Distinct immune profiles among patients with primary effusion lymphoma and other kaposi sarcoma herpesvirus-associated diseases

Abstract

Abstract Introduction Kaposi sarcoma herpesvirus (KSHV), also known as human herpesvirus 8 (HHV8), causes Kaposi sarcoma (KS), multicentric Castleman disease (MCD) and primary effusion lymphoma (PEL). KS, MCD and PEL are distinct pathologic entities. KSHV-associated inflammatory cytokine syndrome (KICS) is associated with KSHV viremia and requires the exclusion of MCD and PEL. These KSHV-associated diseases (KAD) can occur alone or concurrently in people with HIV or other individuals with other immunodeficiencies (Cesarman E, et al. Blood 2022). Circulatory cytokine dysregulation is a key component of MCD, PEL and KICS pathogenesis. However, there are limited data on the immune profiles and how these differ between patients with different KADs at the time of active disease. Methods We evaluated baseline characteristics and flow cytometry data from 63 participants with active KAD who were enrolled on prospective treatment studies [KS and MCD or KS and KICS: pomalidomide with liposomal doxorubicin (NCT02659930), PEL+/-KS+/-MCD: EPOCH-R2 (NCT02911142), or KICS+/-KS (NCT01419561)]. The T-cell, myeloid-derived suppressor cell (MDSC) and monocyte populations were analyzed using peripheral blood mononuclear cells (PBMCs) collected at baseline prior to study treatment. KSHV-specific T-cell responses at baseline were evaluated via a whole KSHV proteome interferon-gamma ELISpot assay employing ~7,500 overlapping 15mer peptides into peptide pools that represented 83 KSHV open reading frames. Assays were performed using fresh PBMCs that were collected. A response was considered positive if > 40 spot-forming units (SFU)/106 cells were detected. The breadth of T-cell responses was determined by the number of antigens eliciting interferon-gamma responses, and the intensity of responses was determined by the total SFU/106 cells at each time point. Wilcoxon rank sum tests were used to analyze significant differences between KAD for flow cytometry and KSHV-specific T-cell responses. Results Among 63 participants with KAD, 20 participants had KS alone, 21 had KICS and KS (KICS+KS), 5 had MCD and KS (MCD+KS), and 17 participants had PEL+/-KS+/-MCD (PEL group). In the PEL group, 65% of participants had concurrent KS and 35% had active MCD at baseline. All but two participants had HIV infection, which was well-controlled (median HIV viral load 33 copies/ml). When comparing across the different KAD groups, flow cytometry differences were most notable between those within the PEL group as compared to those with KS alone. Participants in the PEL group had higher expression of PDL-1+ in monocytic (CD14+CD15-, P=0.02) MDSCs compared to those with KS alone. At baseline, participants in the PEL group had a lower percentage of CD16+CD14- cells, termed non-classical monocytes, compared to participants with KS alone (P=0.04). In an analysis between those with MCD+KS and KICS+KS, participants with KICS+KS presented with elevated classical monocytes (CD14+CD16-, P=0.02), while participants with MCD+KS displayed elevated non-classical monocytes (CD16+CD14-, P=0.03). With regards to T-cell characterization, there were no differences noted in the CD4+ and CD8+ T cell counts by KAD diagnosis. Participants in the PEL group had elevated CD8+ PD1+ expression compared to participants with KS alone (P=0.0007). Overall, analyses of the KSHV-specific T-cell responses across all groups were heterogenous with no evidence of immunodominance. There were no differences in the breadth or intensity of KSHV-specific T-cell responses between KAD groups at the time of active disease. Conclusions Although KSHV-specific T-cell responses were similar across KADs, immune profiling revealed distinct differences by KAD diagnosis. In particular, PEL was associated with increased markers of immune exhaustion and inflammation, including elevated PD-L1+MDSCs and CD8+PD-1+T cells. Monocyte subsets also distinguished MCD+KS from KICS+KS, suggesting that innate immune dysregulation differs across these conditions. These findings support the role of immune profiling in understanding KAD pathogenesis to identify disease-specific therapeutic targets.