Abstract 3132: Granulocyte-Colony Stimulating Factor after Acute Myocardial Infarction Preferentially Mobilizes CD34+ Mononuclear Cells over CD45-/CD34-Mononuclear Cells: Results from the STEMMI Trial
Introduction: Granulocyte-colony stimulating factor (G-CSF) therapy after a STEMI does not affect left ventricular systolic function when compared to placebo. In contrast, direct infusion of bone marrow cells into the coronary arteries seems to improve left ventricular ejection fraction.
Hypothesis: G-CSF only mobilizes subsets of stem cells from the bone marrow and this explains the diverse results of the clinical trials.
Methods: The STEMMI trial population was analyzed. The trial included 78 patients (62 men; 56 [SD 8] years) with STEMI treated with primary PCI < 12 hours after symptom onset. The patients were randomized to double blind treatment with G-CSF (10 μg/kg/day) or placebo for 6 days. The concentration of circulating CD34− and putative mesenchymal stem cells (MSC) quantified as CD45-/CD34- cells was measured by flow cytometry. Left ventricular function was evaluated at baseline and after 6 months with cardiac MRI. An independent core laboratory analyzed the MRI data.
Results: The myocardium was exposed, during 7 days to 25x109 G-CSF mobilized CD34+ cells compared to 3x109 cells in the patients receiving placebo (p < 0.001). The fraction of CD34+ cells/leukocyte increased during G-CSF treatment whereas the fraction of MSC surprisingly decreased. When sub characterizing the putative MSC then most of the subtypes increased in the placebo group. However, cells with the early stem cell marker CD73 or the endothelial marker CD31 had lower increase or even decrease in the G-CSF treated group (p=0.01 and 0.001), whereas cells with the haematopoietic marker CD133 increased most in the G-CSF group (p=0.01). Surprisingly, we found a negative association between the number of MSC and the change in left ventricular EF (95% CI of regression coefficient −11.4 to −2.2, p=0.004), whereas none of the MSC subtypes were independent predictors of the systolic improvement.
Conclusions: 1. G-CSF preferentially mobilizes CD34+ cells over putative MSC. 2. G-CSF causes a shift in the subtypes of MSC in the peripheral blood. 3. The number of circulation MSC seems to affect the change in EF following STEMI. These results could potentially explain the difference in outcome following intracoronary infusion of bone marrow cell solutions and G-CSF mobilization of stem cells.