This study establishes the methodology for monitoring cell retention at the site of transplantation and determining the impact of these injections upon a) the natural change in infarct size, wall motion, and remodeling indices serially for a 12 week period. We have tracked cell retention using the radioactive tracer Indium111 and the fluorescent lypophyllic marker, PKH26, to co-label our cells in vitro. Previously, we have shown that our labeling procedure, at the radioactive doses used, does not affect the survival, proliferation or differentiation of stromal cells . There has been concern that Indium labeling may lead to harmful effects on cell function, but the administered dose per cell was not provided in that publication . SPECT of In111 has allowed us to evalaute cell clearance kinetics, up to 2 weeks, and to correlate these with measures of treatment effect. We observed a rapid loss of111In signal over a two week period post-injection, and this correlated with a small number of PKH positive cells at 12 weeks. Since SPECT could not detect111In signal at 12 weeks post-injection, a direct comparison between111In signal and the number of PKH26 labelled cells was not possible. However, rapid cell loss has previously been described for muscle satellite cells injected into skeletal muscle  and for satellite cells injected into myocardium . We do not know if this observed rapid clearance is the result of cell death caused by the relatively hostile inflammatory environment present in recently infarcted myocardium used in our model, the migration of cells away from the injection site, or if the kinetics described only apply to the cell lines used. We did not, in these experiments, quantify retained cell numbers and therefore, we are not able to correlate these with treatment effect.
While there was some evidence that BMMCs can transdifferentiate into cardiomyocytes at 3 weeks post-injection (Figure 6), this was a relatively rare event. We did not observe BMMC or stromal cell-derived cardiomyocytes in any of our treated dogs at 12 weeks.
Our study demonstrates a therapeutic effect from the injection of autologous BMMC's into the peri-infarct region after 3 hrs of ischemia and 3 hrs of reperfusion, when compared to controls or stromal cell-treated animals. Previous studies have demonstrated improved neovascularization following transplantation of monocytes in a murine model , a reduction in infarct size with increased angiogenesis but no change in regional function in a porcine model , alteration of LV remodeling indices in a rat model , and even improved angiogenesis and cardiac function when cells were retroperfused through the cardiac veins of pigs subject to a left anterior descending occlusion .
Stromal cells have been claimed to produce superior myocardial regeneration in rats [69, 70], and a porcine model , but only increased angiogenesis with no improvement in scar reduction in a rat model . These inconsistent literature reports leave uncertainty as to whether cell therapy does provide reproducible evidence of benefit in large animal models. It is our hope that the concurrent use of MR to monitor changes in scar shrinkage, and correlation of this with both cell retention kinetics and quantitative measure of cell retention (in future experiments), will help to provide more concrete evidence to support claims of treatment efficacy. Infarct shrinkage was the parameter, which, in our hands, was associated with the most consistent pattern of evolution, and the least inter-animal variation through to 6 weeks. Beyond that point, the curves began to diverge with a treatment effect with the BMMC's. We would suggest that future studies focus on this index parameter as a gauge of treatment response. Further, our study provides a framework for planning imaging studies to monitor the effect of cell therapy. We recommend early post-infarct imaging, a repeat at 6 weeks, and then at the end-point of treatment, which may be 12 weeks or longer. The pattern and degree of initial infarct shrinkage may also allow calculation of the necessary sample size needed to establish treatment effect.
Although Orlic claimed a major reduction in the extent of infarct size [6, 8], others have not been able to reproduce these results using similar experimental methods [17, 18]. While the cellular basis for improved cardiac function is still unknown, recent studies suggest that any therapeutic value may involve mechanisms that prevent ventricular dilation, increase myocardial wall thickness (resulting in improved cardiac output) and promote neo-angiogenesis at the site of injury.
We did not see any evidence in our study of cellular differentiation nor of fusion with existing constituents at 12 weeks. Rather, we witnessed the relatively rapid clearance of cells from the injection sites with a half-life of about 5 days, suggesting that these marrow-derived cells could only produce benefit through a transient paracrine effect that persisted beyond their clearance. The rapid loss of cells from injection sites has previously been described for other target tissues, such as skeletal muscle , and cell types such as cardioblasts , and is unlikely to be a consequence of radiation effects. However, Tran claimed no loss of cells from either infarct or normal myocardium from 2 hrs to 7 days after injection in a rat infarct model .
Also, Tran et al recently assessed in a rat model the usefulness of dual-isotope Tc-MIBI perfusion and Indium-oxine cell labeling imaging to gauge the pattern of infarct evolution . They found no significant shrinkage in the size of the perfusion defect between the day of infarction and 1 month, in marked contradistinction to our CMR findings. We used CMR because of its superb spatial resolution [73–77] and it would appear to be a more reliable way of assessing infarct evolution than nuclear medicine-based measures of myocardial perfusion. These depend on sequestration of the perfusion tracer by metabolically intact cells. Our CMR findings in the control animals are consistent with those of Fieno et al who found that between 3 days to 4–8 weeks following infarction, there was a reduction in the extent of CMR signal enhancement to 24 +/- 3% of the original values , or in effect, a 76% reduction in scar size. Our study demonstrated a progressive reduction in infarct size of 75% in the control group at 12 weeks. Studies that look at a single time point, at sacrifice for example, may not appreciate the evolutionary changes that have occurred, and may potentially detect no differences in infarct size reduction with treatment if they have vastly different baseline.