Correlative Cellular Analyses in a Phase I Trial (Safety and Efficacy of Autologous Intracoronary Stem Cell Injections in Total Coronary Artery Occlusions (SEACOAST)) of Autologous Bone Marrow-Derived CD133 Cells.

Correlative laboratory studies were developed in a phase I trial to evaluate the safety of intracoronary injection of escalating doses of bone marrow (BM) CD133+ cells in patients with chronic coronary ischemia. Concurrent with patient cellular therapy, CD133+ cells were phenotyped and tested functionally with endothelial cell colony formation and in vitro and in vivo transmigration. BM (194 ± 11 ml) was isolated from patients meeting study inclusion criteria. CD133+ cells (20 ± 13 x 10⁶, 84 ± 7% purity and 76 ± 7% viability (7AAD)) were isolated using the CliniMACS device (Miltenyi). Contaminating cells following the CliniMACS selection were: < 5% of CD3, CD3^neg/CD56, CD19 (immature/mature), CD14, and CD71 cells with 5% CD61, 8% CD13+ SSC^high. BM, PB (peripheral blood), cord blood (CB)-derived endothelial progenitor cells (EPC) were assessed by a culture assay (StemCell Technologies) scoring early outgrowth CFU-EC. SEACOAST patients yielded significantly less colonies compared to controls of matched PB and BM (donors 28–48 yrs) and CB: normal donor (ND) PB, 65; ND BM, 40; CB, 43; SEACOAST patient PB, 2, SEACOAST patient BM, 1. Transmigration assays were used to evaluate the functionality of selected CD133+ cells to chemotactic agents stromal derived factor-1 (SDF-1) and vascular endothelial growth factor (VEGF). Selected CD133+ cells were recovered, resuspended in DMEM/1% HSA media and after a 37°C incubation for 16–20 hrs, 5 x 10⁴ CD133+ cells were added to transwells (5 mm) for 3 hours. Transmigrated cells were quantitated by flow cytometry using anti-CD45, anti-CD133 antibodies, and Fluorosphere beads. Surface expression on ND BM CD133+ cells of CXCR4 and VEGF-R2 was 0–16.4% and 1.2–4.3%, respectively. Transmigration was effected by 200 ng/ml (range of 16–62%) but not to 10 ng/ml VEGF. For CD133+ cells devoid of the expression of CXCR4, SDF-1-induced transmigration was absent. Expression of CXCR4 and VEGF-R2 on clinical trial patient-selected CD133+ cells was 0–5% and 0–2%, respectively, and transmigration was 5–19% to 200 ng/ml SDF-1 but not to 10 ng/ml VEGF. Patient selected CD133+ cells or PB mononuclear cells (PBMC), ND CD133+ cells, or a vehicle control were injected via a left intraventricular route into NOD/SCID mice with a femoral artery ligation immediately after injury. Doppler flow measurements were obtained weekly for 6 weeks comparing the perfusion ratio of ischemic/healthy limbs. At 28 days, perfusion ratios were statistically higher in study groups receiving ND CD133+ cells (0.51 ± 0.06) compared to controls (0.37 ± 0.03, p=0.025). Mice receiving patient CD133+ cells (0.46 ± 0.04) or PBMC (0.37 ± 0.08) did not show statistically significant improvement over control animals (p= 0.07, p= 0.94, respectively). BM was harvested to assess human engraftment by cytometric analysis. Mice injected with 0.5 x 10⁶ patient BM CD133+ cells showed <0.2% huCD45+ cells compared to 1.6 ± 0.4% ND BM huCD45+. Beyond the demonstrated safety of the delivery of CD133+ cells (>70% purity and >70% viability) to chronic ischemic patients via an intracoronary route, important correlative in vitro and in vivo assays has demonstrated the diminished potency of BM-derived CD133+ cells as compared to CB and ND PB and BM-derived cells.