LAUSR

BACKGROUND Understanding the biological differences between sexes in cancer is essential for personalized treatment and prevention. We hypothesized that the extreme down-regulation of chromosome Y gene expression (EDY) is a signature of cancer risk in men and the functional mediator of the reported association between the mosaic loss of chromosome Y (LOY) and cancer. METHODS We advanced a method to measure EDY from transcriptomic data. We studied EDY across 47 nondiseased tissues from the GTEx project (n = 371) and its association with cancer status across 12 cancer studies from the Cancer Genome Atlas (TCGA) (n = 1,774), and seven other studies (n = 7,562). Associations of EDY with cancer status and presence of loss-off function mutations in chromosome X were tested with logistic regression models, while a Fisher's test was used to assess genome-wide association of EDY with the proportion of copy number gains. All statistical tests were two-sided. RESULTS EDY was likely to occur in multiple nondiseased tissues (P<0.001) and statistically significantly associated with the EGFR tyrosine kinase inhibitor resistance pathway (FDR=0.028). EDY strongly associated with cancer risk in men (OR = 3.66, 95%CI = 1.58, 8.46, P = 0.002), adjusted by LOY and age, and its variability was largely explained by several genes of the non-recombinant region (NRY) whose chromosome X homologs showed loss-of-function mutations that co-occurred with EDY during cancer (OR = 2.82, 95%CI = 1.32, 6.01, P = 0.007). EDY associated with a high proportion of EGFR amplifications (OR = 5.64, 95%CI = 3.70, 8.59, FDR<0.001) and EGFR overexpression, along with SRY hypomethylation and NRY hypermethylation, indicating alternative causes of EDY in cancer other than LOY. EDY associations were independently validated for different cancers and exposure to smoking, and its status was accurately predicted from individual methylation patterns. CONCLUSIONS EDY is a male-specific signature of cancer susceptibility that supports the escape from X-inactivation tumor suppressor hypothesis for genes that protect females with respect to males from cancer risk.