A meta-analysis of HLA peptidome composition in different hematological entities: Entity-specific dividing lines and "pan-leukemia" antigens
Hematological malignancies (HM) are highly amenable targets for immunotherapeutic intervention and may be effectively treated by antigen-specific T-cell based treatment. Recent studies demonstrate that physiologically occurring anti-cancer T-cell responses in certain HM entities target broadly presented non-mutated epitopes. HLA ligands are thus implied as prime targets for broadly applicable and antigen-specific off-the-shelf compounds. With the aim of assessing the presence of common targets shared among different HM which may enable addressing a larger patient collective we conducted a meta-analysis of 83 mass spectrometry-based HLA peptidome datasets (comprising 40,361 unique peptide identifications) across four major HM (19 AML, 16 CML, 35 CLL, and 13 MM/MCL samples) and investigated similarities and differences within the HLA presented antigenic landscape. We found the cancer HLA peptidome datasets to cluster specifically along entity and lineage lines, suggesting that the immunopeptidome directly reflects the differences in the underlying (tumor-)biology. In line with these findings, we only detected a small set of entity-spanning antigens, which were predominantly characterized by low presentation frequencies within the different patient cohorts. These findings suggest that design of T-cell immunotherapies for the treatment of HM should ideally be conducted in an entity-specific fashion.