Where the clone is selected: stage-specific selection from hematopoietic stem cells to plaque macrophages in clonal hematopoiesis

Clonal hematopoiesis of indeterminate potential (CHIP) raises cardiovascular risk, yet the size of a clone in blood does not by itself explain how that clone contributes to disease. We argue that selection acts not once but at three dissociable checkpoints: competition among hematopoietic stem cells, myeloid lineage output, and entry, niche occupancy and local proliferation in tissue. Blood variant allele frequency (VAF), the measurement on which the clinical literature rests, is the product of the first two and therefore does not report the mutant fraction among stem cells; because lineage bias is gene-dependent, VAF-matched comparisons across drivers are not equivalent. Evidence from TET2 , DNMT3A , ASXL1 , JAK2 and Trp53 models shows that an advantage at one checkpoint need not carry to the next, and that the checkpoints separate gene by gene. At checkpoint 1, cardiovascular risk factors and tissue injury act on stem cells both directly and by remodeling the bone marrow niche, but whether cardiovascular stress selects mutant clones specifically through defined niche changes remains unresolved. At checkpoint 3, human cross-sectional data have not demonstrated preferential recruitment of mutant macrophages into cardiovascular tissue and cannot separate recruitment from retention, niche occupancy or local proliferation, and irradiated transplantation chimeras measure how a clone fills a vacated niche rather than how it competes for an occupied one. Existing pharmacology — canakinumab, colchicine, metformin — is read out at checkpoints 1 and 2 alone. Genotype-resolved, spatial and longitudinal measurements are needed to make the framework testable.