Abstract 2719

The outcome of acute myeloid leukemia (AML) is strongly correlated to the extent of minimal residual disease (MRD) at remission. Fusion transcripts, over-expressed genes and mutated genes have been employed in MRD quantification by RQ-PCR and are applicable for 30%–80% of all patients. More recently, flow cytometry (FCM) has been implemented as an additional tool for MRD detection employing strict gating criteria in detection of so-called leukemia associated aberrant immunophenotypes (LAIPs). In line with data from the Schuurhuis group we have identified human Myeloid Inhibitory C-type lectin (hMICL) as a candidate for AML marking at diagnosis and stable expression at relapse in a consecutive study of more than 150 patients. Importantly, we found expression of hMICL in the progenitor populations (defined as CD34+CD38- or CD117+CD38-) in AML cases. This contrasts to its absence on nearly all hematopoietic progenitors in healthy individuals and its lack of expression in acute lymphoblastic leukemia patients. Given that CD123 is also expressed in the vast majority of AML blasts at diagnosis and is characterized as a myeloid progenitor marker we hypothesized that the combination of hMICL and CD123 might serve as widely applicable MRD markers in AML and as such improve sensitivity for this approach. We performed a prospective study encompassing 38 patients with 74 follow up samples. Importantly, the assay was performed in strict parallel with other LAIPs, i.e. with no special provisions for the new markers. This FCM assay correlated strongly (88%) to available data from the most sensitive RQ-PCR marker of a series of molecular aberrations (p=0.37, McNemars chi^2 test), and that, the MRD detection and consequently reduced qualitative discordance resulted from a fourfold reduction of patients with FCM negative results but positive RQ-PCR result (table 1). The quantitative correlation coefficient between RQ-PCR and FCM increased from 0.46 to 0.7 by addition of hMICL and CD123 (figure 1).

When sensitivities of the FCM assay were directly compared to those of validated RQ-PCR assays it was apparent that that of the FCM was inferior. However, it was still in the range in which it strongly predicted treatment failure. Undoubtedly, running more cells and samples and by employing more stringent gating and post-run analyses will further improve this strategy. However, given that treatment failures in AML are apparent by WT1 RQ-PCR with sensitivities at the 1; 1.000 range, the standard hMICL/CD123 assay should suffice in the vast majority of situations. We conclude that high-density expression of both hMICL and CD123 in the vast majority of AML cases at diagnosis and stable expression during the course of the disease combined with high sensitivity holds the potential for wide applicability and improved MRD monitoring by FCM using setup and interpretation not requiring special attention. This assay promises to be of great value in fast and efficient management of all AML patients.

Table 1.

Qualitative concordance between RQ-PCR, rFCM* and tFCM# MRD results

RQ-PCR positive patients
SamplesRQ-PCR+ FCM+RQ-PCR- FCM-RQ-PCR+ FCM-RQ-PCR- FCM+ConcordanceMcNemars chi2
rFCM Vs RQ-PCR 43 16 8 16 3 24/43(56%) p=0,044 
tFCM Vs RQ-PCR 43 28 10 4 1 38/43 (88%) p=0,37 
RQ-PCR negative patients 
 Samples rFCM+ tFCM+ rFCM− tFCM− rFCM+ tFCM− rFCM−/tFCM+ Concordance ND¤ 
rFCM Vs tFCM 23 13 0 8 2 13/23(56%) ND¤ 
RQ-PCR positive patients
SamplesRQ-PCR+ FCM+RQ-PCR- FCM-RQ-PCR+ FCM-RQ-PCR- FCM+ConcordanceMcNemars chi2
rFCM Vs RQ-PCR 43 16 8 16 3 24/43(56%) p=0,044 
tFCM Vs RQ-PCR 43 28 10 4 1 38/43 (88%) p=0,37 
RQ-PCR negative patients 
 Samples rFCM+ tFCM+ rFCM− tFCM− rFCM+ tFCM− rFCM−/tFCM+ Concordance ND¤ 
rFCM Vs tFCM 23 13 0 8 2 13/23(56%) ND¤ 
*

*Routine panel of antigens in routine flow cytometric evaluation of MRD

#

Test panel of antigens in flow cytometric evaluation of MRD including CD123, CD34 and hMICL

¤

Not done.

View Large

Figure 1.
Figure 1. Correlations between RQ-PCR and FCM
View largeDownload slide

Correlations between RQ-PCR and FCM

Figure 1.
Figure 1. Correlations between RQ-PCR and FCM
View largeDownload slide

Correlations between RQ-PCR and FCM

Close modal
Disclosure:

No relevant conflicts of interest to declare.

Topics:
detection of minimal residual disease, leukemia, myelocytic, acute, polymerase chain reaction, antigens, acute lymphocytic leukemia, cd34 antigens, c-type lectins, disease remission, flow cytometry, follow-up

Author notes

*

Asterisk with author names denotes non-ASH members.

© 2010 by The American Society of Hematology
2010
Sign in via your Institution