Advanced-Stage Large-Cell Lymphoma in Children and Adolescents: Results of a Randomized Trial Incorporating Intermediate-Dose Methotrexate a
http://www.100md.com
▲還散笫雖悝◎
the Medical College of Virginia, Richmond, VA
Medical University of South Carolina, Charleston, SC
State University of New York, Syracuse
Roswell Park Memorial Institute, Buffalo, NY
Children's Oncology Group Statistical Office, Gainesville, FL
Maine Bureau of Health, Augusta, ME
Harvard Medical School
Massachusetts General Hospital, Boston, MA
University of Alberta, Edmonton, Alberta
Hospital for Sick Children, Toronto, Ontario, Canada
The University of Texas Health Science Center at San Antonio, TX
ABSTRACT
PATIENTS AND METHODS: From December 1994 to April 2000, we enrolled 180 eligible pediatric patients with stage III/IV large-cell lymphoma (LCL); 90 patients were randomly assigned to the intermediate-dose methotrexate (IDM) and high-dose cytarabine (HiDAC) arm, 85 patients to the APO arm, and five patients directly to the APO arm by study design due to CNS involvement. Planned therapy duration was 12 months.
RESULTS: The 4-year EFS for all patients was 67.4% (SE, 4.2%), and OS was 80.1% (SE, 3.6%) without any significant difference between the two arms. The 4-year EFS and OS were 71.8% (SE, 6.1%) and 88.1% (SE, 4.4%), respectively, for patients with anaplastic large-cell lymphoma, and 63.8% (SE, 10.3%) and 70.3% (SE, 9.0%), respectively, for patients with diffuse large B-cell lymphoma. Only 11 patients required radiation (due to unresponsive bulky disease or CNS involvement). The IDM/HiDAC arm was associated with more toxicity.
CONCLUSION: The efficacy of incorporating IDM/HiDAC in the treatment plan of pediatric and adolescent patients with advanced-stage LCL was inconclusive as to its effect on EFS, regardless of the lymphoma phenotype. It cannot be excluded that with a higher number of patients, one treatment could prove superior and future studies will build on these data.
INTRODUCTION
With the POG data and a modified APO regimen (POG #8615), we found an event-free survival (EFS) of 70% and concluded that the addition of cyclophosphamide did not improve outcome.6 In the current successor study (POG #9315), we used random assignment to assess effects of incorporating into the APO backbone, eight courses of intense antimetabolite therapy (intermediate-dose methotrexate [IDM] given over 24 hours, followed by high-dose cytarabine [HiDAC] by continuous infusion for 48 hours) on overall survival (OS) and EFS.
PATIENTS AND METHODS
Treatment Program
The program consisted of a similar induction and two different maintenance treatment plans and lasted for 12 months for both arms. Patients were randomly assigned upfront to one of the two maintenance arms presented in Figure 1. After the induction phase (doxorubicin 75 mg/m2 day 1 and 22, vincristine 1.5 mg/m2 day 1 and 22, prednisone 40 mg/m2 daily for 28 days, and age-adjusted intrathecal methotrexate on days 1, 8, and 22), patients were treated every 21 days with APO maintenance (doxorubicin 30 mg/m2 and vincristine 1.5 mg/m2 on day 1, prednisone 120 mg/m2 and 6-mercaptopurine 225/mg/m2 days 1 to 5) or APO maintenance alternating with IDM/HiDAC (IDM 1 g/m2 over 24 hours followed by cytarabine 500 mg/m2 bolus, and continuous infusion of cytarabine 60 mg/m2 over 48 hours) every 21 days. Both maintenance arms included intrathecal methotrexate on day one of maintenance cycles 1, 3, and 5, and doxorubicin was substituted with methotrexate after a cumulative dose of 300 mg/m2 was reached (after cycle 5 of the APO arm and cycle 10 of the IDM/HiDAC regimen); the dose of methotrexate was 60 mg/m2 per cycle. Patients with bulky disease at the end of induction (not achieving complete response [CR]; response criteria listed below) were eligible for radiation therapy after biopsy proved viable disease. The total dose to the prescription point was 4,140 cGy in 23 daily fractions (180 cGy/d). For patients in partial remission, most investigators chose to treat them with two cycles of maintenance therapy before performing a biopsy. CNS disease-positive patients were not randomly assigned and were assigned directly to the APO arm. During induction, CNS disease-positive patients received additional intrathecal methotrexate on days 15, 29, and 36, for a total of six intrathecal methotrexate doses during induction. Whole-brain irradiation was started during the first week of maintenance. The total radiation dose was 2,400 cGy in 16 daily fractions (150 cGy/d). These patients also received two additional doses of intrathecal methotrexate on day 1 of maintenance cycles 2 and 4, for a total of five intrathecal methotrexate doses during maintenance. Methotrexate was substituted for doxorubicin at a 33% dose reduction for patients who received cranial irradiation; the dose for these patients was 40 mg/m2 per cycle.
Immunohistochemistry
Immunohistochemistry analysis was performed with a panel of antibodies directed against lymphoid-associated antigens, effective in paraffin sections, and intended to identify lineage: B-lineage antibody L26 (CD20; Dako, Carpinteria, CA); T-lineage antibodies UCHL-1 (CD45Ro), CD3 (Dako), and MT1 (CD43; Biotest, Denville, NJ); Hodgkin's and ALCL antibody Ber-H2 (CD30; Biotest), ALCL-associated antibody ALK-1 (Dako), and histiocyte/macrophage-associated antibody KP1 (CD68; Dako). Antibodies were applied in this order when available material consisted of limited numbers of unstained slides. For cases in which unstained slides were not available, antibody results from treating institutions or reference laboratories of primary diagnosis were accepted if reviewers determined that they were appropriate.
Response Criteria
CR was defined as the disappearance of all evidence of disease from all sites. This was determined by physical examination and appropriate laboratory and imaging studies. Partial response (PR) was determined by at least a 50% decrease in tumor mass by comparing postinduction imaging to imaging at diagnosis. "No response" was defined as failure to achieve PR. The need for radiation therapy was determined at the end of induction (day 42); it could be used earlier only as an emergency intervention due to mediastinal disease with respiratory symptoms.
Statistical Methods
EFS, time from registration to earliest evidence of relapse, progressive disease, second cancer, or death from any cause (treatment failures) were the major end points of the study. The log-rank test was used in comparative analyses. We originally planned to accrue 237 patients. Assuming proportional hazards, the study had 80% power to detect a 2-year EFS rate of 0.87 compared with a 2-year EFS rate of 0.75 in controls, with a significance level of .05. Planned analyses included two interim analyses and one final analysis. EFS and OS curves were constructed by the Kaplan-Meier method with SEs of Peto et al.10,11 Due to the orphan nature of this disease, with relatively few patients, the reader should judge nonsignificant differences as inconclusive. All reported P values are two-sided.
RESULTS
Of 180 eligible patients enrolled, 90 were on the APO arm, including five with CNS involvement who were assigned to the arm by study design, and 90 patients were on the IDM/HiDAC arm (Table 1). There were two patients who had induction failures and two who died early, with a CR/PR rate of 98% on each arm (response rates are depicted in Table 2). Fifty-nine percent of patients presented with peripheral lymphadenopathy (cervical, supraclavicular, axillary, inguinal, or other site), 21% had hepatosplenomegaly, and 55% had mediastinal involvement (61% and 54% of ALCL and B-cell LCL patients, respectively, had mediastinal involvement). Seven percent of patients had skin involvement (10 of 12 had ALCL), and 24% had bone involvement (one-third had primary bone involvement). Forty-one percent of the whole group presented with "B" symptoms (fever > 101∼F or weight loss > 10%), in 47% of ALCL patients and 36% of B-cell LCL patients, respectively. Sixty-three percent had elevated lactate dehydrogenase (76% and 45% of B-cell LCL and ALCL patients, respectively, had increased lactate dehydrogenase).
Phenotype distribution is depicted in Table 3. There were eighty-six cases with ALCL (58 T-cell and 28 null cell phenotype), with 98% CD30+ and 89% ALK+; only 83 patients were randomly assigned, as three patients with CNS involvement were assigned to the APO arm. There were seventy-five cases were B-cell LCL, and 73 were randomized because of two patients having CNS involvement, who were assigned to the APO arm. There was one case of follicular lymphoma, and one of low-grade lymphoma of mucosa-associated tissue (MALT) with increased large cells (the remaining 71 being diffuse large B-cell lymphoma). Ten cases were classified as peripheral T-cell lymphoma, unspecified; none of these expressed CD 30 or ALK, but two expressed CD68. Nine cases did not have immunophenotyping and were categorized as LCL, unclassified.
The 4-year EFS for all patients was 67.4% (SE, 4.2%), and OS was 80.1% (SE, 3.6%; Fig 2). The EFS for IDM/HiDAC was almost identical to that of the APO arm (Fig 3; P = .96).
There were no significant differences in EFS between arms (P = .69, Fig 4) among the randomized diffuse large B-cell lymphoma patients (37 on IDM/HiDAC and 34 on APO); 4-year EFS for APO and IDM/HiDAC patients was 63.8% (SE, 10.3%) and 70.3% (SE, 9.0%), respectively. For randomized T-cell LCL patients (31 on the APO arm and 34 on the IDM/HiDAC arm), the 4-year EFS was 66.1% (SE, 7.3%), and OS was 85.2% (SE, 5.5%), respectively, without significant difference in EFS between the two treatments (P = .31). The 4-year EFS and OS for all ALCL patients (N = 86) were 71.8% (SE = 6.1%) and 88.1% (SE = 4.4%), respectively (Fig 5). For the randomly assigned ALCL patients (38 on APO and 45 on IDM/HiDAC), no significant difference in EFS between the two treatments (P = .70) was found. There was no statistically significant difference in EFS for patients with B-cell versus ALCL (P = .31). Furthermore, there were no differences in EFS for patients with mediastinal (P = .51) or bone involvement (P = .29). Also, patients in PR at the end of induction (n = 49) had EFS similar to those in CR (n = 120; P = .39). There were no significant differences in EFS between patients with stage 3 or stage 4 disease or age younger than 14 years compared with age ≡ 14 (below or above the median); the P values were 0.19 and 0.64, respectively. The distribution of the 59 events (induction failure, relapse, or death) according to treatment arm/histology group is presented in Table 4. However, the data should be viewed as inconclusive for all subgroups, and a larger number of patients could result in statistical significances.
Only 11 of 180 patients received radiation after induction therapy (including patients with CNS disease who were assigned to the APO arm with radiation). Four of the non-CNS patients had biopsy-proven viable disease; the rest of them received radiation based on imaging alone. Five of the 11 were alive and well in continuous complete remission at this writing (one CNS patient and four non-CNS patients).
Most toxicities were hematologic or gastrointestinal. The percent of a particular reported toxicity in an arm was the ratio of the number of patients who experienced the worst toxicity with a grade 3 or higher during the protocol treatment over the total number of patients assessable for toxicity in the arm. The Common Toxicity Criteria defined by the National Cancer Institute were used for grading. Seventy percent of the patients on the IDM/HiDAC arm, compared with 35% on the APO arm, had severe neutropenia and thrombocytopenia. There were 13 documented bacterial infections on the IDM/HiDAC arm, compared with 1% on the APO arm. Nausea and vomiting were reported in 15% and 3%, and mucositis, in 15% and 7.5% on the IDM/HiDAC and APO arms, respectively. During the study, there were no reports of cardiotoxicity or second malignant neoplasms.
DISCUSSION
Previous results of the POG showed that cyclophosphamide was not an essential agent in the treatment program of advanced stage LCL.6 To further improve outcome, an intense antimetabolite therapy was added. Methotrexate and cytarabine have a long track record in lymphoid malignancies15,16, and the POG accumulated experience combining them at these doses.17 Furthermore, use of cranial and local radiation was eliminated, except for a few patients with CNS disease or residual progressive disease after induction.
To date, our study is the only one on which all young patients with advanced-stage diffuse LCL subtypes were treated uniformly. Our data showed that seven cycles of APO alternating with eight cycles of intense antimetabolite therapy did not improve OS or EFS in advanced-stage pediatric LCL, regardless of phenotype (B-cell, T-cell, CD30+, or ALK+). Our preliminary results in smaller subgroups suggested that phenotype-directed therapy might be important in the context of our treatment programs.18,19 However, this analysis, with a longer follow-up, indicates that the intense antimetabolite therapy did not improve EFS for the whole group, and that there were no differences in outcome between the different phenotypes.
Comparing our B-cell LCL data with data of other reports proves difficult because of different inclusion criteria. In French-American-British LMB 1996, the B-cell LCL patients were treated on the same protocol as patients with BL and had a similar outcome of 90% EFS.20 However, the patients staged as "Group B" on these studies (which would comprise most of our advanced-stage patients) might have also included patients with non-advanced-stage LCL (ie, unresected stage I) who have an excellent prognosis, making any comparison problematic. The BFM group also lumps the B-cell LCL with BL.21 Reiter et al reported their data in patients with diffuse large B-cell lymphomas, showing an EFS of approximately 90%, but again, their stratification in the therapeutic groups was different.13 Although the histology, biology, and clinical course of B-cell LCL differs from BL,22 the current trend is to treat advanced-stage B-cell LCL along with BL. The risk of CNS involvement is significantly different for B-cell LCL, and at this time, it is unclear whether these patients need such aggressive CNS therapy. In our study, less than 3% had CNS involvement at diagnosis, and we did not encounter primary CNS failures, despite relatively modest CNS prophylaxis. These patients might not need intensive administration of alkylating agents, which did not improve outcome.6
Our data show that the distribution of subtypes of LCL in children differs from that seen in adults. In adults, most LCLs are diffuse large B-cell lymphoma,23 whereas in pediatric populations, T-cell lymphomas (consisting mostly of ALCL) are more frequent.24,25 In the pediatric age range, most cases with ALCL are associated with the t(2;5) or a variant translocation and NPM-ALK fusion product.26 Among our cases with ALCL, the proportion of null-cell to T-cell was comparable to others.27 Our data, although they showed no improvement with the investigative arm, compared favorably with ALCL data published by others. Brugieres et al and Reiter et al reported the Societe Francaise d'Oncologie Pediatrique and BFM results, which ranged between 54% and 76% EFS.27,28 Data from United Kingdom Children's Cancer Study Group and Massimino et al showed similar outcomes.29,30
The higher OS compared with EFS for both ALCL and B-cell LCL might represent successful salvage therapy with stem-cell transplantation or vinblastine.31
In our previous study, approximately 40% of the relapses occurred in the first 12 months of therapy, and another 45%, in the second year.6 The current study added IDM/HiDAC in an attempt to improve EFS, and the relapse pattern showed that the majority of them occurred in the first year on both arms. Thus, it seems that therapy could be intensified, and other agents may be tested in the context of the APO backbone (ie, the incorporation of vinblastine in the new ALCL study).
The APO regimen is well tolerated and for the most part, does not require hospitalization. It also does not include alkylating agents or epipodophyllotoxins, thus sparing long-term survivors some adverse sequelae. There were no reports of cardiotoxicity (cumulative dose of doxorubicin 300 mg/m2), though it might occur later, especially in female patients.32 Instead of substituting for doxorubicin other drugs that carry their own toxicity, our current data support the use of this backbone regimen for future clinical trials for LCL.
Future studies should focus on molecular heterogeneity of these lymphomas in the pediatric population and compare results to adults with similar pathologic diagnoses.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Katalin Banki, MD, and Clara Finch, MD, for participating in the pathology review, and Rachel Stroman, CRA, for data collection.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Murphy SB: Pediatric lymphomas: Recent advances and commentary on Ki-1-positive anaplastic large-cell lymphomas of childhood. Ann Oncol 5:S31每S33, 1994
2. Magrath IT: Malignant non-Hodgkin's lymphomas in children. Hematol Oncol Clin North Am 1:577每602, 1987
3. Hvizdala EV, Berard C, Callihan T, et al: Nonlymphoblastic lymphoma in children每histology and stage-related response to therapy: A Pediatric Oncology Group study. J Clin Oncol 9:1189每1195, 1991
4. Writing Committee: National Cancer Institute sponsored study of classifications of non-Hodgkin's lymphomas: Summary and description of a working formulation for clinical usage〞The non-Hodgkin's Lymphoma Pathologic Classification Project. Cancer 49:212每2135, 1982
5. Weinstein HJ, Lack EE, Cassady JR: APO therapy for malignant lymphoma of large cell "histiocytic " type of childhood: Analysis of treatment results for 29 patients. Blood 64:422每426, 1984
6. Laver JH, Mahmoud H, Pick TE, et al: Results of a randomized phase III trial in children and adolescents with advanced stage diffuse large cell non-Hodgkin's lymphoma: A Pediatric Oncology Group Study. Leuk Lymphoma 42:399每405, 2001
7. Murphy SB: Classification, staging and end results of treatment of childhood non-Hodgkin's lymphomas: Dissimilarities from lymphomas in adults. Semin Oncol 7:332每339, 1980
8. Harris NL, Jaffe ES, Stein H, et al: A revised European-American classification of lymphoid neoplasms: A proposal from the International Lymphoma Study Group. Blood 84:1361每1364, 1994
9. Jaffe ES, Harris NL, Stein H, et al: (eds): World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France, IARC Press, 2001
10. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 35:457每481, 1958
11. Peto R, Pike MC, Armitage P, et al: Design and analysis of randomized clinical trials requiring prolonged observation of each patient, II: Analysis and examples. Br J Cancer 35:1每39, 1977
12. Patte C: Treatment of mature B-ALL and high grade B-NHL in children. Best Pract Res Clin Haematol 15:695每711, 2002
13. Reiter A, Schrappe M, Tiemann M, et al: Improved treatment results in childhood B-cell neoplasms with tailored intensification of therapy: A report of the Berlin-Frankfurt-Munster Group Trial NHL-BFM 90. Blood 94:3294每3306, 1999
14. Patte C, Auperin A, Michon J, et al: The Societe Francaise d'Oncologie Pediatrique LMB89 protocol: Highly effective multiagent chemotherapy tailored to the tumor burden and initial response in 561 unselected children with B cell lymphomas and L3 leukemia. Blood 97:3370每3379, 2001
15. Wollner N, Exelby P, Lieberman P: Non-Hodgkin's lymphoma in children: A progress report on the original patients treated with the LSA2-L2 protocol. Cancer 44:1990每1994, 1979
16. Edelstein M, Vietti T, Valeriote F: The enhanced cytotoxicity of combination of 1-B-arabinofluranosyl cytosine and methotrexate. Cancer Res 35:1555每1558, 1975
17. Laver JH, Barredo JC, Amylon M, et al: Effects of cranial radiation in children with high risk T cell acute lymphoblastic leukemia: A Pediatric Oncology Group report. Leukemia 14:369每373, 2000
18. Hutchison RE, Berard CW, Shuster JJ, et al: B-cell lineage confers a favorable outcome among children and adolescents with large-cell lymphoma: A Pediatric Oncology Group study. J Clin Oncol 13:2023每2032, 1995
19. Laver JH, Weinstein JH, Hutchison R, et al: Lineage-specific differences in outcome for advanced large cell lymphoma in children and adolescents: Results of a randomized phase III Pediatric Oncology Group trial. Proc Am Soc Hematol 98:363, 2001 (abstr 1455)
20. Patte C, Gerrard A, Auperin A, et al: Results of the randomised international trial FAB LMB 96 for the "intermediate risk" childhood and adolescent B-cell lymphoma: Reduced therapy is efficacious. Proc Am Soc Clin Oncol 22:796, 2003 (abstr 3198)
21. Reiter A, Schrappe M, Parwaresch R, et al: Non-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 13:359每372, 1995
22. Murphy SB, Fairclough DL, Hutchison RE, et al: Non-Hodgkin's lymphomas of childhood: An analysis of the histology, staging, and response to treatment of 338 cases at a single institution. J Clin Oncol 7:186每193, 1989
23. Armitage JO, Vose JM, Linder J, et al: Clinical significance of immunophenotype in diffuse aggressive non-Hodgkin's lymphoma. J Clin Oncol 7:1783每1790, 1989
24. Hutchison RE, Fairclough DL, Holt H, et al: Clinical significance of histology and immunophenotype in childhood diffuse large cell lymphoma. Am J Clin Pathol 95:787每793, 1991
25. Delsol G, Ralfkiaer E, Stein H, et al: Anaplastic large cell lymphoma, in Jaffe ES, Harris NL, Stein H, et al (eds): Tumors of the Hematopoietic and Lymphoid Tissues (WHO Classification of Tumors). Lyon, France, IARC Press, 2001
26. Benharroch D, Meguerian-Bedoyan Z, Lamant L, et al: ALK-positive lymphoma: A single disease with a broad spectrum of morphology. Blood 91:2076每2084, 1998
27. Brugieres L, Deley MC, Pacquement H, et al: CD30(+) anaplastic large-cell lymphoma in children: Analysis of 82 patients enrolled in two consecutive studies of the French Society of Pediatric Oncology. Blood 92:3591每3598, 1998
28. Reiter A, Schrappe M, Tiemann M, et al: Successful treatment strategy for Ki-1 anaplastic large cell lymphoma of childhood: A prospective analysis of 62 patients enrolled in three consecutive Berlin-Frankfurt-Munster group studies. J Clin Oncol 12:899每908, 1994
29. Williams DM, Hobson R, Imeson J, et al: Anaplastic large cell lymphoma in childhood: Analysis of 72 patients treated on UKCCSG chemotherapy regimens. Br J Haematol 117:812每820, 2002
30. Massimino M, Spreafico F, Luksch R, et al: Prognostic significance of p80 and visceral involvement in childhood CD30 anaplastic large cell lymphoma (ALCL). Med Pediatr Oncol 37:97每102, 2001
31. Brugieres L, Quartier P, Le Deley MC, et al: Relapses of childhood ALCL: Treatment results in 41 children〞A report from the FSPO. Ann Oncol 11:53每58, 2000
32. Lipshultz SE, Lipsitz SR, Mone SM, et al: Female sex and drug dose as risk factors for late cardiotoxic effects of doxorubicin therapy for childhood cancer. N Engl J Med 332:1738每1743, 1995(Joseph H. Laver, Jacqueli)
Medical University of South Carolina, Charleston, SC
State University of New York, Syracuse
Roswell Park Memorial Institute, Buffalo, NY
Children's Oncology Group Statistical Office, Gainesville, FL
Maine Bureau of Health, Augusta, ME
Harvard Medical School
Massachusetts General Hospital, Boston, MA
University of Alberta, Edmonton, Alberta
Hospital for Sick Children, Toronto, Ontario, Canada
The University of Texas Health Science Center at San Antonio, TX
ABSTRACT
PATIENTS AND METHODS: From December 1994 to April 2000, we enrolled 180 eligible pediatric patients with stage III/IV large-cell lymphoma (LCL); 90 patients were randomly assigned to the intermediate-dose methotrexate (IDM) and high-dose cytarabine (HiDAC) arm, 85 patients to the APO arm, and five patients directly to the APO arm by study design due to CNS involvement. Planned therapy duration was 12 months.
RESULTS: The 4-year EFS for all patients was 67.4% (SE, 4.2%), and OS was 80.1% (SE, 3.6%) without any significant difference between the two arms. The 4-year EFS and OS were 71.8% (SE, 6.1%) and 88.1% (SE, 4.4%), respectively, for patients with anaplastic large-cell lymphoma, and 63.8% (SE, 10.3%) and 70.3% (SE, 9.0%), respectively, for patients with diffuse large B-cell lymphoma. Only 11 patients required radiation (due to unresponsive bulky disease or CNS involvement). The IDM/HiDAC arm was associated with more toxicity.
CONCLUSION: The efficacy of incorporating IDM/HiDAC in the treatment plan of pediatric and adolescent patients with advanced-stage LCL was inconclusive as to its effect on EFS, regardless of the lymphoma phenotype. It cannot be excluded that with a higher number of patients, one treatment could prove superior and future studies will build on these data.
INTRODUCTION
With the POG data and a modified APO regimen (POG #8615), we found an event-free survival (EFS) of 70% and concluded that the addition of cyclophosphamide did not improve outcome.6 In the current successor study (POG #9315), we used random assignment to assess effects of incorporating into the APO backbone, eight courses of intense antimetabolite therapy (intermediate-dose methotrexate [IDM] given over 24 hours, followed by high-dose cytarabine [HiDAC] by continuous infusion for 48 hours) on overall survival (OS) and EFS.
PATIENTS AND METHODS
Treatment Program
The program consisted of a similar induction and two different maintenance treatment plans and lasted for 12 months for both arms. Patients were randomly assigned upfront to one of the two maintenance arms presented in Figure 1. After the induction phase (doxorubicin 75 mg/m2 day 1 and 22, vincristine 1.5 mg/m2 day 1 and 22, prednisone 40 mg/m2 daily for 28 days, and age-adjusted intrathecal methotrexate on days 1, 8, and 22), patients were treated every 21 days with APO maintenance (doxorubicin 30 mg/m2 and vincristine 1.5 mg/m2 on day 1, prednisone 120 mg/m2 and 6-mercaptopurine 225/mg/m2 days 1 to 5) or APO maintenance alternating with IDM/HiDAC (IDM 1 g/m2 over 24 hours followed by cytarabine 500 mg/m2 bolus, and continuous infusion of cytarabine 60 mg/m2 over 48 hours) every 21 days. Both maintenance arms included intrathecal methotrexate on day one of maintenance cycles 1, 3, and 5, and doxorubicin was substituted with methotrexate after a cumulative dose of 300 mg/m2 was reached (after cycle 5 of the APO arm and cycle 10 of the IDM/HiDAC regimen); the dose of methotrexate was 60 mg/m2 per cycle. Patients with bulky disease at the end of induction (not achieving complete response [CR]; response criteria listed below) were eligible for radiation therapy after biopsy proved viable disease. The total dose to the prescription point was 4,140 cGy in 23 daily fractions (180 cGy/d). For patients in partial remission, most investigators chose to treat them with two cycles of maintenance therapy before performing a biopsy. CNS disease-positive patients were not randomly assigned and were assigned directly to the APO arm. During induction, CNS disease-positive patients received additional intrathecal methotrexate on days 15, 29, and 36, for a total of six intrathecal methotrexate doses during induction. Whole-brain irradiation was started during the first week of maintenance. The total radiation dose was 2,400 cGy in 16 daily fractions (150 cGy/d). These patients also received two additional doses of intrathecal methotrexate on day 1 of maintenance cycles 2 and 4, for a total of five intrathecal methotrexate doses during maintenance. Methotrexate was substituted for doxorubicin at a 33% dose reduction for patients who received cranial irradiation; the dose for these patients was 40 mg/m2 per cycle.
Immunohistochemistry
Immunohistochemistry analysis was performed with a panel of antibodies directed against lymphoid-associated antigens, effective in paraffin sections, and intended to identify lineage: B-lineage antibody L26 (CD20; Dako, Carpinteria, CA); T-lineage antibodies UCHL-1 (CD45Ro), CD3 (Dako), and MT1 (CD43; Biotest, Denville, NJ); Hodgkin's and ALCL antibody Ber-H2 (CD30; Biotest), ALCL-associated antibody ALK-1 (Dako), and histiocyte/macrophage-associated antibody KP1 (CD68; Dako). Antibodies were applied in this order when available material consisted of limited numbers of unstained slides. For cases in which unstained slides were not available, antibody results from treating institutions or reference laboratories of primary diagnosis were accepted if reviewers determined that they were appropriate.
Response Criteria
CR was defined as the disappearance of all evidence of disease from all sites. This was determined by physical examination and appropriate laboratory and imaging studies. Partial response (PR) was determined by at least a 50% decrease in tumor mass by comparing postinduction imaging to imaging at diagnosis. "No response" was defined as failure to achieve PR. The need for radiation therapy was determined at the end of induction (day 42); it could be used earlier only as an emergency intervention due to mediastinal disease with respiratory symptoms.
Statistical Methods
EFS, time from registration to earliest evidence of relapse, progressive disease, second cancer, or death from any cause (treatment failures) were the major end points of the study. The log-rank test was used in comparative analyses. We originally planned to accrue 237 patients. Assuming proportional hazards, the study had 80% power to detect a 2-year EFS rate of 0.87 compared with a 2-year EFS rate of 0.75 in controls, with a significance level of .05. Planned analyses included two interim analyses and one final analysis. EFS and OS curves were constructed by the Kaplan-Meier method with SEs of Peto et al.10,11 Due to the orphan nature of this disease, with relatively few patients, the reader should judge nonsignificant differences as inconclusive. All reported P values are two-sided.
RESULTS
Of 180 eligible patients enrolled, 90 were on the APO arm, including five with CNS involvement who were assigned to the arm by study design, and 90 patients were on the IDM/HiDAC arm (Table 1). There were two patients who had induction failures and two who died early, with a CR/PR rate of 98% on each arm (response rates are depicted in Table 2). Fifty-nine percent of patients presented with peripheral lymphadenopathy (cervical, supraclavicular, axillary, inguinal, or other site), 21% had hepatosplenomegaly, and 55% had mediastinal involvement (61% and 54% of ALCL and B-cell LCL patients, respectively, had mediastinal involvement). Seven percent of patients had skin involvement (10 of 12 had ALCL), and 24% had bone involvement (one-third had primary bone involvement). Forty-one percent of the whole group presented with "B" symptoms (fever > 101∼F or weight loss > 10%), in 47% of ALCL patients and 36% of B-cell LCL patients, respectively. Sixty-three percent had elevated lactate dehydrogenase (76% and 45% of B-cell LCL and ALCL patients, respectively, had increased lactate dehydrogenase).
Phenotype distribution is depicted in Table 3. There were eighty-six cases with ALCL (58 T-cell and 28 null cell phenotype), with 98% CD30+ and 89% ALK+; only 83 patients were randomly assigned, as three patients with CNS involvement were assigned to the APO arm. There were seventy-five cases were B-cell LCL, and 73 were randomized because of two patients having CNS involvement, who were assigned to the APO arm. There was one case of follicular lymphoma, and one of low-grade lymphoma of mucosa-associated tissue (MALT) with increased large cells (the remaining 71 being diffuse large B-cell lymphoma). Ten cases were classified as peripheral T-cell lymphoma, unspecified; none of these expressed CD 30 or ALK, but two expressed CD68. Nine cases did not have immunophenotyping and were categorized as LCL, unclassified.
The 4-year EFS for all patients was 67.4% (SE, 4.2%), and OS was 80.1% (SE, 3.6%; Fig 2). The EFS for IDM/HiDAC was almost identical to that of the APO arm (Fig 3; P = .96).
There were no significant differences in EFS between arms (P = .69, Fig 4) among the randomized diffuse large B-cell lymphoma patients (37 on IDM/HiDAC and 34 on APO); 4-year EFS for APO and IDM/HiDAC patients was 63.8% (SE, 10.3%) and 70.3% (SE, 9.0%), respectively. For randomized T-cell LCL patients (31 on the APO arm and 34 on the IDM/HiDAC arm), the 4-year EFS was 66.1% (SE, 7.3%), and OS was 85.2% (SE, 5.5%), respectively, without significant difference in EFS between the two treatments (P = .31). The 4-year EFS and OS for all ALCL patients (N = 86) were 71.8% (SE = 6.1%) and 88.1% (SE = 4.4%), respectively (Fig 5). For the randomly assigned ALCL patients (38 on APO and 45 on IDM/HiDAC), no significant difference in EFS between the two treatments (P = .70) was found. There was no statistically significant difference in EFS for patients with B-cell versus ALCL (P = .31). Furthermore, there were no differences in EFS for patients with mediastinal (P = .51) or bone involvement (P = .29). Also, patients in PR at the end of induction (n = 49) had EFS similar to those in CR (n = 120; P = .39). There were no significant differences in EFS between patients with stage 3 or stage 4 disease or age younger than 14 years compared with age ≡ 14 (below or above the median); the P values were 0.19 and 0.64, respectively. The distribution of the 59 events (induction failure, relapse, or death) according to treatment arm/histology group is presented in Table 4. However, the data should be viewed as inconclusive for all subgroups, and a larger number of patients could result in statistical significances.
Only 11 of 180 patients received radiation after induction therapy (including patients with CNS disease who were assigned to the APO arm with radiation). Four of the non-CNS patients had biopsy-proven viable disease; the rest of them received radiation based on imaging alone. Five of the 11 were alive and well in continuous complete remission at this writing (one CNS patient and four non-CNS patients).
Most toxicities were hematologic or gastrointestinal. The percent of a particular reported toxicity in an arm was the ratio of the number of patients who experienced the worst toxicity with a grade 3 or higher during the protocol treatment over the total number of patients assessable for toxicity in the arm. The Common Toxicity Criteria defined by the National Cancer Institute were used for grading. Seventy percent of the patients on the IDM/HiDAC arm, compared with 35% on the APO arm, had severe neutropenia and thrombocytopenia. There were 13 documented bacterial infections on the IDM/HiDAC arm, compared with 1% on the APO arm. Nausea and vomiting were reported in 15% and 3%, and mucositis, in 15% and 7.5% on the IDM/HiDAC and APO arms, respectively. During the study, there were no reports of cardiotoxicity or second malignant neoplasms.
DISCUSSION
Previous results of the POG showed that cyclophosphamide was not an essential agent in the treatment program of advanced stage LCL.6 To further improve outcome, an intense antimetabolite therapy was added. Methotrexate and cytarabine have a long track record in lymphoid malignancies15,16, and the POG accumulated experience combining them at these doses.17 Furthermore, use of cranial and local radiation was eliminated, except for a few patients with CNS disease or residual progressive disease after induction.
To date, our study is the only one on which all young patients with advanced-stage diffuse LCL subtypes were treated uniformly. Our data showed that seven cycles of APO alternating with eight cycles of intense antimetabolite therapy did not improve OS or EFS in advanced-stage pediatric LCL, regardless of phenotype (B-cell, T-cell, CD30+, or ALK+). Our preliminary results in smaller subgroups suggested that phenotype-directed therapy might be important in the context of our treatment programs.18,19 However, this analysis, with a longer follow-up, indicates that the intense antimetabolite therapy did not improve EFS for the whole group, and that there were no differences in outcome between the different phenotypes.
Comparing our B-cell LCL data with data of other reports proves difficult because of different inclusion criteria. In French-American-British LMB 1996, the B-cell LCL patients were treated on the same protocol as patients with BL and had a similar outcome of 90% EFS.20 However, the patients staged as "Group B" on these studies (which would comprise most of our advanced-stage patients) might have also included patients with non-advanced-stage LCL (ie, unresected stage I) who have an excellent prognosis, making any comparison problematic. The BFM group also lumps the B-cell LCL with BL.21 Reiter et al reported their data in patients with diffuse large B-cell lymphomas, showing an EFS of approximately 90%, but again, their stratification in the therapeutic groups was different.13 Although the histology, biology, and clinical course of B-cell LCL differs from BL,22 the current trend is to treat advanced-stage B-cell LCL along with BL. The risk of CNS involvement is significantly different for B-cell LCL, and at this time, it is unclear whether these patients need such aggressive CNS therapy. In our study, less than 3% had CNS involvement at diagnosis, and we did not encounter primary CNS failures, despite relatively modest CNS prophylaxis. These patients might not need intensive administration of alkylating agents, which did not improve outcome.6
Our data show that the distribution of subtypes of LCL in children differs from that seen in adults. In adults, most LCLs are diffuse large B-cell lymphoma,23 whereas in pediatric populations, T-cell lymphomas (consisting mostly of ALCL) are more frequent.24,25 In the pediatric age range, most cases with ALCL are associated with the t(2;5) or a variant translocation and NPM-ALK fusion product.26 Among our cases with ALCL, the proportion of null-cell to T-cell was comparable to others.27 Our data, although they showed no improvement with the investigative arm, compared favorably with ALCL data published by others. Brugieres et al and Reiter et al reported the Societe Francaise d'Oncologie Pediatrique and BFM results, which ranged between 54% and 76% EFS.27,28 Data from United Kingdom Children's Cancer Study Group and Massimino et al showed similar outcomes.29,30
The higher OS compared with EFS for both ALCL and B-cell LCL might represent successful salvage therapy with stem-cell transplantation or vinblastine.31
In our previous study, approximately 40% of the relapses occurred in the first 12 months of therapy, and another 45%, in the second year.6 The current study added IDM/HiDAC in an attempt to improve EFS, and the relapse pattern showed that the majority of them occurred in the first year on both arms. Thus, it seems that therapy could be intensified, and other agents may be tested in the context of the APO backbone (ie, the incorporation of vinblastine in the new ALCL study).
The APO regimen is well tolerated and for the most part, does not require hospitalization. It also does not include alkylating agents or epipodophyllotoxins, thus sparing long-term survivors some adverse sequelae. There were no reports of cardiotoxicity (cumulative dose of doxorubicin 300 mg/m2), though it might occur later, especially in female patients.32 Instead of substituting for doxorubicin other drugs that carry their own toxicity, our current data support the use of this backbone regimen for future clinical trials for LCL.
Future studies should focus on molecular heterogeneity of these lymphomas in the pediatric population and compare results to adults with similar pathologic diagnoses.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Katalin Banki, MD, and Clara Finch, MD, for participating in the pathology review, and Rachel Stroman, CRA, for data collection.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Murphy SB: Pediatric lymphomas: Recent advances and commentary on Ki-1-positive anaplastic large-cell lymphomas of childhood. Ann Oncol 5:S31每S33, 1994
2. Magrath IT: Malignant non-Hodgkin's lymphomas in children. Hematol Oncol Clin North Am 1:577每602, 1987
3. Hvizdala EV, Berard C, Callihan T, et al: Nonlymphoblastic lymphoma in children每histology and stage-related response to therapy: A Pediatric Oncology Group study. J Clin Oncol 9:1189每1195, 1991
4. Writing Committee: National Cancer Institute sponsored study of classifications of non-Hodgkin's lymphomas: Summary and description of a working formulation for clinical usage〞The non-Hodgkin's Lymphoma Pathologic Classification Project. Cancer 49:212每2135, 1982
5. Weinstein HJ, Lack EE, Cassady JR: APO therapy for malignant lymphoma of large cell "histiocytic " type of childhood: Analysis of treatment results for 29 patients. Blood 64:422每426, 1984
6. Laver JH, Mahmoud H, Pick TE, et al: Results of a randomized phase III trial in children and adolescents with advanced stage diffuse large cell non-Hodgkin's lymphoma: A Pediatric Oncology Group Study. Leuk Lymphoma 42:399每405, 2001
7. Murphy SB: Classification, staging and end results of treatment of childhood non-Hodgkin's lymphomas: Dissimilarities from lymphomas in adults. Semin Oncol 7:332每339, 1980
8. Harris NL, Jaffe ES, Stein H, et al: A revised European-American classification of lymphoid neoplasms: A proposal from the International Lymphoma Study Group. Blood 84:1361每1364, 1994
9. Jaffe ES, Harris NL, Stein H, et al: (eds): World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France, IARC Press, 2001
10. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 35:457每481, 1958
11. Peto R, Pike MC, Armitage P, et al: Design and analysis of randomized clinical trials requiring prolonged observation of each patient, II: Analysis and examples. Br J Cancer 35:1每39, 1977
12. Patte C: Treatment of mature B-ALL and high grade B-NHL in children. Best Pract Res Clin Haematol 15:695每711, 2002
13. Reiter A, Schrappe M, Tiemann M, et al: Improved treatment results in childhood B-cell neoplasms with tailored intensification of therapy: A report of the Berlin-Frankfurt-Munster Group Trial NHL-BFM 90. Blood 94:3294每3306, 1999
14. Patte C, Auperin A, Michon J, et al: The Societe Francaise d'Oncologie Pediatrique LMB89 protocol: Highly effective multiagent chemotherapy tailored to the tumor burden and initial response in 561 unselected children with B cell lymphomas and L3 leukemia. Blood 97:3370每3379, 2001
15. Wollner N, Exelby P, Lieberman P: Non-Hodgkin's lymphoma in children: A progress report on the original patients treated with the LSA2-L2 protocol. Cancer 44:1990每1994, 1979
16. Edelstein M, Vietti T, Valeriote F: The enhanced cytotoxicity of combination of 1-B-arabinofluranosyl cytosine and methotrexate. Cancer Res 35:1555每1558, 1975
17. Laver JH, Barredo JC, Amylon M, et al: Effects of cranial radiation in children with high risk T cell acute lymphoblastic leukemia: A Pediatric Oncology Group report. Leukemia 14:369每373, 2000
18. Hutchison RE, Berard CW, Shuster JJ, et al: B-cell lineage confers a favorable outcome among children and adolescents with large-cell lymphoma: A Pediatric Oncology Group study. J Clin Oncol 13:2023每2032, 1995
19. Laver JH, Weinstein JH, Hutchison R, et al: Lineage-specific differences in outcome for advanced large cell lymphoma in children and adolescents: Results of a randomized phase III Pediatric Oncology Group trial. Proc Am Soc Hematol 98:363, 2001 (abstr 1455)
20. Patte C, Gerrard A, Auperin A, et al: Results of the randomised international trial FAB LMB 96 for the "intermediate risk" childhood and adolescent B-cell lymphoma: Reduced therapy is efficacious. Proc Am Soc Clin Oncol 22:796, 2003 (abstr 3198)
21. Reiter A, Schrappe M, Parwaresch R, et al: Non-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 13:359每372, 1995
22. Murphy SB, Fairclough DL, Hutchison RE, et al: Non-Hodgkin's lymphomas of childhood: An analysis of the histology, staging, and response to treatment of 338 cases at a single institution. J Clin Oncol 7:186每193, 1989
23. Armitage JO, Vose JM, Linder J, et al: Clinical significance of immunophenotype in diffuse aggressive non-Hodgkin's lymphoma. J Clin Oncol 7:1783每1790, 1989
24. Hutchison RE, Fairclough DL, Holt H, et al: Clinical significance of histology and immunophenotype in childhood diffuse large cell lymphoma. Am J Clin Pathol 95:787每793, 1991
25. Delsol G, Ralfkiaer E, Stein H, et al: Anaplastic large cell lymphoma, in Jaffe ES, Harris NL, Stein H, et al (eds): Tumors of the Hematopoietic and Lymphoid Tissues (WHO Classification of Tumors). Lyon, France, IARC Press, 2001
26. Benharroch D, Meguerian-Bedoyan Z, Lamant L, et al: ALK-positive lymphoma: A single disease with a broad spectrum of morphology. Blood 91:2076每2084, 1998
27. Brugieres L, Deley MC, Pacquement H, et al: CD30(+) anaplastic large-cell lymphoma in children: Analysis of 82 patients enrolled in two consecutive studies of the French Society of Pediatric Oncology. Blood 92:3591每3598, 1998
28. Reiter A, Schrappe M, Tiemann M, et al: Successful treatment strategy for Ki-1 anaplastic large cell lymphoma of childhood: A prospective analysis of 62 patients enrolled in three consecutive Berlin-Frankfurt-Munster group studies. J Clin Oncol 12:899每908, 1994
29. Williams DM, Hobson R, Imeson J, et al: Anaplastic large cell lymphoma in childhood: Analysis of 72 patients treated on UKCCSG chemotherapy regimens. Br J Haematol 117:812每820, 2002
30. Massimino M, Spreafico F, Luksch R, et al: Prognostic significance of p80 and visceral involvement in childhood CD30 anaplastic large cell lymphoma (ALCL). Med Pediatr Oncol 37:97每102, 2001
31. Brugieres L, Quartier P, Le Deley MC, et al: Relapses of childhood ALCL: Treatment results in 41 children〞A report from the FSPO. Ann Oncol 11:53每58, 2000
32. Lipshultz SE, Lipsitz SR, Mone SM, et al: Female sex and drug dose as risk factors for late cardiotoxic effects of doxorubicin therapy for childhood cancer. N Engl J Med 332:1738每1743, 1995(Joseph H. Laver, Jacqueli)