Despite currently available treatments, approximately 60,000 new non-hodgkin lymphoma (NHL) cases and

Transcription

1 Higher Response to Lenalidomide in Relapsed/ Refractory Diffuse Large B-Cell Lymphoma in Nongerminal Center B-Cell Like Than in Germinal Center B-Cell Like Phenotype Francisco J. Hernandez-Ilizaliturri, MD 1 ; George Deeb, MD 2 ; Pier L. Zinzani, MD 3 ; Stefano A. Pileri, MD 3 ; Farhana Malik, MD 1 ; William R. Macon, MD 4 ; Andre Goy, MD 5 ; Thomas E. Witzig, MD 6 ; and Myron S. Czuczman, MD 1 BACKGROUND: There is a need to develop novel therapies for relapsed/refractory diffuse large B-cell lymphoma (DLBCL) and to identify biomarkers predictive for therapeutic response. Lenalidomide was previously shown to induce an overall response rate (ORR) of 28% in patients with relapsed/refractory DLBCL. It is currently unknown if response rates differ between patients with different DLBCL subtypes. METHODS: The authors retrospectively evaluated clinical outcomes of patients with germinal center B-cell like versus nongerminal center B-cell like DLBCL treated with salvage lenalidomide at 4 academic institutions. RESULTS: Forty patients with relapsed/refractory DLBCL were included (24 men; 16 women; median age, 66 years; median of 4 prior treatments, including rituximab chemotherapy). Patients were classified as germinal center B-cell like (n ¼ 23) or nongerminal center B-cell like (n ¼ 17) DLBCL according to the Hans algorithm. The subgroups were similar in terms of stage, international prognostic index score, prior number of treatments, and rituximab resistance. A significant difference in clinical response to lenalidomide was observed in nongerminal center B-cell like versus germinal center B-cell like patients. ORR was 52.9% versus 8.7% (P ¼.006); complete response rate was 23.5% versus 4.3%. Median progression-free survival was 6.2 versus 1.7 months (P ¼.004), although no difference in OS was observed between nongerminal center B-cell like and germinal center B-cell like DLBCL patients. CONCLUSIONS: The data suggest that the 2 major subgroups of patients with DLBCL (germinal center B cell and nongerminal center B cell) have different antitumor responsiveness to lenalidomide in the relapsed/refractory setting. A large international trial (NCT ) has been opened to enrollment in an attempt to prospectively validate these retrospective observations. Cancer 2011;117: VC 2011 American Cancer Society. KEYWORDS: lenalidomide, relapsed/refractory, diffuse large B-cell lymphoma, germinal center B-cell like lymphoma, nongerminal center B-cell like lymphoma, Hans algorithm. Despite currently available treatments, approximately 60,000 new non-hodgkin lymphoma (NHL) cases and 20,000 deaths have been estimated in the United States for The most common type of B-cell NHL is diffuse large B-cell lymphoma (DLBCL), which has an aggressive clinical course. DLBCL can be divided into subgroups with distinct biological characteristics and prognosis using gene expression profiling analysis. 2-4 Initial gene expression profiling studies have identified 3 subgroups of primary DLBCL patients with different biology and clinical outcomes: germinal center B cell-like, activated B cell-like, and type 3. 3,4 There is a growing need to further characterize resistant DLBCL cells at the molecular level in an attempt to identify biomarkers predictive of response to specific salvage therapies, better understand the mechanisms associated Corresponding author: Francisco J. Hernandez-Ilizaliturri, MD, Assistant Professor of Medicine and Immunology, Department of Medical Oncology and Immunology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263; Fax: (716) ; Francisco.hernandez@roswellpark.org 1 Department of Medical Oncology and Immunology, Roswell Park Cancer Institute, Buffalo, New York; 2 Department of Pathology, Roswell Park Cancer Institute, Buffalo, New York; 3 Department of Hematology and Oncological Sciences Lorenzo and Ariosto Seràgnoli, University of Bologna, Bologna, Italy; 4 Department of Anatomic Pathology and Laboratory Medicine and Pathology, College of Medicine, Mayo Clinic, Rochester, Minnesota; 5 Department of Hematology and Oncology, John Theurer Cancer Center at the Hackensack University Medical Center, Hackensack, New Jersey; 6 Department of Hematology and Laboratory Medicine and Pathology, College of Medicine, Mayo Clinic, Rochester, Minnesota Presented in part at the 7th International Workshop on Non-Hodgkin Lymphoma, Barcelona, Spain, October 2-3, 2009, and at the 2010 Annual Meeting of the American Society of Clinical Oncology, Chicago, Illinois, June 2-4, DOI: /cncr.26135, Received: November 23, 2010; Revised: January 19, 2011; Accepted: January 25, 2011, Published online April 14, 2011 in Wiley Online Library (wileyonlinelibrary.com) 5058 Cancer November 15, 2011

2 Lenalidomide Activity in Lymphoma/Hernandez-Ilizaliturri et al with acquired resistance to immunochemotherapy, and identify and develop therapeutic strategies directed against novel targets and/or pathways. Prospective validation of gene expression profiling in frontline therapy for DLBCL is now being incorporated into clinical trials. DLBCL subclassification has been translated into an immunohistochemistry (IHC) approach that demonstrates clinical applicability. 5-7 DLBCL patients are mostly divided into 2 subgroups: patients with germinal center B-cell like or nongerminal center B-cell like (mainly activated B cell-like) DLBCL. However, the differences in the clinical behavior and therapeutic response of patients with relapsed/refractory DLBCL with germinal center B-cell like and nongerminal center B-cell like phenotypes are largely unknown. Lenalidomide (Revlimid; Celgene, Summit, NJ), an immunomodulatory drug, is emerging as an attractive therapeutic option for patients with B-cell lymphoproliferative neoplasms, including DLBCL Studies in lymphoma and multiple myeloma (MM) models have demonstrated that lenalidomide exerts higher antitumor activity than thalidomide, has a unique capacity to stimulate the innate immune system, enhances the antitumor activity of rituximab, and inhibits angiogenesis. 13,14 The effects of lenalidomide appear to be related to the ability of immunomodulatory drug compounds to inhibit tumor necrosis factor-a, vascular endothelial growth factor, and nuclear factor kappa B (NF-KB) activity in tumor cells. 14,15 These compounds have been shown to induce apoptosis, and/or augment natural killer cell cytotoxicity against MM and B-cell lymphoma cells. 13,14,16 Recently, lenalidomide was found to modulate T-cell functions and repair F-actin mediated T-cell immune synapses against primary tumor cells in chronic lymphocytic leukemia, follicular lymphoma, and DLBCL. 17 Two phase 2 clinical trials (NHL-002 and NHL- 003) were previously conducted to evaluate efficacy and safety of lenalidomide monotherapy in relapsed/refractory aggressive lymphoma (including DLBCL). 11,12 In these studies, the overall response rates (ORRs) in patients with DLBCL were 19% to 28%, including 7% to 12% complete responses (CRs)/unconfirmed complete responses. Responses were durable, with a median duration of response in these patients of 4.6 months. In the present report, we studied biomarkers of responsiveness to lenalidomide in patients with relapsed/ refractory DLBCL in an attempt to guide the development of future clinical trials. MATERIALS AND METHODS Study Design We retrospectively reviewed the clinical, radiological, and pathological records, as well as archived lymphoma tissue from patients diagnosed with relapsed/refractory DLBCL treated with lenalidomide monotherapy (from NHL-002 and NHL-003) or in combination with rituximab or dexamethasone from Roswell Park Cancer Institute, Mayo Clinic, University of Bologna, and the John Theurer Cancer Center at the Hackensack University Medical Center. Lenalidomide-treated DLBCL patients were also identified from local registries, the Roswell Park Cancer Institute or John Theurer Cancer Center lenalidomide registry for off-study lenalidomide use, and the clinical database used for NHL-002 and NHL-003 trials (Roswell Park Cancer Institute and Mayo Clinic) and other lenalidomide trials performed at the University of Bologna (Eudract ) and the John Theurer Cancer Center (CC-5013-NHL-005, ClinicalTrials.gov identifier: NCT ). Institutional review board approval was obtained for this investigation. Eudract was a phase 2 study evaluating the safety and efficacy of lenalidomide plus rituximab in patients with DLBCL. CC-5013-NHL-005 was a phase 2 clinical trial evaluating the antitumor activity of lenalidomide plus dexamethasone. Patients with DLBCL, de novo, secondary, or associated with follicular lymphoma diagnosed according to the World Health Organization 2008 classification 18 were included in our final analysis. Demographics and clinical characteristics were collected for each case and included age, sex, race, International Prognostic Index (IPI) score at diagnosis and relapse, histological diagnosis, Ann Arbor stage, performance status, lactate dehydrogenase (LDH) levels, frontline and subsequent therapies (including high-dose chemotherapy [HDC] and autologous stem cell support [ASCS]), history of radiation therapy, response and its duration to last prior therapy, and survival data. Pathological Classification of DLBCL Into Germinal Center B-Cell Like and Nongerminal Center B-Cell Like Subgroups Surgical pathology material and/or pathology reports were reviewed for each patient at Roswell Park Cancer Institute or at Mayo Clinic. At Roswell Park Cancer Institute, all cases of DLBCL were evaluated by IHC for the expression of CD10 (clone NCL-CD10-270; Novocastra, Bannockburn, Ill), B-cell lymphoma 6 (Bcl-6; clone PG- B6p; Dako, Carpentaria, Calif), and interferon regulatory Cancer November 15,

3 factor/multiple myeloma 1 (IRF4/MUM1) oncogene (clone MUM1p, Dako), among other markers. Surgical pathology materials from other participating institutions were submitted to either Roswell Park Cancer Institute or Mayo Clinic for pathological review and IHC. DLBCL case subtyping into germinal center B-cell like or nongerminal center B-cell like subgroups was based on the Hans algorithm. 5 Subtyping was performed on primary diagnostic samples and/or relapsed/refractory samples whenever available. A case was termed positive for each marker if 30% of the lymphoma cells stained positive. In addition, patients with available archived pathological material from Mayo Clinic who were participating in studies NHL-002 and NHL-003 were subjected to pathological review, IHC, and subtyping using the same parameters and antibodies as Roswell Park Cancer Institute. Treatment Plan All 40 patients included in the final analysis of this review received single-agent lenalidomide 25 mg for 21 days of a 28-day cycle. Laboratory studies were performed weekly, and lenalidomide dose was adjusted for subsequent cycles if grade 3 or 4 hematological or nonhematological toxicity was encountered. Patients continued on lenalidomide until disease progression or unacceptable toxicity occurred. All patients underwent imaging studies every 2 or 3 months, with the exception of 1 patient who underwent imaging studies after 4 cycles of lenalidomide and 3 patients who demonstrated rapid clinical progression of disease and went to hospice care after 2 cycles. Clinical Endpoints The efficacy of lenalidomide was determined by ORR, including CR and partial response (PR) rates, progression-free survival (PFS), and overall survival (OS). Clinical response was defined using standard Cheson criteria. 19 Patients enrolled in the lenalidomide clinical trials were evaluated for response every 2 cycles. Patients in the lenalidomide off-study use program were evaluated after 2 or 3 cycles at the discretion of the treating physician. Imaging studies for disease measurement consisted of computed tomography scans of the chest, abdomen, and pelvis, and, if available, positron emission tomography scans. Follow-up for each patient was available until death or last clinic visit (clinical data was collected until May 2010). PFS and OS were recorded for each patient. PFS was defined as the time from enrollment to the time of progressive disease, relapse, or death. OS was defined as the time from study enrollment to death. Differences in endpoints were compared between patients with germinal center B-cell like and nongerminal center B-cell like DLBCL. Statistical Analysis The primary endpoint of this study was to determine whether there was a difference in responsiveness to salvage lenalidomide between germinal center B-cell like and nongerminal center B-cell like DLBCL patients. CR and ORR (CR þ PR) were evaluated for all patients, and 2- sided 95% confidence intervals (CIs) were calculated. The estimate and the 95% CIs of the CR rate, PR rate, and ORR were also computed for all patients. Differences in PFS and OS in patients with germinal center B-cell like versus nongerminal center B-cell like DLBCL were analyzed, and 2-sided 95% CIs were calculated. PFS and OS data were presented as Kaplan-Meier estimates. RESULTS Patient Characteristics We identified 56 patients with relapsed/refractory DLBCL (49 de novo, 5 associated with follicular lymphoma, 2 secondary [ie, transformed]) treated with lenalidomide alone or in combination with rituximab or dexamethasone. Fifty-three cases were classified as germinal center B-cell like or nongerminal center B-cell like DLBCL based on availability of IHC data. Only patients treated with single-agent lenalidomide (n ¼ 40; 27 patients were on a lenalidomide trial; 13 patients received lenalidomide off-study) were included in this report. Table 1 summarizes the demographics of our study population (23 germinal center B cell, 17 nongerminal center B cell). The median number of prior therapies was 4 (range, 2-13). Eleven (27.5%) patients failed HDC and ASCS before lenalidomide therapy. All patients had previously received rituximab-based immunochemotherapy as frontline and/or salvage treatment, and 67.5% were rituximab resistant (defined as progression on their last rituximab chemotherapy regimen or relapse within 6 months after their last rituximab dose). Representative stainings of marker expression used to subtype DLBCL into germinal center B-cell like or nongerminal center B-cell like phenotypes are depicted in Figure 1. Only the DLBCL component was assessed for this subtyping. The germinal center B-cell like and nongerminal center B-cell like subgroups were similar with regard to median age, sex, stage, performance status, IPI score/risk category, prior number of treatments, use of HDC-ASCS, 5060 Cancer November 15, 2011

4 Lenalidomide Activity in Lymphoma/Hernandez-Ilizaliturri et al Table 1. Demographic Characteristics of DLBCL Patients Treated With Single-Agent Lenalidomide Characteristic All Patients GCB Non-GCB Patients Age, y Median Range Age, No. (%) <60 years 10 (25.0) 7 (30.4) 3 (17.6) 60 years 30 (75.0) 16 (69.6) 14 (82.4) Sex, No. (%) Male 24 (60.0) 13 (56.5) 11 (64.7) Female 16 (40.0) 10 (43.5) 6 (35.3) Race, No. (%) Caucasian 38 (95.0) 22 (95.7) 16 (94.1) Other 2 (5.0) 1 (4.3) 1 (5.9) Original histology, No. (%) DLBCL 34 (85.0) 19 (82.6) 15 (88.2) Composite/transformed 6 (15.0) 4 (17.4) 2 (11.8) Prior treatments, No. Median Range Rituximab resistant, No. (%) 27 (67.5) 15 (65.2) 12 (70.6) Stage, No. (%) a I 4 (10.0) 3 (13.0) 1 (5.9) II 4 (10.0) 3 (13.0) 1 (5.9) III 12 (30.0) 8 (34.8) 4 (23.5) IV 20 (50.0) 9 (39.1) 11 (64.7) IPI score risk category, No. (%) Low 11 (27.5) 8 (34.8) 3 (17.6) Low-intermediate 8 (20) 4 (17.4) 4 (23.5) High-intermediate 9 (22.5) 6 (26.0) 3 (17.6) High 12 (30) 5 (21.7) 7 (41.2) Abbreviations: DLBCL, diffuse large B-cell lymphoma; GCB, germinal center B-cell like; IPI, International Prognostic Index. a Stage at the time of lenalidomide therapy. and percentage with rituximab-resistant disease. Although the percentage of patients older than 60 years tended to be higher in the nongerminal center B-cell like (82%) than in the germinal center B-cell like subgroup (70%), this difference was not statistically significant (Table 1). Differences in Response Rates, PFS, and OS All patients completed at least 1 full cycle of lenalidomide therapy, except for 1 patient who died during cycle 1. Patients with germinal center B-cell like DLBCL received a median of 2 (range, 1-21) lenalidomide treatment cycles, whereas patients with nongerminal center B-cell like DLBCL received a median of 4 cycles (range, 1-35) (Table 2). This difference could be related to differences in response rates and duration of response. For the entire cohort, lenalidomide resulted in an ORR of 27.5% (6 CRs, 5 PRs), similar to what has previously been reported. Patients with nongerminal center B-cell like DLBCL had a significantly higher ORR than patients with germinal center B-cell like DLBCL (53% vs 9%; P ¼.006; Table 2 and Fig. 2). Furthermore, more patients with nongerminal center B-cell like DLBCL achieved a CR or PR (5 and 4, respectively) than patients with germinal center B-cell like DLBCL (1 and 1). Twenty-one (53%) patients progressed during lenalidomide treatment; 14 in the germinal center B-cell like subgroup and 7 in the nongerminal center B-cell like subgroup (61% vs 41%). One patient with symptomatic relapsed/refractory nongerminal center B-cell like DLBCL was considered to have progressive disease (PD) based on imaging studies performed after 2 cycles. This patient presented with Cancer November 15,

5 Figure 1. Two cases are shown of diffuse large B-cell lymphoma (DLBCL), with (A, C, E, G) germinal center B-cell like (GCB) and (B, D, F, H) non-gcb phenotypes. In GCB DLBCL, the large lymphoma cells are diffusely and strongly positive for (C) CD10 and (E) B-cell lymphoma 6 (Bcl-6) and are otherwise (G) negative for interferon regulatory factor-4/multiple myeloma-1 (IRF4/ MUM1); only sparse plasma cells are positive for IRF4/MUM1. In contrast, in non-gcb DLBCL, the large lymphoma cells are (D) negative for CD10, (F) partially positive for Bcl-6, and (H) frequently and strongly positive for IRF4/MUM1. (A, B) Hematoxylin & eosin staining; (C-H) immunoperoxidase staining; (A-H) original magnification, Cancer November 15, 2011

6 Lenalidomide Activity in Lymphoma/Hernandez-Ilizaliturri et al Table 2. Response to Lenalidomide Monotherapy All Patients GCB Non-GCB Lenalidomide cycles Median Range Response a CR 6 (15.0) 1 (4.3) 5 (29.4) b PR 5 (12.5) 1 (4.3) 4 (23.5) SD 7 (17.5) 7 (30.4) 0 PD 21 (52.5) 14 (60.9) 7 (41.2) Unknown 1 (2.5) c 0 1 (5.9) c ORR (CR þ PR) 11 (27.5) 2 (8.7) d 9 (52.9) d PFS, mo Mean e 10.8 e 95% CI Median e 6.2 e 95% CI Figure 2. Differences in responses to lenalidomide monotherapy in relapsed/refractory diffuse large B-cell lymphoma with germinal center B-cell like (GCB) versus non-gcb phenotype are shown. Abbreviations: CR, complete response; PR, partial response. Abbreviations: CI, confidence interval; CR, complete response; GCB, germinal left B-cell like; ORR, overall response rate; PD, progressive disease; PET, positron emission tomography; PFS, progression-free survival; PR, partial response; SD, stable disease. As per 1999 Cheson criteria (PET scan excluded). b One patient was originally assessed as SD, but subsequently reclassified as CR by the 2007 Cheson criteria (ie, PET negative for >12 months). c One patient died during the first cycle before imaging studies. d P ¼.006 non-cgb vs GCB. e P ¼.004 non-cgb vs GCB. painful lymphadenopathy and elevated calcium and LDH levels. Hypercalcemia was corrected (calcium mg/ml), and LDH values declined significantly ( IU/L), providing symptomatic disease control. This patient continued on therapy for 4 cycles with transient clinical benefit, as no alternative therapies were available. For the purposes of this report, the patient had PD. DLBCL patients with the nongerminal center B- cell like phenotype had a significantly longer median PFS than those with the germinal center B-cell like phenotype (6.2 months; 95% CI, months vs 1.7 months; 95% CI, months; P ¼.004) (Table 2 and Fig. 3). The median OS of lenalidomide-treated DLBCL was similar in both groups (14 months; 95% CI, months in nongerminal center B-cell like vs 13.5 months; 95% CI 0-33 months in germinal center B-cell like; P ¼.7; Fig. 4). The lack of a significant difference in OS between the 2 subgroups may reflect the relatively small number of patients studied and/or the limited number of active salvage therapies available after completion of lenalidomide treatment. DISCUSSION Our data suggest that relapsed/refractory DLBCL with a nongerminal center B-cell like phenotype (defined by Figure 3. Patients with relapsed/refractory diffuse large B- cell lymphoma with nongerminal center B-cell like (non-gcb) phenotype have a longer progression-free survival (PFS) than patients with GCB phenotype after lenalidomide monotherapy. Hans criteria 5 ) have a higher ORR and CR rate and longer PFS when treated with lenalidomide as compared with those with germinal center B-cell like phenotype. Recent studies using comparative genomic hybridization have demonstrated that patients with germinal center B-cell like lymphoma express a higher number of distinct chromosomal abnormalities than those with activated B-cell like lymphoma, and the 2 subtypes may therefore represent different clinical entities. This finding adds credibility to the concept of 2 different molecular, biological, and clinical subtypes of DLBCL. 20 Cancer November 15,

7 Figure 4. Overall survival (OS) of patients treated with lenalidomide monotherapy was not different between patients with diffuse large B-cell lymphoma with germinal center B- cell like (GCB) or non-gcb phenotypes. Gene expression profiling analysis has been validated to predict the outcome of DLBCL patients receiving frontline cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy. 2,21,22 Other gene expression profiling studies identify different genetic profiles (eg, host inflammatory or stromal gene signatures) that may be associated with responsiveness to frontline rituximab chemotherapy. 23,24 The incorporation of rituximab into the management of DLBCL was associated with improvements in ORR, PFS, and OS. However, this change in therapy requires a re-evaluation of previously accepted biomarkers of response (eg, Bcl-2 expression, IPI, gene expression profiling). The biological, clinical, and therapeutic significance of molecularly classifying DLBCL into germinal center B-cell like and nongerminal center B-cell like subgroups is being prospectively addressed by the Cancer and Leukemia Group B study. In this study, patients with newly diagnosed stage IIB or IV DLBCL are being randomized to rituximab-chop or rituximab plus doseadjusted etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin. This study seeks to validate the predictive value of gene expression profiling in rituximab chemotherapy-treated patients and to identify whether a particular immunochemotherapy regimen is more effective in specific DLBCL subgroups. IHC appears to be easy and practical in differentiating between DLBCL subgroups. Several attempts to subtype DLBCL cases into germinal center B cell and nongerminal center B cell have been made, with algorithms using several markers (eg, CD10, Bcl-6, IRF4/ MUM1), such as the Hans algorithm and the algorithm proposed by Muris et al. 5-7 The Hans algorithm reproduces the gene expression-based classification of DLBCL with a misclassification rate of 20%. 5 Both algorithms have been evaluated as predictors of clinical outcomes in DLBCL patients undergoing frontline therapy with CHOP or rituximab-chop. Nyman et al 25 demonstrated that patients with germinal center B-cell like DLBCL (defined by Hans algorithm) had better clinical outcomes than those with nongerminal center B-cell like DLBCL phenotypes if treated with systemic chemotherapy but not if treated with rituximab chemotherapy. This suggests that the incorporation of rituximab might eliminate the prognostic value of germinal center B-cell like versus nongerminal center B-cell like DLBCL as defined by IHC. 25 These investigators also retrospectively subtyped 88 consecutive rituximab-chop treated DLBCL patients with modified Hans criteria, in which staining for forkhead box P1 was added to the algorithm. 26 Patients with an activated B-cell like phenotype had a significantly worse outcome than patients with a nonactivated B-cell like phenotype (3-year failure-free survival, 63% vs 82%; P ¼.048; OS, 69% vs 85%; P ¼.110). When this patient population was studied by the Muris algorithm, patients with germinal center B-cell like DLBCL exhibited a better clinical outcome after therapy with rituximab-chop than patients with nongerminal center B-cell like DLBCL. 26 The prognostic significance of subclassifying DLBCL by either gene expression profiling or IHC in the salvage setting is largely unknown. Recently, Dunleavy et al 27 addressed the technical issues of IHC versus gene expression profiling in relapsed/refractory DLBCL by classifying patients as germinal center B-cell like or activated B-cell like subtypes using both methods. Patients with the activated B-cell like versus germinal center B- cell like subtype demonstrated a significantly higher response rate to bortezomib plus dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin (83% vs 13%; P <.001) and longer median OS (10.8 vs 3.4 months; P ¼.003). 27 Moreover, gene expression profiling and IHC had a 100% concordance rate in classifying relapsed/refractory DLBCL into germinal center B-cell like and activated B-cell like subtypes. This is in line with our results demonstrating that responses to salvage lenalidomide appear better in patients with nongerminal center B-cell like DLBCL Cancer November 15, 2011

8 Lenalidomide Activity in Lymphoma/Hernandez-Ilizaliturri et al Comparing these data raises some questions: 1) Are differences between germinal center B-cell like and nongerminal center B-cell like DLBCL with respect to response to bortezomib dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin or single-agent lenalidomide target dependent (eg, NF-KB)? 2) Is the response to any salvage chemotherapy worse for the majority of patients with germinal center B-cell like DLBCL refractory to or relapsed after frontline rituximab chemotherapy? Although more studies are necessary to address these questions, we believe that the reasons behind the observed different clinical response between DLBCL subgroups are probably target specific. Bortezomib and lenalidomide are known to both inhibit NF-KB activity in vitro, and activated B-cell like DLBCL is known to be driven by constitutive activation of NF-KB, 28 which supports the hypothesis that targeting the NF-KB-pathway can be clinically relevant in patients with nongerminal center B-cell like (or at least activated B-cell like subtype) DLBCL. A recent IHC study in 59 DLBCL patients with chemosensitive disease undergoing HDC-ASCS demonstrated that germinal center B-cell like and nongerminal center B-cell like patients (defined by Hans criteria) did not differ in risk of progression or OS. 29 Nevertheless, among patients with MUM1-positive DLBCL, there was a trend for a higher risk of progression (P ¼.08). In our cohort of lenalidomide-treated patients, those with nongerminal center B-cell like DLBCL were most likely to benefit from lenalidomide s immunomodulatory effects, which suggests the existence of a pathway targeted by lenalidomide in these patients. One candidate is the transcription regulatory factor IRF4/MUM1. Defects in IRF4/MUM1 expression result in defects in lymphocytes (T and B cell) and terminal differentiation of B cells toward plasma cells. 30 In addition, IRF4/MUM1 expression correlates with poor prognosis in various B-cell lymphoproliferative disorders. IRF4/MUM1 in cooperation with PU-1 enhances the transcription of the monokine induced by interferon-gamma (MIG) gene. 31 MIG protein interacts with the chemokine receptor 3 and induces an autocrine proliferation signal in malignant B cells. 31 IRF4/MUM1 up-regulation of MIG might be mediated by NF-KB activity. 32,33 However, the multiple mechanisms of action associated with immunomodulatory drug compounds suggest that the higher responsiveness of nongerminal center B- cell like patients to lenalidomide may be multifactorial. There are certain limitations of our study that stress the need for further validation. Although the differences in response to lenalidomide between patients with germinal center B-cell like versus nongerminal center B-cell like DLBCL are statistically significant, our sample size is relatively small. Also, this analysis is retrospective, similar to the majority of the above-cited studies, and should therefore be prospectively confirmed in larger populations treated with single-agent lenalidomide before reaching any final conclusions. Moreover, we did not evaluate other biomarkers of response recently identified in the postrituximab era. It is plausible that other biomarkers may more precisely predict responsiveness to lenalidomide. Finally, it remains unclear whether our findings are applicable to frontline lenalidomide treatment. The clinical significance of our findings may prove to be important if prospectively validated. We have identified a potential biomarker of response to lenalidomide with a readily available IHC assay. Moreover, lenalidomide salvage treatment may emerge as an important and effective therapeutic strategy for a high percentage of patients with nongerminal center B-cell like DLBCL. Our hypothesis-driven finding stresses the importance of conducting clinical trials with built-in correlative studies of molecular relevance attempting to identify potential biomarkers of response to specific therapies. A prospective randomized phase 2/3 clinical trial is currently enrolling patients to compare the antitumor activity of lenalidomide versus investigator s choice monotherapy in patients with relapsed/refractory DLBCL who had failed, or were not eligible for, HDC-ASCS (ClinicalTrials.gov Identifier: NCT ). Patients will be risk-stratified according to germinal center B-cell like versus nongerminal center B-cell like subtype using the Hans algorithm. The results of this prospective clinical trial may validate our initial observations and, more importantly, identify a subgroup of patients with DLBCL most likely to clinically benefit from lenalidomide monotherapy in the relapsed/ refractory setting. FUNDING SOURCES The authors received editorial support in the preparation of the manuscript funded by Celgene Corporation. The authors are fully responsible for the content and editorial decisions for this article. CONFLICT OF INTEREST DISCLOSURES F.J.H.-I., A.G., T.E.W., and M.S.C. have received research funding from Celgene. T.E.W. and M.S.C. have a consultant or advisory role with Celgene. The rest of the authors have nothing to disclose. Cancer November 15,

9 REFERENCES 1. Jemal A, Siegel R, Xu J, et al. Cancer statistics, CA Cancer J Clin. 2010;60: Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000;403: 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: Staudt LM. Gene expression profiling of lymphoid malignancies. Annu Rev Med. 2002;53: 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: Muris JJ, Meijer CJ, Vos W, et al. Immunohistochemical profiling based on Bcl-2, CD10 and MUM1 expression improves risk stratification in patients with primary nodal diffuse large B cell lymphoma. J Pathol. 2006;208: Colomo L, Lopez-Guillermo A, Perales M, et al. Clinical impact of the differentiation profile assessed by immunophenotyping in patients with diffuse large B-cell lymphoma. Blood. 2003;101: Galustian C, Dalgleish A. Lenalidomide: a novel anticancer drug with multiple modalities. Expert Opin Pharmacother. 2009;10: Chanan-Khan AA, Cheson BD. Lenalidomide for the treatment of B-cell malignancies. J Clin Oncol. 2008;26: Goy A. New directions in the treatment of mantle cell lymphoma: an overview. Clin Lymphoma Myeloma. 2006;7(suppl 1): S24-S Wiernik PH, Lossos IS, Tuscano JM, et al. Lenalidomide monotherapy in relapsed or refractory aggressive non-hodgkin s lymphoma. J Clin Oncol. 2008;26: Witzig TE, Vose JM, Zinzani PL, et al. An international phase II trial of single-agent lenalidomide for relapsed or refractory aggressive B-cell non-hodgkin s lymphoma. Ann Oncol Jan 12 [Epub ahead of print]. 13. Hernandez-Ilizaliturri FJ, Reddy N, Holkova B, et al. Immunomodulatory drug CC-5013 or CC-4047 and rituximab enhance antitumor activity in a severe combined immunodeficient mouse lymphoma model. Clin Cancer Res. 2005;11: Reddy N, Hernandez-Ilizaliturri FJ, Deeb G, et al. Immunomodulatory drugs stimulate natural killer-cell function, alter cytokine production by dendritic cells, and inhibit angiogenesis enhancing the anti-tumour activity of rituximab in vivo. Br J Haematol. 2008;140: Keifer JA, Guttridge DC, Ashburner BP, et al. Inhibition of NF-kappa B activity by thalidomide through suppression of IkappaB kinase activity. J Biol Chem. 2001;276: Zhu D, Corral LG, Fleming YW, et al. Immunomodulatory drugs Revlimid (lenalidomide) and CC-4047 induce apoptosis of both hematological and solid tumor cells through NK cell activation. Cancer Immunol Immunother. 2008;57: Ramsay AG, Clear AJ, Kelly G, et al. Follicular lymphoma cells induce T-cell immunologic synapse dysfunction that can be repaired with lenalidomide: implications for the tumor microenvironment and immunotherapy. Blood. 2009; 114: Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: IARC Press; Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol. 2007;25: Bea S, Zettl A, Wright G, et al. Diffuse large B-cell lymphoma subgroups have distinct genetic profiles that influence tumor biology and improve gene-expression-based survival prediction. Blood. 2005;106: Shipp MA, Ross KN, Tamayo P, et al. Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nat Med. 2002;8: Lossos IS, Czerwinski DK, Alizadeh AA, et al. Prediction of survival in diffuse large-b-cell lymphoma based on the expression of 6 genes. N Engl J Med. 2004;350: Monti S, Savage KJ, Kutok JL, et al. Molecular profiling of diffuse large B-cell lymphoma identifies robust subtypes including 1 characterized by host inflammatory response. Blood. 2005;105: Lenz G, Wright G, Dave SS, et al. Stromal gene signatures in large-b-cell lymphomas. NEnglJMed.2008;359: Nyman H, Adde M, Karjalainen-Lindsberg ML, et al. Prognostic impact of immunohistochemically defined germinal center phenotype in diffuse large B-cell lymphoma patients treated with immunochemotherapy. Blood. 2007;109: Nyman H, Jerkeman M, Karjalainen-Lindsberg ML, et al. Prognostic impact of activated B-cell focused classification in diffuse large B-cell lymphoma patients treated with R- CHOP. Mod Pathol. 2009;22: 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: Davis RE, Staudt LM. Molecular diagnosis of lymphoid malignancies by gene expression profiling. Curr Opin Hematol. 2002;9: Costa LJ, Feldman AL, Micallef IN, et al. Germinal center B (GCB) and non-gcb cell-like diffuse large B cell lymphomas have similar outcomes following autologous haematopoietic stem cell transplantation. Br J Haematol. 2008;142: Mittrucker HW, Matsuyama T, Grossman A, et al. Requirement for the transcription factor LSIRF/IRF4 for mature B and T lymphocyte function. Science. 1997;275: Uranishi M, Iida S, Sanda T, et al. Multiple myeloma oncogene 1 (MUM1)/interferon regulatory factor 4 (IRF4) upregulates monokine induced by interferon-gamma (MIG) gene expression in B-cell malignancy. Leukemia. 2005;19: Calame KL. Plasma cells: finding new light at the end of B cell development. Nat Immunol. 2001;2: Johnson K, Calame K. Transcription factors controlling the beginning and end of B-cell differentiation. Curr Opin Genet Dev. 2003;13: Cancer November 15, 2011