Heritable Variation in Lymphocyte-Related Traits and Risk of Down Syndrome Acute Lymphoblastic Leukemia: A Mendelian Randomization Study

Introduction: The risk of developing acute lymphoblastic leukemia (ALL) is up to 30-fold higher in children with Down syndrome (DS). While trisomy 21 increases the risk of leukemia, only ~1% of children with DS develop ALL (DS-ALL), suggesting that additional modifying risk factors exist. A recent study in individuals without DS revealed that a genetic propensity for increased lymphocyte production was associated with an increased risk of childhood ALL. Additionally, a previous genome-wide association study (GWAS) of DS-ALL identified associations in genes involved in lymphoid development. Here, we investigated whether genetic variation in lymphocyte-related traits is associated with DS-ALL risk, with the hypothesis that a genetic predisposition for lymphocyte overproduction is associated with an increased risk of DS-ALL.

Methods: We performed Mendelian randomization (MR) analyses to investigate the potential causal relationship between variation in four lymphocyte-related traits – lymphocyte counts, the lymphocytes to monocytes ratio (LMR), the neutrophils to lymphocytes ratio (NLR), and the platelets to lymphocytes ratio (PLR) – and DS-ALL risk. Genetic variants associated with each lymphocyte-related trait (“exposure”) at P < 5x10^-8 and their corresponding effect sizes and standard errors were obtained from GWAS of blood cell traits performed in 413,810 cancer-free individuals of European ancestry in the UK Biobank. A GWAS of DS-ALL (“outcome”) was carried out using germline whole-genome sequencing data from 222 DS-ALL cases and 1,010 individuals without ALL from DS research cohorts, limited to individuals of non-Hispanic White race/ethnicity. We examined consistency across multiple MR methods, including the inverse-variance weighted (IVW), weighted median (WM), MR robust adjusted profile score (RAPS), and MR pleiotropy residual sum and outlier (PRESSO) estimators.

Results: Among the lymphocyte-related traits, we found a significant association between LMR and DS-ALL risk that was consistent across MR estimators. Specifically, each 1-unit increase in LMR was associated with a 41%–53% decrease in ALL risk in children with DS (odds ratio [OR_IVW] = 0.56; 95% CI: 0.35–0.90, p = 0.016; OR_WM = 0.47; 95% CI: 0.21–1.04, p = 0.051; OR_RAPS = 0.59; 95% CI: 0.36–0.97, p = 0.038; OR_PRESSO = 0.55; 95% CI: 0.34–0.88, p = 0.014). There were no statistically significant associations between the remaining blood cell traits (lymphocyte count, PLR, and NLR) and DS-ALL risk, although increased lymphocyte counts conferred a modest increase in risk of DS-ALL (OR_IVW=1.24, p=0.40).

Conclusions: We discovered that genetic predisposition to shifts in the ratio of lymphocytes relative to monocytes may be a novel risk factor for the development of DS-ALL, with a deficit of lymphocytes to monocytes conferring an increased risk of DS-ALL. This observed direction of effect contrasts with previous findings for ALL in the non-DS population, suggesting a distinct etiology of ALL in the context of DS that may be influenced by the profound effects of trisomy 21 on immune function. Replication of our findings in independent DS-ALL studies is underway, and analysis of polygenic risk scores for lymphocyte traits may further improve identification of children with a high risk of DS-ALL.