No Advantage of Dexamethasone Over Prednisolone for the Outcome of Standard- and Intermediate-Risk Childhood Acute Lymphoblastic Leukemia in
http://www.100md.com
《临床肿瘤学》
the Tokyo Children's Cancer Study Group
Departments of Hematology and Epidemiology, National Center for Child Health and Development
Department of Pediatrics, St Luke's International Hospital
First Department of Pediatrics, Toho University
Department of Pediatrics, Tokyo Medical and Dental University, School of Medicine
Department of Pediatrics, Nippon Medical School, Tokyo
Department of Pediatrics, Japanese Red Cross Narita Hospital, Narita
Department of Hematology-Oncology, Chiba Children's Hospital, Chiba
Department of Pediatrics, University of Shinshu, School of Medicine, Matsumoto
Department of Pediatrics, Showa University, School of Medicine, Fujigaoka Hospital
Department of Pediatrics, Yokohama City University, School of Medicine, Yokohama
Department of Pediatrics, Japanese Red Cross Maebashi Hospital, Maebashi
Department of Pediatrics, Dokkyo University School of Medicine, Utsunomiya
Department of Pediatrics, Yamanashi University, School of Medicine, Kofu
Department of Pediatrics, Tokai University, School of Medicine, Isehara
Department of Pediatrics, St Marianna University School of Medicine, Kawasaki
Department of Hematology-Oncology, Tokyo Metropolitan Kiyose Children's Hospital, Kiyose
Department of Hematology-Oncology, Gunma Children's Medical Center, Seta
Department of Hematology-Oncology, Saitama Children's Medical Center, Iwatsuki
Department of Pediatrics, Ibaraki Children's Hospital, Mito, Japan
ABSTRACT
PURPOSE: To evaluate whether dexamethasone (DEXA) yields a better outcome than prednisolone (PRED) in a prospective, randomized, controlled trial for the treatment of childhood acute lymphoblastic leukemia (ALL).
PATIENTS AND METHODS: Two hundred thirty-one standard-risk (SR) patients and 128 intermediate-risk (IR) non–B-cell ALL patients were registered from March 1995 to March 1999. After random assignment in each group, the PRED arm patients received PRED 60 mg/m2 during induction followed by PRED 40 mg/m2 over four intensifications in the SR group and three intensifications in the IR group. DEXA arm patients received DEXA 8 mg/m2 during induction and DEXA 6 mg/m2 during the intensifications. The maintenance phase was continued until week 104.
RESULTS: Event-free survival rates at 8 years in the DEXA and PRED arms were 81.1% ± 3.9% (n = 117) and 84.4% ± 5.2% (n = 114), respectively, in the SR group (P = .217) and 84.9% ± 4.6% (n = 62) and 80.4% ± 5.1% (n = 66), respectively, in the IR group (P = .625). The primary reason for treatment failure was marrow relapse. Only two extramedullary relapses occurred in the DEXA arm compared with seven relapses in the PRED arm. Although complications were more prevalent in the DEXA arm than in the PRED arm, fatal toxicity was rare both groups.
CONCLUSION: DEXA administered at 8 mg/m2 during induction and 6 mg/m2 during intensification showed no advantage over PRED administered at 60 mg/m2 during induction and 40 mg/m2 during intensification in both the SR and IR groups.
INTRODUCTION
Glucocorticoids are essential in the treatment of childhood acute lymphoblastic leukemia (ALL). The use of dexamethasone (DEXA) as an alternative drug for prednisolone (PRED) is an important issue in the treatment of childhood ALL, particularly for patients in lower risk groups. DEXA provides better CNS penetration than PRED,1 and the advantages of DEXA over PRED have been reported in several studies.2-4 In vitro assays show that DEXA is five to six times more cytotoxic to leukemic lymphoblasts than PRED, which is in agreement with the documented anti-inflammatory activities of these drugs.5 Therefore, DEXA is unlikely to produce a more potent cytotoxicity in the currently used dosage of PRED to DEXA ratio of 6 to 7.6
The superiority of DEXA over PRED in reducing CNS relapse is well documented, as is the serious toxicity associated with DEXA. A Cancer and Leukemia Group B study found that children randomly administered DEXA had a lower CNS relapse rate than children administered PRED, although the overall event-free survival (EFS) was similar.2 The Dutch ALL Study VI and Dana-Farber Consortium replaced PRED with DEXA and found better outcomes compared with an historical control.3,4 A randomized study of DEXA versus PRED by the Children's Cancer Group (CCG) found that DEXA yielded a statistically significant and clinically important decrease in the rate of isolated CNS relapse and an increase in EFS.8 A randomized study by the United Kingdom Children's Cancer Group comparing DEXA and PRED found DEXA to be more effective than PRED in the treatment of childhood lymphoblastic leukemia, particularly in preventing CNS relapses without serious toxicity.8 However, DEXA-associated toxicity has been reported in several studies.9-14 The incidence of complications, especially osteonecrosis, was higher in the group of patients receiving DEXA than in the patient group receiving PRED.11,12
In the L95-14 protocol of the Tokyo Children's Cancer Study Group (TCCSG), the hypothesis that DEXA is superior to PRED in preventing CNS relapse and in providing better EFS was tested. Here, we report the treatment outcome of the TCCSG L95-14 protocol, a prospective, randomized, controlled trial that investigated the merits of DEXA versus PRED.
PATIENTS AND METHODS
Patients
The prospective, randomized, controlled trial opened in March 1995 and closed in March 1999. Two hundred thirty-one standard-risk (SR) patients and 128 intermediate-risk (IR) non–B-cell ALL patients were registered. Written informed consent was obtained from the parents or guardians and from the patients as appropriate for their age and conceptual ability.
Diagnosis of ALL was based on the morphologic, biochemical, and flow cytometric features of leukemic cells, including lymphoblast morphology on May- or Wright-Giemsa–stained bone marrow smears, negative staining for myeloperoxidase, and reactivity with monoclonal antibodies to B-lineage–associated or T-lineage–associated lymphoid differentiation antigens, as described previously.15 Remission was defined as the presence of fewer than 5% blasts with recovery of hematopoiesis.
The definition of SR was non-T phenotype, an age of between 1 and 6 years, and a leukocyte count less than 20 x 109/L at diagnosis. IR was defined by any one of the following: an age of 1 to 6 years and a leukocyte count between 20 and 100 x 109/L; an age of 7 to 9 years and a leukocyte count less than 20 x 109/L; or a definition of SR with T markers. Thus, all the patients in the SR and IR groups were less than 10 years old. Patients with cytogenetics of Ph1, 11q23, t(1;19), mediastinal mass, or meningeal infiltration were excluded from both groups and stratified into the high-risk group. The median patient follow-up period was 5.2 years for the SR group and 5.4 years for IR patients.
Treatment Protocol
The protocol was approved by the institutional review board of the participating institutions or the equivalent organization. Remission induction therapy consisted of a standard four-drug regimen with triple intrathecal injections twice during induction. In the IR group, cyclophosphamide 1 g/m2 was added to the SR regimen.
In each group, the patients were randomly assigned into either the PRED or DEXA arm. PRED 60 mg/m2 (no more than 80 mg) or DEXA 8 mg/m2 (no more than 10 mg) was administered in the remission induction phase. PRED 40 mg/m2 or DEXA 6 mg/m2 was used in four intensification phases in the SR group and in three intensification phases in the IR group. Prophylactic cranial radiotherapy (CRT) of 18 Gy was administered only to the IR patients (n = 23) with an initial leukocyte count of more than 50 x 109/L. The other patients in the IR group were treated with high-dose methotrexate instead of CRT (n = 105). A maintenance phase consisting of mercaptopurine and methotrexate was continued until week 104 (Tables 1 and 2; Figs 1 and 2).
Statistical Methods
The duration of EFS was defined as the time from the initiation of therapy to either treatment failure (relapse, death, or diagnosis of secondary cancer) or to the last day when the patient was confirmed to be well. Patients who did not achieve complete remission (CR) after the first induction phase or who died before the confirmation of remission were considered to have experienced treatment failure at day 0. Probability of EFS was estimated by the Kaplan-Meier method16 and was tested for significant difference using the log-rank test. Patients who were inadvertently or intentionally treated by the different regimens were analyzed according to their randomly assigned arm. Three SR and two IR patients in the DEXA arm selected the PRED regimen against the randomized assignment, whereas one IR patient in the PRED arm selected the DEXA regimen against the randomized assignment. All of the analyses were performed according to the intent-to-treat principle.
RESULTS
Patient Characteristics
The presenting characteristics of the 231 SR and 128 IR patients are listed in Table 3. There were no significant differences between the DEXA and PRED arms. Although there was a trend that patients in the SR group of the PRED arm had a higher leukocyte count than those in the DEXA arm, the DNA index values were similar in both groups.
Treatment Outcome
A total of 352 patients (98.1%) achieved first CR. The CR rate was 98.7% in the SR patients, with a rate of 98.3% in the DEXA arm and 99.1% in the PRED arm. The CR rate was 96.9% in the IR patients, with a rate of 95.2% in the DEXA arm and 98.5% in the PRED arm (Table 4).
Two patients (1.7%) in the DEXA arm and one patient (0.9%) in the PRED arm in the SR group died during the induction phase. In the IR group, three patients (4.8%) in the DEXA arm and one patient (1.5%) in the PRED arm died during the induction. All deaths were a result of infection.
The probability (mean ± SE) of EFS for all the patients in the SR group was 82.8% ± 3.2% at 8 years, whereas the EFS of patients in the DEXA and PRED arms was 81.1% ± 3.9% (n = 117) and 84.4% ± 5.2% (n = 114), respectively, which was not significantly different (P = .217). Patients in the IR group had an EFS probability of 82.6% ± 3.5% (n = 128). There was no significant difference in EFS (P = .625) between the DEXA patients (84.9% ± 4.6%; n = 62) and the PRED patients (80.4% ± 5.1%; n = 66; Figs 3 and 4). The major cause of treatment failure was marrow relapse.
In the SR group, there were 18 patients with bone marrow relapse but no patients with extramedullary relapse in the DEXA arm, whereas six patients in the PRED arm showed bone marrow relapse (two had a CNS relapse, three showed testicular relapse, and one developed testicular/marrow relapse). In the IR group, one patient in the DEXA arm developed CNS relapse, one patient showed CNS/marrow relapse, and four patients showed marrow relapse; whereas one patient in the PRED arm suffered CNS/marrow relapse, and seven patients developed marrow relapse and a secondary myelodysplastic syndrome.
Prophylactic CRT of 18 Gy was administered only to IR patients who displayed an initial leukocyte count of more than 50 x 109/L. Twenty-three of 128 IR patients received CRT. The EFS probability for the patients in the DEXA arm not receiving CRT was 85.7% ± 5.0% (n = 51) compared with 81.6% ± 5.6% (n = 54) for the patients in the PRED arm not receiving CRT (P = .680). In contrast, the EFS rates of patients administered CRT were 81.8% ± 11.6% (n = 11) and 75.0% ± 12.5% (n = 12) in the DEXA and PRED arms, respectively, without a significant difference (P = .787). In addition, the differences in site of relapse and relapse rate between the patients in the DEXA and PRED arms who received CRT were not significant.
Toxicity
During the induction therapy, 11 patients in the PRED arm and 19 patients in the DEXA arm developed sepsis (P = .18; Table 5). Fungal infection was seen in one PRED patient and in one DEXA patient from the IR group. Seven patients (1.9%) died of infection during induction; two of these patients were in the PRED arm (Pseudomonas aeruginosa), and five were in the DEXA arm (two P aeruginosa, two Bacillus cereus, and one unknown organism; P = .79). After the induction therapy phase, nine and 11 patients developed sepsis in the PRED and DEXA arms, respectively.
An episode of L-asparaginase–related coagulopathy presenting as CNS hemorrhage and thrombotic stroke was seen in two PRED patients and in four DEXA patients (P = .67). Pancreatitis occurred in two of the DEXA patients (P = .47), and symptomatic osteonecrosis developed in three of the DEXA patients (P = .24). Neuropsychiatric adverse events, such as severe agitation, were observed in three DEXA patients (P = .24). Other rare toxicities observed included a virus-associated hemophagocytosis, which was observed in a DEXA patient, and a Moyamoya disease in one DEXA patient who received CRT. Relatively nonsevere toxicities, such as steroid myopathy, hyperlipidemia, and hyperglycemia, were reported but not statistically analyzed because a specific questionnaire covering these symptoms was not included prospectively in the patient report form. With the exception of the patient suffering osteonecrosis, all patients recovered without serious sequelae.
Two patients in the PRED arm of the IR group died of unidentified encephalopathy during the intensification therapy. In summary, the incidence of severe toxicity was higher in the DEXA arm than in the PRED arm, but the differences were not statistically significant.
DISCUSSION
Corticosteroids are essential in the treatment of ALL.6 The Berlin-Frankfurt-Munster Group, the CCG, and the United Kingdom Acute Lymphoblastic Leukemia Group all recommend the use of DEXA for postinduction therapy.8,17-19 However, the superiority of DEXA over PRED may not always hold true, and the merit of one drug over the other seems to depend on the dosage. In a randomized trial conducted by the CCG, a statistically significant decrease in the rate of isolated CNS relapse with an improved 6-year EFS was noted in patients who received DEXA; the 6-year EFS rate was 85% ± 2% for DEXA patients and 77% ± 2% for PRED patients (P = .002).7 However, the study reported here found no statistically significant difference in EFS between the DEXA and PRED arms in either the SR or IR groups. The clinical outcomes for patients administered the DEXA regimens in the TCCSG and CCG were almost equivalent, although the advantages for the DEXA arm patients presented by the CCG study was not realized in the TCCSG. The EFS of the PRED arms of the TCCSG was better than that of the CCG, which may be a result of the higher dosage of PRED (60 mg/m2 over 31 days with a 7-day taper) used in our study compared with the CCG study (40 mg/m2 over 28 days with a 7- to 10-day taper). It should be noted that the backbone therapy in TCCSG L95-14 was more aggressive than that in CCG 1922 because the former contained doxorubicin during induction, with intermediate- and high-dose methotrexate and high-dose cytarabine.
The sites of relapse in the SR group differed remarkably between the PRED and DEXA arm patients. The replacement of PRED with DEXA apparently reduced the rate of extramedullary relapses, but the rate of bone marrow relapses increased. Extramedullary relapse seemed to be less frequent in the DEXA regimen, which has also been observed by other investigators.3,7 Such a difference was not observed in the IR group in our study. The augmentation of the chemotherapy using cyclophosphamide and high-dose cytarabine in the IR regimen resulted in a low relapse rate in the PRED group, and consequently, it might obscure the expected advantage of using DEXA.
In this study, high EFS rates were obtained in both the SR and IR groups. This may be attributed to the intensive induction treatment (standard three-drug regimen with anthracyclines ± cyclophosphamide). The advantage of using DEXA may not have been strongly evident because of this low overall rate of treatment failure.
There is a tendency towards a higher incidence of notable complications in the DEXA arm than in the PRED arm. Hurwitz et al9 described an increased incidence of Gram-negative bacteremia and induction death in a group of patients who received DEXA during induction compared with historical controls who received PRED. In our study, just as described, septicemia was more frequently observed in DEXA patients than in PRED patients. It should be noted that this was in the context of a five-drug myelosuppressive induction schedule including anthracycline and cyclophosphamide. This was not seen in the CCG 1922 study, which used a three-drug induction schedule (no anthracycline).7 Osteonecrosis is a known hazardous complication of corticosteroids. Mattano et al11 found an incidence of 14% in patients older than 10 years of age compared with 1% in younger patients. Risk factors included age greater than 10 years, female sex, and white race, as well as the number of DEXA-containing delayed-intensification courses. In our current analysis, symptomatic osteonecrosis was found in three DEXA patients (two in the SR group and one in the IR group). None of the patients in the PRED arm had symptomatic osteonecrosis. This was probably because high-risk patients greater than 10 years of age were excluded from the randomized study.
The number of patients required for a P = .05 and a power of .80 was simulated for various differences between the two groups. However, an anticipated difference in EFS or CNS relapse between the PRED and DEXA groups, which was needed in the calculation, was difficult to assume because this study was designed before a difference was reported between them.7 In addition, the number of patients statistically required was expected to far exceed the available number of patients in a few years as the anticipated difference in end points between the two groups should not be large. Therefore, we placed more emphasis on carrying out a solid study to be incorporated into a meta-analysis to yield a firm conclusion. Consequently, the recruitment of the patients was limited to a 4-year period.
We conclude that DEXA 8 mg/m2 during induction and 6 mg/m2 at each of the intensification phases showed no advantage over PRED 60 mg/m2 at induction and 40 mg/m2 at intensifications. Considering the tendency towards a higher incidence of complications in the DEXA arms, we recommend that DEXA should be reserved for a subset of patients who have an increased risk of CNS relapse.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Kaori Itagaki for preparing and refining the protocol data for acute lymphoblastic leukemia in the Tokyo Children's Cancer Study Group. We also thank the pediatricians and nurses who participated in the treatment and follow-up of the patients in this study.
NOTES
Supported in part by a grant from the Children's Cancer Association of Japan.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Balis FM, Lester CM, Chrousos GP, et al: Differences in cerebrospinal fluid penetration of corticosteroids: Possible relationship to the prevention of meningeal leukemia. J Clin Oncol 5:202-207, 1987
Jones B, Freeman AI, Shuster JJ, et al: Lower incidence of meningeal leukemia when prednisone is replaced by dexamethasone in the treatment of acute lymphocytic leukemia. Med Pediatr Oncol 19:269-275, 1991
Veerman AJ, Hahlen K, Kamps WA, et al: High cure rate with a moderately intensive treatment regimen in non-high-risk childhood acute lymphoblastic leukemia: Results of protocol ALL VI from the Dutch Childhood Leukemia Study Group. J Clin Oncol 14:911-918, 1996
Silverman LB, Gelber RD, Dalton VK, et al: Improved outcome for children with acute lymphoblastic leukemia: Results of Dana-Farber Consortium Protocol 91-01. Blood 97:1211-1218, 2001
Ito C, Evans WE, McNinch L, et al: Comparative cytotoxicity of dexamethasone and prednisolone in childhood acute lymphoblastic leukemia. J Clin Oncol 14:2370-2376, 1996
Gaynon PS, Lustig RH: The use of glucocorticoids in acute lymphoblastic leukemia of childhood: Molecular, cellular, and clinical considerations. J Pediatr Hematol Oncol 17:1-12, 1995
Bostrom BC, Sensel MR, Sather HN, et al: Dexamethasone versus prednisone and daily oral versus weekly intravenous mercaptopurine for patients with standard-risk acute lymphoblastic leukemia: A report from the Children's Cancer Group. Blood 101:3809-3817, 2003
Mitchell CD, Kinsey S, Vora AJ, et al: MRC ALL97/99: A randomized comparison of dexamethasone versus prednisolone and thioguanine versus mercaptopurine in the treatment of childhood acute lymphoblastic leukemia. Blood 100:586, 2002 (abstr 586)
Hurwitz CA, Silverman LB, Schorin MA, et al: Substituting dexamethasone for prednisone complicates remission induction in children with acute lymphoblastic leukemia. Cancer 88:1964-1969, 2000
Belgaumi AF, Al-Bakrah M, Al-Mahr M, et al: Dexamethasone-associated toxicity during induction chemotherapy for childhood acute lymphoblastic leukemia is augmented by concurrent use of daunomycin. Cancer 97:2898-2903, 2003
Mattano LA Jr, Sather HN, Trigg ME, et al: Osteonecrosis as a complication of treating acute lymphoblastic leukemia in children: A report from the Children's Cancer Group. J Clin Oncol 18:3262-3272, 2000
Arico M, Boccalatte MF, Silvestri D, et al: Osteonecrosis: An emerging complication of intensive chemotherapy for childhood acute lymphoblastic leukemia. Haematologica 88:747-753, 2003
Dropcho EJ, Soong SJ: Steroid-induced weakness in patients with primary brain tumors. Neurology 41:1235-1239, 1991
Waber DP, Carpentieri SC, Klar N, et al: Cognitive sequelae in children treated for acute lymphoblastic leukemia with dexamethasone or prednisone. J Pediatr Hematol Oncol 22:206-213, 2000
Toyoda Y, Manabe A, Masahiro T, et al: Six months of maintenance chemotherapy after intensified treatment for acute lymphoblastic leukemia of childhood. J Clin Oncol 18:1508-1516, 2000
Kaplan E, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Reiter A, Schrappe M, Ludwig WD, et al: Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients: Results and conclusions of the multicenter trial ALL-BFM 86. Blood 84:3122-3133, 1994
Lange BJ, Bostrom BC, Cherlow JM, et al: Double-delayed intensification improves event-free survival for children with intermediate-risk acute lymphoblastic leukemia: A report from the Children's Cancer Group. Blood 99:825-833, 2002
Hann I, Vora A, Richards S, et al: Benefit of intensified treatment for all children with acute lymphoblastic leukaemia: Results from MRC UKALL XI and MRC ALL97 randomised trials—UK Medical Research Council's Working Party on Childhood Leukaemia. Leukemia 14:356-363, 2000(Shunji Igarashi, Atsushi )
Departments of Hematology and Epidemiology, National Center for Child Health and Development
Department of Pediatrics, St Luke's International Hospital
First Department of Pediatrics, Toho University
Department of Pediatrics, Tokyo Medical and Dental University, School of Medicine
Department of Pediatrics, Nippon Medical School, Tokyo
Department of Pediatrics, Japanese Red Cross Narita Hospital, Narita
Department of Hematology-Oncology, Chiba Children's Hospital, Chiba
Department of Pediatrics, University of Shinshu, School of Medicine, Matsumoto
Department of Pediatrics, Showa University, School of Medicine, Fujigaoka Hospital
Department of Pediatrics, Yokohama City University, School of Medicine, Yokohama
Department of Pediatrics, Japanese Red Cross Maebashi Hospital, Maebashi
Department of Pediatrics, Dokkyo University School of Medicine, Utsunomiya
Department of Pediatrics, Yamanashi University, School of Medicine, Kofu
Department of Pediatrics, Tokai University, School of Medicine, Isehara
Department of Pediatrics, St Marianna University School of Medicine, Kawasaki
Department of Hematology-Oncology, Tokyo Metropolitan Kiyose Children's Hospital, Kiyose
Department of Hematology-Oncology, Gunma Children's Medical Center, Seta
Department of Hematology-Oncology, Saitama Children's Medical Center, Iwatsuki
Department of Pediatrics, Ibaraki Children's Hospital, Mito, Japan
ABSTRACT
PURPOSE: To evaluate whether dexamethasone (DEXA) yields a better outcome than prednisolone (PRED) in a prospective, randomized, controlled trial for the treatment of childhood acute lymphoblastic leukemia (ALL).
PATIENTS AND METHODS: Two hundred thirty-one standard-risk (SR) patients and 128 intermediate-risk (IR) non–B-cell ALL patients were registered from March 1995 to March 1999. After random assignment in each group, the PRED arm patients received PRED 60 mg/m2 during induction followed by PRED 40 mg/m2 over four intensifications in the SR group and three intensifications in the IR group. DEXA arm patients received DEXA 8 mg/m2 during induction and DEXA 6 mg/m2 during the intensifications. The maintenance phase was continued until week 104.
RESULTS: Event-free survival rates at 8 years in the DEXA and PRED arms were 81.1% ± 3.9% (n = 117) and 84.4% ± 5.2% (n = 114), respectively, in the SR group (P = .217) and 84.9% ± 4.6% (n = 62) and 80.4% ± 5.1% (n = 66), respectively, in the IR group (P = .625). The primary reason for treatment failure was marrow relapse. Only two extramedullary relapses occurred in the DEXA arm compared with seven relapses in the PRED arm. Although complications were more prevalent in the DEXA arm than in the PRED arm, fatal toxicity was rare both groups.
CONCLUSION: DEXA administered at 8 mg/m2 during induction and 6 mg/m2 during intensification showed no advantage over PRED administered at 60 mg/m2 during induction and 40 mg/m2 during intensification in both the SR and IR groups.
INTRODUCTION
Glucocorticoids are essential in the treatment of childhood acute lymphoblastic leukemia (ALL). The use of dexamethasone (DEXA) as an alternative drug for prednisolone (PRED) is an important issue in the treatment of childhood ALL, particularly for patients in lower risk groups. DEXA provides better CNS penetration than PRED,1 and the advantages of DEXA over PRED have been reported in several studies.2-4 In vitro assays show that DEXA is five to six times more cytotoxic to leukemic lymphoblasts than PRED, which is in agreement with the documented anti-inflammatory activities of these drugs.5 Therefore, DEXA is unlikely to produce a more potent cytotoxicity in the currently used dosage of PRED to DEXA ratio of 6 to 7.6
The superiority of DEXA over PRED in reducing CNS relapse is well documented, as is the serious toxicity associated with DEXA. A Cancer and Leukemia Group B study found that children randomly administered DEXA had a lower CNS relapse rate than children administered PRED, although the overall event-free survival (EFS) was similar.2 The Dutch ALL Study VI and Dana-Farber Consortium replaced PRED with DEXA and found better outcomes compared with an historical control.3,4 A randomized study of DEXA versus PRED by the Children's Cancer Group (CCG) found that DEXA yielded a statistically significant and clinically important decrease in the rate of isolated CNS relapse and an increase in EFS.8 A randomized study by the United Kingdom Children's Cancer Group comparing DEXA and PRED found DEXA to be more effective than PRED in the treatment of childhood lymphoblastic leukemia, particularly in preventing CNS relapses without serious toxicity.8 However, DEXA-associated toxicity has been reported in several studies.9-14 The incidence of complications, especially osteonecrosis, was higher in the group of patients receiving DEXA than in the patient group receiving PRED.11,12
In the L95-14 protocol of the Tokyo Children's Cancer Study Group (TCCSG), the hypothesis that DEXA is superior to PRED in preventing CNS relapse and in providing better EFS was tested. Here, we report the treatment outcome of the TCCSG L95-14 protocol, a prospective, randomized, controlled trial that investigated the merits of DEXA versus PRED.
PATIENTS AND METHODS
Patients
The prospective, randomized, controlled trial opened in March 1995 and closed in March 1999. Two hundred thirty-one standard-risk (SR) patients and 128 intermediate-risk (IR) non–B-cell ALL patients were registered. Written informed consent was obtained from the parents or guardians and from the patients as appropriate for their age and conceptual ability.
Diagnosis of ALL was based on the morphologic, biochemical, and flow cytometric features of leukemic cells, including lymphoblast morphology on May- or Wright-Giemsa–stained bone marrow smears, negative staining for myeloperoxidase, and reactivity with monoclonal antibodies to B-lineage–associated or T-lineage–associated lymphoid differentiation antigens, as described previously.15 Remission was defined as the presence of fewer than 5% blasts with recovery of hematopoiesis.
The definition of SR was non-T phenotype, an age of between 1 and 6 years, and a leukocyte count less than 20 x 109/L at diagnosis. IR was defined by any one of the following: an age of 1 to 6 years and a leukocyte count between 20 and 100 x 109/L; an age of 7 to 9 years and a leukocyte count less than 20 x 109/L; or a definition of SR with T markers. Thus, all the patients in the SR and IR groups were less than 10 years old. Patients with cytogenetics of Ph1, 11q23, t(1;19), mediastinal mass, or meningeal infiltration were excluded from both groups and stratified into the high-risk group. The median patient follow-up period was 5.2 years for the SR group and 5.4 years for IR patients.
Treatment Protocol
The protocol was approved by the institutional review board of the participating institutions or the equivalent organization. Remission induction therapy consisted of a standard four-drug regimen with triple intrathecal injections twice during induction. In the IR group, cyclophosphamide 1 g/m2 was added to the SR regimen.
In each group, the patients were randomly assigned into either the PRED or DEXA arm. PRED 60 mg/m2 (no more than 80 mg) or DEXA 8 mg/m2 (no more than 10 mg) was administered in the remission induction phase. PRED 40 mg/m2 or DEXA 6 mg/m2 was used in four intensification phases in the SR group and in three intensification phases in the IR group. Prophylactic cranial radiotherapy (CRT) of 18 Gy was administered only to the IR patients (n = 23) with an initial leukocyte count of more than 50 x 109/L. The other patients in the IR group were treated with high-dose methotrexate instead of CRT (n = 105). A maintenance phase consisting of mercaptopurine and methotrexate was continued until week 104 (Tables 1 and 2; Figs 1 and 2).
Statistical Methods
The duration of EFS was defined as the time from the initiation of therapy to either treatment failure (relapse, death, or diagnosis of secondary cancer) or to the last day when the patient was confirmed to be well. Patients who did not achieve complete remission (CR) after the first induction phase or who died before the confirmation of remission were considered to have experienced treatment failure at day 0. Probability of EFS was estimated by the Kaplan-Meier method16 and was tested for significant difference using the log-rank test. Patients who were inadvertently or intentionally treated by the different regimens were analyzed according to their randomly assigned arm. Three SR and two IR patients in the DEXA arm selected the PRED regimen against the randomized assignment, whereas one IR patient in the PRED arm selected the DEXA regimen against the randomized assignment. All of the analyses were performed according to the intent-to-treat principle.
RESULTS
Patient Characteristics
The presenting characteristics of the 231 SR and 128 IR patients are listed in Table 3. There were no significant differences between the DEXA and PRED arms. Although there was a trend that patients in the SR group of the PRED arm had a higher leukocyte count than those in the DEXA arm, the DNA index values were similar in both groups.
Treatment Outcome
A total of 352 patients (98.1%) achieved first CR. The CR rate was 98.7% in the SR patients, with a rate of 98.3% in the DEXA arm and 99.1% in the PRED arm. The CR rate was 96.9% in the IR patients, with a rate of 95.2% in the DEXA arm and 98.5% in the PRED arm (Table 4).
Two patients (1.7%) in the DEXA arm and one patient (0.9%) in the PRED arm in the SR group died during the induction phase. In the IR group, three patients (4.8%) in the DEXA arm and one patient (1.5%) in the PRED arm died during the induction. All deaths were a result of infection.
The probability (mean ± SE) of EFS for all the patients in the SR group was 82.8% ± 3.2% at 8 years, whereas the EFS of patients in the DEXA and PRED arms was 81.1% ± 3.9% (n = 117) and 84.4% ± 5.2% (n = 114), respectively, which was not significantly different (P = .217). Patients in the IR group had an EFS probability of 82.6% ± 3.5% (n = 128). There was no significant difference in EFS (P = .625) between the DEXA patients (84.9% ± 4.6%; n = 62) and the PRED patients (80.4% ± 5.1%; n = 66; Figs 3 and 4). The major cause of treatment failure was marrow relapse.
In the SR group, there were 18 patients with bone marrow relapse but no patients with extramedullary relapse in the DEXA arm, whereas six patients in the PRED arm showed bone marrow relapse (two had a CNS relapse, three showed testicular relapse, and one developed testicular/marrow relapse). In the IR group, one patient in the DEXA arm developed CNS relapse, one patient showed CNS/marrow relapse, and four patients showed marrow relapse; whereas one patient in the PRED arm suffered CNS/marrow relapse, and seven patients developed marrow relapse and a secondary myelodysplastic syndrome.
Prophylactic CRT of 18 Gy was administered only to IR patients who displayed an initial leukocyte count of more than 50 x 109/L. Twenty-three of 128 IR patients received CRT. The EFS probability for the patients in the DEXA arm not receiving CRT was 85.7% ± 5.0% (n = 51) compared with 81.6% ± 5.6% (n = 54) for the patients in the PRED arm not receiving CRT (P = .680). In contrast, the EFS rates of patients administered CRT were 81.8% ± 11.6% (n = 11) and 75.0% ± 12.5% (n = 12) in the DEXA and PRED arms, respectively, without a significant difference (P = .787). In addition, the differences in site of relapse and relapse rate between the patients in the DEXA and PRED arms who received CRT were not significant.
Toxicity
During the induction therapy, 11 patients in the PRED arm and 19 patients in the DEXA arm developed sepsis (P = .18; Table 5). Fungal infection was seen in one PRED patient and in one DEXA patient from the IR group. Seven patients (1.9%) died of infection during induction; two of these patients were in the PRED arm (Pseudomonas aeruginosa), and five were in the DEXA arm (two P aeruginosa, two Bacillus cereus, and one unknown organism; P = .79). After the induction therapy phase, nine and 11 patients developed sepsis in the PRED and DEXA arms, respectively.
An episode of L-asparaginase–related coagulopathy presenting as CNS hemorrhage and thrombotic stroke was seen in two PRED patients and in four DEXA patients (P = .67). Pancreatitis occurred in two of the DEXA patients (P = .47), and symptomatic osteonecrosis developed in three of the DEXA patients (P = .24). Neuropsychiatric adverse events, such as severe agitation, were observed in three DEXA patients (P = .24). Other rare toxicities observed included a virus-associated hemophagocytosis, which was observed in a DEXA patient, and a Moyamoya disease in one DEXA patient who received CRT. Relatively nonsevere toxicities, such as steroid myopathy, hyperlipidemia, and hyperglycemia, were reported but not statistically analyzed because a specific questionnaire covering these symptoms was not included prospectively in the patient report form. With the exception of the patient suffering osteonecrosis, all patients recovered without serious sequelae.
Two patients in the PRED arm of the IR group died of unidentified encephalopathy during the intensification therapy. In summary, the incidence of severe toxicity was higher in the DEXA arm than in the PRED arm, but the differences were not statistically significant.
DISCUSSION
Corticosteroids are essential in the treatment of ALL.6 The Berlin-Frankfurt-Munster Group, the CCG, and the United Kingdom Acute Lymphoblastic Leukemia Group all recommend the use of DEXA for postinduction therapy.8,17-19 However, the superiority of DEXA over PRED may not always hold true, and the merit of one drug over the other seems to depend on the dosage. In a randomized trial conducted by the CCG, a statistically significant decrease in the rate of isolated CNS relapse with an improved 6-year EFS was noted in patients who received DEXA; the 6-year EFS rate was 85% ± 2% for DEXA patients and 77% ± 2% for PRED patients (P = .002).7 However, the study reported here found no statistically significant difference in EFS between the DEXA and PRED arms in either the SR or IR groups. The clinical outcomes for patients administered the DEXA regimens in the TCCSG and CCG were almost equivalent, although the advantages for the DEXA arm patients presented by the CCG study was not realized in the TCCSG. The EFS of the PRED arms of the TCCSG was better than that of the CCG, which may be a result of the higher dosage of PRED (60 mg/m2 over 31 days with a 7-day taper) used in our study compared with the CCG study (40 mg/m2 over 28 days with a 7- to 10-day taper). It should be noted that the backbone therapy in TCCSG L95-14 was more aggressive than that in CCG 1922 because the former contained doxorubicin during induction, with intermediate- and high-dose methotrexate and high-dose cytarabine.
The sites of relapse in the SR group differed remarkably between the PRED and DEXA arm patients. The replacement of PRED with DEXA apparently reduced the rate of extramedullary relapses, but the rate of bone marrow relapses increased. Extramedullary relapse seemed to be less frequent in the DEXA regimen, which has also been observed by other investigators.3,7 Such a difference was not observed in the IR group in our study. The augmentation of the chemotherapy using cyclophosphamide and high-dose cytarabine in the IR regimen resulted in a low relapse rate in the PRED group, and consequently, it might obscure the expected advantage of using DEXA.
In this study, high EFS rates were obtained in both the SR and IR groups. This may be attributed to the intensive induction treatment (standard three-drug regimen with anthracyclines ± cyclophosphamide). The advantage of using DEXA may not have been strongly evident because of this low overall rate of treatment failure.
There is a tendency towards a higher incidence of notable complications in the DEXA arm than in the PRED arm. Hurwitz et al9 described an increased incidence of Gram-negative bacteremia and induction death in a group of patients who received DEXA during induction compared with historical controls who received PRED. In our study, just as described, septicemia was more frequently observed in DEXA patients than in PRED patients. It should be noted that this was in the context of a five-drug myelosuppressive induction schedule including anthracycline and cyclophosphamide. This was not seen in the CCG 1922 study, which used a three-drug induction schedule (no anthracycline).7 Osteonecrosis is a known hazardous complication of corticosteroids. Mattano et al11 found an incidence of 14% in patients older than 10 years of age compared with 1% in younger patients. Risk factors included age greater than 10 years, female sex, and white race, as well as the number of DEXA-containing delayed-intensification courses. In our current analysis, symptomatic osteonecrosis was found in three DEXA patients (two in the SR group and one in the IR group). None of the patients in the PRED arm had symptomatic osteonecrosis. This was probably because high-risk patients greater than 10 years of age were excluded from the randomized study.
The number of patients required for a P = .05 and a power of .80 was simulated for various differences between the two groups. However, an anticipated difference in EFS or CNS relapse between the PRED and DEXA groups, which was needed in the calculation, was difficult to assume because this study was designed before a difference was reported between them.7 In addition, the number of patients statistically required was expected to far exceed the available number of patients in a few years as the anticipated difference in end points between the two groups should not be large. Therefore, we placed more emphasis on carrying out a solid study to be incorporated into a meta-analysis to yield a firm conclusion. Consequently, the recruitment of the patients was limited to a 4-year period.
We conclude that DEXA 8 mg/m2 during induction and 6 mg/m2 at each of the intensification phases showed no advantage over PRED 60 mg/m2 at induction and 40 mg/m2 at intensifications. Considering the tendency towards a higher incidence of complications in the DEXA arms, we recommend that DEXA should be reserved for a subset of patients who have an increased risk of CNS relapse.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Kaori Itagaki for preparing and refining the protocol data for acute lymphoblastic leukemia in the Tokyo Children's Cancer Study Group. We also thank the pediatricians and nurses who participated in the treatment and follow-up of the patients in this study.
NOTES
Supported in part by a grant from the Children's Cancer Association of Japan.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Balis FM, Lester CM, Chrousos GP, et al: Differences in cerebrospinal fluid penetration of corticosteroids: Possible relationship to the prevention of meningeal leukemia. J Clin Oncol 5:202-207, 1987
Jones B, Freeman AI, Shuster JJ, et al: Lower incidence of meningeal leukemia when prednisone is replaced by dexamethasone in the treatment of acute lymphocytic leukemia. Med Pediatr Oncol 19:269-275, 1991
Veerman AJ, Hahlen K, Kamps WA, et al: High cure rate with a moderately intensive treatment regimen in non-high-risk childhood acute lymphoblastic leukemia: Results of protocol ALL VI from the Dutch Childhood Leukemia Study Group. J Clin Oncol 14:911-918, 1996
Silverman LB, Gelber RD, Dalton VK, et al: Improved outcome for children with acute lymphoblastic leukemia: Results of Dana-Farber Consortium Protocol 91-01. Blood 97:1211-1218, 2001
Ito C, Evans WE, McNinch L, et al: Comparative cytotoxicity of dexamethasone and prednisolone in childhood acute lymphoblastic leukemia. J Clin Oncol 14:2370-2376, 1996
Gaynon PS, Lustig RH: The use of glucocorticoids in acute lymphoblastic leukemia of childhood: Molecular, cellular, and clinical considerations. J Pediatr Hematol Oncol 17:1-12, 1995
Bostrom BC, Sensel MR, Sather HN, et al: Dexamethasone versus prednisone and daily oral versus weekly intravenous mercaptopurine for patients with standard-risk acute lymphoblastic leukemia: A report from the Children's Cancer Group. Blood 101:3809-3817, 2003
Mitchell CD, Kinsey S, Vora AJ, et al: MRC ALL97/99: A randomized comparison of dexamethasone versus prednisolone and thioguanine versus mercaptopurine in the treatment of childhood acute lymphoblastic leukemia. Blood 100:586, 2002 (abstr 586)
Hurwitz CA, Silverman LB, Schorin MA, et al: Substituting dexamethasone for prednisone complicates remission induction in children with acute lymphoblastic leukemia. Cancer 88:1964-1969, 2000
Belgaumi AF, Al-Bakrah M, Al-Mahr M, et al: Dexamethasone-associated toxicity during induction chemotherapy for childhood acute lymphoblastic leukemia is augmented by concurrent use of daunomycin. Cancer 97:2898-2903, 2003
Mattano LA Jr, Sather HN, Trigg ME, et al: Osteonecrosis as a complication of treating acute lymphoblastic leukemia in children: A report from the Children's Cancer Group. J Clin Oncol 18:3262-3272, 2000
Arico M, Boccalatte MF, Silvestri D, et al: Osteonecrosis: An emerging complication of intensive chemotherapy for childhood acute lymphoblastic leukemia. Haematologica 88:747-753, 2003
Dropcho EJ, Soong SJ: Steroid-induced weakness in patients with primary brain tumors. Neurology 41:1235-1239, 1991
Waber DP, Carpentieri SC, Klar N, et al: Cognitive sequelae in children treated for acute lymphoblastic leukemia with dexamethasone or prednisone. J Pediatr Hematol Oncol 22:206-213, 2000
Toyoda Y, Manabe A, Masahiro T, et al: Six months of maintenance chemotherapy after intensified treatment for acute lymphoblastic leukemia of childhood. J Clin Oncol 18:1508-1516, 2000
Kaplan E, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Reiter A, Schrappe M, Ludwig WD, et al: Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients: Results and conclusions of the multicenter trial ALL-BFM 86. Blood 84:3122-3133, 1994
Lange BJ, Bostrom BC, Cherlow JM, et al: Double-delayed intensification improves event-free survival for children with intermediate-risk acute lymphoblastic leukemia: A report from the Children's Cancer Group. Blood 99:825-833, 2002
Hann I, Vora A, Richards S, et al: Benefit of intensified treatment for all children with acute lymphoblastic leukaemia: Results from MRC UKALL XI and MRC ALL97 randomised trials—UK Medical Research Council's Working Party on Childhood Leukaemia. Leukemia 14:356-363, 2000(Shunji Igarashi, Atsushi )