Lymphoma Associated Macrophages Predict Survival in Uniformly Treated Patients with Classical Hodgkin Lymphoma.
Pedro Farinha, Sofia Rodrigues, Teresa Fernandes, Alexandra Monteiro, Rodrigo Lopes da Silva, J Salvador, Rita Gerivaz, Fatima Costa, Gilda Ferreira, Carla Lopes, Teresa Lacerda, Jose Ferreira, Isabel Costa, Aida Botelho de Sousa. Centro Hospitalar Lisboa Central, Portugal
Background: Classical Hodgkin Lymphoma (CHL) current therapy cures about 80% of patients. Yet, 20% still dies from progressive disease and many are overtreated. Therapeutic decisions are based on clinical criteria but these imperfectly predict disease behavior and treatment response. A recent gene and protein expression study highlighted the role of microenvironment demonstrating Lymphoma Associated Macrophages (LAM) content to correlate significantly with CHL treatment outcome (Steidl et al. NEJM 2010). Within tumors, macrophages may promote tumor growth and angiogenesis. We aimed to validate these results using whole-sections (WS) of routine diagnostic formalin-fixed-paraffin-embedded (FFPE) biopsies from a cohort of uniformly treated patients in a single institution.
Design: Between Nov1996 and Dec2009, 212 consecutive CHL patients were enrolled on a prospective study receiving Stanford V chemotherapy (mechlorethamine, doxorubicin, etoposide, vincristine, vinblastine, bleomycin, and prednisone) plus involved-field radiotherapy in bulky disease. All patients were treatment naïve and had 15y to 69y. Evaluable FFPE blocks of the diagnostic biopsies were available in 153 patients. LAM content was assessed using WS stained with anti-CD68 (KP1) antibody. Modified Steidl et al scoring criteria (1 <5% LAM; 2 <25% LAM and 3 > 25% LAM) was used. OS and PFS univariate and multivariate analyses were done using SPSS software.
Results: The median age was 29y with median follow-up of the living patients of 6.4y. Clinically, 77 patients had stage IA/IIA, 26 had IB/IIB and 50 III/IV 28%. The estimated 10-year OS and PFS were 85% and 78%, respectively. The IPI (0/1 vs. 2/3/4) was predictive of OS (p=0.01) and PFS (p=0.03). Histological subtypes included 122 nodular sclerosis, 18 mixed celularity, 11 lymphocyte rich and 2 CHL, NOS. There were 54 cases with <5% infiltrating LAM, 84 with <25% and 14 cases with >25%. LAM significantly predicted OS (p=0.003) and PFS (p=0.001). Importantly, cases with limited stage disease (IA/IIA) and low LAM content (<5%) had 100% OS and 97% PFS. A Cox multivariate model with IPI and LAM content showed only LAM to be independent predictor of PFS (p =0.007) and borderline significant for OS (p=0.07).
Conclusions: Increased LAM content is an independent adverse prognostic factor of OS and PFS in CHL. These results using a uniformly treated cohort and WS biopsy are similar to Steidl et al's and reinforce LAM as an important biomarker that maybe use in risk stratification and therapy planning.
Tuesday, March 1, 2011 1:00 PM
Poster Session IV # 154, Tuesday Afternoon