Mobilization of Endogenous CD34+/CD133+ Endothelial Progenitor Cells by Enhanced External Counter Pulsation for Treatment of Refractory Angina

Abstract

Adult stem cell therapy via intramyocardial injection of autologous CD34+ stem cells has been shown to improve exercise capacity and reduce angina frequency and mortality in patients with refractory angina (RA). However, the cost of such therapy is a limitation to its adoption in clinical practice. Our goal was to determine whether the less costly, less invasive, and widely accessible, FDA-approved alternative treatment for RA patients, known as enhanced external counterpulsation (EECP), mobilizes endogenous CD34+ stem cells and whether such mobilization is associated with the clinical benefits seen with intramyocardial injection. We monitored changes in circulating levels of CD34+/CD133+ and CD34+/KDR+ cells in RA patients undergoing EECP therapy and in a comparator cohort of RA patients undergoing an exercise regimen known as cardiac rehabilitation. Changes in exercise capacity in both cohorts were monitored by measuring treadmill times (TT), double product (DP) scores, and Canadian Cardiovascular Society (CCS) angina scores between pre- and post-treatment treadmill stress tests. Circulating levels of CD34+/CD133+ cells increased in patients undergoing EECP and were significant (β = -2.38, p = 0.012) predictors of improved exercise capacity in these patients. CD34+/CD133+ cells isolated from RA patients could differentiate into endothelial cells, and their numbers increased during EECP therapy. Our results support the hypothesis that mobilized CD34+/CD133+ cells repair vascular damage and increase collateral circulation in RA patients. They further support clinical interventions that can mobilize adult CD34+ stem cells as therapy for patients with RA and other vascular diseases.

Keywords: adult stem cells; endothelial progenitor cells (EPCs); endothelial-to-hematopoietic stem cell (HSC) transition; enhanced external counterpulsation (EECP); ischemic coronary artery disease; refractory angina (RA).

Figure 1

Percent changes in DP scores (solid bars) and TT (open bars) and in log₁₀ transformed CD34+/CD133+ counts (solid bars) and CD34+/KDR+ counts (open bars) and occurrence of MACE following EECP and cardiac rehabilitation therapy. (A) The percent change (Δ) in stress test measurements ([post-therapy measurement minus pre-therapy measurements]/pre-therapy measurement × 100) and (B) the percent change (Δ) in log₁₀ transformed cell count measurements ([log₁₀ of mean week 4–7 cell count minus log₁₀ of mean week 0–3 cell count]/log₁₀ of mean week 0–3 cell count × 100) are shown for each patient. In instances where a patient was unable to do a post-therapy stress test, the percent change in TT was recorded as a 100% decrease, and when a patient could not do a pre-therapy stress test but could do a post-therapy test, the percent change was recorded as a 100% increase. The percent change in DP for patients who could not exercise either before or after EECP was calculated as described above based on their resting DP score. Patients who did not complete a post-therapy stress test for non-cardiac related reasons are denoted by (‡) and were not entered in survival analyses. The types of MACE patients experienced are identified as follows: * angioplasty findings of de novo lesion or restenosis; ** heart failure, *** myocardial infarction, and § unstable angina. Deaths are denoted by † (patient R4 had a sudden, non-cardiac-related death). Patients for whom insufficient data were available to calculate treatment-related changes in CD34+/CD133+ and CD34+/KDR+ cell counts are enclosed in brackets [patient #], and those for whom CD34+/KDR+ data were not available are enclosed in braces {patient #}. Percent changes in cell count greater than 200% are indicated in parentheses by the bars (EECP patient n = 37, rehabilitation patient n = 11).

Figure 2

Circulating levels of CD34+/CD133+ increase over the course of EECP therapy in patients who respond with increased exercise capacity. Counts of circulating CD34+/CD133+, CD34+/KDR+, and total CD45+/CD34+ cells from patients undergoing EECP treatment and cardiac rehabilitation therapy are expressed as log₁₀ (cell counts/100,000 mononuclear cells) and are plotted against the treatment periods after being stratified into patients who did (Yes) or did not (No) respond to the therapies with improved exercise capacity. The data were analyzed by repeated measures ANOVA and tested for significant differences between cell counts in the first (weeks 0–3, open circles) and the second half (weeks 4–7, closed circles) of EECP therapy and for differences between baseline cell counts (open circles) and those observed in the second half (middle + end, closed circles) of rehabilitation therapy. Error bars represent the standard error of the mean. EECP patient n = 26, cardiac rehabilitation patient n = 11. * Significant at alpha = 0.05.

Figure 3

EECP responders had larger treatment-related increases in CD34+/CD133+ cell counts than non-responders. The percent change in CD34+/CD133+ and CD34+/KDR+ cell counts between the first and second half of EECP therapy and over the course of cardiac rehabilitation therapy for individual patients are plotted with the median value and IQR (box plot) for each group. Differences between percent changes in EPCs in responder (closed circles) and non-responder (open circles) groups in each treatment were tested for significance by the Mann–Whitney U test on ranked data since these data did not meet the assumptions of normality or equal variance. EECP patient n = 26, cardiac rehabilitation n = 11. * Significant at alpha = 0.05.

Figure 4

RA patients who responded to EECP and cardiac rehabilitation with improved exercise capacity had a reduced risk of MACE. RA patients were stratified according to four criteria: (1) improved TT or not, (2) improved DP score or not, (3) improved CCS score or not, and (4) responder to therapy with improved exercise capacity or not and evaluated for their effect on MACE hazard by Kaplan–Meier survival analysis of the combined EECP and cardiac cohorts and the cohorts separately. In this analysis, only improved exercise capacity had a significant effect on MACE hazard and only for the combined EECP and rehabilitation patient cohorts. In the survival curves shown for the combined cohorts, responders are represented by a solid line and non-responders by a dashed line. Tick marks on the curves denote times at which patients were lost to follow-up. The numbers of patients in the improved (yes) or not improved (no) categories at each 12-month interval are shown in the table below the survival curves.

Figure 5

EECP therapy induces increased numbers of EPC outgrowth colonies. Numbers of outgrowth colonies obtained from CD34+/CD133+ cells isolated from EECP-treated RA patients during therapy (closed symbols) were increased (p = 0.022) over baseline levels and over those of age-matched healthy volunteers (open symbols, p < 0.001). Sequential blood samples taken from EECP-treated RA patients #24 and #26 and healthy control subjects 1 and 2 yielded the number of outgrowth colonies/dL of peripheral blood for the weeks indicated after the baseline samples were taken. The number of outgrowth colonies obtained from single blood samples of EECP-treated RA patient #25 and healthy control subjects 3-6 are denoted by separate symbols at the 0-time point.

Figure 6

Outgrowth colonies of CD34+/CD133+ cells selected from the peripheral blood of EECP patients show endothelial cell characteristics. CD34+/CD133+ cells selected from 7 mL volumes of blood from EECP patients and healthy volunteers following RBC lysis were cultured in modified Hill clonogenic assays as described in the Methods. The total numbers of outgrowth colonies contained in quadruplicate culture wells were visually counted after 2 weeks of incubation and averaged. Examples of initial outgrowth colonies (panel (A)) and the development of a vascular network (arrows, panel (B)) are shown. After 4 weeks in culture, cells were stained for PECAM (denoted by circles in panel (C)) and low-density lipoprotein (denoted by arrows in panel (D)) expression (scale bar = 60 µm).