Role of Antioxidant Gene Polymorphisms in Risk and Prognosis of Chronic Myeloid Leukemia

Chronic myeloid leukemia (CML) is a hematopoietic stem cell disorder of myeloid precursors, characterized by the presence of Philadelphia (Ph) chromosome, which results from a reciprocal chromosomal translocation t (9;22), leading to the BCR-ABL1 fusion gene. The BCR-ABL1 gene can modulate DNA repair mechanisms, cell cycle checkpoints, Bcl2 proteins and enhances reactive oxygen species (ROS) generation, Abstract


Materials and Methods
The present study included 325 samples, out of which 125 are from CML patients and 200 were from age & gender matched controls without a family history of any cancer. The inclusion criteria for patients included Ph+ve CML cases with confirmed diagnosis, on TKI treatment and TKI refractory cases regardless of age, gender or race. The study was approved by the institutional ethics committee and an informed consent was obtained from patients participating in the study. Blood samples (6mL in EDTA vaccutainer) were collected from both CML patients and controls. Genomic DNA was extracted from blood samples using non-enzymatic rapid salting-out method. The purity & concentration of DNA samples were checked on Nanodrop1000 and further these DNA samples were subjected for analysis of SNPs in antioxidant enzyme genes.

Statistical analysis
Chi square and multivariate analysis tests were calculated to test the significance of genotype association with the occurrence of CML and its prognosis. All the p values were two sided and the level of significance was taken as p <0.05. Statistical analyses were performed using the GraphPad Prism software version 6.0 (San Diego, CA) and online VassarStats software. Haplotype and pairwise linkage disequilibrium was calculated using Haploview version 4.2 and cox regression analysis by SPSS version 22 software.

Results
Baseline characteristics (Table 2) The demographic and clinical characteristics of CML patients are presented in Table 2. The median age at diagnosis of CML was 42 years (range 12 to 89 years) and a male preponderance was observed with a male to female ratio of 1 Median follow-up of these patients for a period of 40 median months revealed that 20.8% had optimal response to imatinib and 79.02% of patients lost respone which might be either due to loss of complete hematological response (CHR), complete cytogenetic response (CCyR), major molecular response (MMR) or presence of TKD mutations.
Correlation with CAT -21A/T polymorphism (Table 3) The CAT -21A/T genotyping results revealed that heterozygous AT genotype frequency was observed to be significantly increased in CML patients compared to controls (p=0.037). This polymorphism was significantly assciated with increased risk of CML. With respect to molecular response, homozygous TT genotype and T allele frequencies were elevated in non-responders i.e., patients having higher BCR-ABL1 expression levels (44.70%, 0.705) compared to responders i.e., patients having lower levels (35.0%, 0.625) (p=0.259). Heterozygous AT genotype frequency was found to be slightly increased in TKD mutation carriers (p=0.571) and in deceased group of patients (p=0.548) when compared to respective groups. No differences were found with either of the prognostic risk scores: Sokal, Hasford or EUTOS.
The CAT -21A/T polymorphism showed statistically significant association with risk of CML and conferred Correlation with MPO -463G/A polymorphism (Table 9) The MPO -463G/A polymorphism demonstrated no significant association between cases and controls (p=0.494), nor with either of the confounding variables like molecular response (p=0.465), TKD mutation status (p=0.392), present status (p=0.767) and prognostic risk scores.
Correlation with GSTM1 & GSTT1 null/deletion polymorphism (Table 10) No significant association observed with GSTM1 null polymorphism between cases and controls, molecular response, presence or absence of TKD mutations. Whereas GSTM1 presence genotype (M1) was found to be elevated in deceased group (80.95%) compared to those on follow-up (66.34%) (p=0.289).
With GSTT1 null polymorphism, the GSTT1 null genotype frequency slightly increased in cases compared to controls (22.4%; 16.0%) (p=0.193). When the results are stratified with confounding variables, the GSTT1 null genotype frequency was found to be higher in non-responders (27.05%; 12.5%) and in patients not carrying TKD mutations (26.31%; 10.0%) compared to respective groups. There was no difference was observed between follow-up and deceased group of patients with GSTT1 null genotype.
No significant differences were found between GSTM1 & GSTT1 null/deletion polymorphisms and prognostic  Table 4).

Correlation with CAT -262C/T polymorphism (Table 5)
There was no significant difference observed between cases and controls (p=0.711), molecular response (p=0.865) and presence or absence of TKD mutations (p=0.708) with CAT-262C/T polymorphism. This polymorphism was not assciated with risk of CML. Whereas the homozygous CC genotype and C allele frequencies were found to be elevated in the deceased group (71.42%, 0.857) compared to those patients on follow-up (50.0%; 0.711) (p=0.139). The prognostic risk scores were not associated with this polymorphism.
Correlation with GPX1 -198C/T polymorphism (Table 7) The heterozygous CT genotype and T allele frequencies were significantly increased in CML patients compared to controls (p=<0.0001). With respect to molecular response and TKD mutation status, the heterozygous CT genotype frequency was observed to be significantly elevated in poor molecular responders group (patients having higher BCR-ABL1 levels) (p=0.005), TKD mutation carriers (p=0.114) and in patients of advanced phases (p=0.292) compared to respective groups. With respect to present status, the frequencies of TT genotype and T alleles were found to be slightly increased in deceased group of Gene SNP Primer sequence Product size Restriction enzyme The haplotype and pairwise epistasis among six SNPs did not revealed any significant association, hence data not presented. The linkage disequilibrium (LD) analysis revealed that the two CAT -21A/T (rs1001179) and CAT -262 C/T (rs7943316) exhibited high LD (D'=0.9).
Since the two SNPs are located on chromosome 1, the observed significant LD might be attributed to the physical proximity. None of the other SNP combinations showed significant LD with D'<0.5 ( Figure 1). Cox regression analysis of SNPs with BCR-ABL1 levels revealed no significant association.

Discussion
In the present study, we investigated the association of the genetic variations of the antioxidant enzymes: Catalase is an endogenous antioxidant enzyme involved in ROS neutralizing pathways and prevents    cellular injury from ROS [22]. Two polymorphisms: CAT -21A/T with altered gene expression pattern [23] and CAT -262C/T with lower CAT enzyme activity [24] may alter ROS detoxification and increase oxidative stress, implicating oxidative DNA damage and modulating disease risk [25]. In the present study, the CAT -21A/T polymorphism was significantly associated with increased risk of CML (p=0.037  reported an increased cancer risk with recessive model and homozygote model [26]. This indicates that variant T allele with lower catalase activity and thus increased levels of ROS may contribute to genomic instability and increased risk of cancer. Earlier studies reported no significant associations with the risk of colorectal cancer [27], gastric cancer (GC) and hepatocellular carcinoma (HCC) [13]. In our study, we found no evidence of the CAT -262 C/T polymorphism with CML risk or its association with confounding variables. Our results are in accordance with earlier studies that reported no significant association with risk of hepatocellular carcinoma [28], breast cancer [29], and gastric cancer [30]. Previous other studies showed significant increased risk of cervical cancer [15], breast cancer [31], hepatocellular carcinoma [32] and prostate cancer [33]. Whereas others reported that -262C/T polymorphism was a protective factor with respect to chronic myeloid leukemia [19] and hepatocellular carcinoma susceptibility [14][15][16][17]. GPX1 is a key enzyme of the antioxidative system that detoxifies peroxide radicals and lipid hydroperoxides. The -198C/T (Pro200Leu) polymorphism in GPX1 is associated with reduced enzyme activity [34][35]. Previous studies reported that higher GPX1 activity is required to counterbalance the ROS levels and related damage occurring during initiation or progression of the cancer [36][37][38][39]. We obseved statistically significant association of the homozygous variant TT genotype with CML risk (p=<0.0001). The stratified results of confounding variables presented the significant association of GPX1 -198 C/T polymorphism with poor molecular response (p=0.005) and acquired TKD mutations (p=0.114). In addition, the codominant (CC vs CT and CC vs TT) and dominant (CC vs CT+TT) models conferred increased risk of CML when compared with CC homozygote (p=<0.0001). Our results were in accordance with others findings on breast cancer [39][40][41], bladder cancer [42] and lung cancer [43]. This indicates that the variant Leu allele with reduced enzyme activity might increase ROS levels thereby induced oxidative DNA damage and increased susceptibility to cancer. Whereas other studies failed to find an association of GPX1 -198C/T polymorphism with the risk of CML [19], breast cancer [44][45] and prostate cancer [46]. Glutathione S-transferases (GSTs) are involved in detoxification of a wide range of carcinogens and ROS thereby offering protection against oxidative DNA damage. GST enzymes are polymorphic, which may contribute to the inter-individual variability in the response to oxidative stress suggesting its role in carcinogenesis and risk for cancer. In the present study, the GSTM1 and GSTT1 null/deletion polymorphisms were not associated with risk of CML. Our results are similar with earlier studies on CML [20]. Previous studies on the GSTT1 null polymorphism reported positive association with risk of CML [47][48][49][50] and AML [20]. Earlier studies on GSTM1 null polymorphism showed no association the risk of CML [50], AML [51] and breast cancer [52].
Myeloperoxidase (MPO) is an endogenous oxidant enzyme that activates carcinogens [53]. A single nucleotide polymorphism in the promoter region of the MPO gene, G-463A (rs2333227) has been associated with reduced mRNA expression and transcriptional activity and subsequent decreased metabolic activation of procarcinogens [54]. In the present study, no evidence of MPO -463G/A polymorphism with the risk of CML was observed. Our results were in accordance with earlier studies on ALL [55], AML [56] and breast cancer [57]. Whereas others reported that the A allele with reduced MPO activity and ROS production has been associated with decreased risk of breast cancer [58], lung cancer [59] and prostate cancer [60].
In conclusion, our results suggest that the reduced activity of antioxidant enzymes caused by the CAT -21A/T and GPX1-198C/T polymorphisms might contribute to increased risk of CML. In addition, the GPX1-198C/T polymorphism was associated with poor molecular response and acquired TKD mutations. Hence, the present study indicates that defective antioxidant defense system might have a strong influence on CML susceptibility and TKI (imatinib) response through oxidative stress.

Conflicts of interest
There are no conflicts of interest.