Background: Acute lymphoblastic leukemia (ALL), the most prevalent malignant carcinoma in children, is closely associated with genetic alterations. Susceptible loci derived from conventional association analyses are primarily common variants in noncoding region, which are insufficient to fully explain the disease's genetic etiology. Nowadays, advancements in statistical methods for next-generation sequencing data are continually providing opportunities to detect rare susceptible genetic variations. This study utilized whole-exome sequencing (WES) to identify coding genetic variants critical to the onset of pediatric ALL and their relationship to laboratory tests after long-term survival of these children.

Methods: Peripheral blood samples from 89 ALL patients were collected, along with results of laboratory tests which were done after long-term survival of these children, including routine blood, biochemistry, coagulation, and thyroid function tests. Propensity score matching was used to select controls without tumors at a 1:4 ratio. Initially, an exome-wide association study (EWAS) was conducted in a case-control setting. Following rigorous quality control, variants were categorized as common [minor allele counts (MAC) ≥ 20] or rare [MAC > 1 and minor allele frequency (MAF) < 5%]. A general linear model was employed for the common variants, while the Variant-Set Test for Association using Annotation information omnibus (STAAR-O) test was used to identify susceptible signals among rare variants. Subsequently, we analyzed associations between identified variant sets and laboratory tests after long-term survival of pediatric ALL patiens, adjusting for multiple comparisons using the Bonferroni method.

Results: No common variants were correlated with childhood ALL. However, four sets of rare variants were associated with an increased risk of ALL: missense variants in NOL8 (PSTAAR-O=2.46×10-10), synonymous variants in MFAP3L (PSTAAR-O=3.97×10-8), missense variants in CARD11 (PSTAAR-O=4.18×10-8), and missense variants in OR7G2 (PSTAAR-O=2.20×10-7). Notably, NOL8 missense variants correlated with red blood cell counts (PSTAAR-O=2.83×10-4), while OR7G2 missense variants were linked to thyroid-stimulating hormone levels (PSTAAR-O=9.56×10-3), suggesting a potential impact on thyroid function in ALL suffers. Additionally, NOL8 missense variants showed suggestive associations with mean corpuscular volume (PSTAAR-O=2.04×10-3), mean corpuscular hemoglobin levels (PSTAAR-O=4.77×10-3), erythrocyte distribution width CV (PSTAAR-O =2.55×10-2), and platelet distribution width (PSTAAR-O=3.02×10-2). Similarly, OR7G2 missense variants were suggestively linked to platelet distribution width (PSTAAR-O=7.86×10-3), mean platelet volume (PSTAAR-O=1.18×10-2), platelet large cell ratio (PSTAAR-O=3.96×10-3), international normalized ratios (PSTAAR-O=2.00×10-2), uric acid levels (PSTAAR-O=4.01×10-2), and free triiodothyronine levels (PSTAAR-O= 2.83×10-2).

Conclusion: Through comprehensive analyses based on genotype data and clinical laboratory tests of survivors , we found that variations in the genes NOL8 and OR7G2 were implicated not only in the risk of developing ALL but also in the physiological processes during survival time. Our findings might facilitate more precise risk stratification of ALL patients, and contribute to providing more personalized health monitoring for ALL survivors.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution