Targeted Sequencing Detection of Molecular Minimal Residual Disease and Prognosis in AML
In a study reported in The New England Journal of Medicine, Jongen-Lavrencic et al found that molecular minimal residual disease identified by next-generation sequencing during complete remission was associated with increased risk of relapse and mortality over 4 years of follow-up in patients with acute myeloid leukemia (AML). The presence of persistent mutations associated with age-related clonal hematopoiesis was not associated with prognosis.
Study Details
The study involved assessment of bone marrow or peripheral blood samples collected between 2001 and 2013 from 482 patients aged 18 to 65 years with newly diagnosed AML (n = 428) or refractory disease with an excess of blasts and a Revised International Prognostic Scoring System score > 4.5, indicating high or very high risk of relapse (n = 54). To be included in the study, patients had to be in complete remission or complete remission with incomplete hematologic recovery with less than 5% blast cells in the bone marrow after two cycles of induction chemotherapy. Patients were from clinical trials of treatment according to protocols of the Dutch–Belgian Cooperative Trial Group for Hematology–Oncology (HOVON) or the Swiss Group for Clinical Cancer Research (SAKK).
Targeted next-generation sequencing for mutations in 54 genes present in hematologic malignancies was performed in samples taken at diagnosis and after induction therapy during complete remission. The study endpoints were 4-year rates of relapse, relapse-free survival, and overall survival. Outcomes were analyzed according to the presence of targeted mutations with and without exclusion of persistent DTA mutations (ie, mutations in DNMT3A, TET2, and ASXL1), which are frequently found in age-related clonal hematopoiesis.
Molecular Minimal Residual Disease and Outcomes
An average of 2.9 mutations per patient were detected at diagnosis, with at least 1 mutation that could be a marker of residual disease being present in 430 patients (89.2%). Mutations in NPM1, DNMT3A, FLT3, and NRAS were among the most common identified at diagnosis. Persistent mutations after induction therapy were found in 51.4% of these 430 patients, with the rate of persistence varying widely among genes. DTA mutations were the most common, with persistence rates of 78.7% for DNMT3A, 54.2% for TET2, and 51.6% for ASXL1.
The majority of mutations in RAS pathway genes were absent after induction therapy, with persistence rates of mutations in NRAS, PTPN11, KIT, and KRAS being 4.2%, 7.0%, 13.5%, and 12.5%. Allele frequencies of mutations persisting during complete remission ranged from 0.02% to 47%.
In a training cohort consisting of 283 of the 430 patients with mutations at diagnosis, identification of any persistent mutation during complete remission was associated with increased risk of relapse at 4 years (48.2% vs 32.4%, P = .03), with the correlation appearing to be independent of allele frequency. The presence vs absence of persistent DTA mutations was not significantly associated with higher 4-year relapse rate (P = .29), with the absence of correlation being independent of allele frequency.
Among patients with persistent DTA mutations, presence vs absence of persistent non-DTA mutations was associated with increased risk of relapse (66.7% vs 39.4%, P = .002). Detection vs no detection of persistent non-DTA mutations at any allele frequency was associated with increased relapse risk at 4 years (55.7% vs 34.6%, P = .001).
Among all 430 patients with mutations at diagnosis, persistent non-DTA mutations were detected during complete remission in 28.4%. Presence of these mutations was associated with a higher rate of relapse at 4 years (55.4% vs 31.9%, hazard ratio [HR] = 2.14, P < .001), poorer 4-year relapse-free survival (36.6% vs 58.1%, HR = 1.92, P < .001), and poorer 4-year overall survival (41.9% vs 66.1%, HR = 2.06, P < .001). On multivariate analysis, persistence of non-DTA mutations during complete remission was independently associated with increased relapse rate (HR = 1.89, P < .001), relapse-free survival (HR = 1.64, P = .001), and overall survival (HR = 1.64, P = .003).
Comparison With Flow Cytometry
A comparison of next-generation sequencing with multiparametric flow cytometry for the detection of persistent non-DTA mutations indicated that sequencing provided significant additive prognostic value. In a subgroup of 340 patients with sufficient samples for both analyses, concordance for detection or no detection was found in 69.1% of patients (30 patients with detection and 205 with no detection); persistent non-DTA mutations were detected with sequencing only in 64 patients and with flow cytometry only in 41 patients.
The 4-year relapse rate was 73.3% among patients with positive results on both assays, 52.3% among those with positive findings on sequencing but not on flow cytometry, 49.8% among those with positive findings on flow cytometry but not on sequencing, and 26.7% among those with negative findings on both assays. In multivariate analysis, combined use of the two assays for detection of residual disease was independently associated with rates of relapse (P < .001), relapse-free survival (P < .001), and overall survival (P = .003).
The investigators concluded, “Among patients with AML, the detection of molecular minimal residual disease during complete remission had significant independent prognostic value with respect to relapse and survival rates, but the detection of persistent mutations that are associated with clonal hematopoiesis did not have such prognostic value within a 4-year time frame.”
The study was funded by the Queen Wilhelmina Fund Foundation of the Dutch Cancer Society and the Netherlands Organization for Health Research and Development.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
