The role of surgery in the treatment of pediatric B-cell non-Hodgkin's lymphoma

The Role of Surgery in the Treatment of Pediatric B-Cell Non-Hodgkin’s Lymphoma By Andishe Attarbaschi, Georg Mann, Michael Dworzak, Monika Trebo, Nora Mu¨hlegger, Alfred Reiter, Ernst Horcher, and Helmut Gadner, for the Austrian Cooperative Study Group Vienna, Austria

Background: Within the last 20 years, the role of surgery in the management of pediatric B-cell non-Hodgkin’s lymphoma (B-NHL) has changed substantially. Along with the assignment of risk-adjusted therapy and specific treatment protocols, surgical procedures have been restricted to defined situations including abdominal emergencies, diagnostic biopsy, total tumor extirpation and second-look operations. Methods: The authors retrospectively analyzed the effect of initial surgery and second-look operation on event-free survival (EFS) in 79 patients with B-NHL (abdominal primary, n ⫽ 57; head or neck tumor, n ⫽ 22). Furthermore, the prognostic significance of the stage of disease for the patients analyzed was evaluated. Results: Therapy results showed that the extent of resection did not have a significant influence on EFS and that stage of disease was of prognostic relevance only for patients with head and neck tumors. The number of patients who had a

N

ON-HODGKIN’S LYMPHOMA (NHL) represents the third most frequent cancer in childhood and usually is subdivided into 3 histologic subtypes: (1) B-cell NHL (65%), including both Burkitt’s lymphoma and diffuse large B-cell lymphoma; (2) lymphoblastic (20%); and (3) anaplastic large cell lymphoma (ALCL; 15%).1 Since the early 1980s it is well established that the choice of initial treatment in pediatric NHL depends on whether the disease is localized or disseminated and whether the histologic subtype is lymphoblastic or nonlymphoblastic (B-cell NHL, ALCL). Along with the assignment of specific treatment protocols of relatively short duration event-free survival (EFS) rates for children with B-cell NHL have dramatically increased within the last 20 years. With contemporary

From St Anna Children’s Hospital, Vienna, Austria; the University Children’s Hospital Gießen, Federal Republic of Germany; and the Department of Pediatric Surgery, University of Vienna, Austria. Address reprint requests to Helmut Gadner, MD, St Anna Children’s Hospital, Kinderspitalgasse 6, 1090 Vienna, Austria. Copyright 2002, Elsevier Science (USA). All rights reserved. 0022-3468/02/3710-0018$35.00/0 doi:10.1053/jpsu.2002.35417 1470

second-look operation (n ⫽ 12) was too small for statistical interpretation. Conclusions: Because tumor resectability is determined by the stage of disease, which is the superimposed predictor of prognosis, the influence of the extent of resection on EFS cannot be interpreted independently. Nevertheless, the following conclusions can be drawn: in patients with proven localized disease, total resections should be attempted, if not mutilating, to avoid intensified chemotherapy. However, if only partial resections seem feasible, surgical interventions should be restricted to the least necessary procedures for treating life-threatening local tumor effects and establishing a definite diagnosis. J Pediatr Surg 37:1470-1475. Copyright 2002, Elsevier Science (USA). All rights reserved. INDEX WORDS: B-cell non-Hodgkin’s lymphoma, extent of tumor resection, second-look operation, stage of disease.

risk-directed therapy, approximately 80% of patients with B-cell NHL now become long-term, disease-free survivors.1 Successful treatment of children with B-cell NHL usually requires an interdisciplinary approach. Because B-cell NHL has a high spontaneous cell turnover rate and thus responds readily to systemic combination chemotherapy, surgery as a part of lymphoma therapy has lost its former significance.1-3 Surgical interventions now are needed for complete resection in limited disease, diagnostic biopsies, management of life-threatening local tumor effects, and second-look operations, whereas chemotherapy is essential to achieve and maintain remission and high cure rates. The aim of this study was to analyze the effect of initial surgical procedures and stage of disease on outcome for patients with B-cell NHL treated according to protocols of the Berlin-Frankfurt-Mu¨nster (BFM) Group. Because patients with B-cell NHL often have a stable residual mass at some time during treatment course and guidelines for optimal management in this situation are lacking, we also evaluated the impact of the histologic findings of second-look surgery on the patient’s prognosis.4,5 Journal of Pediatric Surgery, Vol 37, No 10 (October), 2002: pp 1470-1475

SURGERY FOR B-CELL NHL

MATERIALS AND METHODS

Patients

1471

Table 1. Pretreatment sLDH Level of the 57 Patients With Abdominal and 22 Patients With Head and Neck Tumors Pretreatment sLDH

From October 1986 to March 2000, 85 previously untreated Austrian children and adolescents below 20 years of age with B-cell NHL were enrolled into 3 consecutive multicenter trials of the BFM Group (study entry criteria did not differ between the 3 NHL trials): NHL-BFMAustria(A) 86 (n ⫽ 16), NHL-BFM-A 90 (n ⫽ 31), and NHL-BFM-A 95 (n ⫽ 38).6,7 Fifty-seven (67%) patients had an abdominal tumor, 22 (26%) had a tumor confined to the head or neck, and 6 (7%) patients had other primary sites of involvement (mediastinum, n ⫽ 4; bone, n ⫽ 1; axillary lymph nodes, n ⫽ 1). All patients were treated according to the BFM protocols after informed consent was obtained from either the patient’s parents or legal guardians.

Methods Diagnosis and staging. Histologic diagnosis of B-cell NHL was based on the criteria of the updated Kiel-classification in trials BFM-A 86 and BFM-A 90 and on the criteria of the Revised European American Lymphoma (REAL) classification in trial BFM-A 95 and reviewed centrally.8,9 Cytomorphologic diagnosis of B-cell NHL was based on the French-American-British (FAB) criteria, and immunophenotypic as well as cytogenetic analyses were performed as described elsewhere.6,7,10 Staging procedures were performed according to the St Jude Staging System, and restaging procedures were performed after each course of chemotherapy until achieving complete remission (CR).11

Treatment Initial surgery and second-look operation. Patients were candidates for complete surgical resection (1) if they had localized disease; (2) if the operation could be performed easily and safely without any significant delay in beginning chemotherapy, without any functional impairment; and (3) if the histologic examination of suspicious adjacent lymph nodes showed no involvement. In patients without completely resectable disease, initial surgery included the least invasive procedure to establish the diagnosis. Partial resections were defined as surgical procedures leaving gross or minor residuals in situ. Second-look operations were not performed therapeutically, but to differ between necrotic residuals and residuals containing vital tumor cells. Functional impairment and delay in continuing chemotherapy had to be avoided. Chemotherapy. Treatment stratification and protocols used for patients with B-cell NHL in trials BFM-A 86 and BFM-A 90 have been published elsewhere.6,7 Study design and treatment protocols used in trial BFM-A 95 have not been changed significantly, compared with the previous 2 trials and were as follows: Patients were stratified into 4 risk groups (RG). Risk group 1 included all patients with a complete resection of the tumor. Patients enrolled in RG 2 had a lymphoma that could not be completely resected and that was of stage I to III with a serum lactate dehydrogenase (sLDH) level less than 500 U/L. Patients enrolled in RG 3 had a nonresectable lymphoma with either stage III plus sLDH 500 to 1,000 U/L or stage IV disease plus sLDH less than 1,000 U/L. Children enrolled in RG 4 had a nonresectable lymphoma with either stage III or stage IV disease plus sLDH of more than 1,000 U/L. Reference ranges (SI) of sLDH were as follows: less than 1 year, 170 to 580 U/L; 1 to 9 years, 150 to 500 U/L; 10 to 19 years, 120 to 330 U/L.12 In Table 1 the pretreatment sLDH values of all patients are presented. Treatment of patients consisted of 2, 4, 5, and 6 therapy courses in RG 1, 2, 3, and 4, respectively. Therapy courses included the following drugs: cyclophosphamide (200 mg/m2), methotrexate (1 g and 5 g/m2),

Abdominal tumors Abnormal sLDH Normal sLDH Not available Head and neck tumors Abnormal sLDH Normal sLDH Not available

Patients (%)

19 (33) 19 (33) 19 (33) 4 (18) 10 (45) 8 (36)

cytarabine (150 mg and 3 g/m2), ifosphamide (800 mg/m2), etoposide (100 mg/m2), vincristine (1.5 mg/m2), vindesine (3 mg/m2), Adriamycin (25 mg/m2) and dexamethasone (10 and 20 mg/m2). Response evaluation and salvage therapy. In trials BFM-A 86, 90, and 95 incomplete tumor regression, persistence of greater than 5% blast cells in bone marrow (BM) and any tumor cells in the cerebrospinal fluid (CSF) after 2 courses of chemotherapy were considered as initial treatment failure.6,7 In trial BFM-A 86, a second-look operation was indicated for patients with a residual mass after 2 therapy courses. Patients with total resection of the residual tumor continued with chemotherapy. Patients with nonresectable residuals containing vital tumor cells (proven by biopsy) received local radiotherapy at a dose of 30 Gy. Patients neither indicated for complete resection nor for local radiotherapy were candidates for an autologous or allogeneic BM transplantation.6 In trials BFM-A 90 and BFM-A 95, patients with a residual mass after the third (BFM-A 90) and fifth (BFM-A 95) therapy course were scheduled for a second-look operation.7 If the residual mass contained vital tumor cells, patients received high-dose chemotherapy with autologous stem cell rescue, whereas in patients with a necrotic or fibrotic residual, chemotherapy was continued. Local radiotherapy was not performed.

Definition of Relapse A relapse was diagnosed if any extramedullary, histo- or cytologically documented lymphoma reappeared or if more than 5% blast cells were detected in the BM after having achieved CR.6,7

Statistical Analysis The Kaplan-Meier method was used to calculate the probabilities of EFS (pEFS), and comparison of different groups was performed by using the log-rank test.13,14 Event-free survival was defined as the time from diagnosis to the date of the first adverse event (death from any cause, tumor progress or relapse at any site) or to the date of last follow-up. Patients who did not achieve CR were assigned to a failure time of zero.

RESULTS

Overall Results of Trials BFM-A 86, BFM-A 90, and BFM-A 95 Clinical and laboratory features of the patients with B-cell NHL included in trials BFM-A 86, 90, and 95 are presented in Table 2. Fifteen of 16 (94%) patients enrolled in trial BFM-A 86, 30 of 31 (97%) patients enrolled in trial BFM-A 90, and 37 of 38 (97%) patients included in trial BFM-A 95 achieved a complete clinical remission. In each trial only 1 patient relapsed (BM, n ⫽ 1; BM

1472

ATTARBASCHI ET AL

Table 2. Characteristics of the 85 Patients With B-Cell NHL Included in Trials BFM-A 86, 90, and 95 Characteristics

BFM-A 86

BFM-A 90

BFM-A 95

No. of patients Boys Girls Median age Range CNS disease BM involvement Stage of disease Stage I Stage II Stage III Stage IV Histology Burkitt’s lymphoma DLBCL Centrocytic B-NHL Centroblastic B-NHL Immunoblastic B-NHL B-NHL Lymphoblastic B-NHL Non-Burkitt’s lymphoma

16 12 (75%) 4 (25%) 7.3 yr 3.2-15.3 yr 2 (13%) 2 (13%)

31 25 (81%) 6 (19%) 8.9 yr 3.2-19.2 yr 0 4 (13%)

38 32 (84%) 6 (16%) 8.2 yr 1.6-17.3 yr 3 (8%) 5 (13%)

1 (6%) 7 (44%) 4 (25%) 4 (25%)

8 (26%) 5 (16%) 14 (45%) 4 (13%)

5 (13%) 9 (24%) 16 (42%) 8 (21%)

10 (63%) / 1 (6%) 2 (13%) 0 0 3 (19%) 0

24 (77%) / 0 2 (6%) 1 (3%) 3 (10%) 0 1 (3%)

28 (73%) 9 (24%) / / / 1 (3%) / /

Abbreviations: BFM, Berlin-Frankfurt-Mu¨ nster; A, Austria; CNS, central nervous system; BM, bone marrow; DLBCL, diffuse large B-cell lymphoma.

and CNS, n ⫽ 1; local relapse, n ⫽ 1), whereas 4 patients died in trial BFM-A 86 (therapy-related toxicity, n ⫽ 2; refractory disease, n ⫽ 1; death 9 years after diagnosis, cause unknown, n ⫽ 1), 2 in trial BFM-A 90 (relapse, n⫽1, refractory disease, n⫽1), and another 2 in trial BFM-A 95 (infection, n ⫽ 1; relapse, n ⫽ 1). The pEFS at 5 years was 81% ⫾ 10% in trial BFM-A 86 after a median follow-up of 8.0 years, and 94% ⫾ 4% in both studies BFM-A 90 and BFM-A 95 after a median follow-up of 4.7 and 2.3 years, respectively. The Prognostic Role of the Extent of Tumor Resection and of Stage of Disease Patients with an abdominal primary. In 38 of 57 (67%) patients with abdominal disease, an initial laparotomy was performed to establish the diagnosis, whereas no laparotomy was needed in 19 (33%) patients (biopsy, n ⫽ 14; biopsy and CSF, n ⫽ 1; ascites, n ⫽ 1; kidney puncture, n ⫽ 1; pleural effusion and BM, n ⫽ 1; pleural effusion, n ⫽ 1). In 18 (47%) of the 38 patients the initial laparotomy was an urgent operation owing to an abdominal emergency (intussuception, n ⫽ 11; bowel obstruction, n ⫽ 5; appendicitis, n ⫽ 1; intestinal perforation, n ⫽ 1), whereas in the 20 (53%) other patients the laparotomy was an elective procedure. The number of patients with total resection, partial resection, and biopsy only was comparable among the patients with an urgent (3, 11, 4, respectively) or elective operation (4, 10, 6, respectively).

Table 3. Relationship Between the Extent of Resection and Outcome for the 57 Patients With Abdominal and 22 Patients With Head and Neck Tumors Extent of Resection

Abdominal tumors Complete resection Partial resection Laparatomic biopsy Biopsy/puncture Head and neck tumors Complete resection Partial resection Biopsy

Patients No CR Relapse Death

57 7 21 10 19 22 5 1 16

pEFS

0 0 1 1

0 0 0 1

0 2 1 2

1.0 0.95 ⫾ 0.05 0.89 ⫾ 0.10 0.89 ⫾ 0.07

0 0 1

0 0 2

0 0 3

1.0 1.0 0.80 ⫾ 0.10

Abbreviations: CR, complete remission; pEFS, probability of eventfree survival.

Neither of the 7 patients with a complete resection of the abdominal primary tumor (stage I, n ⫽ 6; stage II, n ⫽ 1), of the 21 patients with laparotomy and partial resection, and of the 10 patients with laparotomy and biopsy only, suffered from recurrent disease. The only patient who relapsed was among the 19 patients with no laparotomy. Five patients died including 2 of 21 patients with partial resection (therapy-related toxicity, n ⫽ 1; cause of death unknown, n ⫽ 1), one of 10 patients with a laparotomic biopsy only (refractory disease), and 2 of 19 patients with no laparotomy at all (therapy-related toxicity, n ⫽ 1; infection, n ⫽ 1). The relationship between the extent of resection and outcome is shown in Table 3. Event-free survival rates between the 4 groups of patients did not differ significantly (P ⬎ .05). Among patients with stage I (n ⫽ 7), II (n ⫽ 13), and III (n ⫽ 27) disease, none had recurrent disease. The only patient with an abdominal primary who relapsed had stage IV disease. The pEFS did not differ significantly between the patients with stage I/II and stage III/IV diseases (P ⫽ .38; Table 4). All but one patient with stage I disease had a complete resection of the tumor. However, only 1 of the 40 patients with stage II and III disease had a total tumor extirpation. Neither among the Table 4. Relationship Between the Extent of Disease and Outcome for the 57 Patients With Abdominal and 22 Patients With Head and Neck Tumors Extent of Disease

Abdominal tumors Stage I Stage II Stage III Stage IV Head and neck tumors Stage I Stage II Stage III Stage IV

Patients No CR Relapse Death

57 7 13 27 10 22 6 6 5 5

2 0 0 2 0 1 0 0 0 1

1 0 0 0 1 2 0 0 1 1

5 0 1 3 1 3 0 0 1 2

pEFS

1.0 0.67 ⫾ 0.27 0.89 ⫾ 0.06 0.90 ⫾ 0.09 1.0 1.0 0.75 ⫾ 0.22 0.60 ⫾ 0.22

Abbreviations: CR, complete remission; pEFS, probability of eventfree survival.

SURGERY FOR B-CELL NHL

patients with stage II nor with stage III disease partial resection of the tumor was associated with an improved prognosis as compared with laparotomy and biopsy only or no laparotomy. Patients with a head or neck tumor. Among the 22 patients with a head and neck tumor, total resection of the tumor was performed in 5 patients. They all had stage I disease. One patient had a partial resection (stage II disease), whereas in 16 patients only a biopsy was performed. These 16 patients included all children with stage III and IV disease, only 1 patient with stage I, and 5 of 6 with stage II disease. All 2 patients with recurrent disease (stage III, n ⫽ 1; stage IV, n ⫽ 1) had a biopsy only to gain the diagnosis, and both of them died. Another patient with a biopsy only died from refractory disease. Event-free survival rates between patients with complete resection, partial resection, and biopsy only did not differ significantly (P ⬎ .05; table 3). However, stage of disease was of prognostic value, because stage I/II patients had a significantly better outcome than stage III/IV patients (P ⫽ .03; Table 4). The Prognostic Role of the Histologic Findings of Second-Look Surgery Of the 79 patients included in our study, 12 (15%) had a second-look operation, including 9 with an abdominal primary tumor and 3 patients with a head or neck tumor. In 2 of the 12 patients, no residual mass could be detected; among the other 10 patients, the residual mass was resected completely in 6, and multiple biopsies were taken in 4 patients. In 2 (biopsy, n ⫽ 1; complete resection, n ⫽ 1) of the 10 patients the resected tissue contained vital tumor cells, and in 8 patients the resected material consisted of fibrotic or necrotic tissue. Both patients with tumor persistence received intensified therapy (local radiotherapy, n ⫽ 1; BMT, n ⫽ 1) and were in complete clinical remission at the date of last follow-up. Among the remaining 8 patients with necrotic residuals one (biopsy) had a local relapse and died of refractory disease. Another patient who did not have a residual mass at second-look operation died 9 years after diagnosis; the cause of death remained unknown. DISCUSSION

Between 1986 and 2000, 85 Austrian patients with B-cell NHL have been treated according to the BFM protocols, and 77 (91%) were in complete clinical remission at the date of last follow-up. The pEFS at 5 years was 92% ⫾ 3% for the entire study population. This favorable outcome may suggest that the BFM risk group stratification and treatment strategy of using short and intensive chemotherapy courses offer the patients an excellent chance of EFS.

1473

The purpose of this retrospective study was to reevaluate the role of initial surgery and second-look operation in patients with B-cell NHL, including both patients with an abdominal primary tumor and patients with a tumor of the head or neck. We also analyzed the effect of the extent of disease on EFS. Our data show that patients with localized abdominal (n ⫽ 7) or head and neck disease (n ⫽ 5) and complete tumor extirpation had an excellent prognosis with only 2 (in trial BFM-A 90 and 95) to 3 (in trial BFM-A 86) chemotherapy courses, which neither included high-dose methotrexate (5 g/m2) nor high-dose cytarabine (2 g or 3 g/m2; pEFS, 100%). Despite this favorable outcome, we did not find that among children with abdominal disease, patients with complete tumor resection had a better prognosis than those with incomplete resection. Treatment results between patient cohorts with laparotomy and complete resection, laparotomy and partial resection, laparotomy and biopsy, or no laparotomy at all did not differ significantly. We therefore concluded that in patients with proven localized abdominal disease, total resections may be performed, if not mutilating, so that treatment regimens of higher intensity can be reserved for patients with disseminated disease. The opportunity to reduce the number of chemotherapy courses clearly benefits the patients by avoiding additional periods of acute toxicity with the risk of life-threatening infections and potential late risks of treatment. Because children with partially resected or biopsied tumors are treated identically according to the extent of disease and pretreatment sLDH level, partial resections neither add to survival nor prevent patients from entering the higher risk groups. Thus, in patients without completely resectable disease, surgical interventions should only consist of the examination of easily accessible extraabdominal manifestations to confirm the diagnosis. If a laparotomy is needed for diagnosis, surgery should rely on the least invasive procedure, such as a biopsy. Diagnostic laparoscopy might be an alternative to diagnostic laparotomy because it helps in avoiding unnecessary surgery and delay in beginning chemotherapy and shortens the operative period. When deciding about surgical procedures in patients with abdominal tumors, the pretreatment sLDH level also should be taken into consideration because this laboratory parameter usually is elevated as consequence of tumor cell break down. We found that in 50% of the patients with abdominal disease, the pretreatment sLDH concentration was increased abnormally and suggest that one should always consider B-cell lymphoma in the differential diagnosis of patients with an abdominal mass and increased sLDH level. In patients without extraabdominal manifestations, surgical interventions then

1474

ATTARBASCHI ET AL

should be limited to a diagnostic laparotomy or laparoscopy so that surgical complications can be avoided, and effective chemotherapy can be initiated immediately. Study results were almost similar for the patients with head and neck tumors, indicating that total resections of the tumor should only be attempted in strictly localized disease to reduce chemotherapy. In contrast, in patients without completely resectable disease, surgery may not have a therapeutic role but is necessary only for diagnostic purposes. The initial reports of Magrath et al2 and Kemeny et al3 dealing with the role of surgery in pediatric B-cell NHL suggested that total resection as well as tumor debulking greater than 90% significantly improved survival compared with patients with incomplete resection or biopsy only. Along with the introduction of more potent combination chemotherapy in the early 1980s and concomitant improvement of prognosis for patients with B-cell NHL the role of surgical interventions as a part of therapy changed substantially.15-18 Surgical interventions now are limited to the following situations: (1) complete resection if it can be performed without functional impairment; (2) exploratory and therapeutic laparotomy if patients present with an abdominal emergency (intussusception, intestinal perforation, suspected appendicitis or serious gastrointestinal bleeding); (3) diagnostic biopsy to obtain material for histologic, immunophenotypic, and cytogenetic analyses; (4) second-look operation; and (5) management of chemotherapy complications (perirectal abscesses, gastrointestinal hemorrhage). Radical surgical debulking in patients with extended disease is not recommended anymore. Recent reports of the 1990s still showed that patients with complete resection had a significantly better prognosis compared with patients with incomplete resection or biopsy.4,19,20 This was not the case in our study, suggesting that treatment stratification and protocols used for patients with incomplete resection or biopsy might be so efficient that the possible difference in outcome had been abolished. However, in view of the low power (79 subjects) of our study to detect significant results, these considerations may be speculative. Because the number of patients was too small for statistical interpretation, study results dealing with the role of second-look surgery in B-cell NHL could not be used to clarify prior findings of the BFM group and Helardot et al21 who reported that patients with a residual

mass containing vital tumor cells had an inferior prognosis compared with patients with fibrotic or necrotic residuals.5 Of note, treatment for both patients with active disease had been intensified to prevent disease progression. Extent of disease was of prognostic value for patients with head and neck tumors but not for patients with abdominal tumors. Reasons that might explain this discrepancy are the following. (1) Because of the overlap in definition of abdominal stage II and stage III disease, the group of stage III tumors did not only include patients with widely disseminated disease, but also patients with a more localized disease (ileocecal area) who were defined as stage III by the presence of additional locally restricted manifestations or ascites. These patients may have received an overtreatment that abolished a possible difference in outcome between stage I/II and stage III/IV diseases. (2) However, because the group of stage III tumors of the head and neck only included patients with distant metastases below the diaphragm, treatment results for stage III/IV tumors might not been altered by the inclusion of overstaged patients. Our interpretation of these findings is that the initial tumor mass still appears to be the overriding prognostic marker in B-cell NHL requiring innovative treatment approaches such as the use of anti-CD20 antibody in future clinical trials to further improve the prognosis for patients with advanced disease.22 However, because of the small number of patients analyzed, our results have to be validated on a larger cohort of patients. Conclusively, as tumor resectability is determined by the stage of disease, the influence of the extent of resection on EFS cannot be interpreted independently. Despite this fact and the retrospective character of the current study, we may conclude that surgery as a part of lymphoma therapy still plays an important role in exactly defined situations such as complete resection in localized disease and diagnostic biopsy, whereas chemotherapy represents the cornerstone in treatment of B-cell NHL, which offers to the patients an excellent chance of long-term, disease-free survival. ACKNOWLEDGMENTS The authors thank the principal investigators of the participating centers and are indebted to J. Regelsberger, MS, for documentation, and to U. Po¨ tschger, MS, for data management and statistical analyses.

REFERENCES 1. Sandlund JT, Downing JR, Crist WM: Non-Hodgkin’s lymphoma in childhood. N Engl J Med 334:1238-1248, 1996 2. Magrath IT, Lwanga S, Carswell W, et al: Surgical reduction of tumor bulk in management of abdominal Burkitt’s lymphoma. Br Med J 2:308-312, 1974 3. Kemeny MM, Magrath IT, Brennan MF: The role of surgery in

the management of American Burkitt’s lymphoma and its treatment. Ann Surg 196:82-86, 1982 4. Reiter A, Zimmermann W, Zimmermann M, et al: The role of initial laparotomy and second-look surgery in the treatment of abdominal B-cell non-Hodgkin’s lymphoma of childhood. A report of the BFM Group. Eur J Pediatr Surg 4:74-81, 1994

SURGERY FOR B-CELL NHL

5. Surbone A, Longo DL, DeVita VT, et al: Residual abdominal masses in aggressive non-Hodgkin’s lymphoma after combination chemotherapy: Significance and management. J Clin Oncol 6:18321837, 1988 6. 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 BFM Group. J Clin Oncol 13:359-372, 1995 7. 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 BFM Group Trial NHL-BFM 90.Blood 94:3294-3306, 1999 8. Stansfeld AG, Diebold J, Kapancy Y, et al: Updated Kiel classification for lymphomas. Lancet 1:292-293, 1988 9. Hiddemann W, Longo DL, Coiffier B, et al: Lymphoma classification—The gap between biology and clinical management is closing. Blood 88:4085-4089, 1996 10. Bennet JM, Catovsky D, Daniel MT, et al: Proposals for the classification of the acute leukemias. French-American-British(FAB) cooperative group. Br J Haematol 33:451-458, 1976 11. Murphy SB: Classification, staging and end results of treatment in childhood non-Hodgkin’s lymphoma: Dissimilarities from lymphomas in adults. Semin Oncol 7:332-339, 1980 12. Laboratory medicine, drug therapy and reference tables, in Behrmann RE, Kliegmann RM, and Jenson HB (eds): Nelson Textbook of Pediatrics (ed 16). Philadelphia, PA, Saunders, 2000, p 2179

1475

13. Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958 14. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163170, 1966 15. Whalen TV, La Quaglia MP: The lymphomas: An update for surgeons. Semin Pediatr Surg 6:50-55, 1997 16. Shamberger RC, Weinstein HJ: The role of surgery in abdominal Burkitt’s lymphoma. J Pediatr Surg 27:236-240, 1992 17. Ternberg JL: Changing role of surgery in childhood lymphomas. Semin Surg Oncol 9:541-544, 1993 18. La Quaglia MP, Stolar CJ, Krailo M, et al: The role of surgery in abdominal non-Hodgkin’s lymphoma: Experience from the Children’s Cancer Study Group. J Pediatr Surg 27:230-235, 1992 19. Stovroff MC, Coran AG, Hutchinson RJ: The role of surgery in American Burkitt’s lymphoma in children. J Pediatr Surg 26:12351238, 1991 20. Gahuamble DB, Khamage AS: Limitations of surgery in intraabdominal Burkitt’s lymphoma in children. J Pediatr Surg 30:519-522, 1995 21. Helardot PG, Wakim A, Kalifa C, et al: The place of surgery in the remission assessment of childhood abdominal malignant nonHodgkin’s lymphoma (NHL). Med Ped Oncol 17:322, 1989 (abstr) 22. Feuring-Buske M, Buske C, Unterhalt M, et al: Recent advances in antigen-targeted therapy in non-Hodgkin’s lymphoma. Ann Hematol 79:167-174, 2000