Tumor Necrosis Factor and Lymphotoxin Alfa Genetic Polymorphisms and Outcome in Pediatric Patients With Non-Hodgkin’s Lymphoma: Results From
http://www.100md.com
《临床肿瘤学》
the Department of Pediatric Hematology and Oncology, Children's Hospital, Hannover Medical School, Hannover
NHL-BFM Study Center, Department of Pediatric Hematology and Oncology, University Children's Hospital, Justus-Liebig-University Giessen, Germany
ABSTRACT
PURPOSE: To analyze the association of genetic variation within the tumor necrosis factor (TNF –308 [GA]) and lymphotoxin alfa (LT-a +252 [AG]) genes with outcome in non-Hodgkin's lymphoma of childhood and adolescence.
PATIENTS AND METHODS: Genotyping of the TNF –308 (GA) and LT-a +252 (AG) polymorphisms in patients (n = 488) enrolled onto the German-Austrian-Swiss multicenter trial NHL-BFM 95 from April 1996 to January 2000 was performed by polymerase chain reaction with subsequent restriction fragment length polymorphism analysis on DNA from tumor-free specimen.
RESULTS: In patients with Burkitt's lymphoma (BL) and B-cell acute lymphoblastic leukemia (B-ALL; n = 219, 211 eligible patients), patients carrying at least two variant alleles (high-producer haplotypes) had an increased risk of events: probability of event-free survival (pEFS) at 3 years was 81% (SE = 5%), compared with 91% (SE = 2%) in low-producer haplotypes (P = .018). In BL/B-ALL with high tumor load (lactate dehydrogenase [LDH] 500 U/L; n = 104), pEFS was 69% (SE = 8%) in high-producer versus 85% (SE = 4%) in low-producer haplotypes (P = .05). In multivariate analysis including risk factors for events (LDH 500 U/L, CNS involvement, methotrexate infusion regimen), TNF –308/LT- +252 haplotype kept prognostic relevance: patients with high-producer haplotypes had a 2.34-fold increase in risk of events (P = .048). The TNF –308 (GA) and LT- +252 (AG) polymorphisms were not associated with pEFS in lymphoblastic lymphoma (n = 101), anaplastic large-cell lymphoma (n = 67), or diffuse large B-cell lymphoma (n = 65), nor with therapy-related toxicity.
CONCLUSION: The TNF –308 (GA) and LT-a +252 (AG) polymorphisms were negative prognostic factors in pediatric BL/B-ALL. Among patients with serum LDH 500 U/L, haplotype analysis further determined patients at risk for events.
INTRODUCTION
Tumor necrosis factor (TNF) and lymphotoxin alfa (LT-) are cytokines of the tumor necrosis factor family that function as prominent mediators of immune regulation and inflammation.1-3 Both lymphokines have similar biologic activities, show 35% identity and 50% homology in amino acid sequence, and bind to the same group of cellular TNF receptors.1,2,4 The genes coding for TNF and LT- are located tandemly on the chromosomal region 6p21.3-21.1 and are closely linked to the HLA-B locus within a highly polymorphic region of the major histocompatibility complex.1,3-6
Several single nucleotide polymorphisms of the TNF gene have been described.7,8 Exchange of guanine by adenine at position –308 of the TNF promoter region (TNF2 allele) is associated with higher serum levels of soluble TNF.8-10 Approximately 60% to 70% of the white population are homozygous for the wild-type TNF1 allele, 30% to 40% are heterozygous, and 1.5% to 3% are homozygous for the variant TNF2 allele.11,12 The TNF –308 (GA) polymorphism has been associated with susceptibility to cerebral malaria, lepromatous leprosy, mucocutaneous leishmaniosis, ulcerative colitis, Crohn's disease, fatal meningococcal disease, and septic shock.6,13-17 A polymorphism in the coding region at position +252 of the LT- gene (AG) leads to different alleles of LT-, here referred to as LT- (10.5 kb) for the wild-type allele, and LT- (5.5 kb) for the variant allele.18 The LT- (5.5 kb) allele was associated with higher levels of soluble LT- in patients homozygous for LT- (5.5 kb).18 Within the white population, 40% to 45% are homozygous for the wild-type allele LT- (10.5 kb), 40% to 45% are heterozygous, and approximately 15% are homozygous for the variant allele LT- (5.5 kb).11,12 A strong association between the LT- (5.5 kb) allele and adverse outcome in autoimmune diseases (eg, Crohn's disease) has been described.13,19,20
Biologic actions of both cytokines as well as their association with autoimmune and severe inflammatory disorders have drawn interest to their potential role in the pathogenesis and prognosis of hematologic malignancies.21-27 However, data on the role of TNF –308 (GA) and LT- +252 (AG) polymorphisms in lymphoid malignancies are discussed controversially. Whereas some authors described a higher prevalence of TNF2/LT- (5.5 kb) haplotypes in patients with multiple myeloma compared with controls, others could not confirm these results.21 Although a potential contribution of the TNF –308 (GA) and LT- +252 (AG) polymorphisms to carcinogenic processes remains unclear, it is of interest that so-called high-producer haplotypes (ie, carriers of at least two variant alleles of TNF2 and/or LT- [5.5 kb]) have been associated with a worse prognosis in adult patients with non-Hodgkin's lymphoma (NHL).23 This association was strongest in diffuse large B-cell lymphoma.
Apart from possible associations of TNF and LT- polymorphisms with pathogenesis or prognosis of NHL, their association with severe infections and inflammatory reactions may have an impact on treatment and therapy-related toxicity in patients with NHL, thus delaying therapy and influencing prognosis.16 However, hardly any data exist on the putative role of TNF –308 (GA) and LT- +252 (AG) polymorphisms in chemotherapy-associated toxicity.
The present study examines the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with diagnostic NHL entities and treatment outcome in 488 unselected pediatric patients 18 years of age and younger treated on the multicenter therapy trial NHL-BFM 95, conducted by the Berlin-Frankfurt-Münster (BFM) study group. In addition, particular attention was paid to examine the association of these polymorphisms with therapy-related toxicity to exclude genotype-associated differences in therapy-related toxicity as relevant causes for differences in outcome.
PATIENTS AND METHODS
Patients
In trial NHL-BFM 95, patients were registered from 84 clinics in Austria, Germany, and Switzerland after informed consent was given by the patient, parents or legal guardians according to the Declaration of Helsinki. Approval of the study was obtained from the ethical committee of the principal investigator (A.R.) as well as the participating investigators. From April 1996 to January 2000, 684 patients 18 years of age and younger with newly diagnosed NHL were enrolled. Diagnosis was based on histopathology, cytology, immunology, and immunohistochemistry. NHL subtypes originally diagnosed according to the updated Kiel classification for NHL were reclassified on the basis of the WHO classification of hematologic malignancies.28,29 B-cell acute lymphoblastic leukemia (B-ALL) was diagnosed if the bone marrow smears showed at least 25% of blasts with typical French-American-British L3 morphology and immunology.28,30,31 Tumor slides of all patients were reviewed by the central reference pathology and/or cytology.
Of 684 enrolled patients, 488 patients (71.3%) were genotyped successfully for TNF –308 (GA) and LT- +252 (AG) polymorphisms. Analyses regarding clinical characteristics, distribution of lymphoma entities, and therapy-related toxicity were performed on the total group of 488 genotyped patients. Analyses regarding outcome and prognostic factors were performed on eligible patients only. Thirty-one of 488 genotyped patients were excluded for the following reasons: NHL was second malignancy (n = 1), NHL occurred after solid organ transplantation (n = 8), patient suffered from severe congenital immunodeficiency (n = 8), patient suffered from human immunodeficiency virus infection (n = 1), patient received chemotherapy before diagnosis of NHL (n = 8), therapy according to protocol NHL-BFM 95 was discontinued for nonmedical reasons (n = 1), initial diagnosis of type of NHL was incorrect, patient was assigned to wrong therapy group (n = 2), and patient was treated according to previous therapy protocol (n = 2).
Therapy
Patients with lymphoblastic lymphoma were treated according to an ALL-type protocol with slight modifications, as previously described (Fig 1, only therapy branches standard risk and medium risk are shown).32 Patients with B-cell lymphoma and B-ALL were stratified in four therapy branches of different therapy intensity according to stage and tumor mass at diagnosis, as previously described (Fig 1).33 Patients with anaplastic large-cell lymphoma (ALCL) were stratified according to stage and other risk factors such as skin involvement and lymphohistiocytic subtype (Fig 1). Therapy for B-cell NHL and ALCL (therapy group II and III) consisted of two to six 5-day courses of polychemotherapy, as previously described.33 Radiotherapy was not part of the protocol except for patients with lymphoblastic lymphoma and overt CNS disease.
In trial NHL-BFM 95, duration of methotrexate (MTX) infusion was randomly assigned in therapy group II for B-NHL within each therapy branch: patients received MTX either as 4-hour infusion or as 24-hour infusion with identical leucovorin (racemic folinic acid) rescue in both randomization arms.33
Treatment success was determined by probability of event-free survival (pEFS). Events were defined as follows: Death from any cause, tumor progression, and second malignancy. Progression was defined as growth of an incompletely resolved tumor or as recurrence of tumor at any site proven by biopsy.
Genotype Analysis
Genomic DNA was isolated from all available tumor-free bone marrow or peripheral-blood smears from patients enrolled onto trial NHL-BFM 95 as described before.27,34 Genotyping the TNF –308 (GA) and LT- +252 (AG) polymorphisms was performed by conventional polymerase chain reaction (PCR; Expand High Fidelity PCR System; Roche Diagnostics, Mannheim, Germany) followed by restriction fragment length polymorphism analysis with NcoI (New England Biolabs, Frankfurt, Germany). Primers used for amplification of a 107-bp fragment from the TNF promoter region were as follows: forward 5'-AGGCAATAGGTTTTGAGGGCCAT-3'; reverse 5'-TCCTCCCTGCTCCGATTCCG-3'. By introducing a single base change within the forward primer, an NcoI restriction site was created and used for analysis of the TNF –308 (GA) promoter polymorphism.9 In presence of the TNF1 allele, the amplified 107-bp fragment is cut into two fragments of 20 and 87 bp; the PCR product of the TNF2 allele remains uncut.9
The LT- +252 (AG) polymorphism was analyzed by PCR amplification of a 368-bp fragment using the following primers: forward 5'-CTCCTGCACCTGCTGCCTGGATC-3'; reverse 5'-GAAGAGACGTTCAGGTGGTGTCAT-3'. After NcoI restriction digest, the PCR product amplified from the LT- (10.5 kb) allele remains uncut. In presence of the LT- (5.5 kb) allele, the 368-bp PCR product is cut into two fragments of 133 and 235 bp.23 A random sample of 10% of the patients was genotyped twice; no discordances were observed regarding genotyping results.
Statistical Analysis
Associations of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with clinical characteristics (age, sex, type of NHL, stage, serum lactate dehydrogenase [LDH] at diagnosis, outcome, clinical risk factors for events, therapy-related toxicity, and outcome) were analyzed using the 2 or Fisher's exact test. Prognostic relevance of different parameters was examined by stepwise Cox regression analysis.35,36 Analysis of pEFS was performed according to Kaplan and Meier, with differences compared by the log-rank test.37,38 The pEFS was calculated from the date of diagnosis to the first event or to the date of last follow-up. Relapse, death in continuous complete remission, and second malignancy were regarded as events; failure to achieve remission was regarded as an event on day 1.
Statistical analyses were performed using the SAS program (SAS-PC, version 6.12; SAS Institute, Cary, NC). Follow-up was actualized as of January 1, 2004.
RESULTS
Patients
Of a total 684 pediatric patients enrolled onto trial NHL-BFM 95, 488 patients (71.3%) were genotyped successfully. Of these 488 patients, 101 patients (20.7%) suffered from lymphoblastic lymphoma, 219 patients (44.9%) from Burkitt's lymphoma or B-ALL, 65 patients (13.3%) from diffuse large B-cell lymphoma, 67 patients (13.7%) from ALCL, seven patients (1.5%) from peripheral T-cell lymphoma, and 29 patients (5.9%) from other NHL entities. Clinical characteristics such as tumor stage, tumor load, age, sex, proportion of excluded patients, or randomization in different MTX infusion regimens did not differ between genotyped patients and the remaining study population (data not shown).
Genotype and Clinical Characteristics
Of 488 patients genotyped for the TNF –308 (GA) polymorphism, 356 (73.0%) were homozygous for TNF1, 124 (25.4%) were heterozygous, and eight patients (1.6%) were homozygous for TNF2. Genotyping results of the LT- +252 (AG) polymorphism revealed 216 patients (44.3%) as being homozygous for LT- (10.5 kb), 226 (46.3%) being heterozygous, and 46 patients (9.4%) homozygous for LT- (5.5 kb). The observed genotype distribution was in agreement with the Hardy-Weinberg equilibrium and similar to previously described frequencies in healthy white populations.11
As previously described, a significant association of TNF1 and LT- (10.5 kb) as well as of TNF2 and LT- (5.5 kb) alleles was observed (P = .0001; Table 1). 23,27 As a result of this finding, statistical analysis regarding the association of genotype with clinical characteristics and outcome was performed using haplotypes: patients with at least two variant alleles (TNF2 and/or LT- [5.5 kb]; n = 146) were referred to as high-producer haplotypes ([LT- (10.5/10.5); TNF2/TNF2]; [LT- (5.5/5.5); TNF2/TNF2]; [LT- (5.5/10.5); TNF1/TNF2]; [LT- (5.5/10.5); TNF2/TNF2]; [LT- (5.5/5.5); TNF1/TNF2]; [LT- (5.5/5.5); TNF1/TNF1]). Patients with less than two variant alleles (n = 342) were referred to as low-producer haplotype carriers ([LT- (10.5/10.5); TNF1/TNF1]; [LT- (10.5/5.5); TNF1/TNF1]; [LT- (10.5/10.5); TNF1/TNF2]).
No association could be found between genotype or haplotype for TNF –308 (GA) and LT- +252 (AG) and NHL entities (Table 2). Similarly, there was no association between genotypes and clinical characteristics or previously described risk factors for failure such as sex and stage.32-35,39,40
Genotype Analysis and Outcome
Analysis of genotype and outcome did not show any association of genotype and prognosis for patients with lymphoblastic lymphoma, ALCL, or diffuse large B-cell lymphoma (data not shown). Patient numbers in other subgroups of rare entities of NHL, such as peripheral T-cell lymphoma, were too small to allow for statistical analysis of genotype and outcome.
However, in Burkitt's lymphoma and B-ALL (eligible patients, n = 211), analysis of outcome in relation to haplotypes for TNF –308 (GA) and LT- +252 (AG) revealed that patients with high-producer haplotypes had an increased risk for tumor progression (ie, relapse, progress, tumor-related death): the pEFS at 3 years was 81% (SE = 5%) in patients with high-producer haplotypes versus 92% (SE = 2%) in patients with low-producer haplotypes (P = .018; relative risk, 2.13; Fig 2). Table 3 shows events in patients with Burkitt's lymphoma and B-ALL and demonstrates that the inferior pEFS in patients with high-producer haplotypes was due to a higher proportion of patients suffering from tumor progression.
Among patients with high tumor load at diagnosis (serum LDH 500 U/L; n = 103 eligible patients), haplotype analysis allowed further differentiation of patients at risk for adverse events: prognosis in patients with high tumor load and high-producer haplotypes was significantly worse compared with that of patients with high tumor load and low-producer haplotypes (pEFS of 69% [SE = 8%] v pEFS of 85% [SE = 4%]; P = .05; Fig 3).
In multivariate analysis of prognostic factors for events, including the clinical characteristics of high tumor load (LDH 500 U/L, therapy branch R4/R4), CNS disease, MTX infusion time (4 hours), and TNF –308 (GA) and LT- +252 (AG) haplotype, haplotype was significantly associated with events (P = .048; Table 4). Patients with high-producer haplotypes had a 2.34-fold increase in risk of adverse events. To exclude any bias, the distribution of haplotypes regarding therapy branches and serum LDH was carefully examined. No association of high-producer haplotypes with therapy branch R3/R4 (P = .7) or serum LDH greater than 500 U/L (P = .9) could be found. The effect of haplotype on pEFS in multivariate analysis is therefore independent of high tumor load.
Analysis of toxicity data was performed by correlating the percentage of courses with maximum toxicity (grade 3 and 4) with TNF –308 and LT- +252 geno- and haplotypes. Toxicity analyses were performed for therapy branches R1/R2 with medium-dose MTX (MTX 1g/m2) and R3/R4 with high-dose MTX (MTX 5g/m2), depending on MTX infusion regimen (24-hour v 4-hour infusion time). Analysis of toxicity data did not show any association of genotype or haplotype with therapy-related toxicity in either therapy branch or MTX infusion regimen. To exclude indirect effects of increased therapy-related toxicity on outcome (such as prolongation of therapy owing to infections), intervals between therapy courses were analyzed. However, no differences in intervals between therapy courses could be found between patients with high-producer or low-producer haplotypes. Similarly, toxic deaths did not show any association with geno- or haplotype.
DISCUSSION
To our knowledge, the presented study comprises the largest cohort of uniformly treated pediatric patients with NHL to date that has been examined for polymorphisms in the TNF and LT- genes. Because the vast majority of all newly diagnosed cases of pediatric NHL in Germany are enrolled in multicenter trial NHL-BFM, the presented cohort of patients may be considered as being epidemiologically representative.41
The TNF and LT- genotype frequencies observed in our study were similar to those observed by other authors in healthy white study populations.11,12 The distribution of genotypes within diagnostic subgroups of pediatric NHL did not show any significant difference. Thus the TNF –308 (GA) and LT- +252 (AG) polymorphisms do not seem to play an important role in the pathogenesis of certain NHL entities of childhood and adolescence.
To date, only few studies were published that addressed the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with lymphoid malignancies. Findings from adult studies suggest an increased risk for multiple myeloma and B-cell chronic lymphocytic leukemia in adult patients carrying high-producer haplotypes for TNF –308 and LT- +252.11,25 In a study of adult T-lymphotropic virus type I carriers, genetic polymorphisms leading to enhanced TNF synthesis were associated with an increased risk of T-cell leukemia/lymphoma, suggesting a pathogenetic role of these polymorphisms in T-lymphotropic virus type I–associated NHL.42 However, these lymphoid malignancies are extremely rare in the pediatric population. Furthermore, pathogenetic mechanisms between common adult lymphoproliferative disease entities and those frequently found in pediatric patients might not be comparable.
Data on the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with pediatric lymphoid malignancies are rare. To our knowledge, only two studies have been published to date examining the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with clinical characteristics and risk factors for outcome in pediatric patients with ALL.27,43 In accordance with these studies, we could not detect any association of the investigated genotypes/haplotypes with outcome in pediatric lymphoblastic NHL, an entity of pediatric NHL that shares many clinical and therapeutic characteristics with childhood ALL. Thus these polymorphisms do not seem to influence outcome in pediatric ALL or lymphoblastic NHL. However, outcome in patients with lymphoblastic malignancies is excellent.32 Therefore, we may have failed to demonstrate such an influence because of the excellent prognosis of this entity and a lack of power to discern differences in outcome between groups of genotypes/haplotypes.
In contrast to the findings in patients with lymphoblastic lymphoma, we found a significant association between high-producer TNF –308/LT- +252 haplotypes and reduced outcome in patients with Burkitt's lymphoma and B-ALL. Particularly within the subset of patients with Burkitt's lymphoma/B-ALL and high tumor load at diagnosis (serum LDH 500 U/L), carriers of high-producer haplotypes had a significantly worse prognosis compared with patients with low-producer haplotypes. In multivariate analysis including other risk factors for events (ie, high tumor load [LDH 500 U/L], CNS disease, MTX infusion time of 4 hours, and TNF –308/LT- +252 haplotype), haplotype kept its prognostic significance (relative risk of an event for high-producer compared with low-producer haplotypes = 2.34; P = .048). This finding is of particular clinical interest because successful rescue therapy strategies for relapse or progressive disease in these patients are still not available, and early stratification of therapy intensity seems to be one of the keystones of successful therapy for patients with high tumor load at diagnosis.33
Only few data exist on the association of TNF –308 and LT- +252 haplotypes and prognosis in B-cell malignancies, with ambiguous results. Whereas some authors describe a worse prognosis for patients with high-producer haplotypes, others could not find such an association.23,25,27,44,45 Most of these studies comprise heterogeneously treated groups of adult patients with mature B-cell malignancies and, therefore, may not be comparable to pediatric patients suffering from an entirely different spectrum of B-cell lymphomas. To our knowledge, no data on Burkitt's lymphoma/B-ALL with respect to the TNF –308 (GA) and LT- +252 (AG) polymorphisms have been published yet. The mechanism by which these polymorphisms may influence outcome in B-cell NHL remains unclear. Results from studies examining the association of TNF and LT- gene polymorphisms with severe infections and predisposition to death from sepsis suggest that patients with high-producer haplotypes may exhibit more pronounced inflammatory reactions.7,9,15-17 Thus it could be speculated that in tumor patients, haplotype-associated differences in outcome may be attributable to indirect effects such as differences in therapy-related toxicity, infectious complications during therapy, and consecutive delays in therapy courses, leading to an increased risk of events.14,15,17 These aspects could be especially important for patients with Burkitt's lymphoma and B-ALL who receive a particular intense and toxic polychemotherapy regimen. Therefore, we closely looked for differences in therapy-associated toxicity in the present study. However, no association of high-producer haplotypes with increased occurrence of infections, orointestinal mucositis or other organ-specific toxicity could be observed. Total duration of therapy as well as median interval between therapy courses did not differ significantly between patients with low-producer or high-producer haplotypes, independent of therapy branch or MTX infusion regimen. Therefore, an indirect effect of the investigated genetic polymorphisms on outcome via increased therapy-related toxicity or delays in chemotherapy seems highly unlikely.
Data from experimental studies suggest a direct effect of TNF and LT- polymorphisms on lymphoid malignancies: TNF2 was shown to confer stronger transcriptional activity compared with TNF1. It was suggested that as a result of difference in the DNA/chromatin structure at the polymorphic site, the interaction of transcription factors may be enhanced, leading to the observed stronger transactivation.10 It has also been speculated that TNF2 and high-producer TNF –308/LT- +252 haplotypes may influence normal immune function and hamper endogenous immunologic tumor control.23,46,47 Furthermore, direct effects of TNF on B-cell growth, increasing the risk of uncontrolled expansion of aberrant B cells, have been suggested to play a role.22 Decreased cellular sensitivity to apoptosis-inducing chemotherapeutic agents in the presence of high levels of TNF and LT- —an observation made in in vitro studies—could also play a role, explaining the differences in therapy response.48 However, the location of the TNF and LT- genes within a highly variable region of multiple genes involved in the regulation and modulation of immune response makes it difficult to discern whether the TNF –308 (GA) and LT- +252 (AG) polymorphisms alone are responsible for the observed effects on outcome or whether they merely represent concomitant genetic variations as part of more complex changes and linkage disequilibrium.3,6,49
Even though the mechanisms by which TNF and LT- polymorphisms influence outcome of B-cell neoplasia in childhood and adolescence remain unclear, the data of the present study demonstrate their potential applicability in the clinical evaluation of prognosis in pediatric patients with Burkitt's lymphoma and B-ALL. Especially in patients with high tumor load, high risk for events, and increased risk of therapy-related toxicity, determination of TNF –308 (GA) and LT- +252 (AG) polymorphisms may be useful for the identification of those patients who may be at particularly high risk of treatment failure.
Appendix
Reference laboratories for histopathology and immunohistochemistry. R. Parwaresch, Lymph Node Registry founded by the Society of German Pathologists, Institute of Hematopathology, University of Kiel; A. Feller, Institute of Pathology, University of Lübeck; M.L. Hansmann, Institute of Pathology, University of Frankfurt; P. M?ller, Institute of Pathology, University of Ulm; H.K. Müller-Hermelink, Institute of Pathology, University of Würzburg; H. Stein, Institute of Pathology, University of Berlin, Germany; and I. Simonitsch, Institute of Pathology, University of Vienna, Austria.
Immunophenotyping. W.-D. Ludwig, Berlin, Germany; W. Knapp, Vienna, Austria; F. Niggli, Zürich, Switzerland.
Cytomorphology. A. Reiter, Giessen, Germany; W. Haas, Vienna, Austria; F. Niggli, Zürich, Switzerland.
Principal investigators. R. Mertens (Aachen), R. Angst (Aarau), A. Gnekow (Augsburg), R. Dickerhoff (St Augustin), P. Imbach (Basel), G.F. Wuendisch (Bayreuth), W. D?rffel (Berlin-Buch), G. Henze (Berlin-Charité), U. Bode (Bonn), W. Eberl (Braunschweig), H.-J. Spaar (Bremen), I. Krause (Chemnitz), J.-D. Thaben (Coburg), D. Moebius (Cottbus), W. Wiesel (Datteln), B. Ausserer (Dornbirn), H. Breu (Dortmund), G. Wei?bach (Dresden), W. Kotte (Dresden), U. G?bel (Düsseldorf), W. Weinmann (Erfurt), J.D. Beck (Erlangen), W. Havers (Essen), G. Müller (Feldkirch), B. Kornhuber (Frankfurt), C. Niemeyer (Freiburg), F. Lampert (Gie?en), M. Lakomek (G?ttingen), C. Urban (Graz), H. Reddemann (Greifswald), S. Burdach (Halle), G. Janka-Schaub (Hamburg), K. Welte (Hannover), B. Selle/A. Kulozik (Heidelberg), C. Tautz (Herdecke), N. Graf (Homburg), F.M. Fink (Innsbruck), F. Zintl (Jena), G. Ne?ler (Karlsruhe), H. Wehinger (Kassel), R. Schneppenheim (Kiel), H. Messner (Klagenfurt), M. Rister, (Koblenz), F. Berthold (K?ln), W. Sternschulte (K?ln), C. Schulte-Wissermann (Krefeld), M. Domula (Leipzig), I. Mutz (Leoben), K. Schmitt (Linz), O. Stoellinger (Linz), L. Nobile (Locarno), P. Bucsky (Lübeck), H. Ruetschele (Ludwigshafen), U. Caflisch (Luzern), U. Mittler (Magdeburg), P. Gutjahr (Mainz), O. Sauer (Mannheim), C. Eschenbach (Marburg), W. Tillmann (Minden), K.-D. Tympner (München-Harlaching), R. J. Haas (München), C. Bender-G?tze (München), S. Müller-Weihrich (München), H. Jürgens (Münster), O. Schofer (Neunkirchen), A. Jobke (Nürnberg), G. Eggers (Rostock), R. Geib-K?nig (Saarbrücken), H. Grienberger (Salzburg), H. Haas (Schwarzach), R. Schumacher (Schwerin), F.-J. G?bel (Siegen), R. Ploier (Steyr), A. Feldges (St. Gallen), J. Treuner (Stuttgart), H. Rau (Trier), D. Niethammer (Tübingen), K.-M. Debatin (Ulm), D. Franke (Vechta), H. Gadner (Wien), F. Haschke (Wien), J. Weber (Wiesbaden), D. Dohrn (Wuppertal), J. Kühl (Würzburg), and F. Niggli (Zürich).
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank E. Odenwald for expert work in cytologic diagnosis and U. Meyer for excellent data management. We especially thank all doctors and nurses in participating hospitals for their continuous care for sick children and their excellent cooperation with the NHL-BFM study center.
NOTES
Supported by the Deutsche Krebshilfe, Bonn, Germany, Grant No. M 109/91/Re1, and the Verein zur F?rderung der Behandlung krebskranker Kinder Hannover e.V.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Locksley RM, Killeen N, Lenardo MJ: The TNF and TNF receptor superfamilies: Integrating mammalian biology. Cell 104:487-501, 2001
Chan FK-M, Siegel RM, Lenardo MJ: Signaling by the TNF receptor superfamily and T cell homeostasis. Immunity 13:419-422, 2000
Ruuls SR, Sedgwick JD: Unlinking tumor necrosis factor biology from the major histocompatibility complex: Lessons from human genetics and animal models. Am J Hum Genet 65:294-301, 1999
Nedwin GE, Naylor SL, Sakaguchi AY, et al: Human lymphotoxin and tumor necrosis factor genes: Structure, homology and chromosome localization. Nucleic Acids Res 13:6361-6373, 1985
Spies T, Morton CC, Nedospasov SA, et al: Genes for the tumor necrosis factors alpha and beta are linked to the human major histocompatibility complex. Proc Natl Acad Sci U S A 83:8699-8702, 1986
McGuire W, Hill AVS, Allsopp CEM, et al: Variation in the TNF alpha promoter region associated with susceptibility to cerebral malaria. Nature 371:508-511, 1994
Herrmann S-M, Ricard S, Nicaud V, et al: Polymorphisms of the tumor necrosis factor alpha gene, coronary heart disease and obesity. Eur J Clin Invest 28:59-66, 1998
Knight JC, Udalova I, Hill AVS, et al: A polymorphism that effects OCT-1 binding to the TNF promoter region is associated with severe malaria. Nat Genet 22:145-150, 1999
Wilson AG, di Giovine FS, Blakemore AIF, et al: Single base polymorphism in the tumor necrosis factor alpha (TNF alpha) gene detectable by NcoI restriction of PCR product. Hum Mol Genet 1:353-357, 1992
Wilson AG, Symons JA, McDowell TL, et al: Effects of a polymorphism in the human tumor necrosis factor alpha promoter on transcriptional activation. Proc Natl Acad Sci U S A 94:3195-3199, 1997
Demeter J, Porzsolt F, Ramisch S, et al: Polymorphisms of the tumor-necrosis factor alpha and lymphotoxin-alpha genes in chronic lymphocytic leukemia. Br J Haematol 97:107-112, 1997
Wihlborg C, Sjoberg J, Intaglietta M, et al: Tumor-necrosis factor-alpha cytokine promoter gene polymorphism in Hodgkin's disease and chronic lymphocytic leukaemia. Br J Haematol 104:346-349, 1999
Wilson AG, di Giovine FS, Duff GW: Genetics of tumor necrosis factor-alpha in autoimmune, infectious, and neoplastic diseases. J Inflamm 45:1-12, 1995
Nadel S, Newport MJ, Booy R, et al: Variation in the tumor necrosis factor alpha gene promoter region may be associated with death from meningococcal disease. J Infect Dis 174:878-880, 1996
Mira J-P, Cariou A, Grall F, et al: Association of TNF2, a TNF-alpha promoter polymorphism, with septic shock susceptibility and mortality. JAMA 282:561-568, 1999
Cabrera M, Shaw M-A, Sharples C, et al: Polymorphisms in tumor necrosis factor genes associated with mucocutaneous Leishmaniosis. J Exp Med 182:1259-1264, 1995
Stüber F, Petersen M, Bokelmann F, et al: A genomic polymorphism within the tumor necrosis factor locus influences plasma tumor necrosis factor-alpha concentrations and outcome of patients with severe sepsis. Crit Care Med 24:381-384, 1996
Messer G, Spengler U, Jung MC, et al: Polymorphic structure of the tumor necrosis factor (TNF) locus: An NcoI polymorphism in the first intron of the human TNF-beta gene correlates with a variant amino acid in position 26 and a reduced level of TNF-beta production. J Exp Med 173:209-219, 1991
Koss K, Satsangi J, Fanning GC, et al: Cytokine (TNF-alpha, LT-alpha, and IL-10) polymorphisms in inflammatory bowel disease and normal controls: Differential effects on production and allele frequencies. Genes Immun 1:185-190, 2000
Mulcahy B, Waldron-Lynch F, McDermott MF, et al: Genetic variability in the tumor necrosis factor-lymphotoxin region influences susceptibility to rheumatoid arthritis. Am J Hum Genet 59:676-683, 1996
Mainou-Fowler T, Dickinson AM, Taylor PRA, et al: Tumor necrosis factor gene polymorphisms in lymphoproliferative disease. Leuk Lymphoma 38:547-552, 2000
Warzocha K, Salles G, Bienvenu J, et al: The tumor necrosis factor ligand-receptor system can predict treatment outcome in lymphoma patients. J Clin Oncol 15:499-508, 1997
Warzocha K, Ribeiro P, Bienvenu J, et al: Genetic polymorphisms in the tumor necrosis factor locus influence non-Hodgkin's lymphoma outcome. Blood 91:3574-3581, 1998
Warzocha K, Bienvenu J, Ribeiro P, et al: Plasma levels of tumor necrosis factor and its soluble receptors correlate with clinical features and outcome of Hodgkin's disease patients. Br J Cancer 77:2357-2362, 1998
Davies FE, Rollinson SJ, Rawstron AC, et al: High-producer haplotypes of tumor necrosis factor alpha and lymphotoxin alpha are associated with an increased risk of myeloma and have improved progression-free survival after treatment. J Clin Oncol 18:2843-2851, 2000
Zheng C, Huang D, Bergenbrant S, et al: Interleukin 6, tumor necrosis factor alpha, interleukin 1 beta and interleukin 1 receptor antagonist promoter or coding gene polymorphisms in multiple myeloma. Br J Haematol 109:39-45, 2000
Stanulla M, Schrauder A, Welte K, et al: Tumor necrosis factor and lymphotoxin-alpha genetic polymorphisms and risk of relapse in childhood B-cell precursor acute lymphoblastic leukemia: A case-control study of patients treated with BFM therapy. BMC Blood Disord 1:2, 2001
Harris NL: Principles of the revised European-American Lymphoma Classification (from the International Lymphoma Study Group). Ann Oncol 8:11-16, 1997
Stansfeld AG, Diebold J, Kapanci Y, et al: Updated Kiel classification for lymphomas. Lancet 1:292-293, 1988
Bennett JM, Catorsky D, Daniel MT, et al: Proposal for the classification of the acute leukemias: French-American-British (FAB) Cooperative Groups. Br J Haematol 33:451-458, 1976
Harris NL, Jaffe ES, Stein H, et al: A revised European-American Classification of lymphoid neoplasms: A proposal from the International Lymphoma Study Group. Blood 84:1361-1392, 1994
Reiter A, Schrappe M, Ludwig W-D, et al: Intensive ALL-type therapy without local radiotherapy provides a 90% event-free survival for children with T-cell lymphoblastic lymphoma: A BFM Group report. Blood 95:416-421, 2000
Woessmann W, Seidemann K, Mann G, et al: The impact of methotrexate administration schedule and dose in the therapy of children and adolescents with B-cell neoplasms: A report of the BFM study NHL-BFM 95. Blood 105:948-958, 2005
Stanulla M, Schrappe M, Brechlin AM, et al: Polymorphisms within glutathione S-tranferase genes (GSTM1, GSTT1, GSTP1) and risk of relapse in childhood B-cell precursor acute lymphoblastic leukemia: A case-control study. Blood 95:1222-1228, 2000
Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972
Greenwood M: A report on the natural duration of cancer. Reports on Public Health and Medical Subjects, 1926, pp 1-26
Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966
Reiter A, Schrappe M, Parwaresch R, et al: Non-Hodgkin's lymphomas of childhood and adolescence: Results of a treatment stratified for biologic subtypes and stage—A report of the Berlin-Frankfurt-Münster Group. J Clin Oncol 13:359-372, 1995
Patte C, Auperin A, Michon J, et al: The Societe Francaise d'Oncologie Pediatrique LMB89 protocol: Highly effective multiagent chemotherapy tailored to the tumor burden and initial response in 561 unselected children with B-cell lymphomas and L3 leukemia. Blood 97:3370-3379, 2001
Kaatsch P, Spix C, Michaelis J: 20 years German Childhood Cancer Registry: Annual Report 1999. Mainz, Germany, Druckbetrieb Lindner OHG, 2000
Tsukasaki K, Miller CW, Kubota T, et al: Tumor necrosis factor alpha polymorphism associated with increased susceptibility to development of adult T-cell leukemia/lymphoma in human T-lymphotropic virus type I carriers. Cancer Res 61:3770-3774, 2001
Takeuchi S, Takeuchi N, Tsukasaki K, et al: Genetic polymorphisms in the tumor necrosis factor locus in childhood lymphoblastic leukemia. Br J Haematol 119:985-987, 2002
Neben K, Mytilineos M, Moehler TM, et al: Polymorphisms of the tumor necrosis factor-alpha gene promoter predict for outcome after thalidomide therapy in relapsed and refractory multiple myeloma. Blood 100:2263-2265, 2002
Fitzgibbon J, Grenzelias D, Matthews J, et al: Tumour necrosis factor polymorphisms and susceptibility to follicular lymphoma. Br J Haematol 107:388-391, 1999
Ross ME, Caliguri MA: Cytokine-induced apoptosis of human natural killer cells identifies a novel mechanism to regulate the innate immune response. Blood 89:910-916, 1997
Zheng L, Fisher G, Miller RE, et al: Induction of apoptosis in mature T-cells by tumor necrosis factor. Nature 377, 1995
Kobayashi D, Watanabe N, Yamauchi N, et al: Endogenous tumor necrosis factor as a predictor of doxorubicin sensitivity in leukemic patients. Blood 89:2472-2475, 1997
Carroll MC, Katzman P, Alicot EM, et al: Linkage map of the human major histocompatibility complex including the tumor necrosis factor genes. Proc Natl Acad Sci U S A 84:8535-8539, 1987(Kathrin Seidemann, Martin)
NHL-BFM Study Center, Department of Pediatric Hematology and Oncology, University Children's Hospital, Justus-Liebig-University Giessen, Germany
ABSTRACT
PURPOSE: To analyze the association of genetic variation within the tumor necrosis factor (TNF –308 [GA]) and lymphotoxin alfa (LT-a +252 [AG]) genes with outcome in non-Hodgkin's lymphoma of childhood and adolescence.
PATIENTS AND METHODS: Genotyping of the TNF –308 (GA) and LT-a +252 (AG) polymorphisms in patients (n = 488) enrolled onto the German-Austrian-Swiss multicenter trial NHL-BFM 95 from April 1996 to January 2000 was performed by polymerase chain reaction with subsequent restriction fragment length polymorphism analysis on DNA from tumor-free specimen.
RESULTS: In patients with Burkitt's lymphoma (BL) and B-cell acute lymphoblastic leukemia (B-ALL; n = 219, 211 eligible patients), patients carrying at least two variant alleles (high-producer haplotypes) had an increased risk of events: probability of event-free survival (pEFS) at 3 years was 81% (SE = 5%), compared with 91% (SE = 2%) in low-producer haplotypes (P = .018). In BL/B-ALL with high tumor load (lactate dehydrogenase [LDH] 500 U/L; n = 104), pEFS was 69% (SE = 8%) in high-producer versus 85% (SE = 4%) in low-producer haplotypes (P = .05). In multivariate analysis including risk factors for events (LDH 500 U/L, CNS involvement, methotrexate infusion regimen), TNF –308/LT- +252 haplotype kept prognostic relevance: patients with high-producer haplotypes had a 2.34-fold increase in risk of events (P = .048). The TNF –308 (GA) and LT- +252 (AG) polymorphisms were not associated with pEFS in lymphoblastic lymphoma (n = 101), anaplastic large-cell lymphoma (n = 67), or diffuse large B-cell lymphoma (n = 65), nor with therapy-related toxicity.
CONCLUSION: The TNF –308 (GA) and LT-a +252 (AG) polymorphisms were negative prognostic factors in pediatric BL/B-ALL. Among patients with serum LDH 500 U/L, haplotype analysis further determined patients at risk for events.
INTRODUCTION
Tumor necrosis factor (TNF) and lymphotoxin alfa (LT-) are cytokines of the tumor necrosis factor family that function as prominent mediators of immune regulation and inflammation.1-3 Both lymphokines have similar biologic activities, show 35% identity and 50% homology in amino acid sequence, and bind to the same group of cellular TNF receptors.1,2,4 The genes coding for TNF and LT- are located tandemly on the chromosomal region 6p21.3-21.1 and are closely linked to the HLA-B locus within a highly polymorphic region of the major histocompatibility complex.1,3-6
Several single nucleotide polymorphisms of the TNF gene have been described.7,8 Exchange of guanine by adenine at position –308 of the TNF promoter region (TNF2 allele) is associated with higher serum levels of soluble TNF.8-10 Approximately 60% to 70% of the white population are homozygous for the wild-type TNF1 allele, 30% to 40% are heterozygous, and 1.5% to 3% are homozygous for the variant TNF2 allele.11,12 The TNF –308 (GA) polymorphism has been associated with susceptibility to cerebral malaria, lepromatous leprosy, mucocutaneous leishmaniosis, ulcerative colitis, Crohn's disease, fatal meningococcal disease, and septic shock.6,13-17 A polymorphism in the coding region at position +252 of the LT- gene (AG) leads to different alleles of LT-, here referred to as LT- (10.5 kb) for the wild-type allele, and LT- (5.5 kb) for the variant allele.18 The LT- (5.5 kb) allele was associated with higher levels of soluble LT- in patients homozygous for LT- (5.5 kb).18 Within the white population, 40% to 45% are homozygous for the wild-type allele LT- (10.5 kb), 40% to 45% are heterozygous, and approximately 15% are homozygous for the variant allele LT- (5.5 kb).11,12 A strong association between the LT- (5.5 kb) allele and adverse outcome in autoimmune diseases (eg, Crohn's disease) has been described.13,19,20
Biologic actions of both cytokines as well as their association with autoimmune and severe inflammatory disorders have drawn interest to their potential role in the pathogenesis and prognosis of hematologic malignancies.21-27 However, data on the role of TNF –308 (GA) and LT- +252 (AG) polymorphisms in lymphoid malignancies are discussed controversially. Whereas some authors described a higher prevalence of TNF2/LT- (5.5 kb) haplotypes in patients with multiple myeloma compared with controls, others could not confirm these results.21 Although a potential contribution of the TNF –308 (GA) and LT- +252 (AG) polymorphisms to carcinogenic processes remains unclear, it is of interest that so-called high-producer haplotypes (ie, carriers of at least two variant alleles of TNF2 and/or LT- [5.5 kb]) have been associated with a worse prognosis in adult patients with non-Hodgkin's lymphoma (NHL).23 This association was strongest in diffuse large B-cell lymphoma.
Apart from possible associations of TNF and LT- polymorphisms with pathogenesis or prognosis of NHL, their association with severe infections and inflammatory reactions may have an impact on treatment and therapy-related toxicity in patients with NHL, thus delaying therapy and influencing prognosis.16 However, hardly any data exist on the putative role of TNF –308 (GA) and LT- +252 (AG) polymorphisms in chemotherapy-associated toxicity.
The present study examines the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with diagnostic NHL entities and treatment outcome in 488 unselected pediatric patients 18 years of age and younger treated on the multicenter therapy trial NHL-BFM 95, conducted by the Berlin-Frankfurt-Münster (BFM) study group. In addition, particular attention was paid to examine the association of these polymorphisms with therapy-related toxicity to exclude genotype-associated differences in therapy-related toxicity as relevant causes for differences in outcome.
PATIENTS AND METHODS
Patients
In trial NHL-BFM 95, patients were registered from 84 clinics in Austria, Germany, and Switzerland after informed consent was given by the patient, parents or legal guardians according to the Declaration of Helsinki. Approval of the study was obtained from the ethical committee of the principal investigator (A.R.) as well as the participating investigators. From April 1996 to January 2000, 684 patients 18 years of age and younger with newly diagnosed NHL were enrolled. Diagnosis was based on histopathology, cytology, immunology, and immunohistochemistry. NHL subtypes originally diagnosed according to the updated Kiel classification for NHL were reclassified on the basis of the WHO classification of hematologic malignancies.28,29 B-cell acute lymphoblastic leukemia (B-ALL) was diagnosed if the bone marrow smears showed at least 25% of blasts with typical French-American-British L3 morphology and immunology.28,30,31 Tumor slides of all patients were reviewed by the central reference pathology and/or cytology.
Of 684 enrolled patients, 488 patients (71.3%) were genotyped successfully for TNF –308 (GA) and LT- +252 (AG) polymorphisms. Analyses regarding clinical characteristics, distribution of lymphoma entities, and therapy-related toxicity were performed on the total group of 488 genotyped patients. Analyses regarding outcome and prognostic factors were performed on eligible patients only. Thirty-one of 488 genotyped patients were excluded for the following reasons: NHL was second malignancy (n = 1), NHL occurred after solid organ transplantation (n = 8), patient suffered from severe congenital immunodeficiency (n = 8), patient suffered from human immunodeficiency virus infection (n = 1), patient received chemotherapy before diagnosis of NHL (n = 8), therapy according to protocol NHL-BFM 95 was discontinued for nonmedical reasons (n = 1), initial diagnosis of type of NHL was incorrect, patient was assigned to wrong therapy group (n = 2), and patient was treated according to previous therapy protocol (n = 2).
Therapy
Patients with lymphoblastic lymphoma were treated according to an ALL-type protocol with slight modifications, as previously described (Fig 1, only therapy branches standard risk and medium risk are shown).32 Patients with B-cell lymphoma and B-ALL were stratified in four therapy branches of different therapy intensity according to stage and tumor mass at diagnosis, as previously described (Fig 1).33 Patients with anaplastic large-cell lymphoma (ALCL) were stratified according to stage and other risk factors such as skin involvement and lymphohistiocytic subtype (Fig 1). Therapy for B-cell NHL and ALCL (therapy group II and III) consisted of two to six 5-day courses of polychemotherapy, as previously described.33 Radiotherapy was not part of the protocol except for patients with lymphoblastic lymphoma and overt CNS disease.
In trial NHL-BFM 95, duration of methotrexate (MTX) infusion was randomly assigned in therapy group II for B-NHL within each therapy branch: patients received MTX either as 4-hour infusion or as 24-hour infusion with identical leucovorin (racemic folinic acid) rescue in both randomization arms.33
Treatment success was determined by probability of event-free survival (pEFS). Events were defined as follows: Death from any cause, tumor progression, and second malignancy. Progression was defined as growth of an incompletely resolved tumor or as recurrence of tumor at any site proven by biopsy.
Genotype Analysis
Genomic DNA was isolated from all available tumor-free bone marrow or peripheral-blood smears from patients enrolled onto trial NHL-BFM 95 as described before.27,34 Genotyping the TNF –308 (GA) and LT- +252 (AG) polymorphisms was performed by conventional polymerase chain reaction (PCR; Expand High Fidelity PCR System; Roche Diagnostics, Mannheim, Germany) followed by restriction fragment length polymorphism analysis with NcoI (New England Biolabs, Frankfurt, Germany). Primers used for amplification of a 107-bp fragment from the TNF promoter region were as follows: forward 5'-AGGCAATAGGTTTTGAGGGCCAT-3'; reverse 5'-TCCTCCCTGCTCCGATTCCG-3'. By introducing a single base change within the forward primer, an NcoI restriction site was created and used for analysis of the TNF –308 (GA) promoter polymorphism.9 In presence of the TNF1 allele, the amplified 107-bp fragment is cut into two fragments of 20 and 87 bp; the PCR product of the TNF2 allele remains uncut.9
The LT- +252 (AG) polymorphism was analyzed by PCR amplification of a 368-bp fragment using the following primers: forward 5'-CTCCTGCACCTGCTGCCTGGATC-3'; reverse 5'-GAAGAGACGTTCAGGTGGTGTCAT-3'. After NcoI restriction digest, the PCR product amplified from the LT- (10.5 kb) allele remains uncut. In presence of the LT- (5.5 kb) allele, the 368-bp PCR product is cut into two fragments of 133 and 235 bp.23 A random sample of 10% of the patients was genotyped twice; no discordances were observed regarding genotyping results.
Statistical Analysis
Associations of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with clinical characteristics (age, sex, type of NHL, stage, serum lactate dehydrogenase [LDH] at diagnosis, outcome, clinical risk factors for events, therapy-related toxicity, and outcome) were analyzed using the 2 or Fisher's exact test. Prognostic relevance of different parameters was examined by stepwise Cox regression analysis.35,36 Analysis of pEFS was performed according to Kaplan and Meier, with differences compared by the log-rank test.37,38 The pEFS was calculated from the date of diagnosis to the first event or to the date of last follow-up. Relapse, death in continuous complete remission, and second malignancy were regarded as events; failure to achieve remission was regarded as an event on day 1.
Statistical analyses were performed using the SAS program (SAS-PC, version 6.12; SAS Institute, Cary, NC). Follow-up was actualized as of January 1, 2004.
RESULTS
Patients
Of a total 684 pediatric patients enrolled onto trial NHL-BFM 95, 488 patients (71.3%) were genotyped successfully. Of these 488 patients, 101 patients (20.7%) suffered from lymphoblastic lymphoma, 219 patients (44.9%) from Burkitt's lymphoma or B-ALL, 65 patients (13.3%) from diffuse large B-cell lymphoma, 67 patients (13.7%) from ALCL, seven patients (1.5%) from peripheral T-cell lymphoma, and 29 patients (5.9%) from other NHL entities. Clinical characteristics such as tumor stage, tumor load, age, sex, proportion of excluded patients, or randomization in different MTX infusion regimens did not differ between genotyped patients and the remaining study population (data not shown).
Genotype and Clinical Characteristics
Of 488 patients genotyped for the TNF –308 (GA) polymorphism, 356 (73.0%) were homozygous for TNF1, 124 (25.4%) were heterozygous, and eight patients (1.6%) were homozygous for TNF2. Genotyping results of the LT- +252 (AG) polymorphism revealed 216 patients (44.3%) as being homozygous for LT- (10.5 kb), 226 (46.3%) being heterozygous, and 46 patients (9.4%) homozygous for LT- (5.5 kb). The observed genotype distribution was in agreement with the Hardy-Weinberg equilibrium and similar to previously described frequencies in healthy white populations.11
As previously described, a significant association of TNF1 and LT- (10.5 kb) as well as of TNF2 and LT- (5.5 kb) alleles was observed (P = .0001; Table 1). 23,27 As a result of this finding, statistical analysis regarding the association of genotype with clinical characteristics and outcome was performed using haplotypes: patients with at least two variant alleles (TNF2 and/or LT- [5.5 kb]; n = 146) were referred to as high-producer haplotypes ([LT- (10.5/10.5); TNF2/TNF2]; [LT- (5.5/5.5); TNF2/TNF2]; [LT- (5.5/10.5); TNF1/TNF2]; [LT- (5.5/10.5); TNF2/TNF2]; [LT- (5.5/5.5); TNF1/TNF2]; [LT- (5.5/5.5); TNF1/TNF1]). Patients with less than two variant alleles (n = 342) were referred to as low-producer haplotype carriers ([LT- (10.5/10.5); TNF1/TNF1]; [LT- (10.5/5.5); TNF1/TNF1]; [LT- (10.5/10.5); TNF1/TNF2]).
No association could be found between genotype or haplotype for TNF –308 (GA) and LT- +252 (AG) and NHL entities (Table 2). Similarly, there was no association between genotypes and clinical characteristics or previously described risk factors for failure such as sex and stage.32-35,39,40
Genotype Analysis and Outcome
Analysis of genotype and outcome did not show any association of genotype and prognosis for patients with lymphoblastic lymphoma, ALCL, or diffuse large B-cell lymphoma (data not shown). Patient numbers in other subgroups of rare entities of NHL, such as peripheral T-cell lymphoma, were too small to allow for statistical analysis of genotype and outcome.
However, in Burkitt's lymphoma and B-ALL (eligible patients, n = 211), analysis of outcome in relation to haplotypes for TNF –308 (GA) and LT- +252 (AG) revealed that patients with high-producer haplotypes had an increased risk for tumor progression (ie, relapse, progress, tumor-related death): the pEFS at 3 years was 81% (SE = 5%) in patients with high-producer haplotypes versus 92% (SE = 2%) in patients with low-producer haplotypes (P = .018; relative risk, 2.13; Fig 2). Table 3 shows events in patients with Burkitt's lymphoma and B-ALL and demonstrates that the inferior pEFS in patients with high-producer haplotypes was due to a higher proportion of patients suffering from tumor progression.
Among patients with high tumor load at diagnosis (serum LDH 500 U/L; n = 103 eligible patients), haplotype analysis allowed further differentiation of patients at risk for adverse events: prognosis in patients with high tumor load and high-producer haplotypes was significantly worse compared with that of patients with high tumor load and low-producer haplotypes (pEFS of 69% [SE = 8%] v pEFS of 85% [SE = 4%]; P = .05; Fig 3).
In multivariate analysis of prognostic factors for events, including the clinical characteristics of high tumor load (LDH 500 U/L, therapy branch R4/R4), CNS disease, MTX infusion time (4 hours), and TNF –308 (GA) and LT- +252 (AG) haplotype, haplotype was significantly associated with events (P = .048; Table 4). Patients with high-producer haplotypes had a 2.34-fold increase in risk of adverse events. To exclude any bias, the distribution of haplotypes regarding therapy branches and serum LDH was carefully examined. No association of high-producer haplotypes with therapy branch R3/R4 (P = .7) or serum LDH greater than 500 U/L (P = .9) could be found. The effect of haplotype on pEFS in multivariate analysis is therefore independent of high tumor load.
Analysis of toxicity data was performed by correlating the percentage of courses with maximum toxicity (grade 3 and 4) with TNF –308 and LT- +252 geno- and haplotypes. Toxicity analyses were performed for therapy branches R1/R2 with medium-dose MTX (MTX 1g/m2) and R3/R4 with high-dose MTX (MTX 5g/m2), depending on MTX infusion regimen (24-hour v 4-hour infusion time). Analysis of toxicity data did not show any association of genotype or haplotype with therapy-related toxicity in either therapy branch or MTX infusion regimen. To exclude indirect effects of increased therapy-related toxicity on outcome (such as prolongation of therapy owing to infections), intervals between therapy courses were analyzed. However, no differences in intervals between therapy courses could be found between patients with high-producer or low-producer haplotypes. Similarly, toxic deaths did not show any association with geno- or haplotype.
DISCUSSION
To our knowledge, the presented study comprises the largest cohort of uniformly treated pediatric patients with NHL to date that has been examined for polymorphisms in the TNF and LT- genes. Because the vast majority of all newly diagnosed cases of pediatric NHL in Germany are enrolled in multicenter trial NHL-BFM, the presented cohort of patients may be considered as being epidemiologically representative.41
The TNF and LT- genotype frequencies observed in our study were similar to those observed by other authors in healthy white study populations.11,12 The distribution of genotypes within diagnostic subgroups of pediatric NHL did not show any significant difference. Thus the TNF –308 (GA) and LT- +252 (AG) polymorphisms do not seem to play an important role in the pathogenesis of certain NHL entities of childhood and adolescence.
To date, only few studies were published that addressed the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with lymphoid malignancies. Findings from adult studies suggest an increased risk for multiple myeloma and B-cell chronic lymphocytic leukemia in adult patients carrying high-producer haplotypes for TNF –308 and LT- +252.11,25 In a study of adult T-lymphotropic virus type I carriers, genetic polymorphisms leading to enhanced TNF synthesis were associated with an increased risk of T-cell leukemia/lymphoma, suggesting a pathogenetic role of these polymorphisms in T-lymphotropic virus type I–associated NHL.42 However, these lymphoid malignancies are extremely rare in the pediatric population. Furthermore, pathogenetic mechanisms between common adult lymphoproliferative disease entities and those frequently found in pediatric patients might not be comparable.
Data on the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with pediatric lymphoid malignancies are rare. To our knowledge, only two studies have been published to date examining the association of the TNF –308 (GA) and LT- +252 (AG) polymorphisms with clinical characteristics and risk factors for outcome in pediatric patients with ALL.27,43 In accordance with these studies, we could not detect any association of the investigated genotypes/haplotypes with outcome in pediatric lymphoblastic NHL, an entity of pediatric NHL that shares many clinical and therapeutic characteristics with childhood ALL. Thus these polymorphisms do not seem to influence outcome in pediatric ALL or lymphoblastic NHL. However, outcome in patients with lymphoblastic malignancies is excellent.32 Therefore, we may have failed to demonstrate such an influence because of the excellent prognosis of this entity and a lack of power to discern differences in outcome between groups of genotypes/haplotypes.
In contrast to the findings in patients with lymphoblastic lymphoma, we found a significant association between high-producer TNF –308/LT- +252 haplotypes and reduced outcome in patients with Burkitt's lymphoma and B-ALL. Particularly within the subset of patients with Burkitt's lymphoma/B-ALL and high tumor load at diagnosis (serum LDH 500 U/L), carriers of high-producer haplotypes had a significantly worse prognosis compared with patients with low-producer haplotypes. In multivariate analysis including other risk factors for events (ie, high tumor load [LDH 500 U/L], CNS disease, MTX infusion time of 4 hours, and TNF –308/LT- +252 haplotype), haplotype kept its prognostic significance (relative risk of an event for high-producer compared with low-producer haplotypes = 2.34; P = .048). This finding is of particular clinical interest because successful rescue therapy strategies for relapse or progressive disease in these patients are still not available, and early stratification of therapy intensity seems to be one of the keystones of successful therapy for patients with high tumor load at diagnosis.33
Only few data exist on the association of TNF –308 and LT- +252 haplotypes and prognosis in B-cell malignancies, with ambiguous results. Whereas some authors describe a worse prognosis for patients with high-producer haplotypes, others could not find such an association.23,25,27,44,45 Most of these studies comprise heterogeneously treated groups of adult patients with mature B-cell malignancies and, therefore, may not be comparable to pediatric patients suffering from an entirely different spectrum of B-cell lymphomas. To our knowledge, no data on Burkitt's lymphoma/B-ALL with respect to the TNF –308 (GA) and LT- +252 (AG) polymorphisms have been published yet. The mechanism by which these polymorphisms may influence outcome in B-cell NHL remains unclear. Results from studies examining the association of TNF and LT- gene polymorphisms with severe infections and predisposition to death from sepsis suggest that patients with high-producer haplotypes may exhibit more pronounced inflammatory reactions.7,9,15-17 Thus it could be speculated that in tumor patients, haplotype-associated differences in outcome may be attributable to indirect effects such as differences in therapy-related toxicity, infectious complications during therapy, and consecutive delays in therapy courses, leading to an increased risk of events.14,15,17 These aspects could be especially important for patients with Burkitt's lymphoma and B-ALL who receive a particular intense and toxic polychemotherapy regimen. Therefore, we closely looked for differences in therapy-associated toxicity in the present study. However, no association of high-producer haplotypes with increased occurrence of infections, orointestinal mucositis or other organ-specific toxicity could be observed. Total duration of therapy as well as median interval between therapy courses did not differ significantly between patients with low-producer or high-producer haplotypes, independent of therapy branch or MTX infusion regimen. Therefore, an indirect effect of the investigated genetic polymorphisms on outcome via increased therapy-related toxicity or delays in chemotherapy seems highly unlikely.
Data from experimental studies suggest a direct effect of TNF and LT- polymorphisms on lymphoid malignancies: TNF2 was shown to confer stronger transcriptional activity compared with TNF1. It was suggested that as a result of difference in the DNA/chromatin structure at the polymorphic site, the interaction of transcription factors may be enhanced, leading to the observed stronger transactivation.10 It has also been speculated that TNF2 and high-producer TNF –308/LT- +252 haplotypes may influence normal immune function and hamper endogenous immunologic tumor control.23,46,47 Furthermore, direct effects of TNF on B-cell growth, increasing the risk of uncontrolled expansion of aberrant B cells, have been suggested to play a role.22 Decreased cellular sensitivity to apoptosis-inducing chemotherapeutic agents in the presence of high levels of TNF and LT- —an observation made in in vitro studies—could also play a role, explaining the differences in therapy response.48 However, the location of the TNF and LT- genes within a highly variable region of multiple genes involved in the regulation and modulation of immune response makes it difficult to discern whether the TNF –308 (GA) and LT- +252 (AG) polymorphisms alone are responsible for the observed effects on outcome or whether they merely represent concomitant genetic variations as part of more complex changes and linkage disequilibrium.3,6,49
Even though the mechanisms by which TNF and LT- polymorphisms influence outcome of B-cell neoplasia in childhood and adolescence remain unclear, the data of the present study demonstrate their potential applicability in the clinical evaluation of prognosis in pediatric patients with Burkitt's lymphoma and B-ALL. Especially in patients with high tumor load, high risk for events, and increased risk of therapy-related toxicity, determination of TNF –308 (GA) and LT- +252 (AG) polymorphisms may be useful for the identification of those patients who may be at particularly high risk of treatment failure.
Appendix
Reference laboratories for histopathology and immunohistochemistry. R. Parwaresch, Lymph Node Registry founded by the Society of German Pathologists, Institute of Hematopathology, University of Kiel; A. Feller, Institute of Pathology, University of Lübeck; M.L. Hansmann, Institute of Pathology, University of Frankfurt; P. M?ller, Institute of Pathology, University of Ulm; H.K. Müller-Hermelink, Institute of Pathology, University of Würzburg; H. Stein, Institute of Pathology, University of Berlin, Germany; and I. Simonitsch, Institute of Pathology, University of Vienna, Austria.
Immunophenotyping. W.-D. Ludwig, Berlin, Germany; W. Knapp, Vienna, Austria; F. Niggli, Zürich, Switzerland.
Cytomorphology. A. Reiter, Giessen, Germany; W. Haas, Vienna, Austria; F. Niggli, Zürich, Switzerland.
Principal investigators. R. Mertens (Aachen), R. Angst (Aarau), A. Gnekow (Augsburg), R. Dickerhoff (St Augustin), P. Imbach (Basel), G.F. Wuendisch (Bayreuth), W. D?rffel (Berlin-Buch), G. Henze (Berlin-Charité), U. Bode (Bonn), W. Eberl (Braunschweig), H.-J. Spaar (Bremen), I. Krause (Chemnitz), J.-D. Thaben (Coburg), D. Moebius (Cottbus), W. Wiesel (Datteln), B. Ausserer (Dornbirn), H. Breu (Dortmund), G. Wei?bach (Dresden), W. Kotte (Dresden), U. G?bel (Düsseldorf), W. Weinmann (Erfurt), J.D. Beck (Erlangen), W. Havers (Essen), G. Müller (Feldkirch), B. Kornhuber (Frankfurt), C. Niemeyer (Freiburg), F. Lampert (Gie?en), M. Lakomek (G?ttingen), C. Urban (Graz), H. Reddemann (Greifswald), S. Burdach (Halle), G. Janka-Schaub (Hamburg), K. Welte (Hannover), B. Selle/A. Kulozik (Heidelberg), C. Tautz (Herdecke), N. Graf (Homburg), F.M. Fink (Innsbruck), F. Zintl (Jena), G. Ne?ler (Karlsruhe), H. Wehinger (Kassel), R. Schneppenheim (Kiel), H. Messner (Klagenfurt), M. Rister, (Koblenz), F. Berthold (K?ln), W. Sternschulte (K?ln), C. Schulte-Wissermann (Krefeld), M. Domula (Leipzig), I. Mutz (Leoben), K. Schmitt (Linz), O. Stoellinger (Linz), L. Nobile (Locarno), P. Bucsky (Lübeck), H. Ruetschele (Ludwigshafen), U. Caflisch (Luzern), U. Mittler (Magdeburg), P. Gutjahr (Mainz), O. Sauer (Mannheim), C. Eschenbach (Marburg), W. Tillmann (Minden), K.-D. Tympner (München-Harlaching), R. J. Haas (München), C. Bender-G?tze (München), S. Müller-Weihrich (München), H. Jürgens (Münster), O. Schofer (Neunkirchen), A. Jobke (Nürnberg), G. Eggers (Rostock), R. Geib-K?nig (Saarbrücken), H. Grienberger (Salzburg), H. Haas (Schwarzach), R. Schumacher (Schwerin), F.-J. G?bel (Siegen), R. Ploier (Steyr), A. Feldges (St. Gallen), J. Treuner (Stuttgart), H. Rau (Trier), D. Niethammer (Tübingen), K.-M. Debatin (Ulm), D. Franke (Vechta), H. Gadner (Wien), F. Haschke (Wien), J. Weber (Wiesbaden), D. Dohrn (Wuppertal), J. Kühl (Würzburg), and F. Niggli (Zürich).
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank E. Odenwald for expert work in cytologic diagnosis and U. Meyer for excellent data management. We especially thank all doctors and nurses in participating hospitals for their continuous care for sick children and their excellent cooperation with the NHL-BFM study center.
NOTES
Supported by the Deutsche Krebshilfe, Bonn, Germany, Grant No. M 109/91/Re1, and the Verein zur F?rderung der Behandlung krebskranker Kinder Hannover e.V.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Locksley RM, Killeen N, Lenardo MJ: The TNF and TNF receptor superfamilies: Integrating mammalian biology. Cell 104:487-501, 2001
Chan FK-M, Siegel RM, Lenardo MJ: Signaling by the TNF receptor superfamily and T cell homeostasis. Immunity 13:419-422, 2000
Ruuls SR, Sedgwick JD: Unlinking tumor necrosis factor biology from the major histocompatibility complex: Lessons from human genetics and animal models. Am J Hum Genet 65:294-301, 1999
Nedwin GE, Naylor SL, Sakaguchi AY, et al: Human lymphotoxin and tumor necrosis factor genes: Structure, homology and chromosome localization. Nucleic Acids Res 13:6361-6373, 1985
Spies T, Morton CC, Nedospasov SA, et al: Genes for the tumor necrosis factors alpha and beta are linked to the human major histocompatibility complex. Proc Natl Acad Sci U S A 83:8699-8702, 1986
McGuire W, Hill AVS, Allsopp CEM, et al: Variation in the TNF alpha promoter region associated with susceptibility to cerebral malaria. Nature 371:508-511, 1994
Herrmann S-M, Ricard S, Nicaud V, et al: Polymorphisms of the tumor necrosis factor alpha gene, coronary heart disease and obesity. Eur J Clin Invest 28:59-66, 1998
Knight JC, Udalova I, Hill AVS, et al: A polymorphism that effects OCT-1 binding to the TNF promoter region is associated with severe malaria. Nat Genet 22:145-150, 1999
Wilson AG, di Giovine FS, Blakemore AIF, et al: Single base polymorphism in the tumor necrosis factor alpha (TNF alpha) gene detectable by NcoI restriction of PCR product. Hum Mol Genet 1:353-357, 1992
Wilson AG, Symons JA, McDowell TL, et al: Effects of a polymorphism in the human tumor necrosis factor alpha promoter on transcriptional activation. Proc Natl Acad Sci U S A 94:3195-3199, 1997
Demeter J, Porzsolt F, Ramisch S, et al: Polymorphisms of the tumor-necrosis factor alpha and lymphotoxin-alpha genes in chronic lymphocytic leukemia. Br J Haematol 97:107-112, 1997
Wihlborg C, Sjoberg J, Intaglietta M, et al: Tumor-necrosis factor-alpha cytokine promoter gene polymorphism in Hodgkin's disease and chronic lymphocytic leukaemia. Br J Haematol 104:346-349, 1999
Wilson AG, di Giovine FS, Duff GW: Genetics of tumor necrosis factor-alpha in autoimmune, infectious, and neoplastic diseases. J Inflamm 45:1-12, 1995
Nadel S, Newport MJ, Booy R, et al: Variation in the tumor necrosis factor alpha gene promoter region may be associated with death from meningococcal disease. J Infect Dis 174:878-880, 1996
Mira J-P, Cariou A, Grall F, et al: Association of TNF2, a TNF-alpha promoter polymorphism, with septic shock susceptibility and mortality. JAMA 282:561-568, 1999
Cabrera M, Shaw M-A, Sharples C, et al: Polymorphisms in tumor necrosis factor genes associated with mucocutaneous Leishmaniosis. J Exp Med 182:1259-1264, 1995
Stüber F, Petersen M, Bokelmann F, et al: A genomic polymorphism within the tumor necrosis factor locus influences plasma tumor necrosis factor-alpha concentrations and outcome of patients with severe sepsis. Crit Care Med 24:381-384, 1996
Messer G, Spengler U, Jung MC, et al: Polymorphic structure of the tumor necrosis factor (TNF) locus: An NcoI polymorphism in the first intron of the human TNF-beta gene correlates with a variant amino acid in position 26 and a reduced level of TNF-beta production. J Exp Med 173:209-219, 1991
Koss K, Satsangi J, Fanning GC, et al: Cytokine (TNF-alpha, LT-alpha, and IL-10) polymorphisms in inflammatory bowel disease and normal controls: Differential effects on production and allele frequencies. Genes Immun 1:185-190, 2000
Mulcahy B, Waldron-Lynch F, McDermott MF, et al: Genetic variability in the tumor necrosis factor-lymphotoxin region influences susceptibility to rheumatoid arthritis. Am J Hum Genet 59:676-683, 1996
Mainou-Fowler T, Dickinson AM, Taylor PRA, et al: Tumor necrosis factor gene polymorphisms in lymphoproliferative disease. Leuk Lymphoma 38:547-552, 2000
Warzocha K, Salles G, Bienvenu J, et al: The tumor necrosis factor ligand-receptor system can predict treatment outcome in lymphoma patients. J Clin Oncol 15:499-508, 1997
Warzocha K, Ribeiro P, Bienvenu J, et al: Genetic polymorphisms in the tumor necrosis factor locus influence non-Hodgkin's lymphoma outcome. Blood 91:3574-3581, 1998
Warzocha K, Bienvenu J, Ribeiro P, et al: Plasma levels of tumor necrosis factor and its soluble receptors correlate with clinical features and outcome of Hodgkin's disease patients. Br J Cancer 77:2357-2362, 1998
Davies FE, Rollinson SJ, Rawstron AC, et al: High-producer haplotypes of tumor necrosis factor alpha and lymphotoxin alpha are associated with an increased risk of myeloma and have improved progression-free survival after treatment. J Clin Oncol 18:2843-2851, 2000
Zheng C, Huang D, Bergenbrant S, et al: Interleukin 6, tumor necrosis factor alpha, interleukin 1 beta and interleukin 1 receptor antagonist promoter or coding gene polymorphisms in multiple myeloma. Br J Haematol 109:39-45, 2000
Stanulla M, Schrauder A, Welte K, et al: Tumor necrosis factor and lymphotoxin-alpha genetic polymorphisms and risk of relapse in childhood B-cell precursor acute lymphoblastic leukemia: A case-control study of patients treated with BFM therapy. BMC Blood Disord 1:2, 2001
Harris NL: Principles of the revised European-American Lymphoma Classification (from the International Lymphoma Study Group). Ann Oncol 8:11-16, 1997
Stansfeld AG, Diebold J, Kapanci Y, et al: Updated Kiel classification for lymphomas. Lancet 1:292-293, 1988
Bennett JM, Catorsky D, Daniel MT, et al: Proposal for the classification of the acute leukemias: French-American-British (FAB) Cooperative Groups. Br J Haematol 33:451-458, 1976
Harris NL, Jaffe ES, Stein H, et al: A revised European-American Classification of lymphoid neoplasms: A proposal from the International Lymphoma Study Group. Blood 84:1361-1392, 1994
Reiter A, Schrappe M, Ludwig W-D, et al: Intensive ALL-type therapy without local radiotherapy provides a 90% event-free survival for children with T-cell lymphoblastic lymphoma: A BFM Group report. Blood 95:416-421, 2000
Woessmann W, Seidemann K, Mann G, et al: The impact of methotrexate administration schedule and dose in the therapy of children and adolescents with B-cell neoplasms: A report of the BFM study NHL-BFM 95. Blood 105:948-958, 2005
Stanulla M, Schrappe M, Brechlin AM, et al: Polymorphisms within glutathione S-tranferase genes (GSTM1, GSTT1, GSTP1) and risk of relapse in childhood B-cell precursor acute lymphoblastic leukemia: A case-control study. Blood 95:1222-1228, 2000
Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972
Greenwood M: A report on the natural duration of cancer. Reports on Public Health and Medical Subjects, 1926, pp 1-26
Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966
Reiter A, Schrappe M, Parwaresch R, et al: Non-Hodgkin's lymphomas of childhood and adolescence: Results of a treatment stratified for biologic subtypes and stage—A report of the Berlin-Frankfurt-Münster Group. J Clin Oncol 13:359-372, 1995
Patte C, Auperin A, Michon J, et al: The Societe Francaise d'Oncologie Pediatrique LMB89 protocol: Highly effective multiagent chemotherapy tailored to the tumor burden and initial response in 561 unselected children with B-cell lymphomas and L3 leukemia. Blood 97:3370-3379, 2001
Kaatsch P, Spix C, Michaelis J: 20 years German Childhood Cancer Registry: Annual Report 1999. Mainz, Germany, Druckbetrieb Lindner OHG, 2000
Tsukasaki K, Miller CW, Kubota T, et al: Tumor necrosis factor alpha polymorphism associated with increased susceptibility to development of adult T-cell leukemia/lymphoma in human T-lymphotropic virus type I carriers. Cancer Res 61:3770-3774, 2001
Takeuchi S, Takeuchi N, Tsukasaki K, et al: Genetic polymorphisms in the tumor necrosis factor locus in childhood lymphoblastic leukemia. Br J Haematol 119:985-987, 2002
Neben K, Mytilineos M, Moehler TM, et al: Polymorphisms of the tumor necrosis factor-alpha gene promoter predict for outcome after thalidomide therapy in relapsed and refractory multiple myeloma. Blood 100:2263-2265, 2002
Fitzgibbon J, Grenzelias D, Matthews J, et al: Tumour necrosis factor polymorphisms and susceptibility to follicular lymphoma. Br J Haematol 107:388-391, 1999
Ross ME, Caliguri MA: Cytokine-induced apoptosis of human natural killer cells identifies a novel mechanism to regulate the innate immune response. Blood 89:910-916, 1997
Zheng L, Fisher G, Miller RE, et al: Induction of apoptosis in mature T-cells by tumor necrosis factor. Nature 377, 1995
Kobayashi D, Watanabe N, Yamauchi N, et al: Endogenous tumor necrosis factor as a predictor of doxorubicin sensitivity in leukemic patients. Blood 89:2472-2475, 1997
Carroll MC, Katzman P, Alicot EM, et al: Linkage map of the human major histocompatibility complex including the tumor necrosis factor genes. Proc Natl Acad Sci U S A 84:8535-8539, 1987(Kathrin Seidemann, Martin)