Comparisons between new and established methods for analysis of response to treatment in acute leukemia

Abstract

Acute leukemia is a severe hematologic malignancy that affects both adults and children. Although prognosis has improved over recent decades, relapse remains one of the major challenges. A key reason for relapse is that small amounts of leukemic cells, in numbers well below the sensitivity of routine morphology, have survived treatment and remain in the bone marrow: measurable residual disease (MRD). Presence of MRD after initial treatment is one of the strongest predictors of relapse, and sensitive MRD analysis is cru-cial for identifying patients at risk of relapse. MRD is routinely assessed at treatment of acute leukemia, using various methods depending on the specific leukemia subtype. How-ever, none of the current techniques is fully optimal; many require further evaluation, and for some patients, no applicable method is available. This thesis aims to improve MRD diagnostics in acute leukemia by evaluating existing methods and developing novel tech-niques, with the goal of contributing to improved clinical management and, ultimately, a better prognosis for patients with acute leukemia. The main method used for MRD analysis in this thesis was deep sequencing, a highly sensitive next generation sequencing-based technique that quantifies mutations present in leukemic cells at diagnosis. Other molecular methods, including reverse transcription quantitative polymerase chain reaction (RT-qPCR), were also applied, as well as flow cytometry for comparison and complementary MRD assessment. The material consisted of blood and bone marrow samples analyzed at diagnosis and during and after treatment, from adult and pediatric patients with acute mye-loid leukemia (AML), and from pediatric patients with precursor B-cell acute lympho-blastic leukemia (pre-B ALL). Paper I shows that RT-qPCR of ETV6::RUNX1 fusion transcript can be used as an alternative and valuable complementary MRD method to flow cytometry in children with pre-B ALL. Paper II shows that there is a strong correlation between RNA- and DNA-based methods for MRD analysis of NPM1 mutations in adults with AML, and that DNA-based methods can complement, or replace, RT-qPCR. Paper III shows that deep sequencing of mutated NPM1 during and after treatment predicts relapse and poorer survival in adult AML. Paper IV shows that deep sequencing of FLT3-ITD is a highly sensitive method for MRD detection in pediatric AML, enabling monitoring of treatment response and early relapse detection. In conclusion, this thesis contributes new knowledge about MRD analysis in acute leukemia, highlighting the clinical value of deep sequencing. The implementation of these findings in clinical practice could support more precise risk stratification, guide treatment decisions, and may contribute to improved prognosis for patients with acute leukemia.