- © 2006 by American Society of Clinical Oncology
Impact of Cranial Radiotherapy on Central Nervous System Prophylaxis in Children and Adolescents With Central Nervous System–Negative Stage III or IV Lymphoblastic Lymphoma
- Birgit Burkhardt,
- Wilhelm Woessmann,
- Martin Zimmermann,
- Udo Kontny,
- Josef Vormoor,
- Wolfgang Doerffel,
- Georg Mann,
- Guenter Henze,
- Felix Niggli,
- Wolf-Dieter Ludwig,
- Dirk Janssen,
- Hansjoerg Riehm,
- Martin Schrappe and
- Alfred Reiter
- From the Department of Pediatric Hematology and Oncology, Children's University Hospital, Giessen; Department of Pediatric Hematology and Oncology, Children's University Hospital, Freiburg; Department of Pediatric Hematology and Oncology, Children's University Hospital, Münster; Department of Pediatric Hematology and Oncology; and Department of Hematology, Oncology, and Tumor Immunology, Robert-Rössle-Clinic, Helios Klinikum Berlin-Buch; Department of Pediatric Hematology and Oncology, Children's University Hospital; Charité Medical School, Berlin; Department of Hematopathology and Lymph Node Registry, University Hospital; Children's University Hospital, University Hospital Schleswig-Holstein, Campus Kiel, Kiel; Department of Pediatric Hematology and Oncology, Medizinische Hochschule, Hannover, Germany; Department of Pediatric Hematology and Oncology, St. Anna Children's Hospital, Vienna, Austria; and Department of Pediatric Hematology and Oncology, Children's University Hospital Zurich, Switzerland.
- Address reprint requests to Alfred Reiter, MD, NHL-BFM-Study Center, Justus-Liebig-University, Department of Pediatric Hematology and Oncology, Feulgenstr 12, D-35385 Gieβen, Germany; e-mail: alfred.reiter{at}paediat.med.uni-giessen.de
Abstract
Purpose In the Non-Hodgkin's Lymphoma–Berlin-Frankfurt-Munster (NHL-BFM) 95 trial, we tested, against the historical control of the combined trials NHL-BFM90 and NHL-BFM86, whether prophylactic cranial radiotherapy (PCRT) can be omitted for CNS-negative patients with stage III or IV lymphoblastic lymphoma (LBL) with sufficient early response.
Patients and Methods Apart from the removal of PCRT in NHL-BFM95, the chemotherapy of the three trials was identical except for the amount of l-asparaginase and daunorubicin during induction. The therapy in NHL-BFM95 was accepted to be noninferior when compared with trials NHL-BFM90/86 if the lower limit of the one-sided 95% CI for the difference in the 2-year probability of event-free-survival (pEFS) between target patients of NHL-BFM95 and the historical controls of NHL-BFM90/86 did not exceed −14%. The target patient group consisted of stage III and IV patients who were CNS negative and responded well to induction therapy.
Results The number of target patients was 156 in NHL-BFM95 (median age, 8.6 years; range, 0.2 to 19.5 years) and 163 in NHL-BFM90/86 (median age, 8.4 years; range, 0.6 to 16.6 years). For the target group, the pEFS rates at 2 and 5 years were 86% ± 3% and 82% ± 3%, respectively, in NHL-BFM95 (median follow-up time, 5.1 years; range, 2.1 to 9.1 years) compared with 91% ± 2% and 88% ± 3%, respectively in NHL-BFM90/86 (median follow-up time, 10.7 years; range, 5 to 15.4 years). The lower limit of the one-sided 95% CI for the difference in pEFS was −11% at 2 years and −13% at 5 years. In NHL-BFM95, one isolated and two combined CNS relapses occurred compared with one combined CNS relapse in NHL-BFM90/86. Five-year disease-free-survival rate was 88% ± 3% in NHL-BFM95 compared with 91% ± 2% in NHL-BFM90/86.
Conclusion For CNS-negative patients with stage III or IV LBL and sufficient response to induction therapy, treatment without PCRT may be noninferior to treatment including PCRT.
INTRODUCTION
Combination chemotherapy with or without local radiotherapy results in event-free survival (EFS) rates of 64% to 90% for children suffering from lymphoblastic lymphoma (LBL).1-8 One unresolved issue is the role of prophylactic cranial radiotherapy (PCRT) in the treatment of patients with advanced-stage disease. PCRT has been used for prevention of CNS relapses in some studies2,5,9 and not used in other studies.1,4,6,7,10 Favorable results were achieved in our previous Non-Hodgkin's Lymphoma (NHL) –Berlin-Frankfurt-Munster (BFM) 90 and NHL-BFM86 trials, even for patients with advanced-stage T-cell LBL,5 with an intensive acute lymphoblastic leukemia (ALL) –type strategy including PCRT. However, CRT may carry significant late risks such as neuropsychological deficits,11-16 mood disturbances,17 short stature,18-22 and secondary malignancies (SMN).23-27 Therefore, in the subsequent trial NHL-BFM95, we tested against the historical control of trials NHL-BFM90/86 whether PCRT can be safely omitted for CNS-negative patients with stage III or IV disease. Chemotherapy differed only slightly between the three consecutive studies.
PATIENTS AND METHODS
Patients
From April 1995 to March 2001, 236 patients up to 19 years of age with newly diagnosed LBL were enrolled onto NHL-BFM95 from 76 centers in Austria, Germany, and Switzerland after informed consent of patients and/or guardians. Of these patients, 198 were assessable as protocol patients. Thirty-eight patients were excluded from evaluation for the following reasons: diagnosis of LBL was not confirmed, patients had a peripheral T-cell lymphoma or a not further classified T-cell lymphoma (n = 23, eight events), NHL was an SMN (n = 4, one toxic death), previous cytostatic treatment (n = 5, no events), clinic was not participating in the study (n = 3, no event), continuation of treatment abroad (n = 1, lost to follow-up), and discontinuation of treatment before day 33 of induction as a result of religious conviction of the parents (n = 2, two events). For the study objective concerning the impact of PCRT, additional eligibility criteria were as follows: stage III or IV disease, CNS negative, sufficient response at day 33 (see Therapy for definition), and no receipt of PCRT.
Diagnostic Work-Up
Patients were classified according to criteria of the updated Kiel classification28 and the WHO classification of hematologic malignancies.29 The diagnosis was centrally reviewed in more than 90% of patients.
Staging included peripheral-blood and bone marrow (BM) aspiration smears, CSF, serum lactate dehydrogenase analysis, ultrasound, x-ray, computed tomography or magnetic resonance imaging, and skeletal scintigraphy. The St Jude staging system30 was used. Criteria for CNS involvement were more than 5 cells/μL CSF and identifiable blasts in CSF on cytospin preparations, and/or cerebral infiltrates on computed tomography/magnetic resonance imaging, and/or cranial nerve palsy not caused by extracerebral lymphoma manifestations.
Therapy
In NHL-BFM95, patients with stage III or IV LBL received induction protocol I (Table 1) followed by the extracompartment protocol M, reintensification protocol II, and maintenance (mercaptopurine 50 mg/m2 daily and methotrexate [MTX] 20 mg/m2 weekly, both orally) for up to a total therapy duration of 24 months. In case of respiratory impairment as a result of mediastinal mass, patients received one to two doses of cyclophosphamide 200 mg/m2, and the lumbar puncture on day 1 was withheld until the patient was stabilized. In protocol M, MTX therapy (5 g/m2) was administered as described previously.5 CNS-positive patients received two additional doses of intrathecal MTX at days 18 and 27 of induction and received CRT after reintensification. The dose was 12 Gy for patients in their second year of life and 18 Gy in older children (children < 1 year of age were not irradiated). Patients with identifiable blasts in CSF-cytospin preparation but less than 5 cells/μL in CSF were not considered CNS positive but received two additional doses of intrathecal MTX at days 18 and 27. For boys with testicular involvement, orchiectomy was not performed, and irradiation (20 Gy) of testes was to be confined to biopsy-proven persistent infiltration of testis after protocol M.
Treatment Protocol for NHL-BFM95 for Advanced-Stage Lymphoblastic Lymphoma
Response to treatment was evaluated on day 33 of induction in NHL-BFM90 and NHL-BFM95 and at day 40 in NHL-BFM86. Sufficient response was defined as at least 70% tumor regression, less than 5% BM blasts, and no CNS disease. For patients with insufficient response at day 33, treatment was to be intensified according to the high-risk branch of trial ALL-BFM95, with local radiotherapy (30 Gy) and allogeneic blood stem-cell transplantation.31
Apart from PCRT, the treatments in the three trials NHL-BFM86, NHL-BFM90, and NHL-BFM95 were identical with the following modifications in the induction protocol I: in NHL-BFM86, the dose of daunorubicin was 40 mg/m2 for 4 days compared with 30 mg/m2 for 4 days in NHL-BFM90 and NHL-BFM95. Treatment intensity over time was increased by condensation of induction phase I from 42 days in NHL-BFM86 to 35 days in NHL-BFM90 and NHL-BFM95. In NHL-BFM95, patients received in induction Escherichia coli l-asparaginase 5,000 U/m2 for 8 days (Kyowa, Japan), whereas in NHL-BFM90 and NHL-BFM86, induction E coli l-asparaginase 10,000 U/m2 for 8 days (Bayer, Leverkusen, Germany) was administered.5,9
Statistical Analysis
The purpose of the study was to test, against the historical controls of the combined trials NHL-BFM90/86, whether PCRT can be omitted for CNS-negative patients with stage III or IV LBL and sufficient response at day 33 of induction. The primary end point was the one-sided 95% CI for the difference of probability of EFS (pEFS) at 2 years between the test group and the control group. Taking into account the serious late risks of PCRT and a rescue chance for patients suffering from CNS relapses,32-34 treatment without PCRT was considered noninferior to therapy including PCRT if the lower limit of the one-sided 95% CI for the difference in pEFS at 2 years between the target group of NHL-BFM95 treated without PCRT and the historical control group of combined trials NHL-BFM90/86 treated with PCRT did not exceed −14%. This test was planned as a per-protocol analysis; that is, only patients allocated to the target group (stage III or IV LBL, CNS negative, and at least 70% tumor reduction at day 33 of induction) who received no PCRT were included. The expected number of patients was 150 in the target group of NHL-BFM95 and 160 in the historical control group of combined trials NHL-BFM90/86. With these expected numbers of patients, the power to prove noninferiority of treatment without PCRT was estimated to be 0.80 if pEFS is 0.75 in the control group of combined trials NHL-BFM90/86 with treatment including PCRT (type I error = 5%). Survival functions were calculated according to the Kaplan-Meier method,35 with differences compared using the log-rank test.36 The 95% CI for the estimate of EFS was calculated using the bootstrap method.37 EFS was calculated from the date of diagnosis to the date of the first event (death from any cause, tumor progression, or SMN) or to the date of last follow-up. Disease-free survival (DFS) was calculated from the date of diagnosis to tumor progression or relapse. Overall survival was calculated from the time of diagnosis to death. Patients lost to follow-up were censored at the time of their last follow-up examination. Differences in the distribution of individual parameters among patient subsets were analyzed using the χ2 test or Fisher's exact test. The statistical analysis was carried out using the SAS statistical program (SAS-PC, version 6.04; SAS Institute, Inc, Cary, NC). Follow-up data were updated as of September 2004.
RESULTS
Patient Characteristics
One hundred ninety-eight protocol LBL patients were enrolled onto NHL-BFM95. The immunophenotypes were precursor B cell, T cell, and not available in 52, 140, and six patients, respectively. Six, 16, 123, and 46 patients were diagnosed with stage I, II, III, and IV CNS-negative disease, respectively. Seven patients were CNS positive. Of 169 stage III or IV CNS-negative patients, six received PCRT, and one received different chemotherapy. Six patients had insufficient response at day 33. Thus, 156 (79%) of 198 LBL patients in NHL-BFM95 were defined as the target group of NHL-BFM95 for the evaluation of the study question (Table 2).
Patients Enrolled Onto Trials NHL-BFM86, NHL-BFM90, and NHL-BFM95: Treatment Results of Patients With Stage III or IV CNS-Negative Lymphoblastic Lymphoma Compared With Treatment Results of the Target Group (patients with stage III or IV CNS-negative lymphoblastic lymphoma and sufficient response to treatment at day 33 not receiving PCRT)
In NHL-BFM90, 136 LBL patients were enrolled. The immunophenotypes were precursor B cell, T cell, and not available in 24, 106, and six patients, respectively. Seven, 11, 92, and 22 patients had stage I, II, III, and IV CNS-negative disease, respectively. Four patients were CNS positive (Table 2). Of the 114 stage III or IV CNS-negative patients, three received different chemotherapy. Two patients had insufficient response at day 33. Thus, 109 patients (80%) were eligible as historical controls for the study objective (Table 2). Of the 63 LBL patients in NHL-BFM86, five, 53, and five patients had the immunophenotypes of precursor B cell, T cell, and not available, respectively. One, three, 43, and 12 patients had stage I, II, III, and IV CNS-negative disease, respectively. Four patients were CNS positive. Of the 55 stage III or IV CNS-negative patients, one had insufficient response after the first 5 weeks of induction. Thus, 54 patients (86%) were eligible as historical controls for the study objective (Table 2).
The clinical characteristics of the target patients of NHL-BFM95 and controls of NHL-BFM90/86 were comparable (Table 3). However, in NHL-BFM95, fewer patients had mediastinal tumors, which corresponded to a higher percentage of precursor B-cell LBL patients compared with NHL-BFM90/86.
Characteristics and Treatment Results of Target Patients Assessable for the Study Objective (treatment without prophylactic cranial radiotherapy in NHL-BFM95) and the Historical Controls of NHL-BFM90/86
Outcome
The median follow-up time was 5.1 years (range, 2.1 to 9.1 years) for the target patients of NHL-BFM95 and 10.7 years (range, 5.0 to 15.4 years) for the historical controls of NHL-BFM90/86. Two-year pEFS rate was 86% ± 3% for the target group of NHL-BFM95 and 91% ± 2% for the controls in trials NHL-BFM90/86 (Fig 1A). The lower limit of the 95% CI for the difference in pEFS in NHL-BFM95 compared with NHL-BFM90/86 at 2 years was −11%. This was greater than the predetermined maximal acceptable loss of −14%. The pEFS rate at 5 years was 82% ± 3% for the target group of NHL-BFM95 compared with 88% ± 3% for NHL-BFM90/86 (lower limit of the 95% CI for the loss in pEFS = −13%). Adverse events are listed in Table 3. In NHL-BFM95, three patients died of mediastinal tumor complications before beginning therapy and at days 2 and 6 of therapy. Two patients died of infection (Rota virus and Pseudomonas) during reintensification. In NHL-BFM90, one patient died of mediastinal mass compression before treatment. In NHL-BFM86, one patient died during initial thoracotomy, one patient died at day 39 of GI hemorrhage, and one patient died during reintensification of hepatic failure (Table 3). The cumulative incidences of isolated CNS relapse, combined CNS relapse, non-CNS relapse, SMN, and toxic death were analyzed for the three trials (Figs 2A, 2B, and 2C). Although not statistically significant, the cumulative incidence at 5 years of non-CNS tumor failure was lowest in NHL-BFM90, whereas the cumulative incidence of CNS failure (isolated and combined) was similar in all three trials. In NHL-BFM95, there was a trend for earlier occurrence and a higher incidence of SMN in the first 5 years compared with NHL-BFM90 and NHL-BFM86.
Outcome of target patients in the Non-Hodgkin's Lymphoma (NHL) –Berlin-Frankfurt-Munster 95 trial (stage III or IV, CNS negative, sufficient response at induction day 33, and no prophylactic cranial radiotherapy) compared with the historical control of combined trials NHL-BFM90/86. (A) Event-free survival (pEFS) at 2 and 5 years, (B) disease-free survival (pDFS) at 5 years, and (C) overall survival (pOS) at 5 years. SE, standard error.
Cumulative incidences of isolated CNS relapses, combined CNS relapses, non-CNS relapses, second malignancies, and death unrelated to disease for (A) Non-Hodgkin's Lymphoma (NHL) –Berlin-Frankfurt-Munster 86 trial, (B) NHL-BFM90, and (C) NHL-BFM95. pEFS, probability of event-free survival; SE, standard error.
The 5-year probability of DFS (pDFS) rate was 88% ± 3% for the target group of NHL-BFM95 compared with 91% ± 2% for the historical control group of NHL-BFM90/86 (Fig 1B). Of 18 patients suffering from tumor failure in the target group of NHL-BFM95, 14 died of lymphoma, and four survived for a median of 3 years. For comparison, all 14 patients with tumor failure in NHL-BFM90/86 died. The overall survival probability of the target group in NHL-BFM95 was not statistically significantly different compared with the historical controls of NHL-BFM90/86 (Fig 1C).
When the results for the target group of NHL-BFM95 are compared with historical controls of NHL-BFM90 and NHL-BFM86 separately, then the pEFS at 5 years is highest in NHL-BFM90 (Fig 3). However, the difference is not statistically significant.
Event-free survival at 5 years in target patients of Non-Hodgkin's Lymphoma (NHL) –Berlin-Frankfurt-Munster (BFM) 95 trial (stage III or IV, CNS negative, sufficient response at induction day 33, and no prophylactic cranial radiotherapy) compared with the historical controls of NHL-BFM90 and NHL-BFM86 separately.
For T-cell LBL patients, the pEFS rate at 5 years was 80% ± 4% (n = 126), 92% ± 3% (n = 96), 82% ± 6% (n = 49), and 88% ± 3% (n = 145) for trials NHL-BFM95, NHL-BFM90, NHL-BFM86, and combined NHL-BFM90/86, respectively. The difference in pEFS rate between T-cell LBL patients in NHL-BFM95 compared with NHL-BFM90 was statistically significant (P = .03). The pDFS rate at 5 years for T-cell LBL patients was 87% ± 3%, 93% ± 3%, 89% ± 5%, and 91% ± 2% for trials NHL-BFM95, NHL-BFM90, NHL-BFM86, and combined NHL-BFM90/86, respectively. The difference in pDFS rate between trials was not statistically significant in all comparisons. For precursor B-cell LBL patients, the pEFS and pDFS rates at 5 years were 92% ± 5% and 96% ± 4% in NHL-BFM95 (n = 26), respectively, compared with 85% ± 10% and 93% ± 7% in combined NHL-BFM90/86 (n = 14), respectively (P > .05 in all comparisons).
To exclude a selection bias introduced by the definition of the target group, we analyzed pEFS and pDFS for all patients with stage III or IV CNS-negative disease. For these patients, pEFS was significantly lower in NHL-BFM95 compared with NHL-BFM90 (Fig 4A). However, pDFS at 5 years was not statistically significantly different between NHL-BFM95 and NHL-BFM90/86 (Fig 4B).
Outcome at 5 years for all lymphoblastic lymphoma patients with stage III or IV, CNS negative disease in Non-Hodgkin's Lymphoma (NHL) –Berlin-Frankfurt-Munster trials 95, 90, and 86. (A) Event-free survival; (B) disease-free survival.
DISCUSSION
In our previous study NHL-BFM90, an intensive ALL-type therapy, including CNS-preventative treatment based on corticosteroids, intrathecal MTX, intravenous high-dose MTX, and PCRT, provided a 90% EFS rate even for patients with advanced-stage T-cell LBL.5 CRT, however, carries significant late risks such as neurologic deficits, short stature, and especially SMN.11-27 Hence, the question of whether PCRT can be completely omitted from the treatment of CNS-negative patients without jeopardizing patient outcome is one important issue in the optimization of therapy. Therefore, in our trial NHL-BFM95, we tested, against the historical controls of the combined trials NHL-BFM90/86, whether chemotherapy alone, without PCRT, is not inferior to treatment including PCRT for CNS-negative patients with advanced-stage disease. Patients with insufficient response at day 33 of induction still received PCRT in NHL-BFM95. This decision was based on the observation of an increased risk of CNS relapses in ALL patients with initial poor response in the ALL-BFM trials.38
We are aware of the drawbacks of our study using a historical control design. Because of the rarity of the disease, however, either an unduly long period would have been necessary for accrual of a sufficient number of patients for a randomized trial within the BFM group or the study would require an international cooperation. At the time of planning NHL-BFM95, most other study groups did not include PCRT in their treatment protocols.1,4,6,7,10
We believe our study contains useful information for further optimization of treatment of patients with advanced-stage LBL. With 5-year pEFS rates of 80% ± 3% for the total group of 198 LBL patients and of 78% ± 3% for patients with advanced-stage disease, the overall results of NHL-BFM95 still compare favorably with other studies.1,3,4 As in our previous trials, NHL-BFM90 and NHL-BFM86,5,9 treatment outcome did not significantly differ between patients with stage IV disease and patients with stage III disease. For stage III or IV CNS-negative LBL patients with sufficient initial response at day 33 of induction, according to the predetermined criteria, treatment without PCRT in NHL-BFM95 was noninferior to treatment including PCRT in the trials NHL-BFM90/86.
However, our results have to be interpreted critically. When planning NHL-BFM95, we decided to compare the test group of patients in NHL-BFM95 with the historical control group of the combined trials NHL-BFM90/86 to increase the number of control patients. At that time, we did not yet know about the extraordinarily favorable final results of LBL patients in NHL-BFM90, which exceeded the results of NHL-BFM86, although not significantly. If we compare the pEFS and pDFS of the target group in NHL-BFM95 separately with pEFS and pDFS of the historical control groups of NHL-BFM90 and NHL-BFM86, the 5-year pEFS of the test group of NHL-BFM95 is comparable to the pEFS in NHL-BFM86 but lower than the pEFS in NHL-BFM90, especially for T-cell LBL patients. However, the probability of 5-year DFS was similar, even for the T-cell LBL patients. The incidence of CNS relapses was comparable between the test group of NHL-BFM95 and the control groups of NHL-BFM90 and NHL-BFM86. However, there was at least a trend for a lower cumulative incidence of non-CNS relapses in the target patients of NHL-BFM90 compared with the patients of NHL-BFM95 and NHL-BFM86.
The chemotherapy of the three consecutive trials was similar but not identical concerning the first phase of induction. The dose of daunorubicin was reduced from 40 mg/m2 for 4 days in NHL-BFM86 to 30 mg/m2 for 4 days in NHL-BFM90, and the dose-intensity over time of the combination chemotherapy was increased in NHL-BFM90 compared with NHL-BFM86 by condensation of induction therapy by 1 week. From NHL-BFM90 to NHL-BFM95, E coli l-asparaginase (Bayer) 10,000 U/m2 for 8 days was substituted by E coli l-asparaginase (Kyowa) 5,000 U/m2 for 8 days. The decision was based on pharmacokinetic and pharmacodynamic findings showing similar complete l-asparagine depletion with E coli l-asparaginase 10,000 U/m2 produced by Bayer compared with E coli l-asparaginase 5,000 U/m2 produced by Kyowa when applied in 3-day intervals in ALL patients.39,40 However, apart from deprivation of the malignant cells from l-asparagine, alternative mechanisms of action of l-asparaginase, such as induction of apoptosis via signal transduction, were reported.41 Therefore, the impact of the reduction of the dose of l-asparaginase in NHL-BFM95 on the cumulative incidence of non-CNS tumor failure is difficult to determine. There is only one published study so far testing the impact of l-asparaginase on treatment outcome in T-cell ALL and T-cell NHL. In this Pediatric Oncology Group trial, patients receiving weekly l-asparaginase 25,000 U/m2 during maintenance had a superior outcome compared with patients not receiving l-asparaginase.4 Because, in our trial NHL-BFM95, the dose of l-asparaginase was reduced in the first phase of induction, this could have resulted in a higher rate of patients with insufficient response at day 33 of induction compared with the preceding trials NHL-BFM90/86. To exclude such selection bias, we analyzed pEFS and pDFS for all patients with stage III or IV CNS-negative disease. The pDFS at 5 years was not statistically significantly different for these patients when NHL-BFM95 was compared with NHL-BFM90 and NHL-BFM86. However, an impact of the omission of PCRT on the increased incidence of non-CNS relapses in NHL-BFM95 compared with NHL-BFM90 cannot be completely excluded. In acute leukemia patients, an increase in BM relapses but not CNS relapses was observed after omission of PCRT.42,43 It is notable that, in three of the four late relapses observed in our T-cell LBL patients in the target group of NHL-BFM95 who did not receive PCRT, the BM was involved. Thus, whether the increased rate of non-CNS tumor failure in NHL-BFM95 compared with NHL-BFM90 was a result of the omission of PCRT or a result of the reduction of the dose of l-asparaginase in the induction therapy cannot be clearly differentiated. In the ongoing European intergroup trial on childhood and adolescent LBL, which is based on the strategy of NHL-BFM90, E coli l-asparaginase (Kyowa) 10,000 U/m2 was chosen.
The major rationale for testing whether PCRT can be omitted from our treatment strategy was to avoid the late risks of PCRT, especially SMN.25 However, in most studies analyzing the impact of PCRT on the risk of SMN, doses of 18 Gy or higher were used. It remains questionable whether conclusions from such studies can be transferred to PCRT 12 Gy as applied in NHL-BFM90. Remarkably, our data so far give no evidence for a diminished risk of SMN after omission of PCRT. On the contrary, the cumulative incidence of SMN was higher in the target patients of NHL-BFM95, who did not receive PCRT, than in the control group of NHL-BFM90/86, who did receive PCRT. Although the difference is still not statistically significant, one has to take into account the shorter follow-up time of the patients in NHL-BFM95. The reasons, especially for the occurrence of secondary acute myeloid leukemia as early as 4 months from diagnosis of LBL, remain obscure. Concerning this matter, recent reports describing immature T-cell ALL with myeloid potential and biologic characteristics similar to acute myeloid leukemia may be of special interest.44,45
For explicit clarification of the impact of PCRT 12 Gy on the long-term outcome of patients with advanced-stage LBL, a randomized trial would be necessary. Because of the rarity of the disease, this would require an international cooperation.
Appendix
Study Committee of Trial NHL-BFM95
W. Doerffel, Berlin; W. Ebell, Berlin; N. Graf; Homburg; H. Gadner, Vienna; G. Henze, Berlin; G. Janka-Schaub, Hamburg; T. Klingebiel, Frankfurt; St. Mueller-Weihrich, Munich; I. Mutz, Leoben; H.J.R. Parwaresch, Kiel; A. Reiter, Giessen; H. Riehm, Hannover; G. Schellong, Muenster; M. Schrappe, Hannover; and F. Zintl, Jena.
Reference Laboratories
Histopathology and immunohistochemistry. R. Parwaresch, Lymph Node Registry founded by the Society of German Pathologists, Institute of Hematopathology, University Kiel; A. Feller, Institute of Pathology, University Luebeck; M.L. Hansmann, Institute of Pathology, University of Frankfurt; P. Moeller, Institute of Pathology, University of Ulm; H.K. Mueller-Hermelink, Institute of Pathology, University Wuerzburg; H. Stein, Institute of Pathology, Free University Berlin; and I. Simonitsch, Institute of Pathology, University Vienna, Austria.
Immunophenotyping. W.-D. Ludwig, Berlin; W. Knapp, Vienna; and F. Niggli, Zurich.
Cytomorphology. A. Reiter, Giessen; W. Haas, Wien; and F. Niggli, Zuerich.
Principal Investigators
R. Mertens (Aachen); R. Angst (Aarau); A. Gnekow (Augsburg); R. Dickerhoff (St Augustin); P. Imbach (Basel); G.F. Wuendisch (Bayreuth); W. Doerffel (Berlin-Buch); G. Henze (Berlin, Charité); U. Bode (Bonn); W. Eberl (Braunschweig); I. Krause (Chemnitz); D. Moebius (Cottbus); Th. Wiesel (Datteln); B. Ausserer (Dornbirn); H. Breu (Dortmund); G. Weiβbach/M. Suttorp (Dresden, University); W. Kotte (Dresden); W. Weinmann (Erfurt): J.D. Beck (Erlangen); W. Havers (Essen); G. Mueller (Feldkirch); B. Kornhuber/Th. Klingebiel (Frankfurt); C. Niemeyer (Freiburg); F. Lampert/A. Reiter (Gieβen); M. Lakomek (Goettingen); C. Urban (Graz); H. Reddemann (Greifswald); St. Burdach (Halle); H. Riehm/K. Welte (Hannover); B. Selle/A. Kulozik (Heidelberg); N. Graf (Homburg); C. Tautz (Herdecke); F.M. Fink (Innsbruck); F. Zintl (Jena); G. Nessler/W. Dupuis (Karlsruhe); H. Wehinger (Kassel); R. Schneppenheim/M. Suttorp (Kiel); H. Meβner (Klagenfurt); M. Rister (Koblenz); F. Berthold (Koeln); W. Sternschulte (Koeln); M. Domula (Leipzig); I. Mutz (Leoben); G. Schmitt (Linz); O. Stoellinger (Linz); L. Nobile (Locarno); P. Bucsky (Luebeck); H. Ruetschle/B. Selle (Ludwigshafen); U. Caflisch (Luzern); U. Mittler (Magdeburg); W. Scheurlen (Mannheim); H. Christiansen (Marburg); W. Tillmann (Minden); S. Mueller-Weihrich (Muenchen, Technical University); C. Bender-Goetze (Muenchen); H. Juergens (Muenster); O. Schofer (Neunkirchen); A. Jobke (Nuernberg); U. Schwarzer (Nuernberg); G. Eggers (Rostock); R. Geib-Koenig (Saarbruecken); H. Grienberger (Salzburg); H. Haas (Schwarzach); F.J. Goebel (Siegen); R. Ploier (Steyr); J. Treuner (Stuttgart); R. Schumacher (Schwerin); A. Feldges/J. Greiner (St. Gallen); H. Rau (Trier); D. Niethammer (Tuebingen); K. Debatin (Ulm); H. Gadner (Wien St. Anna Kinderspital); F. Haschke (Wien); A. Dohrn (Wuppertal); J. Kuehl (Wuerzburg); and F. Niggli (Zurich)
Pathologists
H. Mittermeyer (Aachen); B. Stamm (Aarau); R. Backmann (Augsburg); H. Ohnacker (Basel); M. Stolte (Bayreuth); H. Stein (Berlin); W. Schneider (Berlin); H.J. Foedisch, (Bonn); R. Donhuisen (Braunschweig); J.O. Habeck (Chemnitz); P. Stosiek (Cottbus); E.W. Schwarze (Dortmund); M. Mueller (Dresden); V. Becker (Erlangen); L.D. Leder (Essen); M.L. Hansmann (Frankfurt); H.E. Schäfer (Freiburg); W. Schultz (Gieβen); E. Kunze (Goettingen); C. Schmid (Graz); G. Lorenz (Greifswald); F.W. Rath (Halle); F. Kreipe (Hannover); F. Otto (Heidelberg); K. Remberger (Homburg); G. Mikuz (Innsbruck); D. Katenkamp (Jena); W. Gusek (Karlsruhe); O. Klinge (Kassel); R. Parwaresch (Kiel); W. Wagner (Klagenfurt); F. deLeon (Koblenz); H.P. Dienes (Koeln); C. Wittekind (Leipzig); G. Leitner (Leoben); M. Weber (Linz); E. Pedrinis (Locarno); K. Wegener (Ludwigshafen); A. Feller (Luebeck); J.O. Grebbers (Luzern); A. Roesser (Magdeburg); U. Bleyl (Mannheim); C. Thomas (Marburg); E. Jehn (Minden); K. Wurster (Muenchen, Technical University); U. Loehrs (Muenchen); W. Boecker (Muenster); P.H. Wuensch (Nuernberg); F. Hofstädter (Regensburg); H. Nizze (Rostock); H. Mitschke (Saarbruecken); O. Dietze (Salzburg); A. Hittmair (Schwarzach); G. Wittstock (Schwerin); D. Kunde (Siegen); A. Diener (St. Gallen); J. Feichtinger (Steyr); A. Bosse (Stuttgart); H. Maeusle (Trier); P. Kaiserling (Tuebingen); O. Haferkamp (Ulm); Th. Radasckiewicz (Wien); G. Fischer (Wilhelmshaven); H.K. Mueller-Hermelink (Wuerzburg); and T. Stallmach (Zurich).
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Author Contributions
Conception and design: Birgit Burkhardt, Martin Zimmermann, Wolfgang Doerffel, Georg Mann, Guenter Henze, Felix Niggli, Hansjoerg Riehm, Martin Schrappe, Alfred Reiter
Administrative support: Alfred Reiter
Provision of study materials or patients: Udo Kontny, Josef Vormoor, Wolfgang Doerffel, Georg Mann, Guenter Henze, Felix Niggli, Wolf-Dieter Ludwig, Dirk Janssen, Hansjoerg Riehm, Martin Schrappe, Alfred Reiter
Collection and assembly of data: Birgit Burkhardt, Wilhelm Woessmann, Georg Mann, Wolf-Dieter Ludwig, Dirk Janssen, Hansjoerg Riehm, Alfred Reiter
Data analysis and interpretation: Birgit Burkhardt, Martin Zimmermann, Udo Kontny, Josef Vormoor, Wolfgang Doerffel, Georg Mann, Guenter Henze, Martin Schrappe, Alfred Reiter
Manuscript writing: Birgit Burkhardt, Martin Zimmermann, Udo Kontny, Georg Mann, Guenter Henze, Felix Niggli, Wolf-Dieter Ludwig, Dirk Janssen, Martin Schrappe, Alfred Reiter
Final approval of manuscript: Birgit Burkhardt, Wilhelm Woessmann, Martin Zimmermann, Udo Kontny, Josef Vormoor, Wolfgang Doerffel, Georg Mann, Guenter Henze, Felix Niggli, Wolf-Dieter Ludwig, Dirk Janssen, Hansjoerg Riehm, Martin Schrappe, Alfred Reiter
Acknowledgments
We acknowledge the expert work of Edelgard Odenwald, Beatrice Puttkammer (cytomorphology), Ulrike Meyer, Nora Muehlegger (data management), and the external data safety and monitoring committee: Catherine Patte, MD, Villejuif, France; Maria-Gracia Valsecchi, PhD, Milano, Italy; Ian Magrath, MD, PhD, Bethesda, MD; and Joerg Michaelis, MD, Mainz, Germany.
Footnotes
-
Supported by the Deutsche Krebshilfe (Bonn, Germany) Grant No. M 109/91/Re1.
Presented in part at the 45th Annual Meeting of the American Society of Hematology, San Diego, CA, December 6-9, 2003.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
- Received April 7, 2005.
- Accepted October 21, 2005.
