Effect of Aging on Human Mesenchymal Stem Cell Therapy in Ischemic Cardiomyopathy Patients

Abstract

BACKGROUND

The role of patient age on the efficacy of mesenchymal stem cell (MSC) therapy in ischemic cardiomyopathy (ICM) is controversial.

OBJECTIVE

We sought to determine whether the therapeutic effect of culture-expanded MSCs persists even in older subjects.

METHODS

Patients with ICM who received MSCs via transendocardial stem cell injection (TESI) as part of the TAC-HFT (n = 19) and POSEIDON (N = 30) clinical trials were divided into 2 age groups: <60 versus ≥60 years. Functional capacity was measured by 6-minute walk distance (6MWD) and quality of life using the Minnesota Living with Heart Failure Questionnaire (MLHFQ) score, measured at baseline, 6 months, and 1-year post-TESI. Various cardiac imaging parameters, including absolute scar size, were compared at baseline and 1 year post-TESI.

RESULTS

Mean 6MWD was similar at baseline and increased at 1 year post-TESI in both groups: 48.5 ± 14.6 m (p = 0.001) for the younger and 35.9 ± 18.3 m (p = 0.038) for the older participants (p = NS between groups). The older group exhibited a significant reduction in MLHFQ score (−7.04 ± 3.54; p = 0.022), while the <60 age group had a borderline significant reduction (−11.22 ± 5.24; p = 0.058) from baseline (p = NS between groups). While there were significant reductions in absolute scar size from baseline to 1 year post-TESI, the effect did not differ by age.

CONCLUSION

MSC therapy via TESI in ICM patients improves 6MWD and MLHFQ score and reduces MI size. Importantly, age did not impair response.

INTRODUCTION

Based on preclinical studies and clinical trials, bone marrow-derived mesenchymal stem cells (MSCs) (1-3) have been shown to mitigate left ventricular (LV) remodeling associated with acute (2,4,5) myocardial infarction (MI) and chronic (1,6-8) ischemic cardiomyopathy (ICM). While the data are encouraging, evidence suggesting a deleterious effect of aging on autologous MSC transplantation has been highly controversial (9). Telomere length and shortening play crucial roles in the cellular molecular aging process (10,11) and there is a strong correlation between human MSC (hMSC) proliferative capacity and telomere length, in culture and with donor age (12). In addition to their diminished proliferative potential, aging hMSCs tend to have a compromised homing capability (13-16). Accordingly, these age-related impairments suggest that MSC therapy might produce a reduced effect when the cells are derived from older individuals.

While some proponents believe advanced stem cell donor age results in diminished function (17-21), other studies raise a clinically relevant issue as to whether recipient age is a crucial factor limiting response to cell therapy (22-24). This has led to the notion that MSC therapy outcome depends not only on stem cell age, and thus function, but also recipient age and comorbidities (9,22). Indeed, reduced responsiveness as a function of donor age would be a major limitation to the emerging development of cell therapy for heart disease, given the increasing incidence and morbidity of heart disease with age (25). Here, we tested the hypothesis that improvements in functional capacity, quality of life (QOL), and reverse remodeling by transendocardial injection of hMSCs in patients with ICM is actually preserved with recipient age. Our data here derive from the phase I/II randomized trials of TAC-HFT (Transendocardial Autologous Cells in Ischemic Heart Failure) (26) and POSEIDON (Percutaneous Stem Cell Injection Delivery Effects on Neomyogenesis) (27).

METHODS

Data from the TAC-HFT and POSEIDON were collected in a similar fashion in a central electronic data system. All ICM patients who received hMSCs from these trials were pooled together and dichotomized into 2 age groups, <60 years and ≥60 years. The associations between age and both clinical and imaging parameters were assessed. P values <0.05 were considered significant. Comprehensive statistical methods can be found in the Online Supplement.

RESULTS

A total of 49 patients who received hMSCs from both trials are included in this analysis. Thirty patients received hMSCs in the POSEIDON trial, of which 11 patients (36.7%) were <60 years old and 19 patients (63.3%) were ≥60 years old. In the TAC-HFT trial, a total of 19 patients received hMSCs, with 12 (63.2%) younger than 60 years of age and 7 patients (36.8%) in the ≥60 years age group. Average age at transplant was 51.95 (±7.33; 32.48 to 59.91 range) in the <60 years age group and 68.86 (±4.51; 62.84 to 79.01 range) in the older group. Mean time from MI to cell therapy was 6.26 ± 6.42 years for younger patients and 15.43 ± 9.23 years in the older group (p = 0.0002).

Baseline characteristics are shown and compared between age groups in Table 1. The majority of the cohort was male (89.8%). A borderline statistically significant difference was observed between age groups for the baseline 6MWD test (p = 0.0561). Scar size as percent of LV mass was significantly different between age groups at baseline (p = 0.0041). No other statistically significant differences were observed for demographic characteristics, MLHFQ, or other cardiac imaging parameters.

TABLE 1.

Baseline Characteristics

	Age Group
	<60 years (n = 23)	≥60 years (n = 26)	p value
Age at Transplant, year
	51.95 (7.33)	68.86 (4.51)	<0.0001
Time from MI to therapy, year
	6.26 (6.42)	15.43 (9.23)	0.0002
Sex
Male	21 (42.9%)	23 (46.9%)
Female	2 (4.1%)	3 (6.1%)	1.000
Race
White	12 (24.5%)	19 (38.8%)
European	0 (0.0%)	1 (2.0%)	0.1547
White North American	3 (6.1%)	1 (2.0%)
Western European	0 (0.0%)	1 (2.0%)
Black	2 (4.1%)	0 (0.0%)
Indian/South Asian	0 (0.0%)	1 (2.0%)
Filipino (Pilipino)	1 (2.0%)	0 (0.0%)
Native American	0 (0.0%)	1 (2.0%)
White Caribbean	5 (10.2%)	2 (4.1%)
Ethnicity
Hispanic or Latino	9 (18.4%)	4 (8.2%)
Not Hispanic or Latino	14 (28.6%)	20 (40.8%)	0.0629
Unknown	0 (0.0%)	2 (4.1%)
6MWD
	418.30 (71.57)	372.12 (93.01)	0.0561
MLHFQ Total Score
	42.33 (28.84)	31.58 (27.81)	0.2013
Scar size as absolute value
	26.93 (15.35)	21.47 (13.29)	0.2080
Scar size as % of LV mass
	22.09 (13.55)	11.79 (6.06)	0.0041
Ejection fraction
	31.96 (6.22)	29.49 (12.65)	0.3945
EDV
	289.36 (81.65)	274.40 (86.66)	0.5525
ESV
	199.50 (68.53)	199.29 (86.15)	0.9930
SI
	0.50 (0.07)	0.47 (0.11)	0.3077

Values are mean (SD) or n (%).

6MWD = 6-minute walking distance; EDV = end-diastolic volume; EF = ejection fraction; ESV = end-systolic volume; LV = left ventricular; MI = myocardial infarction; MLHFQ = Minnesota Living with Heart Failure Questionnaire; SD = standard deviation; SI = sphericity index.

There was a statistical trend towards reduced functional capacity at baseline in the older age group, with 6MWD at baseline (418.30 ± 14.92 meters vs. 372.12 ± 18.24 meters, <60 years vs. ≥60 years, respectively; p = 0.056). A repeated measures model was used to estimate 6MWD at 6 months and 1 year post-TESI, which is shown in the Central Illustration. The 6MWD increased significantly over time in both the younger and older age groups (p = 0.001 and p = 0.038, respectively). A repeated measures model adjusting for baseline 6MWD showed no significant difference in 6MWD between age groups across time (p = 0.5621). Using the estimates from the model, we tested whether there were differences between age groups at each of the follow-up time points. The estimated difference at 6 months post-TESI was -1.01 (95% confidence interval [CI]: -35.51 to 33.49; p = 0.9538) and at 1 year 18.61 (95% CI: 15.95 to 53.17; p = 0.2882).

QOL measured by the MLHFQ was compared between age groups. A repeated measures model was used to estimate MLHFQ total score at 6 months and 1 year post-TESI, which is displayed in Figure 1. Patients <60 years of age showed an improvement in MLHFQ total score over time, although of only borderline significance (p = 0.0580), while those ≥60 years of age exhibited a more significant improvement in MLHFQ total score over time (p = 0.0220). A repeated measures model adjusting for baseline MLHFQ total score showed no significant difference in total score between age groups over time (p = 0.5859). Estimated differences between groups at 6 months and 1 year post-TESI were not significant (2.74; 95% CI: -5.74 to 11.23; p = 0.5237 and -0.72; 95% CI: -9.51 to 8.07; p = 0.8711, respectively). When 6MWD and MLHFQ were analyzed using the repeated measures model using age as a continuous covariate, neither outcome demonstrated a significant change with increase in age (p = 0.137 and p = 0.535, respectively).

FIGURE 1. Patient Quality of Life: Minnesota Living with Heart Failure Questionnaire.

In this graphic representation of estimated mean Minnesotta Living With Heart Failure Questionnaire, the total score values and individual patient values in each age group at each time point are depicted, using a repeated measures model. Both age groups’ total scores improved in a parallel fashion at 6 months post-TESI. Both groups plateaued by 1 year post-TESI with similar mean total scores in between-group comparison (p = 0.524 and p = 0.871 at 6 months and 1 year post-TESI, respectively; between-group comparison). Abbreviation as in the Central Illustration.

Next we examined the impact of age on reduction in MI scar size. Both age groups had a similar absolute scar size at baseline. MI size was 26.93 ± 15.35 g in the <60 years age group and 21.47 ± 13.29 g in the ≥60 years age group (p = 0.208). When testing within group changes from baseline to 1 year post-TESI, patients <60 and ≥60 years of age both had a significant decrease in absolute scar size (p < 0.0001 and p < 0.0001, respectively). Furthermore, this percent change over time was not significantly different between groups (-33.44 ± 5.41% in the younger versus -32.89 ± 5.25% in the older age group; p = 0.945) (Figure 2A and Figure 3). Scar size as a percentage of LV mass was significantly higher at baseline in the younger group (22.09 ± 13.55 g) when compared to older patients (11.79 ± 6.06 g; p = 0.004). However, at 1 year post-TESI, the percent change from baseline was significant within both age groups (p = 0.0013 in the younger age group and p < 0.0001 in the older age group). There was no significant difference in scar size as percent of LV mass when comparing percent change from baseline to 1 year post-TESI between age groups (p = 0.197) (Figure 2B).

FIGURE 2. Changes in Cardiac Structure and Function 1 Year Post-MSC Injection.

(A) Neither age group demonstrates significant improvement in EF. (B) Both younger and older patients show a significant decrease in absolute scar size, with no difference at 1 year post-TESI. (C) Scar size as a percentage of LV mass decreases in both age groups, and do not differ at 1 year. (D and E) Sphericity index and EDV significantly improve only in the older age group; however, there are no between-group changes 1 year post-TESI. (F) The older age group shows a trend in decreased ESV; neither ESV or EF were significantly different between groups at 1 year post-TESI. *< 0.05 within-group repeated measures ANOVA. EDV = end-diastolic volume; EF = ejection fraction; ESV = end-systolic volume; LV = left ventricular; MSC = mesenchymal stem cell; other abbreviation as in the Central Illustration.

FIGURE 3. Impact of TESI on Scar Reduction.

(A) Short-axis views of the midventricular area of a younger patient's heart show delayed enhancement delineated at the anterior and septal walls. After MSC injection, scar size in <60-year-old patient decreased from 30.9 g at baseline to 21.2 g at 1 year with a -31.4% reduction. (B) Short-axis views of the midventricular area of a ≥60 year-old patient's heart, with delayed enhancement shown at the anterior and septal walls. Delayed tissue enhancement corresponds to scarred tissue and is depicted brighter than the nonscarred tissue (automatically detected and delineated with red using the full width at half maximum technique). Red, green, and white lines demarcating the endocardial, epicardial contours, and borders of the segments, respectively, were drawn manually. Extent of scar is represented by blue arrows. After MSC injection, scar size in the ≥60-year-old patient decreased from 36.2 g at baseline to 24.5 g at 1 year, with a similar -32.3% reduction in MI size. MI = myocardial infarction; other abbreviations as in Figures 1 and 3.

Cardiac computed tomography (CT) or magnetic resonance imaging (CMR) measured end-diastolic volume (EDV) and sphericity index (SI) were similar between age groups at baseline (p = 0.553 and p = 0.508, respectively). At 1 year post-TESI, within-group changes in EDV (p = 0.024) and SI (p < 0.0001) significantly decreased in the ≥60 years group, but no significant change was observed in the younger group. Neither EDV nor SI, as a percent change from baseline, differed between age groups at 1 year post-TESI (p = 0.434 and p = 0.077, respectively) (Figures 3C and D). Although mean ejection fraction (EF) and end-systolic volume (ESV) between groups were not different at baseline (p = 0.395 and p = 0.993, respectively) or at 1 year post-TESI (p = 0.143 and p = 0.417, respectively), there was a borderline significant decrease over time in ESV within the older age group (p = 0.054), that was not replicated in the younger group. There was no significant increase in EF in either age group. Linear regression analyses, using age as a continuous variable, did not indicate any significant association between cardiac structure or function and age (Table 2).

TABLE 2.

Association Between Cardiac Structure and Age Using a Linear Regression Model

Cardiac Imaging Parameter	Regression Parameter Estimate	Standard Error	p Value
Scar size as absolute value	−0.25	0.38	0.5009
Scar size as % of LV mass	−0.58	0.39	0.1501
EF*	—	—	0.5414
(EDV	−0.25	0.18	0.1639
(ESV	−0.28	0.26	0.2777
SI	−0.19	0.18	0.2831

^*Due to the highly skewed distribution of ejection fraction, a non-parametric Spearman Rank Correlation Coefficient test was used to determine if a significant association was present. Correlation coefficient between age and EF = 0.0957.

Abbreviations as in Table 1.

DISCUSSION

The major new finding of this study shows that therapeutic responses to culture-expanded MSCs are not impaired in subjects of older age. This is particularly important to the emerging field of cell therapy for chronic heart failure due to ischemic cardiomyopathy, a disorder that increases dramatically in incidence with age (28). If cell therapy responses were impaired with age, this would impact patients at greatest risk. Our findings suggest that culture expansion of MSCs overcomes any limitation that endogenous cells might have and age of the host is not a limiting factor. These data support the developing this strategy for individuals of advanced age and, thus, they boast major public health implications.

Here, we analyzed efficacy outcomes, from the TAC-HFT (26) and POSEIDON (27) trials, to test whether older patients with chronic ICM receiving transendocardial mesenchymal stem cell therapy have impaired outcomes relative to young individuals. Notably, improvements in functional capacity were evident at 6 months following injection and persisted to 1 year after TESI to similar degrees in both age groups. To date, while several studies have examined whether donor cell age and function influence responses to cell therapy (17,19,20), no study has tested the hypothesis that recipient age diminishes the efficacy of MSC transplantation. Moreover, while previous studies examining aging and stem cell potency have tested bone marrow mononuclear cells (29-31) and peripheral blood progenitor cells (32), this relationship in culture-expanded mesenchymal stem cells has not previously been explored.

The 6-minute walk test has been widely used to assess functional capacity in patients with advanced heart failure (33), and is an independent predictor of all-cause mortality (34). We found a significant improvement in 6MWD in both age groups (Central Illustration), a result that parallels the overall results of TAC-HFT and POSEIDON. More importantly, the change from baseline at both 6 months and 1 year following TESI did not differ significantly between age groups. Such an improvement in functional capacity highlights the test's prognostic value (35) and strongly suggests that older cell recipients functionally recover just as well as younger patients.

Findings from the TAC-HFT and POSEIDON trials suggest that MSCs reduce infarct size over time. We found a similar significant decrease in absolute scar size irrespective of age groups at 1 year post-TESI (Figure 2 and Figure 3). These results corroborate findings from the POSEIDON trial where both autologous and allogeneic MSCs reduced infarct size over time. It is known that MSCs secrete anti-fibrotic matrix metalloproteases (36) via paracrine signaling. Although the reason remains mechanistically unclear, our study demonstrates that aging in stem cell recipients does not appear to influence the anti-fibrotic effects of MSC therapy. Interestingly, scar size as a percentage of LV mass in the younger age group was almost twice as large as the older age group at baseline (Table 1). This cannot be accounted for except possibly for the small sample size used in the study, which was further narrowed when focusing on the <60 years of age group. More importantly, the change from baseline to 1 year post-injection was similar between the two groups.

Quality of life in each age group was similar at baseline and comparisons of MLHFQ total score changes at both 6 months and 1 year from baseline did not differ significantly between groups (Figure 1). The questionnaire has been deemed a valid and effective instrument in measuring QOL in heart failure patients (37). Composed of 21 items that sum up to a total score, ranging from 0 (no effect) to 105 (strong effect of heart failure on daily life), the MLHFQ is a commonly used assessment tool in heart failure studies (38,39). Our analysis demonstrated a significant improvement in MLHFQ score in older patients and a trend, with borderline significance, towards an improved total score in patients <60 years of age.

Past studies (1,2) established the role of MSC therapy in LV reverse remodeling. We found that left ventricular chamber volumes between age groups were similar at baseline as well as 1 year following injection, albeit a significant improvement in EDV only in patients, aged ≥60 years. Correspondingly, sphericity index was reduced in the older age group, despite being the same between groups at baseline and 1 year post-TESI. While these findings may not completely corroborate those in POSEIDON and TAC-HFT, they do raise the important concept of recipient age not having an influence on MSC therapy efficacy. Improvements in EF have been inconsistent throughout clinical trials of stem cell therapy (40) and we did not find significant increases in either age group. Still, 1 year following injection, EF between older and younger age groups was not different. Notwithstanding these data, it is important to note that infarct size is a stronger predictor of future adverse cardiac events than EF (41).

STUDY LIMITATIONS

This study has several limitations that warrant mention. First, this work has a relatively small sample size, which may limit conclusions from certain measurements such as time to therapy. While a formal power or sample size calculation was not performed prospectively given that the patient population originates from a fixed cohort, we note that the sample size in each of the age groups would have 84% power to detect a difference in the distance walked in 6 minutes from 33 to -10 using a 2-sided alpha of 0.05, a relatively large difference. Second, because this analysis is a composite of 2 different clinical trials, data from 2 different imaging modalities are employed; multidetector CT scanning in POSEIDON and both cardiac CT and CMR in TAC-HFT. We corrected for this by calculating percent changes for cardiac imaging measurements. Finally, the issue of biological versus chronological age merits comment. Researchers generally feel that biological age predominates over chronological age and can be assessed with molecular assays such as telomere length (10-12). We did not incorporate this assessment, as telomere length assays were not part of either the POSEIDON or TAC-HFT study design. Importantly, whereas most telomere studies have correlated cellular and chronological age of donor cells, our study does not incorporate cellular characteristics of donor cells, rather the chronological age of recipients. Future studies are planned to measure telomere length in both donors and recipients.

CONCLUSIONS

In conclusion, our study suggests that recipient age does not reduce the effects of MSC therapy in patients with ischemic cardiomyopathy. Importantly, comparisons of 6-minute walk test and absolute scar size between age groups did not differ. Our findings document for the first time the relationship between advanced age and clinical outcomes in heart failure and show an important preservation in responses to cell therapy in a group of recipients of advanced age. These data support ongoing clinical trials on cell-based therapy and the need for future clinical investigation of MSC use in older age groups.

PERSPECTIVES.

Competency in Medical Knowledge: Aging patients respond similarly to younger patients after receiving human mesenchymal stem cell therapy for ischemic cardiomyopathy. Functional capacity, quality of life, and several cardiac function parameters may improve in these patients, even long after myocardial infraction, highlighting their ability to yield to mesenchymal stem cell anti-fibrotic and pro-myogenic effects.

Translational Outlook: Older patients may be eligible in future clinical stem cell therapy trials. Future randomized studies are required to further demonstrate equivalence in responsiveness between older and younger recipients.

CENTRAL ILLUSTRATION Patient Functional Capacity: 6-Minute Walking Distance.

This graphic representation shows estimated mean 6-minute walking distance values and individual patient values in each age group at each time point, using a repeated measures model. Both age groups increased similarly until 6 months post-TESI. Although the older age group plateaus by 1 year post-TESI, average distances between age groups do not differ significantly (p = 0.954 and p = 0.288 at 6 months and 1 year post-TESI, respectively; between-group comparison). Abbreviations: TESI = transendocardial stem cell injection.

ABBREVIATIONS

6MWD

6-minute walk distance

EDV

end-diastolic volume

EF

ejection fraction

ESV

end-systolic volume

hMSC

human mesenchymal stem cell

LV

left ventricular

MI

myocardial infarction

MLHFQ

Minnesota Living with Heart Failure Questionnaire

SI

sphericity index

TESI

transendocardial stem cell injection