Five-Gene Model to Predict Survival in Mantle-Cell Lymphoma Using Frozen or Formalin-Fixed, Paraffin-Embedded Tissue Elena Hartmann, Vero `nica Ferna `ndez, Victor Moreno, Joan Valls, Luis Herna ´ndez, Francesc Bosch, Pau Abrisqueta, Wolfram Klapper, Martin Dreyling, Eva Hoster, Hans Konrad Mu ¨ller-Hermelink, German Ott, Andreas Rosenwald, and Elı ´as Campo From the Institute of Pathology, Univer- sity of Wu ¨ rzburg, Wu ¨ rzburg; Institute of Pathology, University of Kiel, Kiel; Department of Internal Medicine III, University of Munich, Munich, Germ- any; Hematopathology Section, Depart- ments of Pathology and Hematology, Hospital Clinic, Institut d’Investigacions Biomediques August Pi i Sunyer, University of Barcelona; and Unit of Biostatistics and Bioinformatics, Cancer Epidemiology Service, Institut Catala ` d'Oncologia-Institut d'Investigacio ´ de Bellvitge, Barcelona, Spain. Submitted April 4, 2007; accepted May 14, 2008; published online ahead of print at www.jco.org on July 7, 2008. Supported by the Spanish Ministry of Education and Science (predoctoral fellowship to V.F.); the Interdisciplinary Center for Clinical Research, University of Wuerzburg, Germany (A.R. and E.H.); FIS 01/3046 (L.H.); European Grant No. 503351 (European Mantle Cell Lymphoma Network, A.R., E.C., G.O., W.K., and M.D.); the Lymphoma Research Foundation (A.R. and E.C.); Grant No. CICYT SAF 05/5855 from the Spanish Ministry of Science (E.C.); and Grant No. ISCIII-RETIC RD06/0020 from the Spanish Ministry of Health (E.C. and V.M.). Both E.H. and V.F. contributed equally to this work. E.C. and A.R. are cosenior authors of this study. Authors’ disclosures of potential con- flicts of interest and author contribu- tions are found at the end of this article. Corresponding author: Andreas Rosenwald, MD, Institute of Pathology, University of Wu ¨ rzburg; Josef- Schneider-Str 2, 97080 Wu ¨ rzburg, Germany; e-mail: rosenwald@mail .uni-wuerzburg.de. © 2008 by American Society of Clinical Oncology 0732-183X/08/2630-4966/$20.00 DOI: 10.1200/JCO.2007.12.0410 A B S T R A C T Purpose Despite the common underlying translocation t(11;14) involving cyclin D1 that is present in nearly all cases of mantle-cell lymphoma (MCL), the clinical course of the disease is highly variable. The aim of the present study was to develop a quantitative gene expression– based model to predict survival in newly diagnosed patients with MCL that involves a minimum number of genes and is applicable to fresh-frozen and formalin-fixed, paraffin-embedded (FFPE) tumor samples. Patients and Methods The expression of 33 genes with potential prognostic and pathogenetic impact in MCL was analyzed using quantitative reverse-transcription polymerase chain reactions (qRT-PCR) in a low-density array format in frozen tumor samples from 73 patients with MCL. Multivariate Cox methods and stepwise algorithms were applied to build gene expression-based survival predic- tors. An optimized five-gene model was subsequently applied to FFPE tumor samples from 13 patients with MCL from the initial series and to 42 independent MCL samples. Results The optimized survival predictor was composed of the five genes RAN, MYC, TNFRSF10B, POLE2, and SLC29A2 and was validated for application in FFPE tissue samples. It allowed the survival prediction of patients with MCL with widely disparate clinical outcome and was superior to the immunohistochemical marker Ki-67, an established prognostic factor in MCL. Conclusion We here present a validated qRT-PCR– based test for survival prediction in patients with MCL that is applicable to fresh frozen as well as to FFPE tissue specimens. This test may prove useful to guide individualized treatment approaches for patients with MCL. J Clin Oncol 26:4966-4972. © 2008 by American Society of Clinical Oncology INTRODUCTION Mantle-cell lymphoma (MCL) is a distinctive B-cell non-Hodgkin’s lymphoma characterized by the translocation t(11;14)(q13;32) and the overexpres- sion of cyclin D1. 1-3 The clinical course of patients with MCL can be highly variable. In general, MCL shows an aggressive clinical behavior with poor re- sponse to current therapeutic approaches, 2,4 and the median survival of patients with MCL after diagno- sis is between 3 and 5 years. However, some patients succumb to their disease in less than 6 months, whereas others survive more than 10 years. The cur- rent treatment approaches for patients with MCL are rather uniform and rarely reflect the widely vary- ing clinical behavior. However, before more individ- ualized and risk-adapted treatment strategies can be tested in prospective clinical trials, it is a prerequisite to develop molecular tests for newly diagnosed pa- tients with MCL that reliably predict their clinical course and that are applicable in a routine diagnos- tic setting. In recent years, many attempts have been made to identify clinical, histologic, and molecular mark- ers that allow the stratification of patients with MCL into different risk groups. Morphologically, two ma- jor variants of MCL have been identified, namely the classical and the blastoid variants, and blastoid MCL cases show a worse clinical outcome. 5,6 Genetically, alterations affecting the cell cycle machinery and the DNA damage response pathways seem to be the hallmarks of MCL. 1-3 Secondary genetic alterations in MCL frequently target key regulators of these pathways and are also associated with inferior clini- cal outcome. 7-12 The strongest predictor of survival in patients with MCL was identified by gene expression profil- ing in a large series of MCL. 13 In particular, a gene JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T VOLUME 26  NUMBER 30  OCTOBER 20 2008 4966 © 2008 by American Society of Clinical Oncology Downloaded from ascopubs.org by 35.175.195.222 on June 14, 2022 from 035.175.195.222 Copyright © 2022 American Society of Clinical Oncology. All rights reserved.