Session: 621. Lymphomas: Translational—Molecular and Genetic: Single Cell, Microenvironment, and Treatment Resistance
Hematology Disease Topics & Pathways:
Research, Translational Research
Consecutive HIV-infected patients or transplant recipients with treatment-naïve NHL were included and compared with immunocompetent patients with diffuse large B-cell lymphomas (DLBCL). A whole exome sequencing (WES) was carried out in parallel on tumor and on PBMC for TMB assessment and a whole genome RNA sequencing was performed on frozen tumor biopsies. The neoepitopes derived from the somatic tumor variants (called in WES and RNAseq data) were predicted in silico using a bioinformatic pipeline including pVACseq and NetMHC and filtered according to: VAF≥10%, gene expression≥1 TPM and affinity binding score ≤500 nM. Neoepitopes specific T cell responses were assessed by ELISPOT assays after a 10 days PBMC co-culture with peptides from the Top 50 best predicted autologous neoepitopes (positivity defined as spot forming cells >50/106 cells). For the TME study, the profiling of cell type abundance and the T-cell receptor (TCR) sequencing were performed with deconvolution tools (EPIC) and MiXCR respectively. All statistical tests used R.
Sixty-seven patients were included so far: 33 post-transplant lymphoproliferative disorders (PTLD) and 16 HIV-positive NHL, 70% of whom having DLBCL, compared to 18 immunocompetent (IC)-DLBCL. Diseases were systemic in 73% and localized in CNS in 27% cases. The median age was 58 years (range 21-85) and 70% were male. Immuno-molecular data are available for 57 patients so far. The TMB was 3.7/Mb for all patients but was lower in EBV-positive NHL (2.7/Mb) compared to the negative ones (4/Mb) (p=0.012, t-test) (Fig 1) without difference according to the ID status. The overall median number of neoepitopes per tumor predicted from the non-immunoglobulin (Ig) variants was 113 (range 11-720) but was lower in EBV-positive NHL (49 vs 243, p= 0.04, t-test), correlating with the TMB (r=0.8, p<0.0001). Most neoepitopes were MHC-class II restricted (ratio MHC-class II/ MHC- class I= 4.1) independently of the EBV status. The median number of neoepitopes predicted from the Ig heavy chain (IgH) genes was 17.5 (range, 6-36) with a MHC-class II/MHC-class I ratio of 1.4. Neoepitopes specific T cell responses were detectable among 71% cases out of 14 patients tested so far, independently of ID status or CNS localization (Fig 2). All 3 positive responses against Ig-derived neoepitopes out of 5 tested cases were directed against the IgH, MHC-class II restricted and immunodominant compared to those against the non-Ig neoepitopes. There was no shared immunogenic neoepitopes between patients. The TME study showed a higher frequency of CD4+ and CD8+ T cells mean frequency in EBV-positive compared to EBV-negative NHL (11% vs 4% and 10% vs 2% respectively, p<0.03).The intra-tumoral TCR repertoire distribution was mostly polyclonal with a median frequency of dominant clones of 8% (range 1-61) and a lower repertoire diversity, assessed on the ratio of clonotypes number/ total TCR-β reads number, in EBV-positive NHL (29) than in EBV-negative ones (37) (p=0.02), without difference according to the immune status.
Our data demonstrate that both immunogenomics and TME of NHL from immuno-compromized patients are preferentially influenced by the EBV status, with lower numbers of tumor neoepitopes correlating with lower TMB, and a lower intra-tumoral TCR diversity in EBV-positive versus EBV-negative NHL. We further demonstrated that anti-tumor immune responses can be mounted despite ID status or CNS localization, especially against MHC-class II presented Ig-variants, and might be targeted for therapeutic strategies.
Disclosures: Leblond: Abbvie, Beigene, Roche, Amgen, Lilly AstraZeneca, Janssen, Gilead: Honoraria, Membership on an entity's Board of Directors or advisory committees. Sylvain: gilead: Consultancy, Honoraria; Novartis: Consultancy, Honoraria.
See more of: Oral and Poster Abstracts