Macrophages have been implicated in the pathogenesis of classical Hodgkin lymphoma (cHL) and have been suggested to have a negative impact on outcome. Most studies addressing the role of macrophages in cHL have relied on identification of macrophages by generic macrophage antigens, e.g., CD68. We have therefore conducted an in situ analysis of macrophage polarization in a series of 100 pediatric cHL (pcHL) cases using double staining immunohistochemistry, combining CD68 or CD163 with pSTAT1 (M1-like) or CMAF (M2-like). M1- or M2-polarised microenvironment was defined by an excess of one population over the other (>1.5). Expression of STAT1 and LYZ genes was also evaluated by RT-qPCR. Patients 1.5 was associated with higher numbers of CD68+pSTAT1+ (P=0.025) and CD163+pSTAT1+ macrophages (P 1.5 was associated with better OS (P= 0.037). In conclusion, macrophage polarization in pcHL correlates with prevalent local T cell response and may be influenced by the EBV-status of neoplastic cells. Besides, M1-like and M2-like macrophages displayed differential effects on outcome in pcHL.

Epstein-Barr virus (EBV) is associated to the etio-pathogenesis of an increasing number of tumors. Detection of EBV in pathology samples is relevant since its high prevalence in some cancers makes the virus a promising target of specific therapies. RNA in situ hybridization (RISH) is the standard diagnostic procedure, while polymerase chain reaction (PCR)-based methods are used for strain (EBV type-1 or 2) distinction. We performed a systematic comparison between RISH and PCR for EBV detection, in a group of childhood B-cell Non-Hodgkin lymphomas (NHL), aiming to validate PCR as a first, rapid method for the diagnosis of EBV-associated B-cell NHL. EBV infection was investigated in formalin fixed paraffin-embedded tumor samples of 41 children with B-cell NHL, including 35 Burkitt's lymphoma (BL), from Rio de Janeiro, Brazil, by in situ hybridization of EBV-encoded small RNA (EBER-RISH) and PCR assays based on EBNA2 amplification. EBV genomes were detected in 68% of all NHL. Type 1 and 2 accounted for 80% and 20% of EBV infection, respectively. PCR and RISH were highly concordant (95%), as well as single- and nested-PCR results, allowing the use of a single PCR round for diagnostic purposes. PCR assays showed a sensitivity and specificity of 96% and 100%, respectively, with a detection level of 1 EBV genome in 5,000–10,000 EBV-negative cells, excluding the possibility of detecting low-number EBV-bearing memory cells. We describe adequate PCR conditions with similar sensitivity and reliability to RISH, to be used for EBV diagnostic screening in high grade B-NHL, in "at risk" geographic regions.

Hodgkin lymphoma (HL) shows a bimodal distribution with a first peak in developing countries during childhood. The causative role and prognostic significance of Epstein-Barr virus (EBV) association in patients with HL is controversial. Our aim was to perform a comparative study of EBV association in 2 Latin American pediatric HL series, and to correlate it with patient's survival. Epstein-Barr encoded RNAs in situ hybridization and latent membrane protein 1 immunohistochemistry were performed on formalin-fixed, paraffin-embedded HL biopsies from 176 pediatric patients from 2 public institutions from Argentina and Southeast Brazil. Mixed cellularity subtype was prevalent in Argentine HL (Arg HL) (52%) and nodular sclerosis subtype in Brazilian HL (BR HL) (83%). EBV expression was detected in 52% of cases, namely 54% Arg HL and 48% Br HL. EBV was significantly associated with mixed cellularity subtype in both populations. In Arg HL, EBV positivity was significantly higher in patients ≤10 years (P=0.0011). Event-free survival did not attain statistical significance neither in Arg HL (P=0.5317), nor in Br HL (P=0.8321). Our results do not support EBV association stated for pediatric HL in developing countries. Correlation of younger age with EBV infection only in Argentine patients might be related to a different age background. In our pediatric series, EBV status cannot be used as prognostic factor.