Clonal Evolution of Therapy-Related Myeloid Neoplasm Analyzed Sequentially By Targeted Deep Sequencing Using Bone Marrow Cells

Background:

Therapy-related myeloid neoplasm (tMN) is a late complication of cytotoxic chemotherapy and radiotherapy. Because patients with malignant lymphoma (ML) tend to receive various types of chemotherapy, the developmental mechanism of tMN remains unclear.In myelodysplastic syndrome (MDS), clonal hematopoiesis of indeterminate potential (CHIP) has recently been consideredas a disease background,and preleukemic mutations have been detected as CHIP in peripheral blood samples of some patients with tMN before chemotherapy. Nevertheless, the clonal evolution from CHIP-related genetic modifications to tMN remains to be validated by the sequential analysis of bone marrow (BM) samples.

Materials and methods:

We performed targeted deepsequencing (TDS) using BM samples and ML tissues of five patients with tMN after treatment for ML between 2005 and 2017 to examine the sequentialclonal evolution of tMN. For DNA sequencing with MiSeq, we used the Myeloid Neoplasms Panel with 50 known mutation target genes of hematopoietic diseases. This study was approvedby the Institutional Review Board of Tokyo Medical University (no. 2151).

Results:

TDS revealed variedclonal evolution patterns in three patients. In patient 1, a small clone having NRAS, SETBP1, and SMC1A mutations was detectedat the diagnosis of diffuse large B-cell lymphoma [DLBCL; variant allele frequency (VAF): 0.07, 0.05, and 0.07, respectively], whereas other clones having TP53 and TET2 mutations emerged as a dominant clone at tMN by clonal sweeping.In patient 2, a clone having RUNX1 and TET2 mutations (VAF: 0.41 and 0.40, respectively) was detected 6 years before the diagnosis of nodular sclerosis classical Hodgkin lymphoma (NSCHL). The clone seemed to acquired NRAS and NF1 mutations at the NSCHL and tMN diagnoses (VAF: 0.31 and 0.38, respectively).This pattern is similar to the clonal evolution of MDS to secondary acute myeloid leukemia. In patient 3, we detected a clone having a DNMT3A mutation as the primary clone at DLBCL diagnosis; however, this clone was not detectedat the tMN diagnosis. Moreover, a clone having a SF3B1 mutation that was detected as a minor clone (VAF, 0.07) at the DLBCL diagnosis seemingly acquired the TP53 mutation and evolved as a dominant clone at tMN.This pattern demonstrates a clonalevolution from CHIP to tMN.We detected no common nonsynonymous mutations with BM mononuclear cells (BMMCs) in ML tissues of patients with tMN, excluding patient 2. The same RUNX1, TET2, and NRAS mutations observed in BMMCs were detectedat the NSCHL diagnosis in patient 2, however we speculated that these mutations were derived from nonmalignant T cells.

Conclusion:

This study suggests that the primaryclone as CHIP at the ML diagnosis was not always responsible for tMN, and clones that resisted chemotherapy or radiotherapy may have evolved into tMN by several clonal evolution patterns, including clonal sweeping and linear evolution.