P1227: comprehensive molecular subtyping of diffuse large b-cell lymphoma cell lines and association with tafasitamab activity

Topic: 20. Lymphoma Biology & Translational Research Background: Tafasitamab (tafa), an anti-CD19 immunotherapy that enhances antibody dependent cellular cytotoxicity (ADCC) and phagocytosis, achieved a 57.5% objective response rate in combination with lenalidomide in the phase 2 L-MIND trial (NCT02399085), and received accelerated US approval and conditional authorization in Europe for treatment of patients with relapsed or refractory diffuse large B-cell lymphoma (DLBCL) ineligible for autologous stem cell transplant. DLBCL is a heterogeneous disease with subtypes characterized by cell of origin (COO) and/or driver mutation signatures; these are associated with differential response to standard immunochemotherapy. Identification of biomarkers predictive of response and more efficacious combination therapies with tafa may provide further benefit to patients with limited or incomplete responses. Aims: We performed a comprehensive characterization of DLBCL cell lines and examined whether tafa activity is associated with COO, specific molecular subtypes, and patients’ immune fitness. Methods: An extensive literature review was performed to assign DLBCL cell lines to COO subtypes—activated B-cell like (ABC) or germinal center B-cell like (GCB). Genomic profiles of DLBCL cell lines were downloaded from DepMap version 22Q1. The weights for the COO model were computed using a 15-gene expression signature (Scott et al. Blood. 2014) on frozen tissue data from Lenz et al (N Engl J Med. 2008) to score each DepMap DLBCL cell line. Additionally, key mutations associated with DLBCL molecular subtypes (MCD, N1, A53, BN2, ST2, EZB) defined in Wright et al (Cancer Cell. 2020) were compiled from DepMap data. Sensitivity data on DLBCL cell lines screened against BTK inhibitors (AVL-292, acalabrutinib, ibrutinib) were obtained from the PRISM repurposing dataset. Results: Genomic profiles of DLBCL cell lines included gene expression (n=30), copy number (n=21), mutation (n=32), and fusion data (n=30). Cell lines were characterized as COO subtypes ABC (n=4), GCB (n=21), and unclassified (n=5). Although there was general agreement between COO signature and the literature, we noted inconsistencies in 2 DLBCL cell lines. Interestingly, with exception of OCI-Ly3, all profiled cell lines had features of multiple molecular subtypes. In vitro, tafa ADCC was independent of DLBCL subtypes in the cell lines evaluated, whereas BTK inhibitor activity was enriched in cell lines carrying genetic alterations associated with MCD and N1 subtypes. In addition, tafa activity did not depend on TP53 status. To extend COO analyses in cell lines and account for impact of patient immune fitness on response, we tested tafa activity in an ex vivo ADCC assay using peripheral NK cells collected at baseline from patients in the L-MIND study. No strong correlation between tafa-mediated ADCC and COO was observed, although NK cells from patients with GCB tumors demonstrated more activity in this assay. Interestingly, patients with higher peripheral NK cell counts at baseline had increased PFS, suggesting that patients’ immune fitness contributes to response to tafa treatment. Summary/Conclusion: Consistent with L-MIND clinical data, tafa was active in both GCB and non-GCB DLBCL cell lines. Additionally, tafa activity was not associated with a specific molecular subtype or mutations in TP53. Biomarker analyses are ongoing in tafa clinical trials (topMIND [NCT04809467], frontMIND [NCT04824092], firmMIND [NCT05429268]) to confirm these findings and provide further insight into genomic and immune parameters that may play a role in response to tafa. Keywords: B cell lymphoma, Immunotherapy, Cancer, CD19