Background: The multidrug resistance gene MDR1 is highly expressed in hematopoietic stem cells and may protect them from toxic insults, therefore alterations in MDR1 may predispose to acute myeloid leukemia (AML). Glutathione-S-transferase (GST) null alleles may also influence leukemia predisposition. We therefore studied patients (pts) with AML and controls of two ethnic groups for polymorphisms (SNPs) in these genes. Our patients are younger (mean, 46.1 yrs) than the mean of AML pts in the literature, Arab pts being significantly younger (41.4 yrs) than Jewish pts (48.7 yrs) (p<0.002). We asked whether genetic background predisposes to AML in these young pts.

Methods: We studied 84 Jewish pts and 42 Arab pts with de novo AML and 109 Jewish controls and 91 Arab controls. We studied three SNPs in the MDR1 gene: T-129C (which may reduce gene expression), G2677T (leading to an ala893ser substitution, which increases duodenal MDR1 activity) and C3435T, which is a synonymous substitution. We also studied GST null alleles (GST T1 and M1). Polymorphisms were studied using PCR-based techniques. Survival was analyzed using Kaplan-Meier analysis. SNP genotype frequencies were analyzed using Chi Square and Student’s t-test.

Results: G2677T was found to be associated with AML: the T allele (serine) was more frequent in Arab controls than in Arab pts. Thus, it seems to exert a protective effect against developing AML among Arabs (p<0.01). For both ethnic groups combined, the TT genotype at nt 2677 was associated with a significantly older age at diagnosis, compared to the GG/TG genotypes (mean 57 versus 46 yrs, p<0.03). In contrast, none of the other individual MDR1 polymorphisms were significantly associated with a protective effect in either ethnic group, nor were they found to be associated with any difference in age at diagnosis. Using the PHASE 2.1 program to create possible combinations of pairs of SNPs and all 3 SNPs combined for each ethnic group, we found a statistically significant different distribution of combinations between Arab AML pts and Arab controls for all possible combinations (p< 0.01–0.001), and between Arab and Jewish controls (p<0.01–0.001). In contrast, no such difference was found between Jewish AML pts and Jewish controls (p<0.2–1). Statistically significant differences between Arab and Jewish AML pts were found for 2 combinations: G2677T and C3435T together and all 3 SNPs combined (p<0.01 and p<0.025, respectively). Survival was not influenced by any of the genotype profiles of MDR1 SNPs in either ethnic group. For GST alleles, we did not find a difference in the frequency of either GST T1 null, GST M1 null or double null genotypes when comparing AML pts to controls of the same ethnic groups.

Conclusions: We conclude that MDR1 polymorphisms may protect against de novo AML in Arabs, most significantly the 2677 polymorphic T allele. This effect was not found for Jews. The T allele of 2677 was also associated with an older age at diagnosis in both ethnic groups. GST null genotypes did not predispose to the development of de novo AML in the ethnic groups studied. Neither MDR1 nor GST polymorphisms influenced the outcome of AML in our patient population.

Author notes

Corresponding author