Identifying individuals with primary central nervous system tumors at risk for hereditary cancer syndromes using the Utah Population Database.

Abstract

10532 Background: CNS tumors are the most common solid tumors and the deadliest cancers in children. Approximately 10% of children with a CNS tumor harbor a hereditary cancer syndrome (HCS), but many will not be tested for a HCS. The Utah Population Database (UPDB) contains comprehensive cancer registry data for Utah families and can determine multigenerational cancer pedigrees across an archive of 5.8 million individuals. Early identification of HCSs results in improved cancer surveillance and outcomes, reducing the impact of CNS tumors in children. We hypothesize that the UPDB can identify children and families with HCSs not previously identified. Methods: We queried the UPDB for individuals ages 0-39 diagnosed with a primary CNS tumor (malignant and benign) between 1966-2017 and generated cancer pedigrees of 3 generations or more for probands, extending to at least third-degree relatives. Specialized software calculated a familial standardized incidence ratio (FSIR) to determine families with excess clustering of CNS tumors. Clinical cancer genetics experts reviewed pedigrees to confirm patterns of HCS. Results: We identified 4,634 CNS tumors in 4,550 individuals, of whom 2,233 (49%) reside in high-quality pedigrees containing ≥2 grandparents, at least 1 from both maternal and paternal sides. To identify families with excess clustering of CNS tumors, we selected pedigrees with an FSIR P< 0.05 and ≥2 affected patients, resulting in 161 high-risk families with a mean of 170 (median 96) relatives per pedigree of 3-6 generations. Among these 161 families, there were 2,017 unique relatives (first-third degree) of CNS probands with 2,355 tumors (any site), for a per pedigree average of 14.7 tumors in 12.5 relatives. Review of the 10 highest risk pedigrees indicated that 4 meet HCS criteria, including Li-Fraumeni (n = 2), von Hippel-Lindau (n = 1), and rhabdoid tumor predisposition (n = 1). Conclusions: The UPDB can produce multigenerational cancer pedigrees that identify individuals and families at risk of harboring a HCS who warrant germline testing. These findings should encourage clinicians to perform thorough family history screening and to always consider workup for associated HCSs.