Clonal Dynamics and Clinical Implications of Post-Remission Clonal Hematopoiesis in Acute Myeloid Leukemia (AML)

Introduction

While clonal hematopoiesis (CH) can precede the development of AML, preleukemic CH can persist after attaining remission. In rare cases, non-leukemic CH can also emerge after remission. Long-term clonal dynamics of persistent CH or emerging CH (together, post-remission CH) in AML patients during consolidation therapies or maintenance therapies is not well understood. Furthermore, the clinical implications of post-remission CH have not been systematically studied. Since AML patients receive various types of post-remission consolidation therapies, including allogeneic stem cell transplant (allo-SCT), analysis of clonal dynamics of post-remission CH offers unique opportunity to study the response of CH to therapies. Such knowledge may be translated into the development of therapeutic strategy targeting CH. Here, we analyzed multiple longitudinal samples from AML patients after remission and analyzed clonal behavior of post-remission CH during consolidation and maintenance therapies.

Methods

We studied 164 AML patients who attained complete remission (CR) after induction therapies and analyzed baseline and CR marrow by targeted deep sequencing of 295 genes (median 403x depth). Among the patients who were identified to have post-remission CH, we analyzed multiple longitudinal marrows and tracked clonal dynacmis of post-remission CH. We defined "persistent CH" when patients had mutations that were originally detected in AML and persisted after CR with variant allele frequency (VAF) > 2.5%. On the other hand, we defined " emerging CH" when the patients had new mutations arising after attaining CR.

Results

Among the 164 AML patients, 79 (48%) patients were found to have post-remission CH at first CR. Of those, we were able to analyze multiple post-remission marrows in 54 patients (median 6 samples/patient, IQR 4-7.75),of which 44 (66%) had persistent CH and 1 (1.9%) had emerging CH, and 9 (17%) had both. The median age of these 54 patients cohort was 59 (IQR: 51-68). 34 (63%) patients attained CR after intensive chemotherapies (IC, idarubicin with high-dose cytarabine based), whereas 20 (37%) patients received low-intensity chemotherapies as induction (hypomethylating agent [HMA]-based N=3, low dose cytarabine-based N=17). All 34 patients treated with IC induction subsequently received consolidation chemotherapies with high-dose cytarabine (HDAC) regimens for median 2 cycles (IQR 2-3.25). Among the 20 patients treated with low-intensity chemotherapies, all patients received consolidation therapies with HMA-based regimen. A subset of patients received maintenance therapies under clinical trials: lenalidomide (N = 1), venetoclax (N = 2), and nivolumab (N = 1). 29 patients underwent allo-SCT (23 with myeloablative conditioning [MAC], and 6 with reduced-intensity conditioning [RIC]). The most frequently mutated genes for persistent CH were DNMT3A (N = 23), followed by TET2 (N = 17), IDH2 (N = 10), and SRSF2 (N = 10). For emerging CH, we observed emergence of JAK2 (N = 2), RUNX1 (N = 2), TET2 (N = 2), BRAF (N = 1), BCOR (N = 1), TP53 (N = 1), KDM6A (N = 1), and NRAS (N = 1).

Longitudinal clonal analysis revealed that post-remission CH persisted in 52 of 54 (96%) patients during consolidation and maintenance therapies. Neither cytarabine based chemotherapies, HMA, venetoclax, lenalidomide, and nivolumab, reduced the VAF of post-remission CH, except in 2 (4%) patients, in which post-remission CH (IDH2 and TET2 mutations) were cleared with high dose cytarabine regimen. Although consolidation or maintenance therapies did not reduce post-remission CH, 17 out of 19 (89.4%) patients who underwent for allo-SCT had clearance of post-remission CH with both MAC and RIC.

Overall, post-remission CH did not impact the risk of relapse or overall survival. Furthermore, post-remission CH did not affect long-term blood counts (neutrophil counts, hemoglobin, and platelet counts).

Conclusion

Post-remission CH persisted long-term in AML patients during consolidation and maintenance therapies without having significant impact on blood counts, relapse, and survival. Therapies with cytarabine-based chemotherapies and HMA did not affect the clonal size of post-remission CH, suggesting that these therapies may not be useful to treat CH in pre-leukemic context either. The novel therapeutic approach is warranted to target CH.