Longitudinal genomic analysis of clonal hematopoiesis and tumor-associated mutations in acute myeloid leukemia Free

[INTRODUCTION] Clonal hematopoiesis (CH) refers to a clonal expansion of blood cells due to a genetic mutation in hematopoietic stem cells resulting in some degree of proliferative advantage. It is known that the presence of CH increases the risk of developing hematologic malignancies such as acute myelogenous leukemia (AML) and myelodysplastic syndrome (MDS). CH in patients with leukemia is known to persist after complete remission, but the prognostic impact of residual CH remains unclear. In this study, we aimed to elucidate the clinical significance of CH by analyzing longitudinal genomic data from patients with AML.

[METHODS] We recruited 372 AML patients whose genomic information was linked to detailed clinical information in the HM-SCREEN-Japan project in collaboration with multiple Japanese high-volume centers involved in hematology and oncology. Among the 39 cases (10.5%) with longitudinal data from two or more time points, we analyzed the genomic data of 36 cases with no missing clinical information. Known germline variants were excluded by referencing public databases such as gnomAD and jMorp. Subsequently, clonal hematopoiesis (CH) and tumor-associated mutations (MT) were distinguished according to the following criteria:

Variants that were detected at all time points during the clinical course and exhibited no more than a twofold change in variant allele frequency (VAF).

Variants that disappeared at any time point during follow-up or showed a VAF change exceeding twofold.

Based on the above definitions, the 36 cases were classified into three groups: (1) those with CH only (CH-only group), (2) those with MT only (MT-only group), and (3) those with both CH and MT (MT-with-CH group). The clinical characteristics of these groups were then compared in detail.

[RESULTS] The number of cases in the CH-only, MT-only, and MT-with-CH groups was 2, 21, and 13, respectively. The median age in each group was 77, 51, and 58 years. The number of patients classified as having adverse risk according to the ELN criteria was 0 (0%), 2 (10%), and 4 (31%) in the respective groups. The number of patients who underwent hematopoietic stem cell transplantation was 2 (100%), 5 (24%), and 7 (54%), respectively.

The median overall survival (mOS) for all 36 cases was 28.4 months [19.1–not reached (NR)]. When stratified by group, the mOS was NR in the CH-only group, 34.1 months [15.8–NR] in the MT-only group, and 21.5 months [16.2–24.8] in the MT-with-CH group.

Similarly, the median progression-free survival (mPFS) for the entire cohort was 11.7 months [7.3–27.2]. When stratified by group, the mPFS was 27.2 months [6.1–NR] in the MT-only group, 7.7 months [6.0–11.7] in the MT-with-CH group, and 0.6 months [0.6–NR] in the CH-only group.

To further explore the relationship between CH and MT, we performed a network analysis by integrating all CH–MT pairs observed across the cases. The results suggested that CH was relatively strongly associated with certain tumor-associated mutations, such as FLT3 and NRAS. When examining the frequency of gene mutations in each group, FLT3-ITD was observed in 6 cases (29%) in the MT-only group and 8 cases (62%) in the MT-with-CH group, while NRAS mutations were found in 0 cases (0%) and 3 cases (23%) in the respective groups.[CONCLUSION] In this study, we distinguished between clonal hematopoiesis (CH) and tumor-associated mutations (MT) by analyzing longitudinal genomic data from patients with AML. Although the median age was generally comparable between the MT-only and MT-with-CH groups, the latter tended to have a shorter median overall survival. A relatively strong association between CH and driver gene mutations such as FLT3 and NRAS was observed in network analysis, and these mutations were more frequently detected in the MT-with-CH group. These differences in genetic background may have contributed to the observed disparity in clinical outcomes. Although the number of available cases in this study was limited, our future aim is to accumulate a larger cohort to enable statistical quantification of the impact of CH on overall survival and relapse risk.