Prognostic Significance of the Null Genotype of Glutathione S

Leukemia (2002) 16, 203–208  2002 Nature Publishing Group All rights reserved 0887-6924/02 $25.00 www.nature.com/leu Prognostic signiﬁcance of the null genotype of glutathione S-transferase-T1 in patients with acute myeloid leukemia: increased early death after chemotherapy T Naoe1, Y Tagawa1, H Kiyoi1, Y Kodera2, S Miyawaki3, N Asou4, K Kuriyama5, S Kusumoto6, C Shimazaki7, K Saito8, H Akiyama9, T Motoji10, M Nishimura11, K Shinagawa12, R Ueda13, H Saito14 and R Ohno15

1Department of Infectious Diseases, Nagoya University School of Medicine, Nagoya, Japan; 2Department of Medicine, Japanese Red Cross Nagoya First Hospital, Nagoya, Japan; 3Department of Medicine, Saiseikai Maebashi Hospital, Maebashi, Japan; 4Second Department of Internal Medicine, Kumamoto University School of Medicine, Kumamoto, Nagasaki, Japan; 5Department of Hematology, Atomic Disease Institute Nagasaki University School of Medicine, Nagasaki, Japan; 6First Department of Internal Medicine, Saitama Medical School, Saitama, Japan; 7Second Department of Medicine, Kyoto Prefectural University of Medicine, Kyoto, Japan; 8Third Department of Internal Medicine, Dokkyo University School of Medicine, Tochigi, Japan; 9Hematology Division, Tokyo Metropolitan Komagome Hospital, Tokyo, Japan; 10Department of Hematology, Tokyo Women’s Medical University, Tokyo, Japan; 11Second Department of Internal Medicine, Chiba University School of Medicine, Chiba, Japan; 12Department of Medicine, Okayama University Medical School, Okayama, Japan; 13Second Department of Internal Medicine, Nagoya City University School of Medicine, Nagoya, Japan; 14Nagoya National Hospital, Nagoya, Japan; and 15Aichi Cancer Center, Nagoya, Japan

We investigated the prognostic significance of genetic myeloperoxidase (MPO), the products of which are associated polymorphism in glutathione-S transferase mu 1 (GSTM1), glut- with drug metabolism as well as with detoxication, in de novo athione-S transferase theta 1 (GSTT1), NAD(P)H:quinone oxidoreductase (NQO1) and myeloperoxidase (MPO), the products acute myeloid leukemia (AML). Cytochrome P450 (CYP) 3A4 of which are associated with drug metabolism as well as with is also involved in metabolizing anti-cancer drugs but was detoxication, in 193 patients with de novo acute myeloid leuke- not studied due to lack of polymorphism in the Japanese mia (AML) other than M3. Of the patients, 64.2% were either population.8 homozygous or heterozygous for GSTT1 (GSTT1+), while 35.8% showed homozygous deletions of GSTT1 (GSTT1−). The GSTT1− group had a worse prognosis than the GSTT1+ group (P = 0.04), whereas other genotypes did not affect the outcome. Materials and methods Multivariate analysis revealed that GSTT1− was an independent prognostic factor for overall survival (relative risk: 1.53; P = Patients 0.026) but not for disease-free survival of 140 patients who achieved complete remission (CR). The rate of early death after the − The subjects included 193 patients with previously non- initiation of chemotherapy was higher in the GSTT1 group than treated de novo AML other than M3, who were registered for the GSTT1+ group (within 45 days after initial chemotherapy, P = 0.073; within 120 days, P = 0.028), whereas CR rates and the AML protocols conducted by the Japan Adult Leukemia relapse frequencies were similar. The null genotype of GSTT1 Study Group (JALSG). Twenty-six, 39 and 128 patients were might be associated with increased toxicity after chemo- treated with the AML-87, AML-89 and AML-92 protocols, therapy. respectively.9–11 AML samples were provided through 16 hos- Leukemia (2002) 16, 203–208. DOI: 10.1038/sj/leu/2402361 pitals (about a quarter of JALSG institutes) for the purpose of Keywords: glutathione S-transferase; polymorphism; acute molecular study after informed consent. AML was diagnosed myeloid leukemia; prognosis according to the FAB classification, which was evaluated by a central review committee. Patients whose performance status was 3 or 4 (ECOG classification) were excluded from Introduction the registration. In the AML-87 study,9 induction therapy consisted of daily In acute leukemia, prognosis depends on biological character- behenoyl cytarabine (BHAC) 200 mg/m2, daily 6-mercaptopu- istics including morphology (FAB classification), immuno- rine (6-MP) 70 mg/m2, daily prednisolone (PRD) 40 mg/m2 phenotype, peripheral white blood cell (WBC) counts, and daunorubicin (DNR) 40 mg/m2 on days 1 to 3, and if chromosomal alterations, and many other molecular and necessary on days 7, 8 and 11. The therapy was continued 1–3 phenotypic markers of leukemia cells. Host-sided factors for a 10- to 12-day period until the bone marrow became including the patient’s age and performance status are also severely hypoplastic with less than 5% blasts. In the AML-89 2–5 associated with the prognosis. Genetic polymorphisms study,10 patients were randomized to receive induction ther- have been examined focusing on individual differences in apy that included BHAC (200 mg/m2 by 3 h infusion) or cytar- 6 pharmaco-dynamics, response, and the side-effects of drugs. abine (AraC, 80 mg/m2 by continuous infusion). BHAC or However, it remains to be elucidated whether genetic poly- AraC, and 6-MP 70 mg/m2 were administrated for 10 to 12 morphisms influence the prognosis of leukemia after chemo- days, and DNR 40 mg/m2 was given on days 1 to 4, and if therapy. Recently it was reported that the null genotype of necessary, on days 10 to 12 in addition to the above schedule glutathione-S transferase theta 1 (GSTT1) was associated with for AML-87. In the AML-92 study,11 patients were randomized 7 prognosis in childhood acute myeloid leukemia (AML). to receive BHAC-DM similar to the AML-87 protocol with or In this study, we analyzed the prognostic significance of without etoposide (ETP, 100 mg/m2 for 5 days). After complete gene polymorphism of GSTT1, glutathione-S transferase mu remission (CR) was achieved, three courses of consolidation 1 (GSTM1), NAD(P)H:quinone oxidoreductase (NQO1) and chemotherapy and six courses of intensification chemotherapy were given. Patients 60 years or older received about two-thirds of the dosage of each drug throughout the study Correspondence: T Naoe, Department of Infectious Diseases, Nagoya University School of Medicine, Tsurumai-cho 65, Showa-ku, Nagoya period. 466-8560, Japan; Fax: +81-52-744-2801 CR was defined as less than 5% blasts in normo-cellular Received 27 June 2001; accepted 12 October 2001 bone marrow with normal levels of peripheral neutrophils and GSTT1 genotype and early death in AML T Naoe et al 204 platelet counts. Overall survival was calculated from the first were 5-ACACAACTGTGTTCACTAGC-3 and 5-CAACTTCATC day of therapy to death. Disease-free survival (DFS) for CACGTTCACC-3. The MPO gene polymorphism located 463 patients who had achieved CR was measured from the date bp upstream of exon 1 in the promoter region was exam- of CR to relapse or death. Relapse-free survival was defined ined.12 Briefly, a 567-bp DNA fragment was amplified using as the time from the date of CR to relapse or death from pro- 5Ј-AGGCCAATTGGGTCATCTTTACTC-3Ј and 5Ј-GACGGTT gressive disease, censoring deaths from other causes. Patients ATCTTGCTCTGTT-3Ј, with a second amplification using 5Ј- who underwent bone marrow transplantation (BMT) were AGGAACCCTGGATAAACAGTGTAACC-3Ј and 5Ј-GCCTCTA censored from the date of BMT. GCCACATCATCAATT-3Ј. The reaction comprised 30 cycles of 94°C for 30 s, 55°C for 2 min and 72°C for 2 min. An additional cycle was performed at 72°C for 10 min. The final Genotyping PCR products were digested with AciI (New England Biolabs). In cases of G/G and A/A at 463, two fragments (189 and 154 The genotyping procedure for NQO1, GSTM1 and GSTT1 bp) and one fragment (343 bp) were obtained, respectively. was described previously.8 Briefly, for the amplification of the NQO1 gene fragment, the sense and anti-sense primers were 5-AGTGGCATTCTGCATTTCTGTG-3 and 5-GATGGACTT GCCCAAGTGATG-3, respectively. The amplification was car- Statistical analysis ried out in a thermocycler (model 9600; Applied Biosystems, Foster City, CA, USA) with an initial denaturation step (8 min, 95°C), followed by 35 cycles consisting of three steps: 94°C Survival probabilities were estimated by the Kaplan–Meier for 30 s, 56°C for 1 min and 72°C for 2 min. An additional method, and differences in the distributions between the cycle was performed at 72°C for 10 min. The amplified frag- genotypes were evaluated using the log-rank test. The prog- ments were digested with HinfI endonuclease (New England nostic significance of the clinical variables was assessed using Biolabs, Beverly, MA, USA) and analyzed on agarose gel the Cox proportional hazards model. These statistic analyses electrophoresis. The paired primers for GSTM1 and GSTT1 were performed with StatView software (Abacus Concepts, were 5-GAACTCCCTGAAAAGCTAAAGC-3 and 5-GTTGGG Berkeley, CA, USA). For all analyses, P values were two-tailed, CTCAAATATACGGTGG-3, and 5-TTCCTTACTGGTCCTCAC and a P value of less than 0.05 was considered significant. ATCTC-3 and 5-TCACCGGATCATGGCCAGCA-3, respect- Since CR rates, overall survival and DFS according to ively. The presence or absence of GST genes was determined induction therapy (AML-87, -89 and -92) did not show any by a differential PCR in which the ␤-globin gene was co- differences, the data of the three studies were combined and amplified in the same reaction tube. The primers for ␤-globin analyzed.

Table 1 Clinical and molecular characteristics of 193 patients with de novo AML except M3 according to GSTT1 genotype

Factora GSTT1+ GSTT1− P n = 124 n = 69

Clinical characteristic Age (y) median 50.5 47 0.56 Sex F 51 30 0.75 M7339 FAB M0 0 3 0.13 M1 33 13 M2 51 30 M4 28 17 M5 10 3 M6 2 2 M7 0 1 WBC median 24.4 24.3 0.8 (× 106/l) Karyotype favorable 20 13 0.59 intermediate 77 39 unfavorable 10 6 ND 17 11 Germ line polymorphism GST-M1 present 48 29 0.65 absent 76 40 NQO1 P/P 58 31 0.47 P/S 49 24 S/S 17 14 MPOb A/A 1 0 0.49 A/G 18 11 G/G 105 57 Therapy AML-87 16 10 0.54 AML-89 28 11 AML-92 80 48

aFactors. Analysis of frequencies was performed using Fisher’s exact test for 2 × 2 tables or Pearson’s ␹2 test for larger tables. Differences in median variables in age and WBC counts were analyzed with the Wilcoxon rank-sum test. bMPO not ampliﬁed in one case.

Leukemia GSTT1 genotype and early death in AML T Naoe et al 205

Figure 2 Accumulation curves of relapse according to GSTT1 genotype. The frequency of relapse at 50 months was similar in the GSTT1+ and GSTT1− groups (53.2% vs 54.2%, P = 0.39 by the log- rank test).

Table 3 Clinical outcome of 193 patients with de novo AML except M3 after induction therapy

GSTT1+ GSTT1− P n = 124 (%) n = 69 (%)

Outcome CR 94 (75.8) 46 (66.7) 0.17 failure 30 (24.2) 23 (33.3) Death Figure 1 Kaplan–Meier curves of overall survival and DFS accord- within 45 11 (8.9) 12 (17.4) 0.073 ing to GSTT1 genotype. The GSTT1− group (n = 69) had a worse over- daysa all survival than the GSTT+ group (n = 124). The predicted overall within 60 13 (10.5) 14 (20.3) 0.054 survival rates at 50 months were 33.7% and 22.1% in the GSTT1+ days and GSTT1− groups, respectively. DFS in 140 patients who achieved within 120 17 (13.7) 19 (27.5) 0.028 CR did not differ between GSTT1+ (n = 94) and GSTT1− (n = 46) groups days (P = 0.22). aDays after initiation of induction therapy. Table 2 Unfavorable prognostic factors for overall survival in 193 patients with de novo AML ties. There was no association among the four genotypes Factor P value RR 95% CI (GSTT1, GSTM1, NQO1 and MPO) (Table 1). At a median follow-up time of 50 months, 64 (33.2%) − Clinical Age у 60 0.0008 1.93 1.32–2.84 patients were alive. The GSTT1 group had a worse overall + characteristic FAB (not M2 vs 0.15 1.36 0.90–2.06 survival than the GSTT1 group (P = 0.04, by the log-rank test, M2) Figure 1). The predicted overall survival rates at 50 months WBC Ͼ 100 × 106/l 0.06 1.55 0.98–2.44 were 33.7% and 22.1% in the GSTT1+ and GSTT1− groups, Karyotype 0.015 2.60 1.21–5.59 respectively (Figure 1). DFS in 140 patients who achieved CR (unfavorable vs + − favorable) was compared between GSTT1 and GSTT1 groups. How- ever, there was no significant difference (Figure 1). Germ line GSTM1 null 0.47 0.88 0.61–1.26 Multivariate analysis of overall survival revealed that an age polymorphism GSTT1 null 0.026 1.52 1.05–2.18 − of 60 or more, unfavorable karyotype, and GSTT1 were inde- NQO1 (P/P vs S/S) 0.96 0.99 0.59–1.64 × MPO (G/A vs G/G) 0.91 0.97 0.59–1.60 pendent factors for a poor prognosis, and WBC over 100 106/l was a marginal factor (P = 0.06) (Table 2). Multivariate RR, relative risk; CI, confidence interval. analysis of DFS in 150 patients who achieved CR showed that age and karyotype were significant factors, whereas GSTT1− was not significantly associated with a poor prognosis (P = Results 0.19, data not shown). The frequency of relapse at 50 months was similar in the GSTT1+ and GSTT1− groups (53.2% vs On genotyping, 124 of 193 patients (64.2%) were found to 54.2%, P = 0.39 by the log-rank test, Figure 2). be either homozygous or heterozygous for GSTT1 (GSTT1+), To further analyze the significance of the GSTT1 polymor- while 69 patients (35.8%) showed homozygous deletions of phism, clinical outcome was compared between the GSTT1+ GSTT1 (GSTT1−). The GSTT1 genotype was related neither to and GSTT1− groups (Table 3). The CR rate after initial induc- age, sex, FAB subtype, WBC counts, nor karyotype abnormali- tion therapy was lower in the GSTT1− group than the GSTT1+

Leukemia GSTT1 genotype and early death in AML T Naoe et al 206 group, although the difference was not significant (66.7% vs Discussion 75.8%, P = 0.17). Since early death after initial chemotherapy is one of the factors worsening CR and overall survival rates, GSTs are a family of enzymes that play an important role in we studied how many patients died within 45, 60 and 120 detoxification by catalyzing the conjugation of many hydro- days after the initiation of remission induction in each group. phobic and electrophilic compounds with reduced gluta- The rate of early death was higher than in the GSTT1− group thione.13,14 Human GSTs are categorized into four main that GSTT1+ group (17.4% vs 8.9% within 45 days, P = 0.073; classes: alpha (A), mu (M), pi (P) and theta (T). Each class has 20.3% vs 10.5% within 60 days, P = 0.054; 27.5% vs 13.7% a different substrate specificity. Among GST genes, GSTM1 within 120 days, P = 0.028). Bleeding and infection were and GSTT1 have null-allele variants that are commonly found 15,16 common causes of early death in both groups. The GSTT1 in the population and result in a lack of enzyme activity. genotype did not significantly affect the neutrophilic recovery A homozygous defect of GSTM1 or GSTT1 is associated with after chemotherapy (data not shown). Notably, respiratory fail- an increased risk of lung, breast and bladder cancers either ure (n = 3) and cardiac arrhythmia (n = 1) occurred within 45 alone or in combination with other intrinsic or environmental 17–19 days only in the GSTT1− group. factors. The GSTT1 null genotype was reported to increase the risk of myelodysplastic syndrome and acute leu- The GSTM1, NQO1 and MPO genotypes did not influence 7,20,21 the prognosis (Figure 3, Table 2). kemia, although other studies failed to confirm these findings.22,23 In this study, we showed that the GSTT1− genotype was associated with a worse prognosis than the GSTT1+ genotype mainly due to increased early death after initial chemotherapy. Recently Davies et al24 reported that in childhood AML, the GSTT1− genotype was associated with a poor prognosis, and that the frequency of death in remission was increased by the GSTT1− genotype. Our study, in adult patients with AML, essentially confirms this observation and further examined the reason why the GSTT1− genotype wors- ened the prognosis. We noticed that CR, DFS, and relapse rates were not significantly different between the GSTT1+ and GSTT1− groups. The rate of death among patients in remission was similar for the two genotypes (P = 0.25, by the ␹2 test), which is in contrast to the previous paper.24 Notably, the rate of early death was higher for the GSTT1− genotype than the GSTT1+ genotype. Although the number of cases examined was limited, respiratory failure and cardiac arrhythmia were noted within 45 days only in the GSTT1− group. Since the GSTT1 enzyme potentially metabolizes chemotherapeutic agents, both increased responsiveness and toxicity might be expected in the GSTT1− group. Our study clearly indicated that the null genotype of GSTT1 was associated with increased toxicity of chemotherapy but not with a reduced rate of relapse. In malignancies other than leukemia, a GSTM1-null and GSTT1-null genotype was reportedly associated with poor prognosis due to unresponsiveness to primary chemotherapy,25 as distinct from our finding. The significance of GST enzyme may be different in each malignancy and therapy. It is unknown whether GSTT1 enzyme is actually involved in the metabolism of chemotherapeutic agents such as DNR, AraC and BHAC used in AML patients. According to the litera- ture, carcinogens such as methyl chloride, mono-epoxybutane and di-epoxybutane are substrates for GSTT1.26,27 Lympho- cytes with the GSTT1− genotype acquire chromosomal aber- rations more sensitively than the GSTT1+ genotype when exposed to specific mutagenic substrates.18,28–31 Importantly, the increased expression of GST has been shown to be associated with resistance to a range of cytotoxic drugs, including alkylating agents,32 anthracyclines,33 and nitrosoureas.34 Glut- athione (GSH)-related enzymes and the GSH-conjugate export pump are associated with cellular resistance to anti-cancer drugs.35 However, the biological significance of GSTT1 enzyme among GST family is not fully elucidated. There are a number of polymorphic genes whose products are involved in drug metabolism that potentially influence outcome of chemotherapy. In this study, however, only GSTT1 genotype Figure 3 Kaplan–Meier curves of overall survival according to GSTM1, NQO1 and MPO genotypes. had an impact but not GSTM1, MPO and NQO1. A larger clinical study should be carried out to further confirm the sig-

Leukemia GSTT1 genotype and early death in AML T Naoe et al 207 nificance of the GSTT1 genotype in addition to a biological rabine in combination induction and consolidation therapy, and study of the products. with or without ubenimex after maintenance/intensification ther- In conclusion, we report evidence of individual differences apy in adult acute myeloid leukemia. The Japan Leukemia Study Group. J Clin Oncol 1996; 14: 204–213. of outcome after chemotherapy. To further improve chemo- 11 Miyawaki S, Tanimoto M, Kobayashi T, Minami S, Tamura J, therapeutic outcome, a search for the genes influencing the Omoto E, Kuriyama K, Hatake K, Saito K, Kanamaru A, Oh H, outcome of anti-leukemia therapy is required and the Ohtake S, Asou N, Sakamaki H, Yamada O, Jinnai I, Tsubaki K, dose/schedule of chemotherapy should be determined based Takeyama K, Hiraoka A, Matsuda S, Takahashi M, Shimazaki C, on the individual differences using the gene polymorphisms. Adachi K, Kageyama S, Ohno R, for the Japan Adult Leukemia Study Group. No beneficial effect from addition of etoposide to daunorubicin, cytarabine, and 6-mercaptopurine in individualized induction therapy of adult acute myeloid leukemia: the JALSG- Acknowledgements AML92 study. Japan Adult Leukemia Study Group. Int J Hematol 1999; 70: 97–104. This study was supported by a Grant-in Aid (No. 9–2) from 12 Cascorbi I, Henning S, Brockmoller J, Gephart J, Meisel C, Muller the Japanese Ministry of Health and Welfare. We are grateful JM, Loddenkemper R, Roots I. Substantially reduced risk of cancer of the aerodigestive tract in subjects with variant-463A of the mye- to members of the Japan Adult Leukemia Study Group for pro- loperoxidase gene. Cancer Res 2000; 60: 644–649. viding patients’ samples. We also thank Ms Yoko Kudo for 13 Wilce M, Parker M. Structure and function of glutathione S-trans- preparing the manuscript. ferases. Biochim Biophys Acta 1994; 1205: 1–18. 14 Hayes J, Pulford D. The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to References cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 1995; 30: 445–600. 15 Board PG. Biochemical genetics of glutathione-S-transferase in 1 Appelbaum F. Molecular diagnosis and clinical decisions in adult man. Am J Hum Genet 1981; 33: 36–43. acute leukemia. Semin Hematol 1999; 36: 401–410. 16 Pemble S, Schroeder KR, Spencer SR, Meyer DJ, Hallier E, Bolt 2 Lowenberg B, Downing J, Burnett A. Acute myeloid leukemia. N HM, Ketterer B, Taylor JB. Human glutathione S-transferase theta Engl J Med 1999; 341: 1051–1062. (GSTT1): cDNA cloning and the characterization of a genetic 3 Kiyoi H, Naoe T, Nakano Y, Yokota S, Minami S, Miyawaki S, polymorphism. Biochem J 1994; 300: 271–276. Asou N, Kuriyama K, Jinnai I, Shimazaki C, Akiyama H, Saito K, 17 Rebbeck T. Molecular epidemiology of the human glutathione S- Oh H, Motoji T, Omoto E, Saito H, Ohno R, Ueda R. Prognostic transferase genotypes GSTM1 and GSTT1 in cancer susceptibility. implication of FLT3 and N-RAS gene mutations in acute myeloid Cancer Epidemiol Biomarkers Prev 1997; 6: 733–743. leukemia. Blood 1999; 93: 3074–3080. 18 Hengstler JG, Arand M, Herrero ME, Oesch F. Polymorphisms of 4 Kiyoi H, Naoe T, Yokota S, Nakao M, Minami S, Kuriyama K, N-acetyltransferases, glutathione S-transferases, microsomal epox- Takeshita A, Saito K, Hasegawa S, Shimodaira S, Tamura J, Shima- ide hydrolase and sulfotransferases: influence on cancer suscepti- zaki C, Matsue K, Kobayashi H, Arima N, Suzuki R, Morishita H, Saito H, Ueda R, Ohno R. Internal tandem duplication of FLT3 bility. Rec Res Cancer Res 1998; 154: 47–85. associated with leukocytosis in acute promyelocytic leukemia. 19 Strange R, Fryer A. The glutathione S-transferases: influence of Leukemia Study Group of the Ministry of Health and Welfare polymorphism on cancer susceptibility. IARC Sci Publ 1999; 148: (Kohseisho). Leukemia 1997; 11: 1447–1452. 231–249. 5 Baudard M, Beauchamp-Nicoud A, Delmer A, Rio B, Blanc C, 20 Chen CL, Liu Q, Pui CH, Rivera GK, Sandlund JT, Ribeiro R, Evans Zittoun R, Marie JP. Has the prognosis of adult patients with acute WE, Relling MV. Higher frequency of glutathione S-transferase myeloid leukemia improved over years? A single institution experi- deletions in black children with acute lymphoblastic leukemia. ence of 784 consecutive patients over a 16-year period. Leukemia Blood 1997; 89: 1701–1707. 1999; 13: 1481–1490. 21 Chen H, Sandler DP, Taylor JA, Shore DL, Liu E, Bloomfield CD, 6 Boddy AV, Ratain MJ. Pharmacogenetics in cancer etiology and Bell DA. Increased risk for myelodysplastic syndromes in individ- chemotherapy. Clin Cancer Res 1997; 3: 1025–1030. uals with glutathione transferase theta 1 (GSTT1) gene defect. Lan- 7 Davies SM, Robison LL, Buckley JD, Radloff GA, Ross JA, Peren- cet 1996; 347: 295–297. tesis JP. Glutathione S-transferase polymorphisms in children with 22 Preudhomme C, Nisse C, Hebbar M, Vanrumbeke M, Brizard A, myeloid leukemia: a Children’s Cancer Group study. Cancer Epi- Lai JL, Fenaux P. Glutathione S transferase theta 1 gene defects in demiol Biomarkers Prev 2000; 9: 563–566. myelodysplastic syndromes and their correlation with karyotype 8 Naoe T, Takeyama K, Yokozawa T, Kiyoi H, Seto M, Uike N, Ino and exposure to potential carcinogens. Leukemia 1997; 11: T, Utsunomiya A, Maruta A, Jin-nai I, Kamada N, Kubota Y, 1580–1582. Nakamura H, Shimazaki C, Horiike S, Kodera Y, Saito H, Ueda 23 Crump C, Chen C, Appelbaum FR, Kopecky KJ, SchwartzSM, R, Wiemels J, Ohno R. Analysis of genetic polymorphism in Willman CL, Slovak ML, Weiss NS. Glutathione S-transferase theta NQO1, GST-M1, GST-T1 and CYP3A4 in 469 Japanese patients 1 gene deletion and risk of acute myeloid leukemia. Cancer Epide- with therapy-related leukemia/myelodysplastic syndrome and de miol Biomarkers Prev 2000; 9: 457–460. novo acute myeloid leukemia. Clin Cancer Res 2000; 6: 4091– 24 Davies SM, Robison LL, Buckley JD, Tjoa T, Woods WG, Radloff 4095. GA, Ross JA, Perentesis JP. Glutathione S-transferase polymor- 9 Ohno R, Kobayashi T, Tanimoto M, Hiraoka A, Imai K, Asou N, phisms and outcome of chemotherapy in childhood acute myeloid Tomonaga M, Tsubaki K, Takahashi I, Kodera Y, Yoshida M, leukemia. J Clin Oncol 2001; 19: 1279–1287. Murakami H, Naoe T, Shimayama M, Tsukada T, Takeo T, Tesh- 25 Howells RE, Redman CW, Dhar KK, Sarhanis P, Musgrove C, Jones ima H, Onozawa Y, Fujimoto K, Kuriyama K, Horiuchi A, Kimura PW, Alldersea J, Fryer AA, Hoban PR, Strange RC. Association of I, Minami S, Miura Y, Kageyama H, Tahara T, Masaoka T, Shirak- glutathione S-transferase GSTM1 and GSTT1 null genotypes with awa S, Saito H. Randomized study of individualized induction clinical outcome in epithelial ovarian cancer. Clin Cancer Res therapy with or without vincristine, and of maintenance-intensifi- 1998; 4: 2439–2445. cation therapy between 4 or 12 courses in adult acute myeloid 26 Russo J, Russo I. Biological and molecular bases of mammary car- leukemia. AML-87 Study of the Japan Adult Leukemia Study cinogenesis. Lab Invest 1987; 57: 112–137. Group. Cancer 1993; 71: 3888–3895. 27 Wiencke J, Pemble S, Ketterer B, Kelsey K. Gene deletion of 10 Kobayashi T, Miyawaki S, Tanimoto M, Kuriyama K, Murakami glutathione S-transferase theta: correlation with induced genetic H, Yoshida M, Minami S, Minato K, Tsubaki K, Ohmoto E, Oh H, damage and potential role in endogenous mutagenesis. Cancer Jinnai I, Sakamaki H, Hiraoka A, Kanamaru A, Takahashi I, Saito Epidemiol Biomarkers Prev 1995; 4: 253–259. K, Naoe T, Yamada O, Asou N, Kageyama S, Emi N, Matsuoka 28 Kelsey KT, Wiencke JK, Ward J, Bechtold W, Fajen J. Sister-chrom- A, Tomonaga M, Ohno R, for the Japan Adult Leukemia Study atid exchanges, glutathione S-transferase theta deletion and cyto- Group. Randomized trials between behenoyl cytarabine and cyta- genetic sensitivity to diepoxybutane in lymphocytes from buta-

Leukemia GSTT1 genotype and early death in AML T Naoe et al 208 diene monomer production workers. Mutat Res 1995; 335: ID, Harris AL. Reduced topoisomerase II and elevated alpha class 267–273. glutathione S-transferase expression in a multidrug resistant CHO 29 Norppa H, Hirvonen A, Jarventaus H, Uuskula M, Tasa G, Ojajarvi cell line highly cross-resistant to mitomycin C. Biochem Pharma- A, Sorsa M. Role of GSTT1 and GSTM1 genotypes in determining col 1992; 43: 685–693. individual sensitivity to sister chromatid exchange induction by 33 Chao CC, Huang YT, Ma CM, Chou WY, Lin-Chao S. Over- diepoxybutane in cultured human lymphocytes. Carcinogenesis expression of glutathione S-transferase and elevation of thiol pools 1995; 16: 1261–1264. in a multidrug-resistant human colon cancer cell line. Mol Pharm- 30 Pelin K, Hirvonen A, Norppa H. Inﬂuence of erythrocyte gluta- acol 1992; 41: 69–75. thione S-transferase T1 on sister chromatid exchanges induced by 34 Smith MT, Evans CG, Doane-Setzer P, Castro VM, Tahir MK, Man- diepoxybutane in cultured human lymphocytes. Mutagenesis nervik B. Denitrosation of 1,3-bis(2-chloroethyl)-1-nitrosourea by 1996; 11: 213–215. class mu glutathione transferases and its role in cellular resistance 31 Xu X, Wiencke JK, Niu T, Wang M, Watanabe H, Kelsey KT, Chris- in rat brain tumor cells. Cancer Res 1989; 49: 2621–2625. tiani DC. Benzene exposure, glutathione S-transferase theta homo- 35 Loe DW, Stewart RK, Massey TE, Deeley RG, Cole SP. ATP-depen- zygous deletion, and sister chromatid exchanges. Am J Ind Med dent transport of aﬂatoxin B1 and its glutathione conjugates by 1998; 33: 157–163. the product of the multidrug resistance protein (MRP) gene. Mol 32 Hoban PR, Robson CN, Davies SM, Hall AG, Cattan AR, Hickson Pharmacol 1997; 51: 1034–1041.

Leukemia