High-risk childhood acute lymphoblastic leukemia in first remission treated with novel intensive chemotherapy and allogeneic transplantation

Leukemia (2013) 27, 1497 1503 & 2013 Macmillan Publishers Limited All rights reserved 0887-6924/13 www.nature.com/leu ORIGINAL ARTICLE High-risk childhood acute lymphoblastic leukemia in first remission treated with novel intensive chemotherapy and allogeneic transplantation GM Marshall 1,2,15, L Dalla Pozza 3,15, R Sutton 1,ANg 3, HA de Groot-Kruseman 4, VH van der Velden 5, NC Venn 1, H van den Berg 6, ESJM de Bont 7, R Maarten Egeler 8, PM Hoogerbrugge 9, GJL Kaspers 10, MB Bierings 11, E van der Schoot 6, J van Dongen 5, T Law 1, S Cross 3, H Mueller 2, V de Haas 4, M Haber 1,TRévész 12, F Alvaro 13, R Suppiah 14, MD Norris 1 and R Pieters 5 Children with acute lymphoblastic leukemia (ALL) and high minimal residual disease (MRD) levels after initial chemotherapy have a poor clinical outcome. In this prospective, single arm, Phase 2 trial, 111 Dutch and Australian children aged 1 18 years with newly diagnosed, t(9;22)-negative ALL, were identified among 1041 consecutively enrolled patients as high risk (HR) based on clinical features or high MRD. The HR cohort received the AIEOP-BFM (Associazione Italiana di Ematologia ed Oncologia Pediatrica (Italy) Berlin-Frankfurt-Münster ALL Study Group) 2000 ALL Protocol I, then three novel HR chemotherapy blocks, followed by allogeneic transplant or chemotherapy. Of the 111 HR patients, 91 began HR treatment blocks, while 79 completed the protocol. There were 3 remission failures, 12 relapses, 7 toxic deaths in remission and 10 patients who changed protocol due to toxicity or clinician/parent preference. For the 111 HR patients, 5-year event-free survival (EFS) was 66.8% (±5.5) and overall survival (OS) was 75.6% (±4.3). The 30 patients treated as HR solely on the basis of high MRD levels had a 5-year EFS of 63% (±9.4%). All patients experienced grade 3 or 4 toxicities during HR block therapy. Although cure rates were improved compared with previous studies, high treatment toxicity suggested that novel agents are needed to achieve further improvement. Leukemia (2013) 27, 1497 1503; doi:10.1038/leu.2013.44 Keywords: childhood; acute lymphoblastic leukemia; MRD testing; bone marrow transplant; drug toxicity INTRODUCTION Acute lymphoblastic leukemia (ALL) is the commonest childhood malignancy, with a cure rate 480%. 1,2 However, cure rates for the 10% of ALL patients identified as high risk (HR) using minimal residual disease (MRD) assays are only 20 50% and are equally poor after relapse. 3 9 Surrogate markers of in vivo treatment response during early treatment phases, such as MRD testing, are now the most powerful predictors of relapse. 6,7,10 12 In a study of the mechanism of relapse in ALL patients, we have previously shown that the relapsing clone is present at diagnosis as a minor population among most relapsing patients. 13 Moreover, an earlier Australian and New Zealand Children s Haematology and Oncology Group (ANZCHOG) trial, randomising ALL patients at 12 months from diagnosis to more intensive therapy, failed to reduce subsequent relapse risk. 4 These observations suggested that stratifying HR patients to different therapy early in first complete remission (CR1) may be a more effective strategy to prevent relapse. Strategies used to reduce relapse in HR patients have involved new agents, 14 intensification of existing chemotherapy and allogeneic stem cell transplantation (SCT) in CR1. 15 17 We developed a novel treatment regimen using existing cytotoxic agents to assess the effect of sequential courses of highly myelo-suppressive chemotherapy on clinical outcome and residual leukemic burden in patients with HR ALL, followed by SCT or further chemotherapy. Features identifying HR status were identical to those used in the recently published AIEOP-BFM (Associazione Italiana di Ematologia ed Oncologia Pediatrica (Italy) Berlin- Frankfurt-Münster ALL Study Group) ALL 2000 protocol. 6,7 Each of our three HR treatment blocks was designed around chemotherapy regimens of proven value in relapsed or adult ALL: etoposide and cyclophosphamide; 18,19 mitoxantrone and cytarabine; 20,21 and, fludarabine, idarubicin and cytarabine. 22,23 PATIENTS AND METHODS Patients A total of 1041 children aged 1 18 years with newly diagnosed, t(9:22) negative, ALL were prospectively enrolled: 504 patients from 1 October 2002 to 31 December 2009 on the ANZCHOG ALL Study 8 and 537 patients from 1 October 2004 to 31 December 2009 on the Dutch Childhood 1 Children s Cancer Institute Australia for Medical Research, Lowy Cancer Research Centre, UNSW, Sydney, New South Wales, Australia; 2 Centre for Children s Cancer and Blood Disorders, Sydney Children s Hospital, Randwick, Sydney, New South Wales, Australia; 3 Children s Hospital at Westmead, Sydney, New South Wales, Australia; 4 The Dutch Childhood Oncology Group (DCOG) for the Dutch Childhood Leukemia Study Group and the Australian and New Zealand Children s Haematology and Oncology Group (ANZCHOG), The Hague, The Netherlands; 5 Erasmus Medical Centre, Rotterdam, The Netherlands; 6 Emma Children Hospital AMC, University of Amsterdam, The Netherlands; 7 Beatrix Children s Hospital, University of Groningen, Groningen, The Netherlands; 8 Leiden University Medical Center, Leiden, The Netherlands; 9 Department of Human Genetics, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands; 10 VU University Medical Center, Amsterdam, The Netherlands; 11 Department of Hemato-oncology, Utrecht University Children s Hospital, Utrecht, The Netherlands; 12 Women s and Children s Hospital, Adelaide, South Australia, Australia; 13 John Hunter Children s Hospital, Newcastle, New South Wales, Australia and 14 Mater Children s Hospital, Brisbane, Queensland, Australia. Correspondence: Professor GM Marshall, Centre for Children s Cancer and Blood Disorders, Sydney Children s Hospital, High Street, Randwick, Sydney 2031, New South Wales, Australia. E-mail: g.marshall@unsw.edu.au 15 These authors contributed equally to this work. Received 24 November 2012; revised 3 February 2013; accepted 7 February 2013; accepted article preview online 14 February 2013; advance online publication, 8 March 2013

1498 Oncology Group (DCOG) ALL10 protocol. They were treated at 14 child cancer centres in Australia, New Zealand and the Netherlands. A total of 111 (10.7%) patients (48 Australian and 63 Dutch patients) were eligible for enrolment. This analysis is based on a snapshot of the data taken on 31 December 2010. Ethics approval was obtained from institutional ethics committees and written informed consent was obtained from patients, parents or legal guardians. The study data were reviewed annually by a data safety monitoring committee. HR features Criteria for HR stratification were identical to those used in the AEIOP-BFM ALL 2000 study. 6,7 Patients were identified as HR if the peripheral blood blast count response after day 8 of oral prednisone alone was 41.0 10 9 /l (poor prednisone response (PPR)), the MRD level in the bone marrow aspirate at day 79 after initial diagnosis was 45 10 4, t(4;11) was present in the diagnosis marrow or there was an M3 marrow (425% blasts) at day 33. Treatment All patients received Pre-phase with prednisone alone plus one intrathecal dose of, then Protocol IA and IB according to AEIOP-BFM ALL 2000. 6,7 HR patients then received three novel 7-week blocks of intensive chemotherapy (Table 1). Thereafter, HR patients with a suitable donor (n ¼ 57) were then eligible for an allogeneic SCT in CR1 if they had one of the following features: an MRD level 45 10 4 in the bone marrow aspirate at day 79 after diagnosis, no CR at day 33, PPR and either M3 marrow at day 15 or T-cell phenotype or white blood cell count (WBC) 4100 10 9 /l (high WBC (HWBC)). HR patients without these criteria or who were lacking a suitable donor (n ¼ 22) then repeated the three HR chemotherapy blocks, followed by AEIOP-BFM ALL 2000 Protocol II and oral maintenance chemotherapy with (20 mg/m 2 ) and 6-mercaptopurine (50 mg/m 2 ) for a total treatment duration of 2 years from diagnosis. The HR protocol was amended in April 2004 because of excessive toxicity in patients not receiving SCT. Thereafter, the myelosuppressive chemotherapy doses for HR blocks 5 (mitoxantrone and cytarabine) and 6 (fludarabine and cytarabine) were reduced by 50%. Thus, patients (n ¼ 12) enrolled between October 2002 and April 2004 received twice the doses listed in Table 1 of mitoxantrone, cytarabine, fludarabine and cytarabine. Prophylactic cranial irradiation at a dose of 12 Gy was given to HR patients receiving chemotherapy alone during Protocol II but only if 43 years of age for Dutch patients. Those patients receiving SCT had cranial irradiation as a component of the total body irradiation (TBI) given during conditioning. Supportive care All patients received prophylactic oral co-trimoxazole and antifungal therapy during HR therapy. Dutch patients also received prophylactic ciprofloxacin. Granulocyte colony-stimulating factor was administered to all patients for 10 14 days from 24 h after the completion of each HR block. As a precaution, prophylactic peripheral blood stem cell collection and storage was recommended at the end of HR block 1. Allogeneic stem cell transplant procedures A total of 57 patients underwent SCT after three HR blocks. SCT procedures were not mandated as part of the protocol. From the ANZCHOG cohort, 22 patients underwent SCT. Donor sources were: matched sibling (10), unrelated umbilical cord blood (9), matched unrelated adult (2), or a mismatched family donor (1). All patients were conditioned with a regimen utilising TBI at a dose of 12 Gy in 6 fractions over 3 days. The most commonly used conditioning regimen was TBI, cyclophosphamide and thiotepa (13), followed by TBI, cyclophosphamide and fludarabine (5). Both matched unrelated adult donor stem cell sources were CD-34 selected. Three of the SCTs were performed using allogeneic peripheral blood as the stem cell source. Among the DCOG cohort, 35 of 45 patients who completed HR3 underwent SCT according to protocol DCOG ALL10. As stratification in ALL10 was based on the BFM-2000 study, the guidelines for SCT also follow the ALL SZT-BFM 2003 protocol. Dutch patients were conditioned with TBI, cyclophosphamide and etoposide. Donor sources were: matched sibling (16), matched unrelated donor (14), mismatched unrelated donor (4), or a mismatched family donor (1). Stem cell sources were: bone marrow (25), umbilical cord blood (6), or peripheral blood (4). Table 1. Chemotherapy protocol a for HR ALL patients HR block 1 (days 1 49) Triple intrathecal b Age-related Day 1 6-Mercaptopurine 50 mg/day Days 1 14 Etoposide 350 mg/day Days 15 17 Cyclophosphamide 1200 mg/day Days 15 17 Vincristine 1.5 mg/dose Days 22 and 29 L-Asparaginase 10 000 U/ dose Days 22, 25, 29 and 32 HR block 2 (days 49 98) Cytarabine 1500 mg/day Days 15 19 Mitoxantrone 5.25 mg/day Days 15 19 Vincristine 1.5 mg/dose Days 22 and 29 L-Asparaginase 10 000 U/ dose Days 22, 25, 29 and 32 HR block 3 (days 98 147) Idarubin 6 mg/day Days 15 17 Cytarabine 1500 mg/day Days 15 19 Fludarabine 22.5 mg/day Days 15 19 HR block 4 (days 147 196) Etoposide 275 mg/day Days 15 17 Cyclophosphamide 900 mg/day Days 15 17 Vincristine 1.5 mg/dose Days 22 and 29 L-Asparaginase 10 000 U/ dose Days 22, 25, 29 and 32 HR Block 5 (days 196 246) Same as HR block 2 HR block 6 (days 246 295) Cytarabine 1500 mg/day Days 15 19 Fludarabine 22.5 mg/day Days 15 19 Abbreviations: ALL, acute lymphoblastic leukemia; HR, high risk. a All doses were calculated as per m 2. b Triple intrathecal medication consisted of age-related doses of cytarabine, hydrocortisone and. MRD assessments MRD was measured in bone marrow samples by real-time quantitative PCR for immunoglobulin and T-cell receptor gene rearrangements, as previously described. 24 26 The data were interpreted using EuroMRD guidelines, 27 with quality control by the EuroMRD group. At least one MRD test with quantitative range of 5 10 4 5 10 5, and sensitivity of 1 10 4 1 10 5, was achieved for 90 of the 91 patients. We designed individual probes in four cases. Statistical analysis The primary aim of the study was to prospectively evaluate the clinical and molecular outcomes of HR ALL patients in first remission treated with HR block therapy and either SCT or chemotherapy. A secondary aim was to assess MRD levels following each HR block. The event-free survival (EFS) for N ¼ 111 eligible HR patients was defined as the time from diagnosis to first event (an event was defined as remission failure, relapse, withdrawal of consent, secondary malignancy or death from any cause) or to the date of Leukemia (2013) 1497 1503 & 2013 Macmillan Publishers Limited

last follow-up. The EFS for N ¼ 91 patients who received three HR blocks was defined as the time from diagnosis to first event or withdrawal from study or to date of last follow-up. Overall survival (OS) was defined as the time from diagnosis to the point of death from any cause. For EFS and OS, the primary analyses were Kaplan Meier and log-rank tests. Multivariate Cox regression was used to assess the prognostic significance of HR features. Adverse events were graded according to the NCI Common Terminology Criteria for Adverse Events (CTCAE). 28 Safety was summarised for all the patients by the number of toxic effects during the trial scored at CTCAE grade 3 or higher and the proportion of patients having a serious adverse event. Each toxic effect was placed into one of the seven toxicity groups: haematological, infectious, gastro-intestinal, skin, renal, cardiac and neurological. Because a patient may have experienced both grade 3 and 4 effects from the same toxicity group during the same HR block, only the highest grade of toxicity was reported for each patient. Some of the clinical trial costs were funded by grants from Cancer Council New South Wales and Cancer Institute New South Wales, while some of the MRD assay costs were funded by the Australian National Health and Medical Research Council (NHMRC). These funding bodies had no role in study design, data collection, data analysis, data interpretation or writing of this report. The authors had full access to the data and final responsibility for the decision to submit for publication. The study was registered with the NHMRC s Australian Clinical Trials Registry and had a reference number of ACTRN12607000302459. RESULTS HR patient characteristics A total of 111 children with newly diagnosed t(9;22)-negative ALL, aged 1 18 years, were consecutively identified using clinical or MRD criteria as HR by the end of Protocol I and were eligible to receive the HR chemotherapy (Table 1). However, 20 eligible 1499 patients did not begin the HR chemotherapy, or were not assessable, due to: death in induction (1) failure to remit (3), treatment protocol changed due to toxicity (3) or clinician/ parent preference (5), and missing toxicity data (8) (Figure 1). The remaining 91 patients commenced HR chemotherapy blocks. Clinical features that were more frequent among the HR cohort at diagnosis, compared with the entire ALL patient cohort, were male gender, age 410 years, T-cell and HWBC (Table 2). The majority of patients were stratified to HR therapy because of high MRD at day 79 (33%), PPR þ T-cell (±HWBC) (34%) or PPR alone (16%) (Table 3). Treatment Of the 91 patients who commenced HR block therapy, 87 completed the intended first 3 HR blocks. Four patients who commenced HR block therapy received only HR block 1, or HR blocks 1 and 2 alone, then switched therapy due to clinician/ parent preference (3) or prolonged myelosuppression and infection (1) (Figure 1). After completing all three HR blocks, 8 patients changed protocol due to clinician/parent preference (2), severe respiratory syncytial virus infection during SCT conditioning (1), central nervous system relapse (1), anaphylaxis to PEG- Asparaginase (1), intra-cerebral haemorrhage with thrombocytopenia (1), severe Escherichia coli sepsis (1) or prolonged myelosuppression and infection (1). Of the 79 patients completing protocol therapy, 57 underwent SCT and 22 continued chemotherapy alone. Thirteen of the 22 patients who received chemotherapy alone, who had been assigned to receive a SCT, did not receive SCT due to the absence 111 HR patients 3 remission failures 1 death in induction 8 with missing data 3 treated on a different protocol due to induction toxicity 5 treated on a different protocol due to clinician/parent preference 91 patients commenced HR block therapy 1 relapse 6 treated on a different protocol due to HR block toxicity 5 treated on a different protocol due to clinician/parent preference 79 patients completed protocol therapy 57 SCT 22 chemotherapy 3 toxic deaths 9 relapses 4 toxic deaths 2 relapses 61 patients in first remission Figure 1. Patient flow diagram for 111 HR ALL patients eligible for enrolment. HR, high risk of relapse. & 2013 Macmillan Publishers Limited Leukemia (2013) 1497 1503

1500 of a suitable donor or clinician/parent preference. Most of the 57 SCT patients had either high MRD (47%), or PPR þ T-cell þ HWBC (49%) as HR features, while chemotherapy alone patients had PPR alone (27%), PPR þ T-cell (45%), or t(4:11) (14%). Toxicity All 91 patients treated with HR blocks experienced at least one episode of grade 3 or 4 hematological toxicity requiring transfusions and hospitalisation for fever and neutropenia Table 2. Clinical and laboratory features of patients starting HR therapy (n ¼ 91) at diagnosis, compared with the total ALL cohort (n ¼ 1041) Patient category HR ALL ALL P-value a Patient number (n ¼ 91) % (n ¼ 1041) % Age 2 9 years 49 53.8 747 71.8 Age 410 years 34 37.4 230 22.1 o0.001 Gender Male 68 74.7 583 56.0 o0.001 Gender Female 23 25.3 458 44.0 WCCo50 10 9 /l 44 48.4 845 b 81.6 o0.001 WCC4100 10 9 /l 34 37.4 103 b 10.0 o0.001 B cell 47 51.6 898 86.7 o0.001 T cell 44 48.4 143 13.3 o0.001 Abbreviations: ALL, acute lymphoblastic leukemia; HR, high risk; WCC, white blood cell count. a P-values derived using Pearson s chi-squared test. b There were 6 ALL patients with missing data. Table 3. Event-free survival (EFS) and overall survival (OS) for the 91 patients starting HR block therapy by their HR clinical and molecular features Criteria for HR N 5-Year EFS 5-Year OS PPR only 15 87.5% (±11.7) 100% High MRD at day 79 30 63% (±9.4) 69.8% (±9.1) PPR þ T-cell (±HWBC) 31 80.2% (±7.3) 79.7% (±7.5) PPR þ MRD 6 83.3% (±15.2) 83.3% (±15.2) t(4;11) only 3 100% 100% No CR at day 33 5 60% (±21.9) 60% (±21.9) t(4;11) þ MRD 1 100% 100% TOTAL 91 76.5% (±4.6) 79.6% (±4.5) Abbreviations: CR, complete remission; HR, high risk; HWBC, high white blood cell count; MRD, minimal residual disease; PPR, poor prednisone response. (Table 4). Intensive care admission was required for one patient after HR1, three after HR2 and one after HR3. Most patients experienced a grade 3 or 4 gastrointestinal toxicity during HR1 (mucositis, vomiting or diarrhoea necessitating augmented nutrition). The intended duration of each HR block was 49 days. The median duration of HR1 was 48 days, compared with 49 days for HR2 and 54 days for HR3. Clinical outcomes The 5-year EFS for all 111 eligible HR patients was 66.8% (±5.5) and OS 75.6% (±4.3) (Figure 2a) at 52 months (15 94) median follow-up. For the 91 patients who received three HR blocks, 5-year EFS was 76.5% (±4.6) and OS 79.6% (±4.5) (Figure 2b). Among the 91 patients completing three HR blocks, there were 12 relapses (1 after three HR blocks but before SCT) and 7 deaths in remission. HR patients with high MRD at day 79 alone (n ¼ 30) had a 5-year EFS of 63% (±9.4) and OS of 69.8% (±9.1), while those 15 patients with PPR alone had an EFS of 87.5% (±11.7) and an OS of 100% (Table 2). Univariate analysis of HR features for OS demonstrated that patients with PPR alone had a significantly better outcome, compared with patients with either MRD alone (P ¼ 0.04) or no CR at day 33 (P ¼ 0.01). However, a comparison of EFS for patients with PPR alone to patients with either MRD alone (P ¼ 0.06) or no CR at day 33 (P ¼ 0.09) only approached statistical significance. Nine of the 57 patients who underwent SCT after three HR blocks relapsed (8 bone marrow, 1 CNS). Only 3 of 9 relapsed patients are alive and free of disease, 7 36 months after relapse. Three patients who underwent SCT died from SCT-related complications. HR patients treated with SCT had a 5-year EFS of 73.3% (±7.8) and OS of 82.7% (±5.3). Among the 22 HR patients who received chemotherapy alone, there were 2 relapses and 4 late toxic deaths in remission, resulting in a 5-year EFS of 70.5% (±10.3) and an OS 76.4% (±9.3). MRD responses and HR blocks All 91 HR patients received HR blocks in the same order and so it is not possible to directly compare the anti-leukemic effect of individual HR treatment blocks. The most significant cytoreductive effect on MRD followed HR1 (Figure 3). HR1 treatment caused an MRD reduction of 41 log in 45% of patients, compared with only 3% after HR2 and 11% after HR3. A small number of patients showed an increase in MRD following either HR2 (n ¼ 2) or HR3 (n ¼ 3). Among patients who received chemotherapy alone, HR4 6 blocks correlated with no significant changes in MRD levels, however, the patient numbers were too small to draw conclusions about the effectiveness of HR4 6. MRD was negative in 75% of all patients after three HR blocks (Figure 3). Table 4. Percentage (n) of patients who had at least one grade 3 or 4 treatment toxicity following each HR chemotherapy block Chemotherapy block HR1 (n ¼ 91) HR2 (n ¼ 89) HR3 (n ¼ 84) Toxicity grade 3 4 3 4 3 4 Haematological 2.2 (2) 97.8 (89) 2.2 (2) 97.8 (87) 9.5 (8) 91.7 (77) Infection 54.9 (50) 2.2 (2) 69.7 (62) 2.2 (2) 67.9 (57) 2.4 (2) Skin 3.3 (3) 1.1 (1) 4.5 (4) 0 2.3 (2) 0 GIT 40.7 (37) 16.5 (15) 33.7 (30) 25.8 (23) 32.1 (27) 16.7 (14) Renal 0 0 0 0 0 0 Cardiac 0 0 0 1.1 (1) 0 0 Neurological 2.2 (2) 0 2.2 (2) 0 0 0 Abbreviations: GIT, gastro-intestinal tract; HR, high risk. Leukemia (2013) 1497 1503 & 2013 Macmillan Publishers Limited

OS 100 EFS 1501 Percent survival 80 60 40 20 75.6% 66.8% Subjects at risk: 104 89 81 75 70 68 0 0 10 20 30 40 50 60 Time to event/death (months) 100 80 OS EFS 79.6% Percent survival 60 40 20 76.5% Subjects at risk: 90 78 72 71 71 71 0 0 10 20 30 40 50 60 Time to event/death (months) Figure 2. Kaplan Meier EFS and OS for HR ALL patients treated on the protocol. (a) EFS and OS for all 111 patients consecutively identified as HR. (b) EFS and OS for the 91 of 111 HR patients who commenced HR block chemotherapy. DISCUSSION We report here a prospective, multicentre, single-arm, phase 2 trial for children with HR ALL in first remission treated in Australia, New Zealand and the Netherlands, showing an improved cure rate compared with other published series in similarly defined HR ALL patients but high treatment toxicity. 6,7 The use of MRD measurements during each treatment component indicated that the highest level of cyto-reduction occurred after HR block 1. Our results compare favourably with most reports of HR ALL cohorts, but differences in the definition of HR status make direct comparisons difficult. 29 32 Our criteria for HR treatment stratification were identical to those used for patients treated on the AIEOP-BFM ALL 2000 protocol. Two recent reports on the outcomes of ALL patients treated on the AIEOP-BFM ALL 2000 protocol demonstrated 7-year EFS of 46.6% for HR B-lineage ALL (n ¼ 189) and of 49.8% for T-lineage ALL (n ¼ 97), with median patient follow-ups of 4.0 and 5.6 years, respectively. 6,7 Apart from an unexpectedly low EFS for HR patients on BFM 90 (31.9%), outcomes of HR patients in BFM 86, 95 and 2000 have been similar (45.3% 53.2%) despite changes in risk criteria and therapy. 5 Clinical outcomes for HR patients in our study were higher, with the 5-year EFS of 66.8% for all 111 eligible HR patients at a 4.4 year median follow-up. The treatment of HR patients on our protocol and AIEOP-BFM ALL 2000 were designed to be very similar, apart from HR block therapy. The differences in HR therapy were: (i) dexamethasone in Figure 3. (a) The proportion of patients with detectable MRD levels in bone marrow during HR block therapy. MRD was detected at a range of levels (0 or negative to 41 10 2 ) in bone marrow collected just before each HR block of chemotherapy. Also listed are the number (n) of patients for whom bone marrow samples were available for MRD testing at each time point. TP2 signifies the time point at the end of Protocol IB and the commencement of the first HR block of chemotherapy. (b) MRD levels of all the patients have been plotted as log values at each HR block, and a line graph has been drawn plotting the calculated mean values and s.e.m. for each HR block. We have arbitrarily assigned 10 6 as the MRD level for positive levels p10 6, which could not be quantitated, and negative values a value of 10 8 for the purposes of this line graph. each HR block in the AIEOP-BFM protocol; (ii) the use of 2 g of cyclophosphamide in the AIEOP-BFM HR1, compared with 3.6 g in our HR1; (iii) the use of 4.0 g of cytarabine in the AIEOP-BFM HR1, compared with 1.05 g of etoposide in our HR1; (iv) the use of vindesine, daunorubicin and ifosfamide in the AIEOP-BFM HR2, compared with cytarabine 7.5 g and mitozantrone in our HR2; and (v) the use of 8.0 g cytarabine and 0.5 g of etoposide in the AIEOP- BFM HR3, compared with 7.5 g of cytarabine, 18 mg of idarubicin and 112.5 mg of fludarabine in our HR3. Although it is difficult to compare individual components of a multi-agent chemotherapy program without random patient assignment, it may be concluded that our HR blocks were effective in the absence of dexamethasone and that 3.6 g cyclophosphamide combined with etoposide in HR1 may be more effective than 2.0 g of cyclophosphamide and cytarabine. Our study also provided greater dose intensity compared with AIEOP-BFM ALL 2000, as our HR patients essentially received chemotherapy at 2 weekly intervals, compared with 3 weekly intervals for the AIEOP-BFM ALL 2000 HR protocol. Moreover, the 5-year EFS of 63% for HR patients with high MRD in our study is much higher than the 3-year relapse free survival of 25% seen in BFM 95 for patients defined by the same MRD criteria. 3 & 2013 Macmillan Publishers Limited Leukemia (2013) 1497 1503

1502 HR block treatment led to considerable toxicity in all the patients, as exemplified by the high incidence of grade 4 toxicities, intensive care admission, prolonged hospitalisation and 6.3% incidence of death in remission. In addition, we believe the 9% (10/111) incidence of eligible patients declining to be treated on protocol therapy, or withdrawing after commencing HR block therapy, highlighted concerns of the parents and clinicians about treatment toxicity. Future HR protocols will seek to reduce the number and duration of HR blocks and improve infection prevention. Our study was not designed to address the role of SCT in HR ALL. The use of SCT in HR ALL in first remission has correlated with reduced relapse in some studies but not others. 7,15,16,33 35 The higher proportion of patients in our study receiving SCT (63%), compared with HR ALL patients treated on AIEOP-BFM ALL 2000 (57%), correlated with a better overall EFS. Patients in our study receiving SCT had a similar cure rate to those treated with chemotherapy alone; however, the latter group had a higher proportion with PPR or t(4:11) alone and no other HR features. Thus, we cannot conclude that SCT was a necessary factor in the high EFS achieved in our cohort. The therapeutic strategy of using surrogate markers of in vivo treatment response during early treatment phases, before clinical relapse and consequently intensifying therapy is now possible with MRD measurement. The improved cure rate seen in our study demonstrated the utility of this approach and confirm the role of MRD as a powerful prognostic factor in ALL. 5 7,14 CONFLICT OF INTEREST The authors declare no conflict of interest. ACKNOWLEDGEMENTS We thank Prof Martin Schrappe and the I-BFM for their support. This work was supported by the National Health and Medical Research Council of Australia, Cancer Council New South Wales, Steven Walter Children s Cancer Foundation, Cancer Institute New South Wales and Leukemia Foundation. AUTHOR CONTRIBUTIONS Conception and design: LDP, GMM, MDN and RP. Administrative support: RS, AN, HG, SC, HM and VH. Provision of study materials or patients: LDP, RS, NCV, VHV, HB, ESJMB, RME, PMH, GJLK, ES, JD, TL, MDN, MH, TR, A, RS, RP and GMM. Collection and assembly of data: GMM, LDP, RS, AN, HG-K, NCV, VHV, HM, VH, MDN, MH, TR, FA and RP. Data analysis and interpretation: GMM, LDP, RS, AN, HG-K, NCV, VHV, SC, HM, VH, MDN, MH, TR, FA, RS and RP. Manuscript writing: GMM, LDP, RS, AN and RP. Final approval of manuscript: All the authors. REFERENCES 1 Schrappe M, Nachman J, Hunger S, Schmiegelow K, Conter V, Masera G et al. Educational symposium on long-term results of large prospective clinical trials for childhood acute lymphoblastic leukemia (1985-2000). Leukemia 2010; 24(2): 253 254. 2 Pui CH, Campana D, Pei D, Bowman WP, Sandlund JT, Kaste SC et al. Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med 2009; 360: 2730 2741. 3 Van Dongen JJM, Seriu T, Panzer-Grümayer ER, Biondi A, Pongers-Willemse MJ, Corral L et al. Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood. Lancet 1998; 352: 1731 1738. 4 Marshall GM, Haber M, Kwan E, Zhu L, Ferrara D, Xue C et al. Importance of minimal residual disease testing during the second year of therapy for children with acute lymphoblastic leukaemia. J Clin Oncol 2003; 21(4): 704 709. 5 Möricke A, Reiter A, Zimmermann M, Gadner H, Stanulla M, Dördelmann M et al. Risk-adjusted therapy of acute lymphoblastic leukemia can decrease treatment burden and improve survival: treatment results of 2169 unselected pediatric and adolescent patients enrolled in the trial ALL-BFM 95. Blood 2008; 111(9): 4477 4489. 6 Conter V, Bartram CR, Valsecchi MG, Schrauder A, Panzer-Grümayer R, Möricke A et al. Molecular response to treatment redefines all prognostic factors in children and adolescents with B-cell precursor acute lymphoblastic leukemia: results in 3184 patients of the AIEOP-BFM ALL 2000 study. Blood 2010; 115: 3206 3214. 7 Schrappe M, Valsecchi MG, Bartram CR, Schrauder A, Panzer-Grümayer R, Möricke A et al. Late MRD response determines relapse risk overall and in subsets of childhood T-cell ALL: results of the AEIOP-BFM-ALL 2000 Study. Blood 2011; 118(8): 2077 2084. 8 Tallen G, Ratei R, Mann G, Kaspers G, Niggli F, Karachunsky A et al. Long-term outcome in children with relapsed acute lymphoblastic leukemia after time-point and site-of-relapse stratification and intensified short-course multidrug chemotherapy: results of Trial ALL-REZ BFM 90. J Clin Oncol 2010; 28(14): 2339 2347. 9 Parker C, Waters R, Leighton C, Hancock J, Sutton R, Moorman AV et al. Effect of mitoxantrone on outcome of children with first relapse of acute lymphoblastic leukaemia (ALL R3): an open-label randomised trial. Lancet 2010; 376: 2009 2017. 10 Coustan-Smith E, Sancho J, Behm FG, Hancock ML, Razzouk BI, Ribeiro RC et al. Prognostic importance of measuring early clearance of leukemic cells by flow cytometry in childhood acute lymphoblastic leukemia. Blood 2002; 100: 52 58. 11 Basso G, Veltroni M, Valsecchi MG, Dworzak MN, Ratei R, Silvestri D et al. Risk of relapse of childhood acute lymphoblastic leukaemia is predicted by flow cytometric measurement of residual disease on day 15 bone marrow. J Clin Oncol 2009; 27(31): 5168 5174. 12 Borowitz MJ, Devidas M, Hunger SP, Bowman WP, Carroll AJ, Carroll WL et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia and its relationship to other prognostic factors: a Children s Oncology Group study. Blood 2008; 111(12): 5477 5485. 13 Choi S, Henderson MJ, Kwan E, Beesley AH, Sutton R, Bahar AY et al. Relapse in children with acute lymphoblastic leukaemia involving selection of a pre-existing drug resistant sub-clone. Blood 2007; 110(2): 632 639. 14 Schultz KR, Bowman WP, Aledo A, Slayton WB, Sather H, Devidas M et al. Improved early EFS with imatinib in philadelphia chromosome-positive acute lymphoblastic leukemia: a Children s Oncology Group study. J Clin Oncol 2009; 27: 5175 5181. 15 Ribera J-M, Ortega J-J, Oriol A, Bastida P, Calvo C, Pérez-Hurtado JM et al. Comparison of intensive chemotherapy, allogeneic, or autologous stem-cell transplantation as postremission treatment for children with very high risk acute lymphoblastic leukemia: PETHEMA ALL-93 Trial. J Clin Oncol 2007; 25(1): 16 24. 16 Schrauder A, Reiter A, Gadner H, Niethammer D, Klingebiel T, Kremens B et al. Superiority of allogeneic hematopoietic stem-cell transplantation compared with chemotherapy alone in high-risk childhood T-cell acute lymphoblastic leukemia: results from ALL-BFM 90 and 95. J Clin Oncol 2006; 24(36): 5742 5749. 17 Aricó M, Schrappe M, Hunger SP, Carroll WL, Conter V, Galimberti S et al. Clinical outcome of children with newly diagnosed philadelphia chromosome positive acute lymphoblastic leukemia treated between 1995 and 2005. J Clin Oncol 2010; 28(31): 4755 4761. 18 Brown RA, Herzig RH, Wolff SN, Frei-Lahr D, Pineiro L, Bolwell BJ et al. etoposide and cyclophosphamide without bone marrow transplantation for resistant hematologic malignancy. Blood 1990; 76(3): 473 479. 19 Morland BJ, Shaw PJ.. Induction toxicity of a modified Memorial Sloan-Kettering New York II protocol in children with relapsed adult lymphoblastic leukemia: a single institution study. Med Ped Oncol 1996; 27: 139 144. 20 Feldman EJ, Albert DS, Arlin Z, Ahmed T, Mittelman A, Baskind P et al. Phase I clinical and pharmacokinetic evaluation of high-dose mitoxantrone in combination with cytarabine in patients with acute leukemia. J Clin Oncol 1993; 11(10): 2002 2009. 21 Arlin ZA, Feldman EJ, Finger LR, Ahmed T, Mittelman A, Cook P et al. Short course high-dose mitoxantrone with high-dose cytarabine is effective therapy for adult lymphoblastic leukemia. Leukemia 1991; 5(8): 712 714. 22 Visani G, Tosi P, Zinzani PL, Manfroi S, Ottaviani E, Cenacchi A et al. FLAG (fludarabine, cytarabine, G-CSF) as second line therapy for acute lymphoblastic leukemia with myeloid antigen expression: in vitro and in vivo effects. Eur J Haem 1996; 56(5): 308 312. 23 McCarthy AJ, Pitcher LA, Hann IM, Oakhill A.. FLAG (fludarabine, high-dose cytarabine and G-CSF) for refractory and relapsed high risk acute leukemia in children. Med Ped Oncol 1999; 34(2): 411 415. 24 Flohr T, Schrauder A, Cazzaniga G, Panzer-Grümayer R, van der Velden V, Fischer S et al. Minimal residual disease-directed risk stratification using real-time quantitative PCR analysis of immunoglobulin and T-cell receptor gene rearrangements in the international multicenter trial AIEOP-BFM ALL 2000 for childhood acute lymphoblastic leukemia. Leukemia 2008; 22: 771 782. 25 Sutton R, Bahar AY, Kwan E, Giles JE, Venn NC, Tran S et al. Improving minimal residual disease detection in precursor-b-all based on immunoglobulin kappa and heavy chain gene rearrangements. Leukemia 2008; 22: 2265 2267. Leukemia (2013) 1497 1503 & 2013 Macmillan Publishers Limited

26 Sutton R, Venn NC, Tolisano J, Bahar AY, Giles JE, Ashton LJ et al. Clinical significance of minimal residual disease at day 15 and at the end of therapy in childhood acute lymphoblastic leukaemia. Brit J Haem 2009; 146: 292 299. 27 van der Velden VHJ, Cazzaniga G, Schrauder A, Hancock J, Bader P, Panzer-Grumayer ER et al. Analysis of minimal residual disease by Ig/TCR gene rearrangements: guidelines for interpretation of real-time quantitative PCR data. Leukemia 2007; 21: 604 611. 28 Trotti A, Colevas AD, Setser A, Rusch V, Jaques D, Budach V et al. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 2003; 13(3): 176 181. 29 Nachman J, Sather HN, Cherlow JM, Sensel MG, Gaynon PS, Lukens JN et al. Response of children with high-risk acute lymphoblastic leukemia treated with and without cranial irradiation: a report from the Children s Cancer Group. J Clin Oncol 1998; 16: 920 930. 30 Aricò M, Valsecchi MG, Conter V, Rizzari C, Pession A, Messina C et al. Improved outcome in high-risk childhood acute lymphoblastic leukemia defined by prednisone-poor response treated with double Berlin-Frankfurt-Muenster protocol II. Blood 2002; 100: 420 426. 31 Moghrabi A, Levy DE, Asselin B, Barr R, Clavell L, Hurwitz C et al. Results of the Dana-Farber Cancer Institute ALL Consortium Protocol 95-01 for children with acute lymphoblastic leukemia. Blood 2007; 109: 896 904. 32 Seibel NL, Steinherz PG, Sather HN, Nachman JB, Delaat C, Ettinger LJ et al. Early postinduction intensification therapy improves survival for children and adolescents with high-risk acute lymphoblastic leukemia: a report from the Children s Oncology Group. Blood 2008; 111: 2548 2555. 33 Bordigoni P, Vernant JP, Souillet G, Gluckman E, Marininchi D, Milpied N et al. Allogeneic bone marrow transplantation for children with acute lymphoblastic leukemia in first remission: a cooperative study of the Groupe d Etude de la Greffe de Moelle Osseuse. J Clin Oncol 1989; 7: 747 753. 34 Balduzzi A, Valsecchi MG, Uderzo C, De Lorenzo P, Klingebiel T, Peters C et al. Chemotherapy versus allogeneic transplantation for very high-risk childhood acute lymphoblastic leukaemia in first complete remission: comparison by genetic randomisation in an international prospective study. Lancet 2005; 366: 635 642. 35 Satwani P, Sather H, Ozkaynak F, Heerema NA, Schultz KR, Sanders J et al. Allogeneic bone marrow transplantation in first remission for children with ultra-high-risk features of acute lymphoblastic leukemia: a Children s Oncology Group study report. Biol Blood Marrow Transplant 2007; 13: 218 227. 1503 & 2013 Macmillan Publishers Limited Leukemia (2013) 1497 1503