Diffuse Large B-Cell Lymphoma: Clinical and Biological Characterization and Outcome According to the Nodal or Extranodal Primary Origin
http://www.100md.com
《临床肿瘤学》
the Departments of Hematology and Pathology, Hospital Clínic, Institut de Recerca Biomèdica August Pi i Sunyer, Barcelona, Spain
ABSTRACT
PURPOSE: To study the main clinicobiologic features, response, and outcome of patients with diffuse large B-cell lymphoma (DLBCL) according to the primary site, lymph node, or different extranodal organs of the disease.
PATIENTS AND METHODS: We included 382 patients consecutively diagnosed with DLBCL in a single institution during a 13-year period. Morphology, immunophenotyping, proliferation index, differentiation profile, bcl-2/JH rearrangement, and clinical characteristics were analyzed according to the primary site of the lymphoma.
RESULTS: Sites of the disease were: lymph node, 222 cases (58%); Waldeyer's ring (WR), 42 (11%); and extranodal sites, 118 (31%), including GI tract in 45 cases. Primary extranodal cases, particularly GI, showed a bcl-6 expression more frequently than nodal cases. Patients with primary WR or GI lymphomas presented with early-stage disease, no marrow infiltration, normal serum lactate dehydrogenase, and low- to low/intermediate-risk international prognostic index (IPI) more frequently than the remainder. Complete response (CR) rate was 63%, with WR and GI lymphomas having a higher CR rate (85% and 80%, respectively) than the other groups. In the whole series, 5-year overall survival (OS) was 52%. Patients with WR or GI lymphomas showed better OS (5-year OS: 77% and 68%, respectively) than patients with nodal or other extranodal sites. In the multivariate analysis, IPI, bulky disease, and 2-microglobulin were the main variables to predict OS; no nodal or extranodal site maintained their prognostic value.
CONCLUSION: In the present series, the primary site of disease was associated with particular clinicopathologic features and outcome, though the latter largely depended on other factors.
INTRODUCTION
Approximately one third of the non-Hodgkin's lymphomas arise in tissues different from the lymph node, and for this reason, they are usually termed extranodal lymphomas.1-4 It has been shown recently that during the last two decades the incidence of lymphomas has increased, and that extranodal lymphomas increased more rapidly than nodal.3,5 The study of extranodal lymphomas as a group, regarding etiopathogenesis, biologic features, clinical characteristics, and outcome, has been the subject of several publications.3,4,6-12 However, most of them include heterogeneous series of patients in which several histologic subgroups are usually merged. The latter makes it difficult to reach conclusions, since it is not clear whether the clinicobiologic differences between nodal and extranodal are due to the site of the disease, or merely show differences in the particular spectrum of histologies of each site. On the other hand, most studies refer only to clinical aspects or are based on a relatively small number of patients. Therefore, clinicobiologic studies in histologically homogeneous series of patients are warranted.
Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin's lymphoma in Western countries, representing approximately 30% of the total.13 Between one third and one quarter of DLBCLs have a primary extranodal origin. Although considered a single entity in the Revised European-American Lymphoma (REAL)/WHO classification, DLBCL most likely includes different clinicopathologic entities that are currently difficult to separate using standard procedures.14-16 In fact, morphology, immunophenotyping, and genetics are heterogeneous among DLBCLs, and, more recently, microarray technology has been able to distinguish two groups of DLBCL according to the "germinal-center" or "postgerminal-center" (also called "activated") genetic signatures.17-19 Particular clinical and biologic characteristics have been suggested for DLBCLs arising in the lymph node, versus those in extranodal sites, including alterations in some genes such as bcl-2,20 bcl-6,20,21 c-myc,20 rel,22,23 and fas.24,25 Based on these data, a separate genetic origin for nodal and extranodal lymphomas has been claimed, and it has been suggested that these forms should be regarded as separate nosological entities. However, there is no consensus on this aspect, and in the WHO classification, it is suggested that the information regarding the nodal or extranodal primary site should be collected for investigational purposes only.15,26
The aim of the present study was to analyze the main clinicobiologic features at diagnosis, as well as the response to therapy and the outcome of a large series of patients diagnosed with DLBCL and homogeneously treated and followed-up in a single institution, according to the primary site (either lymph node or different extranodal territories) of the disease.
PATIENTS AND METHODS
Patients
Three hundred eighty-two patients consecutively diagnosed with a DLBCL between September 1987 and September 2000, and followed up in a single institution were selected for the present study. The cases with recognized disease phase of a follicular lymphoma or another type of indolent lymphoma with subsequent transformation into a DLBCL were not included. In addition, immunodeficiency-associated tumors, primary mediastinal, intravascular, and primary effusion lymphomas, were also excluded from the study.
Median age of the patients was 60 years (range, 14 to 94 years), and the male-female distribution was 188:194. The main characteristics of the patients at diagnosis are listed in Table 1. Advanced stage (Ann Arbor III or IV) was observed in 181 patients (47%), and any extranodal involvement, in 270 patients (71%), including bone marrow infiltration in 71 cases (19%). One hundred ninety-six (55%) of 357 patients with available data presented with high serum lactate dehydrogenase (LDH) levels, whereas the distribution according to the International Prognostic Index (IPI)27 was the following: low-risk, 135 patients (37%); low/intermediate, 71 patients (20%); high/intermediate, 73 patients (20%); high-risk, 84 patients (23%); and nonassessable, 19 cases. The main initial and evolutive variables, including the histologic parameters indicated below, were recorded and analyzed for prognosis.
Staging maneuvers included patient history and physical examination (including Waldeyer's ring area); CBCs and serum biochemistry, including LDH and 2-microglobulin (2m) levels; computed tomography (CT) scan of chest, abdomen, and pelvis; as well as bone marrow biopsy. The distinction between nodal and extranodal was not taken into account in planning the treatment. Three hundred sixty-nine patients were treated with combination chemotherapy: adriamycin-containing regimens in 344 patients (90%; in most cases cyclophosphamide, doxorubicin, vincristine, and prednisone) and combination chemotherapy without adriamycin in the remaining 25 patients. Post-therapy restaging consisted of the repetition of the previously abnormal tests and/or biopsies. Response was assessed according to conventional criteria. Overall, 240 patients achieved a complete response (CR; 63%); 44 patients, a partial response (12%); and 94 patients failed to respond to treatment. In four cases, the response was not assessable. After a median follow-up of 6.5 years (range, 0.2 to 15.3 years) for surviving patients, 189 patients had died. The 5-year and 10-year overall survival rates were 52% (95% CI, 47% to 57%) and 46% (95%CI, 40% to 52%), respectively.
Nodal/Extranodal Definitions
Lymphomas presenting in extranodal organs with no or only minor lymph node involvement were considered primary extranodal. Lymphomas with lymph node involvement clinically dominant, as well as those presenting at the spleen, were considered as primary nodal. Lymphomas arising in the Waldeyer's ring, though considered nodal, were analyzed separately. Bone marrow was considered an extranodal site. Finally, those lymphomas with extensive disease involving both nodal and extranodal sites were considered nodal.
Histologic Features
The diagnosis of DLBCL was based in all cases on the criteria established in the WHO classification.15 Morphologic and immunophenotypic studies were performed in 203 patients in whom adequate histologic material was available. For the morphologic analysis, histologic slides were reviewed in all the cases by two different observers (L.C. and E.C.). Morphologic subclassification was performed as previously described.28 The panel of monoclonal antibodies included antibodies against the following antigens: CD20, CD79a, CD3, CD5, CD10, MUM1, CD138, bcl-2, bcl-6, Ki-67, p53, and p27. The antibodies and the immunohistochemical conditions of use have been previously described.28 Tumor proliferation was analyzed by two different methods: (1) analysis of mitotic index, considering a cutoff of 25 mitoses per 10 high-power fields, and (2) proliferative index assessed by Ki-67 immunostaining.
The patients were clustered into four groups according to their immunophenotypic profile in order to assess the pattern of differentiation28: germinal center origin with CD10 positivity (GC-CD10+): CD10+/bcl-6+/MUM1–/CD138–; germinal center origin–CD10 negative (GC-CD10–): CD10-/bcl-6+/MUM1–/CD138–; postgerminal center origin (pGC): CD10–/bcl-6±/MUM1+/CD138–; and plasmablastic: CD10–/bcl-6–/MUM1+/CD138+.
Detection of the bcl-2/JH rearrangement involving the bcl-2 major breakpoint region (MBR) and minor cluster region (mcr) was performed in 73 cases by a polymerase chain reaction technique.28
Statistical Analysis
Categorical data were compared using Fisher's exact test for a two-sided P value, whereas for ordinal data, nonparametric tests were used. Bonferroni correction for multiple comparisons was applied when necessary. The multivariate analysis of the variables predicting response was performed by using a logistic regression. The actuarial survival analysis was performed according to the method described by Kaplan and Meier,29 and the curves were compared by the log-rank test.30 The multivariate analysis for survival was performed by using the stepwise proportional hazards model (Cox).31
RESULTS
Distribution and Clinical Features
One hundred eighteen patients (31%) presented at a primary extranodal site, and 42 patients (11%), at Waldeyer's ring, while the remaining 222 patients (58%) had a primary nodal DLBCL, including one at the spleen. The distribution according to the extranodal site was as follows: GI tract (45 cases, 12% of the overall series; including gastric lymphoma in 37 cases); liver (seven cases; 2%); soft tissue (17 cases; 4.5%); CNS (10 cases; 2.5%); bone (nine cases; 2%); skin (five cases; 1%); breast (five cases); kidney (five cases); testis/ovary (five cases); thyroid (three cases); bone marrow (two cases); and lung, prostate, uterus, and pericardium (one case each). Finally, one patient showed a double extranodal lymphoma (bilateral breast and ovary).
The main clinical characteristics at diagnosis according to the primary site of the lymphoma are detailed in Table 1. Patients with Waldeyer's ring DLBCL presented more frequently with early-stage disease; ambulatory performance status; absence of "B" symptoms; bulky disease and bone marrow infiltration; normal serum albumin, LDH, and 2m levels; and low- or low/intermediate-risk IPI (P < .01 in all cases v the other groups). GI DLBCL was also early-stage, with absence of bone marrow infiltration, normal serum LDH, and had low- or low/intermediate-risk IPI more so than the other groups (P < .03 in all cases). Finally, the whole group of patients with nodal DLBCL, including those with Waldeyers' ring DLBCL, showed bulky disease (36% v 23%; P = .02), bone marrow infiltration (23% v 10%; P = .004), and high serum LDH levels (58% v 47%; P = .05) more frequently than the extranodal lymphomas.
Morphology and Immunophenotypic Profile
The 172 DLBCL patients in whom adequate material was available to fully assess morphology were classified as: centroblastic lymphoma, 57 cases (33%); centroblastic polymorphic with less than 50% immunoblasts, 23 cases (13%); centroblastic polymorphic with 50% immunoblasts, 16 cases (9%); immunoblastic, 15 cases (9%); other, five cases (3%; including two patients with plasmablastic, two with a pleomorphic, and one patient with a T-cell–rich DLBCL). The remaining 56 cases were not otherwise classifiable. The proportion of DLBCL with immunoblastic features ( 50% immunoblastic cells) was 33% (24 of 72) for primary nodal lymphomas, 38% (five of 13) for Waldeyer's ring lymphomas, and 23% (seven of 31) for extranodal lymphomas (including 13% [two of 15] in the GI area, 75% [three of four] in soft tissue, and 25% [one of four] in bone).
The immunohistochemical expression of the main single antigens according to the origin of the lymphoma is summarized in Table 2. Patients with primary extranodal lymphoma more frequently presented with bcl-6 expression than those with a lymphoma of nodal origin (38 of 43 [88%] v 71 of 100 [71%]), respectively; P = .03). No significant differences were found with regard to the remainder antigens. In the 130 patients with available data, the distribution according to the differentiation profile was the following: GC-CD10+ profile, 30 cases (23%); GC-CD10– profile, 38 cases (29%); pGC profile, 62 cases (48%). The differentiation profile according to the primary site of the lymphoma is presented in Table 2. Of note, DLBCLs from the GI area expressed bcl-6 in most cases (94%), with a GC-CD10– differentiation profile in 56% of cases.
bcl-2 expression is also detailed in Table 2. No significant differences were found among the groups. Nine of the 73 patients in whom the analysis was performed presented with the bcl-2/JH rearrangement (five at MBR and three at mcr). The distribution by primary sites was as follows: six of 47 primary nodal DLBCLs (MBR, four patients; mcr, two patients), 0 of six Waldeyer's ring, 0 of five GI, and three of 15 from other primary extranodal sites (bone, one patient [MBR]; breast, one patient [mcr]; and kidney, one patient [mcr]).
The mitotic index and the Ki-67 expression, as well as p53 expression did not significantly differ among the different primary sites. Finally, p27 expression was more frequently negative (< 25% of cells) in the DLBCLs from Waldeyer's area (11 of 12 patients [92%]), than in the other nodal (44 of 80 [55%]; P = .02) and primary extranodal lymphomas (22 of 38 [58%]; P = .04).
Response to Treatment and Outcome: Prognostic Analysis
No differences were found in the treatment given to the patients according to the primary nodal or extranodal origin of the lymphoma. CR rate was 63%, with this figure being 61% and 68% for primary nodal and extranodal cases, respectively (P = .2; Table 3). The CR rate was higher in Waldeyer's ring (35 of 41, 85%) and GI lymphomas (36 of 45, 80%) than in the other groups (P = .002 and P = .01, respectively). The following variables predicted for CR achievement: age younger than 60 years; ambulatory performance status (Eastern Cooperative Oncology Group [ECOG] score < 2); absence of B symptoms; absence of bulky disease; early Ann Arbor stage; no bone marrow involvement; normal hemoglobin, platelet count, serum albumin, LDH, and 2m levels; as well as the IPI. No single antigen expression or differentiation profile predicted for CR. In the logistic regression analysis, ambulatory performance status (ECOG score < 2; P < .001; relative risk [RR], 3.0), age younger than 60 years (P < .001; RR, 2.6), normal serum LDH (P = .04; RR, 1.8), primary Waldeyer's ring (P = .04; RR, 4.2), and primary GI lymphomas (P = .04; RR, 3.3) were the most important variables predicting CR achievement.
Seventy-four of 240 CR patients eventually relapsed, with a risk of relapse at 5 years of 29% (95% CI, 23% to 35%). The risk of relapse according to the primary site of the lymphomas is detailed in Table 3 and in Figure 1.
Five-year overall survival (OS) of the entire series was 52% (95% CI, 47% to 57%). Unfavorable variables predicting OS were: age older than 60 years, presence of B symptoms, poor performance status (ECOG score 2), presence of bulky disease, advanced Ann Arbor stage (III-IV), extranodal involvement at two or more sites, bone marrow involvement, anemia (hemoglobin < 12 g/L), thrombocytopenia (platelet count < 100 x 109/L), high erythrocyte sedimentation rate (> 40 mm/h), low serum albumin levels, high serum LDH levels, and high 2m. In addition, bcl-2 protein expression predicted a poor OS. Finally, as expected, the IPI also had a high value to predict OS (P < .0001). In Table 3 and Figure 2, OS according to the primary site of the lymphoma is listed and plotted, respectively. Waldeyer's ring lymphomas showed the best OS (5-year OS 77%; 95% CI, 63% to 91%; P = .007 v the other groups). As a whole, primary nodal and extranodal lymphomas had a 5-year OS of 49% and 59%, respectively (P = .1). The group of GI lymphomas showed a remarkably good OS (68% at 5 years; 95% CI, 54% to 82%; P = .02 v the other groups), whereas the same value was lower for the lymphomas arising at the other sites. However, when the analysis of OS according to the primary site (nodal, Waldeyer's ring, GI, other extranodal) was performed in patients with either favorable or unfavorable IPI, no significant differences were observed according to the site of presentation, as shown in Figures 3A and B. Main prognostic factors for OS did not differ among the different sites of presentation.
To further assess the clinical interest of primary site of presentation, a multivariate analysis was performed, including all the significant variables predicting OS in the univariate analysis (namely, age < 60 v > 60 years, B symptoms, performance status [ECOG score < 2 v 2], bulky disease, Ann Arbor stage [I-II v III/IV], extranodal involvement [fewer than two sites v two or more sites], bone marrow infiltration, and serum LDH [normal v high]; serum albumin [low v normal]; and 2m [normal v high]; erythrocyte sedimentation rate was excluded due to the relatively high quantity of missing data), along with the primary site of the lymphoma (nodal, Waldeyer's ring, GI, and other extranodal sites). In this model, poor performance status (P = .037; RR, 1.7), bulky disease (P = .01; RR, 1.8), and high serum 2m (P = .002; RR, 2) maintained its prognostic value to predict poor OS. In addition, IPI (low- v low/intermediate- v high/intermediate- v high-risk) was incorporated into the analysis (whereas age, performance status, stage, extranodal involvement, and serum LDH were excluded as redundant variables). In this model, with 217 patients, IPI (P = .002; RR, 1.4), bulky disease (P = .006; RR, 1.8), and 2m (P = .01; RR, 1.8) were the most important factors in predicting OS. No primary site of the lymphoma (nodal, Waldeyer's ring, GI, and other extranodal sites) maintained independent value in the multivariate analyses. Finally, a further study was carried out in the 130 patients in whom all the biologic variables, including bcl-2 expression and the differentiation profile, were available, in addition to the clinical features. Only IPI (P = .04; RR, 1.4) and bulky disease (P = .03; RR, 2.3) showed prognostic importance for OS in this group.
DISCUSSION
The diversity in the clinical presentation, morphology, immunophenotype, and genetic and molecular alterations strongly suggests that DLBCL is a heterogeneous group of aggressive B-cell lymphomas rather than a single clinicopathologic entity.14-16 Such heterogeneity has been recognized in the REAL and WHO classifications.14,15 However, the different approaches to separate nosological entities within DLBCL have failed, owing in part to the lack of reproducibility of histologic criteria. The primary site of the lymphoma, either the lymph node or different extranodal territories, has been suggested as a criterion that might separate two different groups of DLBCL, nodal and extranodal, with particular clinicobiological characteristics and different natural history.3,4,12 However, the current classification of the lymphomas is largely based on the clinicopathologic features and, at present, does not take into consideration the primary site of the lymphoma, regarding it simply as additional information.15,26
The consideration of a lymphoma as primary nodal or extranodal is a controversial issue.11 Patients with purely nodal or extranodal involvement are easily classified. In some studies, only localized extranodal lymphomas have been defined as primary extranodal.6,32,33 However, since of course extranodal lymphomas can also disseminate, this restrictive criterion leads to an incomplete picture of these lymphomas. For this reason, those cases with "clinically relevant" extranodal involvement are usually considered as extranodal.2-4,11,12 The cases with extensive disease, involving both nodal and extranodal areas, are difficult to categorize. In the present report, following previous publications,11 these cases were included among the nodal lymphomas. However, this may represent a bias against the nodal group. On the other hand, a second controversial issue is the consideration of particular territories as nodal or extranodal, such as Waldeyer's ring, spleen, or bone marrow. In general, Waldeyer's ring, whose lymphoid tissue is similar to that of the lymph node, is currently included among nodal areas, whereas bone marrow lymphomas are usually regarded as extranodal. All of the above issues could perfectly explain the disparate results found in different series. Therefore, the generic study of DLBCL as either nodal or extranodal may be inappropriate. The analysis of the characteristics of each particular primary site would be more informative.
During the last decade, cytogenetic and molecular analyses of different lymphoid neoplasms have demonstrated a close relationship between particular genetic alterations and the clinicopathologic manifestations of the disease, as well as their value as prognostic markers.17-19,34,35 In this setting, it has been suggested that genetic differences between nodal and extranodal DLBCLs may exist. Bcl-2/JH rearrangement, found in the majority of follicular lymphomas, is present in up to one third of DLBCL cases. It has been described that this alteration is more frequent in nodal than in extranodal DLBCLs.20 In the present series, however, the proportion of bcl-2/JH rearrangement in nodal and extranodal cases was similar (13% v 15%), though the number of patients tested (N = 73) was relatively low. On the other hand, in agreement with previous descriptions, the expression of the bcl-2 protein was basically independent from the presence of bcl-2/JH rearrangement.28,36 In addition, no significant differences were found in bcl-2 expression according to the primary localization of the disease. Recently, our group published the data of a small subset of these patients with DLBCL herein analyzed in whom comparative genomic hybridization (CGH) studies were performed.37 18q gains, a frequent genetic alteration seen in 20% of the cases, were observed more frequently in DLBCL of primary nodal origin than in extranodal (92% v 53%, respectively). 18q gains and amplifications, that were poor risk factor, did not show a correlation with the bcl-2/JH rearrangement, nor with the bcl-2 protein overexpression. Furthermore, single gene alterations at c-myc,20 bcl-6,20,21 rel,22,23 and fas24,25 have been described more frequently in extranodal DLBCLs. However, the biologic features found in small series should be confirmed in large series of patients with DLBCL. Thus, the relationship between rel alterations and extranodal lymphomas, suggested some time ago,22 has not been confirmed.38
More recently, DLBCLs have been separated by means of microarrays in two different groups according to the germinal-center or postgerminal-center genetic signature, with different clinicobiologic features and outcome.17-19,34,35 However, there is still no information available about the gene expression profile according to the primary site of the lymphoma. In this study, using immunophenotyping, we defined three groups, germinal-center CD10+, germinal-center CD10–, and postgerminal-center that mimic the genetic profiles.28 In this setting, patients with primary extranodal DLBCL, particularly GI, showed a germinal-center CD10– profile more frequently than nodal lymphomas. Nevertheless, the study of the gene profile for DLBCLs from different sites is still warranted.
In the present series, the groups of nodal and extranodal DLBCLs were not homogeneous from the clinical standpoint. Lymphomas arising in two specific sites (one nodal, Waldeyer's ring; one extranodal, GI) showed very favorable features at diagnosis (eg, early stage, absence of bone marrow infiltration, normal serum LDH, and low-risk IPI), whereas the DLBCLs arising in the remaining areas (lymph nodes or other extranodal sites) presented with poorer characteristics. It is of note that no GI patient, and only two of 42 Waldeyer's ring patients had bone marrow infiltration, and that 72% of Waldeyer's ring patients had a low-risk IPI (Table 1). In terms of response to therapy, risk of relapse and OS, Waldeyer's, and GI lymphomas also showed a notably better outcome than that of the other groups. Thus, it is not the nodal or extranodal presentation, but some specific sites that may be related to particular clinicobiologic features and to the outcome of the patients. On the other hand, with regard to the prognostic factors for OS, none of the primary sites showed an independent predictive weight when standard variables were included in multivariate analyses. In fact, IPI, bulky disease, and serum 2m were the most important factors for OS, irrespective of the primary site of the lymphoma (Figs 3A and B).
In conclusion, in the present series of a large number of DLBCL patients from a single institution, the primary site of the disease was associated with particular clinicopathologic features and with the outcome, although the latter largely depended on other prognostic variables such as IPI or 2m.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by grants FIS-IP03/0473, Spanish Ministry of Health; "Red de Centros de Investigación del Cáncer y de Grupos para el Estudio de Linfomas, Instituto Carlos III–FIS", and the Jose Carreras Foundation Against Leukemia (EM-P/04 and CR-P/04).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Harris NL: Mature B-cell neoplasms, in EJ Jaffe, NL Harris, H Stein et al (eds): WHO Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France, IARCPress, 2001, pp 121-126
Zucca E, Cavalli F: Extranodal lymphomas. Ann Oncol 11:S219-S222, 2000 (suppl 3)
Zucca E, Roggero E, Bertoni F, et al: Primary extranodal non-Hodgkin's lymphomas, part 1: Gastrointestinal, cutaneous and genitourinary lymphomas. Ann Oncol 8:727-737, 1997
Zucca E, Roggero E, Bertoni F: Primary extranodal non-Hodgkin's lymphomas, part 2: Head and neck, central nervous system and other less common sites. Ann Oncol 10:1023-1033, 1999
Groves FD, Linet MS, Travis LB, et al: Cancer surveillance series: Non-Hodgkin's lymphoma incidence by histologic subtype in the United States from 1978 through 1995. J Natl Cancer Inst 92:1240-1251, 2000
Freeman C, Berg JW, Cutler SJ: Occurrence and prognosis of extranodal lymphomas. Cancer 29:252-260, 1972
Rudders RA, Ross ME, DeLellis RA: Primary extranodal lymphoma: Response to treatment and factors influencing prognosis. Cancer 42:406-416, 1978
Otter R, Gerrits WBJ, Sandt MMVD, et al: Primary extranodal and nodal non-Hodgkin's lymphoma: A survey of a population-based registry. Eur J Cancer Clin Oncol 25:1203-1210, 1989
d'Amore F, Christensen BE, Brincker H, et al: Clinicopathological features and prognostic factors in extranodal non-Hodgkin lymphomas. Eur J Cancer 27:1201-1208, 1991
Newton R, Ferlay J, Beral V, et al: The epidemiology of non-Hodgkin's lymphoma: Comparison of nodal and extra-nodal sites. Int J Cancer 72:923-930, 1997
Krol ADG, leCessie S, Snijder S, et al: Primary extranodal non-Hodgkin's lymphoma (NHL): The impact of alternative definitions tested in the Comprehensive Cancer Centre West population-based NHL registry. Ann Oncol 14:131-139, 2003
Moller MB, Pedersen NT, Christensen BE: Diffuse large B-cell lymphoma: Clinical implications of extranodal versus nodal presentation—A population-based study of 1575 cases. Br J Haematol 124:151-159, 2004
A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin's lymphoma: The Non-Hodgkin's Lymphoma Classification Project. Blood 89:3909-3918, 1997
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
Gatter KC, Warnke RA: Diffuse large B-cell lymphoma, in Jaffe ES, Harris NL, Stein H (eds): WHO Classification of Tumors: Pathology and Genetics of Tumors of Haematopoietic and Lymphoid Tissues. Lyon, France, IARCPress, 2001, pp 171-174
Pileri SA, Dirnhofer S, Went Ph, et al: Diffuse large B-cell lymphoma: One or more entities Present controversies and possible tools for its subclassification. Histopathology 41:482-509, 2002
Alizadeh AA, Elsen MB, Davis RE, et al: Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 403:503-511, 2000
Shipp MA, Ross KN, Tamayo P, et al: Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nature Med 8:68-74, 2002
Rosenwald A, Wright G, Chan WC, et al: The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 346:1937-1947, 2002
Kramer MH, Hermans J, Wijburg E, et al: Clinical relevance of BCL2, BCL6, and MYC rearrangements in diffuse large B-cell lymphoma. Blood 92:3152-3162, 1998
Muramatsu M, Akasaka T, Kadowaki N, et al: Rearrangement of the BCL6 gene in B-cell lymphoid neoplasms: Comparison with lymphomas associated with BCL2 rearrangement. Br J Haematol 93:911-920, 1996
Houldsworth J, Mathews S, Rao PH, et al: REL proto-oncogene is frequently amplified in extranodal diffuse large cell lymphoma. Blood 87:25-29, 1996
Rao PH, Houldsworth J, Dyomina K, et al: Chromosomal and gene amplification in diffuse large B-cell lymphoma. Blood 92:234-240, 1998
Gronbaek K, Straten PT, Ralfkiaer E, et al: Somatic Fas mutations in non-Hodgkin's lymphoma: Association with extranodal disease and autoimmunity. Blood 92:3018-3024, 1998
Wohlfart S, Sebinger D, Gruber P, et al: FAS (CD95) mutations are rare in gastric MALT lymphoma but occur more frequently in primary gastric diffuse large B-cell lymphoma. Am J Pathol 164:1081-1089, 2004
Harris NL, Jaffe ES, Diebold J, et al: The World Health Organization classification of neoplasms of the hematopoietic and lymphoid tissues: Report of the Clinical Advisory Committee meeting -Airlie House, Virginia, November 1997. Hematol J 1:53-66, 2000
The International Non-Hodgkin's Lymphoma Prognostic Factors Project: A predictive model for aggressive non-Hodgkin's lymphoma. N Engl J Med 329:987-994, 1993
Colomo L, López-Guillermo A, Perales M, et al: Clinical impact of the differentiation profile assessed by immunophenotyping in patients with diffuse large B-cell lymphoma. Blood 101:78-84, 2003
Kaplan GL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Peto R, Pike MC: Conservatism of the approximation (O-E)2/E in the log-rank test for survival data or tumour incidence data. Biometrics 29:759-784, 1973
Cox DR: Regression models and life tables. J R Stat Assoc 34:187-220, 1972
Dawson IP, Cornes JS, Morson BC: Primary malignant lymphoid tumours of the intestinal tract: Report of 37 cases with a study of factors influencing the prognosis. Br J Surg 49:80-89, 1961
Gospodarowicz MK, Sutcliffe SB: The extranodal lymphomas. Semin Radiat Oncol 5:281-300, 1995
Lossos IS, Czerwinski DK, Alizadeh AA, et al: Prediction of survival in diffuse large-B-cell lymphoma based on the expression of six genes. N Engl J Med 350:1828-1837, 2004
Sáez AI, Sáez AJ, Artiga MJ, et al: Building an outcome predictor model for diffuse large B-cell lymphoma. Am J Pathol 164:613-622, 2004
Gascoyne RD, Adomat SA, Krajewski S, et al: Prognostic significance of bcl-2 protein expression and bcl-2 gene rearrangement in diffuse aggressive non-Hodgkin's lymphoma. Blood 90:244-251, 1997
Beà S, Colomo L, López-Guillermo A, et al: Clinicopathologic significance and prognostic value of chromosomal imbalances in diffuse large B-cell lymphomas. J Clin Oncol 22:3498-3506, 2004
Houldsworth J, Olshen AB, Cattorelli G, et al: Relationship between REL amplification, REL function, and clinical and biologic features in diffuse large B-cell lymphomas. Blood 103:1862-1868, 2004(Armando López-Guillermo, )
ABSTRACT
PURPOSE: To study the main clinicobiologic features, response, and outcome of patients with diffuse large B-cell lymphoma (DLBCL) according to the primary site, lymph node, or different extranodal organs of the disease.
PATIENTS AND METHODS: We included 382 patients consecutively diagnosed with DLBCL in a single institution during a 13-year period. Morphology, immunophenotyping, proliferation index, differentiation profile, bcl-2/JH rearrangement, and clinical characteristics were analyzed according to the primary site of the lymphoma.
RESULTS: Sites of the disease were: lymph node, 222 cases (58%); Waldeyer's ring (WR), 42 (11%); and extranodal sites, 118 (31%), including GI tract in 45 cases. Primary extranodal cases, particularly GI, showed a bcl-6 expression more frequently than nodal cases. Patients with primary WR or GI lymphomas presented with early-stage disease, no marrow infiltration, normal serum lactate dehydrogenase, and low- to low/intermediate-risk international prognostic index (IPI) more frequently than the remainder. Complete response (CR) rate was 63%, with WR and GI lymphomas having a higher CR rate (85% and 80%, respectively) than the other groups. In the whole series, 5-year overall survival (OS) was 52%. Patients with WR or GI lymphomas showed better OS (5-year OS: 77% and 68%, respectively) than patients with nodal or other extranodal sites. In the multivariate analysis, IPI, bulky disease, and 2-microglobulin were the main variables to predict OS; no nodal or extranodal site maintained their prognostic value.
CONCLUSION: In the present series, the primary site of disease was associated with particular clinicopathologic features and outcome, though the latter largely depended on other factors.
INTRODUCTION
Approximately one third of the non-Hodgkin's lymphomas arise in tissues different from the lymph node, and for this reason, they are usually termed extranodal lymphomas.1-4 It has been shown recently that during the last two decades the incidence of lymphomas has increased, and that extranodal lymphomas increased more rapidly than nodal.3,5 The study of extranodal lymphomas as a group, regarding etiopathogenesis, biologic features, clinical characteristics, and outcome, has been the subject of several publications.3,4,6-12 However, most of them include heterogeneous series of patients in which several histologic subgroups are usually merged. The latter makes it difficult to reach conclusions, since it is not clear whether the clinicobiologic differences between nodal and extranodal are due to the site of the disease, or merely show differences in the particular spectrum of histologies of each site. On the other hand, most studies refer only to clinical aspects or are based on a relatively small number of patients. Therefore, clinicobiologic studies in histologically homogeneous series of patients are warranted.
Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin's lymphoma in Western countries, representing approximately 30% of the total.13 Between one third and one quarter of DLBCLs have a primary extranodal origin. Although considered a single entity in the Revised European-American Lymphoma (REAL)/WHO classification, DLBCL most likely includes different clinicopathologic entities that are currently difficult to separate using standard procedures.14-16 In fact, morphology, immunophenotyping, and genetics are heterogeneous among DLBCLs, and, more recently, microarray technology has been able to distinguish two groups of DLBCL according to the "germinal-center" or "postgerminal-center" (also called "activated") genetic signatures.17-19 Particular clinical and biologic characteristics have been suggested for DLBCLs arising in the lymph node, versus those in extranodal sites, including alterations in some genes such as bcl-2,20 bcl-6,20,21 c-myc,20 rel,22,23 and fas.24,25 Based on these data, a separate genetic origin for nodal and extranodal lymphomas has been claimed, and it has been suggested that these forms should be regarded as separate nosological entities. However, there is no consensus on this aspect, and in the WHO classification, it is suggested that the information regarding the nodal or extranodal primary site should be collected for investigational purposes only.15,26
The aim of the present study was to analyze the main clinicobiologic features at diagnosis, as well as the response to therapy and the outcome of a large series of patients diagnosed with DLBCL and homogeneously treated and followed-up in a single institution, according to the primary site (either lymph node or different extranodal territories) of the disease.
PATIENTS AND METHODS
Patients
Three hundred eighty-two patients consecutively diagnosed with a DLBCL between September 1987 and September 2000, and followed up in a single institution were selected for the present study. The cases with recognized disease phase of a follicular lymphoma or another type of indolent lymphoma with subsequent transformation into a DLBCL were not included. In addition, immunodeficiency-associated tumors, primary mediastinal, intravascular, and primary effusion lymphomas, were also excluded from the study.
Median age of the patients was 60 years (range, 14 to 94 years), and the male-female distribution was 188:194. The main characteristics of the patients at diagnosis are listed in Table 1. Advanced stage (Ann Arbor III or IV) was observed in 181 patients (47%), and any extranodal involvement, in 270 patients (71%), including bone marrow infiltration in 71 cases (19%). One hundred ninety-six (55%) of 357 patients with available data presented with high serum lactate dehydrogenase (LDH) levels, whereas the distribution according to the International Prognostic Index (IPI)27 was the following: low-risk, 135 patients (37%); low/intermediate, 71 patients (20%); high/intermediate, 73 patients (20%); high-risk, 84 patients (23%); and nonassessable, 19 cases. The main initial and evolutive variables, including the histologic parameters indicated below, were recorded and analyzed for prognosis.
Staging maneuvers included patient history and physical examination (including Waldeyer's ring area); CBCs and serum biochemistry, including LDH and 2-microglobulin (2m) levels; computed tomography (CT) scan of chest, abdomen, and pelvis; as well as bone marrow biopsy. The distinction between nodal and extranodal was not taken into account in planning the treatment. Three hundred sixty-nine patients were treated with combination chemotherapy: adriamycin-containing regimens in 344 patients (90%; in most cases cyclophosphamide, doxorubicin, vincristine, and prednisone) and combination chemotherapy without adriamycin in the remaining 25 patients. Post-therapy restaging consisted of the repetition of the previously abnormal tests and/or biopsies. Response was assessed according to conventional criteria. Overall, 240 patients achieved a complete response (CR; 63%); 44 patients, a partial response (12%); and 94 patients failed to respond to treatment. In four cases, the response was not assessable. After a median follow-up of 6.5 years (range, 0.2 to 15.3 years) for surviving patients, 189 patients had died. The 5-year and 10-year overall survival rates were 52% (95% CI, 47% to 57%) and 46% (95%CI, 40% to 52%), respectively.
Nodal/Extranodal Definitions
Lymphomas presenting in extranodal organs with no or only minor lymph node involvement were considered primary extranodal. Lymphomas with lymph node involvement clinically dominant, as well as those presenting at the spleen, were considered as primary nodal. Lymphomas arising in the Waldeyer's ring, though considered nodal, were analyzed separately. Bone marrow was considered an extranodal site. Finally, those lymphomas with extensive disease involving both nodal and extranodal sites were considered nodal.
Histologic Features
The diagnosis of DLBCL was based in all cases on the criteria established in the WHO classification.15 Morphologic and immunophenotypic studies were performed in 203 patients in whom adequate histologic material was available. For the morphologic analysis, histologic slides were reviewed in all the cases by two different observers (L.C. and E.C.). Morphologic subclassification was performed as previously described.28 The panel of monoclonal antibodies included antibodies against the following antigens: CD20, CD79a, CD3, CD5, CD10, MUM1, CD138, bcl-2, bcl-6, Ki-67, p53, and p27. The antibodies and the immunohistochemical conditions of use have been previously described.28 Tumor proliferation was analyzed by two different methods: (1) analysis of mitotic index, considering a cutoff of 25 mitoses per 10 high-power fields, and (2) proliferative index assessed by Ki-67 immunostaining.
The patients were clustered into four groups according to their immunophenotypic profile in order to assess the pattern of differentiation28: germinal center origin with CD10 positivity (GC-CD10+): CD10+/bcl-6+/MUM1–/CD138–; germinal center origin–CD10 negative (GC-CD10–): CD10-/bcl-6+/MUM1–/CD138–; postgerminal center origin (pGC): CD10–/bcl-6±/MUM1+/CD138–; and plasmablastic: CD10–/bcl-6–/MUM1+/CD138+.
Detection of the bcl-2/JH rearrangement involving the bcl-2 major breakpoint region (MBR) and minor cluster region (mcr) was performed in 73 cases by a polymerase chain reaction technique.28
Statistical Analysis
Categorical data were compared using Fisher's exact test for a two-sided P value, whereas for ordinal data, nonparametric tests were used. Bonferroni correction for multiple comparisons was applied when necessary. The multivariate analysis of the variables predicting response was performed by using a logistic regression. The actuarial survival analysis was performed according to the method described by Kaplan and Meier,29 and the curves were compared by the log-rank test.30 The multivariate analysis for survival was performed by using the stepwise proportional hazards model (Cox).31
RESULTS
Distribution and Clinical Features
One hundred eighteen patients (31%) presented at a primary extranodal site, and 42 patients (11%), at Waldeyer's ring, while the remaining 222 patients (58%) had a primary nodal DLBCL, including one at the spleen. The distribution according to the extranodal site was as follows: GI tract (45 cases, 12% of the overall series; including gastric lymphoma in 37 cases); liver (seven cases; 2%); soft tissue (17 cases; 4.5%); CNS (10 cases; 2.5%); bone (nine cases; 2%); skin (five cases; 1%); breast (five cases); kidney (five cases); testis/ovary (five cases); thyroid (three cases); bone marrow (two cases); and lung, prostate, uterus, and pericardium (one case each). Finally, one patient showed a double extranodal lymphoma (bilateral breast and ovary).
The main clinical characteristics at diagnosis according to the primary site of the lymphoma are detailed in Table 1. Patients with Waldeyer's ring DLBCL presented more frequently with early-stage disease; ambulatory performance status; absence of "B" symptoms; bulky disease and bone marrow infiltration; normal serum albumin, LDH, and 2m levels; and low- or low/intermediate-risk IPI (P < .01 in all cases v the other groups). GI DLBCL was also early-stage, with absence of bone marrow infiltration, normal serum LDH, and had low- or low/intermediate-risk IPI more so than the other groups (P < .03 in all cases). Finally, the whole group of patients with nodal DLBCL, including those with Waldeyers' ring DLBCL, showed bulky disease (36% v 23%; P = .02), bone marrow infiltration (23% v 10%; P = .004), and high serum LDH levels (58% v 47%; P = .05) more frequently than the extranodal lymphomas.
Morphology and Immunophenotypic Profile
The 172 DLBCL patients in whom adequate material was available to fully assess morphology were classified as: centroblastic lymphoma, 57 cases (33%); centroblastic polymorphic with less than 50% immunoblasts, 23 cases (13%); centroblastic polymorphic with 50% immunoblasts, 16 cases (9%); immunoblastic, 15 cases (9%); other, five cases (3%; including two patients with plasmablastic, two with a pleomorphic, and one patient with a T-cell–rich DLBCL). The remaining 56 cases were not otherwise classifiable. The proportion of DLBCL with immunoblastic features ( 50% immunoblastic cells) was 33% (24 of 72) for primary nodal lymphomas, 38% (five of 13) for Waldeyer's ring lymphomas, and 23% (seven of 31) for extranodal lymphomas (including 13% [two of 15] in the GI area, 75% [three of four] in soft tissue, and 25% [one of four] in bone).
The immunohistochemical expression of the main single antigens according to the origin of the lymphoma is summarized in Table 2. Patients with primary extranodal lymphoma more frequently presented with bcl-6 expression than those with a lymphoma of nodal origin (38 of 43 [88%] v 71 of 100 [71%]), respectively; P = .03). No significant differences were found with regard to the remainder antigens. In the 130 patients with available data, the distribution according to the differentiation profile was the following: GC-CD10+ profile, 30 cases (23%); GC-CD10– profile, 38 cases (29%); pGC profile, 62 cases (48%). The differentiation profile according to the primary site of the lymphoma is presented in Table 2. Of note, DLBCLs from the GI area expressed bcl-6 in most cases (94%), with a GC-CD10– differentiation profile in 56% of cases.
bcl-2 expression is also detailed in Table 2. No significant differences were found among the groups. Nine of the 73 patients in whom the analysis was performed presented with the bcl-2/JH rearrangement (five at MBR and three at mcr). The distribution by primary sites was as follows: six of 47 primary nodal DLBCLs (MBR, four patients; mcr, two patients), 0 of six Waldeyer's ring, 0 of five GI, and three of 15 from other primary extranodal sites (bone, one patient [MBR]; breast, one patient [mcr]; and kidney, one patient [mcr]).
The mitotic index and the Ki-67 expression, as well as p53 expression did not significantly differ among the different primary sites. Finally, p27 expression was more frequently negative (< 25% of cells) in the DLBCLs from Waldeyer's area (11 of 12 patients [92%]), than in the other nodal (44 of 80 [55%]; P = .02) and primary extranodal lymphomas (22 of 38 [58%]; P = .04).
Response to Treatment and Outcome: Prognostic Analysis
No differences were found in the treatment given to the patients according to the primary nodal or extranodal origin of the lymphoma. CR rate was 63%, with this figure being 61% and 68% for primary nodal and extranodal cases, respectively (P = .2; Table 3). The CR rate was higher in Waldeyer's ring (35 of 41, 85%) and GI lymphomas (36 of 45, 80%) than in the other groups (P = .002 and P = .01, respectively). The following variables predicted for CR achievement: age younger than 60 years; ambulatory performance status (Eastern Cooperative Oncology Group [ECOG] score < 2); absence of B symptoms; absence of bulky disease; early Ann Arbor stage; no bone marrow involvement; normal hemoglobin, platelet count, serum albumin, LDH, and 2m levels; as well as the IPI. No single antigen expression or differentiation profile predicted for CR. In the logistic regression analysis, ambulatory performance status (ECOG score < 2; P < .001; relative risk [RR], 3.0), age younger than 60 years (P < .001; RR, 2.6), normal serum LDH (P = .04; RR, 1.8), primary Waldeyer's ring (P = .04; RR, 4.2), and primary GI lymphomas (P = .04; RR, 3.3) were the most important variables predicting CR achievement.
Seventy-four of 240 CR patients eventually relapsed, with a risk of relapse at 5 years of 29% (95% CI, 23% to 35%). The risk of relapse according to the primary site of the lymphomas is detailed in Table 3 and in Figure 1.
Five-year overall survival (OS) of the entire series was 52% (95% CI, 47% to 57%). Unfavorable variables predicting OS were: age older than 60 years, presence of B symptoms, poor performance status (ECOG score 2), presence of bulky disease, advanced Ann Arbor stage (III-IV), extranodal involvement at two or more sites, bone marrow involvement, anemia (hemoglobin < 12 g/L), thrombocytopenia (platelet count < 100 x 109/L), high erythrocyte sedimentation rate (> 40 mm/h), low serum albumin levels, high serum LDH levels, and high 2m. In addition, bcl-2 protein expression predicted a poor OS. Finally, as expected, the IPI also had a high value to predict OS (P < .0001). In Table 3 and Figure 2, OS according to the primary site of the lymphoma is listed and plotted, respectively. Waldeyer's ring lymphomas showed the best OS (5-year OS 77%; 95% CI, 63% to 91%; P = .007 v the other groups). As a whole, primary nodal and extranodal lymphomas had a 5-year OS of 49% and 59%, respectively (P = .1). The group of GI lymphomas showed a remarkably good OS (68% at 5 years; 95% CI, 54% to 82%; P = .02 v the other groups), whereas the same value was lower for the lymphomas arising at the other sites. However, when the analysis of OS according to the primary site (nodal, Waldeyer's ring, GI, other extranodal) was performed in patients with either favorable or unfavorable IPI, no significant differences were observed according to the site of presentation, as shown in Figures 3A and B. Main prognostic factors for OS did not differ among the different sites of presentation.
To further assess the clinical interest of primary site of presentation, a multivariate analysis was performed, including all the significant variables predicting OS in the univariate analysis (namely, age < 60 v > 60 years, B symptoms, performance status [ECOG score < 2 v 2], bulky disease, Ann Arbor stage [I-II v III/IV], extranodal involvement [fewer than two sites v two or more sites], bone marrow infiltration, and serum LDH [normal v high]; serum albumin [low v normal]; and 2m [normal v high]; erythrocyte sedimentation rate was excluded due to the relatively high quantity of missing data), along with the primary site of the lymphoma (nodal, Waldeyer's ring, GI, and other extranodal sites). In this model, poor performance status (P = .037; RR, 1.7), bulky disease (P = .01; RR, 1.8), and high serum 2m (P = .002; RR, 2) maintained its prognostic value to predict poor OS. In addition, IPI (low- v low/intermediate- v high/intermediate- v high-risk) was incorporated into the analysis (whereas age, performance status, stage, extranodal involvement, and serum LDH were excluded as redundant variables). In this model, with 217 patients, IPI (P = .002; RR, 1.4), bulky disease (P = .006; RR, 1.8), and 2m (P = .01; RR, 1.8) were the most important factors in predicting OS. No primary site of the lymphoma (nodal, Waldeyer's ring, GI, and other extranodal sites) maintained independent value in the multivariate analyses. Finally, a further study was carried out in the 130 patients in whom all the biologic variables, including bcl-2 expression and the differentiation profile, were available, in addition to the clinical features. Only IPI (P = .04; RR, 1.4) and bulky disease (P = .03; RR, 2.3) showed prognostic importance for OS in this group.
DISCUSSION
The diversity in the clinical presentation, morphology, immunophenotype, and genetic and molecular alterations strongly suggests that DLBCL is a heterogeneous group of aggressive B-cell lymphomas rather than a single clinicopathologic entity.14-16 Such heterogeneity has been recognized in the REAL and WHO classifications.14,15 However, the different approaches to separate nosological entities within DLBCL have failed, owing in part to the lack of reproducibility of histologic criteria. The primary site of the lymphoma, either the lymph node or different extranodal territories, has been suggested as a criterion that might separate two different groups of DLBCL, nodal and extranodal, with particular clinicobiological characteristics and different natural history.3,4,12 However, the current classification of the lymphomas is largely based on the clinicopathologic features and, at present, does not take into consideration the primary site of the lymphoma, regarding it simply as additional information.15,26
The consideration of a lymphoma as primary nodal or extranodal is a controversial issue.11 Patients with purely nodal or extranodal involvement are easily classified. In some studies, only localized extranodal lymphomas have been defined as primary extranodal.6,32,33 However, since of course extranodal lymphomas can also disseminate, this restrictive criterion leads to an incomplete picture of these lymphomas. For this reason, those cases with "clinically relevant" extranodal involvement are usually considered as extranodal.2-4,11,12 The cases with extensive disease, involving both nodal and extranodal areas, are difficult to categorize. In the present report, following previous publications,11 these cases were included among the nodal lymphomas. However, this may represent a bias against the nodal group. On the other hand, a second controversial issue is the consideration of particular territories as nodal or extranodal, such as Waldeyer's ring, spleen, or bone marrow. In general, Waldeyer's ring, whose lymphoid tissue is similar to that of the lymph node, is currently included among nodal areas, whereas bone marrow lymphomas are usually regarded as extranodal. All of the above issues could perfectly explain the disparate results found in different series. Therefore, the generic study of DLBCL as either nodal or extranodal may be inappropriate. The analysis of the characteristics of each particular primary site would be more informative.
During the last decade, cytogenetic and molecular analyses of different lymphoid neoplasms have demonstrated a close relationship between particular genetic alterations and the clinicopathologic manifestations of the disease, as well as their value as prognostic markers.17-19,34,35 In this setting, it has been suggested that genetic differences between nodal and extranodal DLBCLs may exist. Bcl-2/JH rearrangement, found in the majority of follicular lymphomas, is present in up to one third of DLBCL cases. It has been described that this alteration is more frequent in nodal than in extranodal DLBCLs.20 In the present series, however, the proportion of bcl-2/JH rearrangement in nodal and extranodal cases was similar (13% v 15%), though the number of patients tested (N = 73) was relatively low. On the other hand, in agreement with previous descriptions, the expression of the bcl-2 protein was basically independent from the presence of bcl-2/JH rearrangement.28,36 In addition, no significant differences were found in bcl-2 expression according to the primary localization of the disease. Recently, our group published the data of a small subset of these patients with DLBCL herein analyzed in whom comparative genomic hybridization (CGH) studies were performed.37 18q gains, a frequent genetic alteration seen in 20% of the cases, were observed more frequently in DLBCL of primary nodal origin than in extranodal (92% v 53%, respectively). 18q gains and amplifications, that were poor risk factor, did not show a correlation with the bcl-2/JH rearrangement, nor with the bcl-2 protein overexpression. Furthermore, single gene alterations at c-myc,20 bcl-6,20,21 rel,22,23 and fas24,25 have been described more frequently in extranodal DLBCLs. However, the biologic features found in small series should be confirmed in large series of patients with DLBCL. Thus, the relationship between rel alterations and extranodal lymphomas, suggested some time ago,22 has not been confirmed.38
More recently, DLBCLs have been separated by means of microarrays in two different groups according to the germinal-center or postgerminal-center genetic signature, with different clinicobiologic features and outcome.17-19,34,35 However, there is still no information available about the gene expression profile according to the primary site of the lymphoma. In this study, using immunophenotyping, we defined three groups, germinal-center CD10+, germinal-center CD10–, and postgerminal-center that mimic the genetic profiles.28 In this setting, patients with primary extranodal DLBCL, particularly GI, showed a germinal-center CD10– profile more frequently than nodal lymphomas. Nevertheless, the study of the gene profile for DLBCLs from different sites is still warranted.
In the present series, the groups of nodal and extranodal DLBCLs were not homogeneous from the clinical standpoint. Lymphomas arising in two specific sites (one nodal, Waldeyer's ring; one extranodal, GI) showed very favorable features at diagnosis (eg, early stage, absence of bone marrow infiltration, normal serum LDH, and low-risk IPI), whereas the DLBCLs arising in the remaining areas (lymph nodes or other extranodal sites) presented with poorer characteristics. It is of note that no GI patient, and only two of 42 Waldeyer's ring patients had bone marrow infiltration, and that 72% of Waldeyer's ring patients had a low-risk IPI (Table 1). In terms of response to therapy, risk of relapse and OS, Waldeyer's, and GI lymphomas also showed a notably better outcome than that of the other groups. Thus, it is not the nodal or extranodal presentation, but some specific sites that may be related to particular clinicobiologic features and to the outcome of the patients. On the other hand, with regard to the prognostic factors for OS, none of the primary sites showed an independent predictive weight when standard variables were included in multivariate analyses. In fact, IPI, bulky disease, and serum 2m were the most important factors for OS, irrespective of the primary site of the lymphoma (Figs 3A and B).
In conclusion, in the present series of a large number of DLBCL patients from a single institution, the primary site of the disease was associated with particular clinicopathologic features and with the outcome, although the latter largely depended on other prognostic variables such as IPI or 2m.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by grants FIS-IP03/0473, Spanish Ministry of Health; "Red de Centros de Investigación del Cáncer y de Grupos para el Estudio de Linfomas, Instituto Carlos III–FIS", and the Jose Carreras Foundation Against Leukemia (EM-P/04 and CR-P/04).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Harris NL: Mature B-cell neoplasms, in EJ Jaffe, NL Harris, H Stein et al (eds): WHO Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France, IARCPress, 2001, pp 121-126
Zucca E, Cavalli F: Extranodal lymphomas. Ann Oncol 11:S219-S222, 2000 (suppl 3)
Zucca E, Roggero E, Bertoni F, et al: Primary extranodal non-Hodgkin's lymphomas, part 1: Gastrointestinal, cutaneous and genitourinary lymphomas. Ann Oncol 8:727-737, 1997
Zucca E, Roggero E, Bertoni F: Primary extranodal non-Hodgkin's lymphomas, part 2: Head and neck, central nervous system and other less common sites. Ann Oncol 10:1023-1033, 1999
Groves FD, Linet MS, Travis LB, et al: Cancer surveillance series: Non-Hodgkin's lymphoma incidence by histologic subtype in the United States from 1978 through 1995. J Natl Cancer Inst 92:1240-1251, 2000
Freeman C, Berg JW, Cutler SJ: Occurrence and prognosis of extranodal lymphomas. Cancer 29:252-260, 1972
Rudders RA, Ross ME, DeLellis RA: Primary extranodal lymphoma: Response to treatment and factors influencing prognosis. Cancer 42:406-416, 1978
Otter R, Gerrits WBJ, Sandt MMVD, et al: Primary extranodal and nodal non-Hodgkin's lymphoma: A survey of a population-based registry. Eur J Cancer Clin Oncol 25:1203-1210, 1989
d'Amore F, Christensen BE, Brincker H, et al: Clinicopathological features and prognostic factors in extranodal non-Hodgkin lymphomas. Eur J Cancer 27:1201-1208, 1991
Newton R, Ferlay J, Beral V, et al: The epidemiology of non-Hodgkin's lymphoma: Comparison of nodal and extra-nodal sites. Int J Cancer 72:923-930, 1997
Krol ADG, leCessie S, Snijder S, et al: Primary extranodal non-Hodgkin's lymphoma (NHL): The impact of alternative definitions tested in the Comprehensive Cancer Centre West population-based NHL registry. Ann Oncol 14:131-139, 2003
Moller MB, Pedersen NT, Christensen BE: Diffuse large B-cell lymphoma: Clinical implications of extranodal versus nodal presentation—A population-based study of 1575 cases. Br J Haematol 124:151-159, 2004
A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin's lymphoma: The Non-Hodgkin's Lymphoma Classification Project. Blood 89:3909-3918, 1997
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
Gatter KC, Warnke RA: Diffuse large B-cell lymphoma, in Jaffe ES, Harris NL, Stein H (eds): WHO Classification of Tumors: Pathology and Genetics of Tumors of Haematopoietic and Lymphoid Tissues. Lyon, France, IARCPress, 2001, pp 171-174
Pileri SA, Dirnhofer S, Went Ph, et al: Diffuse large B-cell lymphoma: One or more entities Present controversies and possible tools for its subclassification. Histopathology 41:482-509, 2002
Alizadeh AA, Elsen MB, Davis RE, et al: Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 403:503-511, 2000
Shipp MA, Ross KN, Tamayo P, et al: Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nature Med 8:68-74, 2002
Rosenwald A, Wright G, Chan WC, et al: The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 346:1937-1947, 2002
Kramer MH, Hermans J, Wijburg E, et al: Clinical relevance of BCL2, BCL6, and MYC rearrangements in diffuse large B-cell lymphoma. Blood 92:3152-3162, 1998
Muramatsu M, Akasaka T, Kadowaki N, et al: Rearrangement of the BCL6 gene in B-cell lymphoid neoplasms: Comparison with lymphomas associated with BCL2 rearrangement. Br J Haematol 93:911-920, 1996
Houldsworth J, Mathews S, Rao PH, et al: REL proto-oncogene is frequently amplified in extranodal diffuse large cell lymphoma. Blood 87:25-29, 1996
Rao PH, Houldsworth J, Dyomina K, et al: Chromosomal and gene amplification in diffuse large B-cell lymphoma. Blood 92:234-240, 1998
Gronbaek K, Straten PT, Ralfkiaer E, et al: Somatic Fas mutations in non-Hodgkin's lymphoma: Association with extranodal disease and autoimmunity. Blood 92:3018-3024, 1998
Wohlfart S, Sebinger D, Gruber P, et al: FAS (CD95) mutations are rare in gastric MALT lymphoma but occur more frequently in primary gastric diffuse large B-cell lymphoma. Am J Pathol 164:1081-1089, 2004
Harris NL, Jaffe ES, Diebold J, et al: The World Health Organization classification of neoplasms of the hematopoietic and lymphoid tissues: Report of the Clinical Advisory Committee meeting -Airlie House, Virginia, November 1997. Hematol J 1:53-66, 2000
The International Non-Hodgkin's Lymphoma Prognostic Factors Project: A predictive model for aggressive non-Hodgkin's lymphoma. N Engl J Med 329:987-994, 1993
Colomo L, López-Guillermo A, Perales M, et al: Clinical impact of the differentiation profile assessed by immunophenotyping in patients with diffuse large B-cell lymphoma. Blood 101:78-84, 2003
Kaplan GL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Peto R, Pike MC: Conservatism of the approximation (O-E)2/E in the log-rank test for survival data or tumour incidence data. Biometrics 29:759-784, 1973
Cox DR: Regression models and life tables. J R Stat Assoc 34:187-220, 1972
Dawson IP, Cornes JS, Morson BC: Primary malignant lymphoid tumours of the intestinal tract: Report of 37 cases with a study of factors influencing the prognosis. Br J Surg 49:80-89, 1961
Gospodarowicz MK, Sutcliffe SB: The extranodal lymphomas. Semin Radiat Oncol 5:281-300, 1995
Lossos IS, Czerwinski DK, Alizadeh AA, et al: Prediction of survival in diffuse large-B-cell lymphoma based on the expression of six genes. N Engl J Med 350:1828-1837, 2004
Sáez AI, Sáez AJ, Artiga MJ, et al: Building an outcome predictor model for diffuse large B-cell lymphoma. Am J Pathol 164:613-622, 2004
Gascoyne RD, Adomat SA, Krajewski S, et al: Prognostic significance of bcl-2 protein expression and bcl-2 gene rearrangement in diffuse aggressive non-Hodgkin's lymphoma. Blood 90:244-251, 1997
Beà S, Colomo L, López-Guillermo A, et al: Clinicopathologic significance and prognostic value of chromosomal imbalances in diffuse large B-cell lymphomas. J Clin Oncol 22:3498-3506, 2004
Houldsworth J, Olshen AB, Cattorelli G, et al: Relationship between REL amplification, REL function, and clinical and biologic features in diffuse large B-cell lymphomas. Blood 103:1862-1868, 2004(Armando López-Guillermo, )