Back

Trans-ancestral genetic study of diabetes mellitus risk in survivors of childhood cancer: a report from the St. Jude Lifetime Cohort and the Childhood Cancer Survivor Study

Im, C.; Neupane, A.; Baedke, J. L.; Delaney, A.; Dixon, S. B.; Chow, E. J.; Mostoufi Moab, S.; Richard, M. A.; Gramatges, M. M.; Lupo, P. J.; Sharafeldin, N.; Bhatia, S.; Armstrong, G. T.; Hudson, M. M.; Ness, K. K.; Robison, L. L.; Yasui, Y.; Wilson, C. L.; Sapkota, Y.

2023-06-05 genetic and genomic medicine
10.1101/2023.06.02.23290868 medRxiv
Show abstract

Type 2 diabetes mellitus (T2D) is an established late effect of treatment for childhood cancer. Leveraging detailed cancer treatment and whole-genome sequencing data among survivors of childhood cancer of European (EUR) and African (AFR) genetic ancestry in the St. Jude Lifetime Cohort (N=3,676; 304 cases), five novel diabetes mellitus (DM) risk loci were identified with independent trans-/within-ancestry replication, including in 5,965 survivors of the Childhood Cancer Survivor Study. Among these, common risk variants at 5p15.2 (LINC02112), 2p25.3 (MYT1L), and 19p12 (ZNF492) modified alkylating agent-related risks across ancestry groups, but AFR survivors with risk alleles experienced disproportionately greater risk of DM (AFR, variant ORs: 3.95-17.81; EUR, variant ORs: 2.37-3.32). Novel risk locus XNDC1N was identified in the first genome-wide DM rare variant burden association analysis in survivors (OR=8.65, 95% CI: 3.02-24.74, P=8.1x10-6). Lastly, a general-population 338-variant multi-ancestry T2D polygenic risk score was informative for DM risk in AFR survivors, and showed elevated DM odds after alkylating agent exposures (quintiles: combined OREUR=8.43, P=1.1x10-8; ORAFR=13.85, P=0.033). This study supports future precision diabetes surveillance/survivorship care for all childhood cancer survivors, including those with AFR ancestry.

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.