Clonal Hematopoiesis Across the Cancer Treatment Continuum: A Scoping Review of Clonal Selection, Evolution, and Clinical Outcomes

Background: Clonal hematopoiesis (CH) is characterized by the proliferation of blood cell clones that possess acquired genetic modifications. While therapy-related CH refers to CH detected in the setting of prior or current treatment in cancer populations, therapy-selected CH demands demonstration that treatment preferentially expanded or maintained an extant clone; therapy-induced CH requires evidence of de novo treatment-associated mutation supported by serial sampling and mutational-signature analyses. Objective: To map evidence distinguishing therapy-related, therapy-selected and therapy-induced CH across cancer treatment, including clonal dynamics, genomic features, clinical outcomes and surveillance gaps. Methods: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidance, PubMed/MEDLINE, Embase, Scopus, Web of Science Core Collection, Cochrane Library and Google Scholar were searched from inception through 1 June 2026, and updated on 15 September 2026. Primary evidence mapping was limited to original human studies. Reviews and conceptual sources were used only for context or reference tracking and were ineligible for the primary evidence map. Narrative synthesis was employed as critical variation existed in sequencing platforms, variant allele frequency thresholds, timepoints of sampling, filtering strategies, and outcome definitions. Results: Of 684 records identified, 213 duplicates were removed and 471 records were screened. Across both pathways, 121 full texts were assessed, 87 excluded and 34 primary reports mapped; contextual citations were outside the included set. Contextual references were used for definitions and interpretation and were excluded from the included-report count. Six domains were deductively established and inductively refined during charting: terminology; therapy-associated mechanisms of clonal selection/expansion; genomic architecture; longitudinal clonal dynamics; clinical consequences; and study design and methodological heterogeneity. Treatment-associated selection or expansion of pre-existing TP53 and PPM1D clones was the most consistently reported mechanism of therapy-related CH, while CHEK2 was inconsistently detected. Pediatric cohorts and mutational-signature analyses also supported de novo mutagenesis with therapy in select contexts. Progression to therapy-related myeloid neoplasm was the most consistent outcome across studies; however, cytopenia, CAR-T-therapy-associated toxicity, cardiovascular events and mortality were inconsistently reported. Conclusion: Current evidence supports a spectrum of mechanisms ranging from selection of pre-existing clones to therapy-induced de novo mutation. Existing literature cannot support universal CH screening recommendations or guide treatment decisions. Prospective serial analyses harmonized on definition, employing sensitive baseline assays with paired germline controls and tumor-contamination checks, are needed to justify clinical implementation and assessment of fixed outcomes.