CD36 – a marker for drug sensitive acute megakaryoblastic leukemia

S. Gurbuxani

DOI: https://doi.org/10.1080/10428190600763579

2006-01-01

Abstract:Significant progress has been made in the treatment of acute myeloid leukemia (AML) in children in the last three decades. As a consequence, various cooperative group trials carried out after 1995 have reported overall survival rates as high as 60% (reviewed in [1]). However, current chemotherapy protocols are associated with significant, frequently life-threatening toxicity. Furthermore, pediatric AML is a heterogeneous disease with a remarkably diverse biology. Clinically this manifests as variable risk factors and different treatment outcomes. Therefore, a key step in improving therapy for pediatric AML is to identify the various biologic subgroups and tailor specific therapy for these subgroups. Sensitive disease can then be treated with less intense chemotherapy regimens as a means to lessen the mortality and morbidity associated with high dose regimens, without compromising the overall outcome. In this issue of Leukemia & Lymphoma, Savaşan and colleagues [2] describe CD36 as a molecule that may be used to identify a subgroup of non-Down Syndrome (non-DS) children with acute megakaryoblastic leukemia (AMKL) that have chemosensitive disease. The authors compared the immunophenotypic profile of 17 Down Syndrome AMKL (DS-AMKL), 16 non-DS AMKL and 242 non-AMKL cases and then investigated sensitivity to Ara-C and Daunorubicin (DNR) In doing so, they identified some interesting correlations. They confirmed the presence of high CD36 expression in all DS-AMKL patients and identified a subset of non-DS AMKL that expressed high levels of CD36 as well. These CD36 positive non-DS AMKL patients were similar to the DS-AMKL patients with respect to expression of some of the other antigens such as CD11b, CD7, CD38, and CD56, as well as sensitivity to Ara-C and DNR. DS-AMKL has several unique attributes, such as expression of truncated GATA-1 (GATA1s) [3], higher cure rates when compared to non-DS AMKL [4], and increased in vitro sensitivity of blasts to Ara-C and DNR that may be related to GATA1s expression [5]. The possibility that CD36 can be used to identify a sub-group of non-DS AMKL that resemble DS-AMKL is particularly interesting in view of a recent report that used gene expression profiling to identify distinct molecular phenotypes of AMKL [6]. Bourquin and colleagues found that unsupervised consensus clustering of the gene expression profiles of non-DS AMKL subdivides these leukemias into two clusters. One of these, identified as cluster I, indeed had some similarities with DS-AMKL. These included the expression levels of molecules like KLF, GATA1, TRF2 and CD36. While the data in the two papers are difficult to compare due to the differences in the methodology used, it is possible that both groups of investigators have identified the same biologic subset of non-DS AMKL. Although the association between CD36 expression and chemosensitivity in vitro, and response to therapy in vivo has been elegantly demonstrated by Savaşan and colleagues [2], the study also raises several interesting questions. What is the actual contribution, if any, of CD36 in bestowing a drugsensitive phenotype in these blasts? This is a relevant question especially since the authors did not find CD36 expression in non-AMKL AML to be associated with increased in vitro sensitivity to Ara-C or DNR. In addition, although the authors speculate extensively on the role of GATA1 in transcriptional regulation of CD36, experimental evidence that this might indeed be the case is missing. Additional studies designed to address these questions as well as