Skip to main content
NCBI home page
Clinical and Translational Science logoLink to Clinical and Translational Science
. 2026 Jun 12;19(6):e70639. doi: 10.1111/cts.70639

Severe Thiopurine‐Induced Myelosuppression in a Pediatric Acute Lymphoblastic Leukemia Patient With the NUDT15 *1/*6 Genotype: A Brief Report

Jillian Fry 1,2,✉, Erin C Boone 2, Wendy Y Wang 2, Shobana John 2, Lisa A Lansdon 3,4,5, Midhat S Farooqi 3,5, Byunggil Yoo 3, Andrea Gaedigk 2,5, Keith August 1,5, Terrie G Flatt 1,5, Laura B Ramsey 2,5
PMCID: PMC13263158  PMID: 42286409

ABSTRACT

Variants in TPMT and NUDT15 genes that affect thiopurine metabolism can guide personalized dosing to minimize toxicity. Decreased or no appreciable NUDT15 activity demonstrates impaired breakdown of active thiopurine metabolites which can lead to severe adverse events including potentially life‐threatening myelosuppression. Recently, the NUDT15*6 allele was re‐classified from having uncertain function to no function by the Clinical Pharmacogenetics Implementation Consortium. Here, we present a pediatric patient with a NUDT15*1/*6 genotype who experienced significant thiopurine‐induced myelosuppression. The patient is a 4‐year‐old female with standard risk‐average precursor B‐cell acute lymphoblastic leukemia treated per AALL1731. Exome sequencing determined TPMT*1/*1 (normal metabolizer) and NUDT15*1/*6 (indeterminate metabolizer). After 75 mg/m2/day mercaptopurine, myelosuppression necessitated a three‐week delay prior to the next phase of therapy. With 60 mg/m2/day thioguanine in a subsequent phase of therapy, she was admitted twice for fever and neutropenia. In the final phase of therapy, mercaptopurine at standard dosing of 75 mg/m2/day caused severe neutropenia and thrombocytopenia with elevated metabolite levels that required stopping mercaptopurine. After neutrophil recovery, a trial of 50% standard dosing was not tolerated and her dose was reduced to 27%, which was tolerated. Among 339 patients with cancer, sequencing revealed an allele frequency of 0.88% for NUDT15*6. In a larger cohort sequenced for suspicion of rare genetic disease, the frequency of NUDT15*6 was 0.15% among 1011 unrelated individuals. This brief report supports the recent update that patients with the NUDT15*1/*6 genotype should be classified as intermediate metabolizers.

Keywords: chemotherapy, oncology, pediatric, pharmacogenomics, toxicity

Study Highlights

  • What is the current knowledge on the topic?

    The treatment of pediatric acute lymphoblastic leukemia includes the use of thiopurines mercaptopurine and thioguanine, and genetic variants in TPMT and NUDT15 that affect thiopurine metabolism can guide dosing. NUDT15 deficiency can cause myelosuppression when affected patients are treated with thiopurines. Recently, the NUDT15*6 allele was re‐classified from having uncertain function to no function by the Clinical Pharmacogenetics Implementation Consortium.

  • What question did this study address?

    What is the activity of the NUDT15*6 allele?

  • What does this study add to our knowledge?

    This brief report supports the determination that the NUDT15*6 allele has no activity, and patients with the NUDT15*1/*6 genotype have a higher risk of toxicity compared to normal metabolizers and, thus, should be treated as intermediate metabolizers.

  • How might this change clinical pharmacology or translational science?

    Patients with a NUDT15*1/*6 genotype may need to be monitored more closely for toxicity and require dose reductions while taking thiopurines compared to patients genotyped as NUDT15*1/*1.

1. Introduction

Thiopurines, including thioguanine (6TG) and mercaptopurine (6MP), are the cornerstone of therapy for pediatric acute lymphoblastic leukemia (ALL). In the complex thiopurine metabolic pathway, 6TG and 6MP are methylated by thiopurine methyltransferase (TPMT) to an inactive metabolite, leaving less parent drug available for formation of active thioguanine nucleotides (TGNs), which cause leukemia cell death and myelosuppression [1, 2]. Thus, patients with TPMT alleles associated with decreased or no enzyme function are more susceptible to severe myelosuppression due to very high TGNs if given standard thiopurine doses [3, 4]. Nudix hydroxylase 15 (NUDT15) similarly converts cytotoxic thiopurine metabolites to less toxic metabolites. NUDT15 genetic variants resulting in an enzyme with decreased or no activity also lead to more active metabolites available for incorporation into DNA for myelosuppression and thiopurine intolerance [5].

Thus, genetic variants in TPMT and NUDT15 warrant testing to inform thiopurine dosing. TPMT and NUDT15 genetic testing examines the DNA sequence for genetic variation, which are reported using star (*) allele nomenclature per the TPMT nomenclature website [6] and Pharmacogene Variation Consortium (PharmVar) [7, 8]. These star alleles, or haplotypes, consist of one or more variants, and their function informs a person's predicted metabolic capacity, or phenotype status. Metabolizer status is taken into consideration to individualize drug dosing to minimize toxicity [9]. The NUDT15*6 allele is defined by a six‐base pair repeat insertion in exon 1 (NM_018283.4:c.55_56insGAGTCG, rs746071566) (Figure 1) causing an in‐frame insertion of two amino acids, p.V18_V19insGV. This allele was recently re‐classified from having uncertain function to no function by the Clinical Pharmacogenetics Implementation Consortium (CPIC) [11].

FIGURE 1.

FIGURE 1

Open in a new tab

Illustration of selected NUDT15 alleles including those having an insertion (c.55_56insGAGTCG, p.V18_V19insGV) or deletion (c.50delGAGTCG, p.V17_18GV) of a 6‐bp the repeat unit both of which are described by rs746071566. The NUDT15*3 allele has c.415C > T (p.R139C, rs116855232) with the *3.002 suballele (formerly NUDT15*2) also having the 6‐bp insertion. The NUDT15*1 reference sequence has three 6‐bp repeat motifs while the NUDT15*3.002 suballele and *6 have four repeats and the NUDT15*9 allele has two repeats. Positions for the insertion may be displayed differently as compared to previous publications as detailed in reference [10].

Here, we present a patient who was genotyped as TPMT*1/*1 and NUDT15*1/*6 and experienced significant thiopurine‐induced myelosuppression. Our finding adds to the body of evidence supporting that NUDT15*6 should be classified as a no function allele and NUDT15*1/*6 as intermediate metabolizer (IM). We also present frequency data for the NUDT15*6 allele among patients with clinical sequencing.

2. Methods

This patient was enrolled in the Children's Mercy Research Institute Biorepository (Tumor Bank) study, which is approved by the Children's Mercy Institutional Review Board. The frequency of NUDT15*6 was determined in this cohort of patients with clinical short‐read whole exome sequencing (SRS) data (n = 339). Libraries for SRS were prepared with the Illumina TruSeq PCR‐Free kit and enriched using the IDT xGen Exome Research Panel v2, then sequenced on a NovaSeq6000 platform to ≥ 300× targeted depth. Sequencing, primary analysis, alignment, and variant calling using DRAGEN v3.10.4‐v4.3.6 were performed by the Children's Mercy Center for Genomic Medicine (Kansas City, MO). The NUDT15 gene was filtered from the variant call files using BCFtoolsv1.18 [12] and star alleles were assigned based on PharmVar Nomenclature.

The frequency of the NUDT15*6 allele was also calculated from participants enrolled in the Genomic Answers for Kids (GA4K) Program, which is approved by the Children's Mercy Institutional Review Board. In the GA4K cohort, PacBio HiFi long‐read sequencing (LRS, targeted coverage > 25×) was used to identify and phase single nucleotide variants (SNVs) and structural variants using standard PacBio pipelines (pbmm2 v2.1.4.0, DeepVariant v1.0, WhatsHap v1.0, pbsv v2.4.0, and HIFI‐CNV v1.0.1). NUDT15 star alleles were assigned with pb‐StarPhase v2.0.1 for subjects with ≥ 10 reads, enabling accurate identification of the NUDT15*1/*6 diplotype. Self‐identified race, ethnicity, and relatedness were collected at enrollment and accessed via electronic health records.

3. Results

We present a 4‐year‐old female with standard risk‐average precursor B‐cell ALL treated per the Children's Oncology Group protocol AALL1731 [13]. Similar to prior protocols, AALL1731 includes 6MP at 75 mg/m2 daily for 28 days in Consolidation (second phase of therapy), 6TG at 60 mg/m2 daily for 14 days in Delayed Intensification (fourth phase), and 6MP 75 mg/m2 daily continuously in Maintenance (final phase) [13]. At diagnosis, SRS reported TPMT*1/*1 (Normal Metabolizer, NM) and NUDT15*1/*6 (Indeterminate Metabolizer) using the genotype–phenotype translation table that was current at that time.

Per protocol, her first exposure to thiopurines occurred in Consolidation, when 6MP was started at standard dosing of 75 mg/m2/day (Figure 2). Though she did not have myelosuppression necessitating holding 6MP or dose reductions, she developed neutropenia thereafter, which delayed the start of Interim Maintenance by 3 weeks. For 3 weeks, her absolute neutrophil count (ANC) ranged between 0.22 × 103/mcL and 0.34 × 103/mcL. She did not have thrombocytopenia during that time.

FIGURE 2.

FIGURE 2

Open in a new tab

White blood cell (WBC) count and absolute neutrophil count (ANC) in the Maintenance phase of chemotherapy with time points marked where mercaptopurine (6MP) was held, resumed, or reduced in dose (A). 6‐TGN and 6‐MMP metabolites in the Maintenance phase of chemotherapy with time points marked where mercaptopurine (6MP) was held, resumed, or reduced in dose (B). During Maintenance, 6MP dosing was adjusted based on ANC and platelet counts, per protocol.

Her second exposure to thiopurines occurred in Delayed Intensification, at which time she received 6TG on Days 29 through 42 at standard dosing of 60 mg/m2/day. She also received cyclophosphamide 1000 mg/m2 on Day 29, intrathecal methotrexate 12 mg on Day 29, and cytarabine 75 mg/m2/dose daily on Days 29–32 and Days 36–39. On Day 39, she was admitted for fever and severe neutropenia with ANC 0.00 × 103/mcL. In addition, she had anemia with Hgb 6.5 g/dL and thrombocytopenia with platelets 31 × 103/mcL. Per protocol, 6TG was not held. She was admitted again for fever and severe neutropenia on Day 45, immediately after completing 6TG. At that time, her ANC was 0.00 × 103/mcL. She also had anemia with Hgb 9.8 g/dL and thrombocytopenia with platelets 1 × 103/mcL. During this 10‐day admission, she was also found to have a peritonsillar phlegmon requiring an additional 10 days of antibiotics after discharge.

In Maintenance Cycle 1, 6MP was started at standard dosing of 75 mg/m2/day. On Cycle 1 Day 29, her ANC and platelets had decreased from 1.56 × 103/mcL to 1.28 × 103/mcL and from 329 × 103/mcL to 94 × 103/mcL, respectively, but she did not require any dose reductions. At that time, 6‐thioguanine nucleotides (6‐TGN) were at the upper end of normal at 445 pmol/8 × 108 RBC (target range 235–450 pmol/8 × 108 RBC), and 6‐methylmercaptopurine (6‐MMP) was elevated at 9200 pmol/8 × 108 RBC (target ≤ 5700 pmol/8 × 108 RBC). On Cycle 1 Day 47, she developed increased bruising and was found to have severe neutropenia with ANC 0.03 × 103/mcL and thrombocytopenia with platelets 5 × 103/mcL. 6‐TGN were relatively unchanged at 440 pmol/8 × 108 RBC, but 6‐MMP had significantly increased to 16,177 pmol/8 × 108 RBC. Liver function tests were within normal limits during this time. There were no other apparent causes of myelosuppression at that time, including recent infections or concern for relapse. Though the Naranjo Adverse Drug Reaction Probability Scale is not well validated for chemotherapy polypharmacy, assessment for this event results in a “Probable” attribution for 6MP and “Possible” for methotrexate [14]. Throughout Maintenance, the family verbalized the correct doses of oral chemotherapy and denied any missed doses. Given the degree of myelosuppression, 6MP was held. Her ANC and platelets did not recover until Cycle 1 Day 64, at which time 6MP was restarted at 50% of the original dose.

At the start of Maintenance Cycle 2, her ANC had risen to 2.43 × 103/mcL and platelets to 251 × 103/mcL. At that time, 6‐TGN and 6‐MMP had both normalized to 179 pmol/8 × 108 RBC and 1570 pmol/8 × 108 RBC, respectively. No dose changes were made at that time. However, on Maintenance Cycle 2 Day 57, her ANC and platelets had decreased to 0.43 × 103/mcL and 152 × 103/mcL, respectively. 6‐TGN and 6‐MMP were relatively stable at 193 pmol/8 × 108 RBC and 1247 pmol/8 × 108 RBC, respectively. Given her severe neutropenia, 6MP was held. On Maintenance Cycle 2 Day 63, her ANC had decreased further to 0.12 × 103/mcL, as well as her platelets to 142 × 103/mcL. For this reason, 6MP continued to be held. On Maintenance Cycle 2 Day 78, her ANC and platelets had recovered to 1.58 × 103/mcL and 205 × 103/mcL, respectively. Thus, 6MP was restarted at 36% of the original dose.

On Maintenance Cycle 3 Day 1, her ANC and platelets had decreased to 0.69 × 103/mcL and 187 × 103/mcL, respectively. Given the decrease in her ANC so quickly after restarting, 6MP was reduced to 27% of the original dose. She has not received allopurinol during therapy except for the first few days of Induction while monitoring closely for tumor lysis syndrome. She did receive trimethoprim‐sulfamethoxazole throughout treatment for prophylaxis per supportive care guidelines.

Among 1011 unrelated individuals with LRS data enrolled in the GA4K research study, there were 3 individuals that carried the NUDT15*6 allele. All were NUDT15*1/*6 and identified as White. Thus, the frequency of the NUDT15*6 allele in the overall cohort was 0.15%, and 0.24% among the White unrelated individuals. In a separate cohort of 339 cancer patients with SRS, there were six individuals with the NUDT15*6 allele all of which were NUDT15*1/*6. The frequency of the NUDT15*6 allele in this cohort was 0.88%; five of these individuals were White Non‐Hispanic and one was White Hispanic.

4. Discussion

Thiopurines and their active metabolites that interfere with the growth of cancer cells are an integral part of therapy for pediatric ALL and will likely remain so, even with the addition of targeted agents. When patients with impaired NUDT15 activity are given the same doses of thiopurines as patients with normal activity, they are at risk of developing toxicity resulting in treatment delays, severe infections, and dose reductions. In turn, treatment delays can increase the risk of relapse. Based on our observations, the NUDT15*6 allele seems to have little to no enzyme activity, and patients with the NUDT15*1/*6 genotype should be treated as IMs, as recently updated by CPIC. Patients with this genotype likely have a higher risk of toxicity compared to those having two normal function alleles (NM) and may benefit from dose reductions to minimize toxicities.

Though it is difficult to definitively attribute this patient's severe neutropenia during Delayed Intensification only to 6TG at a time when she was receiving other myelosuppressive agents, she continued to have severe neutropenia during Maintenance, when the only significant myelosuppressive medications given were 6MP and oral methotrexate. Per protocol, oral methotrexate was also held whenever 6MP was held in Maintenance. However, the dose of methotrexate was not reduced when resumed. Now in Maintenance Cycle 6, the patient is currently continuing to tolerate mercaptopurine at 28% dosing and methotrexate at 100% dosing without toxicity, suggesting that the neutropenia is more likely related to 6MP than methotrexate. In addition, the 6‐TGN concentrations in Delayed Intensification were near the upper limit of what our reference lab considers normal but were above the limit of 400 pmol/8 × 108 RBCs that some other reference labs consider the upper limit of normal. Previous reports demonstrate that the 6‐TGN concentrations associated with leukopenia in NUDT15 IMs may be lower than in NUDT15 NMs. This may be due to increased incorporation of the 6‐TGN into the DNA in NUDT15 IMs, such that equivalent 6‐TGN concentrations are associated with increased toxicity in NUDT15 IMs compared to NMs [15, 16]. This suggests that DNA‐TG may be a better predictor of toxicity than 6‐TGN concentrations in patients with NUDT15 intermediate or poor metabolizer phenotypes.

The inability of this patient to tolerate full doses of thiopurines is consistent with prior reports of NUDT15*6 carriers that were not able to tolerate the standard dose [17, 18, 19], the recent Dutch Pharmacogenetic Working Group guideline that categorizes NUDT15*6 as “strongly reduced function” [20], and the recent CPIC guideline that categorizes NUDT15*1/*6 individuals as IMs. Prior reports have also found that patients having a NUDT15*6 allele exhibit increased incidence of adverse reactions and increased therapy interruption compared to NUDT15*1/*1 patients [18, 21, 22, 23, 24, 25]. In addition, in vitro studies have demonstrated that NUDT15.6 exhibits decreased enzymatic activity and reduced protein stability compared to NUDT15.1 [26, 27].

Many haplotypes have been described for the NUDT15 gene, of which several have a 6‐bp insertion or deletion (Figure 1). Of note, when the NUDT15*6‐defining insertion is found in cis with c.415C > T (p.R139C), the resulting haplotype is now designated NUDT15*3.002 (PharmVar recently merged NUDT15*2 with *3) [10]. In cases when both variants are heterozygous, targeted panel genotyping and SRS cannot determine whether they are in cis or in trans. Consequently, the diplotype may be either NUDT15*1/*3 (variants on the same allele) or NUDT15*3/*6 (variants on opposite alleles) (Figure 1). Within the SRS cohort, six individuals had a NUDT15*1/*3 (legacy *1/*2) diplotype. However, some or all of these patients may have the alternate NUDT15*3/*6 genotype, which would further increase the frequency of the NUDT15*6 allele in this cohort. Since LRS is able to resolve the ambiguity in whether a patient is NUDT15*1/*3 or *3/*6, frequencies determined by LRS are more accurate compared to those determined by targeted sequencing or short‐read sequencing, which may have underestimated the frequency of NUDT15*6 in the past. Our study includes the NUDT15*6 allele frequencies from both SRS in cancer patients (0.88%) and LRS in individuals with suspected genetic disorders (0.15%), with most individuals having the NUDT15*6 allele being White, non‐Hispanic. The observed 0.24% frequency among unrelated White individuals with LRS is in line with what has been previously reported for individuals of European ancestry at 0.28% [28].

As in any retrospective analysis of toxicity, there are limitations to this report. It is not possible for us to definitively attribute the myelosuppression experienced by this patient to thiopurines, as the patient was also taking other myelosuppressive drugs concurrently. While our cohorts for sequencing are large, they are not very diverse and do not have exposure to thiopurines to assess the function of the NUDT15*6 allele in a larger cohort.

In conclusion, this brief report demonstrates that a large dose reduction was necessary in a patient with TPMT *1/*1 and NUDT15*1/*6, supporting the recently updated translation of this NUDT15 diplotype as IM rather than indeterminate in the future. Patients carrying NUDT15*1/*6 should be closely monitored for thiopurine‐related toxicities and may require significant dose reductions.

Author Contributions

All authors wrote the manuscript; all authors performed the research; J.F., L.A.L., M.S.F., L.B.R., S.J., E.C.B. and A.G. analyzed the data; J.F., E.C.B., K.A., L.B.R. and A.G. designed the research.

Funding

The authors have nothing to report.

Conflicts of Interest

L.B.R. has received grant support and performed consulting for B.T.G. Specialty Pharmaceuticals. All other authors have nothing to declare.

Acknowledgments

We would like to thank the families for participating in the Children's Mercy Research Institute Biorepository (Tumor Bank) and Genomic Answers for Kids programs, which are made possible by the generous gifts to Children's Mercy Research Institute at Children's Mercy Kansas City. Genomic Answers for Kids provided salary support for research infrastructure and staff. Pacific Biosciences provided in‐kind contribution for the high‐fidelity genome sequencing reagents. We also thank the leadership and administrative staff of the GA4K project for their continued support.

References

  • 1. Evans W. E., “Pharmacogenetics of Thiopurine S‐Methyltransferase and Thiopurine Therapy,” Therapeutic Drug Monitoring 26, no. 2 (2004): 186–191. [DOI] [PubMed] [Google Scholar]
  • 2. Sandborn W. J., “Pharmacogenomics and IBD: TPMT and Thiopurines,” Inflammatory Bowel Diseases 10, no. 1 (2004): S35–S37. [DOI] [PubMed] [Google Scholar]
  • 3. Weinshilboum R., “Inheritance and Drug Response,” New England Journal of Medicine 348 (2003): 529–537. [DOI] [PubMed] [Google Scholar]
  • 4. Ford L. T. and Berg J. D., “Thiopurine S‐Methyltransferase (TPMT) Assessment Prior to Starting Thiopurine Drug Treatment; a Pharmacogenomic Test Whose Time Has Come,” Journal of Clinical Pathology 63 (2010): 288–295. [DOI] [PubMed] [Google Scholar]
  • 5. Yang J. J., Landier W., Yang W., et al., “Inherited NUDT15 Variant Is a Genetic Determinant of Mercaptopurine Intolerance in Children With Acute Lymphoblastic Leukemia,” Journal of Clinical Oncology 33 (2015): 1235–1242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Appell M.L., “TPMT nomenclature committee (TPMT alleles),” (2025), https://liu.se/en/research/tpmt‐nomenclature‐committee.
  • 7. Gaedigk A., Casey S. T., Whirl‐Carrillo M., Miller N. A., and Klein T. E., “Pharmacogene Variation Consortium: A Global Resource and Repository for Pharmacogene Variation,” Clinical Pharmacology and Therapeutics 110 (2021): 542–545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Yang J. J., Whirl‐Carrillo M., Scott S. A., et al., “Pharmacogene Variation Consortium Gene Introduction: NUDT15,” Clinical Pharmacology and Therapeutics 105 (2019): 1091–1094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Maillard M., Schwab M., Whirl‐Carrillo M., et al., “Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for Thiopurine Dosing Based on TPMT and NUDT15 Genotypes: 2025 Update,” Clinical Pharmacology and Therapeutics 119 (2026): 916–927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. PharmVar , “NUDT15,” (2026), https://www.pharmvar.org/gene/NUDT15.
  • 11. ClinPGx , NUDT15 Allele Functionality Table (ClinPGx, 2026), https://files.cpicpgx.org/data/report/current/allele_function_reference/NUDT15_allele_functionality_reference.xlsx. [Google Scholar]
  • 12. Danecek P., Bonfield J. K., Liddle J., et al., “Twelve Years of SAMtools and BCFtools,” GigaScience 10 (2021): giab008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Gupta S., Rau R. E., Kairalla J. A., et al., “Blinatumomab in Standard‐Risk B‐Cell Acute Lymphoblastic Leukemia in Children,” New England Journal of Medicine 392 (2025): 875–891. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Naranjo C. A., Busto U., Sellers E. M., et al., “A Method for Estimating the Probability of Adverse Drug Reactions,” Clinical Pharmacology and Therapeutics 30 (1981): 239–245. [DOI] [PubMed] [Google Scholar]
  • 15. Zhu X., Chao K., Li M., et al., “Nucleoside Diphosphate‐Linked Moiety X‐Type Motif 15 R139C Genotypes Impact 6‐Thioguanine Nucleotide Cut‐Off Levels to Predict Thiopurine‐Induced Leukopenia in Crohn's Disease Patients,” World Journal of Gastroenterology 38 (2019): 5850–5861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Moriyama T., Nishii R., Lin T. N., et al., “The Effects of Inherited NUDT15 Polymorphisms on Thiopurine Active Metabolites in Japanese Children With Acute Lymphoblastic Leukemia,” Pharmacogenetics and Genomics 6 (2017): 236–239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Wang D. S., Yu C. H., Lin C. Y., et al., “Childhood Acute Lymphoblastic Leukemia Mercaptopurine Intolerance Is Associated With NUDT15 Variants,” Pediatric Research 89 (2021): 217–222. [DOI] [PubMed] [Google Scholar]
  • 18. Fan P. O., Leung K. T., Chan K. Y., et al., “ABCC4, ITPA, NUDT15, TPMT and Their Interaction as Genetic Predictors of 6‐Mercaptopurine Intolerance in Chinese Patients With Acute Lymphoblastic Leukemia,” Pediatric Hematology and Oncology 39 (2022): 254–266. [DOI] [PubMed] [Google Scholar]
  • 19. Yu C. H., Chang Y. H., Wang D. S., et al., “Determination of NUDT15 Variants by Targeted Sequencing Can Identify Compound Heterozygosity in Pediatric Acute Lymphoblastic Leukemia Patients,” Scientific Reports 10 (2020): 14400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Coenen M. J., Nijenhuis M., Soree B., et al., “Dutch Pharmacogenetics Working Group (DPWG) Guideline for the Gene‐Drug Interaction Between TPMT/NUDT15 and Thiopurines,” European Journal of Human Genetics 34 (2026): 152–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Yi E. S., “NUDT15 Variants Cause Hematopoietic Toxicity With Low 6‐TGN Levels in Children With Acute Lymphoblastic Leukemia,” Cancer Research and Treatment 50 (2018): 872–882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Schaeffeler E., Jaeger S. U., Klumpp V., et al., “Impact of NUDT15 Genetics on Severe Thiopurine‐Related Hematotoxicity in Patients With European Ancestry,” Genetics in Medicine 21 (2019): 2145–2150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Suzuki S., Uchiyama K., Motoi Y., et al., “Analysis of the NUDT15 Gene and Metabolites of Azathioprine in Japanese Patients With Inflammatory Bowel Disease,” BMC Gastroenterology 23 (2023): 239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Nguyen M. L., Hoang A. V., Duong B. T., and Phung N. T. N., “The Influence of NUDT15 Variants on 6‐Mercaptopurine‐Induced Neutropenia in Vietnamese Pediatric Acute Lymphoblastic Leukemia,” Human Genetics and Genomics Advances 4 (2023): 100183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Walker G. J., Harrison J. W., Heap G. A., et al., “Association of Genetic Variants in NUDT15 With Thiopurine‐Induced Myelosuppression in Patients With Inflammatory Bowel Disease,” JAMA 321 (2019): 773–785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Moriyama T., Nishii R., Perez‐Andreu V., et al., “NUDT15 Polymorphisms Alter Thiopurine Metabolism and Hematopoietic Toxicity,” Nature Genetics 48 (2016): 367–373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Rehling D., Zhang S. M., Jemth A. S., et al., “Crystal Structures of NUDT15 Variants Enabled by a Potent Inhibitor Reveal the Structural Basis for Thiopurine Sensitivity,” Journal of Biological Chemistry 296 (2021): 100568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. ClinPGx , NUDT15 Frequency Table (ClinPGx, 2026), https://files.cpicpgx.org/data/report/current/frequency/NUDT15_frequency_table.xlsx. [Google Scholar]

Articles from Clinical and Translational Science are provided here courtesy of Wiley

RESOURCES