Back

Combining genome-wide studies of breast, prostate, ovarian and endometrial cancers maps cross-cancer susceptibility loci and identifies new genetic associations

Kar, S. P.; Lindstroem, S.; Hung, R.; Lawrenson, K.; Schmidt, M.; O'Mara, T.; Glubb, D.; Tyrer, J.; Schildkraut, J.; Chang-Claude, J.; Alsulimani, A.; Anton, F.; Beeghly-Fadiel, A.; Bjorge, L.; Bodelon, C.; Brauch, H.; Burghaus, S.; Campa, D.; Carney, M.; Chen, Z.; Daly, M.; du Bois, A.; Ekici, A.; Ewing, A.; Fasching, P.; Flanagan, J. M.; Gawelko, J.; Giles, G.; Hamilton, R.; Harris, H.; Heitz, F.; Hildebrandt, M.; Hillemanns, P.; Huang, R.-Y.; Imaz, L.; Irmejs, A.; Jakubowska, A.; Jensen, A.; John, E.; Kannisto, P.; Karlan, B.; Khusnutdinova, E.; Kiemeney, L.; Kjaer, S.; Klapdor, R.; Kleiblo

2020-06-19 genetics
10.1101/2020.06.16.146803 bioRxiv
Show abstract

We report a meta-analysis of breast, prostate, ovarian, and endometrial cancer genome-wide association data (effective sample size: 237,483 cases/317,006 controls). This identified 465 independent lead variants (P<5x10-8) across 192 genomic regions. Four lead variants were >1Mb from previously identified risk loci for the four cancers and an additional 23 lead variant-cancer associations were novel for one of the cancers. Bayesian models supported pleiotropic effects involving at least two cancers at 222/465 lead variants in 118/192 regions. Gene-level association analysis identified 13 shared susceptibility genes (P<2.6x10-6) in 13 regions not previously implicated in any of the four cancers and not uncovered by our variant-level meta-analysis. Several lead variants had opposite effects across cancers, including a cluster of such variants in the TP53 pathway. Fifty-four lead variants were associated with blood cell traits and suggested genetic overlaps with clonal hematopoiesis. Our study highlights the remarkable pervasiveness of pleiotropy across hormone-related cancers, further illuminating their shared genetic and mechanistic origins at variant- and gene-level resolution.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.