Genetic Polymorphisms of Drug-Metabolizing Enzymes Involved in 6-Mercaptopurine-Induced Myelosuppression in Thai Pediatric Acute Lymphoblastic Leukemia Patients

Abstract

Genetic polymorphisms of thiopurine S-methyltransferase (TPMT) and nucleoside diphosphate-linked moiety X-type motif 15 ( NUDT15 ) genes have been proposed as key determinants of 6-mercaptopurine (6-MP)-induced myelosuppression in pediatric acute lymphoblastic leukemia (ALL). In the present study, genotypes of TPMT and NUDT15 were investigated in 178 Thai pediatric patients with ALL by the TaqMan SNP genotyping assay and DNA sequencing. The frequency of TPMT*3C was 0.034. Among NUDT15 variants, NUDT15*3 is the most common variant with the allele frequency of 0.073, whereas those of NUDT15*2 , NUDT15*5 , and NUDT15*6 variants were 0.022, 0.011, and 0.039. These data suggest that a high proportion of Thai pediatric ALL patients may be at risk of thiopurine-induced myelosuppression.

Introduction

Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer worldwide. ¹ Chemotherapy for this disease consists of multiple drugs and several treatment protocols. The 6-mercaptopurine (6-MP) is a key chemotherapeutic agent in the maintenance phase of ALL treatment; however, this drug is frequently associated with life-threatening myelosuppression. ²

The 6-MP is an inactive drug and undergoes bioactivation by hypoxanthine-guanine phosphoribosyltransferase (HGPRT) to the active metabolite, 6-thioinosine monophosphate (6-TIMP). ³ The 6-TIMP is subsequently converted via multiple steps to active 6-thioguanine nucleotides (6-TGN) metabolites including 6-thioguanine monophosphate (6-TGMP), 6-thioguanine diphosphate (6-TGDP), and 6-thioguanine triphosphate (6-TGTP). These 6-TGN metabolites are then incorporated into DNA leading to cell apoptosis and immunosuppression. ³ Some portion of 6-MP, however, undergoes detoxification to inactive methylated metabolites, 6-methyl mercaptopurine (6-MMP) by thiopurine S-methyltransferase (TPMT), and 6-thiouric acid (6-TU) by xanthine oxidase (XO). ³ In addition, prevention of 6-TGTP incorporation into DNA can occur via dephosphorylation of 6-TGTP to 6-TGMP by the nucleoside diphosphate-linked moiety X-type motif 15 (NUDT15). ⁴

To date, more than 40 variants of the TPMT gene have been reported. ⁵ Among them, TPMT*3A alleles (c.460G > A and c.719A > G), TPMT*3C (c.719A > G), and TPMT*2 (c.238G > C) are the common variant alleles in Caucasians, whereas only the TPMT*3C allele (c.719A > G) is the most prevalent variant allele in Asian populations. ⁶ Activities of TPMT enzymes coding from these variant alleles are lower than the wild-type enzymes. ⁷ Several studies have demonstrated that individuals who carry these variant alleles may be at a higher risk of severe myelosuppression when treated with 6-MP or 6-MP derivatives. ^{7 8}

In contrast to TPMT , NUDT15 genetic polymorphism is more common in Asian populations than in Caucasian populations. Phenotypes of more than 10 variant alleles of NUDT15 have been extensively characterized. ^{4 9} The common variant alleles are NUDT15*2 (c.36_37insGGAGTC; c.415C > T), NUDT15*3 (c.415C > T), NUDT15*4 (c.416G > A), NUDT15*5 (c.52G > A), and NUDT15*6 (c.36_37insGGAGTC). ⁴ Several lines of evidence strongly suggest that NUDT15 variants particularly, NUDT15* 3, NUDT15*5 , and NUDT15*6 are strongly associated with myelosuppression caused by 6-MP or 6-MP derivatives. ^{10 11 12 13 14 15}

Compared with European countries, the successful outcomes of ALL therapy in Thailand are still low and myelosuppression caused by 6-MP is one of the main obstacles for the treatment of this disease. ¹⁶ Information about the polymorphisms of these key genetic markers is necessary for estimating the prevalence of Thai pediatric ALL patients who may be at a higher risk of 6-MP induced myelosuppression. Thus, the present study was aimed to elucidate the distribution of TPMT and NUDT15 polymorphisms in Thai pediatric ALL patients.

Materials and Methods

Study Population

One hundred seventy-eight Thai pediatric patients aged 1 to 15 years who had been diagnosed with ALL at Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, Thailand between 2018 and 2019, were enrolled in the study. These patients were categorized into risk groups and underwent chemotherapy according to the National protocol developed by the Thai Pediatric Oncology Group (ThaiPOG). ¹⁷ Written informed consent was obtained from subject's parents or guardians after receiving information about experimental procedures and purposes of the study. Approval for this study was obtained from the Ethics Committee for Human Research of Khon Kaen University (HE601062).

Detection of NUDT15 and TPMT Variants

Genomic DNA was purified from peripheral blood leucocytes using a DNA extraction kit (QIAamp DNA Blood mini kit, Germany). Genotyping of TPMT*3C , NUDT15*3 , and NUDT15*5 was performed using TaqMan SNP genotyping assays on a QuantStudio 6 Flex Machine (The Applied Biosystems, Massachusetts, United States). Other variants of NUDT15 including NUDT15*2 , NUDT15*4 , and NUDT15*6 to NUDT15*11 were determined by amplification of exon 1 and exon 3 of the NUDT15 gene as previously described, ⁴ and the DNA fragments were subsequently subjected for DNA sequencing by using a ABI Prism 3730XL DNA sequencer (Thermo Fisher Scientific Inc., Massachusetts, United States).

Statistical Analysis

Genotype and allele frequency data were calculated by counting alleles. Goodness of fit between observed and estimated genotype frequencies, according to the Hardy–Weinberg equilibrium, was determined by the Chi-square test. Allele frequencies in different ALL populations were compared using the Z-test. A p -value of less than 0.05 was considered statistically significant.

Results

Characteristics of the Study Population

A total of 178 ALL patients consisting of 93 males (52.25%) and 85 females (47.75%) were enrolled in this study. The median age at diagnosis was 6 years, and ranged from 1 to 15 years. The distribution of the age of these patients was less than 5 years in each segment, 57 patients (32.02%); 5 to 10 years, 82 patients (46.07%); and more than 10 years, 39 patients (21.91%). According to immunophenotype, 104 patients (58.43%) were of B cell lineage, 25 patients (14.04%) were of T cell lineage, 3 patients (1.69%) were positive for the Philadelphia chromosome and 46 patients (25.84%) were not specified. Based on the disease risk groups recommended by the ThaiPOG protocol, 96 patients (53.93%) were of standard risk, 66 patients (37.08%) were of high risk and 16 patients (8.99%) were of very high risk.

Thiopurine S-Methyltransferase Genetic Polymorphism

Results for TPMT*3C genotypes are shown in Table 1 . Based on the Hardy–Weinberg equilibrium, all of the observed genotype frequencies were not statistically different from the actual observed values. The allele frequency of the TPMT*3C allele observed in this study was 0.034 (95% confidence interval [CI]: 0.018–0.058; Table 2 ). In comparison to other Asian ALL populations, the allele frequency of TPMT*3C in Thai pediatric ALL patients observed in the present study was comparable except for the Korean ALL population. The frequency of this variant allele, however, was approximately 17-fold higher than that found in the European ALL population.

Table 1. Genotype frequencies of thiopurine S-methyltransferase in Thai pediatric acute lymphoblastic leukemia patients.

Genotype	Number of subjects	%Observed genotype frequency (95% CI)	%Expected genotype frequency	p -Value
TPMT*3C
*1/*1 a	166	93.26 (88.52–96.47)	93.37	0.89
*1/*3C	12	6.74 (3.53–11.48)	6.52
*3C/*3C	0	0 (0)	0.11

Abbreviations: CI, confidence interval; TPMT, thiopurine S-methyltransferase.

^a Referring to subjects who did not carry TPMT*3C variant.

Table 2. Comparative allele frequency of thiopurine S-methyltransferase among various acute lymphoblastic leukemia populations.

Populations	Number of subjects	Allele frequency
		*1	*3C
Thai (this study)	178	0.966 (0.942–0.982)	0.034 (0.018–0.058)
Thai 20	75	0.947	0.053
Thai 21	102	0.975	0.025
Korean 28	244	0.986	0.012 a
Korean 29	139	0.978	0.007 a
Chinese 23	105	0.971	0.029
Chinese 24	404	0.984	0.016
Japanese 25	60	0.983	0.017
Japanese 26	71	0.979	0.014
Japanese 13	95	0.979	0.021
Japanese 27	119	0.987	0.013
European 30	203	0.924 a	0.002 a
African-Americans 22	248	0.964	0.024

Note: Data in parentheses represent 95% confidence interval.

^a p  < 0.05 (different from this study).

Nucleoside Diphosphate-Linked Moiety X-Type Motif 15 Genetic Polymorphism

Table 3 summarizes the genotype frequencies of NUDT15 variant alleles in Thai ALL patients. Of 10 variant allele genotypes, none of the study subjects carried NUDT15*4 or NUDT15*7 through NUDT15*11 . Moreover, results obtained from DNA sequencing revealed that there were no novel variants in the exons 1 and 3 of the NUDT15 gene detected in this study population.

Table 3. Genotype frequencies of NUDT15 in Thai pediatric acute lymphoblastic leukemia patients .

Genotype	Number of subjects	%Observed genotype frequency (95% CI)	%Expected genotype frequency	p -Value
NUDT15*1
*1/*1 a	128	71.91 (64.70–78.38)	73.88	0.09
NUDT15*2
*1/*2	8	4.49 (1.96–8.66)	4.39	0.95
*2/*2	0	0 (0)	0.05
NUDT15*3
*1/*3	25	14.04 (9.30–20.03)	13.06	0.60
*3/*3	0	0 (0)	0.49
NUDT15*5
*1/*5	4	2.25 (0.62–5.65)	2.22	0.99
*5/*5	0	0 (0)	0.01
NUDT15*6
*1/*6	12	6.74 (3.53–11.48)	7.56	0.34
*2/*6	1	0.56 (0.01–3.09)	0.15

Note: Data in parentheses represent 95% confidence interval.

^a Referring to subjects who did not carry any of the investigated NUDT15 variants.

Based on the Hardy–Weinberg equilibrium, the actual observed genotype frequencies of NUDT15 were not statistically different from the estimated frequencies. The NUDT15*3 is the most common variant allele with an allele frequency of 0.073 (95% CI: 0.048–0.105), followed by NUDT15*6 (0.039; 95% CI: 0.022–0.065) and NUDT15*2 (0.022; 95% CI: 0.010–0.044).

The NUDT15 allele frequencies detected in the present study as well as those previously reported from other ALL populations are summarized in Table 4 . Compared with other Asian ALL populations, the frequency of the most common variant allele— NUDT15*3 —was quite comparable. In contrast, the frequency of this variant allele was much higher than those reported in European and Hispanic-American, Native-American, and African ALL populations. Compared with Japanese, the frequency of NUDT15*6 allele in Thai ALL patients was more than 10-fold higher, whereas the frequency of NUDT15*2 was more than 2-fold lower. The allele frequency of NUDT15*5 in the study population was comparable to those reported in other Asian ALL patients.

Table 4. Comparative allele frequency of NUDT15 among various acute lymphoblastic leukemia populations .

Populations	Allele frequency
	Number of subjects	*1	*2	*3	*4	*5	*6
	DNA variant	Wild-type	c.36_37insGGAGTC; c.415C > T	c.415C >T	c.416 G   >   A	c.52G   >   A	c.36_37insGGAGTC
	Protein variant	Wild-type	p.Val18_Val19insGlyVal/ p.Arg139Cys	p.Arg139Cys	p.Arg139His	p.Val18Ile	p.Val18_Val19insGlyVal
Thai (this study)	178	0.855 (0.816–0.891)	0.022 (0.010–0.044)	0.073 (0.048–0.105)	0 (0–0.016)	0.011 (0.003–0.029)	0.039 (0.022–0.065)
Thai 21	102	0.873	NA	0.127 a	NA	NA	NA
Thai 31	82	0.915	NA	0.085	NA	NA	NA
Korean 28	244	0.898	0.012	0.072	0.008	0.008	0.002 a
Chinese 32	60	0.875	NA	0.125	NA	NA	NA
Chinese 24	404	0.884	NA	0.116 a	NA	NA	NA
Japanese 4	32	0.828	0.063	0.109	0	0	0
Japanese 33	138	0.848	0.051 a	0.069	0	0.029	0.003 a
Singaporean 4	79	0.867	0.032	0.089	0	0.013	0 a
European 14	205	0.998 a	NA	0.002 a	NA	NA	NA
Hispanic American 14	222	0.959 a	NA	0.041 a	NA	NA	NA
Native American 4	159	0.934 a	0.028	0.019 a	0.019 a	0	0 a
African 14	93	1.000 a	NA	0 a	NA	NA	NA

Note: Data in parentheses represent 95% CI. NA represents data not available.

^a p  < 0.05 (different from this study).

Discussion

The 6-MP is a key chemotherapeutic drug used for the treatment of pediatric ALL; however, myelosuppression remains an important adverse reaction to this agent. Recently, genetic polymorphisms of TPMT and NUDT15 have been proposed to be involved with an increased risk of myelosuppression in patients treated with thiopurine drugs, including 6-MP and its prodrug, azathioprine. ^{4 8 9 14 15 18 19} To the authors' knowledge, this is the first study evaluating the common TPMT variant and all known loss-of-function NUDT15 variants in Thai pediatric ALL patients.

Demographic and clinical data of 178 pediatric ALL patients enrolled in the present study were consistent to those of 486 pediatric ALL patients previously reported by the ThaiPOG. ¹⁶ Allele frequency of TPMT*3C in this study was comparable with those previous reports in Thais ^{20 21} and African-Americans. ²² Compared with other Asian ALL populations, ^{13 23 24 25 26 27} the frequency of TPMT*3C in this Thai pediatric ALL population was somewhat higher but did not reach statistical significance except with the Korean population. ^{28 29} In addition, the allele frequency TPMT*3C in our study population was extremely higher than those reported in European ALL patients ³⁰ ( Table 2 ). Among 10 NUDT15 variants investigated, NUDT15*3 is the most common variant in Thai ALL population, followed by NUDT15*6 , NUDT15* 2, and NUDT15*5 variants. Other known loss-of-function alleles were not detected in this study population. Compared with other Asian ALL populations, ^{4 21 24 28 31 32 33} the frequency of the NUDT15* 3 in Thai pediatric ALL patients was comparable, whereas NUDT15*6 was significantly higher. In contrast, the allele frequency of NUDT15*3 observed in this study was higher than those reported in European, ¹⁴ Hispanic-American, ¹⁴ Native-American, ⁴ and African ¹⁴ ALL populations ( Table 4 ). It is noteworthy that the risk of thiopurine-induced myelosuppression appears to be higher among Asian populations compared with Caucasian populations. ^{14 15} The high frequency of the NUDT15 variants may, in part, explain the higher risk of thiopurine-induced myelosuppression observed in Asian populations. ^{14 15}

It has been reported that Thai patients who carried the TPMT*1/*3C genotype were at about a 14-fold higher risk of thiopurine-induced myelosuppression compared with patients with the homozygous wild-type allele. ³⁴ A recent study in Thai pediatric ALL patients revealed that patients who carried NUDT15*3 , the most common variant allele, were at a 7- to 14-fold higher risk of 6-MP and the 6-MP dose in this variant group was significantly lower compared with the homozygous wild-type group. ³¹ A higher risk of 6 MP-induced myelosuppression was also reported in patients who carried NUDT15*2 , NUDT15*4 , NUDT15*5 , and NUDT15*6 . ^{4 35} Recently, the Clinical Pharmacogenetics Implementation Consortium Guidelines have recommended that physicians should adjust the starting dose of thiopurine drugs based on the patient's TPMT and NUDT15 genotypes. ³⁶ It should be noted that four out of 178 patients (2.25%; 95% CI: 0.62–5.65) carried TPMT*3C together with one NUDT15 variant allele. The severity of myelosuppression in patients who carried the loss-of-functional alleles of both TPMT and NUDT15 genes needs to be further evaluated.

In conclusion, this is the first study evaluating the common TPMT variants together with all known loss-of-function NUDT15 variants in Thai pediatric ALL patients. The TPMT*3C as well as NUDT15 variants with low/intermediate activity were present in a high frequency in Thai pediatric ALL patients. These data suggest that a high proportion of Thai pediatric ALL patients may at risk of thiopurine-induced myelosuppression. Genotyping of these genes prior to thiopurine prescribing may need to be considered to ensure the proper dosage adjustment. A current investigation of the impact of NUDT15 as well as TPMT variants on thiopurine-induced myelosuppression as well as proper dose adjustment in Thai pediatric ALL subjects is underway.

Funding Statement

Funding This work was supported by grants from the Thailand Center of Excellence for Life Sciences (grant number TC-12/63) and the Program Management Unit for Human Resources & Institutional Development, Research and Innovation (grant number 630000050064). Scholarship support from Graduate School, Khon Kaen University through the Research Fund for Supporting Lecturer to Admit High Potential Student to Study and Research on His Expert Program Year 2019 (grant number 621H219).