Association of Y-chromosomal gr/gr deletions with testicular germ cell tumor: whole-genome analysis of 198,306 individuals
Choi, S.; Rocca, M. S.; Vinanzi, C.; Pluta, J.; Kuzbari, Z.; Loveday, C.; Allen, S.; Torr, B.; Weathers, B.; Anson-Cartwright, L.; Feldman, D. R.; Gietema, J. A.; Gonzalez-Neira, A.; Hamilton, R. J.; Krausz, C.; Moirano, G.; Nead, K. T.; Nsengimana, J.; Poynter, J. N.; Vaughn, D. J.; Kanetsky, P. A.; Nathanson, K. L.; Ferlin, A.; Turnbull, C.; Rowlands, C. F.
Show abstract
PurposeGermline deletions affecting the Y-chromosomal gr/gr region were reported in 2005 as associated with susceptibility to testicular germ cell tumor (TGCT), a highly heritable tumor type that is the most common cancer type affecting adult men under the age of 45. Attempts to replicate this association have been equivocal, primarily due to limited power. MethodsHere, we compare and validate two computational approaches to gr/gr deletion calling in high-, low- and ultra-low-coverage whole genome sequencing data, applying these to two datasets from UK Biobank and the TECAC consortium. We generate dataset-specific effect size estimates for the gr/gr deletion-TGCT association using Firths bias-reduced logistic regression across a total of 198,306 men of European-like ancestry (2231 with and 196,075 without TGCT). ResultsUpon random-effects meta-analysis of estimated effect sizes in the two datasets, we found no significant association between gr/gr deletion status and TGCT risk (combined odds ratio=1.24, 95% CI=0.74-2.07, p=0.42), nor upon stratification of seminoma and non-seminoma/mixed histological subtypes. ConclusionOur analysis suggests gr/gr deletion status alone is likely not predictive of TGCT risk in population-scale analyses of European-like individuals; however, the importance of other proposed determinants of gr/gr deletion impact, including Y-haplogroups and semen phenotype, remains unexplored at scale.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.