Oncology and Haematology.(Statistical Data Included)

G65 INVESTIGATION OF MYC GENE AMPLIFICATION IN PAEDIATRIC BRAIN TUMOURS

C.D. Rajapakse, S. Ward, P. Finn, T.J. Warr. Department of Neurological Surgery, Institute of Neurology, Queen Square, London, UK

Background: Brain tumours are the second most common malignancy in childhood, and carry a poor prognosis. Identification of tumour-specific genetic changes may provide insights into novel therapies. Myc oncogenes are involved in regulation of gene transcription, and activation of these genes has been implicated in tumourigenesis.

Aims: To demonstrate amplification of the C-myc and N-myc oncogenes, located on 8q and 2p respectively in 14 paediatric brain tumours.

Methods: Four high grade astrocytomas, seven ependymomas and three medulloblastomas were selected as demonstrating gains at Sq and/or 2p by Comparative Genomic Hybridisation analysis. Interphase Fluorescence In Situ Hybridisation utilising markers for N-myc and centromere 2 and C-myc and centromere 8 was used to determine amplification and identify the copy number of these oncogenes.

Results: Ten tumours demonstrated amplification of the myc oncogenes. Amplification of the N-myc gene alone occurred in 3 ependymomas. Amplification of the C-myc gene alone occurred in 1 high grade astrocytoma, 1 medulloblastoma and 3 ependymomas. In 2 high grade astrocytomas there was amplification of both the C-myc and N-myc oncogenes. Five tumours demonstrated sub-populations of cells with triploidy of chromosome 2, and 6 tumours demonstrated triploidy of chromosome 8. One tumour (a medulloblastoma) demonstrated an increased copy number of chromosome 8 without amplification of C-myc.

Conclusions: N-myc and C-myc amplifications occur in high grade astrocytomas, medulloblastomas and ependymomas in the paediatric population. The pattern of amplification varied within and between the tumours studied. Further study is needed to correlate these tumour-specific genetic changes with phenotype.

G66 STATISTICAL ANALYSES PROVIDE SUPPORT FOR TWO INFECTIOUS MECHANISMS IN THE AETIOLOGY OF CHILDHOOD BRAIN TUMOURS

R.J.Q. McNally, O.B. Eden, F.E. Alexander, A.M. Kelsey, J.M. Birch. CRC Poediarric and Familial Cancer Research Group, Royal Manchesrer Children's Hospital, UK

Aims: (i) To use Manchester Children's Tumour Registry data (1954-1998) to test predictions of space-time clustering patterns which might arise as a result of environmental exposures. (ii) To distinguish between hypotheses relating to post-natal exposures or pre-natal events by using locations and dates of birth as well as at diagnosis.

Methods: Knox tests for space-time interactions between cases were applied with fixed thresholds of close in space, [less than]5km, and close in time, [less than]1 year apart. Both places and times of birth and diagnosis were utilised. Tests were repeated replacing geographical distance with distance to the Nth nearest neighbour (Nearest Neighbour [NN] threshold analysis). N was chosen such that the mean distance was 5km. Data were also examined by a second order procedure based on K-functions. Analyses were applied to 15 diagnostic subgroups.

Results: Diagnostic subgroups involving astrocytoma and ependymoma showed significant evidence of space-time clustering at place and time of diagnosis. Two diagnostic sub-groups involving pilocytic astrocytoma and ependymoma showed significant evidence of spacetime clustering at place and time of birth.

Conclusions: The results are consistent with a role for infections or other localised environmental causes involving two different mechanisms, one acting around the time of birth, with a variable latent period, for cases of pilocytic astrocytoma and ependymoma, and the other acting around the time of diagnosis, with a short constant latent period, for older cases (aged 5-14 years) of astrocytoma and ependymoma.

G67 GEOGRAPHIC MOBILITY FOLLOWING CANCER TREATMENT IN YORKSHIRE

R.G. Feltbower[1], R.C. Parslow[1], A.W Glaser[2].

(1.)Paediatric Epidemiology Group, Unit of Epidemiology & Health Services Research, University of Leeds, LS2 9LN;

(2.)Paediatric Oncology & Haematology, Children's Day Hospital, St James's University Hospital, Leeds LS9 7TF UK

Maximising re-integration into society is a major goal of childhood cancer units. Geographic mobility has been used as an objective proxy for this and here we present the first population-based study.

Data was extracted from …