Results of Dana-Farber Cancer Institute Consortium protocols for children with newly diagnosed acute lymphoblastic leukemia ( )

Transcription

1 (2000) 14, Macmillan Publishers Ltd All rights reserved /00 $ Results of Dana-Farber Cancer Institute Consortium protocols for children with newly diagnosed acute lymphoblastic leukemia ( ) LB Silverman 1,2, L Declerck 3, RD Gelber 3, V Kimball Dalton 1, BL Asselin 4, RD Barr 5, LA Clavell 6, CA Hurwitz 7, A Moghrabi 8, Y Samson 9, MA Schorin 10, JM Lipton 11, HJ Cohen 12 and SE Sallan 1,2 1 Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA; 2 Division of Hematology/Oncology, Children s Hospital and Department of Pediatrics, Harvard Medical School, Boston, MA; 3 Department of Biostatistical Science, Dana-Farber Cancer Institute, Boston, MA; 4 Department of Hematology/Oncology, University of Rochester Medical Center, Rochester, NY, USA; 5 Department of Pediatrics, McMaster University, Hamilton, Ontario, Canada; 6 San Jorge Children s Hospital, Puerto Rico; 7 Maine Children s Cancer Program, The Barbara Bush Children s Hospital at Maine Medical Center, Portland, ME, USA; 8 Hospital Ste Justine, Montreal, Quebec, Canada; 9 Le Centre Hospitalier de L Université, Laval, Quebec, Canada; 10 Department of Pediatrics, Oschner Medical Institutions, New Orleans, LA; 11 Schneider Children s Hospital, New Hyde Park, NY; 12 Department of Pediatrics, Stanford University School of Medicine, Stanford, CA, USA The Dana-Farber Cancer Institute (DFCI) ALL consortium has been conducting clinical trials in childhood acute lymphoblastic leukemia (ALL) since The treatment backbone has included intensive, multi-agent remission induction, early intensification with weekly, high-dose asparaginase, cranial radiation for the majority of patients, frequent vincristine/ corticosteroid pulses during post-remission therapy, and for high-risk patients, doxorubicin during intensification. Between 1981 and 1995, 1255 children with newly diagnosed ALL were evaluated on four consecutive protocols: ( ), ( ), ( ) and ( ). The 5-year event-free survival (EFS) rates (± standard error) for all patients by protocol were as follows: 74 ± 3% (81-01), 78 ± 3% (85-01), 77 ± 2% (87-01) and 83 ± 2% (91-01). The 5-year EFS rates ranged from 78 to 85% for patients with B-progenitor phenotype retrospectively classified as NCI standard-risk, 63 82% for NCI high-risk B-progenitor patients, and 70 79% for patients with T cell phenotype. Results of randomized studies revealed that neither high-dose methotrexate during induction (protocol 87-01) nor high-dose 6-mercaptopurine during intensification (protocol 91-01) were associated with improvement in EFS compared with standard doses. Current studies continue to focus on improving efficacy while minimizing acute and late toxicities. (2000) 14, Keywords: acute lymphoblastic leukemia; childhood; asparaginase; doxorubicin Introduction The Dana-Farber Cancer Institute (DFCI) began conducting randomized clinical trials in childhood ALL in Studies performed during the 1970s demonstrated improved eventfree survival (EFS) for children who received doxorubicin (in addition to vincristine and prednisone) during remission induction therapy, 1 and for those who received weekly, highdose asparaginase during the post-remission intensification phase. 2 In 1981, DFCI and several other institutions throughout the United States and Canada formed the DFCI ALL Consortium. Features of DFCI ALL Consortium protocols have included intensive, four- to five-agent remission induction, early intensification with weekly, high-dose asparaginase, cranial radiation for the majority of patients, frequent pulses of vincristine and corticosteroid during continuation therapy, and for highrisk patients, doxorubicin during intensification. The therapeutic success over the last two decades has been accompanied by recognition of acute toxicities and late effects Correspondence: LB Silverman, Dana-Farber Cancer Institute, 44 Binney Street, Boston, MA 02115, USA; Fax: Received 15 August 2000; accepted 7 September 2000 in long-term survivors. Acute toxicities experienced by patients treated on DFCI ALL consortium protocols include asparaginase-related allergic events and pancreatitis, as well as fractures from corticosteroids. 3 5 Late toxicities include asymptomatic echocardiographic abnormalities related to anthracycline exposure, 6 osteonecrosis from corticosteroids, 7 as well as short stature and learning disabilities of varying severity, presumably secondary to central nervous system (CNS)-directed therapy. 8 The focus of randomized studies has been to improve efficacy while minimizing acute and late toxicities. In this report, we review the results of four consecutive clinical trials conducted between 1981 and 1995 by the DFCI ALL consortium. Patients and methods Patients Between 1981 and 1995, 1255 children aged 0 18 years with newly diagnosed ALL (excluding mature B cell ALL) were evaluated on four consecutive DFCI ALL consortium protocols and were evaluable for results of treatment. Patients were enrolled on the following protocols: ( , n = 289), ( , n = 220), ( , n = 369) and ( , n = 377). Informed consent was obtained from parents or guardians prior to instituting therapy. Patients were enrolled from the following DFCI ALL consortium institutions: DFCI/Children s Hospital ( ), Maine Medical Center/Maine Children s Cancer Program ( ), University of Rochester Medical Center ( ), Ochsner Clinic, New Orleans ( ), University of Massachusetts ( ), University of Puerto Rico, San Juan ( ), Eastern Maine Medical Center ( ), McMaster University Medical Center ( ), Mount Sinai Medical Center ( ), San Jorge Children s Hospital, San Juan ( ), Hospital Sainte Justine, Montreal ( ), and Le Centre Hospitalier de L Université, Laval, Quebec ( ). All protocols were approved by the institutional review boards of each participating institution. Risk group classification Risk group was determined at the time of diagnosis (Table 1). On all protocols between 1981 and 1995, standard risk (SR) was defined as follows: age between 2 and 9 years, initial white blood cell (WBC) count /l, B-progenitor

2 2248 Table 1 Risk group criteria on DFCI ALL consortium protocols ( ) Standard risk Very high risk High risk ( ) ( ) ( ) All of the Either or both of Any of the following the following following Age 2 to 9 years 12 months All others WBC /l /l a All others CNS disease b Absent Present Phenotype non-t cell T cell Mediastinal Absent Present mass t(9;22) Absent Present a Criterion used only. b See text for definition. WBC, white blood cell count; CNS, central nervous system. phenotype, absence of mediastinal mass, absence of Philadelphia-chromosome and absence of CNS leukemia (see definition below). Between 1985 and 1991 (protocols and 87-01), a very high risk (VHR) group was defined that included patients with presenting WBC counts /l and/or age 12 months at diagnosis. Between 1991 and 1995 (protocol 91-01), only infants (age 12 months) were considered VHR. All patients not classified as SR or VHR were considered high risk (HR). CNS leukemia was defined as the presence of leukemic blast cells on a cytospin preparation in conjunction with five WBC/high powered field (hpf) (CNS-3) from 1981 to 1987 (protocols and 85-01); after 1987 (protocols and 91-01), CNS leukemia was defined as the presence of leukemic blast cells on a cytospin preparation regardless of CSF WBC count (CNS-2 and CNS-3). Therapy Details of therapy have been previously published, 3,4,9,10 and are summarized in Table 2. In general, each protocol consisted of five phases of therapy: investigational window, remission induction, CNS treatment, intensification, and continuation. VHR patients received an additional month of intensification therapy (VHR intensification) immediately after remission induction. Investigational windows consisted of a single chemotherapeutic agent administered at the time of diagnosis 3 5 days prior to the initiation of multi-agent remission induction, and were designed to assess initial response. Results have been previously reported Two hundred and eighty-nine patients were registered on protocol ( ). All patients were eligible and evaluable for follow-up. Seventy-seven patients were randomized to receive a single dose of methotrexate either 4 gm/m 2 (high-dose) or 40 mg/m 2 (low-dose) during an investigational window (4 days prior to the initiation of remission induction therapy). 14 There was a single dose of doxorubicin during remission induction (45 mg/m 2 ). Only HR patients received doxorubicin during intensification (cumulative dose 345 mg/m 2 ). All patients received cranial radiation (2800 cgy for HR patients and 1800 cgy for SR patients). Males with T cell ALL (n = 24) also received bilateral testicular radiation (2400 cgy). All patients received 20 weeks of high-dose asparaginase during the intensification phase. Total duration of therapy for all patients was 24 months from the time of CR. On protocol 85-01, the HR group was subdivided into HR and VHR categories (Table 1). Two hundred and twenty-five patients were registered between 1985 and Five patients were considered ineligble due to incorrect diagnosis (n = 4) and subsequent decision to treat with another protocol (n = 1). Therapy was identical to its predecessor except for the following: (1) The investigational window consisted of a single dose of E. coli L-asparaginase (randomized to either IU/m 2 or 2500 IU/m 2 ) administered 5 days prior to remission induction therapy. 11 (2) There were two consecutive daily doses of doxorubicin (30 mg/m 2 /dose) and one dose of lowdose methotrexate during remission induction. (3) Cumulative dose of doxorubicin for HR/VHR patients was 360 mg/m 2. (4) The dose of cranial radiation for HR/VHR was decreased to 2400 cgy (2200 cgy for those between 12 and 24 months). (5) No testicular radiation was administered. (6) VHR patients received an additional cycle of intensification (including highdose cytarabine and high-dose methotrexate) immediately after remission induction (Table 2). Three hundred and eighty-one patients were registered on protocol between 1987 and Ten patients were ineligible due to incorrect diagnosis and two were cancelled because family declined all treatment and follow-up, leaving 369 evaluable patients. Therapy differed from protocol as follows: (1) Patients were randomized to receive one of three types of asparaginase as a single dose during the 5-day investigational window (either E. coli IU/m 2, Erwinia IU/m 2 or PEG 2500 IU/m 2 ). 11,12 (2) Patients were randomized to receive either high- or low-dose methotrexate during remission induction (dosed as in protocol 81-01). (3) SR patients were treated without cranial radiation. (4) The dose of cranial radiation was decreased to 1800 cgy for HR patients. (5) Patients receiving cranial radiation were randomized to either daily (conventional) or twice-daily (hyperfractionated) schedules. (6) Beginning in 1989, patients with t(9;22) were treated with allogeneic bone marrow transplant in first remission. Because of a higher than expected incidence of CNS relapse in SR boys, all boys who had received SR treatment and remained in CCR were recalled to receive additional therapy in Forty of 60 eligible SR boys received 1 year of additional therapy, including a month of remission induction, 1800 cgy cranial radiation, and then cycles of continuation therapy, continuing until 1 year from the completion of reinduction or 2 years of CCR (for those patients recalled within the first year from their diagnosis). Three hundred and eighty-six children were registered on protocol ( ). Nine patients were ineligible due to incorrect diagnosis (n = 6), pre-treatment with other anti-leukemia therapy (n = 2) and absence of signed parental consent prior to therapy (n = 1). The remaining 377 evaluable patients received therapy identical to protocol 87-01, except for the following: (1) VHR patients consisted only of those 12 months at diagnosis (no WBC criteria). (2) Patients were randomized to receive one of four corticosteroid regimens during a 3-day investigational window (prednisone 40 mg/m 2 /day, or dexamethasone 6, 18 or 150 mg/m 2 /day. 13 (3) All patients received high-dose methotrexate during remission induction. (4) Dexamethasone was substituted for prednisone during the intensification and continuation phases (Table 2). (5) Patients received 30 rather than 20 weeks of asparaginase during intensification. (6) SR boys received 1800 cgy cranial radiation; SR girls continued to be treated without radiation. (7) Patients were randomized to receive either PEG or E. coli asparaginase during intensification. (8) Patients were randomized to receive either high-dose i.v. 6- mercaptopurine (1000 mg/m 2 /dose over 20 h) or conventional

3 Table 2 Therapy on DFCI ALL Consortium protocols ( ) Protocol Investigational window Mtx 1 dose (randomized) Asp 1 dose (randomized) Asp 1 dose (randomized) Steroid 3 days (randomized) (3 5 days) Induction (4 weeks) Vcr 1.5 mg/m 2 q week Same as 81-01, except: Same as 85-01, except: Same as in 87-01, except: all Mtx 4 g/m 2 Pred 40 mg/m 2 /day Dox 30 mg/m 2 2 doses Mtx 40 mg/m 2 or 4 g/m 2 Dox 45 mg/m 2 1 dose Mtx 40 mg/m 2 1 dose (randomized) IT ara-c (dosed by age) CNS treatment (3 weeks) SR: cranial XRT 1800 cgy SR: same as SR: IT mtx/ara-c SR girls: same as SR patients in HR: cranial XRT 2800 cgy HR/VHR: cranial XRT 2400 cgy 2 /week 2 weeks, then SR boys: same as HR patients in IT meds a for both IT meds a for both q 18 weeks HR/VHR: same as HR: cranial XRT 1800 cgy, b with IT meds a Intensification ( 6 9 q 3-week cycles months) SR: Vcr 2.0 mg/m 2 i.v. q 3 week SR: same as SR: same as SR: same as 87-01, d except 6MP 50 mg/m 2 /day p.o. 14 days Dex 6 mg/m 2 /day p.o. 5 days (instead of Pred 40 mg/m 2 /day p.o. 5 days pred) Mtx 30 mg/m 2 i.v./i.m. q week Asp IU/m 2 i.m. q week HR: same as SR patients, except HR: same as 81-01, except HR: same as HR: same as d, except Dex Pred dose (120 mg/m 2 /day p.o. 5 cumulative Dox 360 mg/m 2 18 mg/m 2 /day p.o. days (instead of pred) days), and Dox 30 mg/m 2 q 3 week (no MTX) VHR: same as HR, except 1 VHR: same as c VHR: same as c until cumulative dose of 345 mg/m 2 additional cycle with high-dose arac/mtx c Continuation (until 2 q 3 week cycles years CCR) SR: same as intensification, except no same as same as SR: same as 81-01, except Dex instead of asparaginase pred (as above) HR: same as SR patients, except pred dose HR: same as SR patients, except Dex dose (as above) (as above) ara-c, cytarabine; Asp, asparaginase; Dex, dexamethasone; Dox, doxorubicin; IT, intrathecal; Mtx, methotrexate; Pred, prednisone; Vcr, vincristine; 6MP, 6-mercaptopurine; SR; standard risk; HR, high risk; VHR, very high risk. Maximum dose vincristine 2.0 mg. Dosing per kilogram for patients with BSA 0.6 m 2 (m 2 dosing divided by 30). Intensification: weekly asparaginase 20 weeks on all protocols except (30 weeks). a IT meds with cranial XRT: IT mtx/ara-c 2 /week 2 weeks, then q 18 weeks. b XRT randomization on and 91-01: cranial XRT 10 days: 180 cgy q d or 90 cgy bid. c VHR intensification cycle: immediately after remission induction: HD mtx 130 mg/kg i.v. over 1 h on wks 1 and 2 (with leucovorin rescue); HD ara-c 100 mg/kg i.v. q 12 h 6 doses on wk 3; asp 825 IU/kg i.m. wks 2 and 3; vcr 0.05 mg/kg i.v. q week 4, 6MP 1.3 mg/kg/d p.o. days 1 14; IT mtx 6 mg/dose wks 1 and 2. d Intensification randomizations on 91-01: asparaginase (E. coli IU/m 2 q week or PEG 2500 IU/m 2 q other week), 6-MP (1000 mg/m 2 /dose i.v. over 20 hours on weeks or 50 mg/m 2 /d p.o. 2 weeks); HR only: dox (30 mg/m 2 /dose delivered as 48-h continuous infusion or i.v. bolus). 2249

4 2250 Figure 1 Event-free survival, cumulative incidence of any CNS relapse and of isolated CNS relapse for 289 patients treated on protocol ( ). Median follow-up was 15.4 years. oral 6-mercaptopurine (50 mg/m 2 /day) on weeks 1 and 2 of each 3-week cycle during the first year of post-remission therapy. (9) HR patients were randomized to receive doxorubicin either as an i.v. bolus or as a 48-h continuous infusion during intensification. Immunophenotype and cytogenetics Bone marrow cells from diagnostic aspirates were examined for cell surface antigens using standard indirect immunofluorescence assays and cultured for cytogenetic analyses, as previously described. 10 Statistical analysis Outcome events were death during induction therapy, failure to achieve complete remission (defined as persistent leukemia at day 52 after diagnosis), death during remission, relapse and diagnosis of second malignancy. Event-free survival (EFS) was measured from the date of complete remission to the first outcome event or, if no such events occurred, until the date of last contact; induction failure and induction death were considered events at time zero. Overall survival (OS) was the time from the start of treatment to death from any cause. EFS and OS were estimated using the Kaplan Meier method, 15 and Greenwood s formula was used to calculate standard errors. 16 Univariate analyses of differences in EFS were conducted with logrank tests. 17 Cumulative incidence functions of isolated and any CNS relapse were constructed by the method of Kalbfleish and Prentice for patients who achieved a complete remission. 18 An isolated CNS relapse was defined as a CNS relapse without relapse at other site or other type of failure. Any CNS relapse was defined as an isolated CNS relapse or a CNS relapse in conjunction with another type of relapse or failure. Similar definitions were used in defining isolated and any testicular relapse. In the estimation of cumulative incidence functions, all other failures were considered competing events. The databases for analyses of protocols and 85- Table 3 Protocol treatment results according to presenting features ( ) Presenting feature No. of 5-year 8-year 10-year patients EFS EFS EFS (±s.e.) (±s.e.) (±s.e.) Overall ± 3 71 ± 3 69± 3 Non-T lineage Standard ± 3 76 ± 3 74± 3 High ± 5 62 ± 5 59± 5 T lineage Standard ± ± ± 13 High ± 8 76 ± 8 76± 8 Age at diagnosis 1 year years ± 3 73 ± 3 72± 3 10 years ± 6 68 ± 6 67± 6 WBC (10 9 /l) ± 3 75 ± 4 73± ± 5 67 ± 5 67± ± ± ± ± 7 69 ± 7 66± 7 Lineage B lineage ± 3 70 ± 3 68± 3 T lineage ± 7 77 ± 7 77± 7 CNS leukemia CNS ± ± ± 14 Other ± 3 71 ± 3 69± 3 Sex Male ± 4 68 ± 4 66± 4 Female ± 4 74 ± 4 73± 4 t(9;22) present 6 17 ± all other patients ± 3 72 ± 3 71± 3 DFCI risk group Standard ± 3 81 ± 4 79± 4 High/very high ± 4 65 ± 4 63± 4 Methotrexate High-dose ± 6 82 ± 6 82± 6 Low-dose ± 7 64 ± 8 64± 8 Median follow-up: 15.4 years.

5 01 were frozen in June 2000, for in March 2000 and for in November Results Protocol ( ) Of the 289 evaluable patients, 278 entered CR (96%). Median follow-up was 15.4 years. Of the 200 patients who remained in CR, 193 (97%) were followed for at least 5 years, 182 (91%) for at least 8 years and 168 (84%) for at least 10 years. The 5- year EFS (±s.e.) for all patients was 74 ± 3% (Figure 1). Overall survival rates at 5 years, 8 years and 10 years were 81 ± 2%, 77 ± 3% and 75 ± 3%, respectively. The cumulative incidence of isolated CNS relapse at 5 years was 4.5 ± 1.2% and that of any CNS relapse (isolated or combined) was 6.3 ± 1.4%. The cumulative incidence of isolated or combined testicular relapse in male patients at 5 years was 0.6 ± 0.6% (observed in only one patient). Five patients experienced a second malignancy as a first event (glioma, astrocytoma, glioblastoma, meningioma, and germ cell tumor). Remission death rate was 3.5%. Outcome by presenting features is shown in Table 3. Seventy-eight patients participated in a randomized comparison of high- vs low-dose methotrexate as part of an investigational window. There was a trend toward improved EFS with high-dose methotrexate that was not statistically significant (P = 0.10). Protocol ( ) Of the 220 evaluable patients, 217 entered CR (99%). Median follow-up was 12.0 years. Of the 171 patients who remained in CR, 162 (95%) were followed for at least 5 years, 139 (81%) for at least 8 years and 118 (69%) for at least 10 years. The 5- year EFS (±s.e.) for all patients was 78 ± 3% (Figure 2). Overall survival rates at 5, 8 and 10 years were 84 ± 2%, 81 ± 3% and 81 ± 3%, respectively. The cumulative incidence of isolated CNS relapse at 5 years was 2.8 ± 1.1% and that of any CNS relapse (isolated or combined) was 3.7 ± 1.3%. The cumulative incidence of isolated or combined testicular relapse in male patients at 5 years was 0.9 ± 0.9% (observed in only one patient). Two patients experienced a second malignancy as a first event (meningioma and AML). Remission death rate was 3.6%. Outcome by presenting features is shown in Table 4. Protocol ( ) Of the 369 evaluable patients, 356 entered CR (96%). Median follow-up was 9.0 years. Of the 276 patients who remained in CR, 267 (97%) have been followed for at least 5 years and 213 (77%) for at least 8 years. The 5-year EFS (±s.e.) for all patients was 77 ± 2% (Figure 3). Overall survival rates at 5, 8 and 10 years were 88 ± 2%, 85 ± 2% and 84 ± 2%, respectively. The cumulative incidence of isolated CNS relapse at 5 years was 4.1 ± 1.0% and that of any CNS relapse (isolated or combined) was 5.7 ± 1.2%. The cumulative incidence of isolated or combined testicular relapse in male patients at 5 years was 0.9 ± 0.7% (observed in two patients). Two patients experienced a second malignancy as a first event (both AML). Remission death rate was 2.0%. Methotrexate dose (high vs low) during remission induction was not associated with significant differences in EFS (8-year EFS of 77 ± 3% for highdose vs 73 ± 3% for low-dose; P = 0.48). Outcome by presenting features is shown in Table 5. Protocol ( ) Of the 377 evaluable patients, 370 entered CR (98%). Median follow-up was 4.9 years. Of the 312 patients who remained in CR, 306 (98%) have been followed for at least 3 years and 233 (75%) for at least 4 years. The 5-year EFS (±s.e.) for all patients was 83 ± 2% (Figure 4). The overall survival rate at 5 years was 88 ± 2%. The cumulative risk of isolated CNS relapse at 5 years was 1.1 ± 0.5% and that of any CNS relapse (isolated or combined) was 3.1 ± 0.9%. The cumulative incidence of isolated or combined testicular relapse in male 2251 Figure 2 Event-free survival, cumulative incidence of any CNS relapse and of isolated CNS relapse for 220 patients treated on protocol ( ). Median follow-up was 12.0 years.

6 2252 Table 4 Protocol treatment results according to presenting features ( ) Presenting feature No. of 5-year 8-year 10-year patients EFS EFS EFS (±s.e.) (±s.e.) (±s.e.) Overall ± 3 77 ± 3 77± 3 Non-T lineage Standard ± 3 83 ± 3 83± 3 High ± 6 70 ± 6 70± 6 T lineage Standard 6 83 ± ± ± 15 High ± ± ± 13 Age at diagnosis 1 year ± ± ± years ± 3 80 ± 3 80± 3 10 years ± 7 71 ± 7 71± 7 WBC (10 9 /l) ± 3 84 ± 4 84± ± 5 76 ± 5 76± ± 9 83 ± 9 83± ± ± ± 10 Lineage B lineage ± 3 78 ± 3 78± 3 T lineage ± ± ± 10 CNS leukemia CNS ± ± ± 25 Other ± 3 78 ± 3 78± 3 Sex Male ± 4 74 ± 4 74± 4 Female ± 4 81 ± 4 81± 4 t(9;22) present all others ± 3 78 ± 3 78± 3 DFCI risk group Standard ± 4 87 ± 4 87± 4 High/very high ± 4 72 ± 4 72± 4 Median follow-up: 12.0 years. patients at 5 years was 1.5 ± 0.9% (observed in three patients). No patients have been diagnosed with a second malignancy. Remission death rate was 3.2%. Dosing of 6-MP during the first year of therapy was not associated with significant differences in EFS (5-year EFS of 85 ± 3% for high-dose and % for low-dose, P = 0.90). Outcome by presenting features is shown in Table 6. Prognostic factors For each protocol, we performed univariate analyses to determine if the following factors were significantly associated with EFS: DFCI risk group, age, WBC count, sex and immunophenotype. None of these features was prognostically significant on protocol DFCI risk group was a significant risk factor on protocols (P 0.01) and (P 0.01), but not on (P = 0.32) and (P = 0.24). Age at diagnosis was prognostically significant only on protocols (P 0.01) and (P = 0.04), and presenting WBC only on protocols (P 0.01) and (P 0.01). Male sex was associated with inferior outcome on protocol (P 0.01), but not on any other protocol. Inferior outcome of males on protocol was due to the high incidence of CNS relapses in SR boys who did not receive cranial radiation. Immunophenotype was not a significant risk factor on any protocol during this treatment era. Table 5 Protocol treatment results according to presenting features ( ) Presenting feature No. of 5-year 8-year 10-year patients EFS EFS EFS (±s.e.) (±s.e.) (±s.e.) Overall ± 2 74 ± 2 74± 2 Non-T lineage Standard ± 3 77 ± 3 77± 3 High ± 4 70 ± 5 70± 5 T lineage Standard ± ± ± 13 High ± 8 77 ± 8 77± 8 Age at diagnosis 1 year 8 50 ± ± ± years ± 2 77 ± 2 77± 2 10 years ± 5 66 ± 6 66± 6 WBC (10 9 /l) ± 3 83 ± 3 83± ± 4 64 ± 4 64± ± 6 83 ± 6 83± ± 7 62 ± 7 62± 7 Lineage B lineage ± 2 75 ± 2 75± 2 T lineage ± 7 74 ± 7 74± 7 CNS leukemia CNS ± ± ± 35 Other ± 2 75 ± 2 75± 2 Sex Male ± 3 70 ± 3 70± 3 Female ± 3 81 ± 3 81± 3 t(9;22) present absent ± 2 76 ± 2 76± 2 DFCI risk group Standard ± 3 77 ± 4 77± 4 High/very high ± 3 73 ± 3 73± 3 Methotrexate High-dose ± 3 77 ± 3 77 ± 3 Low-dose ± 3 73 ± 3 73 ± 3 Median follow-up: 9.0 years. Discussion We have demonstrated 5-year EFS rates ranging from 74 to 83% on four consecutive DFCI ALL consortium protocols conducted between 1981 and During that time, the backbone of the treatment regimen remained relatively unchanged, including a four- to five-agent remission induction, high-dose weekly asparaginase intensification for all patients, cranial radiation for the majority of patients, doxorubicin during intensification for HR patients (up to a cumulative dose of mg/m 2 ), and frequent pulses of vincristine and corticosteroid during post-remission therapy. Remission death rates remained constant throughout the treatment era (2 3.6%), and were primarily due to infectious complications. The cumulative incidence of second malignancies on DFCI ALL consortium protocols for all patients treated between 1972 and 1995 was 2.7%, 19 comparable to others 20,21 and lower than on epipodophyllotoxin-containing regimens. 22,23 The cumulative incidence of testicular relapses amongst male patients ( %) was also quite low. The most intensive therapy was reserved for HR and VHR patients. The outcome of subsets of HR/VHR patients, such as those with T cell immunophenotype (5-year EFS ranging from 70 to 79%) and adolescents (5-year EFS 70 to 79%), compared

7 2253 Figure 3 Event-free survival, cumulative incidence of any CNS relapse and of isolated CNS relapse for 369 patients treated on protocol ( ). Median follow-up was 9.0 years. Figure 4 Event-free survival, cumulative incidence of any CNS relapse and of isolated CNS relapse for 377 patients treated on protocol ( ). Median follow-up was 4.9 years. favorably to contemporaneously conducted clinical trials EFS for all HR/VHR patients improved between 1981 and 1995 (5-year EFS 66% on protocol vs 81% on 91-01). The reasons for this improvement may be multifactorial, including: (1) improved outcome for subsets of HR/VHR patients. We have previously reported that the addition of high-dose cytarabine and high-dose methotrexate to the regimen for infants (beginning in 1985) significantly improved EFS. 9 Similarly, relapse rates for patients with t(9;22) decreased after such patients were routinely treated with allogeneic stem cell transplants during first remission beginning in (2) Intensification of therapy for all HR/VHR patients, such as the use of dexamethasone and a more prolonged asparaginase intensification on protocol 91-01; and (3) improvements in supportive care, such as the initiation of rou- tine Pneumocystis carinii pneumonia prophylaxis in Because HR outcomes have improved, we have decreased intensity for some of these patients on our current protocols by adopting NCI age and WBC criteria to risk-stratify patients. 29 With these new criteria, 20% of patients who previously would have been considered HR are treated on the less intensive SR arm on current DFCI ALL consortium protocols. The 5-year EFS for SR patients remained relatively unchanged during this treatment era (5-year EFS 87 89%), with the exception of protocol (5-year EFS 78%). The inferior SR EFS on protocol was due to excessive CNS relapses in SR boys who were treated without cranial radiation (data not shown), consistent with the finding that males with B-precursor ALL fare worse than girls, and may require more

8 2254 Table 6 Protocol treatment results according to presenting features ( ) Presenting feature No. of 5-year 8-year 10-year patients EFS EFS EFS (±s.e.) (±s.e.) (±s.e.) Overall ± 2 N/A N/A Non-T lineage Standard ± 2 N/A N/A High ± 4 T Lineage Standard ± 0 N/A N/A High ± 9 Age at diagnosis 1 year 7 71 ± 17 N/A N/A 1 9 years ± 2 10 years ± 5 WBC (10 9 /l) ± 3 N/A N/A ± ± ± 6 Lineage B lineage ± 2 N/A N/A T lineage ± 8 CNS leukemia CNS ± 22 N/A N/A Other ± 2 Sex Male ± 3 N/A N/A Female ± 3 t(9;22) present 6 50 ± 20 N/A N/A absent ± 2 DFCI risk group Standard ± 3 N/A N/A High/very high ± 3 6-MP High-dose ± 3 N/A N/A Low-dose ± 3 Median follow-up: 4.9 years. N/A, not available due to insufficient follow-up. intensive therapy. 30 Other than protocol 87-01, sex has not been a significant prognostic factor on DFCI ALL consortium protocols; however, based upon the results of protocol 87-01, cranial radiation was re-instituted for SR boys (but not girls) on the subsequent protocol, Others have successfully eliminated cranial radiation in lower-risk patients by substituting it with high-dose antimetabolite therapy and/or intensive intrathecal chemotherapy, changes that were not made when radiation was eliminated on protocol On our current protocol 95-01, we are comparing the relative efficacy and toxicity of intensive intrathecal therapy (without radiation) and 1800 cgy cranial radiation in SR boys. With the exception of SR boys on protocol 87-01, the cumulative incidence of CNS relapses remained low throughout the treatment era despite decreasing intensity of CNSdirected therapy. The dose of cranial radiation for HR patients decreased from 2800 to 1800 cgy, and radiation was eliminated altogether for SR girls. In an attempt to reduce the potential late effects of cranial radiation, we compared two radiation fractionation schedules (hyperfractionated and conventional); longer follow-up is needed to determine relative late effects. Others have utilized alternative CNS treatment strategies, including the use of high-dose antimetabolite therapy, intensive intrathecal chemotherapy and lower radiation doses (12 Gy) While such treatments may offer comparable efficacy as 1800 cgy cranial radiation, the relative long-term neuropsychological sequelae of the various CNS treatment strategies remains unsettled. We have begun exploring the effect of various systemic therapies on neurocognitive status in long-term survivors, and have found that patients who received both high-dose methotrexate and cranial radiation had more marked impairments than those who received either therapy alone. 8 Also, patients who received dexamethasone during post-remission therapy were at increased risk for developing neurocognitive late effects compared with those who received prednisone. 36 To improve EFS for all patients, we conducted two randomized trials studying high-dose methotrexate during remission induction (protocol 87-01) and high-dose i.v. 6-MP during intensification (protocol 91-01). On protocol 81-01, we had observed a trend toward improved EFS in patients randomized to high-dose methotrexate during the investigational window (5-year EFS high-dose vs low-dose: 82% vs 72%, P = 0.10). That randomization included only 77 patients and had not been designed to detect differences in EFS; therefore, all patients on protocol were randomized to receive either high-dose or low-dose methotrexate during remission induction. In this larger trial, we did not observe a significant difference in EFS between the two arms (P = 0.48). Similarly, on protocol 91-01, we did not observe any improvement in EFS associated with the use of high-dose i.v. 6-MP during the first year of post-remission therapy (P = 0.90). These findings are in contradistinction to other reports which have suggested that high doses of methotrexate and/or 6-MP were associated with an improved outcome, 37,38 and suggests that high-dose antimetabolite therapy is less beneficial within the context of the asparaginase- and anthracycline-based DFCI ALL consortium regimens. Prognostic factors on DFCI ALL consortium protocols have included DFCI risk group, age, and presenting WBC. However, in our most recently completed protocol (91-01), none of these factors, including risk group, were prognostically significant, suggesting the need to identify novel risk factors. We have previously reported the poor prognosis of patients with persistent leukemia at the end of 1 month of therapy and have altered our definition of induction failure to include such patients. 10 Others have reported on the prognostic significance of more sensitive early response measures, including marrow response after 7 14 days of multi-agent remission induction, 39 peripheral blast counts after 1 week of corticosteroids, 26 and minimal residual disease levels during and at the completion of remission induction therapy. 40,41 We are currently investigating early response and minimal residual disease measures within the context of DFCI ALL consortium protocols. 42 We are also conducting a prospective study to determine the prognostic significance of TEL/AML-1 gene fusion, 43,44 and attempting to identify other biologically distinct leukemia subtypes. Reduction in late effects of therapy was a major focus of the protocols conducted between 1981 and As noted above, longer follow-up is necessary to determine whether hyperfractionated radiation was associated with a lower incidence of neurocognitive deficits. Similarly, the results of the doxorubicin randomization (continuous infusion vs bolus) conducted on protocol also await longer follow-up. Preliminary results suggest that continuous infusion doxorubicin may not exert a substantial cardioprotective effect in this patient population. 45 On our current protocol, we have low-

9 ered the cumulative dosage of doxorubicin for high-risk patients to 300 mg/m 2, and are testing in a randomized trial whether a cardioprotectant agent, dexrazoxane, reduces the incidence of late cardiac dysfunction. In addition to mitigating the toxicities from currently available therapies, we are also exploring the development of novel therapies, such as the stimulation of a patient s own leukemia-specific immunity, 46 which might improve efficacy while avoiding some of the toxicities associated with conventional chemotherapy. In summary, approximately 75% of patients with newly diagnosed ALL treated on DFCI ALL consortium protocols between 1981 and 1995 are long-term, event-free survivors. To improve upon these results, our current studies focus on identifying biologically distinctive subsets of patients, optimizing available therapies and developing novel treatments. Our goal is to define patient-specific treatment regimens that will provide the least toxic, most efficacious paths to cure. Acknowledgements This study was supported in part by grants from the National Institute of Health (CA and CA 06516). We thank the patients, families, physicians, nurses, data managers and all others who participated in these trials. We acknowledge the fundamental contributions of Molly Schwenn MD, as well as Kristin Barrett, Mia Donnelly, Jennifer Peppe, Joyce Su, Sharon Thornhill, Stacy Waters, Mary Young and Guangyong Zou. References 1 Sallan SE, Camitta BM, Cassady JR, Nathan DG, Frei E. Intermittent combination chemotherapy with adriamycin for childhood acute lymphoblastic leukemia: clinical results. Blood 1978; 51: Sallan SE, Hitchcock-Bryan S, Gelber R, Cassady JR, Frei E III, Nathan DG. Influence of intensive asparaginase in the treatment of childhood non-t cell acute lymphoblastic leukemia. Cancer Res 1983; 43: Clavell LA, Gelber RD, Cohen HJ, Hitchcock-Bryan S, Cassady JR, Tarbell NJ, Blattner SR, Tantravahi R, Leavitt P, Sallan SE. Fouragent induction and intensive asparaginase therapy for treatment of childhood acute lymphoblastic leukemia. New Engl J Med 1986; 315: Schorin MA, Blattner S, Gelber RD, Tarbell NJ, Donnelly M, Dalton V, Cohen HJ, Sallan SE. Treatment of childhood acute lymphoblastic leukemia: results of Dana-Farber Cancer Institute/Children s Hospital Acute Lymphoblastic Consortium Protocol J Clin Oncol 1994; 12: Halton JM, Atkinson SA, Fraher L, Webber C, Gill GJ, Dawson S, Barr RD. Altered mineral metabolism and bone mass in children during treatment for acute lymphoblastic leukemia. J Bone Miner Res 1996; 11: Lipshultz SE, Colan SD, Gelber RD, Perez-Atayde AR, Sallan SE, Sanders SP. Late cardiac effects of doxorubicin therapy for acute lymphoblastic leukemia in childhood. New Engl J Med 1991; 324: Strauss AJ, Su JT, Kimball-Dalton VM, Gelber RD, Sallan SE, Silverman LB. Increased corticosteroid-induced bony morbidity in older children with acute lymphoblastic leukemia. Proc Am Soc Clin Oncol 2000; 19: 583a. 8 Waber DP, Tarbell NJ, Fairclough D, Atmore K, Castro R, Isquith P, Lussier F, Romero I, Carpenter PJ, Schiller M, Sallan SE. Cognitive sequelae of treatment in childhood acute lymphoblastic leukemia: cranial radiation requires an accomplice. J Clin Oncol 1995; 13: Silverman LB, McLean TW, Gelber RD, Donnelly MJ, Gilliland DG, Tarbell NJ, Sallan SE. Intensified therapy for infants with acute lymphoblastic leukemia: results from the Dana-Farber Cancer Institute consortium. Cancer 1997; 80: Silverman LB, Gelber RD, Young ML, Dalton VK, Barr RD, Sallan SE. Induction failure in acute lymphoblastic leukemia of childhood. Cancer 1999; 85: Asselin BL, Whitin JC, Coppola DJ, Rupp IP, Sallan SE, Cohen HJ. Comparative pharmacokinetic studies of three asparaginase preparations. J Clin Oncol 1993; 11: Asselin BL, Kreissman S, Coppola DJ, Bernal SD, Leavitt PR, Gelber RD, Sallan SE, Cohen HJ. Prognostic significance of early response to a single dose of asparaginase in childhood acute lymphoblastic leukemia. J Pediatr Hematol Oncol 1999; 21: Schwartz C, Thompson B, Chilton D, Gelber R, Cohen H, Sallan S. Preliminary analysis of DFCI steroid window. Proc Am Soc Clin Oncol 1995; 14: Niemeyer CM, Reiter A, Riehm H, Donnelly M, Gelber RD, Sallan SE. Comparative results of two intensive treatment programs for childhood acute lymphoblastic leukemia: the Berlin Frankfurt Munster and Dana-Farber Cancer Institute protocols. Ann Oncol 1991; 2: Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: Greenwood M. The natural duration of cancer. Reports on Public Health and Medical Subjects 33. Her Majesty s Stationery Office: London, Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1996; 50: Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. John Wiley: New York, Kimball Dalton VM, Gelber RD, Li F, Donnelly MJ, Tarbell NJ, Sallan SE. Second malignancies in patients treated for childhood acute lymphoblastic leukemia. J Clin Oncol 1998; 16: Neglia JP, Meadows AT, Robison LL, Kim TH, Newton WA, Ruymann FB, Sather HN, Hammond GD. Second neoplasms after acute lymphoblastic leukemia in childhood. New Engl J Med 1991; 325: Loning L, Zimmermann M, Reiter A, Kaatsch P, Henze G, Riehm H, Schrappe M. Secondary neoplasms subsequent to Berlin Frankfurt Munster therapy of acute lymphoblastic leukemia in childhood: significantly lower risk without cranial radiotherapy. Blood 2000; 95: Winick NJ, McKenna RW, Shuster JJ, Schneider NR, Borowitz MJ, Bowman WP, Jacaruso D, Kamen BA, Buchanan GR. Secondary acute myeloid leukemia in children with acute lymphoblastic leukemia treated with etoposide. J Clin Oncol 1993; 11: Pui CH, Ribeiro RC, Hancock ML, Rivera GK, Evans WE, Raimondi SC, Head DR, Behm FG, Mahmoud MH, Sandlund JT, Crist WM. Acute myeloid leukemia in children treated with epipodophyllotoxins for acute lymphoblastic leukemia. New Engl J Med 1991; 325: Shuster JJ, Falletta JM, Pullen DJ, Crist WM, Humphrey GB, Dowell BL, Wharam MD, Borowitz M. Prognostic factors in childhood T cell acute lymphoblastic leukemia: a Pediatric Oncology Group study. Blood 1990; 75: Rivera GK, Raimondi SC, Hancock ML, Behm FG, Pui CH, Abromowitch M, Mirro J Jr, Ochs JS, Look AT, Williams DL, Murphy SB, Dahl GV, Kalwinsky DK, Evans WE, Kun LE, Simone JV, Crist WM. Improved outcome in childhood acute lymphoblastic leukaemia with reinforced early treatment and rotational combination chemotherapy. Lancet 1991; 337: Reiter A, Schrappe M, Ludwig WD, Hiddemann W, Sauter S, Henze G, Zimmermann M, Lampert F, Havers W, Niethammer D, Odenwald E, Ritter J, Mann G, Welte K, Gadner H, Riehm H. Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients. Results and conclusions of the multicenter trial ALL-BFM 86. Blood 1994; 84: Janka-Schaub GE, Harms D, Goebel U, Graubner U, Gutjahr P, Haas RJ, Juergens H, Spaar HJ, Winkler K. Randomized comparison of rotational chemotherapy in high-risk acute lymphoblastic leukaemia of childhood follow-up after 9 years. Coall Study Group. Eur J Pediatr 1996; 155: Arico M, Valsecchi MG, Camitta B, Schrappe M, Chessells J, Baruchel A, Gaynon P, Silverman L, Janka-Schaub G, Kamps W, Pui CH, Masera G. Outcome of treatment in children with Philadel- 2255

10 2256 phia chromosome-positive acute lymphoblastic leukemia. New Engl J Med 2000; 342: Smith M, Arthur D, Camitta B, Carroll AJ, Crist W, Gaynon P, Gelber R, Heerema N, Korn EL, Link M, Murphy S, Pui CH, Pullen J, Reamon G, Sallan SE, Sather H, Shuster J, Simon R, Trigg M, Tubergen D, Uckun F, Ungerleider R. Uniform approach to risk classification and treatment assignment for children with acute lymphoblastic leukemia. J Clin Oncol 1996; 14: Shuster JJ, Wacker P, Pullen J, Humbert J, Land VJ, Mahoney DH Jr, Lauer S, Look AT, Borowitz MJ, Carroll AJ, Camitta B. Prognostic significance of sex in childhood B-precursor acute lymphoblastic leukemia: a Pediatric Oncology Group Study. J Clin Oncol 1998; 16: Tubergen DG, Gilchrist GS, O Brien RT, Coccia PF, Sather HN, Waskerwitz MJ, Hammond GD. Prevention of CNS disease in intermediate-risk acute lymphoblastic leukemia: comparison of cranial radiation and intrathecal methotrexate and the importance of systemic therapy: a Childrens Cancer Group report. J Clin Oncol 1993; 11: Pullen J, Boyett J, Shuster J, Crist W, Land V, Frankel L, Iyer R, Backstrom L, van Eys J, Harris M, Ravindranath Y, Sullivan M. Extended triple intrathecal chemotherapy trial for prevention of CNS relapse in good-risk and poor-risk patients with B-progenitor acute lymphoblastic leukemia: a Pediatric Oncology Group study. J Clin Oncol 1993; 11: Conter V, Arico M, Valsecchi MG, Rizzari C, Testi AM, Messina C, Mori PG, Miniero R, Colella R, Basso G, Rondelli R, Pession A, Masera G. Extended intrathecal methotrexate may replace cranial irradiation for prevention of CNS relapse in children with intermediate-risk acute lymphoblastic leukemia treated with Berlin Frankfurt Munster-based intensive chemotherapy. The Associazione Italiana di Ematologia ed Oncologia Pediatrica. J Clin Oncol 1995; 13: Pui CH, Mahmoud HH, Rivera GK, Hancock ML, Sandlund JT, Behm FG, Head DR, Relling MV, Ribeiro RC, Rubnitz JE, Kun LE, Evans WE. Early intensification of intrathecal chemotherapy virtually eliminates central nervous system relapse in children with acute lymphoblastic leukemia. Blood 1998; 92: Schrappe M, Reiter A, Henze G, Niemeyer C, Bode U, Kuhl J, Gadner H, Havers W, Pluss H, Kornhuber B, Zintl F, Ritter J, Urban C, Niethammer D, Riehm H. Prevention of CNS recurrence in childhood ALL: results with reduced radiotherapy combined with CNS-directed chemotherapy in four consecutive ALL-BFM trials. Klin Padiatr 1998; 210: Waber DP, Carpentieri SC, Klar N, Silverman LB, Schwenn M, Hurwitz CA, Mullenix PJ, Tarbell NJ, Sallan SE. Cognitive sequelae in children treated for acute lymphoblastic leukemia with dexamethasone or prednisone. J Pediatr Hematol Oncol 2000; 22: Abromowitch M, Ochs J, Pui CH, Fairclough D, Murphy SB, Rivera GK. Efficacy of high-dose methotrexate in childhood acute lymphocytic leukemia: analysis by contemporary risk classifications. Blood 1988; 71: Camitta B, Mahoney D, Leventhal B, Lauer SJ, Shuster JJ, Adair S, Civin C, Munoz L, Steuber P, Strother D, Kamen BA. Intensive intravenous methotrexate and mercaptopurine treatment of higher-risk non-t, non-b acute lymphocytic leukemia: a Pediatric Oncology Group study. J Clin Oncol 1994; 12: Gaynon PS, Desai AA, Bostrom BC, Hutchinson RJ, Lange BJ, Nachman JB, Reaman GH, Sather HN, Steinherz PG, Trigg ME, Tubergen DG, Uckun FM. Early response to therapy and outcome in childhood acute lymphoblastic leukemia: a review. Cancer 1997; 80: Cave H, van der Werff ten Bosch J, Suciu S, Guidal C, Waterkeyn C, Otten J, Bakkus M, Thielemans K, Grandchamp B, Vilmer E. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer Childhood Cooperative Group. New Engl J Med 1998; 339: Panzer-Grumayer ER, Schneider M, Panzer S, Fasching K, Gadner H. Rapid molecular response during early induction chemotherapy predicts a good outcome in childhood acute lymphoblastic leukemia. Blood 2000; 95: Donovan JW, Ladetto M, Zou G, Neuberg D, Poor C, Bowers D, Gribben JG. Immunoglobulin heavy-chain consensus probes for real-time PCR quantification of residual disease in acute lymphoblastic leukemia. Blood 2000; 95: McLean TW, Ringold S, Neuberg D, Stegmaier K, Tantravahi R, Ritz J, Koeffler HP, Takeuchi S, Janssen JW, Seriu T, Bartram CR, Sallan SE, Gilliland DG, Golub TR. TEL/AML-1 dimerizes and is associated with a favorable outcome in childhood acute lymphoblastic leukemia. Blood 1996; 88: Loh ML, Silverman LB, Young ML, Neuberg D, Golub TR, Sallan SE, Gilliland DG. Incidence of TEL/AML1 fusion in children with relapsed acute lymphoblastic leukemia. Blood 1998; 92: Lipshultz SE, Sallan SE, Giantris AL, Lipsitz SR, Dalton V, Colan SD. Forty-eight hour continuous infusion doxorubicin infusion is not cardioprotective in children assessed 18 months later: the DFCI ALL protocol. Proc Am Soc Clin Oncol 1998; 17: 528a. 46 Cardoso AA, Veiga JP, Ghia P, Afonso HM, Haining WN, Sallan SE, Nadler LM. Adoptive T cell therapy for B cell acute lymphoblastic leukemia: preclinical studies. Blood 1999; 94: