Pediatric cancer predisposition: improving effective recognition and care

Abstract

In the Netherlands, approximately 600 children are newly diagnosed with cancer each year. The increased awareness of genetic predisposition has been driven by the development of questionnaires and eHealth tools that help clinicians select children who are more likely to have an underlying genetic condition, as well as the introduction of paired tumor-normal next-generation sequencing (NGS). This technological advancement has made whole exome and whole genome germline data from many children with cancer available. However, the diagnostic value of extensive CPS gene sequencing among all children with cancer when implemented in routine care is unknown. The research presented in Part I of this thesis addresses this gap. Chapter 2 describes the experience of a national children’s cancer center in diagnosing CPSs before the implementation of NGS. Chapter 3 presents a prospective comparison between clinical selection-based genetic testing and extensive phenotype-agnostic germline sequencing in an unselected cohort of children with cancer. Over two years, CPS gene sequencing was offered to all newly diagnosed children in addition to standard phenotype-driven testing. This study shows that both approaches are complementary, with various syndromes detected only by one method. The phenotype-driven approach identified most CPSs with established causal associations with the child’s neoplasm type. CPS gene panel sequencing revealed additional CPSs, many of which had uncertain causality in relation to the child’s neoplasm, but some with clinical utility. Chapter 4 reports on the establishment of the precision medicine program iTHER in the Netherlands for children and adolescents with primary very high-risk, relapsed, or refractory pediatric tumours. Part II of this thesis is about DICER1 syndrome, a genetic disorder that predisposes individuals to a wide spectrum of tumors. Chapter 5 reports on the first consecutive series of DICER1 mutation testing in children previously diagnosed with CPAM. In this cohort, no pathogenic DICER1 variants were identified, suggesting that CPAM is not associated with DICER1. To improve care for individuals with DICER1 syndrome, surveillance protocols are needed. Chapter 6 presents a DICER1 surveillance protocol developed through a literature review and consensus meetings with representatives of the SIOPE Host Genome Working Group (HGWG) and CanGene-CanVar Clinical Guideline Working Group. In Chapter 7, the results of the work presented in this thesis are discussed, and suggestions on how to proceed in the field of pediatric cancer predisposition are provided. Based on the findings of this thesis, I propose optimizing germline genetic care in children with cancer by using a stepwise approach: i) a limited actionable gene panel at the time of cancer diagnosis, ii) clinical evaluation to select patients for further genetic testing, and iii) options for extensive germline genetic testing in a research setting.I expect that the continued increase of available molecular tumor and germline data from children with cancer, combined with clinical phenotypic information and improved clinical selection tools will offer many new opportunities for unraveling the etiology of childhood cancer in the coming years. The challenge remains to translate this into effective care for the child and family.