Background: Cytotoxic therapy increases the risk of clonal hematopoiesis (CH) defined by somatic mutations of cancer-associated genes in blood stem cells. CH is associated with secondary blood cancers and other adverse health events, although the minimum clone size at which these risks become evident and factors that could mitigate them are unknown. To explore these questions, we examined associations of CH patterns with demographic and treatment-related factors using highly sensitive, duplex UMI-based error-corrected sequencing (ECS) of whole blood DNA samples from breast cancer survivors enrolled in the Reach for Health (RFH) study, a 2x2 factorial randomized controlled trial of metformin and weight loss.

Methods: Whole blood DNA was isolated from banked samples collected at 2 time points (baseline and 6 months) from 332 RFH participants. Targeted ECS of 70 myeloid malignancy-associated genes was conducted to identify somatic mutations with variant allele frequency (VAF) >0.2%. Likely germline polymorphisms and variants with <5 mutant consensus read families were excluded. Analysis of mutation frequency, number, and type; VAF distribution and stability; and co-mutation were performed. Demographics and clinical data, including pre-enrollment cancer treatment, were correlated with mutation patterns.

Results: 1035 mutations in 37 genes were identified across 305 RFH participants (91.9%) of which 78 (23.5%) had a maximum VAF ≥2% and 28 (8%) with maximum VAF ≥10%. 826 mutations were detected at both time points with 102 and 107 found only at baseline or 6-months, respectively. Mutation VAF was highly correlated across time points (R2=0.99). The most frequently mutated gene was DNMT3A in 68% of RFH participants followed by TET2, PPM1D, TP53, and ASXL1. Mean number of mutations per individual was 3.1 (range 0-19), which increased with age. The mean number of mutations was significantly lower in RFH participants who prior to study enrollment received no cytotoxic therapy (1.4) compared to chemotherapy alone (2.6), radiation alone (2.6), or both (3.0) with p-values of 0.0467, 0.0093, and 0.0002, respectively. RFH participants who had received both chemotherapy and radiation were more likely to have mutations in PPM1D and TP53 compared to participants who had received neither treatment (Χ2 p-values <0.0001 and 0.014, respectively). Average clone size increased with age and was greatest among RFH participants who had received both chemotherapy and radiation. Analysis of CH and associations with study interventions and clinical outcomes is ongoing and will be presented at the meeting.

Conclusions: Our findings show that CH is extremely common among breast cancer survivors at VAF > 0.2%, and that prevalence of CH and mean number of mutations increases with age. Higher-risk mutations in PPM1D and TP53 were enriched in RFH participants who had received both chemotherapy and radiation, as were mutations with greater VAF.

Citation Format: Soo J. Park, Zhan (Jack) Rong, Laura Williamson, Timothy Luger, Emily Johnson, Sheri Hartman, Christopher J. Gibson, R. Coleman Lindsley, Dorothy D. Sears, Rafael Bejar. Clonal hematopoiesis profiles in breast cancer survivors using error-corrected sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1435.