Childhood Cancer Susceptibilty Genes

Genetic Susceptibility Team

Section: Section of Cancer Genetics

Some childhood cancer genes result in familial clustering of cancers or are associated other abnormalities in the child. However, some mutations result in childhood cancer in an otherwise healthy child. Such genes/mutations can be difficult to identify but may increase the risk of adverse effects of therapies or of a second malignancy and are therefore important to recognise. We are using a variety of strategies to identify childhood cancer susceptibility genes, including investigation of unusual childhood cancer cases and high-throughput mutational screening of candidate genes in isolated cases.

Factors Associated with Childhood Tumours (FACT) study

Through collaboration with the United Kingdom Childhood Cancer Study Group (UKCCSG) and the Childhood Cancer Registry (CCRG) we have initiated the FACT study to identify the genetic factors that increase the risk of childhood cancer. We are collecting clinical information and samples from childhood cancer cases throughout the UK. We are focussing on cases in which genetic factors may be acting. This includes cases with more than one cancer, a family history of cancer, medical conditions in addition to cancer and/or excessive treatment toxicity. We use the clinical data to clarify the clinical conditions that are associated with an increased risk of childhood cancer. We use the samples in analyses to identify genes that cause childhood cancer. In the first instance our analyses are focussing on genes involved in DNA repair pathways and childhood overgrowth conditions, as some genes in these pathways are known to be involved in childhood cancer. We will analyse any genes we identify in constitutional DNA from a broad series of childhood tumours to evaluate the prevalence, penetrance and phenotype of mutations.

Familial Wilms tumour genes

Approximately 1-3% of Wilms tumour cases cluster within families. We have collected 45 families from across the world. All have been screened for mutations in WT1, which account for only two families, both of which include urogenital abnormalities in males. One family is due to biallelic BRCA2 mutations. We have mapped a familial Wilms tumour gene, FWT1, to chromosome 17q21 and are undertaking research to identify the causative gene. A second putative familial Wilms tumour gene, FWT2, has been proposed on 19q13. However, heterogeneity linkage analysis in our series does not provide strong evidence for the existence of this gene and suggests that the majority of families are due to unknown genes, which we are trying to identify.

Clinical guidelines for surveillance of individuals at-risk of Wilms tumour

Wilms tumour usually occurs in otherwise healthy children, but a small number occur in children with genetic syndromes. Regular surveillance in children thought to be at increased risk of Wilms tumour has become widespread in UK, USA and parts of Europe. However, the potential risks and benefits of surveillance are finely balanced and there is no clear evidence that screening reduces mortality or morbidity. The rarity of Wilms tumour associated conditions effectively precludes an appropriately powered, prospective, randomised study to evaluate efficacy of surveillance. We therefore formed a working group to review the available evidence and to produce pragmatic recommendations for tumour surveillance in children at risk of Wilms tumour. These recommendations have been broadly supported by Geneticists, Paediatric oncologists and Radiologists and are available in full in the ‘additional information’ section.

Neuroblastoma susceptibility genes

Neuroblastoma is an embryonal tumour originating from neural crest cells and is one of the most common solid tumours of childhood. Recently, constitutional mutations in PHOX2B have been shown to confer an increased risk of neuroblastoma. To date, reported PHOX2B mutations occurred in families with additional clinical features such as Hirschsprung disease or congenital central hypoventilation syndrome, either in the index case or relatives. The contribution of PHOX2B mutations to neuroblastoma cases without additional features is unclear. To address this we sequenced PHOX2B in constitutional DNA from 86 individuals with non-syndromic neuroblastoma and identified two mutations. These mutations are outside the domains typically mutated in PHOX2B syndromes, supporting evidence that different PHOX2B mutations are associated with distinct phenotypes. We have also shown that NSD1 haploinsufficiency is associated with an increased risk of neuroblastoma. In collaboration with Carmel McConville and Andy Pearson we are initiating studies to identify further genes that confer susceptibility to neuroblastoma.