CXCR-4: homing in on Flt3

These findings potentially explain why some AML specimens engraft better than others in NOD/SCID mice, and they suggest a possible role for CXCR-4 in the pathogenesis of relapsed AML.

Acute myeloid leukemia (AML) is a disease resulting from the neoplastic transformation of a stem cell, probably in a manner similar to that of all human cancers.¹  Treatment with intensive cytotoxic chemotherapy can induce a remission in most patients with de novo AML, but the relapse rate remains quite high. Some molecular features of AML have been identified that are particularly useful in predicting the probability of relapse, most notably classical cytogenetic abnormalities (eg, 8;21, 15;17, 7-, etc),²  and, more recently, the presence of a FMS-like tyrosine kinase-3/internal tandem duplication (Flt3/ITD) mutation. In fact, there are now numerous studies comprising over 5000 AML cases that demonstrate that Flt3/ITD mutations are the most common molecular abnormality in AML and their presence predicts for relapse.³  The exact mechanism as to why these molecular abnormalities cause treatment failure is not known, but, regardless of the molecular basis, the disease relapses because of persistence of the leukemia stem cells within the patient. Thus, these elusive cells should be the focus of laboratory efforts directed at understanding (and treating) AML.

One useful system for the study of leukemia stem cells is the nonobese diabetic-severe combined immunodeficiency (NOD/SCID) murine human/chimera engraftment model.⁴  A few years ago, Rombouts et al⁵  made the salient observation that AML specimens harboring Flt3/ITD mutations showed much higher engraftment potential in NOD/SCID mice. Following up on this finding, this same group now reports that the mechanism for this phenomenon may be mediated through the chemokine receptor, CXCR-4. CXCR-4 is a G-protein-coupled receptor that, together with its cognate ligand stromal cell derived factor-1 (SDF-1), appears to play an essential role in hematopoietic homing and engraftment.

In this issue of Blood, Rombouts and colleagues (page 550) examined the CD34⁺ cells in 90 AML patient samples, analyzing them for CXCR-4 expression. They found that those cases in which a high proportion of the CD34⁺ fraction expressed CXCR-4 were far more likely to relapse. As part of a multivariate analysis, they examined CXCR-4 expression in CD34⁺ cells with other variables known to predict relapse, including Flt3 mutation status. As expected, the presence of Flt3/ITD mutations by themselves predicted for relapse. However, when CXCR-4 expression in CD34⁺ cells was included in the multivariate analysis, the presence of Flt3/ITD mutations dropped out as a predictive factor. Were CXCR-4 and Flt3/ITD mutations somehow related? Surprisingly, 19 of 22 Flt3/ITD samples showed high CXCR-4 expression in the CD34⁺ fraction. Perhaps not so surprisingly (given its apparent role in the homing of hematopoietic cells to the bone marrow), CXCR-4 expression in the AML samples appeared to correlate with the engraftment of samples in NOD/SCID mice, thus providing the answer to the puzzle posed by their earlier findings.

Like any good study, these findings only raise new questions and, of course, will need to be confirmed prospectively by others. If constitutive Flt3 activity is somehow inducing cell surface expression of CXCR-4, however, the implication is that the Flt3 mutations may be exerting their negative prognostic influence exactly where we expect them to, at the level of the leukemic stem cell.