Prevalence of germline mutations in POLE and POLD1 in hereditary cancer

Pilar Mur1,2,3, Sandra García-Mulero4, Jesús del Valle1,2,3, August Vidal5, Marta Pineda1,3,2, Giacomo Cinnirella6,1, Edgar Martín-Ramos7, Sami Belhadj1,2, Pau M. Muñoz-Torres1,2, Matilde Navarro1,2,3, Judith Balmaña8, Joan Brunet9,3,2, Victor Moreno4,10,11, Rebeca Sanz-Pamplona4,10, Rosa Aligué7, Gabriel Capellá1,2,3, Conxi Lázaro1,2,3, Laura Valle1,2,3

1Hereditary Cancer Program, Catalan Institute of Oncology, IDIBELL, Hospitalet de Llobregat, Spain. 2Program in Molecular Mechanisms and Experimental Therapy in Oncology (Oncobell), Hospitalet de Llobregat, Spain. 3Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), Madrid, Spain. 4Unit of Biomarkers and Susceptibility, Cancer Prevention and Control Program, Catalan Institute of Oncology, IDIBELL, Hospitalet de Llobregat, Spain. 5Department of Pathology, Bellvitge University Hospital, IDIBELL, Hospitalet de Llobregat, Spain. 6PhD Program in Translational Biomedicine, University of Catania, Catania, Italy. 7Department of Biomedical Sciences, School of Medicine, University of Barcelona, IDIBAPS, Barcelona, Spain. 8Medical Oncology Department, Valld’Hebron University Hospital, Valld’Hebron Institute of Oncology (VHIO), Barcelona, Spain. 9Hereditary Cancer Program, Catalan Institute of Oncology, IDIBGi, Girona, Spain. 10Centro de Investigación Biomédica en Red de Epidemiologia y Salud Pública (CIBERESP), Madrid, Spain. 11Department of Clinical Sciences, Faculty of Medicine, University of Barcelona, Barcelona, Spain


Objectives: Germline mutations in the exonuclease domain of polymerases POLE and POLD1 predispose to adenomatous polyps, colorectal cancer (CRC), endometrial tumors and other extracolonic malignancies. Tumors and polyps with exonuclease domain mutations exhibit an increased mutation rate (hypermutation) and a specific associated mutational signature. Here we aim to assess the prevalence and tumor spectrum of POLE/D1 pathogenic variants in hereditary cancer.

Methods: POLE and POLD1 genes were sequenced in 2,813 unrelated probands referred for genetic counseling: 2,309 hereditary cancer patients prospectively subjected to a multi-gene panel, and 504 patients without mutations in known cancer-predisposing genes, selected according to their phenotypic characteristics. Co-segregation analyses, case-control allele frequencies, yeast-based assays and tumor mutational analyses (mutation burden and mutational signatures) were performed for variant interpretation.

Results and conclusions: We identified 12 novel/rare missense variants within the exonuclease domain, 6 loss-of-function and 13 missense predicted deleterious variants outside the exonuclease domain. Pending interpretation of the clinical impact of loss-of-function and non-exonuclease domain variants, two variants in the exonuclease domain were classified as likely pathogenic, 4 as (likely) benign, and 6 as variants of unknown significance. In the studied prospective cohort, the frequency of (likely pathogenic or VUS) exonuclease domain variants in POLE/D1 in hereditary cancer was 0.43% (10/2,309), 1% (6/601) when only CRC and polyposis were considered. The distribution of variants among different syndromes and the challenges for variant interpretation will be shown, both relevant for routine genetic diagnostics.