and 2 (p = 0.91), respectively. In osteosarcomas, groups 1 and 2's RR were 0.87 and 0.85 (p = 0.35). Mesothelioma's RR in groups 1 and 2 were 3.00 and 2.45 (p = 0.23). Group 1 and 2 had respective brain cancer mortality rates of 1.27 and 2.04 per 100,000 person-years of observation compared to group 3, which had a lower rate of 1.04 per 100,000 person-years of observation.
Amist all the reports stating that SV40 DNA has been found in osteosarcomas, mesotheliomas, ependymomas and other brain cancers, a cause for concern has erupted in patients previously exposed to polio vaccines contaminated with SV40. This cohort study reports that no difference can be found in incidence rates. It is difficult to make determinations of risk based on this study for a number of reasons. SEER and the Connecticut Tumor Registry all have only a small localized area which they survey. Thus, bias are built in the medical care, work-up and reporting which are inherent to that region. A major complicating factor in interpreting the results is that it is unknown how many of the patients reported on by the study had vaccination. For that matter, the amount of SV40 viral titers were not known. The cohort of patients in the were all-comers, vaccinated or not. No difference was seen in the ependymomas or osteosarcoma incidences, but mesothelioma incidence and brain cancer related mortality seemed to be influenced by "possible" exposure to SV40, without significant "p" values. As alluded to by the authors,this may be related to increased follow-up in the "exposed" patients relative to the unexposed. Median follow-up was not documented.
Based on the JAMA article, there appears to be no significant effect of "exposure" to SV40 in increasing malignancy incidence, but proof of significant exposure is lacking. Documentation of the viral titers and identification of which patients were actually exposed would lend more credence to the contention. SV40's role as pathogen in animal models is documented. Based on these data, increased vigilance must be maintain in this patient population.