• autologous stem cell transplantation;
  • chemotherapy;
  • diffuse large B-cell lymphoma;
  • immunohistochemistry;
  • immunotherapy;
  • IPI score


  1. Top of page
  2. Abstract
  6. Acknowledgements


The development of gene expression profiling and tissue microarray techniques have provided more information about the heterogeneity of diffuse large B-cell lymphoma (DLBCL), enabling categorization of DLBCL patients into 3 prognostic groups according to cell origin (but independently from the International Prognostic Index [IPI] score): germinal center (GCB), activated B-cell (ABC), and not classified (NC) diffuse large B-cell lymphoma. This study investigated the role of immunohistochemical discrimination between GCB and ABC&NC-DLBCL subtypes in identifying those high-risk patients who may benefit from a more aggressive first-line therapeutic approach.


From February 2003 to August 2006, 45 newly diagnosed DLBCL patients, with IPI≥2, were considered eligible for this study: 13 had a GCB, 8 an ABC, and 24 a NC-DLBCL. GCB patients received 6 courses of rituximab, cyclophophosphamide, doxorubicin, vinicristine, and prednisone (R-CHOP) chemotherapy, with a subsequent, autologous stem cell transplantation in case of partial response. All ABC and NC-DLBCL patients received 6 R-CHOP cycles and autologous stem cell transplantation.


Complete response rate for each treatment arm was 84.6% for GCB and 89.7% for ABC&NC-DLBCL (P = .50), with a continuous complete response rate of 81.8% and 84.6%, respectively (P = .59). Projected 4-year overall survival is 100% for GCB and 82% for ABC&NC patients (P = .12). Progression-free survival is 77% and 79% (P = .7), respectively.


The autologous stem cell transplantation consolidation in the ABC&NC-DLBCL subtypes induced the same rate of complete response (and similar progression-free survival rate) compared with GCB-DLBCL. In ABC&NC-DLBCL patients the authors observed a complete response rate of 89.7% vs. 84.6% in the GCB-DLBCL subset, without any significant difference in progression-free survival rate. Cancer 2010. © 2010 American Cancer Society.

Diffuse large B-cell lymphoma (DLBCL) is the commonest form of non-Hodgkin lymphoma (NHL), accounting for approximately 30% of new lymphoma diagnoses in adult patients.1 Durable remissions can be achieved in 40% to 50% of patients after anthracycline-based first-line chemotherapy, although DLBCL has been proven to be one of the most chemotherapy-responsive malignant neoplasms. This observation suggests DLBCL is a heterogeneous entity, affecting morphology, clinical behavior, and outcome.2

Currently, the International Prognostic Index (IPI) is the most widely acknowledged strategy to stratify DLBCLs. This process takes into account 5 clinical variables: advanced age, poor performance status (Eastern Cooperative Oncology Group [ECOG]≥2), elevated serum lactate-dehydrogenase (LDH) levels, advanced stage lymphoma (III or IV, according to Ann Arbor criteria), and involvement of more than 2 extranodal sites.3 These parameters differentiate patients into 4 risk groups—low (IPI 0), low-intermediate (IPI 1), high-intermediate (IPI 2-3), and high (IPI 4-5)—each with a different impact on prognosis. This prognostic index yields treatments that are calibrated more precisely to the individual patient. However, the IPI alone inadequately recognizes the biologic and genetic differences that actually make DLBCL a heterogeneous disease.4

Gene expression profiling techniques have provided a powerful tool to effectively explore such biological variability. In 2000, Alizadeh et al first identified 2 distinct types of DLBCL using gene expression profiling on the basis of tumor proliferation rate, differentiation state of the tumor B-cells, and host response.5 In particular, the former expressed a gene pattern close to that of normal germinal center B elements (so-called GCB-DLBCL), whereas the latter had a gene expression profiling comparable to in vitro activated peripheral blood B cells (accordingly termed ABC-DLBCL). Such a distinction also turned out to be important on prognostic grounds: patients with GCB-DLBCL had a significantly better overall survival than those with ABC-DLBCL after first-line therapy with comparable multiagent regimens. This observation suggested that the 2 forms of DLBCL should be considered distinct clinical entities, expressing disparate behaviors among patients in the same IPI-risk category.5 Rosenwald et al subsequently recognized another cluster of DLBCL that was termed “type 3” or “not classified” (NC). This subgroup is quite heterogeneous and ill-defined at the molecular level.6 Like the ABC group, the NC subgroup has a poor outcome. The 2 categories are often collectively referred to as “non-GCB.”

The application of gene expression profiling techniques is not easy because of the need for fresh or frozen material and the costs for gene chips and related instruments. Therefore, attempts have been made to develop tools that may achieve the same results.

In 2004, by adopting the tissue microarray (TMA) technology, Hans et al4 claimed that the immunohistochemical determination of CD10, Bcl-6, and MUM1 (now termed IRF4) could replace the GCB and “nonGCB”-DLBCL classification originally proposed by Alizadeh et al. Specifically, CD10+, Bcl-6+, MUM1/IRF4each corresponded to GCB-DLBCLs, whereas all the other marker combinations (CD10, Bcl-6+, MUM1/IRF4+, and CD10, Bcl-6, MUM1/IRF4+) indicated “non-GCB” phenotypes. One year later, suspecting the original algorithm might have been too simplistic, a much more complex classification system was published by the same authors.7

In 2005, we used TMA to analyze the phenotypic profile of 68 homogeneously treated de novo DLBCL patients with nodal presentation.8 The markers used were CD10, Bcl-6, IRF4, CD30, CD138, IRTA-1, and Bcl-2. The cases were subdivided into three groups termed GCB, ABC, and NC (Table 1), in accordance with nomenclature proposed by Rosenwald et al6. The “GCB” group's profile was consistent with B-cells entering into the formation of or on the way to leaving the germinal center. The “ABC” group was thought to consist of nongerminal center B-cells showing features of activation (CD30+) or plasmablastic/plasmacellular differentiation (CD138/IRF4+). Finally, the “NC” subtype included a spectrum of cases either of putative germinal center or postgerminal center B-cell derivation—all possessing Bcl-2 positivity, a characteristic previously described as a poor prognostic indicator.9 The complete remission (CR), relapse-free survival (RFS), and overall survival (OS) rates were 89%, 86%, 91%; 53%, 41%, 38%; and 73%, 63%, 66%, respectively, for the 3 groups. Interestingly, our approach turned out to be much more predictive than the Hans algorithm.

Table 1. Immunohistochemical Patterns Adopted to Further Classify DLBCL (see text for explanation)
  • a

    Possible combinations: CD10+/Bcl-6+/IRF4; CD10/Bcl-6+/IRF4+; CD10/Bcl-6+/IRF4; CD10/Bcl-6/IRF4+.

GCB-DLBCLCD10+, or CD10+ and Bcl-6+, or CD10, Bcl-6+, MUM1CD10+, Bcl−6+, IRF4 or CD10±, Bcl-6+ and IRF4±, and always Bcl-2, CD30, IRTA-1, and CD138
Non-GCB/ ABC-DLBCLCD10 and Bcl-6, or CD10, Bcl-6+, MUM1+CD10, Bcl-6, Bcl2+ and CD138+/IRF4+/IRTA1 or CD30+/IRTA−1/IRF4 or CD30±/IRTA−1+/IRF4
NC-like DLBCL Variable CD10, Bcl−6, IRF4 staininga and always Bcl−2+ and CD30, IRTA1, and CD138

Herein, we report a phase 2 multicenter study in which patients with de novo DLBCL and intermediate to high IPI score (≥2) were treated with chemoimmunotherapy with or without autologous stem cell transplantation according to their immunophenotype (assessed according to the criteria above). The purpose of the study was to evaluate the role of immunohistochemistry in identifying those high-risk patients who may benefit from a more aggressive, first-line, therapeutic approach.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Design of the Study

All the patients with a histology-proven diagnosis of DLBCL and an IPI value equal to or higher than 2 were further subdivided into GCB, ABC, and NC-DLBCL, according to previously published criteria.8 The ABC and NC groups were combined in light of the survival figures noted above. (The ABC&NC phenotype was deemed unfavorable as a prognostic factor, apart from IPI.) Thus, the GCB-DLBCL patients were to receive 6 courses of rituximab, cyclophophosphamide, doxorubicin, vinicristine, and prednisone (R-CHOP) chemotherapy, with restaging at the end and subsequent follow-up if in CR; or high-dose chemotherapy with autologous stem cell transplantation if in partial response (PR). ABC-DLBCL and NC-DLBCL patients were to receive a first-line of 6 R-CHOP cycles, followed by stem cell mobilization and harvest, after 1 or 2 cycles of chemotherapy according to ifosfamide, epirubicin, and etoposide (IEV) regimen.10 These patients would then receive high-dose chemotherapy with autologous stem cell transplantation, independently from the response obtained after the first line (Fig. 1, 2, 3).

thumbnail image

Figure 1. Treatment algorithm is presented according to DLBCL immunohistochemical categories.

Download figure to PowerPoint

thumbnail image

Figure 2. Overall survival of the entire cohort of patients is plotted according to GCB- and ABC&NC-DLBCL immunohistochemical profile.

Download figure to PowerPoint

thumbnail image

Figure 3. Progression-free survival is presented according to immunohistochemical profile.

Download figure to PowerPoint

All the patients received complete tumor staging at baseline via positron-emission tomography (PET) and computed tomography (CT) scans. Complete blood sampling, including LDH evaluation, and bone marrow biopsy were performed for each patient before chemotherapy. Performance status and vital parameters were registered at each visit. A physical examination, along with complete blood sampling, were performed before each course of chemotherapy.

Re-evaluation was performed after R-CHOP administration and, when applicable, after IEV chemotherapy and autologous stem cell transplantation, with PET and CT scans; bone marrow biopsy was also repeated in patients who had a lymphomatous infiltrate at diagnosis.

R-CHOP chemotherapy was administered intravenously by using standard dosages every 21 days; rituximab was given over a 4-hour infusion on Day 1 of each cycle. Midcycle treatment with filgrastim (G-CSF) or prophylactic antibiotic therapy was permitted when neutrophil count was lower than 1.5 × 109/L, or when the patient developed febrile neutropenia.

IEV chemotherapy was administered (3-day intravenous infusion) 30 to 40 days after the conclusion of the R-CHOP therapy in ABC and NC-DLBCL patients. Standard doses of ifosfamide, epirubicin, and etoposide were used (2500 mg/m2, 100 mg/m2, and 150 mg/m2, respectively), with intravenous mesna rescue to reduce ifosfamide-related urothelial toxicity. Filgrastim was administered for at least 10 days after IEV therapy concluded to support stem cell mobilization in peripheral blood. CD34+ peripheral blood stem cells were evaluated with a cytofluorimetric assay after 10 days of G-CSF subcutaneous therapy.

Thirty to 40 days after stem cell mobilization, the patient was admitted to hospital for the autologous stem cell transplantation procedure. bis-chloronitrosourea, or carmustine [BCNU], etoposide, cytarabine, and melphalan (BEAM) conditioning regimen (BCNU, 300 mg/m2 on Day 1; cytarabine, 400 mg/m2 and etoposide, 200 mg/m2 from Day 2 to Day 5; melphalan, 140 mg/m2 on Day 6) were administered intravenously for 6 consecutive days, followed by stem cell infusion on Day 7. Blood transfusion, granulocyte colony-stimulating factor (G-CSF), and antibiotic therapy were allowed during cytopenia.

All patients were notified of the investigational nature of this study and provided written, informed consent in accordance with institutional guidelines, including the Declaration of Helsinki.

Patient Characteristics

From February 2003 to August 2006, 45 patients were considered eligible for study enrollment at 10 major Italian institutions: Bologna, Siena, Napoli, Cagliari, Nocera Inferiore, Roma La Sapienza University, Roma Campus Bio-medico, Roma Sant'Eugenio, Roma Tor Vergata, and Roma Regina Elena. All patients had a diagnosis of DLBCL and were subsequently classified into GCB-DLBCL (n = 13), ABC-DLBCL (n = 8), and NC-DLBCL (n = 24). Twenty-seven patients were male, and 18 were female, with a median age at diagnosis of 48 years (range, 20-62 years of age). Complete patient demographic and clinical data are listed in Table 2.

Table 2. Patients' Characteristics According to DLBCL Subtypes
  1. GCB-DLBCL indicates germinal center diffuse large B-cell lymphoma; ABC-DLBCL, activated B-cell diffuse large B-cell lymphoma; NC-DLBCL, not classified diffuse large B-cell lymphoma; LDH, lactate dehydrogenase; B-symptoms, fever, weight loss, night sweats, and pruritus sine materia.

No. of patients13824
Median age50 (24-62)48 (20-59)47 (24-60)
 II3 (23.1%)2 (25.0%)3 (12.5%)
 III2 (15.4%)3 (37.5%)5 (20.8%)
 IV8 (61.5%)3 (37.5%)16 (66.7%)
LDH elevation10 (76.9%)7 (87.5%)18 (75.0%)
Performance status
 02 (15.4%)2 (25.0%)4 (16.7%)
 101 (12.5%)0
 210 (76.9%)5 (62.5%)19 (79.1%)
 31 (7.7%)01 (4.2%)
 26 (46.1%)5 (62.5%)13 (54.2%)
 34 (30.8%)3 (37.5%)10 (41.7%)
 43 (23.1%)01 (4.1%)
B-symptoms2 (15.4%)3 (37.5%)12 (50%)
Bulky disease4 (30.8%)5 (62.5%)7 (29.2%)
Extranodal disease8 (61.5%)3 (37.5%)13 (54.2%)
 >2 sites301
BM involvement4 (30.8%)05 (20.8%)

Immunohistochemistry Evaluation

Sections obtained from each patient were tested with specific antibodies against CD10, CD20, CD30, CD79a, CD138, Bcl-2, Bcl-6, IRF4/MUM1, and IRTA1. Bound antibodies were then delineated by the alkaline phosphatase-antialkaline phosphatase complex technique. According to criteria used by Hans et al4, cases were considered positive when 30% or more of the tumor cells were stained with an antibody. Immunohistochemical classification of cases proceeded on the basis of protocol reported by Zinzani et al (see Table 1).8 Two expert hemolymphopathologists (FB and SAP) performed the analysis, and discordant results were discussed collectively at a multihead microscope.

Statistical Analysis

The primary endpoint was efficacy through the measurement of response. Sample size estimation was carried out by Fleming single-stage procedure.11, 12 Previous experience shows that response rate, adjusted for the response criteria as in par, has been 50%. Defining π0 as the proportion of response below the treatment that does not warrant further investigations, and πa as the proportion of responses warranting a phase 3trial, we set π0 = 0.6 and πa = 0.8. The number of patients required, given a type I error (α) at.05 two-tailed and a power of 1−β = 80%, is 45, and the number of successes (responses) is 33. If, at the end of the trial, at least 33 responses (successes) are observed, the treatment will be accepted for a phase 3 trial.13

Study Endpoints

The primary endpoint was efficacy through the measurement of response. Complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) were defined according to International Working Group recommendations.14, 15 Response rates are expressed in percentages with an intention-to-treat modality. Secondary endpoints were overall survival (OS) and progression-free survival (PFS). OS was calculated from trial entry to the time of latest visit or death. PFS was determined as the time from registration to the first observation of progressive disease or death from any cause. OS and PFS curves were plotted by the Kaplan-Meier method,16 and then compared by using the log-rank test.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Forty-one patients of the 45 enrolled in the study were able to complete the entire protocol of treatment, as discussed before. Two patients, both belonging to the NC-DLBCL subgroup, died during treatment because of rapid disease expansion and progression: 1 of them died before induction therapy with R-CHOP was concluded, and the other died immediately after R-CHOP, before mobilization and harvesting of peripheral blood stem cells. Both patients presented with stage IV disease (the former because of stomach involvement, the latter with bone marrow infiltration) and an IPI score of 3. Two patients, 1 with a NC-DLBCL and 1 with an ABC-DLBCL, both in CR after R-CHOP, refused to proceed with autologous stem cell transplantation and withdrew their consent.

Response to Treatment According to Subgroups

All 13 patients with a GCB-DLBCL diagnosis completed the course of treatment. Eleven (84.6%) showed a CR after 6 cycles of R-CHOP, and were considered ineligible for further treatment. Two (15.4%) patients had a PET- and CT-documented PR; each underwent stem cell mobilization, chemotherapy, and subsequent autologous stem cell transplantation. One (50%) patient converted his PR to a CR after high-dose chemotherapy (PR was still observed after IEV therapy, before high-dose consolidation). The other patient (50%) showed disease progression after IEV chemotherapy and at the end of the entire course of treatment.

Among the 8 patients with ABC-DLBCL, 6 (75%) achieved a CR after R-CHOP, and 2 (25%) were in PR. High dose therapy with autologous stem cell transplantation consolidated the previously observed complete responses and converted both PRs into CRs.

Among the 24 patients with NC-DLBCL, 23 were evaluable for interim response to treatment after R-CHOP, and 1 died before early PET or CT evaluation, as previously explained. Seventeen (70.8%) patients obtained a CR, 5 (20.8%) were in PR, and 1 (4.2%) showed disease progression. Twenty-one patients were administered high-dose chemotherapy, with 18 (75%) CRs, 1 (4.2%) PR, and 2 (8.4%) PDs at the end of treatment. In particular, 14 of the 17 patients in CR after R-CHOP still maintained a CR. One (5.9%) showed disease progression; 1 died before transplantation; and 1 withdrew consent (see above for details). Four of the 5 patients in PR converted to a CR, and 1 showed a PD. The only patient with a PD after R-CHOP was able to improve his status to a PR.

When we consider the ABC&NC patients as a whole, 26 of the 29 (89.7%) who completed the treatment schedule achieved a complete response. No statistically significant differences were seen between the GCB-DLBCL and the ABC&NC-DLBCL subgroups in terms of CR (P = .50), as calculated with Fisher exact test (see Table 3).

Table 3. Patients' Response and Survival Details According to GCB- and ABC&NC-Immunohistochemical Profile
  • a

    CCR indicates continuous complete response patients.

CR11/13 (84.6%)26/29 (89.7%).50
CCRa9/11 (81.8%)22/26 (84.6%).59
4-Year OS100%82%.12
4-Year PFS77%79%.70

Treatment responses are summarized in Figure 1.

Follow-Up Analysis

The median follow-up for the entire patient cohort was 32 months (range, 6-59 months). Median PFS was29 months (range, 6-59 months).

All the patients (13of 13) with GCB-DLBCL are alive, 11 in CR and 2 with PD. Median overall survival was 30 months (range, 25-49 months). The CR rate was 92.3% at the end of treatment, compared with 84.6% before autologous stem cell transplantation. High-dose therapy was, thus, able to enhance the CR rate by 7.7%. Two (15.4%) patients in CR after R-CHOP experienced relapses after 9 and 12 months, respectively. Both were administered further therapy, the former undergoing an autologous stem cell transplantation but maintaining a PD status, the latter achieving a second CR. Both patients are alive, with an OS of 36 and 27 months, respectively. PFS for this group of patients is 27.1 months (range, 8.7-48.7 months).

All the patients with ABC-DLBCL are alive, with a median OS of 31 months (range, 17-48 months). Six patients are in CR and 2 in PD. The CR rate was 100% after the entire treatment protocol. Two (25%) patients relapsed after 13 and 23 months. No further therapy was performed, and these individuals were still alive after 18 and 28 months, respectively. Median PFS was 31 months (range, 13-48 months).

At the end of this study, 19 of 24 (79.2%) patients with NC-DLBCL were alive. Eighteen were in CR after therapy; thus, the CR rate was 75%. High-dose therapy was able to improve the CR rate of 4.2%. At the latest contact, 16 patients were in CR, 1 in PR, and 2 with PD, with a PFS rate of 66.7%. Two patients died prematurely, before the conclusion of the treatment schedule. Two more patients died because disease progression occurred after the completion of treatment: 1 of them was in CR, the other was in PR (see next paragraph for details). One patient in CR committed suicide. The OS rate was 79.2%; median OS was 32 months (range, 6-59 months), with a median PFS of 29 months (range, 6-59 months).

Considering all the patients with an ABC&NC histology, 27 of 32 (84.4%) were alive as of this writing, with 22 in CR, 1 in PR, and 4 with PD. The CR rate after the scheduled treatment was 81.3%; the median OS was 32 months (range, 6-59 months), and the median PFS was 30 months (range, 6-59 months).

Analysis of Response Failure

The only GCB-DLBCL patient who showed a PD at the end of the entire course of treatment presented with stage IV B, IPI 4 disease, with muscle and pleural involvement, mediastinal bulk, high LDH levels at presentation, and a severely impaired performance status (3, according to ECOG). This patient had a PR after R-CHOP, and a PD at the end of treatment. After a follow-up of 27 months, the patient was still alive, with a maintained PD.

The patient with NC-DLBCL who had a PD after high-dose therapy, but was in CR after R-CHOP, presented with stage II A, IPI 2 disease, with a performance status of 2 and a LDH elevation. After obtaining a CR after IEV chemotherapy, the patient's disease rapidly progressed, and this patient died 12 months after diagnosis. The other patient with NC-DLBCL, whose disease progressed at the end of treatment, had a PR after R-CHOP and stage IV A, IPI 2 disease, with bone marrow involvement and LDH elevation. After autologous stem cell transplantation, the patient underwent allogeneic bone marrow transplantation; achieving a CR; she was still alive at 31.5 months after diagnosis.

The patient in CR after transplantation, who died because of progressive disease, presented with stage IV A, IPI 2 disease, with adrenal involvement and a performance status of 2. Progression occurred after nearly 2 years of complete response, with an OS of 26 months after diagnosis.

Details for these patients are summarized in Table 4.

Table 4. Unfavorable Outcome Patients Analysis
  1. GCB indicates germinal center diffuse large B-cell lymphoma; NC, not classified diffuse large B-cell lymphoma; LDH, lactate dehydrogenase; PS, performance status; IPI, International Prognostic Index; B-symptoms, fever, weight loss, night sweats, and pruritus sine materia; BM, bone marrow; PR, partial response; CR, complete response; PD, progressive disease; ABMT, autologous bone marrow transplantation.

Sex, ageWoman, 24Male, 56Woman, 55Male, 55
LDH elevationYesYesYesNo
Extranodal sitesMuscle, pleuraNoNoAdrenal
Bulky diseaseMediastinumNoNoNo
BM involvementNoNoYesNo
Early responsePRCRPRCR
StatusAlive, PDDeadAlive, CRDead
Overall survival27.512.331.525.8


  1. Top of page
  2. Abstract
  6. Acknowledgements

Early phase 2 and 3 trials in the 1990s demonstrated the efficacy of the CHOP regimen in curing approximately one-third of DLBCL patients, with no statistically significant differences in survival with second- and third-generation CHOP-like regimens (eg, methotrexate, bleomycine, doxorubicin, cyclophosphamide, vincristine, dexamethasone, with leucovorin rescue [m-BACOD], prednisone, doxorubicin, cyclophosphamide, etoposide, cytarabine, bleomycin, vincristine, methotrexate, with leucovorin rescue [ProMACE-CytaBOM], and methotrexate, doxorubicin, cyclophosphamide, vincristine, bleomycin, prednisone, with leucovorin rescue [MACOP-B]).17-19 More recently, the addition of rituximab to CHOP (R-CHOP) has shown a significant improvement in high-grade lymphoma patients' survival; in elderly20-21 as well as in young, low-risk patients treated with R-CHOP-like regimens.22

High-dose therapy, followed by autologous stem cell transplantation, efficiently cures nearly half the patients with a resistant or relapsed, aggressive non-Hodgkin lymphoma. However, trials using up-front, high-dose therapy and autologous transplantation after an initial course of induction therapy have not shown solid and reliable results.23 Some studies have shown little or no benefit from up-front intensification of doses.24-27

Increased interest in the molecular pathogenesis of DLBCL and, in particular, the ability to discriminate diverse subgroups of this disease through gene expression profiling and TMA technique, has shifted attention to early assessment of prognostic biomarkers. Alizadeh et al were the first to demonstrate the existence of distinct types of DLBCL by gene expression profiling, indicating a different treatment outcome for each of them.5 Subsequent studies attempted to discover specific genetic alterations or aberrant protein expression, in the context of such a heterogeneous disease, to create prognostic models and, more importantly, to reveal new therapeutic implications and targets. This issue appeared to be urgent, as significant heterogeneity could be observed even within each IPI score category.

Lossos et al28 constructed a model on the basis of 6 key genes, each evaluated on a multivariate analysis. In a homogeneously treated cohort of DLBCL patients, they demonstrated that the expression of genes correlated with a “germinal-center” or “lymph-node” signature (LMO2, BCL6, FN1) may be predictive of a better survival, whereas the expression of genes correlated with an “activated B-cell” signature (BCL2, CCND2, SCYA3) was indicative of a shorter survival. Recently, Lenz et al9 demonstrated how DLBCL patients' survival may be influenced by the characteristics of the tumor microenvironment, consisting of immune cells, fibroblasts, and newly formed blood vessels. They identified “stromal-1” and “stromal-2” signatures on the basis of gene expression profiling; the former consisting of genes normally expressed in mesenchymal tissues, the latter associated with increased tumor blood vessel formation. Besides the original distinction between GCB and ABC-DLBCL, the “stromal-1” signature was associated with increased survival, and the “stromal-2” with reduced survival.

The prognostic significance of the molecular distinction of DLBCL in the 2 major molecular subtypes, GCB and non-GCB, was tested in some clinical trials, whereby patients were required to undergo a risk-adapted therapy. Van Imhoff et al demonstrated that the GCB signature, contextually with the expression of BCL2, favorably impacted survival of high-risk (IPI≥2) DLBCL patients treated with high-dose and autologous stem cell transplantation as first-line therapy.29 Nyman et al showed that GCB-DLBCL patients, independent of their IPI class risk, displayed a significantly higher overall and failure free survival than the non-GCB patients when treated with standard chemotherapy. However, survival rates appeared more variable for those patients treated with R-CHOP or CHOP-like regimens.30 This indicates that the GCB phenotype does not predict clinical outcome in immunochemotherapy-treated DLBCL patients and that rituximab seems to eliminate the prognostic value of immunohistochemistry (but not of IPI risk stratification). Veelken et al described the outcome of 60 DLBCL patients treated in the era before rituximab and indicated that the immunohistochemical subcategorization adds very little clinical value. Moreover, they stated that intensified, up-front therapy could mitigate the impact of IPI-associated risk.31

On the basis of previous experience and risk-group definition, taking into account possible histogenetic derivation as well as BCL2 expression,8 we evaluated the role of immunohistochemical discrimination between GCB and ABC&NC-DLBCL subtypes in identifying those high-risk patients who may benefit from a more aggressive first-line therapeutic approach. We also analyzed the clinical outcome of a risk-adapted therapy consisting of R-CHOP induction for GCB-DLBCL and up-front consolidation with high-dose therapy and stem cell transplantation for ABC and NC-DLBCL. Globally, the autologous stem cell transplantation consolidation in the ABC&NC-DLBCL subtypes induces the same rate of CR and a similar PFS rate compared with GCB-DLBCL data. Concerning the ABC&NC-DLBCL patients, we observed a CR rate of 89.7% versus 84.6% for the GCB-DLBCL subset, without any significant difference in PFS curves (79% vs 77%, respectively, with a median follow-up of 32 months).

Conversely, Dunleavy et al32 recently reported that bortezomib enhances the activity of chemotherapy (dose-adjusted-EPOCH, containing etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin [DA-EPOCH]) in ABC but not GCB-DLBCL subtypes. These preliminary results can open the door to incorporating new targeted drugs in conventional chemotherapeutic regimens with the aim of increasing response in the different molecular subtypes of DLBCL patients. In fact, Nowakowski et al33 recently reported the preliminary results of a phase 1/2 trial, whereby they evaluated potential improvement in therapy results by the addition of lenalidomide to the conventional R-CHOP therapy in untreated DLBCL and grade 3 follicular lymphoma patients. Moreover, an alternative therapy could include the oral mTOR inhibitor everolimus (RAD001) as maintenance therapy. An international trial to investigate single-agent activity of everolimus in relapsed/refractory DLBCL has recently begun.34


  1. Top of page
  2. Abstract
  6. Acknowledgements

The authors thank Maurizio Martelli (Department of Cellular Biotechnology and Hematology, La Sapienza University, Rome, Italy), Giuseppe Avvisati (Division of Hematology, Campus Bio-medico University, Rome, Italy), Maria Giuseppina Cabras (Division of Hematology, Cagliari Hospital, Cagliari, Italy), Maria Concetta Petti (Department of Hematology, Regina Elena Hospital, Rome, Italy), Alberto Fabbri (Department of Internal Medicine and Hematology, Le Scotte Hospital, Siena, Italy), and Brunangelo Falini (Institute of Hematology, Policlinico Monteluce Silvestrini, University of Perugia, Perugia, Italy) for their help in patients' enrollment and collection of data. Special thanks to Enrico Derenzini, Cinzia Pellegrini, Letizia Gandolfi, Lisa Argnani, and Pier Paolo Piccaluga (Institute of Hematology and Medical Oncology L. e A. Seràgnoli, Policlinico Sant'Orsola-Malpighi, University of Bologna, Bologna, Italy) for data analysis and results interpretation.


  1. Top of page
  2. Abstract
  6. Acknowledgements

This work was partly supported by BolognAIL (Bologna, Italy) and the Italian Association for Cancer Research (A.I.R.C., Milan, Italy).


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    Swerdlow S, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissue Systems. 4th ed. Lyon, France: IARC; 2008.
  • 2
    Abramson JS, Shipp MA. Advances in the biology and therapy of diffuse large B-cell lymphoma: moving toward a molecularly targeted approach. Blood. 2005; 106: 1164-1174.
  • 3
    The International Non-Hodgkin's Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin's lymphoma. N Engl J Med. 1993; 329: 987-994.
  • 4
    Hans CP, Weisenburger DD, Greiner TC, et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood. 2004; 103: 275-282.
  • 5
    Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000; 403: 503-511.
  • 6
    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. 2002; 346: 1937-1947.
  • 7
    Hans CP, Weisenburger DD, Greiner TC, et al. Expression of PKC-beta or cyclin D2 predicts for inferior survival in diffuse large B-cell lymphoma. Mod Pathol. 2005; 18: 1377-1384.
  • 8
    Zinzani PL, Dirnhofer S, Sabattini E, et al. Identification of outcome predictors in diffuse large B-cell lymphoma. Immunohistochemical profiling of homogeneously treated de novo tumors with nodal presentation on tissue micro-arrays. Haematologica. 2005; 90: 341-347.
  • 9
    Lenz G, Wright G, Dave SS, et al. Stromal gene signatures in large-B-cell lymphomas. N Engl J Med. 2008; 359: 2313-2323.
  • 10
    Zinzani PL, Barbieri E, Visani G, et al. Ifosfamide, epirubicin, and etoposide (IEV) therapy in relapsed and refractory high-grade non-Hodgkin's lymphoma and Hodgkin's disease. Haematologica. 1994; 79: 508-512.
  • 11
    Fleming TR. One-sample multiple testing procedure for phase 2 clinical trials. Biometrics. 1982; 38: 143-151.
  • 12
    Machin D, Campbell MJ. Statistical Tables for the Design of Clinical Trials. Oxford, UK: Blackwell; 1987.
  • 13
    A'Hern RP. Sample size tables for exact single-stage phase 2 designs. Stat Med. 2001; 20: 859-866.
  • 14
    Cheson BD, Horning SJ, Coiffier B, et al. Report of an International Workshop to Standardize Response Criteria for non-Hodgkin's lymphomas. J Clin Oncol. 1999; 17: 1244-1253.
  • 15
    Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol. 2007; 25: 579-586.
  • 16
    Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc. 1958; 53: 457-481.
  • 17
    Gordon LI, Harrington D, Andersen J, et al. Comparison of a second-generation combination chemotherapeutic regimen (m-BACOD) with a standard regimen (CHOP) for advanced diffuse non-Hodgkin's lymphoma. N Engl J Med. 1992; 327: 1342-1349.
  • 18
    Cooper IA, Wolf MM, Robertson TI, et al. Randomized comparison of MACOP-B with CHOP in patients with intermediate-grade non-Hodgkin's lymphoma. The Australian and New Zealand Lymphoma Group. J Clin Oncol. 1994; 12: 769-778.
  • 19
    Fisher RI, Gaynor ER, Dahlberg S, et al. A phase III comparison of CHOP vs. m-BACOD vs. ProMACE-CytaBOM vs. MACOP-B in patients with intermediate- or high-grade non-Hodgkin's lymphoma: results of SWOG-8516 (Intergroup 0067), the National High-Priority Lymphoma Study. Ann Oncol. 1994; 5 (suppl): S91-S95.
  • 20
    Coiffier B, Lepage E, Briere J, et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large B-cell lymphoma. N Engl J Med. 2002; 346: 235-242.
  • 21
    Habermann TM, Weller EA, Morrison VA, et al. Phase III trial of rituximab-CHOP (R-CHOP) vs. CHOP with a second randomization to maintenance rituximab (MR) or observation in patients 60 years of age and older with diffuse large B-cell lymphoma (DLBCL). Blood. 2003; 102: 6.
  • 22
    Pfreundschuh M, Trumper L, Gill D, et al. First analysis of the completed Mabthera International (MInT) trial in young patients with low-risk diffuse large B-cell lymphoma (DLBCL): addition of rituximab to a CHOP-like regimen significantly improves outcome of all patients with the identification of a very favourable subgroup with IPI = 0 and no bulky disease. Blood. 2004; 104: 48.
  • 23
    Haioun C, Lepage E, Gisselbrecht C, et al. Survival benefit of high-dose therapy in poor-risk aggressive non-Hodgkin's lymphoma: final analysis of the prospective LNH87-2 protocol —a Groupe d'Etude des Lymphomes de l'Adulte study. J Clin Oncol. 2000; 18: 3025-3030.
  • 24
    Gisselbrecht G, Lepage E, Molina T, et al. Shortened first-line high-dose chemotherapy for patients with poor-prognosis aggressive lymphoma. J Clin Oncol. 2002; 20: 2472-2479.
  • 25
    Martelli M, Gherlinzoni F, De Renzo A, et al. Early autologous stem-cell transplantation versus conventional chemotherapy as front-line therapy in high-risk, aggressive non-Hodgkin's lymphoma: an Italian multicenter randomized trial. J Clin Oncol. 2003; 21: 1255-1262.
  • 26
    Kaiser U, Uebelacker I, Abel U, et al. Randomized study to evaluate the use of high-dose therapy as part of primary treatment for “aggressive” lymphoma. J Clin Oncol. 2002; 20: 4413-4419.
  • 27
    Verdonck LF, van Putten WL, Hagenbeek A, et al. Comparison of CHOP chemotherapy with autologous bone marrow transplantation for slowly responding patients with aggressive non-Hodgkin's lymphoma. N Engl J Med. 1995; 332: 1045-1051.
  • 28
    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. 2004; 350: 1828-1837.
  • 29
    van Imhoff GW, Boerma E-JG, van der Holt B, et al. Prognostic impact of germinal center-associated proteins and chromosomal breakpoints in poor-risk diffuse large B-cell lymphoma. J Clin Oncol. 2006; 24: 4135-4142.
  • 30
    Nyman H, Adde M, Karjalainen-Lindsberg M-L, et al. Prognostic impact of immunohistochemically defined germinal center phenotype in diffuse large B-cell lymphoma patients treated with immunochemotherapy. Blood. 2007; 109: 4930-4935.
  • 31
    Veelken H, Vik Dannheim S, Schulte Moenting J, Martens UM, Finke J, Schmitt-Graeff A. Immunophenotype as prognostic factor for diffuse large B-cell lymphoma in patients undergoing clinical risk-adapted therapy. Ann Oncol. 2007; 18: 931-939.
  • 32
    Dunleavy K, Pittaluga S, Czuczman MS, et al. Differential efficacy of bortezomib plus chemotherapy within molecular subtypes of diffuse large B-cell lymphoma. Blood. 2009; 113: 6069-6076.
  • 33
    Nowakowski GS, LaPlant B, Habermann T, et al. A phase I/II trial of lenalidomide and RCHOP (R2CHOP) in patients with newly diagnosed diffuse large B-cell (DLBCL) and follicular grade 3 lymphoma. Blood. 2009; 114: 666.
  • 34
    Reeder C, Gornet MK, Habermann TM, et al. A phase 2 trial of the oral mTOR inhibitor everolimus (RAD001) in relapsed aggressive non-Hodgkin lymphoma (NHL). Blood. 2007; 110.