Learn more: PMC Disclaimer | PMC Copyright Notice
Postinduction Dexamethasone and Individualized Dosing of Escherichia Coli L-Asparaginase Each Improve Outcome of Children and Adolescents With Newly Diagnosed Acute Lymphoblastic Leukemia: Results From a Randomized Study—Dana-Farber Cancer Institute ALL Consortium Protocol 00-01
Associated Data
Abstract
Purpose
We assessed the toxicity and efficacy of dexamethasone and a novel dosing method of Escherichia coli L-asparaginase (EC-Asnase) in children and adolescents with newly diagnosed acute lymphoblastic leukemia (ALL).
Patients and Methods
Patients achieving complete remission (CR) on Dana-Farber Cancer Institute ALL Consortium Protocol 00-01 were eligible for random assignment to 1) dexamethasone or prednisone, administered as 5-day pulses, every 3 weeks, and 2) weekly EC-Asnase, administered as a 25,000 IU/m2 fixed dose (FD) or individualized dose (ID) starting at 12,500-IU/m2, adjusted every 3 weeks based on nadir serum asparaginase activity (NSAA) determinations.
Results
Between 2000 and 2004, 492 evaluable patients (ages 1 to 18 years) enrolled; 473 patients (96%) achieved CR. Four hundred eight patients (86%) participated in the corticosteroid randomization and 384 patients (81%) in the EC-Asnase randomization. With 4.9 years of median follow-up, dexamethasone was associated with superior 5-year event-free survival (EFS; 90% v 81% for prednisone; P = .01) but higher rates of infection (P = .03) and, in older children, higher cumulative incidence of osteonecrosis (P = .02) and fracture (P = .06). ID EC-Asnase had superior 5-year EFS (90% v 82% for FD; P = .04), but did not reduce the frequency of asparaginase-related toxicity. Multivariable analysis identified both dexamethasone and ID EC-Asnase as independent predictors of favorable EFS.
Conclusion
There was no overall difference in skeletal toxicity by corticosteroid type; dexamethasone was associated with more infections and, in older children, increased incidence of osteonecrosis and fracture. There was no difference in asparaginase-related toxicity by EC-Asnase dosing method. Dexamethasone and ID EC-Asnase were each associated with superior EFS. Monitoring NSAA during treatment with EC-Asnase may be an effective strategy to improve outcome in pediatric ALL.
INTRODUCTION
Corticosteroids and L-asparaginase are universal components of chemotherapy regimens for childhood acute lymphoblastic leukemia (ALL), although both are associated with significant adverse effects. Interest in substituting dexamethasone for prednisone in ALL treatment arose from studies suggesting that dexamethasone had more potent in vitro antileukemia activity, higher free-plasma levels, and enhanced CSF penetration.1,2 Results of some randomized trials have indicated that dexamethasone was associated with superior event-free survival (EFS), whereas others have found no difference in outcome.3–6 Dexamethasone has been associated with increased toxicities, including higher rates of osteonecrosis, especially in older children and adolescents.7 Thus, the optimal corticosteroid preparation and dosing are unknown, and may differ for patient subgroups.
Similarly, optimal dosing for asparaginase remains unclear. Asparaginase-associated toxicities, such as allergy, pancreatitis, and thrombosis, are a significant source of morbidity and may increase relapse risk in patients unable to receive all intended doses.8 It has been suggested that serum asparaginase activity levels ≥ 0.10 IU/mL are required for therapeutic asparagine depletion.9–14 However, interpatient variability in asparaginase activity levels is high among children receiving a fixed dose (FD; based on body surface) of Escherichia coli L-asparaginase (EC-Asnase).9,15,15 We hypothesized that an individual dosing regimen, monitoring nadir serum asparaginase activity (NSAA) and adjusting the EC-Asnase dose to maintain a therapeutic level, would improve tolerability and outcome.
Dana-Farber Cancer Institute ALL Consortium Protocol 00-01 focused on optimizing the use of corticosteroids and EC-Asnase in children and adolescents with newly diagnosed ALL. The primary objectives of this study were to determine the relative toxicity, tolerability, and efficacy of 1) dexamethasone and prednisone administered during postinduction treatment and 2) weekly intramuscular EC-Asnase administered as the standard FD and a pharmacokinetically guided individualized dose (ID). In this article, we report the findings of Protocol 00-01.
PATIENTS AND METHODS
Patients
Patients ages 1 to 18 years with newly diagnosed ALL (excluding mature B-cell ALL) were eligible for enrollment. The protocol was approved by the institutional review boards of each participating institution (Appendix Table A1, online only).16 Informed consent was obtained from parents or guardians for each patient before enrollment.
Risk Groups
Patients were stratified into risk groups for treatment. Standard risk was defined as: patients ages 1 to 9.99 years, presenting WBC count less than 50,000/μL, B-precursor phenotype, no mediastinal mass, and diagnostic spinal fluid without blasts (CNS1) or with blasts but fewer than five leukocytes per high-power field. All other patients were defined as high risk. Patients with Philadelphia chromosome (Ph+) were considered to be high risk but were eligible for allogeneic hematopoietic stem-cell transplantation in first complete remission (CR). Patients with MLL gene rearrangement received an additional postinduction intensification cycle and then continued as high risk.17
Therapy
Details of treatment are summarized in Table 1. All patients received prednisone during induction. Patients with persistent morphologic leukemia after 1 month were removed from study and received alternative therapy.18 Corticosteroids were permanently discontinued in the setting of symptomatic, radiographically confirmed osteonecrosis and were temporarily held for bone fracture (resumed when the fracture healed). EC-Asnase was held until resolution of mild/moderate pancreatitis or thrombosis and was permanently discontinued after severe pancreatitis. Patients with allergic reactions to EC-Asnase of any severity, including local reactions, were switched to Erwinia asparaginase (25,000 IU/m2) two times per week.16 Patients were switched to weekly intramuscular polyethylene glycol–asparaginase on allergy to or unavailability of Erwinia asparaginase. Asparaginase was permanently discontinued after allergy to all available preparations. Therapy for all patients was discontinued after 24 months of continuous CR.
Table 1.
Therapy on DFCI-ALL Consortium Protocol 00-01
Phase/Duration | Treatment |
---|---|
Induction | VCR 1.5 mg/m2 once per week (maximum 2 mg), days 0, 7, 14, 21 |
4 weeks | Prednisone 40 mg/m2 per day, days 0 to 28 |
DOX 30 mg/m2/dose, days 0 and 1; (HR patients: with dexrazoxane 300 mg/m2/dose) | |
MTX 4 mg/m2 (8-24 hours after doxorubicin) with leucovorin rescue | |
EC-Asnase 25,000 IU/m2 IM × one dose | |
IT cytarabine, day 0,*† IT MTX/cytarabine/hydrocortisone × one dose, day 14; IT on day 28 (counting as first IT of CNS phase) | |
CNS therapy | VCR 2 mg/m2 (maximum 2 mg) day 1; 6-MP 50 mg/m2/d orally at bedtime, × 14 days |
3 weeks | SR patients: IT MTX/cytarabine/hydrocortisone two times per week × four doses |
HR patients: IT MTX/cytarabine two times per week × four doses; also DOX 30 mg/m2 day 1, with dexrazoxane 300 mg/m2; cranial irradiation‡ | |
Intensification | Every 3 week cycles: |
30 weeks | VCR 2 mg/m2 (maximum 2 mg), day 1; 6-MP 50 mg/m2 per day orally at bedtime × 14 doses; MTX 30 mg/m2 (1 mg/kg if < 0.6 m2) IV or IM once per week |
Corticosteroid, randomized: | |
Dexamethasone 6 mg/m2 per day, divided two times per day (3 mg/m2 per dose), days 1-5 of each 3-week cycle, or | |
Prednisone 40 mg/m2 per day divided two times per day (20 mg/m2 per dose), days 1-5 of each 3-week cycle | |
EC-Asnase randomized: | |
Fixed-dosing: 25,000 IU/m2 IM, once per week × 30 weeks, or | |
Individualized dosing: 12,500 IU/m2 IM (starting dose)§, once per week × 30 weeks |
NSAA (IU/mL) | Change in subsequent doses |
< 0.025 | Increase by 80%, send urgent EC-Asnase antibody |
0.025 to < 0.05 | Increase by 60% |
0.05 to < 0.08 | Increase by 40% |
0.08 to < 0.1 | Increase by 20% |
0.1 to < 0.14 | No change |
0.14 to < 0.20 | Decrease by 20% |
> 0.20 | Decrease by 40% |
IT MTX/cytarabine/hydrocortisone at start of a cycle every 9 weeks × 6 doses, then every 18 weeks through completion of therapy (at start of a cycle) | |
HR patients: same as above, except higher corticosteroid dose (prednisone 120 mg/m2 per dose, days 1-5, or dexamethasone 18 mg/m2 per dose, days 1-5), DOX 30 mg/m2 day 1 of each cycle (cumulative dose 300 mg/m2) with dexrazoxane 300 mg/m2/dose, no weekly MTX IV/IM until doxorubicin completed, and IT therapy of MTX/cytarabine every 18 weeks | |
Continuation | Every 3 week cycles: |
74 weeks | SR: same as intensification except no asparaginase |
HR: same as SR patients, including lower corticosteroid dose of dexamethasone 6 mg/m2 per day and prednisone 40 mg/m2 per day, days 1-5 |
Abbreviations: 6-MP, 6-mercaptopurine; ALL, acute lymphoblastic leukemia; bid, twice per day; DFCI, Dana-Farber Cancer Institute; DOX, doxorubicin; EC-Asnase, E coli asparaginase; HR, high risk; IM, intramuscular; IT, intrathecal; IV, intravenous; MTX, methotrexate; NSAA, nadir serum asparaginase activity; SR, standard risk; VCR, vincristine.
Randomizations
Patients who achieved CR were eligible for two randomized comparisons. Patients who declined participation in random assignments and Ph+ patients were directly assigned to receive prednisone and FD EC-Asnase.
Corticosteroid Randomization
Standard-risk patients were randomly assigned to receive either dexamethasone 6-mg/m2/day or prednisone 40 mg/m2/day, administered as 5-day pulses every 3 weeks beginning at week 7, until the completion of therapy. High-risk patients received either dexamethasone 18-mg/m2/day or prednisone 120 mg/m2/day during the 30-week intensification phase, and then the same corticosteroid dose as standard-risk patients during the continuation phase.8,19
Asparaginase Randomization
During intensification, patients were scheduled to receive 30 weekly doses of IM EC-Asnase. Patients were randomly assigned to receive either FD (25,000 IU/m2) or ID, based on monitoring NSAA. The initial dose for the ID arm was 12,500 IU/m2, with subsequent doses adjusted to maintain NSAA between 0.10 and 0.14 IU/mL using the algorithm in Table 1.
Samples for monitoring NSAA were obtained from all patients just before administering the second and fourth EC-Asnase doses and every 3 weeks thereafter. Procedures for sample collection, processing, storage, and determining NSAA have been previously described.16 Only samples obtained 7 days after the prior dose were considered evaluable for dose adjustment and statistical analysis. EC-Asnase antibodies were assayed as previously described in samples obtained every 6 weeks from all patients and when NSAA was below the level of detectability (< 0.025 IU/mL) in ID patients.20,21 ID patients with NSAA less than 0.1 IU/mL on successive determinations despite dose adjustment or when coupled with EC-Asnase antibody positivity were considered to have “silent inactivation” and were switched to Erwinia- or polyethylene glycol–asparaginase. Clinical allergic reaction was the only planned indication for switching asparaginase preparation for FD patients.
Toxicity Assessment
Toxicity data were prospectively collected. For the corticosteroid randomization, the primary toxicity end points were symptomatic osteonecrosis and fracture, confirmed by radiographic imaging. A diagnosis of bacteremia or fungemia was based on positive blood culture; invasive fungal disease was diagnosed by radiographic imaging, plus/minus culture results. Asparaginase-associated toxicities included allergy, thrombosis (radiographically confirmed), and pancreatitis. Pancreatitis cases were classified by duration of clinical signs/symptoms as mild/moderate (< 72 hours in duration) or severe (≥ 72 hours).
Minimal Residual Disease
End-induction marrow aspirates were collected prospectively for patients with B-precursor immunophenotype and were analyzed for minimal residual disease by a polymerase chain reaction–based methodology.22 Results were not used to adjust treatment.
Statistical Methods
Descriptive statistics are presented as frequencies and percentages. The 5-year cumulative incidence (CumInc) of osteonecrosis and fracture were estimated using the CumInc utility in the cmprsk package in R and tested using the Gray test, with relapse and death in remission identified as competing risks. Frequencies of infection and asparaginase toxicities were identified and analyzed with the Fisher's exact test. The study was designed to test for differences in the rate of asparaginase-related toxicity between FD and ID and of bony morbidity (osteonecrosis and fracture) between dexamethasone and prednisone. Four hundred three patients were calculated to provide 90% power to detect a difference between 36% and 21% in asparaginase-related toxicity using a two-sided .05 level test and 80% power to detect a difference between 16% and 28% incidence of bony morbidity using a two-sided .05 level test.
For overall protocol results, EFS was defined as the time from diagnosis to first outcome event (induction failure, induction death, death during remission, or relapse). Overall survival (OS) was calculated from the time of diagnosis to death from any cause. EFS and OS were estimated using Kaplan-Meier methodology, and the Greenwood formula was used to construct 95% CI, which were truncated at 100%. The log-rank test was used to compare EFS and OS between groups. Multivariable regression analysis was conducted using Cox proportional hazards models (Appendix, online only).
Analyses of both randomizations were by intention-to-treat for all patients achieving CR. Because the analysis of patients participating in the randomizations included only those alive and in CR at the start of the intensification phase, a landmark approach defined at this time-point was used for EFS estimates. EFS and OS were planned secondary end points of the randomized comparisons.
No adjustments were made for performing multiple statistical tests. P values are two-sided and values less than .05 were considered significant.
RESULTS
Patient Characteristics
Between September 2000 and December 2004, 498 patients (ages 1 to 18 years) were enrolled, 492 of whom were considered evaluable (five were subsequently found ineligible and one withdrew consent; Fig 1; Table 2). Median follow-up was 4.9 years.
Dana-Farber Cancer Institute Acute Lymphoblastic Leukemia (ALL) Consortium Protocol 00-01 flow diagram. Patients with newly diagnosed ALL were enrolled (N = 498); 492 patients were considered evaluable. Of these 492 patients, 473 patients achieved complete remission (CR). Of the 473 patients who achieved CR, 408 patients (86%) participated in the corticosteroid random assignment and 384 (81%) in the asparaginase random assignment. Ph, Philadelphia chromosome. (*) Includes 13 patients with persistent leukemia at end of first month and three patients who did not recover blood counts by day 49.
Table 2.
Patient Characteristics and 5-Year EFS Survival Rates of Children and Adolescents With Newly Diagnosed ALL Treated on DFCI ALL Consortium Protocol 00-01 (N = 492)
Characteristic | Total No. of Patients | 5-Year EFS (%) | 95% CI | P |
---|---|---|---|---|
All evaluable patients | 492 | 80 | 76 to 84 | |
DFCI risk group | .001 | |||
Standard risk | 282 | 85 | 80 to 89 | |
High risk | 210 | 74 | 68 to 80 | |
Age, years | .008 | |||
1-9.99 | 380 | 83 | 79 to 87 | |
10-18 | 112 | 72 | 63 to 80 | |
Median | 4.75 | |||
Range | 1-17.8 | |||
WBC count, ×109/L | .007 | |||
< 50,000 | 391 | 83 | 78 to 87 | |
≥ 50,000 | 101 | 72 | 63 to 81 | |
Median | 11.4 | |||
Range | 0.9-865.1 | |||
Sex | .04 | |||
Male | 264 | 77 | 71 to 82 | |
Female | 228 | 85 | 80 to 90 | |
Immunophenotype | .01 | |||
B lineage | 443 | 82 | 78 to 85 | |
T cell | 49 | 69 | 56 to 82 | |
CNS status at diagnosis | .68 | |||
CNS 1 | 410 | 80 | 76 to 84 | |
CNS 2 | 60 | 84 | 74 to 93 | |
CNS 3 | 17 | 81 | 62 to 100 | |
Missing | 5 | |||
Down syndrome | .34 | |||
Yes | 13 | 92 | 76 to 100 | |
No | 479 | 80 | 76 to 84 | |
End-induction MRD* | 186 | < .0001 | ||
Low | 168 | 87 | 81 to 92 | |
High | 18 | 61 | 35 to 79 | |
Cytogenetics† | 405 | |||
TEL/AML1 | 98 | 91 | 85 to 97 | |
High hyperdiploid | 110 | 88 | 81 to 95 | |
t(1;19) | 14 | 100 | ||
t(9;22) | 14 | 31 | 6 to 56 | |
MLL rearrangement | 14 | 48 | 21 to 75 |
NOTE. Bone marrow cells from diagnostic aspirate samples were examined for cell-surface antigens using direct immunofluorescence assays and were cultured for cytogenetic analyses. Screening for the following translocations by FISH and/or PCR was performed at each participating institution when diagnostic marrow sample was available: TEL/AML1, MLL gene rearrangements, and BCR-ABL.
Abbreviations: ALL, acute lymphoblastic leukemia; DFCI, Dana-Farber Cancer Institute; EFS, event-free survival; FISH, fluorescent in situ hybridization; MRD, minimal residual disease; PCR, polymerase chain reaction.
Overall Results
Four hundred seventy-three patients (96%) achieved CR, 13 patients (3%) had persistent leukemia at the end of the first month, and three patients (0.6%) died during induction (two patients from infection/sepsis and one from necrotizing colitis). Three patients (0.6%) failed to recover blood counts by day 49 of treatment and were removed from protocol. Of those who achieved CR, 69 patients (15%) relapsed and five patients (1%) died while in first CR as a result of infection/sepsis (n = 3) and complications of hematopoietic stem-cell transplantation (two Ph+ patients). No second malignant neoplasms have been reported. The estimated 5-year EFS and OS for all 492 patients were 80% (95% CI, 76% to 84%) and 91% (95% CI, 88% to 93%), respectively (Fig 2A).
(A) Event-free survival (EFS) and overall survival (OS) for all patients. With a median follow-up of 4.9 years, the 5-year EFS for 492 evaluable patients was 80% (95% CI, 76% to 84%) and the 5-year OS was 91% (95% CI, 88% to 93%). (B) EFS results of corticosteroid randomization. The 5-year EFS for patients randomly assigned to prednisone was 81% (95% CI, 75% to 87%) compared with 90% (95% CI, 85% to 94%) for those randomly assigned to dexamethasone (P = .01). (C) EFS results of asparaginase randomization. The 5-year EFS for patients randomly assigned to fixed-dosing of asparaginase was 82% (95% CI, 75% to 87%) compared with 90% (95% CI, 84% to 94%) for those randomly assigned to individualized dosing (P = .04).
Of the 473 patients who achieved CR, 408 patients (86%) participated in the corticosteroid randomization (prednisone, 207; dexamethasone, 201) and 384 patients (81%) participated in the asparaginase randomization (FD, 195; ID, 189). Presenting characteristics did not differ significantly between treatment arms.
Corticosteroid Randomization
Osteonecrosis.
Of the 408 randomly assigned patients, 23 patients (6%) had symptomatic osteonecrosis (six patients at more than one site). Osteonecrosis was more frequently diagnosed in patients 10 to 18 years old than in younger patients (5-year CumInc, 14% v 3.5%; P = .0004). Five-year osteonecrosis CumInc did not differ by sex (P = .15).
The overall incidence of osteonecrosis did not differ by corticosteroid type (P = .25; Appendix Table A2, online only). However, for patients 10 to 18 years, dexamethasone was associated with a higher risk of osteonecrosis (5-year CumInc, 23% v 5% for patients randomly assigned to prednisone; P = .02). Osteonecrosis CumInc did not differ by corticosteroid type in children 1 to 9.99 years old (P = .43). For high-risk patients ages 1 to 9.99 years randomly assigned to dexamethasone, osteonecrosis CumInc was 3%, similar to that of standard-risk patients and significantly less than older high-risk patients (5-year CumInc, 23%; P < .01). For high-risk patients randomly assigned to prednisone, there was no significant difference in osteonecrosis CumInc in those patients 1 to 9.99 years old (0%) versus patients 10 to 18 years old (5%; P = .92).
Fracture.
Fifty-six patients (14%) were diagnosed with a bone fracture (10 patients with more than one fracture). Five-year CumInc in patients ages 10 to 18 years was 19% compared with 12% for younger patients (P = .06). There was no difference in fracture CumInc by corticosteroid randomization (P = .89). For patients 10 to 18 years old, fracture CumInc for those patients randomly assigned to dexamethasone was 29% compared with 10% for those randomly assigned to prednisone (P = .06).
Infection.
Postinduction infections were documented in 60 patients (15%), including 12 (13%) of 91 patients ages 10 to 18 years and 48 (15%) of 317 patients ages 1 to 9.99 years (P = .74). The infection rate differed by corticosteroid type; 22 patients (11%) randomly assigned to prednisone had ≥ one infection compared with 38 patients (19%) randomly assigned to dexamethasone (P = .03; Appendix Table A2). The infection rate was higher for patients 10 to 18 years old in the dexamethasone arm compared with the prednisone arm (P = .004), but for patients 1 to 9.99 years old, infection rate did not significantly differ by corticosteroid type (P = .27). There was one infection-related death in a child randomly assigned to prednisone (from invasive fungal disease) and no infection-related deaths among patients on the dexamethasone arm.
EFS and OS.
The estimates of EFS and OS included only patients who were alive and in CR at the start of intensification. The 5-year EFS was 81% (95% CI, 75% to 87%) for patients on the prednisone arm and 90% (95% CI, 85% to 94%) for dexamethasone-arm patients (P = .01; Fig 2B). The trend toward superior EFS with dexamethasone was observed in subsets defined by age, risk group, phenotype, and other characteristics (Table 3). Sites of relapse are listed in Appendix Table A3 (online only). There was no significant difference in 5-year OS (prednisone arm, 94%; dexamethasone arm, 95%; P = .31).
Table 3.
Five-Year Event-Free Survival for the Randomly Assigned Study Arms
Patient Group | Corticosteroid Random Assignment | Asparaginase Random Assignment | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Dexamethasone | Prednisone | Fixed Dose | Individualized Dose | |||||||||
No. of Patients | 5-Year EFS (%) | 95% CI | No. of Patients | 5-Year EFS (%) | 95% CI | No. of Patients | 5-Year EFS (%) | 95% CI | No. of Patients | 5-Year EFS (%) | 95% CI | |
All evaluable patients | 201 | 90 | 85 to 94 | 207 | 81 | 75 to 87 | 195 | 82 | 75 to 87 | 189 | 90 | 84 to 94 |
Risk group | ||||||||||||
SR | 114 | 89 | 83 to 95 | 121 | 84 | 76 to 91 | 110 | 86 | 76 to 92 | 115 | 89 | 81 to 93 |
HR | 87 | 91 | 84 to 97 | 86 | 78 | 69 to 87 | 85 | 78 | 66 to 86 | 74 | 93 | 84 to 97 |
Age, years | ||||||||||||
1-9.99 | 153 | 91 | 86 to 95 | 164 | 82 | 76 to 89 | 151 | 85 | 77 to 90 | 147 | 90 | 83 to 94 |
10-18 | 48 | 88 | 77 to 98 | 43 | 77 | 64 to 89 | 44 | 72 | 55 to 84 | 42 | 90 | 76 to 96 |
WBC count, ×109/L | ||||||||||||
< 50,000 | 159 | 91 | 86 to 96 | 163 | 84 | 77 to 90 | 152 | 84 | 76 to 89 | 155 | 89 | 83 to 94 |
≥ 50,000 | 42 | 87 | 76 to 98 | 44 | 73 | 59 to 87 | 43 | 77 | 61 to 87 | 34 | 93 | 75 to 98 |
Sex | ||||||||||||
Male | 100 | 87 | 79 to 94 | 89 | 78 | 69 to 86 | 112 | 77 | 67 to 84 | 90 | 87 | 77 to 92 |
Female | 101 | 93 | 88 to 99 | 118 | 86 | 79 to 94 | 83 | 90 | 81 to 95 | 99 | 93 | 86 to 97 |
Immunophenotype | ||||||||||||
T cell | 22 | 96 | 87 to 100 | 17 | 65 | 42 to 87 | 20 | 74 | 49 to 88 | 15 | 100 | |
B lineage | 179 | 89 | 84 to 94 | 190 | 83 | 77 to 94 | 175 | 83 | 76 to 89 | 174 | 89 | 83 to 93 |
Cytogenetics | ||||||||||||
TEL/AML1 | 43 | 100 | 40 | 85 | 73 to 97 | 41 | 88 | 77 to 99 | 45 | 95 | 89 to 100 | |
High hyperdiploid* | 44 | 95 | 87 to 100 | 52 | 83 | 71 to 95 | 47 | 88 | 77 to 98 | 38 | 89 | 76 to 100 |
Abbreviations: EFS, event-free survival; HR, high risk; SR, standard risk.
Asparaginase Randomization
NSAA and asparaginase antibodies.
Data pertaining to NSAA and asparaginase antibody determinations are summarized in Table 4. A total of 2,536 evaluable NSAA samples were obtained from 95% of the 384 patients randomly assigned to the asparaginase arm. In both arms, median NSAA at each time-point increased progressively during the initial 6 weeks of treatment and then reached steady-state (not shown). Overall median steady-state NSAA was significantly higher on the FD arm. The percentage of NSAA determinations within the 0.10 to 0.14 IU/mL target range was similar for both arms, but there were fewer samples with high NSAA (> 0.14 IU/mL) on the ID arm. The percentage of patients who tested positive for EC-Asnase antibodies was similar for the two arms (P = .24).
Table 4.
Asparaginase Toxicities, Activity, Antibody Positivity, and Relationship With EFS
Characteristic | Fixed Dose | Individualized Dose | P | ||
---|---|---|---|---|---|
No. of Patients | % | No. of Patients | % | ||
Patients randomly assigned | 195 | 189 | |||
EC-Asnase dose, IU/m2 | 25,000 | 17,500* | |||
Any toxicity | 63 | 32 | 59 | 31 | .83 |
EC-Asnase clinical allergy | 39 | 20 | 40 | 21 | .80 |
Pancreatitis | 10 | 5 | 12 | 6 | .66 |
Thrombosis | 16 | 8 | 13 | 7 | .70 |
Completed ≥ 25 week asparaginase | 172 | 88 | 164 | 86 | .76 |
Patients with evaluable NSAA data | 186 | 95 | 180 | 95 | |
No. of evaluable NSAA samples | 1,235 | 1,301 | |||
Overall steady-state NSAA, IU/mL | < .001 | ||||
Median | 0.11 | 0.08 | |||
Range | < 0.025-1.78 | < 0.025-0.74 | |||
Maximum NSAA in each patient | .05 | ||||
≥ 0.10 IU/mL | 150 | 81 | 129 | 72 | |
< 0.10 IU/mL | 36 | 19 | 51 | 28 | |
EC-Asnase antibody determinations | 180 | 181 | .24 | ||
Positive test | 64 | 36 | 76 | 42 | |
Asparaginase preparation changed from EC-Asnase | 43 | 22 | 64 | 34 | .01 |
EC-Asnase clinical allergy | 39 | 20 | 39 | 21 | |
Silent inactivation | N/A | — | 19 | 10 | |
Other | 4 | 2 | 6 | 3 | |
Overall 5-year EFS, % | .04 | ||||
EFS | 82 | 90 | |||
95% CI | 75 to 87 | 84 to 94 | |||
5-year EFS of other subsets, % | |||||
Maximum NSAA ≥ 0.10 IU/mL | |||||
EFS | 85 | 90 | |||
95% CI | 78 to 90 | 83 to 95 | |||
Maximum NSAA < 0.10 IU/mL | .16 | ||||
EFS | 73 | 91 | |||
95% CI | 52 to 86 | 79 to 97 | |||
Maximum NSAA < 0.10 IU/mL, did not change asparaginase preparation | .058 | ||||
EFS | 76 | 78 | |||
95% CI | 38 to 92 | 35 to 94 | |||
Changed asparaginase for silent inactivation | .99 | ||||
EFS | N/A | 95 | |||
95% CI | 68 to 99 |
Abbreviations: EC-Asnase, E coli asparaginase; EFS, event-free survival; N/A, not applicable; NSAA, nadir serum asparaginase activity.
Toxicity.
There was no difference in the frequency of asparaginase-associated toxicity between the two arms (Table 4) or in the proportion of patients completing ≥ 25 weeks of asparaginase (FD, 88%; ID, 87%).
Significantly more patients on the ID arm were switched from EC-Asnase to another asparaginase preparation (Table 4). The percentage of patients for whom the preparation was changed in response to clinical allergy was similar for the two arms (FD, 20%; ID, 21%). However, the preparation was switched for 19 ID patients (10%) because of silent inactivation. Seventeen of these patients (90%) had at least one NSAA ≥ 0.1 IU/mL after switching preparation. Conversely, 18 FD patients (9%) never had NSAA ≥ 0.1 IU/mL with EC-Asnase, never developed clinical allergies, and never changed preparations.
EFS and OS.
Twenty-nine patients (15%) on the FD arm experienced relapse compared with 17 ID patients (9%). Five-year EFS was 82% (95% CI, 75% to 87%) for FD and 90% (95% CI, 84% to 94%) for ID patients (P = .04; Fig 2C). The trend toward superior EFS with ID was observed in subsets defined by age, risk group, phenotype, and other characteristics (Table 3). There was no significant difference in 5-year OS by arm (FD, 93%; 95% CI, 88% to 96%; ID, 96%; 95% CI, 92% to 98%; P = .18).
For patients with at least one NSAA ≥ 0.1 IU/mL with EC-Asnase, 5-year EFS was 85% (95% CI, 78% to 90%) for the FD arm compared with 90% for the ID arm (95% CI, 83% to 95%; P = .16). For patients with persistently low NSAA with EC-Asnase (maximum NSAA < 0.1 IU/mL), 5-year EFS was 73% (n = 36; 95% CI, 52% to 86) on the FD arm and 91% on the ID arm (n = 51; 95% CI, 79% to 97%; P = .058). Five-year EFS for patients with persistently low EC-Asnase NSAA who never changed preparations was similar on both arms (FD: n = 18; 76%; 95% CI, 38% to 92%; ID: n = 12; 78%; 95% CI, 35% to 94%; P = .99). Five-year EFS for the 19 ID patients who switched to a different preparation for silent inactivation was 95% (95% CI, 68% to 99%).
We also explored the outcome of patients based on asparaginase antibody status. There was no significant difference in 5-year EFS in antibody-negative patients who never developed EC-Asnase clinical allergy (FD: 83%; 95% CI, 72% to 90%; ID: 88%; 95% CI, 78% to 92%; P = .38). For the 54 FD patients who developed antibody-positivity but never developed EC-Asnase clinical allergy, 5-year EFS was 84% (95% CI, 70% to 92%) compared with 93% (95% CI, 82% to 97%) for similar patients on the ID arm (n = 63; P = .12). Of note, zero (0%) of 54 antibody-positive/EC-Asnase allergy–negative patients on the FD arm changed preparations, compared with 22 (35%) of 63 of such patients on the ID arm, including all 19 patients who were switched for silent inactivation (Appendix).
Randomization Interactions
We investigated differences in CumInc of osteonecrosis and of fracture between the randomized asparaginase arms and found no statistically significant differences. No significant differences in asparaginase toxicities (clinical allergy, pancreatitis, or thrombosis) were found based on corticosteroid randomization.
A total of 361 patients participated in both randomizations. When proportional hazard regression modeling was restricted to these patients, dexamethasone and ID EC-Asnase were both independent predictors of favorable EFS, without indication of an interaction (dexamethasone: hazard ratio, 0.49; P = .02; ID: hazard ratio, 0.52; P = .04).
DISCUSSION
Our study demonstrates that postinduction dexamethasone and ID of L-asparaginase improve EFS in pediatric patients with newly diagnosed ALL.
A novel aspect of our trial was the investigation of ID of EC-Asnase.23 We demonstrate the feasibility of adjusting the dose of EC-Asnase based on measurement of NSAA in the context of a large multi-institutional study. ID was associated with superior EFS, but did not decrease asparaginase-associated toxicity, increase the proportion of patients able to complete the planned treatment course of asparaginase, or lead to higher (potentially more therapeutic) steady-state NSAA with EC-Asnase.
One possible explanation for this result was that silent inactivation was prospectively identified on the ID but not the FD arm. ID patients with silent inactivation switched asparaginase preparations, but on the FD arm preparation was changed only for clinical allergy. FD patients with low EC-Asnase NSAA (maximum < 0.1 IU/mL) who never switched preparations had a 5-year EFS of 76% compared with 95% for the ID patients who switched preparations for silent inactivation. We and others have previously demonstrated that patients with EC-Asnase hypersensitivity and antibody-positivity achieved therapeutic NSAA after switching preparations.16,24 Furthermore, we demonstrate that patients with silent inactivation also achieve therapeutic NSAA with change in preparation (17 of 19 ID patients with silent inactivation had at least one therapeutic NSAA after switching preparations). Our results suggest that prospectively monitoring for the development of silent inactivation (and not just clinical allergy) and changing asparaginase preparation may improve outcome. In our trial that is currently enrolling, we are monitoring NSAA in real time and switching asparaginase preparation in patients with presumed silent inactivation (two consecutive nondetectable NSAA measurements).
Our results also support the use of dexamethasone instead of prednisone during postinduction treatment phases. There is controversy regarding whether dexamethasone is superior to prednisone. Two trials showed dexamethasone superiority3,4 whereas others demonstrated no differences in outcome by corticosteroid type.5,6 We demonstrate that dexamethasone was associated with better EFS in all examined patient subsets.
The beneficial impact of dexamethasone in reducing relapse risk comes at a cost of increased toxicity, primarily in older children and adolescents. Osteonecrosis can lead to severe pain, joint damage, and articular collapse and can necessitate invasive procedures.7,25–31 Dexamethasone was associated with an increased risk of osteonecrosis and fracture only in patients ages 10 to 18 years. Neither corticosteroid type nor cumulative dose of corticosteroid affected the rate of osteonecrosis in younger patients (ages 1 to 9.99 years). We and others have reported that dexamethasone was associated with higher rates of infection and death when given during induction in conjunction with an anthracycline.32,33 In this article, we demonstrate that dexamethasone was associated with a higher risk of infection during postinduction treatment, particularly in older children and adolescents, but that risk did not result in a higher toxic death rate.
A limitation of this study is that detailed subgroup analyses of EFS and OS were limited by sample size. In addition, analyses of outcome based on NSAA and asparaginase antibody status were exploratory.
In summary, there was no overall difference in skeletal toxicity by corticosteroid type. Dexamethasone was associated with a higher infection rate and, in older children, increased incidence of osteonecrosis and fracture. In addition, we found no difference in asparaginase-related toxicity by EC-Asnase dosing method. Dexamethasone and pharmacokinetically guided individualized dosing of asparaginase were each associated with favorable EFS. Importantly, dexamethasone was associated with a better outcome in younger patients (1 to 10 years old) without any increased risk in skeletal toxicity. Considering its beneficial impact in reducing relapse risk, we continue to use dexamethasone in postinduction therapy for all age groups. Future studies should focus on minimizing the toxicity of dexamethasone in older pediatric patients without compromising efficacy. Our results suggest that pharmacokinetic monitoring during treatment with EC-Asnase to identify patients with silent inactivation may be an effective strategy to improve the outcome of children and adolescents with ALL.
Acknowledgment
Presented in part at the 51st American Society of Hematology Annual Meeting and Exposition, December 5–8, 2009, New Orleans, LA. We thank the patients, families, physicians, nurses, research coordinators, and all others who participated in this trial.We acknowledge the important contributions of Virginia Dalton, PNP, MSN, Annette Dalton, Jessica Comras, Laura Belden, and Emily Swaim. We acknowledge the important contributions of Nathan Kaplan, BS, and Anita Y. Bahar, MS, in the assessment of end-induction minimal residual disease.
Appendix
Additional details regarding statistical methods.
Multivariable models utilized included indicators for both randomizations and an interaction term. In addition, stepwise models were explored including risk group, sex, T-cell disease, testicular involvement, presence of mediastinal mass, Down syndrome, age 10 or older, age 15 or older, WBC 50,000/μL or higher, white race, and Hispanic ethnicity.
Additional details regarding the results of the asparaginase randomization.
Of the 93 patients on the fixed-dose (FD) arm and the 82 patients on the individualized-dose (ID) arm who had both an antibody determination and a nadir serum asparaginase activity (NSAA) sample and did not develop a clinical allergy to Escherichia coli L-asparaginase, had no silent allergy (antibody negative), and did not switch preparations for other reasons, 81 patients (90%) on the FD arm and 72 patients (90%) on the ID arm achieved an NSAA ≥ 0.10. No difference in event-free survival was detected between these groups (P = .27). The 5-year event-free survival estimates were 84% for the FD arm versus 89% for the ID arm.
Of the 19 patients on the ID arm who switched asparaginase preparation because of silent allergy, all had a positive antibody measurement and seven achieved a maximum NSAA less than 0.025. Of the 236 patients on either arm who had adequate NSAA levels (≥ 0.10) and did not have their asparaginase preparation switched, 82 patients (35%) had at least one positive antibody determination.
An exploration of the impact of the presence of positive asparaginase antibodies on the rate of osteonecrosis and of fracture did not find a statistically significant association with the presence of asparaginase antibodies and these outcomes.
Table A1.
DFCI ALL Consortium Protocol 00-01 Participating Institutions
Institution | Location |
---|---|
Dana-Farber Cancer Institute/Boston Children's Hospital | Boston, MA |
Columbia University Medical Center/Morgan Stanley Children's Hospital of New York-Presbyterian | New York, NY |
Hospital Sainte Justine | Montreal, Quebec, Canada |
Le Centre Hospitalier de L'Universite, Laval | Quebec City, Quebec, Canada |
Maine Children's Cancer Program | Scarborough, ME |
McMaster Children's Hospital | Hamilton, Ontario, Canada |
San Jorge Children's Hospital | San Juan, Puerto Rico |
Tulane Hospital for Children | New Orleans, LA |
University of Rochester Medical Center | Rochester, NY |
Abbreviations: ALL, acute lymphoblastic leukemia; DFCI, Dana-Farber Cancer Institute.
Table A2.
Skeletal Toxicity and Infections in the Dexamethasone and Prednisone Study Arms (N = 408)
Dexamethasone (%) | Prednisone (%) | P | |
---|---|---|---|
5-year cumulative incidence of osteonecrosis | |||
All randomly assigned patients | 7 | 4 | .25 |
1-9.99 years old | 3 | 4 | .43 |
10-18 years old | 23 | 5 | .02 |
SR < 10 years old | 3 | 6 | .23 |
HR < 10 years old | 3 | 0 | .29 |
Male | 4 | 4 | .92 |
Female | 10 | 5 | .16 |
5-year cumulative incidence of bone fracture | |||
All randomly assigned patients | 14 | 14 | .89 |
1-9.99 years old | 10 | 9 | .25 |
10-18 years old | 29 | 10 | .06 |
% of patients with infection | |||
All randomly assigned patients | 19 | 11 | .03 |
1-9.99 years old | 18 | 13 | .27 |
10-18 years old | 23 | 2 | .004 |
NOTE. Sites of osteonecrosis were as follows: hip (n = 8), knee (n = 6), shoulder (n = 2), ankle or foot (n = 13), and hand or wrist (n = 2). Site of fracture were as follows: the leg (n = 20), arm (n = 15), foot/toe (n = 21), hand/finger (n = 9), vertebrae (n = 4), pelvis (n = 2), and other sites (n = 3). Osteonecrosis was most often first diagnosed during the continuation phase of treatment (74%), with 9% first diagnosed during intensification, and 17% after the completion of treatment. Fractures were most often first diagnosed in the continuation phase (61%), compared with 14% occurring during intensification, and 25% after the completion of treatment. In an exploratory analysis, development of osteonecrosis or fracture did not appear to be associated with an increased risk of relapse. Infections consisted of 95 episodes of bacteremia, four of fungemia, and three of invasive fungal disease (one case of invasive fungal disease on the prednisone arm, two on the dexamethasone arm).
Abbreviations: HR, high risk; SR, standard risk.
Table A3.
Sites of Relapse for Dexamethasone and Prednisone Randomized Groups
Outcome | Dexamethasone (n = 201) | Prednisone (n = 207) | ||
---|---|---|---|---|
No. of Patients | % | No. of Patients | % | |
Total No. of relapses | 18 | 9 | 34 | 16 |
BM only | 11 | 18 | ||
CNS only | 2 | 4 | ||
BM and CNS | 3 | 8 | ||
Testicular only | 1 | 1 | ||
BM and testicular | 0 | 1 | ||
BM and other | 1 | 1 | ||
Other only | 0 | 1 | ||
Remission deaths | 0 | 2 | ||
Continuous remission | 183 | 171 | ||
5-year cumulative incidence | ||||
Any relapse | 10.1* | 17.7* | ||
Isolated CNS relapse | 1.3 | 2.1 | ||
Isolated BM relapse | 6.2 | 9.9 | ||
BM and CNS relapse | 1.6 | 3.6 |
Abbreviation: BM, bone marrow.
Footnotes
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
Clinical trial information: NCT00165178.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment or Leadership Position: None Consultant or Advisory Role: Lewis B. Silverman, Enzon Pharmaceuticals (C), EUSA Pharma (C) Stock Ownership: None Honoraria: None Research Funding: Jeffrey G. Supko, EUSA Pharma; Mary V. Relling, Sigma-Tau Pharmaceuticals, Enzon Pharmaceuticals; Stephen E. Sallan, Enzon Pharmaceuticals, EUSA Pharma Expert Testimony: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Jeffrey G. Supko, Barbara L. Asselin, Uma H. Athale, Luis A. Clavell, Marshall Schorin, Harvey J. Cohen, Stephen E. Sallan, Lewis B. Silverman
Provision of study materials or patients: Barbara L. Asselin, Uma H. Athale, Luis A. Clavell, Kara M. Kelly, Marshall Schorin, Lewis B. Silverman
Collection and assembly of data: Lynda M. Vrooman, Jeffrey G. Supko, Jane O'Brien, Uma H. Athale, Kara M. Kelly, Jeffery L. Kutok, Bruno Michon, Mary V. Relling, Stephen E. Sallan, Lewis B. Silverman
Data analysis and interpretation: Lynda M. Vrooman, Kristen E. Stevenson, Jeffrey G. Supko, Suzanne E. Dahlberg, Uma H. Athale, Kara M. Kelly, Jeffery L. Kutok, Caroline Laverdière, Steven E. Lipshultz, Marshall Schorin, Harvey J. Cohen, Donna S. Neuberg, Stephen E. Sallan, Lewis B. Silverman
Manuscript writing: All authors
Final approval of manuscript: All authors
Support
Supported in part by Grants No. 5 P01CA068484 and P30 21765 from the National Cancer Institute, by the American Lebanese Syrian Associated Charities, and by the Linde Family Foundation.