Francisco Mazorra

PURPOSE: Despite major advances in the treatment of classic Hodgkin's lymphoma (cHL), approximately 30% of patients in advanced stages may eventually die as result of the disease, and current methods to predict prognosis are rather unreliable. Thus, the application of robust techniques for the identification of biomarkers associated with treatment response is essential if new predictive tools are to be developed. EXPERIMENTAL DESIGN: We used gene expression data from advanced cHL patients to identify transcriptional patterns from the tumoral cells and their nonneoplastic microenvironment, associated with lack of maintained treatment response. Gene-Set Enrichment Analysis was used to identify functional pathways associated with unfavorable outcome that were significantly enriched in either the Hodgkin's and Reed-Sternberg cells (regulation of the G2-M checkpoint, chaperones, histone modification, and signaling pathways) or the reactive cell microenvironment (mainly represented by specific T-cell populations and macrophage activation markers). RESULTS: To explore the pathways identified previously, we used a series of 52 formalin-fixed paraffin-embedded advanced cHL samples and designed a real-time PCR-based low-density array that included the most relevant genes. A large majority of the samples (82.7%) and all selected genes were analyzed successfully with this approach. CONCLUSIONS: The results of this assay can be combined in a single risk score integrating these biological pathways associated with treatment response and eventually used in a larger series to develop a new molecular outcome predictor for advanced cHL.

// Ana Batlle-López 1 , Sonia González de Villambrosía 1 , Mazorra Francisco 1 , Sefora Malatxeberria 1 , Anabel Sáez 2 , Carlos Montalban 3 , Lydia Sánchez 4 , Juan F. Garcia 5 , Eva González-Barca 6 , Andrés López-Hernández 7 , MC Ruiz-Marcellan 7 , Manuela Mollejo 8 , Carlos Grande 9 , Kristy L. Richards 10 , Eric D. Hsi 11 , Alexandar Tzankov 12 , Carlo Visco 13 , Zijun Y. Xu-Monette 14 , Xin Cao 14 , Ken H. Young 14 , Miguel Ángel Piris 1 , Eulogio Conde 1 , Santiago Montes-Moreno 1 1 Departments of Haematology and Pathology, Hospital Marques de Valdecilla, and IDIVAL, Santander, Spain 2 Biobanco del Sistema Sanitario Público de Andalucía, Granada, Spain 3 MD Anderson Cancer Center, Madrid, Spain 4 Biotechnology Programme, Histology and Immunohistochemistry Core Unit, Spanish National Cancer Research Centre (CNIO), Madrid, Spain 5 Pathology, MD Anderson Cancer Center, Madrid, Spain 6 Department of Haematology, Hospital de Bellvitge (ICOIRO), Barcelona, Spain 7 Departments of Pathology and Haematology, Hospital Vall d’Hebron, Barcelona, Spain 8 Hospital Virgen de la Salud, Toledo, Spain 9 Hematology, Hospital 12 de Octubre, Madrid, Spain 10 Department of Hematology-Oncology, University of North Carolina School of Medicine, Chapel Hill, NC, USA 11 Department of Clinical Pathology, Cleveland Clinic, Cleveland, OH, USA 12 University Hospital, Basel, Switzerland 13 Department of Hematology, San Bortolo Hospital, Vicenza, Italy 14 Department of Hematopathology, University of Texas MD Anderson Cancer Center, Houston, TX, USA Correspondence to: Ana Batlle López, email: anabatllelopez@gmail.com Keywords: MYC, BCL2, BCL6, non-GCB and GCB, DLBCL Received: August 23, 2015      Accepted: February 02, 2016      Published: February 19, 2016 ABSTRACT Diffuse large B cell lymphoma (DLBCL) is a heterogeneous group of aggressive lymphomas that can be classified into three molecular subtypes by gene expression profiling (GEP): GCB, ABC and unclassified. Immunohistochemistry-based cell of origin (COO) classification, as a surrogate for GEP, using three available immunohistochemical algorithms was evaluated in TMA-arranged tissue samples from 297 patients with de novo DLBCL treated by chemoimmunotherapy (R-CHOP and R-CHOP-like regimens). Additionally, the prognostic impacts of MYC , BCL2 , IRF4 and BCL6 abnormalities detected by FISH, the relationship between the immunohistochemical COO classification and the immunohistochemical expression of MYC, BCL2 and pSTAT3 proteins and clinical data were evaluated. In our series, non-GCB DLBCL patients had significantly worse progression-free survival (PFS) and overall survival (OS), as calculated using the Choi, Visco-Young and Hans algorithms, indicating that any of these algorithms would be appropriate for identifying patients who require alternative therapies to R-CHOP. Whilst MYC abnormalities had no impact on clinical outcome in the non-GCB subtype, those patients with isolated MYC rearrangements and a GCB-DLBCL phenotype had worse PFS and therefore might benefit from novel treatment approaches.