PERSONALIZATION OF 6-MERCAPTOPURINE THERAPY FOR CHILDREN WITH ACUTE LYMPHOBLASTIC LEUKEMIA. Issue 1 (14th December 2015)

• Record Type: Journal Article
• Authors: Milošević, Goran; Dokmanović, Lidija; Krstovski, Nada; Lazić, Jelena; Rodić, Predrag; Pavlović, Sonja; Zukić, Branka; Kotur, Nikola; Janić, Dragana
• Is Part Of: Archives of disease in childhood. Volume 101:Issue 1(2016)
• Page Start: e1
• Page End: e1
• Publication Date: 2015-12-14
• Subjects: ESDP; Children -- Diseases -- Periodicals; Infants -- Diseases -- Periodicals; 618.920005
• DOI: 10.1136/archdischild-2015-310148.13
• Languages: English
• ISSNs: 0003-9888

Abstract

Abstract : Introduction: Treatment of acute lymphoblastic leukemia (ALL) in children has been significantly improved over past few decades. Optimal use of known antileukemic drugs has lead to four-fold increase of overall survival. Nevertheless, therapy failure occurs in 15–20% of patients. Individualized use of cytostatic drugs can help us reducing toxicity-related deaths and improve therapy outcome. Materials and methods: We studied multidrug resistance protein 4 gene (ABCC4) and inosin triphosphate pyrophosphatase (ITPA) gene polymorphisms to assess their influence on occurrence and intensity of 6-MP toxicity during maintenance phase of ALL treatment in children as well as on treatment outcome. Results: Thiopurine S-methyltransferase (TPMT) genotype was known for all patients before start of therapy and doses were adjusted accordingly. Five patients (7%) were carriers of ITPAc.94C>A genotype. Heterozygote and homozygote carriers of variant alleles for ABCC4 rs9516519 T>G, genotype were represented in 35% and 4% of our patients, respectively. We did not show correlation of variant alleles with increased number of off-therapy weeks neither with number of episodes of leukopenia. Cross-validated AUC for each of genetic markers individually showed low predictive power (cross-validated AUC around 0.5) for predicting toxicity. However, combination of two genetic markers and known TPMT gentotype improved the predictability (cross-validated AUC was 0.65). Conclusions: Our results did not show correlation with toxicity using individual genetic markers. However, in the settings of known TPMT genotype, we showed the potential of using all three genetic markers (including TPMT) in predicting who might be in danger of having increased toxicity.