Long-Term Follow-Up of Short Intensive Multiagent Chemotherapy

Home , Asparaginase

Leukemia (2002) 16, 594–600  2002 Nature Publishing Group All rights reserved 0887-6924/02 $25.00 www.nature.com/leu Long-term follow-up of short intensive multiagent chemotherapy without high-dose methotrexate (‘Orange’) in children with advanced non-lymphoblastic non-Hodgkin’s lymphoma:a Children’s Cancer Group Report MS Cairo1, MD Krailo2, M Morse3, RJ Hutchinson4, RE Harris5, CR Kjeldsberg6, ME Kadin7, E Radel8, LJ Steinherz9, E Morris1, JL Finlay10 and AT Meadows11

1Children’s Hospital of New York, Columbia University, New York, NY, USA; 2Keck School of Medicine, University of Southern California, Los Angeles, CA, USA; 3Children’s Hospital of Denver, Denver, CO, USA; 4Mott Children’s Hospital, University of Michigan, Ann Arbor, MI, USA; 5Children’s Hospital Medical Center, University of Cincinnati, Cincinnati, OH, USA; 6University of Utah Medical Center, Salt Lake City, UT, USA; 7Beth Israel Hospital, Boston, MA, USA; 8Monteﬁore Medical Center, Bronx, NY, USA; 9Memorial Sloan-Kettering Cancer Center, New York, NY, USA; 10New York University, New York, NY, USA; and 11Children’s Hospital of Philadelphia, Philadelphia, PA, USA

Despite prolonged therapy (18 months), children with advanced erson et al2 demonstrated that COMP continued to be signifi- non-lymphoblastic, non-Hodgkin’s lymphoma (NHL) treated on cantly better than LSA -L for undifferentiated lymphoma previous Children’s Cancer Group (CCG) trials achieved less 2 2 than a 60% 5-year event-free survival (EFS). In this study we (Burkitt’s and non-Burkitt’s). In an attempt to improve on the piloted a shorter but more intensive protocol (‘Orange’) to results of the previous study (CCG-551), CCG initiated a ran- determine the feasibility, safety, and efficacy of this alternative domized, group-wide study testing whether the addition of treatment approach. Thirty-nine children received a CHOP- daunomycin to COMP was more effective than COMP alone based induction, etoposide/ifosfamide consolidation, DECAL in the treatment of disseminated non-lymphoblastic, non- (dexamethasone, etoposide, cisplatin, cytosine arabinoside Hodgkin’s lymphoma (CCG-503). This protocol utilized pro- (Ara-C) and L-asparaginase) intensification, and either one or two similar but less intense maintenance courses. Patients longed and less intense therapy over an 18-month period. were stratified to standard-risk (5 months) vs high-risk However, Chilcote et al3 demonstrated that the addition of (7 months) treatment. High risk was defined as either bone mar- daunomycin to COMP (D-COMP) did not improve the 5-year row disease, CNS disease, mediastinal mass у one-third tho- EFS (56 ± 4% vs 56 ± 5%) (D-COMP vs COMP). у racic diameter at T5 and/or LDH 2 times institutional upper A subsequent series of CCG pilot studies led to the develop- limits of normal. All other patients were considered to be standard risk. Results were compared with the previous CCG NHL ment of this pilot phase II study (‘Orange’ regimen). Finlay study (CCG-503). Sixteen and 23 patients were considered et al4 treated 68 children with advanced non-lymphoblastic standard- vs high-risk, respectively. The 5-year EFS and overall lymphoma with a three-cycle induction lasting 9 weeks survival (OS) were 77 ± 7% and 80 ± 7%, respectively. The 5- (cyclophosphamide, adriamycin, vincristine, prednisone) year EFS and OS were significantly better in the standard- vs (intensive CHOP), intensive consolidation and intrathecal tri- high-risk subgroups (100% vs 61 ± 11%) (P Ͻ 0.003) and (100% vs 65 ± 11%) (P Ͻ 0.01), respectively. Lactate dehydrogenase ple chemotherapy, and five 9-week courses of maintenance (LDH) у2 × normal (NL) was associated with significantly therapy. The intensive CHOP induction utilized in that study poorer outcomes (LDH у2 × NL vs Ͻ2 × NL) (5-year EFS: 55 ± resulted in a 97% complete response rate (CR).4 Results of 12% vs 100%) (P Ͻ 0.0004). This CCG hybrid regimen, ‘Orange’, CCG-8605, a retrieval protocol designed for children with of short and more intensive therapy resulted in a significant recurrent solid tumors and/or lymphomas, demonstrated that improvement in outcomes compared with the previous CCG trial of more prolonged but less intense therapy. This regimen the combination of ifosfamide and etoposide was an active that deletes high-dose methotrexate, if confirmed in a larger combination in the reinduction therapy of recurrent/refractory trial, could be considered as an alternative treatment approach non-Hodgkin’s lymphoma (NHL) (J Miser, CCG-8605 Final in children without high tumor burdens (LDH Ͻ2 × NL) and Report, November 1997). Children with recurrent large cell Murphy stage III disease. lymphoma treated on this retrieval regimen had 20% and 40% Leukemia (2002) 16, 594–600. DOI: 10.1038/sj/leu/2402402 CR and PR rates, respectively. Children with recurrent small Keywords: children; non-lymphoblastic lymphoma; large cell lymphoma; Burkitt’s lymphoma; short intensive chemotherapy non-cleaved lymphoma had 11% and 55% CR and PR rates, respectively, with this retrieval regimen. More recently, an alternative combination of chemotherapy was piloted for Introduction recurrent childhood NHL (CCG-5912) (DECAL: dexame- thesone, etoposide, cisplatin, cytosine arabinoside and L- + The treatment for advanced childhood non-lymphoblastic, asparaginase). This regimen resulted in CR PR rates in chil- non-Hodgkin’s lymphoma has changed dramatically over the dren with recurrent large cell lymphoma of 44% and 16%, past 20 years. Anderson et al1 initially reported successful respectively, and in recurrent small non-cleaved lymphoma of 5 results with the use of a four-drug regimen (COMP: cyclophos- 23% and 15%, respectively. phamide, vincristine, methotrexate, and prednisone) in a Chil- Based on the results of these three pilot studies, we devised dren’s Cancer Group (CCG) study (CCG-551). In this original a CCG hybrid regimen (‘Orange’) to incorporate an intensive report, they demonstrated that COMP was significantly better CHOP-based induction (CCG-552), ifosfamide/etoposide (VP- than a 10-drug combination (modified LSA2-L2) (57% vs 28% 16) consolidation (CCG-8605), and DECAL intensification 2-year event-free survival (EFS). With a longer follow-up, And- (CCG-5912) for short but more intense therapy (5 months in standard-risk and 7 months in high-risk) in children with advanced non-lymphoblastic NHL. In particular, this protocol Correspondence: MS Cairo, PO Box 60012, Arcadia, CA 91066-6012, omitted high-dose methotrexate (HDMTX) because of poten- USA; Fax: 626/445-4334 Presented in part at the American Society of Hematology (ASH), San tial added co-morbidity to the incidence and severity of tumor Francisco, CA, 2000. lysis syndrome, which often accompanies childhood non-lym- Received 16March 2001; accepted 13 December 2001 phoblastic lymphoma at presentation. We report the long-term Childhood non-lymphoblastic NHL intensive TX MS Cairo et al 595 follow-up with respect to feasibility, safety, and long-term out- Table 1 Orange regimen (CCG hybrid) come of this pilot phase II group-wide CCG trial (CCG-5911). Induction Cyclophosphamide 1200 mg/m2/d days 0, 1 (i.v.) Vincristine 2.0 mg/m2/d days 0, 7, 14 (i.v.) Materials and methods Doxorubicin 50 mg/m2/d day 0 (i.v.) Methylprednisolone 250 mg/m2/q 6 h × 4 day 1 (i.v.) Patient selection Prednisolone 60 mg/m2/d days 2–21 (PO) IT MTX (SR) Age adjusted day 0 Patient entry for this pilot trial was open to all institutions of IT ARA-C + HC (HR)a Age adjusted days 0, 14, (CNS−) + a the CCG between December 1991 and December 1993. IRB IT ARA-C HC (HR) Age adjusted days 0, 3, 7, 10, 14 (CNS+) approval for the protocol was obtained by all participating institutions prior to patient entry and informed consent was Consolidation obtained from patients and/or parents according to local insti- Ifosfamide 1800 mg/m2/d days 0–4 (i.v.) tutional procedures. All newly diagnosed and untreated VP-16 100 mg/m2/d days 0–4 (i.v.) patients aged 21 years or less with biopsy-proven advanced IT MTX (SR) Age adjusted day 0 + a non-lymphoblastic NHL at diagnosis were eligible for the IT ARA-C HC (HR) Age adjusted days 0, 14 (HR CNS−) study. IT MTX, ARA-C, HC (HR)a Age adjusted days 0, 7, 14 (HR The eligibility criteria for advanced disease were similar but CNS+) not identical to stages III and IV Murphy classification6 but Cranial radiation (1800 cGy) (CNS+) had to include at least one of the following: nodal disease that had extended beyond two lymphatic adjacent regions; Intensification 2 intrathoracic (mediastinum or pleura) disease; incompletely Dexamethasone 10 mg/m /q 12 h days 0, 1 (p.o.) Ara-C 3 gm/m2/over 3 h/q 12 h days 0, 1 resected primary disease of the GI tract; blasts in a bone mar- (i.v.) row aspirate or a bone marrow biopsy positive for lymphoma; VP-16 100 mg/m2/over 3 h/q 12 h days 0, or presence of lymphoma cells with at least five white blood 1 (i.v.) cells (WBC)/mm3 in the cerebrospinal fluid (CSF). Submission L-ASP 25,000 IU/m2 day 1 (i.m.) of representative diagnostic slides was required. These slides, Cisplatin 90 mg/m2/over 6 h day 1 (i.v.) when received, were reviewed to confirm the diagnosis of IT MTX Age adjusted day 0 IT ARA-C + HC Age adjusted day 0 (HR CNS+ or non-lymphoblastic NHL. Both the institutional and review −) pathologists were asked to classify tumors according to the International Working Formulation.7 Maintenanceb At the time of study entry, patients were identified as having Cyclophosphamide 1.0 g/m2/d day 0 (i.v.) either high-risk or standard-risk disease. High-risk involved Vincristine 2.0 mg/m2/d day 0 (i.v.) 2 × any or all of the following characteristics: (1) bone marrow Methylprednisolone 250 mg/m /q 6 h 4 day 0 (i.v.) Doxorubicin 25 mg/m2/d day 0 (i.v.) involvement; (2) CNS involvement defined as the presence of 2 3 Prednisolone 60 mg/m /d days 1–4 (PO) lymphoma cells on cytospin with at least five WBC/mm in IT MTX Age adjusted day 0 the CSF or the presence of cranial neuropathy; (3) mediastinal IT ARA-C + HC Age adjusted day 0 (HR CNS+ or mass of at least one-third of the thoracic diameter at the level −) of T5 or thoracic node involvement by lymphoma; or (4) Ifosfamide 1800 mg/m2/d days 21–25 (i.v.) 2 serum lactate dehydrogenase (LDH) at least twice the insti- VP-16 100 mg/m /d days 21–25 (i.v.) Dexamethasone 10 mg/m2/q 12 h days 42, 43 (p.o.) tutional upper limit of normal. All other patients were con- ARA-C 3 gm/m2/3 h inf/q 12 h day 42 (i.v.) sidered to be standard-risk. Patients classified as high-risk VP-16 100 mg/m2/3 h inf/q 12 h day 42 received more intense CNS therapy and one additional course (i.v.) of maintenance. L-ASP 25000 IU/m2 day 43 (i.m.) Cisplatin 90 mg/m2/6 h inf day 43 (i.v.)

Treatment plan aAmendment to this schedule in March 1992. bStandard risk – one course/high risk (HR) – two courses. cCNS−, central nervous system negative; CNS+, central nervous Patients received CHOP-based induction therapy with cyclo- system positive. phosphamide plus mesna, vincristine, doxorubicin, methylprednisolone and prednisolone, and intrathecal therapy (Table Study design 1). Consolidation consisted of 5 days of ifosfamide, mesna support, etoposide, and intrathecal therapy. Patients with The study was initially designed as a feasibility study. Patients CNS-positive disease received cranial irradiation during con- were allocated within high- and standard-risk strata to receive solidation to a dose of 1800 cGy. Intensification consisted of ‘Orange’ therapy. The plan was to enroll approximately 35 dexamethasone, cytosine arabinoside (Ara-C), etoposide, L- patients on this treatment regimen and, if six or more deaths asparaginase, cisplatin, and intrathecal therapy. Maintenance in remission were noted on this regimen, enrollment was to therapy consisted of three maintenance courses in standard- be stopped and the treatment designated unacceptable with risk patients: first course – cyclophosphamide, vincristine, respect to toxicity. methylprednisolone, doxorubicin, and prednisolone; second course – ifosfamide plus mesna support and etoposide; third course – dexamethasone, Ara-C/etoposide, L-asparaginase, Cardiacevaluation and cisplatin with intrathecal therapy. Patients defined as being high-risk received the same three maintenance courses Serial cardiac evaluations were required before each anthra- repeated again (six total). cycline dose to detect any cardiotoxicity. Cardiac evaluations

Leukemia Childhood non-lymphoblastic NHL intensive TX MS Cairo et al 596 consisted of an electrocardiogram and either echocardiogram Table 2 Patient demographics or radionuclide cardiac angiogram (RNCA) with evaluation of systolic left ventricular performance. Cardiac evaluations were Age requested at baseline and before day 0 of each of the mainte- Median 8 (years) Range 2–20 (years) nance courses and at 4 and 12 months after completion of therapy. Sex: M/F 27/12 Stage (Murphy’s) III: 32 Methods of analysis IV: 5 LDH у2NL Toxicity and feasibility analysis: Each patient who entered 26 (67%) a particular course of therapy was evaluated for any grade III or IV toxicity as described in the National Cancer Institute Histology Large cell: 12 (NCI) common toxicity criteria. The time required to complete Small non-cleaved: 22 each of the courses was compared with the planned course Other: 5 length. The last course was omitted from the planned course BM/CNS involvement length analysis because the time at which the end of therapy 5:2 evaluation was to be made was left to the discretion of the treating physician.

Event-free and overall survival Survival analysis: All data reported by February 2000 were used in the analysis. EFS was taken to be the time from study There have been 10 patient events in the 39 assessable entry to relapse, progressive disease, documentation of resist- patients. There have been three toxic deaths, two sepsis and ant disease after the first course, death without evidence of one CNS toxicity, five relapses (four have died) and two sec- disease progression, second malignancy, or last contact.8 ondary leukemias (one death). The 5-year EFS and overall sur- Patients who experienced a relapse, progressive disease, had vival were estimated to be 77 ± 7% and 80 ± 7%, respectively documentation of resistant disease after the first course, died (Figure 1a and b). For the standard-risk vs high-risk subgroups, without evidence of disease progression, or had a second the 5-year EFS and overall survival were 100% vs 61 ± 11% malignancy were considered to have suffered an adverse (P Ͻ 0.003) (Figure 2) and 100% vs 65 ± 11% (P Ͻ 0.01), event; otherwise, the patient was censored at last contact. Sur- respectively. vival (S) was taken to be the time from study entry to death. A death, regardless of cause, was considered an event; otherwise, the patient was censored at last contact. Prognosticvariables Estimates of survival9 were obtained by the method of Kaplan and Meier. Significance levels for comparisons were The following variables were analyzed in a univariate analysis calculated using the log-rank statistic. Relative risks for (age, sex, Murphy stage, LDH level, histology, bone marrow adverse events and associated 95% confidence intervals were or CNS involvement) and demonstrated that LDH у2 × upper calculated from the proportional hazards regression model. normal institutional values was a poor prognostic factor (5-year EFS: 55 ± 12% vs 100%) (P Ͻ 0.0004) (Figure 3) (5- year OS: 60 ± 11% vs 100%) (P Ͻ 0.002). The 5-year OS was not significantly different by staging util- Results izing St Jude’s staging classification (stage III vs IV) (5-year OS: stage III vs IV – 82 ± 7% vs 60 ± 7%, respectively).6 Although Patient demographics the number of patients was small, there was no difference between patients with bone marrow involvement (M1 (у5% CCG-5911 was open for entry between December 1991 and blasts) and M2 (6–24% blasts) vs M3 (у25% blasts)) or CNS December 1993. positive vs negative (5-year OS: M1 and M2 vs M3) (83 ± 7% Forty-two (42) patients were entered on study to receive vs 50 ± 35%, respectively) (CNS+ vs CNS-) (50 ± 35% vs 81 ‘Orange’ therapy. Of these, three were ineligible: one patient ± 7%, respectively). had lymphoblastic lymphoma, one patient had acute lympho- cytic leukemia, and one patient received therapy prior to local IRB approval of the treatment protocol. The remaining 39 Pathology patients were included in the outcome analysis. Sixteen patients were considered to be standard-risk and 23 patients All histopathological material was classified by the insti- were considered high-risk. There were 27 males and 12 tutional pathologist according to the International Working females (2.2:1.0) and the median age was 8 years (range 2– Formulation. Twelve of the 39 cases reviewed by the insti- 20 years). Murphy staging classification included one each tutional pathologists were classified as large cell lymphomas stages I and II, 32 stage III patients, and five patients with stage (three anaplastic). Twenty-two of the 39 cases were classified IV disease.6 There were 12 patients with large cell lymphoma as small non-cleaved cell lymphoma and five otherwise not and 27 with small non-cleaved lymphoma; 67% of patients specified. Large cell lymphomas had outcomes similar to had LDH у2 times upper normal institutional values, five other histological types (5-year EFS, large cell lymphomas vs patients had bone marrow involvement, and two patients had other – 92 ± 8% vs 70 ± 9%, respectively) and 5-year OS 92 CNS disease (Table 2). ± 8 vs 74 ± 9%, respectively). The risk for adverse events with

Leukemia Childhood non-lymphoblastic NHL intensive TX MS Cairo et al 597

Figure 3 Event-free survival by serum lactate dehydrogenase at study entry: two or more times institutional normal vs less than two times institutional normal (P Ͻ 0.0004).

Burkitt’s type is similar to that for patients with other small non-cleaved histological subtypes (non-Burkitt’s). Central immunophenotyping and central cytogenetics were not man- dated on this study.

Toxicity

Three patients died in remission while being treated with the ‘Orange’ regimen. Despite administration of G-CSF, two children died of sepsis within 2 months of starting induction therapy. One patient with prior CNS disease died of CNS toxicity (spinal cord necrosis) during administration of intrathecal (IT) therapy. Following modification of IT therapy in March 1992, no further serious toxicities attributable to IT therapy occurred. The Orange regimen was amended to change the intrathecal therapy to high-risk patients. Triple IT therapy was amended Figure 1 Event-free survival (a) and overall survival (b) by the in induction to patients with high risk and no CNS disease method of Kaplan and Meier (n = 36). from days 0 and 2 to only Ara-C and hydrocortisone on days 0 and 14. In patients with CNS disease, triple IT therapy was decreased from days 0, 1, 2, 7, 10, 14 and 18 to days 0, 3, 7, 10 and 14. During consolidation, patients with high-risk disease but who were CNS negative had their triple IT therapy decreased from days 0, 7 and 14 to only IT Ara-C and hydrocortisone on days 0 and 14. Patients with high-risk disease and CNS positive had their triple IT therapy decreased from days 0, 3, 7, 10, 14 to days 0, 7 and 14. The percents of grades III or IV toxicities that were у10% are noted by treatment course in Table 3. Specifically, the incidences of grades III and IV renal toxicity during the induction phase were 2.6% and 0%, respectively. There was no incidence of grade III/IV renal toxicity or ototoxicity during the intensification regimen involving cisplatin in DECAL intensification. The average length and range of days during induction, consolidation, and intensification therapy were 24 (17– 33), 23 (18–40), and 24 (20–38), respectively. Follow-up cardiac evaluations revealed minimal abnormali- ties. Thirty-four patients had sequential follow-up studies. One patient (3%) had an abnormal fractional shortening (FS) on Echo (24%), and three patients had decreases of FS of less Figure 2 Event-free survival comparing standard-risk vs high-risk than 10 percentile points but remained in the normal range. stratification by the method of Kaplan and Meier (P Ͻ 0.003). The one patient with an abnormal FS on follow-up unfortu-

Leukemia Childhood non-lymphoblastic NHL intensive TX MS Cairo et al 598 Table 3 Percentage of patients with grade III or IV toxicity and selected supportive care regimens by treatment course

Induction % Consolidation % Intensiﬁcation % First maintenance % Second maintenance %

ANC р500/mm3 46 29 46 47 55 PLT р25 k/mm3 26 14 43 44 44 HgB р6.5 g/dl 10 22 31 35 30 SGPT у5N 10 3 0 6 10 Total parenteral nutrition 10 6 3 3 0 Parenteral antifungal 31 11 9 6 10

nately had no baseline echocardiogram to determine whether nated disease.10 This finding is further illustrated by the this abnormality was treatment related. differences in outcome of children with abdominal Murphy Two patients experienced secondary malignancies after stage III B-NHL and LDH у500 IU/l between the BFM-NHL enrollment on this study. Both were secondary leukemias 86and BFM-NHL 90 studies. In the BFM-NHL 86study, the classified as acute myelogenous leukemia and occurred only dose of methotrexate was only 500 mg/m2 compared to 5000 in the high-risk group of patients. One patient died and one mg/m2 in the BFM-NHL 90 study. The increase in the HDMTX was still alive at this reporting. Cytogenetics were unfortu- in this subgroup of patients in the BFM-NHL 90 study resulted nately not available. in the 6-year EFS significantly improved from 29 ± 14% to 81 ± 4% (P Ͻ 0.0001), respectively.10,11 Therefore, it appears that, in children with advanced non-lymphoblastic B-NHL Discussion and a high tumor burden, the use of HDMTX is critically important to enhance overall outcome. In this pilot phase II trial of children with advanced non-lym- In other comparisons, the Pediatric Oncology Group phoblastic lymphoma, patients were stratified into two risk (POG) demonstrated that utilizing ‘Total-B’ therapy groups: high risk (у2 times normal institutional LDH values, (fractionated cyclophosphamide and four courses of HDMTX bone marrow and/or CNS involvement) and standard risk (1 g/m2/course)) in children with stage III high-grade small (advanced disease but without elevated LDH, bone marrow non-cleaved lymphomas, the 3-year EFS was 79 ± 6%.12 To or CNS involvement). This risk stratification is similar to the achieve these results, however, four doses of HDMTX (1 g/m2) BFM-NHL 90 study recently reported by Reiter et al.10 In the were required. In comparison, in patients with more advanced BFM stratification, patients with advanced childhood B-NHL small non-cleaved disease (B-ALL or Murphy stage IV), Bow- were stratified into either R-2 (no or incomplete resection, man et al13 reported that Total-B therapy with HDMTX extra-abdominal lymphoma) or abdominal lymphoma and (1 g/m2) resulted in a 4-year EFS of 64 ± 13%, similar to our LDH Ͻ500 U/l. The R-2 group classification was similar to results but less than reported in the BFM-NHL 90 study, sug- our standard-risk classification. The BFM R-3 group was gesting more intensified HDMTX (5 g/m2) may be required in defined as no or incomplete resection, abdominal lymphoma this subset of patients. This is supported by the results of the and LDH у500 U/l, and/or bone marrow, CNS or multifocal French Pediatric Oncology Society (SFOP) LMB-89 study bone involvement. This R-3 group was similar in many reported by Patte et al,14 who utilized HDMTX (8 g/m2)in respects to our high-risk group. One of the major differences group C patients with either bone marrow and/or CNS in therapy, however, was the omission of high-dose metho- involvement and achieved a greater than 80% 3-year EFS. trexate (HDMTX) in the current CCG hybrid ‘Orange’ regimen These French results improved significantly for this subset of compared to the use of 5 g/m2 of HDMTX in R-2 and R-3 patients from the previous LMB-84 study, which utilized only subgroups of BFM-NHL 90. However, the current ‘Orange’ 3 g/m2 of HDMTX in patients with B-ALL and/or Murphy stage regimen that does not utilize methotrexate utilizes 18 g/m2 of IV disease and resulted in only a 3-year EFS of 68 ± 8%.15 ifosfamide and 1.6g/m 2 of etoposide in patients with stan- The incidence of grade III and IV renal toxicity during the dard-risk disease, and these doses of ifosfamide and VP-16are induction phase of this protocol was 2.6% and 0%, respect- currently higher than other pediatric regimens that utilize ively. This compares with an incidence of 21% of acute renal high-dose methotrexate for childhood non-lymphoblastic failure in children with advanced Burkitt’s lymphoma or B- lymphoma.10,11 ALL treated on Total-B therapy.13 In the current pediatric inter- In comparing the standard-risk CCG group and the R-2 BFM national B cell lymphoma study (FAB/LMB-95) (CCG-5961), group, the 5-year EFS was similar (100% vs 96 ± 2%, the incidence of acute renal failure requiring dialysis in respectively). Therefore, in this subgroup of children with patients with advanced Burkitt’s lymphoma or B-ALL is cur- non-lymphoblastic B non-Hodgkin’s lymphoma, Murphy rently at 18% (M Cairo, personal communication). stage III (incompletely resected) but low tumor burden (low Two patients (5%) in the high-risk subgroup treated on this or normal LDH), the outcome is both similar and superb protocol developed secondary acute myelogenous leukemia. (у95% 5-year EFS) utilizing either approach of the two co- They each received 2.3 g/m2 of etoposide (VP-16). Total-B operative groups.10 However, in the current CCG regimen, the therapy in the POG studies did not utilize VP-16and the BFM- use of high-dose methotrexate was omitted. In contrast, in the NHL 90 study utilized only a total of 600 mg/m2 (200 higher risk group in our current study and the R-3 risk group mg/m2/course × 3 courses) in their R-2 and R-3 subgroup of in the BFM-NHL 90 study, the 4-year EFS was only 55 ± 14% patients. These doses of VP-16are substantially less than those in the current study compared to a 6-year EFS of 78 ± 3% in used in our current study. These results suggest the need to the BFM-NHL 90 study.10 This comparison suggests the criti- utilize VP-16at lower doses similar to doses used in BFM- cal importance of HDMTX in children with higher tumor bur- NHL 9010 or possibly to delete VP-16in future studies. Fur- dens (LDH у500 IU/l or у2 times NL) and/or more dissemi- thermore, the role and importance of cisplatin in the intensi-

Leukemia Childhood non-lymphoblastic NHL intensive TX MS Cairo et al 599 fication ‘DECAL’ arm in this current regimen is also unclear. 2 Anderson JR, Jenkin RDT, Wilson JF, Kjeldsberg CR, Sposto R, Little is known about the efficacy of cisplatin as a single agent Chilcote RR, Coccia PF, Exelby PR, Siegel S, Meadows AT, Ham- in children with non-lymphoblastic NHL and it is not used in mond GD. Long-term follow-up of patients treated with COMP or LSA2L2 therapy for childhood non-Hodgkin’s lymphoma: a report other successful childhood non-lymphoblastic lymphoma of CCG-551 from the Childrens Cancer Group. J Clin Oncol 1993; regimens. 11: 1024–1032. Unfortunately, this pilot phase II trial did not mandate per- 3 Chilcote RR, Krailo M, Kjeldsberg C, Kadin M, Steinherz PG, formance of central immunophenotyping and/or cytogenetics. Coccia PF, Morse M, Reaman GH, Hammond GD. Daunomycin Only three patients had anaplastic large cell histology and plus COMP vs COMP therapy in childhood non-lymphoblastic therefore there were too few patients to analyze the effects of lymphomas. ProcAm SocClin Oncol 1991; 10: 289 (Abstr.). 4 Finlay JL, Anderson JR, Cecalupo AJ, Hutchinson RJ, Kadin ME, this regimen in this histological subtype. However, in this Kjeldsberg CR, Provisor AJ, Woods WG, Meadows AT. Dissemi- study and in the previous CCG study (503) in children with nated nonlymphoblastic lymphoma of childhood: a Childrens advanced non-lymphoblastic NHL, histological subtype com- Cancer Group study, CCG-552. Med Pediatr Oncol 1994; 23: parisons between large cell and small non-cleaved cell lym- 453–463. phomas demonstrated a trend for better outcome for large cell 5 Kobrinsky N, Anderson J, Delaat C, Gilchrist G, Meadows A, Miser lymphoma, although not statiscally different. J, Morse M, Shah N. Dexamethasone, etoposide, cisplatin, Ara-C and L-asparaginase (DECAL) for recurrent Hodgkin’s disease (HD) In summary, this pilot phase II trial of short but more inten- and non-Hodgkin’s lymphoma (NHL): a preliminary report from sive multiagent chemotherapy, without high-dose methotrex- the Children’s Cancer Group CCG-5912. ProcAm SocClin Oncol ate, was well tolerated in children with advanced non-lym- 1994; 13: 393 (Abstr.). phoblastic NHL. In a small subgroup with advanced disease 6Murphy SB. Classification, staging and end results of treatment of (Murphy stage III but without a high tumor burden) (LDH Ͻ2 childhood non-Hodgkin’s lymphoma: dissimilarities from lym- × NL) and without disseminated disease (bone marrow and/or phomas in adults. Semin Oncol 1980; 7: 332–339. 7 National Cancer Institute sponsored study of classification of non- central nervous system disease) (BM/CNS), the 5-year EFS was Hodgkin’s lymphomas: summary and description of a working for- 100%. These results in this small subgroup should be con- mulation for clinical usage. Cancer 1982; 49: 2112–2135. firmed in a larger series. If confirmed in a larger series, this 8 Kirkwood B. Essentials of Medical Statistics. Blackwell Scientific subgroup of patients with advanced disease but low tumor Publications: Palo Alto, 1988. burden may benefit from this alternative treatment approach. 9 Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time However, in children with more disseminated disease Data. John Wiley and Sons: New York, 1980. у × 10 Reiter A, Schrappe M, Tiemann M, Ludwig W-D, Yakisan E, Zim- (BM/CNS) and/or high tumor burden (LDH 2 NL), this mermann M, Mann G, Chott A, Ebell W, Klingebiel T, Graf N, approach that omits high-dose methotrexate is inferior to other Kremens B, Muller-Weihrich S, Pluss H-J, Zentl F, Henze G, pediatric NHL cooperative group results and therefore should Riehm H. Improved treatment results in childhood B-cell neo- be abandoned. The minimum dose, if any, of VP-16required plasms with tailored intensification of therapy: a report of the in this regimen needs to be explored in future investigations. Berlin–Frankfurt–Munster Group Trial NHL-BFM 90. Blood 1999; Based on the results of this study, the current CCG-5961 study 94: 3294–3306. 11 Reiter A, Schrappe M, Parwaresch R, Henze G, Muller-Weihrich is exploring the appropriate doses of Ara-C and VP-16in S, Sauter S, Sykora K-W, Ludwig W-D, Gadner H, Riehm H. Non- childhood B-NHL. Hodgkin’s lymphomas of childhood and adolescence: results of a treatment stratified for biologic subtypes and stage – a report of the Berlin–Frankfurt–Munster group. J Clin Oncol 1995; 13: 359–372. 12 Brecher ML, Schwenn MR, Coppes MJ, Bowman WP, Link MP, Acknowledgements Berard CW, Shuster JJ, Murphy SB. Fractionated cyclophosphamide and back to back high dose methotrexate and cytosine arabinoside improves outcome in patients with stage III high grade The authors would like to thank Linda Rahl, Shaun Mason, small non-cleaved cell lymphomas (SNCCL): a randomized trial and Lucia Noll for their editorial assistance in the preparation of the Pediatric Oncology Group. Med Pediatr Oncol 1997; 29: of this manuscript. Grant support from the Division of Cancer 526–533. Treatment, National Cancer Institute, National Institutes of 13 Bowman WP, Shuster JJ, Cook B, Griffin T, Behm F, Pullen J, Link Health, Department of Health and Human Services. Contribu- M, Head D, Carroll A, Berard C, Murphy S. Improved survival for ting Children’s Cancer Group investigators, institutions, and children with B-cell acute lymphoblastic leukemia and stage IV small noncleaved-cell lymphoma: a Pediatric Oncology Group grant numbers are given in the Appendix. study. J Clin Oncol 1996; 14: 1252–1261. 14 Patte C, Leverger G, Rubie H. High cure rate in B-cell (Burkitt’s) leukemia in the LMB 89 protocol of the SFOP (French Pediatric Oncology Society). ProcAm SocClin Oncol 1993; 12: 317a References (Abstr.). 15 Patte C, Philip T, Rodary C, Zucker J-M, Behrendt H, Gentet J-C, 1 Anderson JR, Wilson JF, Jenkin RDT, Meadows AT, Kersey J, Chil- Lamagnere J-P, Otten J, Dufillot D, Pein F, Caillou B, Lemerle J. cote RR, Coccia P, Exelby P, Kushner J, Siegel S, Hammond D. High survival rate in advanced-stage B-cell lymphomas and leuke- Childhood non-Hodgkin’s lymphoma. The results of a randomized mias without CNS involvement with a short intensive polychemo- therapeutic trial comparing a 4-drug regimen (COMP) with a 10- therapy: results from the French Pediatric Oncology Society of a drug regimen (LSA2-L2). N Engl J Med 1983; 308: 559–565. randomized trial of 216children. J Clin Oncol 1991; 9: 123–132.

Leukemia Childhood non-lymphoblastic NHL intensive TX MS Cairo et al 600 Appendix: Participating principal investigators – Children’s Cancer Group

Institution Investigators Grant No.

Group Operations Center, Arcadia, California W Archie Bleyer MD CA 13539 Anita Khayat PhD Harland Sather PhD Mark Krailo PhD Jonathan Buckley MBBS, PhD Daniel Stram PhD Richard Sposto PhD University of Michigan Medical Center, Ann Arbor, Michigan Raymond Hutchinson MD CA 02971 University of California Medical Center, San Francisco, California Katherine Matthay MD CA 17829 University of Wisconsin Hospital, Madison, Wisconsin Jousif Matloub MD CA 05436 Children’s Hospital and Medical Center, Seattle, Washington J Rusell Geyer MD CA 10382 Rainbow Babies and Children’s Hospital, Cleveland, Ohio Eric Kodish MD CA 20320 Children’s National Medical Center, Washington, DC Gregory Reaman MD CA 03888 Children’s Hospital of Los Angeles, Los Angeles, California Paul Gaynon MD CA 02649 Children’s Hospital of Columbus, Columbus, Ohio Frederick Ruymann MD CA 03750 Colubmia Presbyterian College of Physicians and Surgeons, New York Linda Granowetter MD CA 03526 Vanderbilt University School of Medicine, Nashville, Tennessee James Whitlock MD CA 26270 University of Minnesota Health Sciences Center, Minneapolis, Minnesota Joseph Neglia MD CA 07306 Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania Beverly Lange MD CA 11796 Memorial Slona-Kettering Cancer Center, New York, New York Peter Steinherz MD CA 42764 James Whitcomb Riley Hospital for Children, Indianapolis, Indiana Robert Fallon MD CA 13809 University of Utah Medical Center, Salt Lake City, Utah William Carroll MD CA 10198 University of British Columbia, Vancouver, Canada Paul Rogers MD CA 29013 Children’s Hospital Medical Center, Cincinnati, Ohio Robert Wells MD CA 26126 Harbor/UCLA and Miller Children’s Medical Center, Torrance/Long Beach, Jerry Finklestein MD CA 14560 California University of California Medical Center (UCLA), Los Angeles, California Stephen Feig MD CA 27678 University of Iowa Hospitals and Clinics, Iowa City, Iowa Raymond Tannous MD CA 29314 Mayo Clinic and Foundation, Rochester, Minnesota Carola Arndt MD CA 28882 Izaak Walton Killan Hospital for Children, Halifax, Canada Doroth Barnad MD — University of North Carolina, Chapel Hill, North Carolina Stuart Gold MD — University of Medicine and Dentistry of New Jersey, Camden, New Jersey Richard Drachtman MD — Children’s Mercy Hospital, Kansas City, Missouri Maxine Hetherington MD — Wyler Children’s Hospital, Chicago, Illinois James Nachman MD — M.D. Anderson Cancer Center, Houston, Texas Beverly Raney MD — Princess Margaret Hospital, Perth, Western Australia David Baker MD — New York University Medical Center, New York, New York Aaron Rausen MD — Childrens Hospital of Orange County, Orange, California Violet Shen MD —

Leukemia