Somatic mutations, hematopoietic mosaicism, and clonal dynamics in immune-mediated AA. (A) From the polyclonal T-cell repertoire, some clonal T cells specific for some antigen expressed on HSPCs (see “Immune synapsis”) may expand, leading to an oligoclonal antigen-driven T-cell response. (B) The immune synapsis: T cells may recognize through their T-cell receptor–specific antigens, presented on (some) HSPCs within either HLA alleles (peptidic epitopes) or HLA-like molecules (for lipidic epitopes; this is the case as with the GPI anchor presented within CD1d).^6,7  (C) Pathogenic T-cell clones may exert T-cell–mediated cytotoxicity over many HSPCs (via the immune synapsis depicted in the inset), eventually leading to oligoclonal hematopoiesis.^2,4  (D) Different somatic mutations may stochastically occur within individual HSPCs; because of the underlying HSPC oligoclonality, any neutral mutation carried by surviving HSPCs becomes evident (Darwinian selection).⁹  Individual mutations leading to specific functional phenotypes shape the subsequent hematopoietic mosaicism and clonal dynamics through different mechanisms, including immune escape, HSPC fitness, or proliferative advantage. In the absence of somatic mutations, HSPCs may undergo exhaustion (first quadrant). Expansion of clones escaping the immune response may occur through different mechanisms, such as GPI-deficient cells (PNH, second quadrant)⁶  or functional loss of HLA due to 6pLOH (third quadrant) or to other structural HLA gene mutations (ie, B4002⁻ cells).¹  Other somatic mutations may contribute to clonal dominance through distinct specific mechanisms (fifth quadrant)¹⁰ : true malignant transformation for splicing genes, survival/growth advantage, or increased HSPC fitness for epigenetic mutations; unknown (possibly immune escape?) for BCOR-BCORL1 mutations. Ag, antigen; TCR, T-cell receptor. Professional illustration by Somersault18:24.