Human Cancer Genetics Programme

Human Genetics Group

Group Leader:  Javier Benítez
Research highlights
Breast cancer: PARP1 and OGG1 inhibitors in BRCA1 mutation carriers

We have demonstrated that certain missense mutations in BRCA1 seem to make cells more sensitive to Poly (ADP-ribose) Polymerase (PARP) inhibitors than those mutations that give rise to the absence of the protein (frameshift mutations) (T. Valclová, Hum Mol Genet 2016). We are currently investigating the mechanisms underlying these differences with the aim of identifying new markers of sensitivity or resistance to these agents.

In parallel, we recently showed that the Single Nucleotide Polymorphism (SNP) rs2304277, located in the 3’ untranslated region (UTR) of the OGG1 DNA glycosylase gene of the Base Excision Repair pathway (BER), modified cancer risk in patients harbouring mutations in BRCA1 (Osorio A. et al., Plos Genetics, 2014). We have identified that the SNP is associated with a constitutive hOGG1 transcriptional downregulation, which leads to a high genome and telomere instability in those patients harbouring BRCA1 and BRCA2 mutations, thereby explaining the contribution of this polymorphism to cancer risk. This association is most likely explained by a synthetic lethal/sick interaction between these 2 genetic events. (Benitez-Buelga C. et al., Oncotarget, 2016). In order to take an in-depth look at the biological link between BER and the homologous recombination (HR) DNA repair pathway, we tested the pharmacological inhibition of OGG1 in a set of BRCA1 and BRCA2 deficient cancer cell lines. We found that OGG1 inhibition is effective, leading to 1)an accumulation of telomere oxidation (genomic instability), and 2) an alteration in the normal proliferation of BRCA1 deficient cell lines, pointing to a synthetic lethal interaction between OGG1 and BRCA1 (FIGURE 1).

Familial cancer exome project

This project started several years ago with the objective of identifying new high susceptibility genes that explain families with rare tumours as well as deciphering the genetic heterogeneity present in some of them:

  • In 2015, we identified ATP4a as being responsible for families with gastric neuroendocrine tumours. We are currently searching for new genes in two families that cannot be explained by mutations in ATP4a.
  • A second gene, POT1, which was published in 2015 as being associated with familial cardiac angiosarcoma, also explains some families with Li Fraumeni-like syndrome. Analysis of tumour samples by Next Generation Sequencing (NGS) has shown an over-representation of the angiogenic pathway, which may be useful in clinical trials. In collaboration with M. Blasco’s Telomeres and Telomerase Group, we are generating a knock-in mouse with the aim of recapitulating the disease, as well as enabling us to work with antiangiogenic drugs.
  • We are currently investigating a large family with meningiomas across 3 generations. Analysis of the data generated using bioinformatics tools has shown the existence of 2 candidate genes that could be responsible for this familial tumour. We are starting functional studies and are recruiting more families.
  • Ovarian cancer families are rare and are usually associated with breast cancer. We sequenced the exomes of 9 patients from 5 families and identified 33 rare variants in 28 genes potentially implicated in ovarian cancer risk. By conducting a case control association study we narrowed down the number of candidate missense variants to 10. These, together with 5 high-impact variants (protein truncating or splicing variants), will be evaluated in a larger international case control study to finally define their role in ovarian cancer susceptibility. In parallel, we selected a non-described RAD51C missense variant among the identified candidates, and through functional characterisation we were able to determine its pathogenicity and its probable involvement in ovarian cancer risk in one of the families. This finding not only has implications for genetic counselling but also for the potential treatment of affected carriers with PARPi.
  • Breast cancer. We have performed whole-exome sequencing (WES) in 3 BRCAX families (familial breast cancer families negative for mutations in BRCA1/2). One of the families was found to harbour a deleterious mutation in the known breast cancer susceptibility gene ATM. A complete screening of this gene in a set of 400 Spanish breast cancer families showed a prevalence of almost 2% of mutations in ATM, higher than that reported in other populations (Tavera-Tapia et al., BCRT 2016). In another family, we found an excellent candidate breast cancer gene that is currently being screened by targeted NGS in a series of 700 BRCAX families and 700 controls. The third family is still under analysis.
  • Male breast cancer. We performed WES in a male breast cancer family with an apparently recessive model of inheritance. We have found 7 candidate variants that are currently being validated in a series of 1000 male breast cancer cases and 1000 controls; this is undertaken in collaboration with Nick Orr’s lab at the Institute of Cancer Research in London.
  • Testicular cancer. Testicular cancer follows a polygenic model of inheritance. We have studied, by NGS, 35 families with 2 or 3 first degree relatives affected by the disease. The results have been classified according to different inheritance models; different methods of analysis have been conducted in order to select some candidate genes (FIGURE 2 . The candidate variants are currently being genotyped in a set of more than 500 sporadic testicular cancer cases and 500 controls in order to know how many of them could be considered as candidates to be associated with the disease.