Residual circulating tumor DNA as a prognostic biomarker in patients with AML/MDS

In the retrospective setting, residual circulating tumor DNA (ctDNA) can be useful as a prognostic biomarker to identify patients with acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) at high risk of relapse after allogeneic stem cell transplantation (allo-SCT).¹

To assess the prognostic value of ctDNA status after allo-SCT in such patients in the prospective setting, Miho Ogawa and colleagues² examined the utility of this approach in a Japanese multicenter, prospective, observational study (KSGCT1702). The interim results of this study were reported during the Virtual 46th Annual Meeting of the European Society for Blood and Marrow Transplantation (EBMT).

Methods²

The study included patients with AML and MDS planning to undergo allo-SCT after a myeloablative conditioning regimen.

Tumor samples plus control (buccal swab) and matched serum samples were collected at diagnosis, and serum samples were also collected pretransplant and at Day 30, 60, 90, and 120 post allo-SCT. Tumor DNA (bone marrow or peripheral blood) and buccal swab DNA were analyzed by next-generation sequencing (NGS) to identify candidate driver mutations.

One or two allele-specific droplet digital PCR (ddPCR) assays, indicating driver mutation/structural variants for each patient, were designed to monitor characteristic driver mutations in ctDNA. The primary objective of the study was to evaluate the 1-year cumulative incidence of relapse rate (CIR) post allo-SCT, according to residual ctDNA status.

Results²

The study enrolled 60 patients, and 46 underwent allo-SCT. The baseline patient characteristics are reported in Table 1.

Table 1. Patient characteristics²

The median follow-up period was 123 days (25─360); out of 46 patients who underwent allo-SCT, five died without relapse and eight relapsed before Day 360.

Mutations detected by NGS included single nucleotide variants DNMT3A, CEBPA, NRAS, TET2, PTPN11, NPM1, IDH2, TP53, SMC3, RUNX1, JAK2, ASXL2, GATA2, STAG2, IDH1, DDX41, CBL, ASXL1, SETBP1, SF3B1, and FLT3, and structural variants CBFB/MYH11, KMT2A/MLLT3, and KMT2A/AFDN. Patient-specific driver mutations were used to design specific ddPCR assays.

A strong correlation was observed between diagnostic ctDNA and matched tumor DNA from BM with regard to variant allele frequency.

Fourteen patients were available for ctDNA status 3 months post allo-SCT, four with positive and ten with negative ctDNA status. Patients with positive ctDNA status at 3 months post allo-SCT had higher CIR at 10 months: 100% in patients with positive ctDNA status versus 0% in patients with negative ctDNA status.

Conclusion

Different methods, summarized in Table 2, are available to assess measurable residual disease. ctDNA monitoring is a non-invasive method that could predict relapse in patients with AML and MDS who underwent allo-SCT. However, because of the limitations of this study, such as the small sample size and the short observation period, additional studies are needed to confirm the results.

Table 2. Advantages and disadvantages of methods for MRD assessment²