Integrating Next Generation Flow Cytometry and Next-Generation Sequencing for an Enhanced Detection of Measurable Residual Disease in Acute Myeloid Leukemia with Myelodysplasia-Related Gene Mutations

Background:

The diagnosis of acute myeloid leukemia (AML) has evolved significantly with the last classifications of hematopoietic neoplasms (WHO and ICC 5^th edition).^1,2 A key update is the addition of a new AML group with myelodysplasia (MDS)- related molecular alterations, that allows categorization of many patients previously classified as AML-NOS³.

The monitoring of minimal residual disease (MRD) in this AML group using molecular approaches is challenging, due to the unfeasibility of creating a specific PCR assay for each single alteration in the clinical practice, which leaves flow cytometry as the only option for the monitoring of MRD. In order to improve MRD evaluations we proposed an automated Next Generation Flow Cytometry approach (NGF-MRD) that provides a highly sensitive, specific, and reproducible MRD strategy.

However, the molecular alterations in those patients might also be monitored thanks to the advances in Next Generation Sequencing (NGS-MRD), that can detect any mutations and gene rearrangements at very low frequencies. However, the complex subclonal architecture of leukemic hematopoiesis and the prevalence of pre-existing and non-specific mutations complicate the interpretation of NGS-MRD studies.

Aims

• To study the correlation between NGF-MRD and NGS-MRD in AML patients with MDS-related mutations.

• To investigate the phenotypic and molecular characteristics of subpopulations potentially monitorable

Patients and methods

We studied bone marrow samples from 51 patients that reached complete remission after induction (26 after 3+7 protocol, and 25 after 21 days of Aza-Venetoclax). We used a highly sensitive flow cytometry protocol, based on an automatized data-base guided strategy, as previously described⁴. The acquisition was performed on a FACSCANTO II flow cytometer (BD Biosciences). FCS files were analyzed using the Infinicyt Software V 2.0 (Cytognos).

We performed an NGS study following the international recommendations for myeloid pathology at diagnosis ⁵. NGS libraries were synthesized using the QIAseq Targeted Human Myeloid Neoplasms DNA Panel DHS-003Z (Qiagen). The same panel was used for the evaluation of NGS-MRD. This was possible thanks to the use of unique molecular index (UMIs), which allows the reliable detection of variants at very low frequencies.

Data analysis was performed using the CLC Genomics Workbench software and the QCInterpret platform.

The study received ethical approval from the Hospital ethics committee, and all participants provided informed consent in accordance with the Declaration of Helsinki.

Results

• All patients presented at least one mutation, not monitorable by PCR, at the time of diagnosis (Me= 3, range 1-7). In patients considered NGF-MRD positive, the mutations detected at diagnosis were also found in the NGS-MRD study (VAF= 1.3%-46%).

• The correlation analysis revealed a strong association between NGF-MRD and NGS-MRD results (χ2: 18.844, p<0.05). NGF-MRD negative cases often exhibited a disappearance of the initial pathogenic mutations (excluding DNMT3A, TET2, and ASXL1). On the contrary, NGF-MRD+ cases retained any of the initial pathogenic alterations and, additionally, we detected an emerging subclonal TP53 mutation. Interestingly, most of those patients (84.9%) exhibited a high proportion (≥5%) of leukemic cells (LSCs) characterized by a stem cell phenotype CD34+ CD38- CD133++ at diagnosis.

• Discordances between NGF-MRD and NGS-MRD were observed only in 9% of NGF-MRD negative cases that showed the persistence of mutations from diagnosis. Interestingly, the detection of emerging mutations (whose meaning is still unknown) was significantly more frequent in the NGF-MRD negative group.

• In spite of the small patient cohort and the relatively short follow-up period, double-negative cases for both NGF-MRD and NGS-MRD correlate with a particularly favorable outcome (13-month SLE: 95.2% vs. 66.6%, Median SLE non reached).

Conclusions

• The combination of NGF-MRD and NGS-MRD demonstrated high level of concordance and sensitivity in AML patients with MDS-related mutations.

• The dual evaluation of MRD revealed synergistic assessment capabilities, helping to reach safer conclusions.

• The combined use of both techniques refines risk group definitions and helps to select optimal post-remission therapies.

• Further research will be essential to elucidate the clinical implications of emerging alterations.

No relevant conflicts of interest to declare.