Lineage-specific early complete donor chimerism and risk of relapse after allogeneic hematopoietic stem cell transplantation for acute myeloid leukemia

Abstract

Recipient–donor chimerism is routinely analyzed after allogeneic hematopoietic stem cell transplantation (HSCT) to monitor engraftment and graft rejection. For malignancies, chimerism can also be used to screen for disease relapse post-HSCT but methodology and interpretation of results are not standardized and likely depend on underlying diagnosis. We have implemented highly sensitive and accurate methodologies for chimerism analysis for the purpose of improving relapse prediction. Here, we report an exploratory retrospective analysis of clinical routine chimerism results from all 154 HSCTs for acute myeloid leukemia (AML) performed at our center during the years 2015–2020 with the aim of suggesting a clinically useful threshold at which risk of relapse is high. Relapse was not reliably predicted based on single elevated chimerism values obtained before time of overt relapse. However, early complete donor chimerism, here defined as recipient DNA < 0.2% in CD33+ cells in any blood or bone marrow sample taken during the first 60 days after HSCT, correlated inversely with relapse during the observation time (log-rank test P = 0.033). We propose that achievement of complete chimerism determined early after HSCT using sensitive methods can be used for risk-stratification of AML patients.

Early relapse prediction after allogeneic hematopoietic stem cell transplantation for acute lymphoblastic leukemia (ALL) using lineage‐specific chimerism analysis

“…In our previous study of AML this was not the case [10]. Interestingly, relapse was not primarily associated with increased chimerism in the B cell lineage, as would be expected if it acted as a MRD marker, but rather in the T cell lineage.…”

“…However, in both compartments relapse was preceded by a transient increase in CD3 + cell chimerism. To address if this is a general phenomenon observable also in other forms of leukemia, we retrieved similar data from our recently reported study of AML HSCTs [10] and constructed similar plots (Figure 4B,D). In this case, we compared the kinetics of chimerism results in CD3 + cells with cells positive for the myeloid cell marker CD33 and observed subtle signs the same phenomenon.…”

“…Interestingly, relapse was not primarily associated with increased chimerism in the B cell lineage, as would be expected if it acted as a MRD marker, but rather in the T cell lineage. In figure 4, we have plotted chimerism data from HSCTs that resulted in a relapse and also included data from our previous study of AML [10]. The fitted lines suggest that there is often a transient increase in recipient T cell chimerism that precedes the more pronounced increase at the time of relapse of CD19 + and CD33 + chimerism, respectively.…”

“…were 24% and 58% [10], reflecting the worse prognosis associated with Note: Logistic regression and Kaplan-Meier analysis results are presented. p-Values for Kaplan-Meier analysis were calculated using the log-rank test.…”

“…We have recently reported an analysis of chimerism results from the 154 HSCTs performed at our center during the years 2015–2020 for acute myeloid leukemia (AML) [10]. Given the sampling frequency performed in clinical practice during these years it did not appear feasible to predict an emerging relapse in time to intervene, likely due to the rapid disease kinetics of AML relapses.…”