EUVIMEDEUVIMED entdecken
MetaanalyseOpen Access · CC BY-NC-ND 4.0

NUDT15 R139C is a specific but insensitive pre-treatment screen for thiopurine-induced leukopenia: a meta-analysis of eight cohorts from six countries.

Garg Pardeep, Jindal Amita, Bansal Nagma, Bhullar Harkiratjit, Kukkar Latika

SJ Pediatrics and Child Health Africa · 01.11.2026 · 3 (4) · 14

DOI: 10.51168/4bvv6j64 ↗

Kurz gefasst

Metaanalyse in der Zeitschrift SJ Pediatrics and Child Health Africa, online erschienen am 01.11.2026. Verfasst von Garg Pardeep und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Loss-of-function variation in NUDT15 causes thiopurine-induced leukopenia, and genotype-based dosing recommendations exist. Earlier syntheses pooled East Asian cohorts almost exclusively, and the performance of the variant as a screening test in other populations has not been examined. Methodology: A curated set of full-text studies on NUDT15 and thiopurine toxicity held by the authors was analysed, checked against a PubMed and Google Scholar search to July 2026. Studies reporting leukopenia by NUDT15 c.415C>T (rs116855232) genotype in azathioprine- or mercaptopurine-treated patients were eligible. Two-by-two tables were pooled by DerSimonian–Laird random-effects meta-analysis, with accuracy and absolute-risk analyses restricted by study design. Certainty was rated by GRADE. The analysis was not registered, and the search was not exhaustive.

Results

Eight studies (1 702 patients, 355 events) from six countries were included. Carriage was associated with leukopenia (pooled odds ratio 5.70, 95% CI 3.44–9.45; prediction interval 1.48–21.99; moderate certainty) and raised absolute risk by 38.7 percentage points (95% CI 28.2–49.1). Excluding the exposure-matched cohort, pooled sensitivity was 40.3% (95% CI 29.9–51.6) and specificity 89.9% (95% CI 85.6–93.1). Sensitivity was 49.5% in East and Southeast Asian cohorts and 28.0% in Indian cohorts (p=0.009; very low certainty). The number genotyped per excess episode was 12 in East and Southeast Asian populations, 23 in South Asian populations, and about 774 in European populations.

Conclusion

The association holds outside East Asia, but the variant is a rule-in rather than a rule-out marker: a normal R139C result should not be read as permission to relax haematological monitoring. The lower sensitivity in Indian cohorts is hypothesis-generating. Recommendation: Prospective multi-ancestry cohorts genotyping a full NUDT15 panel alongside TPMT, with a common leukopenia definition, are needed before an ancestry-specific testing policy is set.

Zitieren

APA

Pardeep, G., Amita, J., Nagma, B., Harkiratjit, B., & Latika, K. (2026). NUDT15 R139C is a specific but insensitive pre-treatment screen for thiopurine-induced leukopenia: a meta-analysis of eight cohorts from six countries. SJ Pediatrics and Child Health Africa, 3(4), 14. https://doi.org/10.51168/4bvv6j64

Vancouver

Pardeep G, Amita J, Nagma B, Harkiratjit B, Latika K. NUDT15 R139C is a specific but insensitive pre-treatment screen for thiopurine-induced leukopenia: a meta-analysis of eight cohorts from six countries. SJ Pediatrics and Child Health Africa. 2026;3(4):14. doi:10.51168/4bvv6j64

BibTeX
@article{Pardeep2026,
  author = {Pardeep, Garg and Amita, Jindal and Nagma, Bansal and Harkiratjit, Bhullar and Latika, Kukkar},
  title = {NUDT15 R139C is a specific but insensitive pre-treatment screen for thiopurine-induced leukopenia: a meta-analysis of eight cohorts from six countries},
  journal = {SJ Pediatrics and Child Health Africa},
  year = {2026},
  volume = {3},
  number = {4},
  pages = {14},
  doi = {10.51168/4bvv6j64}
}
Kostenlos starten →