Genetic factors in childhood cancer. Associations between tumors in childhood and adulthood, and prevalence of germline TP53 mutations

University dissertation from Department of Oncology, Clinical Sciences, Lund University

Abstract: The etiology of childhood cancer is largely unknown. Approximately 1-10% of all childhood tumors are associated with known cancer predisposition syndromes. However, the contribution may be underestimated due to the failure to detect patients with genetic susceptibility for cancer when relying on known family pattern and anomalies. Growing evidence indicates that patients with genetic susceptibility to cancer may be at higher than normal risk for therapy related cancers. Increased knowledge regarding the importance of hereditary factors in the development of childhood tumors may improve the medical care of such patients by identifying those in need of more individualized treatment. In this thesis, genetic factors, familial cancers, and their associations with childhood cancer have been studied. The general aim was to investigate the importance of hereditary factors in the etiology of childhood cancer and to evaluate possible associations between childhood and adult cancers.

In paper I, a national registry-based cohort of parous women with breast cancer was used to study whether the occurrence of childhood cancer in children affects the survival of mothers with breast cancer. Women who had a child with childhood cancer were found to have shorter survival compared to other parous patients with breast cancer, suggesting that hereditary factors may affect prognosis.

In paper II, the occurrence of childhood cancer in families with hereditary adult cancer syndromes was investigated. Families with BRCA2-associated hereditary breast and ovarian cancer, mismatch repair gene-associated hereditary non-polyposis colorectal cancer and CDKN2A-associated malignant melanoma were found to have a higher occurrence of childhood cancer compared to the general population. No increased occurrence of childhood cancer was found in families with BRCA1-associated breast and ovarian cancers.

In paper III, the incidence of childhood and adult cancer was evaluated in the extended families of patients with childhood cancer, and the frequency of germline TP53 mutations in families with multiple childhood tumors was investigated. The relatives of patients with childhood cancer were found to have an increased incidence of childhood and adult cancers, particularly of the breast and prostate. Breast and prostate cancers were observed at earlier than average ages. No germline TP53 mutations were found in families with multiple childhood tumors, which exclude TP53 mutations as a contributor to the observed excess of childhood tumors.

In paper IV, a population-based material was used to confirm the prevalence of germline TP53 mutations in children with adrenocortical tumors, choroid plexus tumors and early childhood rhabdomyosarcomas and investigate whether these may be early manifestations of Li-Fraumeni syndrome (LFS). Germline TP53 mutations were found in few children with adrenocortical tumors and rhabdomyosarcomas. No mutations were found in children with choroid plexus tumors. Furthermore, neither the family history nor the observed tumor spectra in the relatives of most children with these rare tumors were suggestive of LFS. These data suggest that most children, particularly those with choroid plexus tumors or rhabdomyosarcomas, do not present early manifestations of LFS. Nevertheless, an increased cancer incidence, particularly for certain adult tumors, was found in the relatives of children with choroid plexus tumors and rhabdomyosarcomas, which suggests that other syndromes or predisposing factors may exist.

In summary, this thesis adds new data suggesting that hereditary factors play a role in the development of childhood tumors. In addition, these factors may also increase the risk for adult tumors, modify the onset age of common adult tumors, and affect breast cancer prognosis. Our findings further support the need for future studies regarding the importance of genetic susceptibility to childhood cancer, particularly in families with multiple childhood tumors. Also the associations between tumors of childhood and adulthood in the same family should be further studied.