The Role of B Regulatory Cells in the Immunological Escape of Tumor Cells in Hodgkin Lymphoma

Abstract Hodgkin and Reed-Sternberg (HRS) cells in classical Hodgkin Lymphoma (HL) are surrounded by a rich inflammatory infiltrate which aids in their survival and escape from cytotoxic CD8+ T cells (CTLs) and Natural Killer cells (NKs). Within HL environment, T regulatory cells (Tregs) directly suppress the activity of CTLs and NKs, enhancing the tolerance against HRS cells. B regulatory cells (Bregs) have been shown to support the differentiation of Tregs through IL-10 production; thus, we hypothesized that they could have a role in the pathophysiology of HL. We evaluated 30 classic HL patients (M/F: 18/12; median age, 31 years, range 15-62) and 5 healthy controls (HC) for circulating peripheral blood (PB) Bregs, Tregs, CTLs, NKs, and NKTs. Twenty-four of them were new-diagnosed patients (NwHL) and 6 received a previous diagnosis of HL but were in complete remission (CR) for more than 12 months (PvHL). NwHL patients were divided according to the International Prognostic Score (IPS) and the Ann-Arbor Staging System. All subjects were treated following the ABVD protocol (doxorubicin 25 mg/m2, bleomycin 10 U/m2, vinblastine 6 mg/m2, and dacarbazine 375 mg/m2). Flow cytometry was performed on heparinized PB samples with a 5-color Beckman Coulter Cytomics FC500 flow cytometer. Breg (CD19+CD24+), Treg (CD3+CD4+CD25+), CTL (CD3+CD8+), NK (CD3-CD56+), and NKT (CD3+CD56+) levels were measured simultaneously with the PET/CT evaluations, ie at diagnosis, at the end of the second ABVD administration, at the end of treatment, and at 6 and/or 12 months off-therapy. Moreover, Breg levels were compared to IPS and Ann-Arbor staging groups, and also were correlated to the erythrocyte sedimentation rate (ESR) and to the absolute lymphocyte count (ALC). We found decreased circulating Bregs in NwHL and PvHL patients compared to controls (0.39% vs 0.875% vs 1.813%, respectively, p<0.0001). In addition, we found decreased CTLs in NwHL compared to PvHL and HC (2.815% vs 4.057% vs 5.780%, respectively, p=0.0228). Thus, all HL patients showed lower numbers of Bregs, but there were no differences between patients of different IPS or stage. Besides, lower Breg levels were correlated with higher ESR values (r2=0.3193, p=0.0117), while no correlation was found for ALC (r2=0.0414, p=0.3898). The levels of Tregs, NKs, and NKTs were not significantly different between patients and controls at diagnosis. As a result, the greater reduction of Bregs compared to Tregs in PB caused a larger Bregs/Tregs ratio in NwHL and PvHL patients compared to controls (1.014 and 1.003 vs. 5.975, p=0.0094). Further reduction of circulating Bregs was observed in 58% of NwHL patients at the end of the second ABVD administration for patients experiencing both CR and partial remission (PR) (based on PET/CT scans) and in 80% at the end of treatment (71% CR, 29% PR). Interestingly, post-treatment Breg levels of NwHL were initially lower than that of PvHL patients and controls (NwHL post-treatment 0.4062% vs PvHL 0.8750% vs HC1.813%, p=0.0030) but normalized after 6 months off-therapy (1.444%, p=0.6189). Similar trends were observed for CTLs (NwHL post-treatment 4.615% vs NwHL 6 months off therapy 5.603% vs PvHL4.057% vs HC 5.780%, p=0.7558) and Bregs/Tregs ratio (0.9613 vs 5.975, p=0.0678). These data suggest that the B cell depletion phase during chemotherapy may interrupt the positive feedback between B and T cell compartments and the normalization of Bregs and CTLs after treatment may be linked to the restoration of a normal immune response, supporting by the achievement of CR in all patients. Our preliminary data suggest involvement of Bregs in the escape and survival of HRS cells during active disease. Peripheral blood may mirror disease activity in lymphoid tissues. Thus, the decrease of circulating Bregs may be related to the recruitment of these cells to the tumor site; amplification of the Bregs/Tregs ratio may result in a greater Breg-dependent Treg activation with subsequent inhibition of CTL and NK function. Additionally, the normalization of Bregs and the Bregs/Tregs ratio after chemotherapy could be used to predict disease remission. While larger prospective studies are required to validate these results, we present intriguing evidence of the involvement of Bregs in the pathophysiology of HL. Disclosures No relevant conflicts of interest to declare.