Management of pediatric non-Hodgkin's lymphoma

Management of pediatric non-Hodgkin's lymphoma

Management Lymphoma of Pediatric Non-Hodgkin’s J.-Y.Blay, D. Louis, E. Bouffet, P. Thiesse, P. Biron, M. C. Favrot, M. Brunat-Mentigny, T. Philip ...

1MB Sizes 0 Downloads 10 Views

Management Lymphoma

of Pediatric Non-Hodgkin’s

J.-Y.Blay, D. Louis, E. Bouffet, P. Thiesse, P. Biron, M. C. Favrot, M. Brunat-Mentigny,

T. Philip

S UMMA R Y. Over the past 15 years, significant progress has been made in the understanding of the molecular mechanisms involved in malignant transformation of lymphocytes as well as in the management and treatment of non-Hodgkin’s lymphoma (NHL) in childhood. Cyto-histological classifications and immunophenotyping of different types of NHL have contributed to the characterisation of three major subtypes of NHL in children i.e. Burkitt’s lymphoma (BL), lymphoblastic lymphoma (LL) and large cell lymphoma (LCL). Precise staging of the disease at diagnosis is necessary before the onset of the treatment and should be performed as quickly as possible. Presence of bone marrow and central nervous system (CNS) involvement are major prognosis criteria. In most cases, surgery has no therapeutic role and is required only for diagnostic procedures. Similarly, several studies have demonstrated that irradiation of various sites including the CNS does not improve survival. Thus, NHL patients are usually treated with chemotherapy alone. BL and LL have distinct clinical presentations and require completely different chemotherapy protocols. After comparable induction phases with intensive chemotherapy regimens, the former is usually treated with a short consolidation phase while the latter receives a long lasting consolidation consisting of intermittent chemotherapy for at least one year. The prognosis of stage I-II, and III-IV bone marrow negative NHL of children is excellent with respectively 95% and 75% long term survival. However, patients with concomittent CNS and bone marrow involvement in both histological subtypes have a considerably worse prognosis with only 30% long term survival. Treatment of relapse is currently under investigation and usually includes high dose chemotherapy regimen followed by bone marrow transplantation in patients experiencing a response after second line conventional dose chemotherapy. Cure after relapse is achieved in 2540% of the cases.

Since 1955, the understanding of the molecular mechanisms involved in the transformation of normal lymphocytes into lymphoma cells has considerably improved. ‘-’ In parallel, considerable progress were made in the classification and treatment of nonHodgkin’s lymphoma (NHL) in children.‘j NHL are J.-Y. Blay, E. Boufiet, P. l’biesse, P. Biioo, M.C. Favrot, M. Bmnat-Mentigny, T. Philip, Pediatric Oncology Department, Cen-

tre Lkon Bbrard, 28 rue Lahnec, 69008 Lyon Cedex 03, D. Louis, Surgery Department, Hospital Debrousse, Rue Soeur Bouvier 69005 Lyon, France. Reprint requests to: T. Philip. Blood Reviews (1991) 5,90-91 0 1991 Longman GroupUK Ltd

constantly of the diffuse type and belong to the high grade lymphoma of the Working formulation.7 Today, NHL in childhood can be cured with chemotherapy alone in more than 75% of the cases. However, treatment of NHL requires a precise cytological diagnosis and clinical staging in order to identify patients with bad prognosis criteria. Current concepts in the management of the three major types of NHL of children, i.e. Burkitt’s lymphoma (BL) lymphoblastic lymphoma (LL) and large cell lymphoma (LCL), will be presented.

BLOOD

Burkitt’s Lymphoma Burkitt’s lymphoma (BL) was characterized 30 years ago by Dennis Burkitt, an English surgeon, who reported on an unusual childhood tumor in Africa, which now bears his name.’ This tumor had the epidemiological features of an infectious disease.2 Subsequently, using electron microscopy, Epstein, Achong and Barr described a previously unclassified herpes-like virus from a BL cell line ‘EBl’,3*4 which was then supposed to cause BL. It is now well known that Epstein Barr virus (EBV) is a cofactor rather than a cause of BL.’ Subsequently, non-African forms of BL were discovered in Europe and in the USA. BL is the most frequent histological subtype of NHL of children in European Countries.6

Morphology, Immunology, Cytogenetics, and Molecular Biology of Burkitt’s Lymphoma

In the International Working Formulation (IWF),7 BL is recognised as a small, non-cleaved, diffuse high grade lymphoma and characterized by a monotonous proliferation of cells, lo-25 microns in diameter, with round to oval nuclei, and 2-5 prominent basophilic nucleoli. The cytoplasm is deeply basophilic, scanty, with several clear vacuoles, some of which contain neutral lipid as recognised by specific stains. Mitotic features are seen in approximately 4% of the neoplastic cells, and in association with this high growth fraction, one often observes numerous pycnotic cells and ‘starry sky’ macrophages containing nuclear debris.’ BL is a monoclonal proliferation as indicated by the selective expression of one of the two Ig light chains, kappa or lambda. BL is derived from normal B-cells. achieving an antigen independant stage of maturation. Most often, surface IgM is expressed, rarely associated with IgD. Cells expressing IgG or IgA with or without IgM are exceptional, and BL cells only express intracytoplasmic IgM in 30% of the cases with or without light chain. All BL express HLA DR and most CD lo,* but not the terminal deoxyribonucleotide transferase, clearly differentiating BL from acute common lymphoblastic leukemia.g The majority of cell lines derived from African and North African BL express EBV receptors (CD21) and may derive from the follicular center cell of lymph nodes.* BL malignant cells possess non-random chromosomal translocation involving constantly 8q24 in both endemic and non-endemic disease.“-12 This translocation is directed most often to Ig light or heavy chain on chromosome 14 -t(8;14) (q24;q32), chromosome 22 -t@q24;22qll)- or chromosome 2 -t(2p12;8q24)-(10-13). Five percent of cases have no specific translocation but all bear a 6q- marker. Rearrangements of chromosomes 1 and 7 have also been described.12 The breakpoint on chromosome 14 (14q32) corresponds to the gene of the heavy chain of immunoglobulin. The light chain lambda gene sits

REVIEWS

91

on break point 22q12, light chain kappa gene on break point 2~12. In 1982, it was demonstrated that BL cells expressed either kappa or lambda light chain if translocation (8;14) was present, but only kappa if chromosome 2, and only lambda if chromosome 22 were involved.” Dalla Favera showed that the translocation involved the proto oncogene c-myc located on 8q24, fused to the regulatory sequences of Ig chain genes.” This translocation results in an abnormal expression of c-myc, involved in the process of malignant transformation. Clinical Characteristics and Epidemiology

BL predominantly occurs in the first two decades of life. The clinical and epidemiological features will be described separately for endemic and non-endemic cases, although common points may be found. Sex ratio is 2-3 boys/l girl in both endemic and epidemic regions. 13*14 The endemic regions of BL include Equatorial Africa and New Guinea whereas North Africa and South America are regions of intermediate incidence.g The annual incidence of African BL is 2.2-3.8 cases/100 000 inhabitants, and peak incidence is between 5 and 8 years of age in central Africa, and 4 to 5 years of age in north Africa. Initial clinical findings include jaw lesion (58%), abdominal disease is observed also in 58%, involving the retroperitoneum, mesentary and omentum more often than the bowel itself so that presentation as intussusception, perforation or intestinal obstruction is rare. Peripheral lymph node involvement is uncommon (less than 1%). Bone marrow involvement is present initially in 12% of cases, and CNS in 30%, the latter is often associated with extra-abdominal tumor, rarely with bone marrow infiltration. CNS involvement is documented by CSF cellularity, cranial nerve palsies, or epidural mass with cord compression.13 The annual incidence of non-African BL in children below 16 years of age in western countries is about 0.1-0.3 cases / 100 000, twenty to forty times lower than in endemic regions.6 BL represents 40 to 45% of non Hodgkin malignant lymphomas in children. There is no clear relationship between the incidence of BL, geographical or climatical conditions, and/or viral association.14 The most common initial presentation6sr4 is abdominal tumor (70-90%) rapidly growing, with a high incidence of abdominal complications: intussusception, gut obstruction and perforation (20%). Peripheral lymph node presentation is unfrequent (20%) and jaw tumor rare (7-18%). Involvement of the pharynx is reported in approximately 5-10% of the cases, and exclusively leukemic presentation with circulating tumor cells but without associated mass lesion is specific for non-endemic BL. Bone marrow infiltration associated with other organs involvement is more frequent than in endemic patients (20%).

92

MANAGEMENT

OF PEDIATRIC NON-HODGKIN’S

LYMPHOMA

CNS infiltration occurs as meningeal infiltration, cranial nerve palsies, or epidural mass with cord compression in 20% of the cases. Two out of three patients with CNS involvement also present with bone marrow involvement.6*‘4 Initial Diagnosis and Staging of Burkitt’s Lymphoma The treatment of BL is an oncologic emergency. It requires prompt diagnosis and staging before initiation of a specific treatment. Diagnosis is usually performed by cytological examination thin needle aspiration. In every case, initial investigations should include: 1. A complete clinical examination searching for any superficial tumor and/or localization that could easily be biopsied or aspirated and for any neurological dysfunction, especially facial nerve palsies, skin hypoesthesia and pyramidal symptoms. 2. Rapid and simple imaging, such as abdominal ultrasonography and chest X-ray, avoiding any sophisticated investigation that would delay initiation of therapy. A three dimension measurement of the tumour should be obtained with ultrasonography. 3. Two bone marrow smears, with cytogenetical and immunological assessment. 4. Evaluation of the tumor burden should include uric acid level, urea, creatinine, potassium, calcium, phosphorous, LDH and B2 microglobulinemia. Lumbar puncture and examination of CSF cytospin must be done also before initiation of therapy. Several different clinical situations may occur: 1. In a child with extensive disease, BL is immediately suspected, and staging procedures are limited to tumour aspiration, CSF cytology, and bone marrow aspiration. Formal histology is not required when cytology is available for diagnosis. Results should be obtained within few hours, and chemotherapy started immediately in the first 24 hours. Tumor lysis syndrome is a major potential complication during induction therapy. l5 The use of a cytoreductive regimen containing low dose of chemotherapeutic agents before the beginning of induction therapy, such as the COP regimen in the SPOP protocol LMB81, associated with hyperhydration and uricase has rendered this complication very unfrequent. 2. BL may present as a local disease and may be small, sometimes accessible to complete and easy surgical removal. Staging in such cases is not an emergency and should be very extensive since underestimation may lead to sub-optimal treatment9 3. The diagnosis may be suspected only at time of surgery for abdominal emergency such as acute intussuseption. Resection of the mass should be considered only if it is likely to be complete, and not to delay initiation of chemotherapy. Aggressive surgery (debulking) should be avoided in the modem treatment of BL.i6 Early therapy is always essential, but its intensity

should be modulated according to the initial staging. Several staging systems have been described. At the present time, Murphy’s classification” is widely used and is based upon a more precise definition of locoregional tumor extension, and distant metastases (Table 1). The role of extensive multi-organ involvement (such as kidney or liver) as opposed to limited but unresectable disease can improve the clinical usefulness of the Murphy classification in stage III patients. 17*i8 Distinction between stage IV due to CNS or bone marrow invasion is also of value for predicting survival. 18--2o Treatment of Burkitt’s Lymphoma BL is an extremely chemosensitive tumor. For these reasons, large debulking surgery should be avoided. The potential morbidity associated with surgery is no longer acceptable.16 Initial surgery should be restricted to four well defined situations: 1. Biopsy for initial diagnosis when no other means (eg: examination of ascitic fluid, marrow, CSF, pleural effusion) are available. 2. Acute abdominal complications. 3. Complete removal of localized disease but excluding mutilating surgery. 4. Secondary surgery is also indicated for patients who still have a residual mass after induction therapy. A necrotic mass is found in 213 of these patients but residual viable tumour is an indication for intensification of chemotherapy.21 A role for radiation therapy to prevent local relapse has not been confirmed in recent studies.g*22 Moreover late sequelae may be expected and local radiation is no longer part of treatment, except in some cases of cord compression, although even in this situation chemotherapy is usually effective.

Table 1

Staging of childhood lymphoma according to Murphy’s classification Stage I Stage II

Stage III

Stage IV

Single tumor (extranodal). Single anatomical area (nodal) excluding mediastinum or abdomen. Single tumor (extranodal) with regional lymph node involvment. Primary gastrointestinal lymph node with or without involvment of associated mesenteric nodes only, grossly completely resected. On the same side of the diaphragm: a) two or more nodal areas b) two single (extranodal) tumors with or without regional lymph node involvement. On both side of the diaphragm: a) two or more nodal areas b) two or more extranodal tumors. All primary intrathoracic tumors. All extensive primary intra abdominal disease. All primary paraspinal or epidural tumors regardless of their sites. Any of the above with initial CNS or bone marrow involvement (<25%).

BLOOD

Local control is a minor problem in the treatment of BL, since l/3 of relapses are local, l/3 occur in the CNS, and the rest are multiorgan. At the present time, standard treatment for advanced BL comprises a high doses chemotherapy induction treatment followed by consolidation by sequential treatment. The complete treatment is usually performed in 4-5 months. The first evidence demonstrating that B- and T-cell lymphomas require distinct treatments was brought in 1983.23 In this study, the authors reported that the COMP regimen induced significantly better results than the LSA2L2 regimen for B lymphoma, although less than 50% of children with stage III and TV disease had long survival.23 In France, in 1980, the COPAD regimen, including Vincristine, Cyclophosphamide, Adriamycin and Prednisone, was accepted as the standard treatment for BL. This combination resulted in a 90% cure rate in stage I and II BL, but gave poor results in advanced disease with only 40 and 32% cure rate respectively in stage III and IV.2o In 1981, the LMB 0281 protocol was initiated. It was an intensive nine drugs regimen lasting one year with CNS directed treatment based on high doses methotrexate without CNS irradiation. The overall and event free survival was close to 75% survival in both stage III and IV patients without CNS involvement. l9 The CNS chemoprop h y laxis was efficient with an isolated CNS relapse rate of 1% and although long term follow up studies are needed to detect possible late effects of HDMTX, chemotherapy alone is now the standard therapy to avoid CNS relapses in BL.19 However toxic death occured in 10% of the cases. Based on these results, a randomized trial was conducted to study the possibility of reducing toxicity by shortening to 4 months the length of treatment without jeopardizing the survival rates. Induction treatment was unchanged and was followed for those patients in CR by a random allocation to either five additional courses of chemotherapy or two courses. Both treatments were equivalent with regard to event free survival (EFS) but morbidity was lower in the short arm.24 This multicenter study has thus demonstrated that advanced B-cell lymphoma including patients with B-ALL can be cured by a 4 month six drug chemotherapy regimen without irradiation. Several other studies have demonstrated the efficacy of short pulse regimen for B-cell lymphoma. Table 2 summarizes the current treatment for BL in France. In contrast, children with both initial CNS disease and bone marrow involvement have had a 20-30% cure rate even with modern regimens.23 The use of crania-spinal radiotherapy and consolidation, with pulses of high dose Arac was recently reported by the SFOP group with a 75% disease free survival rate which compares favorably with the results achieved by massive therapy (BACT or BEAM) with autologous bone marrow rescue.25q26 Patients with histologically documented residual tumor after four courses of chemotherapy have had their outcome dramatically improved with elective

Table 2 Chemotherapy regimen CNS-Burkitt’s lymphoma COP Vincristine 1 mg/m2 Cyclophosphamide 300 mg/m2 Prednisone 2 mg/kg Methotrexate 15 mg/m2 5 to I days res COPADEM 1 Vincristine 2 mg/m2 (max: 2 mg) Cyclophosphamide 250 mg/m2/12 Prednisone 2 mg/kg Doxorubicin 60 mg/mZ Methotrexate 15 mg/m2 Methotrexate 3 g/m2 Folinic acid 15 mg/6 h 2 to 3 weeks rest COPADEMZ Idem supra except for Vincristine 2 mg/m2 (max: 2 mg) Cyclophosphamide 500 mg/m2/12 2 to 3 weeks rest CYM Cytarabine 100 mg/m2 Methotrexate 3 g/m2 Folinic acid I5 mg/6 h Methotrexate 15 mg/m2 Cytarabine 30 mg/m2 2 to 3 weeks rest CYM (idem supra) 2 to 3 weeks rest Maintenance Therapy Vincristine 2 mg/m2 (max: 2 mg) Cyclophosphamide 500 mg/m2 Prednisone 2 mg/kg Doxorubicin 60 mg/m2 Methotrexate 15 mg/m2 Methotrexate 3 g/m2 Folinic acid 15 mg/6 h IV: intravenous; central nervous

REVIEWS

93

for Stage II and Stage II/IV

h

h

IV IV IV IT

DI Dl Dll7 Dl

IV IV IV IV IT IV IV

Dl D2-4 Dll6 D2 Dl and D6 DI D2-4

IV IV

Dl and D6 D2-4

IV IV IV IT IT

D2-6 Dl D224 D2 D6

IV IV PO IV IT IV IV

D2 DZ-3 Dl-5 D2 D2 Dl D2-4

PO: per OS; IT: intrathecal; system

D: day; CNS:

intensification of chemotherapy followed by autologous bone marrow rescue leading to a 70% cure rate as reported by the SFOP group.21 Studies are underway to assess if the response to the first course is a predictive factor of subsequent further outcome (SFOP LMB 89 study). Management of Relapses During the 1980s three questions were raised concerning the management of relapsing BL: 1. May dose effect be used in BL? After the demonstration of a dose response relationship by Appelbaum using the BACT regimen,27 several alternative protocols were proposed either increasing BCNU dose,29-30 as used in the modified BACT IGR schedule, or incorporating new agents in an attempt to reduce the toxicity of the BACT combination (Etoposide and Melphalan) as proposed in the BEAM protocol.25,28 The efficacy of total body irradiation has not been demonstrated in BL. Moreover the short term toxicity (pneumonitis and encephalopathy) and the expected late sequelae are arguments against this technique for children in whom cure is expected. 2. Which type of relapse benefits from massive

94

MANAGEMENT

OF PEDIATRIC

NON-HODGKIN’S

LYMPHOMA

therapy? Resistant relapses (patients who failed to respond to conventional chemotherapy preceding ABMT) may show responses but no cure. This category of patients remains an important group in which to consider phase I-II studies. By contrast, at least 25% of patients with sensitive relapses may be cured even when relapsing from modern protocols provided such a treatment is performed without delay.28,29. 3. Which type of marrow rescue should be used? The choice between autologous or allogenic BMT is still a matter of debate. The possible contamination of the bone marrow by malignant cells is an argument for allogenic BMT when a donor is available. However, recent reports have demonstrated the efficiency of immunomagnetic depletion to remove tumors cells from bone marrow.3o The antileukemic effect of graft versus host disease as seen in leukemia is not demonstrated in lymphomas although recent reports from Chicago and Seattle may renew interest in allogenic transplants.31 One challenge for the 1990s will be to increase salvage rates in previously heavily treated patients. Trials are under study to evaluate new combination of alkylating agents and also the possible value of adjuvant immunotherapy in this setting. Late Effects of Chemotherapy in Long Surviving Patients Two main late effects can be observed in patients who have completed therapy: reproductive dysfunction and a risk of second malignancies. The reproductive dysfunction in male is much more severely altered than in females.32 Several reports have demonstrated the major dose dependant toxicity of alkylating agents on male fertility. Male patients receiving more than 9 g/m2 of cyclophosphamide have a particularly high risk of sterility. 32 Women treated before the age of 20 years of age who did not receive abdominal irradiation usually have normal reproductive function. The risk of a second malignancy for children is not as great as in children with solid tumors or Hodgkin’s disease. The risk is higher in patients treated with alkylating agents.3s BL has become a curable disease in more than 80% of the patients. Further advances in management will include: avoiding the use of toxic drugs such as alkylating agents in good prognosis group in order to prevent sterility and secondary malignancies, and early detection of bad risk patients who may require either high dose chemotherapy with marrow rescue, and/or immunotherapy. Lymphohlastic

Lymphoma

of T-Cell Origin

Morphology, Immunology, Cytogenetics and Molecular Biology of Lymphoblastic Lymphoma Lymphoblastic T-cell lymphoma and T-cell acute lymphoblastic leukemias (T-ALL) are closely related

diseases. Both are indistinguishable, either cytologically, histologically or phenotypically. They have a high nuclear to cytoplasm ratio with a basophilic cytoplasm, sometimes associated with nuclear folds (convolution).33 However, contrasting to T-ALL which express early thymocytes markers, most of the LL cells express a T-cell phenotype of late or intermediate thymocytes, including the presence of the enzyme terminal deoxynucleotidyl transferase, involved in the rearrangement of T-cell receptor (TCR), and the presence of various T-cell membrane antigens (CD7, CD2, CD4, CD8, CD3).34 Non-random chromosomal translocations are commonly observed in LL and are likely to contribute to the process of malignant transformation. These translocations involve the TCR alpha or delta genes on chromosome 14(qll) or the TCRP gene on chromosome 7q32, fused to various protooncogenes (Ha-ras, c-myc).33

Epidemiology LL accounts for 35% of childhood lymphoma in Europe and is considered to be a disease of children and young adults. The peak incidence occurs during the second decade. The male/female ratio is 2:l. In contrast with BL, little is known about the etiology of this disease. Patients with immunodeficiency have no increased risks for LL. However, it has been proposed that low doses of radiation may be involved in the pathogenesis of LL.33

Clinical Characteristics Patients with LL commonly present (50-70%) with a mediastinal mass. Associated symptoms may include thoracic pain, dyspnea, sometimes associated with superior vena cava obstruction. Peripheral lymphadenopathy is found in 50-80%, mainly in cervical and supraclavicular sites. In contrast, abdominal involvement is unusual. Bone marrow involvement is frequent further indicating a considerable overlap between LL and ALL. Lymphoma is accepted when less than 25% blasts are found in the marrow, the presence of more than 25% blasts defines a leukemia. CNS involvement is uncommon at diagnosis, except in patients with bone marrow involvement. Meningeal and cranial nerve involvement are the most common features in this case. However, if CNS prophylaxis is not given, CNS involvement is likely to occur during therapy, particularly in patients with primary sites in head and neck.

Initial Diagnosis and Staging Initial staging for LL is similar to BL. In any case, initial investigations should include: A tumor aspirate or biopsy for cytological or histological examination and immunophenotyping of lym-

BLOOD REVIEWS

phoma cells is necessary before the onset of the treatment. 1. A complete clinical examination searching for any superficial tumor and/or localization that could easily be biopsied or aspirated and for any neurological dysfunction, especially facial nerve palsies, skin hypoesthesia and pyramidal symptoms. 2. A chest X-ray is mandatory, associated, if possible, with a computed tomographic scanning of the chest. An echocardiogram may be useful in examining the relation of the mediastinal tumor to the heart. 3. Two bone marrow smears are mandatory to detect involvement which have important implications for treatment. 4. Lumbar puncture for cytological examination of cerebrospinal fluid is also systematically performed before initiation of therapy. 5. Evaluation of the tumor burden should include uric acid level, urea, creatinine, potassium, calcium, phosphorus, LDH and 82 microglobulinemia. The staging system which is commonly used for childhood LL is Murphy’s staging classification also used for BL (Table 1). According to this staging system, thoracic LL, even with no extension in other sites is considered as a stage III. Prognostic criteria in LL comprise tumor burden evaluated by clinical staging, number of circulating lymphoblasts and serum levels of molecules secreted by tumor cells such as LDH. Recent reports suggest that the serum level of soluble IL-2R is a major prognostic marker.35

Treatment of Lymphoblastic Lymphoma Emergency

Treatment. Patients with LL may present initially with complications due to a large mediastinal tumor. Complications may include superior vena caval syndrome, pleuresia, dyspnea, cardiac arrhythmias and tamponnade which may require immediate treatment. Uric acid nephropathy is less frequent than in BL but requires the same treatment.15 The Place of Radiotherapy and Surgery, Surgery has no role in the treatment of LL except in establishing the diagnosis. Combinations of surgery and irradiation alone yielded only a very limited number of long term survivors. The role of radiotherapy in the treatment of LL is now limited. Several recent randomized studies have demonstrated the lack of beneficial effect of radiotherapy in combination with chemotherapy in patients with localized (stage I and II) and extensive disease (stage III and IV).36 The place of radiotherapy in LL appears to be restricted nowadays to the treatment of CNS involvement and of testicular tumors at the diagnosis or relapse. Treatment of CNS involvement is usually performed with a combination of cranial irradiation (18-24 Gy), high dose metho-

trexate and intrathecal chemotherapy and cytosine arabinoside).

95

(methotrexate

Chemotherapy. Chemotherapy protocols used for the treatment of children with LL are close to those used for the treatment of ALL in children. In contrast, adult patients are usually treated with regimen similar to those used for the treatment of intermediate grade lymphoma. Whereas the first trials of single or multiagent therapy were shown, in the early seventies, to be unable to provide long term survivors, significant progress have been achieved since 1976 in terms of complete remission rates and survival rate.33 The LSA2L2 protocol developed in the Memorial Sloan Kettering Cancer Center by Wollner et al is an intensive regimen combining ten different drugs administered in three phases, induction, consolidation and a prolonged maintenance, including CNS prophylaxis by intrathecal injections of methotrexate.37 This protocol (and its different versions) induces a 90% complete response rate with 60-90% long term survival in patients with stage III and IV disease. In contrast, the COMP regimen, a standard regimen for BL, produced significantly worse results in a randomized tria1.23 Similarly, the BFM protocol, initially developed for the treatment of ALL, produces an excellent response rate with a 78% disease free survival with a 48 month follow up in patients with stage III or IV disease. 33 However, more recent regimens, not based on an ALL protocol may have a comparable efficacy in randomized trials.33 At the present time, regimens derived from ALL protocols (i.e. LSA2L2 or BFM) are generally considered as the standard therapy for stage III/IV LL with a long term survival rate of 60-80%. Patients with histologically documented residual tumor after the phase of induction of chemotherapy have had their outcome improved with intensification of chemotherapy followed by autologous bone marrow rescue. An intensification is also proposed for patients with CNS involvement at the diagnosis who have a significantly poorer prognosis. In patients with limited stage I and II disease, intensive regimens derived from the LSA2L2 without radiotherapy, give excellent results with a 85-90% long term disease free survival.

CNS Prophylaxis. In the absence of prophylaxis the CNS is the site of relapse in 50% of the cases. Several methods for prophylaxis have been used. Due to the severe late adverse effect of cranial irradiation, intrathecal chemotherapy coupled with intermediate or high doses MTX appears to be the treatment of choice for CNS prophylaxis. Management of Relapses

Children relapsing of a LL are generally more resistent to second line chemotherapy. Some authors have proposed ALL relapse regimens as second line chemo-

96

MANAGEMENT

OF PEDIATRIC

NON-HODGKIN’S

LYMPHOMA

therapy for lymphoblastic leukemia on relapse. At the Centre Leon Berard, patients at relapse receive a second line chemotherapy with high doses VP16ARA-C or a MIME regimen. In case of partial or complete response to chemotherapy, they are considered as appropriate candidates for high dose chemotherapy followed by autologous bone marrow transplantation. Patients with a cyclophosphamideTBI conditioning regimen seem to show better survival than those treated with non TBI regimens.38

able results. Sixty five to 80% of patients achieve long term survival. Since both types of protocols give comparable results, most pediatric oncologists consider that LCL, at least of B type should be treated according to the regimen used for BL, whereas LCL of the T-cell type are treated as conventional lymphoblastic lymphoma. 3g Radiotherapy appears to increase toxicity but provides no therapeutic benefit. For these reasons, it has been largely abandoned. Treatment of Relapse

Late Effects They are similar to those mentioned for BL.32 Large Cell Lymphomas

of Children

Morphology, Immunology, Cytogenetics and Molecular Biology of Large Cell Lymphoma Large cell lymphomas (LCL) comprise 1530% of childhood NHL. They represent an heterogenous population still poorly understood, with regard to pathology and behaviour. LCL include probably at least two major categories: (1) large non cleaved lymphoma arising from germinal centers of secondary follicules and (2) immunoblastic lymphomas. As expected, most of the LCL express B-cell markers; however, immunoblastic types sometimes express T-cell markers. Both B- and T-LCL may express the CD30 antigen. Epidemiology The epidemiologic features of LCL are diverse. Some LCL, mainly immunoblastic, may arise in patients with congenital, infectious or iatrogenic immunosuppression. Spontaneous regression after withdrawal of immunosuppression have been observed in the latter case.39 Clinical Presentation LCL may arise in almost any tissue, including abdomen and peritoneum, skin, and brain in immunocompromised patients. Given the wide anatomical distribution of LCL, extensive studies are necessary to determine the precise distribution of the tumor. Staging must include the same biological and morphological examinations as indicated for BL and LL. The stage of the patient is determined according to Murphy’s classification, as for BL and LL. Treatment of Large Cell Lymphoma The optimal treatment of children with LCL has not been generally agreed upon because of the rarity and heterogeneity of these tumors. Protocols used in lymphoblastic lymphoma and BL produce compar-

In case of relapse, a second line chemotherapy can be proposed. LCL of B immunophenotype receive the same regimen than BL in relapse whereas LCL with a T-cell phenotype are generally treated like a LL in relapse. High dose chemotherapy followed by autologous bone marrow transpiantation could be performed in the same circumstances as described for BL.25 Conclusion

Since 1980, the progress of molecular biology techniques have allowed the precise molecular characterisation of gene alterations (oncogens, Ig and TcR genes) involved in the process of malignant transformation of normal lymphocytes. In parallel, treatment of malignant NHL of childhood has dramatically improved. At the present time, the progress in molecular biology have not lead to a modification of the clinical management of NHL of childhood which remains mainly empirical. Precise cyto-histological classification of lymphomas has resulted in the characterisation of low and high risk patients requiring distinct therapeutic approaches. Patients with CNS and BM involvement as well as relapsing patients still have a poor prognosis. The major goals of the future years will be: 1. To increase the cure rates of those patients probably through an intensification of chemotherapy (increase in the dose or intensity of the chemotherapy). 2. To precisely define subgroups of good prognosis patients requesting less aggressive treatment. The precise analysis of gene alterations in lymphoma cells of a given patient may have important clinical applications in this regard. References Burkitt D P 1959 A Sarcoma involving the jaws in african children. British Journal of Surgery 46 218-223 Burkitt D P 1969 Related disease related cause. Lancet ii: 122991231 Epstein M A 1985 Historical backgrounds; Burkitt’s Iymphoma and Epstein Barr virus-In: Lenoir G, O’Conor G. Olwenv C L M, eds. Burkitt’s Lymphoma: a human cancer model. IARC Scientific publications, Lyon, 17-27 Epstein M A, Achong B G, Barr Y M 1964 Virus particles in cultured lymphoblasts from Burkitt’s lymphoma. Lancet 702-703

i:

BLOOD 5. Lenoir G M, Bornkamm G W 1987 Burkitt’s lymphoma, a human cancer model for the study of the multistep development of cancer: proposal for a new scenario. Advance in Viral Oncology 7: 173-206 6 Philip T, Lenoir G M, Bryon P A et al 1982 Burkitt type lymphoma in France among non-Hodgkin malignant lymphomas in Caucasian children. British Journal of Cancer 45: 670-678 7 The National Cancer Institute sponsored study of classification of non-Hodgkin’s lymphomas: summary and description of a working formulation for clinical use. The non Hodgkin’s lymphomas pathologic classification project. 1982 Cancer 49: 2112-2135 J, Dare J F, 8 Favrot M C, Philip I, Philip T, Portoukalian Lenoir G M 1984 Journal of the National Cancer Institute 73: 841-847 In: 9 Magrath I 1990 Small non cleaved cell lymphoma. Magrath I (ed) The Non Hodgkin’s lymphomas. Edward Arnold/Hodder and Sloughton, London: pp 256-278 J L. Bernheim A. Beraer R 1982 10 Lenoir G M. Preud’homme Correlation between immunoglobulin light chain expression and variant translocation in Burkitt’s lymphoma. Nature 293: 474-476 R. Bregni M, Erikson J. Patterson D, Gallo R 11 Dalla-Favera C, Croce C M 1982 Human c-myc oncogene is located on the region of chromosome 8 but it translocated in Burkitt lymphoma cell. Proceedings of the National Academy of Science of the USA 79: 7824-7827 I2 Berger R, Bemheim A 1985 Cytogenetics of Burkitt’s lymphoma and leukemia: a review. In: Lenoir G M et al (eds) Burkitt’s lymphoma, a human cancer model. IARC Scientific publications. Lyon, pp 65-80 13 Nkrumah F K. Olwenv C L M 1985 Clinical features of Burkitt’s lymphoma: the african experience. In: Lenoir G M et al (eds) Burkitt’s lymphoma, a human cancer model. IARC publications, Lyon, pp 87-95 14 Philip T Burkitt’s lymphoma in Europe 1985 In: Lenoir G M et al (eds) Burkitt’s lymphoma, a human cancer model. IARC Scientific publications, Lyon, pp 107-l 18 15 Barriga F, Magrath I, Wilson W 1990 Complications in the management of non Hodgkin’s lymphoma. In: Magrath I (ed) The Non Hodgkin’s lymphomas. Edward Arnold/ Hodder and Sloughton, London, pp 227-238 16 Kaufmann B H, Burgert E H, Banks P M 1987 Abdominal Burkitt’s lymphoma: role of early aggressive surgery. Journal of Pediatric Suraerv 22: 67 I-674 17 Murphy S B 1980 Classification, staging and end results of treatment of childhood non Hodgkin’s lymphoma: dissimilarities from lymphomas in adults. Seminars in Oncology 7: 332-339 18 Philip T, Pinkerton R, Biron P et al 1987 Effective multiagent chemotherapy in children with advanced B cell lymphoma: who remains the high risk patients? British journal of Haematology 65: 159-164 19 Patte C, Philip T, Rodary C et al I984 Improved survival rate in children with stage III and IV B cell non Hodgkin lymphoma and leukemia using multi-agent chemotherapy: results of a study of 114 children from the French pediatric onoclogy society. Journal of Clinical Oncology 4: 1219-1226 20 Patte 6; Bernard A, Caillaux J M. Philip T, Robert A. Rodarv C 1989 Lvmohomes malins non Hodekiniens. In: ‘Cancers de I’enfant’.* Lemerle J (ed) FlammaGon Medecine Science, Paris, pp 288-3 13 21 Philip T. Hartmann 0, Biron P et al 1988 High dose therapy and autologous bone marrow transplantation in partial remission after first line induction therapy for difi‘use non Hodgkin’s lymphoma. Journal of Clinical Oncology 6: 11181124 22 Lemerle J 1985 Treatment of B cell non Hodgkin lymphomas of childhood in Europe: recent and on-going studies. In: Lenoir G M et al (eds) Burkitt’s lymphoma, a human cancer model. IARC Scientific publications, Lyon, pp 383-398 23 Anderson J R, Wilson J F, Jenkin D T 1983 Childhood non Hodgkin’s lymphoma: results of a randomised therapic trial

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

REVIEWS

97

comparing a four drug regimen (COMP) to a ten drug regimen (LSA2L2). New England Journal of Medicine 308: 559-565 Patte C, Philip T, Rodary C et al 199l High survival rate of advanced stage fl cell lymphomas and leukemias without CNS involvement with a short intensive polychemiotherapy: results from the French Pediatric Oncology Society of a randomized trial of 216 children. Journal of Clinical Oncology 9: 123-132 Philip T, Pinkerton R, Hartmann 0, Patte C, Philip I, Favrot M C 1986 The role of massive therapy with ABMT in Burkitt lymphoma. Clinical Hematology 15: 205-218 Gentet J C, Patte C, Quintana E et al 1990 Phase II of study of cytarabine and etoposide in children with refractory or relapsed Non-Hodgkin’s Lymphoma: A study of the SFOP. Journal of Clinical Oncology 8: 661-665 Appelbaum F R, Sullivan K M, Buckner C D et al 1987 Treatment of malignant lymphoma in 100 patients with chemotherapy, total body irradiation and bone marrow transplantation. Journal of Clinical Oncology 5: 1340-1347 Philip T, Hartmann 0, Zucker J M et al 1988 Treatment of relapses in the SFOP LMB 03-84 protocol. Role of bone marrow transplantation as a salvage therapy. XX meeting International Society of Paediatric Oncology, Trondheim abst 103 Philip T. Armitage 0, Spitzer G et al 1987 High dose therapy and autologous bone marrow transplantation after failure of conventional chemotherapy in adults with intermediate-grade or high grade non Hodgkin’s lymphoma. New England Journal of Medicine 316: 1493-1498 Philip I. Favrot M C. Combaret V et al 1986 Use of a liquid cell culture assay to measure the in vitro Burkitt cell elimination from the BM in preclinical and clinical procedure. In: Dicke K, Spitzer G, Zander A R (eds). Proceeding of the second international symposium on ABMT. Houston, Texas 341-347 Appelbaum F, Peterson F, Bolonesi B et al 1990 A prospective comparison of autologous vs. allogenic marrow transplantation for patients with malignant Lymphoma in 1st relapse or 2nd remission. Proceeding of the fifth international symposium on ABMT. Omaha. Nebraska, August abstract no 48, 104 Lemerle J, Dubousset J, Brauner R et al 1989 Sequelles chez les enfants gueris. Siquelles chez les enfants giris. In: Cancers de l’enfant. Lemerle J (ed) Flammarion Medecine Science, Paris. pp 173-217 Sandlund J, Magrath I 1990 Lymphoblastic lymphoma. In: Magrath I (ed) The Non Hodgkin’s lymphomas. Edward Amold/Hodder and Sloughton, London, pp 240-255 Bernard A, Boumsell L, Reinherz E L et al 1981 Cell surface characterisation of malignant T cells from lymphoblastic lymphoma using monoclonal antibodies: evidence for phenotypic differences between malignant T cells from acute lymphoblastic leukemia and lymphoblastic lymphona. Blood 57: 1105-1111 Pui C H, Ip S, Kung P et al 1987 High serum IL-2R levels are related to advanced disease and a poor outcome in childhood non Hodgkin’s lymphoma. Blood 70: 624-628 Link M P, Donaldson S S, Berard C W, Shuster J J. Murphy S B 1990 Results of treatment of childhood localized non hodgkin’s lymphoma with combination chemotherapy with or without radiotherapy. New England Journal of Medicine 322: 1169-I 174 Wollner N, Exelby P R, Lieberman P H 1979 Non Hodgkin’s lymphoma in children. A progress report on the original patients treated with the LSA2L2 protocol. Cancer 44: 1990-1999 Cabanillas F, Jagannath S, Philip T 1990 Management of recurrent or refractory disease. In: Magrath I (ed) The Non Hodgkin’s lymphomas. Edward Arnold/Hodder and Sloughton. London, pp 359-372 Magrath I, Kadin M 1990 Large cell lymphoma in children. In: Magrath I (ed) The Non Hodgkin’s lymphomas. Edward Amold/Hodder and Sloughton. London, 279-292