Abstract T027

Distinct cell state ecosystems for nodular lymphocyte-predominant Hodgkin lymphoma

Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) is a rare lymphoma, and the microenvironment is characterized by a paucity of lymphocyte-predominant (LP) cells surrounded by abundant immune cells. Few studies have explored the microenvironment, and recent single cell sequencing techniques and atlases may shed light on the cell state phenotypes and their prognostic implications for NLPHL. Here we develop a NLPHL-specific cell type gene expression signature matrix with subsequent utilization in a machine learning framework called EcoTyper to identify 34 distinct cell states across 14 cell types for 171 cases of NLPHL. We found evidence of CD8 T-cell exhaustion, M2 polarized macrophages, immune checkpoint genes expressed by follicular T-cells, and three distinct LP cell states that do not segregate with morphologic variant patterns. These cell states co-occur in 3 LP EcoTypes (LPE1 [46% of cohort], LPE2 [25%], and LPE3 [29%]) with LPE3 portending worse freedom from progression in the training (n=109, HR=2.74, P=0.01) and validation cohorts (n=62, HR=2.16, P=0.003) after multivariable adjustment for the LP-international prognostic score. Further, LPE3 appears predictive of worse freedom from progression after single modality but not combined modality therapy in the training and validation cohorts. Using single-nucleus RNA-seq and spatial transcriptomics, we validate the co-occurrence and co-localization of these cell states, respectively. Finally, we reconstructed the B-cell and T-cell receptor repertoires, finding lower diversity for relapse and LPE3 cases. Collectively, identify a new classification of tumor tissue for NLPHL instead of the morphologic variant patterns and show that most patients with NLPHL have a favorable prognosis with a microenvironment characterized by checkpoint immunosuppression and exhausted T cells supporting future trials exploring de-intensification approaches with immune checkpoint inhibitors. Conversely, patients with LPE3 may benefit from upfront combined modality therapy.