An infectious aetiology for childhood brain tumours? Evidence from space-time clustering and seasonality analyses.

To investigate whether infections or other environmental exposures may be involved in the aetiology of childhood central nervous system tumours, we have analysed for space-time clustering and seasonality using population-based data from the North West of England for the period 1954 to 1998. Knox tests for space-time interactions between cases were applied with fixed thresholds of close in space, <5 km, and close in time, <1 year apart. Addresses at birth and diagnosis were used. Tests were repeated replacing geographical distance with distance to the Nth nearest neighbour. N was chosen such that the mean distance was 5 km. Data were also examined by a second order procedure based on K-functions. Tests for heterogeneity and Edwards' test for sinusoidal variation were applied to examine changes of incidence with month of birth or diagnosis. There was strong evidence of space-time clustering, particularly involving cases of astrocytoma and ependymoma. Analyses of seasonal variation showed excesses of cases born in the late Autumn or Winter. Results are consistent with a role for infections in a proportion of cases from these diagnostic groups. Further studies are needed to identify putative infectious agents.

DNA sequences similar to those of simian virus 40 in ependymomas and choroid plexus tumors of childhood.

BACKGROUND: Ependymomas and papillomas of the choroid plexus occur in early childhood. The ubiquitous human polyomaviruses, BK virus and JC virus, have been associated with the induction of these neoplasms in animal models. A related monkey polyomavirus, simian virus 40 (SV40), is highly tumorigenic in rodents and also induces choroid plexus papillomas. METHODS: We tested the possibility that polyomaviruses were associated with these tumors in humans. Tumors from 31 children--20 with choroid plexus neoplasms and 11 with ependymomas--were evaluated for the presence of polyomavirus T-antigen gene sequences by means of amplification with the polymerase chain reaction. RESULTS: Ten of the 20 choroid plexus tumors and 10 of the 11 ependymomas contained amplification products that preferentially hybridized to probes specific for SV40 viral DNA rather than BK or JC viral DNA. In two specimens, DNA sequencing demonstrated that the amplified sequence was identical to the sequence of that region of the SV40 gene. In three other specimens, amplification with SV40-specific primers revealed a 574-bp segment of the SV40 viral gene. In 7 of 11 tumors examined by immunohistochemical staining, viral T antigen was expressed in the nuclei of the neoplastic cells. CONCLUSIONS: Half of the choroid plexus tumors and most of the ependymomas that we studied contained and expressed a segment of T-antigen gene related to SV40. These results suggest that SV40 or a closely related virus may have an etiologic role in the development of these neoplasms during childhood, as in animal models.

Lack of association between BK virus and ependymomas, malignant tumors of pancreatic islets, osteosarcomas and other human tumors.

BK virus (BKV) DNA sequences were not detected in 142 human tumors analyzed by DNA-DNA reassociation kinetics and blot hybridization. The investigation was focused mainly on those rare types of human tumors (ependymomas, choroid plexus papillomas, tumors of pancreatic islets and osteosarcomas) that are induced with highest frequency by BKV in experimental animals. In addition, other tumors of the urinary apparatus and of the central nervous system were analyzed. BKV tumor (T) antigen was not detected in neoplastic tissues, and BKV T antibodies were not found in sera and cerebrospinal fluids from patients with neoplasms. Sequences homologous to BKV DNA were found in normal tissue from a kidney carrying a carcinoma. The neoplastic tissue from the same organ, however, had no sequences homologous to BKV DNA. Such DNA does not belong to BKV but probably to another papovavirus related to BKV.

Integrated and free viral DNA in hamster tumors induced by BK virus.

BK virus (BKV)-induced tumors in hamsters were investigated for the presence of viral DNA by the blot-transfer hybridization technique. Several viral genomes per cell were found in tumor tissues and in their derived cell lines and clones. Most of the detected viral genomes were integrated into the cellular DNA, but some tumors also contained free viral DNA sequences. Integration patterns were different from each other, and many different integration sites were available on the cellular or on the viral DNA or on both. Typical features of integration patterns were found in ependymomas, which were the most frequent (72%) among BKV-induced tumors. Readily detectable viral DNA sequences were only found in neoplastic tissues, but traces of BKV DNA were also present in the apparently normal portion of the brain of an animal that had developed an ependymoma and in the brain (but not in the liver) of another animal 15 days after virus inoculation. A cell line and a single-cell clone derived from a tumor had hybridization patterns considerably simpler that the pattern of the original tumor, lacking several integrated viral genomes and all free viral sequences.

Induction of papillary ependymomas and insulinomas in the Syrian golden hamster by BK virus, a human papovavirus.

Newborn hamsters were inoculated intracerebrally with BK virus. Between 3 and 6 months after inoculation, they developed papillary ependymoma (8 hamsters, 42%) or functional malignant islet cell tumors of the pancreas (insulinoma, 8 hamsters), or both (1 hamster). Both tumors contained an antigen reactive to SV40 T-antibody, suggesting that at least a part of BK virus genomen has been integrated into the tumor cells. No infectious virus was detected in the extract of these tumors.

Particles with RNA of high molecular weight and RNA-directed DNA polymerase in human brain tumors.

We have previously shown that neoplastic cells of human breast cancers, leukemias, lymphomas, and sarcomas contain particles similar to the viruses that have been established as etiologic agents of these diseases in mice. The present paper concerns tumors of the central nervous system for which no suitable animal model or corresponding virus exists. Nevertheless, using the simultaneous detection test, we showed that human brain tumors contain 70S RNA and RNA-directed DNA polymerase encapsulated in a particulate component possessing a density of 1.17 g/ml. These particles satisfy the three diagnostic criteria that characterize RNA tumor viruses of animals. 24 Out of 26 (92%) of the most malignant (glioblastoma and medulloblastoma) brain tumors examined contained these virus-like entities.