Abstract 3061

Allogeneic stem cell transplantation (alloSCT) is a potentially curative treatment for patients (pts) with acute myeloid leukemia (AML). However, approximately 30% of the transplanted AML pts develop a clinical relapse. Relapsing AML often clinically presents with a rapid increase in leukemic cell load, making current treatment options less effective. In order to prevent an overt clinical relapse, early detection of an impending relapse to guide pre-emptive immunotherapeutic interventions is important. After alloSCT mixed chimerism (MC) reflects the persistence of pt hematopoietic cells including the malignant hematopoietic cell population. It has been shown that donor lymphocytes can be administered to AML pts with MC at 6 months post alloSCT with a relative low risk of graft versus host disease (GvHD). However, a high percentage of pts still suffers from relapse in the first 6 months. An option to decrease the incidence of this early relapse is the pre-emptive administration of donor lymphocytes prior to 6 months after alloSCT. However, this is associated with a higher risk of potentially life threatening GvHD. Since chimerism analysis will also reflect the persistence of non-malignant pt cells, donor lymphocyte infusion (DLI) based on MC could cause overtreatment. Therefore, availability of tumor specific and sensitive methods to monitor minimal residual disease (MRD) is necessary. Recently, a quantitative real-time polymerase chain reaction (RQ-PCR) assay to detect NPM1 type A mutations in NPM1 positive AML pts has proven suitable for MRD detection and strong correlations were found between clinical courses and the residual mutation load post-transplant. In the current study, we prospectively analyzed in a cohort of 13 NPM1A positive AML pts whether the NPM1A mutation level and or chimerism analysis in bone marrow (BM) at three months after alloSCT can be used as a guide to select pts for pre-emptive immunotherapeutic intervention. Chimerism analysis was performed using short tandem repeats PCR with a sensitivity of 1–5%. NPM1A levels were quantified by RQ-PCR with a detection limit of 6×10–4. Clinical relapse was defined as >5% blasts in BM. 3/13 Pts died in remission from infectious complications less than 3 months after alloSCT. These pts were MC and NPM1A negative. 2/13 Pts received pre-emptive DLI at 3 months based on high-risk disease alone. In the remaining 8/13 pts, without immunotherapeutic intervention in the first six months after alloSCT, chimerism and NPM1A levels at 3 months after alloSCT were compared with regard to prediction of early clinical relapse. 3/8 Pts were complete donor chimeric without detectable NPM1A and have remained in continuous complete remission. 2/8 Pts with MC and detectable NPM1A at 3 months had frank clinical relapse 14 days and 41 days later, respectively. However, 3/8 pts were MC while the NPM1A mutation was absent, and these pts did not develop a relapse within the first 6 months post alloSCT. To investigate if pre-emptive immunotherapeutic intervention could be guided by NPM1A MRD levels, we analyzed NPM1A levels in 4/13 pts who received pre-planned DLI in complete remission for MC according to institutional guidelines. 3/4 Pts were consistently NPM1A-negative. One of these pts had received two low dose DLIs at 4 and 9 months after alloSCT based on high risk disease and MC, respectively. Despite these interventions, increasing NPM1A levels were detected in five consecutive samples between 12–18 months after alloSCT. Therefore, a third high dose DLI was given at 17 months after alloSCT. Two months after this DLI, the NPM1A mutation became undetectable for the entire follow-up period of three years, and the pt remained alive in complete remission.

In conclusion, these data illustrate that quantitative monitoring of NPM1A is a more sensitive marker to predict relapse early after alloSCT as compared to chimerism analysis. Furthermore, we conclude that pre-emptive immunotherapeutic intervention based on highly sensitive detection of molecular relapse can be effective in eliminating the malignant AML cells and thereby prevent clinical relapses.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution