Leukemia (2003) 17, 1344–1348 & 2003 Nature Publishing Group All rights reserved 0887-6924/03 $25.00 www.nature.com/leu Dose reduction of coadministered 6-mercaptopurine decreases myelotoxicity following high-dose methotrexate in childhood leukemia U Nygaard1 and K Schmiegelow1

1The Pediatric Clinic II, The University Hospital of Copenhagen, Rigshospitalet, Copenhagen, Denmark

High-dose methotrexate (HDM) given concurrently with oral 6- duration in childhood ALL.3,10,11 Another important metabolic mercaptopurine (6 MP) may be followed by myelotoxicity, which pathway of 6MP is the methylation of 6MP and some of its may necessitate treatment interruption and thus interfere with the efficacy of the treatment of childhood acute lymphoblastic metabolites by thiopurine methyltransferase (TPMT). Methylated leukemia (ALL). Through inhibition of purine de novo synthesis products of 6MP as well as MTX polyglutamates inhibit purine and enhancement of the bioavailability, HDM may increase the de novo synthesis and, thus, may enhance the incorporation of incorporation into DNA of 6-thioguanine nucleotides, the 6TGN into DNA.12,13 cytotoxic metabolites of 6 MP. Myelotoxicity following HDM may lead to treatment inter- A total of 26 children diagnosed 3/1996–4/2001 with ALL 2 ruptions of maintenance therapy and a reduction of dose received five courses of HDM (5 g/m /24 h with leucovorin 2,14,15 rescue) at 8 weeks intervals during their first year of intensity, which could affect survival rate. We here report maintenance therapy with oral methotrexate (20 mg/m2/week) that reduction of the dose of concurrently given oral 6MP and 6MP (75 mg/m2/day). The dose of oral 6MP was reduced to a significantly reduces the degree of myelosuppression and risk of median of 51% (75% range: 39–62%, maximum 74%) of the treatment interruptions following HDM. standard dose from 2 weeks prior to until 2 weeks after HDM, because the previous HDM had led to a thrombocyte nadir 9 9 p60 Â 10 /l and/or a neutrophil nadir p0.7 Â 10 /l. Patients and methods The 6MP dose reductions raised the median thrombocyte nadir following HDM from 46 Â 109/l (range: 6–214) to 133 Â 109/l (range: 21–305; Po0.001) and the median neutrophil nadir from Patients 0.5 Â 109/l (range: 0.0–1.4) to 0.9 Â 109/l (range: 0.2–3.2; Po0.001). The effect of 6MP dose reductions was not Between March 1, 1996 and April 1, 2001, 50 patients were significantly related to risk group, gender, age, or thiopurine diagnosed with standard risk (SR) or intermediate risk (IR) non-B- methyltransferase genotype. With 6MP dose reductions, the cell ALL at the University Hospital, Rigshospitalet, Copenhagen, median duration of treatment interruption following HDM was and completed therapy according to the NOPHO ALL-92 reduced from 8 to 0 days (Po0.001). 1 The reduction of 6MP dosage during HDM can significantly program. The SR/IR risk classification has been described reduce the risk of severe myelotoxicity and prevent treatment elsewhere.2 A total of 26 patients were included in the present interruptions. analyses, because they had their dose of oral 6MP reduced to Leukemia (2003) 17, 1344–1348. doi:10.1038/sj.leu.2402990 approximately 50% from 2 weeks before until 2 weeks after at Keywords: myelotoxicity; child; acute lymphoblastic leukemia; 6- least one HDM treatment during maintenance therapy because mercaptopurine; high-dose methotrexate of a thrombocyte nadir p60 Â 109/l and/or an absolute neutrophil count (ANC) nadir p0.7 Â 109/l following their previous HDM. The cohort consisted of seven girls and 19 boys Introduction including 12 cases of SR- and 14 cases of IR-ALL. Their median age was 4.9 years. One patient heterozygous for TPMT Acute lymphoblastic leukemia (ALL) is the most common cancer (mutations G460A and A719G) has developed acute myeloid leukemia, and one patient has experienced a bone-marrow in childhood. High-dose methotrexate (HDM) is an important 16 part of the therapy given to these patients to reduce the risk of relapse. All of the remaining 24 patients are alive in first systemic and extramedullary relapse.1 HDM is commonly given remission. during consolidation therapy with or without concurrent oral 6- mercaptopurine (6MP) and as reinductions during maintenance therapy with daily oral 6MP and weekly oral methotrexate Therapy (MTX) as the backbone. Induction therapy, consolidation therapy, and maintenance We have previously demonstrated that the risk of significant therapy have been described in detail previously.2 Patients bone-marrow suppression is increased if oral 6MP is coadmi- 2 received five courses of HDM at 8-week intervals during their nistered with HDM. MTX may increase the bioavailability of first year of maintenance therapy with oral MTX (20 mg/m2/ 6MP through inhibition of xanthine oxidase, which catabolizes 2 3 week) and 6MP (75 mg/m /day). HDM was given at a dose of 6MP. In addition, several studies of cytotoxicity have indicated 2 4–8 5000 mg/m over 24 h with intrathecal MTX. Starting 36 h from that MTX and 6MP act synergistically. In hemopoietic cells, the beginning of the HDM infusion, leucovorin rescue was given 6MP is anabolized to 6-thioguanine nucleotides (6TGNs), which at a dose of 15 mg/m2 i.v. every 6 h until the serum MTX level are the most important mediators of the cytotoxic effect of 6MP 9 was below 200 nmol/l. If serum MTX was 41000 nmol/l at 42 h, through incorporation into DNA. Erythrocytic accumulation of the leucovorin doses were increased according to the serum 6TGN correlate with the degree of myelotoxicity and remission MTX concentrations.2 A total of 95 HDM treatments given during maintenance therapy were included in the study Correspondence: Dr K Schmiegelow, The Pediatric Clinic II, The University Hospital of Copenhagen, Rigshospitalet, Blegdamsvej 9, including 45 courses with standard 6MP dosage and 50 courses Copenhagen DK-2100, Denmark; Fax: +45 3545 4673 with reduced 6MP dosage. Treatments of HDM with standard Received 3 February 2003; accepted 20 March 2003 6MP dosage were given three times to six patients (HDM High-dose methotrexate–6-mercaptopurine V Nygaard and K Schmiegelow 1345 Table 1 Myelotoxicity following HDM with coadministered 6MP

HDM blocks 6MP Hb Thrombocytes WBC ANC Treatment (mg/m2/day) (mmol/l) ( Â 109/l) ( Â 109/l) ( Â 109/l) interruption (days)

Standard 6MP dosagea 72 (35–116) 5.3 (4.2–6.3) 46 (6–214) 1.1 (0.2–2.5) 0.5 (0.0–1.4) 8 (0–17) HDM no. À3(n=6) 99 5.5 54 1.8 0.8 5 HDM no. À2(n=13) 75 5.4 70 1.5 0.4 4 HDM no. À1(n=26) 72 5.3 25 1.1 0.2 9

Reduced 6MP dosageb 35 (11–64) 6.6c (5.2–8.2) 133c (21–305) 2.1c (1.0–3.9) 0.9c (0.2–3.2) 0c (0–10) HDM no. +1 (n=26) 35 6.4d 131d 2.1d 0.8d 0d HDM no. +2 (n=16) 35 6.7 135 2.0 0.9 0 HDM no. +3 (n=8) 34 6.8 216 2.5 1.2 0

HDM no. +1 is the first HDM with reduced 6MP doses. HDM no. À3toÀ1 with standard 6MP doses are given previous to HDM no. +1. aMedian of the average nadir values during HDM blocks no. À3toÀ1. bMedian of the average nadir values during HDM blocks no. +1 to +3. cPo0.001; paired comparison to HDM blocks with standard 6MP dosage (Wilcoxon’s signed-rank test). dPo0.001; paired comparison to HDM block no. À1 (Wilcoxon’s signed-rank test). HDM=high-dose methotrexate; 6MP=6-mercaptopurine; Hb=hemoglobin; WBC=white blood cell counts; ANC=absolute neutrophil count. treatment no. À3, À2 and À1), twice to seven patients (HDM weeks after the first HDM treatment with reduced 6MP treatment no. À2 and À1) and once (HDM treatment no. À1) to treatment (treatment no. +1) and compared to the average 13 patients. Treatments of HDM with reduced 6MP dosage were dose of 6MP/m2/day from 2 weeks before and until 4 weeks given three times to 8 patients (HDM treatment no. +1, +2 and after the HDM leading to treatment interruption (treatment +3), twice to eight patients (HDM treatment no. +1 and +2), and no. À1). once to 10 patients (HDM treatment no. +1) (Table 1).

6-MP dose reduction Thiopurine methyltransferase

Based on our previous observations showing that the degree of The TPMT genotype was determined using polymerase chain 2 reaction-based methods specific for the mutations G460A and myelotoxicity following HDM was related to the dose of 6MP, 17 patients who developed myelotoxicity following a HDM A719G detecting the TPMT*3A, *3B, *3C, and *3D mutations. treatment (ie a thrombocyte nadir p60 Â 109/l and/or an ANC Six patients were TPMT heterozygotes, all with both mutations nadir p0.7 Â 109/l) had their dose of 6MP reduced at all present. subsequent HDM treatments from 2 weeks before HDM and X 9 until the white blood cell count (WBC) was 1.0 Â 10 /l and Statistics the thrombocyte count X100 Â 109/l, however, at least until 2 weeks after the HDM infusion. This duration of 6MP dose For each patient, we calculated the mean hemoglobin nadir, reduction was chosen as the steady state of 6TGN on daily oral thrombocyte nadir, WBC nadir, ANC nadir, and the mean length 6MP is obtained after approximately 2 weeks and the of treatment interruption following HDM blocks with standard myelotoxicity of HDM occurs between 1 and 3 weeks after doses of 6MP (HDM treatment no. À3toÀ1) as well as HDM. Both the patients and the physicians were aware of the following HDM blocks with reduced 6MP dosage (HDM 6MP dose reductions. The 6MP dose reductions were not done treatment no. +1 to +3). Since HDM was given right at the as part of a clinical trial but on compassionate basis approved by start of maintenance therapy for IR-ALL without preceding oral the Ethical Committee of Copenhagen, Denmark. The NOPHO 6MP, this first HDM was excluded from the analyses. ALL-92 protocol included no recommendations for patients who Spearman’s rank-order correlation analysis was used for experienced significant myelosuppression or treatment interrup- correlations between parameters (r ¼ correlation coefficient). tions following HDM. s The Mann–Whitney U-test and Wilcoxon’s test were applied to The total doses of 6MP that the patients received during the compare the distributions of parameters between subgroups and period with reduced 6MP dosage were compared to standard in matched samples. Data analyses were performed using the 6MP dosage using two separate control materials: SPSS statistical software.18 Two-sided P values o0.05 were regarded as being significant. (1) A historic comparison: Patients elected from our previous study,2 who experienced the same degree of myelotoxicity as the patients included in the present study and/or treatment Results interruption following HDM because of myelotoxicity. Of 46 patients, 37 fulfilled these criteria. Their average dose of The doses of 6MP given during blocks of HDM with standard 6MP/m2/day was calculated by dividing the total cumulated 6MP dosage (HDM no. À3, À2, and À1) are shown in Table 1. prescribed dose/m2 by the number of days from start of The doses were not significantly related to risk group (median: 2 2 maintenance therapy until 4 weeks after the last HDM SR ¼ 79 mg/m /day; IR ¼ 70 mg/m /day; P ¼ 0.07), age (rs ¼ 0.008; treatment. P ¼ 0.97), gender (median: female ¼ 72 mg/m2/day; male ¼ (2) A matched comparison: The average dose of 6MP/m2/day 71 mg/m2/day; P ¼ 0.61), or to TPMT genotype (median: wild was calculated as above from 2 weeks before and until 4 type ¼ 72 mg/m2/day; heterozygote ¼ 66 mg/m2/day; P ¼ 0.46).

Leukemia High-dose methotrexate–6-mercaptopurine V Nygaard and K Schmiegelow 1346 a 300 was significantly reduced (Table 1). This was the case both in a paired comparison of the mean degree of myelotoxicity 250 following HDM blocks with standard 6MP dosage vs reduced

/L) 200 6MP dosage (Table 1, Figure 1), and if only HDM no. À1 was 9 150 compared to HDM no. +1 (Table 1). The mean hemoglobin nadir and thrombocyte nadir were raised for all but one patient 100 100 each (Figure 1a), and the mean WBC nadir was increased for all patients except three. Differential counts with ANCs were 75 available for 25 patients of whom 21 patients experienced less 50 neutropenia while one patient had an unchanged value and three patients experienced more neutropenia following the 25 HDM treatments given with 6MP dose reduction (Figure 1b). Of Thrombocyte nadir (x10 the latter three patients, two fell 0.2  109/l in ANC nadir while 0 one patient fell 0.4  109/l compared to the mean ANC nadir 0 25 50 75 100 125 following HDM blocks with standard 6MP dosage. Two of these 6MP doses during HDM (mg/m2 /day) three patients were TPMT heterozygotes (Figure 1b). The effect of dose reductions was not significantly related to risk group b 3 (thrombocytes: P ¼ 0.56; ANC: P ¼ 0.81), gender (thrombocytes: P ¼ 0.53; ANC: P ¼ 0.69), age (thrombocytes: P ¼ 0.24; ANC: 2 P ¼ 0.97), or TPMT genotype (thrombocytes: P ¼ 0.93; ANC: P ¼ 0.64). /L) 1 9 There were no significant differences between the 23 h serum 1.0 MTX concentrations during HDM blocks with standard 6MP 0.8 dosage (median: 64 590 nM (range: 11 320–11 4070)) compared to the 23 h serum MTX concentrations during HDM blocks with 0.6 reduced doses of 6MP (median: 67 011 nM (range: 22 100– 0.4 53 4725); P ¼ 0.99). The 42 h serum MTX concentrations were ANC nadir (x10 higher during HDM blocks with standard 6MP dosage (median 0.2 42 h MTX with standard 6MP dosage: 757 nM (range: 290– 0.0 12 470); median with reduced 6MP dosage: 394 nM (range: 130– 0 25 50 75 100 125 743); P ¼ 0.001). Similarly, the total dose of leucovorin rescue 6MP doses during HDM (mg/m2/day) was higher in HDM blocks with standard 6MP dosage (median: 120 mg/m2 (range: 60–4559)) compared to HDM blocks with Figure 1 (a) Average thrombocyte nadirs following HDM in reduced 6MP dosage (median: 75 mg/m2 (range: 53–165); relation to coadministered 6MP doses (median dose reduction 51% Po0.001). The 42 h serum MTX concentrations were not related (75% range: 39–62%, maximum 74%)). Each line represents one to bone marrow suppression following HDM (thrombocytes: patient with the start of the line reflecting the reduced 6MP doses and the end of the line the standard 6MP doses. The dashed lines indicate rS ¼À0.11, P ¼ 0.28; ANC: rS ¼À0.14, P ¼ 0.16). patients with TPMT G460A and A719G mutations. (b) Average The duration of treatment interruption following HDM was neutrophil count nadirs following HDM in relation to coadministered significantly shorter when 6MP was reduced prior to HDM. 6MP doses (median dose reduction 51% (75% range: 39–62%, Thus, with 6MP dose reductions, the median treatment maximum 74%)). Each line represents one patient with the start of interruption was reduced from 8 to 0 days (Po0.001) the line reflecting the reduced 6MP doses and the end of the line the (Table 1), and only nine of 26 patients had treatment standard 6MP doses. The dashed lines indicate patients with TPMT G460A and A719G mutations. interruptions following 6MP dose reduction compared to 24 of 26 patients when treated with standard doses of 6MP (Po0.001). The degree of myelotoxicity and the length of treatment According to the historical as well as the matched compar- interruption following blocks of HDM with standard 6MP isons, the overall 6MP doses were significantly reduced during dosage are given in Table 1. The degree of neutropenia HDM with 6MP dose reductions. Thus, the median dose of 6MP following blocks of HDM with standard 6MP dosage was not given from the start of maintenance therapy until 4 weeks after significantly related to TPMT genotype (median ANC: wild the last HDM treatment was 51 mg/m2/day (range: 22–78) type ¼ 0.3  109/l; heterozygote ¼ 0.5  109/l; P ¼ 0.33). There compared to 61 mg/m2/day (range: 3–84) in the historic were no significant differences between HDM no. À3, À2, and comparison (P ¼ 0.04). The median dose of 6MP given from 2 À1 regarding the degree of myelotoxicity or length of treatment weeks before and until 4 weeks after HDM no. À1 was 55 mg/ interruption. m2/day (range: 24–88) compared to 40 mg/m2/day (range: 16– Owing to myelotoxicity and treatment interruption, the doses 80) during HDM treatment no. +1 (P ¼ 0.01). of 6MP during subsequent HDM treatments were reduced with a median of 51% (75% range: 39–62%, maximum 74%) of the doses in previous HDM blocks. This degree of 6MP dose Discussion reduction was related to risk group (median: SR ¼ 55%; IR ¼ 45%, P ¼ 0.02), but not significantly to age (rs ¼À0.03; Although HDM has been used since the 1960s, the optimal way P ¼ 0.89), gender (median: female ¼ 50%; male ¼ 52%; to administer HDM is still debated. HDM is given in P ¼ 0.53), or to TPMT genotype (median: wild type ¼ 52%; combination with oral 6MP in order to increase the antileu- heterozygote ¼ 45%; P ¼ 0.66). kemic efficacy of the treatment; however, this may in part be When the dose of 6MP was reduced from 2 weeks before and counteracted by an increased risk of toxicity with subsequent until at least 2 weeks after HDM, the degree of myelotoxicity treatment interruption. We have previously shown that HDM

Leukemia High-dose methotrexate–6-mercaptopurine V Nygaard and K Schmiegelow 1347 given during maintenance therapy with coadministered oral dose intensity during MT. This approach is now an option in the daily 6MP causes significant pancytopenia.2 Owing to myelo- NOPHO ALL-2000 protocol for patients who have experienced suppression, 42% of the treatments with HDM (5 g/m2/24 h) and an ANC nadir o0.5 Â 109/l and/or a thrombocyte nadir oral 6MP were found to be followed by treatment interruptions o60 Â 109/l following a HDM treatment with standard 6MP of 6MP and low-dose MTX with a median duration of 10 dosage during maintenance therapy. days(2). In addition, fatal infections because of immune It is unlikely that the reduced myelotoxicity subsequent to suppression have been reported to occur in 1–6% of patients 6MP dose reduction found in the present analyses is a random in remission.1,19 To decrease or even prevent myelotoxicity and effect or explained by an increasing tolerance of the bone treatment interruptions following HDM in maintenance therapy marrow to HDM during maintenance therapy. Thus, the degree several approaches could be chosen: of myelotoxicity following HDM at 8-week intervals does not decrease during maintenance therapy.2 In fact, we found an (1) The total dose of MTX could be reduced. However, this increase in the degree of myelotoxicity during HDM treatment approach carries several drawbacks. Higher doses of MTX À3toÀ1 (Table 1). Furthermore, we found no significant increase the accumulation in blasts of MTX polyglutamates and difference in post-HDM myelotoxicity in SR and IR patients in specifically long-chain MTX polyglutamates. Hence, decreased 20,21 spite of a more intensive treatment in IR patients during their MTX doses could reduce the antileukemic effect of MTX. In 1,2 preceding consolidation therapy. addition, MTX dose reduction would limit penetration of MTX We found a delayed excretion of MTX following HDM blocks into the CNS,22 which could increase the risk of CNS relapse. with standard 6MP dosage compared to reduced 6MP dosage. Supporting the clinical significance of high systemic MTX The majority of MTX is bound to albumin in the blood and, exposure, Evans et al23 found that pharmacokinetically guided when HDM is administered, approximately one-third of MTX is MTX dosage to achieve higher MTX concentration in patients metabolized in the liver. It is well known that MTX/6MP with rapid clearance of the drug lead to a better outcome. 28 maintenance therapy causes hepatotoxicity and this may in Further, it is uncertain whether a reduction in MTX dose would 29 part be because of methylated metabolites of 6MP. It is lead to reduced myelotoxicity. Thus, HDM followed by possible that a reduced degree of hepatotoxicity second to the leucovorin rescue within 36 h carries only little myelotoxicity reduced 6MP doses could have caused an improvement in the if given without concurrent oral 6MP.2,22 Thus, with the hepatic clearance of MTX and explain the moderate decrease in leucovorin regimen used in the NOPHO ALL-92 protocol, the 42 h MTX concentrations observed in the present study. median neutrophil and thrombocyte nadirs were only 1.9 Â 109/ However, the MTX concentrations were not related to the l and 240 Â 109/l, respectively, when HDM was given without degree of bone-marrow toxicity. concurrent 6MP.2 Finally, the duration of MTX exposure above a The degree of myelotoxicity following HDM treatments with certain threshold is probably more important than the peak MTX standard doses of 6MP in this cohort of patients were very concentration.24 similar to what we have reported previously with respect (2) The leucovorin rescue could be given earlier. The toxicity to the median nadirs of hemoglobin, thrombocytes, white of HDM is primarily related to the duration of exposure to the 2 blood cells, and neutrophil counts. Even though reduced drug prior to leucovorin administration rather than to the peak 6MP dosage during HDM was able to decrease myelo- MTX concentration.22,24,25 MTX and its polyglutamated meta- toxicity and treatment interruption, the situation is still less bolites inhibit dihydrofolate reductase, thereby lowering cellular than optimal since the efficacy of dose reductions could not pools of reduced folates.12 The lack of reduced folate cofactors be predicted for the individual patient and the overall 6MP is responsible for the inhibition of the synthesis of purines and dose intensity was reduced. Thus, further pharmacokinetic pyrimidines. Thus, leucovorin rescue administered earlier after and biological studies are needed to improve the prediction the start of MTX infusion could be another approach to reduce of the patients who are at risk to develop severe myelo- myelotoxicity and the duration of treatment interruption. toxicity following HDM and to improve individual preventive However, even though this has not been tested in randomized measures. studies, this approach could to lead to a decreased antileukemic The frequency of TPMT heterozygotes (including TPMT efficacy.26 G460A and A719G mutations) in this study was 23% (95% (3) The approach chosen in the present study was to reduce 30,31 range: 9–44%) compared to the expected 8–9%. Although, 6MP doses during HDM. This approach was based on our this indicates that TPMT heterozygosity may be a risk factor for previous observations that significant myelotoxicity following myelotoxicity following HDM with concurrent oral 6MP, larger HDM was seen when the treatment was given with coadminis- studies are needed to explore if TPMT heterozygotes as a rule tered 6MP.2 It was also based on the assumption that MTX, should have their dose of 6MP reduced to HDM. through inhibition of the de novo purine synthesis, increases the On the basis on the present findings, we conclude that levels of phosphoribosyl pyrophosphate, and thereby increases reducing the dose of 6MP from 2 weeks before to 2 weeks after both the build up of 6TGN and their incorporation into DNA, HDM is one way to decrease bone-marrow toxicity and although the latter has not yet been proven in clinical studies.27 treatment interruptions in patients who have previously devel- However, it is not known whether uninterrupted maintenance oped significant myelotoxicity after HDM/6MP. However, therapy with reduced doses of 6MP is superior to full-dose further studies are needed to be able to predict the effect of therapy with intermittent treatment interruptions. A few studies 6MP dose reduction in the single patient. have indicated that reductions of the overall 6MP dose intensity because of treatment interruptions may affect the cure rate.14,15 Since the half-life of the cytosol 6TGN levels is only 3–5 days at Acknowledgements least in erythrocytes (unpublished observations), it is possible that a period of dose reductions of 1 week before HDM and until The commitment and skillful technical assistance of Jannie the nadirs have been passed could be sufficient to prevent Gregers, Kristine Nielsen, and Michael Timm are greatly clinically significant myelotoxicity. If this is the case, such a appreciated. This study has received financial support from the 6MP dose reduction approach may not reduce the overall 6MP Lundbeck Foundation (Grant No. 38/99); the Danish Childrens

Leukemia High-dose methotrexate–6-mercaptopurine V Nygaard and K Schmiegelow 1348 Cancer Foundation; the Childrens Cancer Foundation, Sweden 15 Relling MV, Hancock ML, Boyett JM, Pui CH, Evans WE. (Grant No. 1999/080). Prognostic importance of 6MP dose intensity in acute lympho- blastic leukemia. Blood 1999; 93: 2817–2823. 16 Thomsen JB, Schroder H, Kristinsson J, Madsen B, Szumlanski C, References Weinshilboum R et al. Possible carcinogenic effect of 6MP on bone marrow stem cells. Cancer 1999; 86: 1080–1086. 1 Gustafsson G, Schmiegelow K, Forestier E, Clausen N, Glomstein 17 Otterness D, Szumlanski C, Lennard L, Klemetsdal B, Aarbakke J, A, Jonmundsson G et al. Improving outcome through two decades Park-Hah JO et al. Human thiopurine methyltransferase pharma- in childhood ALL in the Nordic countries: the impact of high-dose cogenetics: gene sequence polymorphisms. Clin Pharmacol Ther methotrexate in the reduction of CNS irradiation. Nordic Society of 1997; 62: 60–73. Pediatric Haematology and Oncology (NOPHO). Leukemia 2000; 18 SPSS statistical software for Windows release 10.0.5.SPSS Inc. 14: 2267–2275. SPSS Inc. 1999. 2 Schmiegelow K, Bretton-Meyer U. 6-Mercaptopurine dosage and 19 Hargrave DR, Hann II, Richards SM, Hill FG, Lilleyman JS, Kinsey pharmacokinetics influence the degree of bone marrow toxicity S et al. Progressive reduction in treatment-related deaths in following high-dose methotrexate in children with acute lympho- Medical Research Council childhood lymphoblastic leukaemia blastic leukemia. Leukemia 2001; 15: 74–79. trials from 1980 to 1997 (UKALL VIII, X and XI). Br J Haematol 3 Balis FM, Holcenberg JS, Zimm S, Tubergen D, Collins JM, Murphy 2001; 112: 293–299. RF et al. The effect of methotrexate on the bioavailability of oral 20 Synold TW, Relling MV, Boyett JM, Rivera GK, Sandlund JT, 6MP. Clin Pharmacol Ther 1987; 41: 384–387. Mahmoud H et al. Blast cell methotrexate-polyglutamate accu- 4Bo¨kkerink JP, Bakker MA, Hulscher TW, De Abreu RA, Schretlen mulation in vivo differs by lineage ploidy and methotrexate dose in ED. Purine de novo synthesis as the basis of synergism of acute lymphoblastic leukemia. J Clin Invest 1994; 94: 1996–2001. methotrexate and 6-mercaptopurine in human malignant lympho- 21 Masson E, Relling MV, Synold TW, Liu Q, Schuetz JD, Sandlund JT blasts of different lineages. Biochem Pharmacol 1988; 37: 2321– et al. Accumulation of methotrexate polyglutamates in lympho- 2327. blasts is a determinant of antileukemic effects in vivo. A rationale 5 Schmiegelow K, Schroder H, Gustafsson G, Kristinsson J, for high-dose methotrexate. J Clin Invest 1996; 97: 73–80. Glomstein A, Salmi T et al. Risk of relapse in childhood acute 22 Schornagel JH, McVie JG. The clinical pharmacology of metho- lymphoblastic leukemia is related to RBC methotrexate and trexate. Cancer Treat Rev 1983; 10: 53–75. mercaptopurine metabolites during maintenance chemotherapy. 23 Evans WE, Relling MV, Rodman JH, Crom WR, Boyett JM, Pui CH. Nordic Society for Pediatric Hematology and Oncology. J Clin Conventional compared with individualized chemotherapy for Oncol 1995; 13: 345–351. childhood acute lymphoblastic leukemia. N Engl J Med 1998; 338: 6 Innocenti F, Danesi R, Di Paolo A, Loru B, Favre C, Nardi M et al. 499–505. Clinical and experimental pharmacokinetic interaction between 24 Wolfrom C, Hartmann R, Fengler R, Bruhmuller S, Ingwersen A, 6MP and methotrexate. Cancer Chemother Pharmacol 1996; 37: Henze G. Randomized comparison of 36-hour intermediate-dose 409–414. versus 4-hour high-dose methotrexate infusions for remission 7 Schmiegelow K, Ifversen M. Myelotoxicity, pharmacokinetics, and induction in relapsed childhood acute lymphoblastic leukemia. J relapse rate with methotrexate/6-mercaptopurine maintenance Clin Oncol 1993; 11: 827–833. therapy of childhood acute lymphoblastic leukemia. Pediatr 25 Rask C, Albertioni F, Bentzen SM, Schroeder H, Peterson C. Hematol Oncol 1996; 13: 433–441. Clinical and pharmacokinetic risk factors for high-dose metho- 8 Giverhaug T, Loennechen T, Aarbakke J. The interaction of 6- trexate-induced toxicity in children with acute lymphoblastic mercaptopurine (6-MP) and methotrexate (MTX). Gen Pharmacol leukemia – a logistic regression analysis. Acta Oncol 1998; 37: 1999; 33: 341–346. 277–284. 9 Waters TR, Swann PF. Cytotoxic mechanism of 6-thioguanine: 26 Borsi JD, Sagen E, Romslo I, Moe PJ. Rescue after intermediate and hMutSalpha, the human mismatch binding heterodimer, binds to high-dose methotrexate: background, rationale, and current DNA containing S6-methylthioguanine. Biochemistry 1997; practice. Pediatr Hematol Oncol 1990; 7: 347–363. 36:2501–2506. 27 Dervieux T, Hancock M, Evans W, Pui CH, Relling MV. Effect of 10 Lilleyman JS, Lennard L. Mercaptopurine metabolism and risk of methotrexate polyglutamates on thioguanine nucleotide concen- relapse in childhood lymphoblastic leukaemia. Lancet 1994; 343: trations during continuation therapy of acute lymphoblastic 1188–1190. leukemia with mercaptopurine. Leukemia 2002; 16: 209–212. 11 Schmiegelow K, Bjo¨rk O, Glomstein A, Gustafsson G, Keiding N, 28 Schmiegelow K, Pulczynska M. Prognostic significance of hepa- Kristinsson J et al. Intensification of 6-mercaptopurine/methotrex- totoxicity during maintenance chemotherapy for childhood acute ate maintenance chemotherapy may increase the risk of relapse for lymphoblastic leukaemia. Br J Haematol 1990; 61: 767–772. some children with acute lymphoblastic leukaemia. J Clin Oncol 29 Dubinsky MC, Lamothe S, Yang HY, Targan SR, Sinnett D, Theoret 2003; 21: 1332–1339. Y et al. Pharmacogenomics and metabolite measurement for 6MP 12 Chabner BA, Allegra CJ, Curt GA, Clendeninn NJ, Baram J, therapy in inflammatory bowel disease. Gastroenterol Enobiot Koizumi S et al. Polyglutamation of methotrexate. Is methotrexate 2000; 4: 705–713. a prodrug? J Clin Invest 1985; 76: 907–912. 30 McLeod HL, Krynetsky EY, Relling MV, Evans WE. Genetic 13 Weinshilboum RM, Otterness DM, Szumlanski CL. Methylation polymorphism of thiopurine methyltransferase and its clinical pharmacogenetics: catechol O-methyltransferase, thiopurine relevance for childhood acute lymphoblastic leukemia. Leukemia methyltransferase, and histamine N-methyltransferase. Annu Rev 2000; 14: 567–572. Pharmacol Toxicol 1999; 39: 19–52. 31 Tai HL, Krynetski EY, Yates CR, Loennechen T, Fessing MY, 14 Schmiegelow K. Prognostic significance of methotrexate and 6- Krynetskaia NF et al. Thiopurine S-methyltransferase deficiency: mercaptopurine dosage during maintenance chemotherapy for two nucleotide transitions define the most prevalent mutant allele childhood acute lymphoblastic leukemia. Pediatr Hematol Oncol associated with loss of catalytic activity in Caucasians. Am J Hum 1991; 8: 301–312. Genet 1996; 58: 694–702.

Leukemia