Infants with acute lymphoblastic leukemia: no evidence for high methotrexate resistance. Leukemia (2002) 16, 949â951. DOI: 10.1038/sj/leu/2402491. TO THE ...
Leukemia (2002) 16, 949–969 2002 Nature Publishing Group All rights reserved 0887-6924/02 $25.00 www.nature.com/leu
CORRESPONDENCE Infants with acute lymphoblastic leukemia: no evidence for high methotrexate resistance Leukemia (2002) 16, 949–951. DOI: 10.1038/sj/leu/2402491
Table 1 Relationship between age and MTX sensitivity as determined by the in situ thymidylate synthase inhibition assay (TSIA)
TO THE EDITOR The antifolate methotrexate (MTX) has contributed significantly to the great improvement in overall survival and central nervous system prophylaxis in patients with acute lymphoblastic leukemia (ALL) in the past 50 years. After transport into the cell, MTX is polyglutamylated with multiple glutamate residues to MTX-polyglutamates (MTXPGs), which have superior intracellular retention. Cellular resistance to MTX might contribute to treatment failure in childhood ALL. C/preB-lineage ALL (⭓1 year) has a favorable prognosis and is in vitro more sensitive to MTX in the TSIA (short exposure) than T-lineage ALL.1 C/preB ALL also have more efficient accumulation of (long chain) MTX-PGs compared to T-ALL and acute myeloid leukemia.2–4 ALL diagnosed in infants less than 1 year of age is closely associated with a number of biological features, especially MLL gene (at chromosome 11q23) rearrangements and the proB (CD10-negative precursor B) immunophenotype, and still have a poor outcome.5 So far, little is known of the pharmacodynamics of MTX in infants with ALL, and the relationship thereof with the other biological characteristics frequent in infant ALL. The question then arises whether the poor prognosis of infants with ALL is associated with cellular resistance to MTX. Lymphoblasts isolated from bone marrow or peripheral blood of 47 infants ⬍1 year with newly diagnosed, untreated ALL from the Dutch Childhood Leukemia Study Group (DCLSG), the German COALL study group, the Berlin–Frankfurt–Mu¨nster (BFM) Study Group and the Pediatric Oncology Group cell bank (POG) were used for this portion of the study. The distribution of important clinical parameters within the infants showed a high association with the proB immunophenotype and translocations involving the MLL gene (11q23, determined by karyotype, RT-PCR, and/or Southern blotting), and significantly higher white blood cell counts at presentation, features typical of infants with ALL. Since MTX cytotoxicity on primary ALL cells cannot be measured with the MTT assay, we used the thymidylate synthase (TS) inhibition assay (TSIA), which correlates strongly with IC50 values for MTX obtained for cell lines in the MTT assay.1 These data were expressed as the concentration of MTX necessary to inhibit 50% of the TS activity (TSI50), compared to the controls incubated without MTX. A large range of TSI50 values for MTX was observed for both the short (3 h, followed by 18 h drug-free) and continuous (21 h) exposure conditions. Infant ALL cells did not differ in MTX sensitivity from those of a reference group of 109 common(c)/preB ALL patients older than 1 year, with overlapping ranges and similar median TSI50 values (Table 1, Figure 1a and b). The total in vitro accumulation of MTX and the pharmacologically important long-chain polyglutamates (MTX-PG4–6, analyzed by HPLC)4 did not differ significantly between the infants and the older c/preB ALL group (Figure 1c and d). Hyperdiploid B-lineage ALL cells accumulate higher levels of MTXpolyglutamates.6 Because hyperdiploidy is found in one fifth of the older children but in no case of infant ALL, a comparison was also made of only those patients in both groups known to have a nonhyperdiploid DNA index, as determined by flow cytometry (hyperdiploidy was defined as a DNA index 1.16–1.35). Infants were as sensitive to MTX in the TSIA compared to non-hyperdiploid c/preB ALL patients ⭓1 year (Table 1). In order to examine the influence of proB immunophenotype and MLL gene rearrangements on infant drug sensitivity, the impact of age was investigated within patients with proB immunophenotype and
Correspondence: NL Ramakers-van Woerden, VU University Medical Center, Department of Pediatric Hematology/Oncology, PO Box 7057, 1007 MB Amsterdam, The Netherlands; Fax: 31 20 444 2422 Received 11 May 2001; accepted 22 January 2002
Age:
All patients short exposure continuous exposure Non-hyperdiploid cases only short exposure continuous exposure ProB immunophenotype only short exposure continuous exposure MLL rearranged only short exposure continuous exposure
⬍1 year TSI50 (n)
⭓1 year TSI50 (n)
P-value
0.366 (38) 0.386 (59) 0.063 (29) 0.064 (62)
0.40 0.74
0.512 (26) 0.490 (31) 0.097 (19) 0.100 (30)
0.31 0.94
0.327 (20) 0.578 (3) 0.055 (19) 0.105 (2)
0.35 0.63
0.394 (29) 0.578 (5) 0.058 (21) 0.094 (4)
0.39 0.60
Median TSI50 values (m) represent the concentration of MTX necessary to inhibit 50% of the control TS activity. For the short exposure, cells were exposed to MTX for 3 h followed by an 18-h drug-free period. The continuous exposure was for 21 h. n = number of patient samples. P-values determined by the Mann– Whitney U test.
MLL rearrangements, respectively. No differences were observed between infants and patients ⭓1 year within these categories (Table 1). However, the number of samples with proB-ALL and/or MLL rearrangements ⭓1 year of age is very small (n = 3 and 5, respectively); these results must therefore be regarded as descriptive. So far, no differences were seen between MLL rearranged (n = 34) and germline (n = 6) cases with the TSIA. It has been suggested that infants with the translocation t(4;11)(q21;q23) (MLL-AF4) might have a particularly poor prognosis.7 We did not observe any difference in MTX resistance between MLL-AF4 positive (n = 16) and other MLL translocated infants (n = 9) with the TS inhibition assay (short and continuous exposure conditions). This small number of patients precludes definite conclusions. Further analysis of MTX-PG accumulation vs age within the proB immunophenotype and MLL gene rearranged groups was not possible. For the in vivo MTX-PG part of this study, samples were analyzed from children treated with MTX as a single drug in an unfront window setting at St Jude Children’s Research Hospital.3 Five infants ⬍1 year were compared with 159 children ⭓1 year of age, diagnosed with non-hyperdiploid B-lineage ALL. Informed consent was obtained from each patient or his/her parents or guardian, as appropriate. At initial diagnosis, patients were randomized to receive either fractionated low-dose (LDMTX; 180 mg/m2, consisting of 30 mg/m2 per os given every 6 h) or high-dose methotrexate monotherapy (HDMTX; 1000 mg/m2 in a 24 h intravenous (i.v.) infusion). Bone marrow samples were obtained at 44 h after the start of LDMTX or HDMTX. The leukemic blast cells were isolated and MTX (ie MTX-Glu1) and its polyglutamates (MTX-Glu2–7) were analyzed by HPLC and radioligand binding-assay.3 The infants had similar accumulation of total and long-chain MTX-PG in their leukemia cells, compared to older children with non-hyperdiploid B-lineage ALL. Moreover, HDMTX achieved higher MTX-PG in leukemia cells than lower-dose MTX among infants (median = 2560 with HDMTX vs 462 pmol/109 cells with LDMTX) and among older children (1501 vs 604 pmol/109 cells). In vivo pharmacokinetic data of MTX in infants are limited. MTX is eliminated by the kidneys, and hence might be expected to have
Correspondence
950
Figure 1 Relationship between age and in vitro MTX sensitivity in childhood B-lineage ALL. Determined by the in situ TS inhibition assay with continuous (a) and short (b) exposure to MTX, and by in vitro total MTX (c) and long-chain (4–6 glutamate residues; d) MTX polyglutamylate accumulation, for infants (⬍1 year) vs older (⭓1 year) childhood patients. The horizontal lines show the median values, the circles represent the individual patient samples. P-values determined by the Mann–Whitney U test.
slower clearance in young infants. Data on the elimination of MTX in infants are somewhat conflicting, and not available in infants under the age of 2–3 months; however it seems that high doses of MTX are adequately tolerated.8 Our data indicating that 1 g/m2 i.v. achieved substantially higher MTX-PGs in infants than LDMTX suggest that higher doses of MTX are warranted in infants. Whether they should receive, or can tolerate, doses ⭓1 g/m2 remains to be shown. Moreover, intensive systemic and intrathecal chemotherapy regimens including HDMTX appear to be as effective as cranial irradiation in the treatment and prevention of central nervous system leukemia in infants with ALL.7 The data from the present study suggest that infant (⬍1 year) ALL cells are not resistant to MTX, and have a sensitivity comparable to c/preB ALL ⭓1 year of age, measured by the indirect in situ TS inhibition assay and by in vitro and in vivo MTX and MTX polyglutamate accumulation. This might also partly explain why recent protocols, that have included the addition of high-dose MTX, appear to improve the outcome of infants with ALL.5 Hence, ALL in infants appears to be relatively sensitive to MTX; resistance to MTX most likely does not contribute to the poor prognosis of infant ALL. Taken together with reports that MTX-treatment is relatively well tolerated by infants, it is important to include MTX in infant ALL treatment schemes at adequately tolerated doses for both systemic and intrathecal treatment.
Acknowledgements This work was supported in part by the Dutch Cancer Society grant VU95–921, by St Jude Cancer Center Grant CA21765, by NIH grants R37CA36401, R01CA78224, R01CA51001, by the American Lebanese Leukemia
Syrian Associated Charities, and by USPHS grants CA32053 and CA29139. We wish to thank the pediatric oncological centers participating in the DCLSG, COALL, BFM and POG study groups. Karyotyping of the Dutch patients was performed by members of the Netherlands Working Party on Cancer Genetics and Cytogenetics at regional cytogenetics centers and for German childhood cases karyotyping was performed in the Oncogenetic Laboratory, Children’s Hospital, University of Giessen, Germany. We thank Prof W-D Ludwig and Prof OA Haas (German and Austrian BFM, respectively) for their efforts in supplying cryopreserved samples of infants with ALL for this study. We also wish to thank Dr J Shuster, POG Statistical Office, University of Florida, Gainsville FL, for providing the clinical information for the POG samples. 1 Department of Pediatric NL Ramakers-van Woerden1 Hematology/Oncology VU University R Pieters2 Medical Center, Amsterdam, The MG Rots1,3 Netherlands; 2Sophia Children’s Hospital, CH van Zantwijk1 Division of Oncology/Hematology, P Noordhuis3 University Hospital Rotterdam, The HB Beverloo4 Netherlands; 3Department of Medical GJ Peters3 Oncology, VU University Medical Center, ER van Wering5 Amsterdam, The Netherlands; 4Department BM Camitta6 of Cell Biology and Genetics, Erasmus C-H Pui7 University, Rotterdam, The Netherlands; MV Relling7 7 5 Dutch Childhood Leukemia Study Group, WE Evans The Hague, The Netherlands; 6Pediatric AJP Veerman1 Oncology Group, The Midwest Children’s Cancer Center and the Medical College of Wisconsin, Milwaukee, WI, USA; and 7 Pharmaceutical Department, St Jude Children’s Research Hospital, Memphis, TN, USA
Correspondence
References 1 Rots MG, Pieters R, Kaspers GJL, Van Zantwijk CH, Noordhuis P, Mauritz R, Veerman AJP, Jansen G, Peters GJ. Differential methotrexate resistance in childhood T-versus common/preB-acute lymphoblastic leukemia can be measured by an in situ thymidylate synthase inhibition assay, but not by the MTT assay. Blood 1999; 93: 1067–1074. 2 Goker E, Lin JT, Trippett T, Elisseyeff Y, Tong WP, Niedzwiecki D, Tan C, Steinherz P, Schweitzer BI, Bertino JR. Decreased polyglutamylation of methotrexate in acute lymphoblastic leukemia blasts in adults compared to children with this disease. Leukemia 1993; 7: 1000–1004. 3 Synold TW, Relling MV, Boyett JM, Rivera GK, Sandlund JT, Mahmoud H, Crist WM, Pui CH, Evans WE. Blast cell methotrexatepolyglutamate accumulation in vivo differs by lineage, ploidy, and methotrexate dose in acute lymphoblastic leukemia. J Clin Invest 1994; 94: 1996–2001. 4 Rots MG, Pieters R, Peters GJ, Noordhuis P, Van Zantwijk CH, Kaspers GJL, Ha¨hlen K, Creutzig U, Veerman AJP, Jansen G. Role of folylpolyglutamate synthetase and folylpolyglutamate hydrolase
5 6
7
8
in methotrexate accumulation and polyglutamylation in childhood leukemia. Blood 1999; 93: 1677–1683. Biondi A, Cimino G, Pieters R, Pui CH. Biological and therapeutic aspects of infant leukemia. Blood 2000; 96: 24–33. Whitehead VM, Vuchich MJ, Lauer SJ, Mahoney D, Carroll AJ, Shuster JJ, Esseltine DW, Payment C, Look AT, Akabutu J. Accumulation of high levels of methotrexate polyglutamates in lymphoblasts from children with hyperdiploid (greater than 50 chromosomes) B-lineage acute lymphoblastic leukemia: a Pediatric Oncology Group study. Blood 1992; 80: 1316–1323. Reaman GH, Sposto R, Sensel MG, Lange BJ, Feusner JH, Heerema NA, Leonard M, Holmes EJ, Sather HN, Pendergrass, TW, Johnstone HS, O’Brien RT, Steinherz PG, Zeltzer PM, Gaynon PS, Trigg ME, Uckun FM. Treatment outcome and prognostic factors for infants with acute lymphoblastic leukemia treated on two consecutive trials of the Children’s Cancer Group. J Clin Oncol 1999; 17: 445–455. Donelli MG, Zucchetti M, Robatto A, Perlangeli V, D’Incalci M, Masera G, Rossi MR. Pharmacokinetics of HD-MTX in infants, children, and adolescents with non-B acute lymphoblastic leukemia. Med Ped Oncol 1995; 24: 154–159.
951
ABL gene fuses BCR during t(20;22) results in Philadelphia-negative, but BCR/ABLpositive chronic myeloid leukemia Leukemia (2002) 16, 951–952. DOI: 10.1038/sj/leu/2402441 TO THE EDITOR Chronic myelogenous leukemia (CML) is usually characterized by the presence of the Philadelphia (Ph) translocation, t(9;22)(q34;q11). BCR-ABL fusion in Ph translocation is well known to be critical in the pathogenesis. The interest in treating this disease is now focused on a novel oncoprotein p210BCR-ABL as a target molecule. On the other hand, a minority of patients with CML have cytogenetically normal leukemic cells. A subgroup in Ph-negative patients showed a rearrangement of the major BCR.1 A previous report, analyzing patients with Ph-negative CML showed M-BCR rearrangement in 11 out of 23 patients.2 Insertion of a small segment of chromosome 9 (9q34) into chromosome 22q11 was suggested to form MBCR in Ph-negative CML.3,4 A unique complex translocation: t(9;12;15)(q34;q12;q21) in a CML patient has been reported to mask the Ph chromosome.5 Fluorescence in situ hybridization (FISH) has become useful to confirm cytogenetic diagnosis in Ph-positive and -negative CML.6 Seong et al,7 using hypermetaphase FISH, reported that two out of six Ph-negative CML patients had ABL insertions into the BCR gene on chromosome 22. In Ph-negative CML, retranslocation between chromosomes 9q+ and 22q− was postulated to result in normal-looking chromosomes 9 and 22.1 We here report a patient with t(20;22)(q13;q11) accompanying BCR-ABL fusion on chromosome 22 which resulted in a normal-looking cytogenetic study by the Giemsa-banding method. A 61-year-old Japanese woman was found to have leukocytosis and thrombocytosis in September 2000. Physical examination showed no hepatosplenomegaly or any other abnormal findings. Laboratory data revealed a white blood cell (WBC) count of 19.2 × 109/l including 0.5% myelocytes, 1% metamyelocytes, 8.5% eosinophils and 7.5% basophils, a red blood cell (RBC) count of 4.23 × 1012/l, hemoglobin (Hb) of 122 g/l, hematocrit (Ht) of 0.384 and platelet (Plt) count of 1067 × 109/l. Biochemical examination revealed lactate dehydrase of 554 U/l in which type 3 and 4 dominated. In addition, vitamin B12 was more than 2000 pg/ml. The neutrophil alkaline phosphatase score was 145. Bone marrow (BM) aspiration demonstrated hypercellular
Correspondence: Y Ito, First Department of Internal Medicine, Tokyo Medical University, 6–7-1 Nishishinjuku, Shinjuku-ku, Tokyo 160– 0023, Japan; Fax: + 81 3 5381 6651 Received 25 July 2001; accepted 20 December 2001
marrow (nuclear cell count (NCC) of 674 × 109/l) and myeloid hyperplasia (myeloyd/erythroid ratio, 12.9) with 2.4% myeloblasts, 1.2% promyelocytes, 13.2% eosinophils, 1.6% basophils and 6.4% erythroblasts. All of the hematologic results were compatible with typical CML. Chromosomal findings of bone marrow cells seemed to be 46,XX in 20 cells (Figure 1a). Based on these data, Ph-negative CML was diagnosed. Detection of the major BCR-ABL mRNA of bone marrow cells from this patient by RT-PCR analysis was performed as previously described,8 using a primer ABL1 (5′-GGC CCA TGG TAC CAG GAG TG-3′) as a reverse transcription primer, followed by the amplification of cDNA with a first primer pair ABL1 and BCR1 (5′-GCT TCT CCC TGA CAT CCG TG-3′) for 35 cycles and a second primer pair ABL2 (5′-GTT TCT CCA GAC TGT TGA CTG-3′) and BCR2 (5′-GGA GCT GCA GAT GCT GAC CAA C-3′) for 35 cycles, respectively. RTPCR analysis detected a 446 bp major BCR-ABL transcript, corresponding to b3a2 fusion (Figure 1c). FISH analysis of BCR-ABL fusion revealed 60.8% positive as compared to 1.9% positive in a normal control. Dual-color FISH analysis demonstrated a BCR-ABL fusion signal on one chromosome 22, a normal green ABL signal and a normal red BCR signal on one chromosome 9 and the other chromosome 22, respectively (Figure 1d). Karyotyping by SKY9 showed t(20;22) without detectable segmental translocation of the 9q34 region (Figure 1b). She was given hydroxyurea for about 2 months, and then was given interferon-alpha after being found positive for major BCR-ABL by RTPCR analysis. Ph-negative and BCR-positive CML patients are now easily analyzed using molecular techniques. The finding of BCR-ABL molecular rearrangements indicates undetected Ph translocation in Ph-negative patients. Cytogenetic study in this patient failed to demonstrate Ph translocation. RT-PCR analysis indicated that this patient had BCRABL rearrangement. This finding was confirmed by dual-color FISH analysis demonstrating a BCR-ABL fusion signal on one chromosome 22. Meanwhile, karyotyping by SKY analysis demonstrated t(20;22) without abnormality in chromosome 9. These results indicate that only a small segment of chromosome 9 including ABL is located on chromosome 22. Inazawa et al1 speculated that retranslocation between chromosome 9 and 22 might result in Ph-negative CML. Other reports suggested insertion of a small segment of chromosome 9 into chromosome 22 to form M-BCR in Ph-negative CML.3,4 Moreover, masked Ph chromosomes in the literature usually have a 9q+ anomaly that inspires further examination of the translocated part on 9q34. However, the current case showed neither 9q+ nor 22q− anomaly using standard G- and Q-bandings. A single cytogenetic event seems to be more plausible than a twoLeukemia
