Unraveling the Transcriptional Landscape of Nodular Lymphocyte-Predominant Hodgkin Lymphoma and T-Cell/Histiocyte Rich Large B-Cell Lymphoma: Impact of Tumor Microenvironment and Checkpoint Gene Expression

Introduction: Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) is a rare B-cell malignancy with a paucity of malignant cells embedded in a diverse tumor microenvironment (TME). The histological composition of the TME is known to influence outcomes, with nodular B-cell rich TME (classical histology, Fan patterns A&B) associated with indolent clinical course, while T-cell infiltration or diffuse growth (variant histology, Fan patterns C-F) may resemble aggressive T/cell-histiocyte rich large B-cell lymphoma (THRLBCL) to which NLPHL can transform. Molecular features of NLPHL and THRLBCL remain to be discovered. To gain further insights into the biology of these lymphomas, we recruited NLPHL and THRLBCL cases as part of the Atlas of Blood Cancer Genomes (ABCG) initiative, a consortium consisting of 26 institutions.

Design: We collected comprehensive clinicopathological, treatment and outcome data from 154 NLPHL and 21 THRLBCL patients, with centralized pathology review performed by a panel of dedicated hematopathologists. Fan patterns of the NLPHL samples were documented in all samples where possible. RNA sequencing was performed on formalin-fixed paraffin-embedded (FFPE) diagnostic tumor samples (n=129 NLPHL; n=17 THRLBCL). Tumor-infiltrating immune cell compositions were enumerated using FARDEEP with signature matrix LM22 from CIBERSORT.

Results: Median age at diagnosis for the patients with THRLBCL and NLPHL was 54 years (interquartile ratio [IQR] 43-65) and 40 years (IQR 24-53), respectively. The majority of patients with NLPHL had limited stage (74%), whereas the patients with THRLBCL predominantly had advanced stage at diagnosis (95%). In both diseases, most patients were males (NLPHL, 69%; THRLBCL, 76%). Spleen involvement was present in 11 NLPHL patients (8%) and histological growth pattern was documented in 112 patients (73%). Out of those, 47 (42%) presented with variant histology and 65 with classical histology. Variant histology was enriched in NLPHL patients with splenic involvement.

During a median follow-up of 6.4 years (IQR 3.8-10.1), 12 NLPHL patients and 9 THRLBCL patients died. Patients with THRLBCL had worse 10-year overall survival (OS, 31% vs 89%; p<0.001) and progression-free survival (PFS, 30% vs 66%; p=0.002) compared to patients with NLPHL. Moreover, NLPHLs with variant histology had inferior 10-year OS compared to NLPHLs with classical histology (81% vs 91%; P=0.03; Figure 1).

According to in silico immunophenotyping, THRLBCL cases had the highest median proportions of M1-macrophages (11.0%) and CD8+ T-cells (17.0%), followed by NLPHLs with variant histology (M1-macrophages, 6.5%; CD8+ T-cells, 14.5%), with the lowest proportion observed in NLPHLs with classical histology (M1-macrophages, 5.2%; CD8+ T-cells, 10.9%). In contrast, the median proportions of naïve B cells followed an opposite trend across subtypes (NLPHL classical histology, 15.5%; NLPHL variant histology, 11.0%; THRLBCL, 2.0%), whereas the median proportion of CD4+ T-cells was similar across both lymphomas (NLPHL classical histology, 33.9%; NLPHL variant histology, 33.9%; THRLBCL, 32.4%).

Despite differences in the overall transcriptomic landscape, gene expression patterns posited THRLBCL and NLPHL as comprising a biological continuum. THRLBCL shared more similarities with NLPHL with variant histology, while NLPHLs with classical histology exhibited upregulation of genes related to cell growth and proliferation through the MYC pathway (P<0.001). Finally, we identified a strong association between THRLBCL and upregulation of checkpoint genes (P<0.001) and gene expression related to inflammatory response (P<0.001) (Figure 2). Furthermore, NLPHLs with a high checkpoint score (above the 75th percentile) were associated with spleen involvement (P<0.001) and worse outcome (5-year PFS 61% vs 87%; P=0.02).

Conclusion: Our study provides the largest comprehensive clinical and transcriptomic analyses of NLPHL and THRLBCL to-date. The findings suggest that the inferior outcomes observed in THRLBCL and in a subset of NLPHL are attributable to a higher degree of inflammation in the TME, increased expression of checkpoint genes, and a high proportion of macrophages.