Original Article

Prevalence and Clinical Implications of Cyclin D1 Expression in Diffuse Large B-Cell Lymphoma (DLBCL) Treated With Immunochemotherapy A Report From the International DLBCL Rituximab-CHOP Consortium Program Chi Young Ok, MD1; Zijun Y. Xu-Monette, PhD1; Alexandar Tzankov, MD2; Dennis P. O’Malley, MD3; Santiago Montes-Moreno, MD4; Carlo Visco, MD5; Michael B. Møller, MD6; Karen Dybkær, PhD7; Attilio Orazi, MD8; Youli Zu, MD9; Govind Bhagat, MD10; Kristy L. Richards, MD11; Eric D. Hsi, MD12; J. Han van Krieken, MD13; Maurilio Ponzoni, MD14; John P. Farnen, MD15; Miguel A. Piris, MD4; Jane N. Winter, MD16; L. Jeffrey Medeiros, MD1; and Ken H. Young, MD, PhD1

BACKGROUND: Cyclin D1 expression has been reported in a subset of patients with diffuse large B-cell leukemia (DLBCL), but studies have been few and generally small, and they have demonstrated no obvious clinical implications attributable to cyclin D1 expression. METHODS: The authors reviewed 1435 patients who were diagnosed with DLBCL as part of the International DLBCL rituximab with cyclophosphamide, hydroxydaunorubicin, vincristine, and prednisone (R-CHOP) Consortium Program and performed clinical, immunohistochemical, and genetic analyses with a focus on cyclin D1. All patients who were cyclin D1-positive according to immunohistochemistry were also assessed for rearrangements of the cyclin D1 gene (CCND1) using fluorescence in situ hybridization. Gene expression profiling was performed to compare patients who had DLBCL with and without cyclin D1 expression. RESULTS: In total, 30 patients (2.1%) who had DLBCL that expressed cyclin D1 and lacked CCND1 gene rearrangements were identified. Patients with cyclin D1-positive DLBCL had a median age of 57 years (range, 16.0-82.6 years). There were 23 males and 7 females. Twelve patients (40%) had bulky disease. None of them expressed CD5. Two patients expressed cyclin D2. Gene expression profiling indicated that 17 tumors were of the germinal center type, and 13 were of the activated B-cell type. Genetic aberrations of B-cell leukemia/lymphoma 2 (BCL2), BCL6, v-myc avian myelocytomatosis viral oncogene homolog (MYC), mouse double minute 2 oncogene E3 ubiquitin protein ligase (MDM2), MDM4, and tumor protein 53 (TP53) were rare or absent. Gene expression profiling did not reveal any striking differences with respect to cyclin D1 in DLBCL. CONCLUSIONS: Compared with patients who had cyclin D1-negative DLBCL, men were more commonly affected with cyclin D1-positive DLBCL, and they were significantly younger. There were no other significant differences in clinical presentation, pathologic features, overall survival, or progression-free survival between these two subgroups of patients with C 2014 American Cancer Society. DLBCL. Cancer 2014;120:1818–29. V KEYWORDS: cyclin D1, pleomorphic mantle cell lymphoma, diffuse large B-cell lymphoma.

INTRODUCTION Diffuse large B-cell lymphoma (DLBCL) is the most common type of B-cell lymphoma in the world.1 DLBCL, as currently defined, is a heterogeneous group of diseases with many morphologic and immunophenotypic variants and molecular subtypes.2 The standard therapy for patients with DLBCL is rituximab with cyclophosphamide, hydroxydaunorubicin, vincristine, and prednisone (R-CHOP); and the 10-year overall survival (OS) and disease-free survival (DFS) rates are 43.5% and 36.5%, respectively.3

Corresponding author: Ken H. Young, MD., PhD, Department of Hematopathology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77230-1439; Phone: (713) 745-2598; Fax: (713) 792-7273; [email protected] 1 Department of Hematopathology, The University of Texas MD Anderson Cancer Center, Houston, Texas; 2Department of Pathology, Basel University Hospital, Basel, Switzerland; 3Clarient Pathology Laboratory, Aliso Viejo, California; 4Marques de Valdecilla University HospitalSantander, Spain; 5Department of Hematology/Oncology, San Bortolo Hospital, Vicenza, Italy; 6Department of Pathology, Odense University Hospital, Odense, Denmark; 7Department of Clinical Medicine/ Hematology, Aalborg University Hospital, Aalborg, Denmark; 8Department of Pathology, Weill Medical College of Cornell University, New York, New York; 9Department of Pathology, The Methodist Hospital, Houston, Texas; 10Department of Pathology, College of Physicians and Surgeons, Columbia University Medical Center and New York Presbyterian Hospital, New York, New York; 11Department of Medicine, University of North Carolina School of Medicine, Chapel Hill, North Carolina; 12 Department of Clinical Pathology, The Cleveland Clinic, Cleveland, Ohio; 13Department of Pathology, Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands; 14Department of Pathology, San Raffaele H. Scientific Institute, Milan, Italy; 15Department of Hematology/Oncology, Gundersen Lutheran Health System, La Crosse, Wisconsin; 16Department of Hematology/Oncology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois

DOI: 10.1002/cncr.28664, Received: January 28, 2014; Accepted: February 10, 2014, Published online March 19, 2014 in Wiley Online Library (wileyonlinelibrary. com)

1818

Cancer

June 15, 2014

DLBCL With Cyclin D1 Expression/Ok et al

Mantle cell lymphoma (MCL) represents 4% to 10% of B-cell lymphomas worldwide and is characterized by t(11;14)(q13;q32) resulting in immunoglobulin heavy locus/ cyclin D1 (IGH/CCND1) fusion, which results in marked over-expression of the cyclin D1 protein.4 MCL is an aggressive lymphoma, and affected patients respond poorly to conventional chemotherapy. MCL can exhibit a wide range of morphologic appearances. A blastoid form occurs in up to 10% of patients, and another histological form, the pleomorphic variant, exhibits morphologic overlap with DLBCL. Distinguishing pleomorphic variant MCL from DLBCL is important because R-CHOP therapy is considered inadequate for patients with MCL; and, in some institutions (including ours), patients with MCL most often receive treatment with a dose-intensified regimen: rituximab combined with fractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone (R-hyper-CVAD) alternating with rituximab in combination with high-dose methotrexatecytarabine (R-MA).5 Previous studies have demonstrated that a subset of patients with DLBCL overexpress cyclin D1 at the immunohistochemical level, and this expression occurs independently of the t(11;14).6-11 There have been only few case reports and case series of cyclin D1 expression in DLBCL. The objectives of the current study were: 1) to determine the frequency of cyclin D1 expression in a large cohort of patients with DLBCL; 2) to comprehensively characterize patients who have DLBCL with cyclin D1 expression; and 3) to estimate of the frequency with which pleomorphic MCL might be misdiagnosed as DLBCL. MATERIALS AND METHODS Patients

In total, 1435 patients with de novo DLBCL were evaluated. All patients were a part of the International DLBCL Rituximab-CHOP Consortium Program12-15 and were diagnosed according to World Health Organization (WHO) classification criteria. Exclusion criteria were transformation from low-grade B-cell lymphoma and immunodeficiency-associated cases, especially human immunodeficiency virus infection. This study was conducted in accordance with the Declaration of Helsinki and was approved by the institutional review boards of all participating institutions. The overall collaborative study was approved by the Institutional Review Board at The University of Texas MD Anderson Cancer Center.12-15 Tissue Microarray and Immunohistochemistry

Hematoxylin-eosin stained slides from each of the 1435 patients with DLBCL were reviewed, and tumor-rich Cancer

June 15, 2014

areas were selected. Tissue microarrays (TMAs) were constructed using a tissue microarrayer (Beecher Instrument, Silver Spring, Md). Immunohistochemical analyses were performed on 4-lm TMA sections using a streptavidinbiotin complex technique with antibodies that were reactive for the following antigens: cluster of differentiation 5 (CD5; a type I transmembrane protein), CD10 (a type II transmembrane protein), CD30 (a type I transmembrane protein), B-cell leukemia/lymphoma 2 (BCL-2), BCL-6, cyclin D1, cyclin D2, cyclin D3, forkhead box P1 (FOXP1), germinal center B-cell–expressed transcript 1 (GCET1), multiple myeloma antigen 1 (MUM1), and nuclear factor-jB (NF-jB) pathway components protein 65 (p65), p50, p52, and c-Rel (a protein encoded by the REL gene); and phosphorylated signal transducer and activator of transcription 3 (pSTAT3) in all cases. Because of tissue exhaustion, staining was not always available for each marker. Antigen expression was scored in 10% increments by assessing the percentage of immunoreactive tumor cells. A cutoff value for each marker was established using an analysis of receiver operating characteristic (ROC) curves and/or X-Tile analysis to achieve maximum specificity and sensitivity as described previously.13 The cutoff scores for these markers were established as follows: 20% for CD30, 30% for CD10 and BCL-6; 50% for cyclin D2 and cyclin D3; 60% for GCET1, MUM1, and FOXP1; and 70% for BCL-2. When an optimal cutoff value could not be determined by ROC curve and/or XTile analyses, cutoff values were established based on a literature review, which indicate cutoff scores of 20% for CD516 and 30% for pSTAT3.17 Any nuclear expression of cyclin D1 and each NF-jB component was considered positive. We also performed cyclin D1 and CD20 double staining in selected cases. Fluorescence in Situ Hybridization for CCND1, BCL2, BCL6, MYC, MDM2, MDM4, and TP53 Sequencing

Fluorescence in situ hybridization (FISH) was performed on paraffin-embedded tissue sections with a locus-specific CCND1 dual-color, break-apart probe (Vysis, Downers Grove, Ill); BCL2 and BCL6 dual-color, break-apart probes (Vysis); v-myc avian myelocytomatosis viral oncogene homolog (MYC) locus-specific IGH/MYC/chromosome enumeration probe 8 (CEP8) tricolor, dual-fusion probes; a locus-specific MYC dual-color, break-apart probe (Vysis); and mouse double minute 2 oncogene E3 ubiquitin protein ligase (MDM2) and MDM4 probes as described previously.13,15 Tissues on the TMA were considered for evaluation if at least 200 tumor cell nuclei per 1819

Original Article

core displayed positive signals. Abnormal FISH signals were recorded as the percentage of cells that had an abnormality. Tumor protein 53 (TP53) exon sequencing with a p53 AmpliChip (Roche Molecular Systems, Pleasanton, Calif) was performed in 618 patients as previously described.12 Cell-of-Origin Classification

Cell-of-origin (COO) classification was established by combining data from gene expression profiling (GEP), which is considered the “gold standard,” and immunohistochemistry (IHC) data. RNA was extracted from each of 583 sections using the HighPure RNA Extraction Kit (Roche Applied Science, Indianapolis, Ind) and was subjected to GEP as previously described.12,13,15 COO classification could be determined in 528 cases by GEP. When COO was not classifiable with GEP, it was determined by IHC according to the Visco-Young and Choi algorithms.13,18 The correlation between GEP and IHC for COO classification was 86% overall. Response Definitions and Statistical Analysis

Clinical and laboratory features were compared using the Fisher exact test for categorical variables and the MannWhitney U test for continuous variables. OS and PFS were calculated from the date of diagnosis to the date of last follow-up or death and from the date of diagnosis to the date of progression or death, respectively. KaplanMeier survival curves were used to estimate OS, and the log-rank test was used to assess differences in survival between groups. All differences with P values < .05 were considered statistically significant. RESULTS Patients’ Characteristics

There were 821 males and 614 females (male to female ratio, 1.3:1), and the median patient age was 62.9 years (range, 3.4-96.0 years). B-symptoms were present in 35.6% of patients. Approximately half of the patients had advanced Ann Arbor stage disease (51.3%), or had an elevated serum lactate dehydrogenase level (49.4%), or had both. The Eastern Cooperative Oncology Group performance status was considered good (performance status