In a Phase 1a escalating clinical trial, autologous mesenchymal stem cell infusion for renovascular disease increases blood flow and the glomerular filtration rate while reducing inflammatory biomarkers and blood pressure.

Abstract

Atherosclerotic renovascular disease (ARVD) reduces tissue perfusion and eventually leads to loss of kidney function with limited therapeutic options. Here we describe results of Phase 1a escalating dose clinical trial of autologous mesenchymal stem cell infusion for ARVD. Thirty-nine patients with ARVD were studied on two occasions separated by three months. Autologous adipose-derived mesenchymal stem cells were infused through the renal artery in 21 patients at three different dose levels (1, 2.5 and 5.0 × 10⁵ cells/kg) in seven patients each. We measured renal blood flow, glomerular filtration rate (GFR) (iothalamate and estimated GFR), renal vein cytokine levels, blood pressure, and tissue oxygenation before and three months after stem cell delivery. These indices were compared to those of 18 patients with ARVD matched for age, kidney function and blood pressure receiving medical therapy alone that underwent an identical study protocol. Cultured mesenchymal stem cells were also studied in vitro. For the entire stem cell treated-cohort, mean renal blood flow in the treated stenotic kidney significantly increased after stem cell infusion from (164 to 190 ml/min). Hypoxia, renal vein inflammatory cytokines, and angiogenic biomarkers significantly decreased following stem cell infusion. Mean systolic blood pressure significantly fell (144 to 136 mmHg) and the mean two-kidney GFR (Iothalamate) modestly but significantly increased from (53 to 56 ml/min). Changes in GFR and blood pressure were largest in the high dose stem cell treated individuals. No such changes were observed in the cohort receiving medical treatment alone. Thus, our data demonstrate the potential for autologous mesenchymal stem cell to increase blood flow, GFR and attenuate inflammatory injury in post-stenotic kidneys. The observation that some effects are dose-dependent and related to in-vitro properties of mesenchymal stem cell may direct efforts to maximize potential therapeutic efficacy.

Graphical Abstract

Introduction

Atherosclerotic renovascular disease (ARVD) commonly reduces kidney blood flow and GFR, in addition to accelerating hypertension ¹. It gradually produces a decline in tissue perfusion and eventually overt tissue hypoxia and inflammatory kidney injury that has been designated “ischemic nephropathy” ^2–5. Results from recent trials indicate that restoring blood flow with endovascular stenting often fails to restore glomerular filtration or reverse inflammatory injury for such individuals ^6–9. Despite the observation that moderate reductions in blood flow can be tolerated with only minor changes in renal oxygenation, progressive ARVD can override compensatory mechanisms and lead to irreversible tissue injury. Because ARVD is usually asymmetric, renal injury in post-stenotic kidneys can be overlooked due to compensatory changes in the contralateral kidneys. Clinical loss of kidney function therefore often becomes evident only after irreversible injury in one affected kidney is amplified by the involvement of the second kidney ¹. There is a pressing need to develop adjunctive treatment regimens to restore kidney perfusion and reverse inflammatory injury in the post-stenotic kidney.

Administration of mesenchymal stem cells (MSC) has emerged in preclinical studies as a candidate to repair the renal microcirculation, attenuate inflammatory injury and restore kidney function ^10–12. The beneficial effects of MSC are thought to be mediated by paracrine effects involving release of soluble mediators and extracellular vesicles. Our initial human data utilizing autologous adipose tissue-derived MSCs infusion in subjects with ARVD disease showed that MSC infusion into one renal artery could be achieved safely with increased tissue oxygenation ¹³.

We undertook this study with three major aims: first, we sought to define dose-related changes after infusion of autologous MSC into a single post-stenotic kidney in terms of kidney perfusion, blood flow, cytokine signaling, and glomerular filtration. Secondly, we sought to compare these effects to subjects with ARVD treated with medical therapy alone over the course of an identical study protocol (depicted in Figure 1). Finally, we sought to search for in-vitro predictors of MSC in-vivo efficacy regarding changes in renal hypoxia.

Figure 1|. (a) Enrolled participants receiving medical therapy (Med Rx-only) alone and participants receiving medical therapy plus mesenchymal stem cell (MSC) at 3 escalating doses.

Two (*) patients with MSC were excluded because of technical issues related to cell delivery (see text), (b) The study protocol for baseline (visit 1) and follow-up (visit 2) after 3 months. Medical therapy was maintained throughout the study protocol. Sodium intake was standardized to 150 mEq/d. On day 3, renal vein sampling was followed by multidetector computed tomography scan (MDCT), after which MSCs were infused into the renal artery without dilation of the stenotic renal artery. BOLD, blood oxygen level dependent; GFR, glomerular filtration rate; MRI, magnetic resonance imaging.

RESULTS:

Results of this study include first, comparison of the demographics and baseline measurements (Table 1 and 2); Second, changes in the affected kidney before and after treatment (Table 2); Third, changes in the contralateral kidney (CLK)over the the follow-up period which serves as an internal control and measure of the potential systemic effects (Table 3); Fourth, systemic effects such as blood pressure, two-kidney clearance and proteinuria (Table 4).

Table 1:

Demographics, kidney function and peak systolic velocity at Baseline

Characteristic	Low Dose (1.0 × 105 cells/kg)	Medium Dose (2.5×105 cells/kg)	High Dose (5.0 × 105 cells/kg)	Medical + MSC treated (Entire cohort)	Medical only treated (Entire cohort)	P-value ¶
Number of subjects (n)	6	7	6	19	18
Age, years	74.5 ± 3.8	73.2 ± 2.7	72 ± 5.6	73.26 ± 4.2	70.3±6.3	0.1
Gender (Male), %	50%	70%	50%	57%	39%	0.2
eGFR, ml/min	42.8 ± 11.3	46.5 ± 11.7	51 ± 11.5	46.7± 11.3	53±16	0.2
Iothalamate clearanc (Both kidneys)	44[39–48]	52[44–64]	57[54–61]	53[43–61]	53[44–66]	0.7
SBP, mm Hg	151 ± 16	140 ± 12	140 ± 12	144 ± 14	136±16	0.14
DBP, mm Hg	74±7	63±15	70±8	69±11	66± 8	0.5
Hemoglobin, g/dl	13.4± 1.7	13.1± 1.2	14.4± 0.7	13.6± 1.3	12.6± 1.2	0.02
Total urine protein, mg/24h	88[62–149]	116[38–286]	101[81–164]	101[57–238]	67[50–96]	0.09
Peak systolic velocity	287[285–289]	287[263–329]	348[275–417]	287[275–343]	326[285–380]	0.4
BMI	30.7±5	30.7 ± 5.7	31.8± 7.7	31±6	27.2±4.2	0.03

Data are presented as mean ±SD for normally distributed values or median [Q1–Q3] for skewed data.

^¶compares MSC-treated and Medical-only-treated entire cohorts. BMI, body mass index; DBP, diastolic blood pressure; SBP, systolic blood pressure

Table 2:

Volume, blood flow and BOLD imaging in the stenotic kidneys (STK)

Single Kidney Measurements		Medical + MSC treated Group			Medical only treated group			Baseline¶	Follow up¥
		Baseline	3 Months	p-value	Baseline	3 Months	p-value	p-value	p-value
STK Kidney volume and Blood flow	Total Kidney Volume, CC	93±39	94±39	0.4	93±42	91±41	03	0.95	0.8
	Cortical volume, CC	55±29	57±28	0.4	60±26	56±27	0.3	0.8	0.97
	Medullary volume, CC	36±13.4	36±13	0.34	33±17	33±17	0.4	0.6	0.5
	Cortical Perfusion, mL/min/cc	2.3[1.4–1.5]	2.6 [1.6–3.2]	0.03	2.3[1.−3.0]	2.2[1.9–2.6]	0.3	0.6	0.48
	Cortical blood flow, mL/min	126[61–199]	156[63–209]	0.02	131[92–211]	111[92–179]	0.18	0.5	0.46
	Medullary perfusion, mL/min/cc	0.65[0.56–0.96]	0.73[0.7–1]	0.06	0.9[0.7–1.2]	0.94[0.7–1.2]	0.3	0.042	0.35
	Medullary blood flow, mL/min	28.6[15–32]	31[17–38]	0.08	24[19–47]	28[16–42]	0.3	0.3	0.99
	Renal blood flow, mL/min	164±99	190±126	0.007	195±132	166±98	0.1	0.4	0.5
STK Hypoxia	Cortical R2*, sec−1	19.9±3	20±3	0.5	19.8±4.1	20.8±5.1	0.14	0.9	0.55
	Fractional hypoxia % R2*> 30,%	11[6–14]	8.3[3–10]	0.01	10.6[8–18]	9 [4–15]	0.5	0.2	0.14
sGFR		20.4±11.8	21.0±12.5	0.233	20.5±10.6	19.8±11.8	0.097	0.984	0.76

Data are presented as mean ±SD for normally distributed values or median [Q1–Q3] for skewed data. All numbers are single-kidney measurements.

^¶compare baseline and follow-up of MSC treated with Medical only treated Entire cohorts respectively. sGFR, single-kidney glomerular filtration.

^¥compare baseline and follow-up of MSC treated with Medical only treated Entire cohorts respectively. sGFR, single-kidney glomerular filtration.

Table 3:

Volume, blood flows and BOLD imaging in the contralateral kidneys (CLK).

Single Kidney		Medical + MSC treated Group			Medical only treated group			Baseline¶	Follow up¥
		Baseline	3 Months	p-value	Baseline	3 Months	p-value	p-value	p-value
CLK Kidney volume and Blood flow	Total Kidney Volume cc	132±35	131±37	0.35	141±38	136±36	0.07	0.5	0.68
	Cortical volume cc	83±25	86±29	0.12	92±27	88±27	0.14	0.3	0.8
	Medullary volume cc	49±14	45±18	0.09	49±16	48±14	0.8	0.9	0.6
	Cortical Perfusion, mL/min/cc	2.4[2.1–2.8]	3[2.3–3.5]	0.007	2.6[2.3–3.4]	2.8[2.4–3.3]	0.44	0.4	0.7
	Cortical blood flow	219±98	267±134	0.004	258±125	260±121	0.8	0.3	0.87
	Medullary perfusion, mL/min/cc	0.70[0.7–1]	0.90[0.80–1.20]	0.02	0.98[0.80–1.10]	1.13[0.90–1.40]	0.046	0.1	0.1
	Medullary blood flow, mL/min	41±17	45±16	0.12	45±18.5	55±23	0.02	0.5	0.14
	Renal blood flow, mL/min	260±110	312±145	0.003	306±133	315±135	0.6	0.3	0.94
CLK Hypoxia	Cortical R2*, sec−1	19±3	18.8±2	0.3	17.8±3.2	18.8±1.9	0.3	0.4	0.99
	Fractional hypoxia % R2*> 30,%	4.4[3.2–7.12]	3[1.4–4.5]	0.09	6.5[5–8.4]	9.3[7.1–10.7]	0.74	0.1	0.003
sGFR		33.1±13.4	35.1±14.3	0.103	34.6±12.4	36.1±13.1	0.31	0.739	0.837

Data are presented as mean ±SD for normally distributed values or median [Q1–Q3] for skewed data. All numbers are single-kidney measurements.

^¶compare baseline and follow-up of MSC treated with Medical only treated Entire cohorts respectively. sGFR, single-kidney glomerular filtration.

^¥compare baseline and follow-up of MSC treated with Medical only treated Entire cohorts respectively. sGFR, single-kidney glomerular filtration.

Table 4:

Blood pressure, GFR (Two kidneys) and 24hr Total urine protein before and 3 months after therapy.

Characteristich		Medical + MSC treated Group (Entire cohort, n=19)			Medical only treated Group (n=18)			Baseline¶
		Baseline	3 Months	p-value	Baseline	3 Months	p-value	p-value
Blood pressure	SBP, mmHg	144±11	136±13	0.04	136±16	135±19	0.8	0.1
	DBP, mmHg	69±13	65±15	0.09	66± 8	65±12	0.6	0.5
GFR (Two kidneys)	GFR, ml/min (Iothalamate)	53±17	56±16	0.053	55±19	58 ±21	0.8	0.7
	eGFR (MDRD)	47±11	50±15	0.026	53±16	55±23	0.35	0.21
Total urine protein, mg/24hr		101[57–238]	78[53–240]	0.3	67[50–97]	58[30–103]	0.5	0.09

The Modification of Diet in Renal Disease Study (MDRD) equation; SBP, systolic blood pressure, DBP, diastolic blood pressure

^¶compares baseline of MSC treated with Medical only treated Entire cohorts.

Demographics, Blood pressure, and GFR:

Demographic characteristics of the MSC treated groups and the medical treatment alone group are summarized in Table 1. Measured GFR (Iothalamate) values were: 53[43–61] ml/min for the entire MSC treated group and 53[44–66] ml/min for the Medical treatment alone group. The three different MSC groups did not differ in terms of age, sex, systolic blood pressure (SBP), diastolic blood pressure (DBP), or eGFR during the baseline measurement period (p>.05).

Kidney volume and blood flow measurements:

Stenotic Kidney (STK) measurements:

Kidney volumes, blood flows, and BOLD measurements are summarized in Tables 2 and 3. For both MSC and Medically treated groups, volumes and blood flow in the stenotic kidneys were lower as compared to the contralateral kidneys. Baseline RBF in STK was 164±99 ml/min as compared with the CLK 260±110 ml/min, (p=0.004), for MSC group. Similarly, baseline RBF in STK was lower than the CLK (195±132 versus 306±133 ml/min, respectively, p=0.0002), for the medically treated group. Baseline kidney volume in STK was 93±39 cc as compared with the CLK 132±35 cc, (p=0.008), for the medically treated group. Similarly, baseline kidney volume in STK was lower than the CLK (93±42 versus 141±38 cc, p=0.008) for the medically treated group.

In the MSC group, cortical blood flow increased from 126[61–199] to 156[63–209] ml/min, p=0.02 after three-months. Similarly, whole kidney renal blood flow increased from 164±99 to 190±126 ml/min, p=0.007. Total kidney volume in the MSC-treated group did not change.

Cortical blood flow in the medically treated group was unchanged from 131[92–211] to 111[92–179] ml/min,p=0.18. Renal blood flow was unchanged 195±132 to166±98 ml/min, p=0.1. Figure 2 illustrates the RBF in STK and CLK at baseline and after three months. Individual changes in RBF in the treated kidneys are illustrated in Supplemental figure1.

Figure 2|. Renal blood flow (ml/min) increased in both the (a) stenotic kidney (STK) and (b) contralateral kidney (CLK) after mesenchymal stem cell (MSC) administration after the 3-month hiatus.

No changes were observed in the Med Rx-only group. ▼, low dose; ●, medium dose; +, high dose.

Contralateral Kidney (CLK) measurements:

In the MSC group, cortical blood flow increased from 219±98 to267±134 ml/min, (p=0.004) after the three-month follow-up period. Similarly, renal blood flow increased from 260±110 to 312±145,p=0.003. In the medically treated group, cortical and renal blood flow did not change.

BOLD MRI measurements:

Stenotic Kidney (STK) measurements:

Measurements of whole kidney levels of hypoxia expressed at the percentage of R2* above 30 sec⁻¹, are summarized for both study groups at baseline and after three months in Figure 3. Fractional hypoxia >30 sec⁻¹ was higher in the STK as compared to the CLK (11 [5.5–13.7] versus 4.4[3.2–7.12] %, p= 0.0183) in the MSC group. Similarly, Fractional hypoxia >30 sec⁻¹ was higher in the STK as compared to the CLK (10.6[8–18] versus 6.5[5–8.4] %, p=0.006) in the medically treated group.

Figure 3|. Fractional hypoxia (expressed as % >30 s⁻¹) was higher in (a) the stenotic kidney (STK) as compared with (b) the contralateral kidney (CLK) at baseline (visit 1).

Fractional hypoxia %>R2*30 s ⁻¹ decreased after 3 months in the STK after mesenchymal stem cell (MSC) administration. No changes in oxygenation were observed in the Med Rx-only group. ▼, low dose; ●, medium dose; +, high dose.

In the MSC group, the fraction of the whole kidney with R2* above 30 sec⁻¹ decreased in the post-stenotic kidneys from (Median (IQR): (11% [5.5–13.7] to 8.3% [3.3–9.7]P=0.01).Fractional hypoxia expressed as % R2* >20 sec⁻¹ decreased from 62% [49.4–66.7] to 53.5% [34.4–65.6], P= 0.01 following MSC administration. Cortical R2* levels did not change consistently. Representative images for a patient from each group are depicted in Figure 4.

Figure 4|. Representative parametric maps of R2* that represent kidney oxygenation obtained by blood oxygen level-dependent magnetic resonance imaging.

(a,b) Transition to lower levels of R2*, especially in medullary zones in the post-stenotic kidney after mesenchymal stem cell (MSC) infusion. (c,d) R2* maps in a post-stenotic kidney treated with Med Rx-only. No change in the level of R2* was identified after 3 months.

In the medically treated group, the fraction of the whole kidney with R2* above 30 sec⁻¹ and cortical hypoxia did not change over the follow-up interval.

Contralateral Kidney (CLK) measurements:

For both MSC and medically treated group, the fraction of the whole kidney with R2* above 30 sec⁻¹ and cortical hypoxia did not change over the three months between studies.

Renal vein angiogenic and inflammatory markers:

Angiogenic Markers:

Stenotic Kidney (STK) measurements:

The levels of angiopoietin-2, VEGF-A, and VEGF-C sampled from the renal vein of the STK in the MSC treated group were higher than the levels from the CLK [(32±46 vs17.3±20 pg/ml/cc, P=0.0277), (1.8±2.3 vs0.94±1.0, P=0.013), and (8.3±5.3 vs 5.4±3.3 pg/ml/cc, P=0.019)], respectively.

In the MSC group, levels of angiopoietin-2, VEGF-A, and VEGF-C: decreased from [(32±46 to 25±32 pg/ml/cc, p=0.039), (1.8±2.3 to 1.3±1.4 pg/ml/cc, p=0.0195) and (8.3±5.3 to 7.2 ±3.9 pg/ml/cc, p=0.021)], respectively. Figure 5 illustrates the changes in the angiogenic markers. These changes were most evident in the subjects that had higher levels of the biomarkers above the levels observed from the CLK, depicted in the light background.

Figure 5 |. Overall levels fell for (a) angiopoietin-2, (b) vascular endothelial growth factor A (VEGF-A), vascular endothelial growth factor C (VEGF-C) (c) from the treated stenotic kidney.

Red dashed lines and shaded zones illustrate the range (mean ± SD) of levels obtained from the contralateral kidney (CLK) for comparison. Changes were most evident in individuals with high baseline levels, whereas no consistent changes were observed in those that were in the CLK range at baseline. MSC, mesenchymal stem cell; V1, visit 1; V2, visit 2.

Contralateral kidney (CLK) measurements:

No significant changes occurred in the level of angiopoietin-2, VEGF-A, and VEGF-C from the CLK over the follow-up period.

Inflammatory Markers:

Stenotic Kidney (STK) measurements:

The level of NGAL sampled from the renal vein of the STK in the MSC treated group was higher than the levels of the CLK (2.4±1.4 and 1.5 ±0.7 ng/ml/cc respectively p=0.0054) and decreased to 2.1±1.0 ng/ml/cc, (p=0.034) after three months.

Levels of NGAL sampled at the renal vein of the STK in the medical treatment alone group were higher than the levels from the CLK (2.0±1.3 and 1.1±0.6 ng/ml/cc respectively p=0.0015). No significant change occurred in the STK 2.0±1.3 to 2.4±1.9 ng/ml/cc, P=0.17.

In the MSC Treated group, levels of IFN-g sampled at the renal vein of the STK were 0.24±0.6 and decreased to 0.17±0.5 pg/ml/cc, p=0.01 over the follow-up period.

In the MSC treated group, levels of TIMP-2 sampled at the renal vein of the STK were higher than the levels of the CLK (1.1±1.04 and 0.5±0.35 ng/ml/cc, respectively p=0.0006) and decreased to 0.77±0.56 ng/ml/cc, p=0.044 over the follow-up period.

In the MSC Treated group, levels of TNFα sampled at the renal vein of the STK were higher than the levels of the CLK 0.17[0.09–0.22] and 0.077[0.062–0.017] pg/ml/cc respectively, p=0.009) and decreased to 0.14[0.09, 0.21] pg/ml/cc, p=0.024, No significant change occurred in the level of TNFα in the CLK over the follow-up period.

In the medical treated TNFα sampled at the renal vein of the STK was higher than the levels of the CLK 0.15[0.10, 0.23] and 0.07[0.05, 0.1] pg/ml/cc respectively, p= 0.007. No significant change occurred in the STK 0.15[0.10, 0.23] to 0.15[0.08, 0.20], p= 0.24 pg/ml/cc. Figure 6 illustrates the changes in these Inflammatory Markers.

Figure 6|. Renal vein levels of inflammatory cytokine markers obtained from the mesenchymal stem cell (MSC)-treated stenotic kidney before (visit 1) and 3 months after MSC infusion (visit 2).

Red dashed lines and transparent red rectangles represent the range (mean + SD) of these biomarkers obtained from the contralateral kidney. Renal vein levels of (a) neutrophil gelatinase-associated lipocalin (NGAL), (b) interferon-gamma (IFN-γ), and (c) tissue inhibitor of metalloproteinases 2 (TIMP-2) fell after 3 months after MSC administration.

Contralateral kidney (CLK) measurements:

No significant changes occurred in the level of NGAL, IFN-g TIMP-2, and TNFα in the CLK over the follow-up period.

Iothalamate clearance (GFR):

Glomerular filtration rates are summarized in Tables 4 and 5. In the MSC-treated group, total GFR for both kidneys tended to increase from 53±17 to 56±16 ml/min, p= 0.053. The increase was greatest in the high dose group (56.7±7 to 64±8 ml/min, p=0.045). GFR did not change in the medically treated group (55±19 to 58 ±21 ml/min, P=0.7). Figure 7 illustrates the dose-response changes in eGFR, DBP, and urine protein. Separately, Figure. 8 depicts line plots for the changes iothalamate GFR, SBP and Urine protein for the highest MSC dose (5.0 × 10⁵ cells/kg).

Table 5:

Blood pressure, GFR and urine protein changes by individual MSC dose.

MSC Dose related response		Low dose group (n=6)			Medium dose group (n=7)			High Dose Group (n=6)
		Baseline	3 Months	p-value	Baseline	3 Months	p-value	Baseline	3 Months	p-value
Blood pressure	SBP, mmHg	151±16	137±18	0.11	140±12	142±16	0.37	140 ±12	128±8	0.005
	DBP, mmHg	74±6	72±11	0.38	63±15	61±14	0.29	70±8	63 ±12	0.1
GFR (Two kidneys)	GFR, ml/min (Iothalamate)	42±15	45±16	0.2	58±20	58±18	0.1	57±7	64±9	0.045
Total urine protein, mg/24hr		88[62–149]	115[56–240]	0.07	116[38–286]	123[70–210]	0.4	101[81–164]	72[45–110]	0.0006

SBP, systolic blood pressure, DBP, diastolic blood pressure

Figure 7|. The dose-response changes in (a) estimated glomerular filtration rate (eGFR), (b) urine protein, and (c) diastolic blood pressure (DBP).

These effects were most evident in participants who received the highest mesenchymal stem cell dose (5.0 × 105 cells/kg).

Figure 8|. The changes in (a) glomerular filtration rate (GFR), (b) systolic blood pressure, and (c) total urine protein/24 h evident in the highest-dose group (5.0 × 105 cells/kg).

These changes were larger than those observed in the lower 2 doses. MSC, mesenchymal stem cell.

Urine Protein:

24 hr. total urine protein (mg/24hr) did not change in the medically treatment cohort, (67 [50–97] to 58[30–103] mg/d,p=0.5), low dose (88[62–149] to 115[56–240] mg/d,p=.07) and medium dose MSC-treated groups, (116[38–286] to 123 [70–210] mg/24hr, p=0.4),. However, 24 hr. total urine proteins decreased in all subjects who received the highest dose of MSC from [150.2±90 to 103±95,mg/24hr, p=0.0006]. Urine Proteins results are summarized in tables 4 and 5.

Systolic Blood pressure:

In the MSC-treated group, systolic blood pressure decreased from 144 ±11 to 136±13 mmHg, p=0.04, this change was most evident in the high dose group (140±12 to 128±8 mmHg, p=0.005). In the medically treated group, no significant change occurred (136±16 to 135±19 mmHg, P=0.8). Blood pressure results are summarized in Tables 4 and 5.

In-vivo Hypoxia attenuation and In-vitro MSC secretion of HGF:

Levels of growth factors secreted per cell measured in the supernatant during MSC culture were as followHGF median[IQR] (pg/ml/cell) 1.96×10⁻³ [1.69×10⁻³, 3.49×10⁻³], EGF(pg/ml/cell) 1.06×10⁻⁵ [6.52×10⁻⁶ 3.67×10⁻⁵], ICAM (ng/ml/cell) 3.8×10⁻⁶ [2.56×10⁻⁶ 6.39×10⁻⁶ ] and VEGF (pg/ml/cell) 4.95×10⁻⁴ [1.72 ×10⁻⁴, 6.0×10⁻⁴].

The amount of HGF secreted per cell correlated with the change in fractional kidney hypoxia (measured as the change in the fraction of the kidney with R2*>20 sec⁻¹ and R2*>30 sec⁻¹), Pearson rho=−0.61 and −0.4, P-value 0.01 and 0.1, respectively as illustrated in Figure.9.

Figure 9|.

Levels of mesenchymal stem cell (MSC)-secreted hepatocyte growth factor (HGF) in vitro correlated with the reduction in fractional hypoxia stenotic kidney hypoxia in vivo after MSC administration.

Discussion:

Our results demonstrate an increase in cortical and whole kidney blood flows in the post-stenotic kidney for subjects with ARVD over a three month period after intra-arterial infusion of autologous adipose-derived MSC. These changes were associated with partial reduction of renal vein inflammatory and angiogenic biomarkers and dose-related changes in GFR and blood pressure. No such changes were observed in a medically-treated group, for whom blood flows and GFR were stable or tended to decrease over the same interval. The change in RBF associated with MSC infusion for the post-stenotic kidney resulted from increased cortical and medullary perfusion as no measurable changes in kidney volumes were identified. Moreover, the contralateral kidneys of these subjects also showed increased cortical and medullary perfusion, as well as whole kidney RBF. Levels of tissue hypoxia within stenotic kidneys measured by BOLD MR fell in MSC treated subjects.

By contrast, the medical treatment only group had no change or a decrease in RBF in the post-stenotic kidney over the course of three months. Those results extend previous studies demonstrating reduced function in stenotic kidneys, which may be partially offset by compensatory hypertrophy in the contralateral kidney ¹⁴. Our previous study demonstrated the safety of intra-arterial infusion of MSC into the renal artery of subjects with ARVD ¹³. It has been suggested that instrumentation of the renal artery itself as part of the procedure might have produced a small rise in blood flow. Our current results argue against that possibility insofar as increased MSC doses produced hemodynamic effects in the untouched contralateral kidneys and dose-related changes in GFR and systemic blood pressure. Both of these observations make a direct effect of arterial manipulation unlikely.

We interpret the rise in cortical and whole-kidney RBF in the MSC treated group to reflect angiogenic and/or reparative capabilities of MSC to form new vessels and restore the microcirculation, consistent with experimental observations in a slowly developing swine model of atherosclerotic renal artery stenosis ¹⁵.

Not surprisingly, many stenotic kidneys demonstrated evident hypoxia, manifest as a larger proportion of the kidney with R2* levels above 30 sec⁻¹ as compared with the contralateral kidneys. We and others have shown that despite remarkably adaptive circulations, reduced blood flows ultimately magnify hypoxia and are associated with increases in inflammatory and injury biomarkers ^16–20. The fraction of the kidney with R2* more than 30 sec⁻¹ decreased by 25% in the STK. Similarly, we interpret the observed decreases in angiopoietin, VEGF-A, and VEGF-C to reflect partial attenuation of hypoxic stimulation, particularly in those kidneys with elevated renal venous biomarker levels.

This interpretation is consistent with other experimental data demonstrating overt tissue hypoxia associated with increased VEGF gene expression that falls after restoration of tissue oxygen ²¹.

For the first time, our study identified modest reductions in biomarkers associated with inflammatory injury in the MSC group. Renal vein levels of NGAL, IFN-γ, TNFα, and TIMP-2 from the STK were decreased three months after the MSC infusion, whereas such changes were not identified in the medically treated group or from the CLK. These changes may be related to the immunomodulatory properties of the stem cells ^{22, 23}. The exact mechanisms by which MSC may affect parenchymal inflammation are not known. Previous studies demonstrate that stem cells are able to impair dendritic cell maturation¹¹, reprogram macrophages from a pro-inflammatory to an anti-inflammatory state ^24,25 and reduce oxidative stress ²⁶. Although the subjects were randomly enrolled in the clinical trial, these observations partially reflect a regression to the mean phenomenon, although the rise in the low-level markers was minimal.

Moreover, MSC may differentiate into renal tubular cells as observed when MSCs were co-cultured for up to seven days with injured murine cortical tubular renal epithelial cells in-vitro ²⁷.

Overall GFR (iothalamate clearance) and eGFR of both kidneys increased modestly in our study, particularly in the high dose group (10%). No consistent change in GFR was identified in the medically treated group over the same follow-up interval.

Systolic blood pressure measured under standardized conditions decreased by 10 mmHg with no change in the antihypertensive medications in MSC treated subjects, whereas no change was identified in the medically treated group. This was associated with a modest decrease in urinary protein.

It is widely recognized that administered MSC do not remain in a single location. Tracking studies using labeled MSC suggest that while a modest level of MSC retention may occur in the post-stenotic kidney (8–12% after one month), as much as a third of that number can be identified in the contralateral kidney. This capacity for cell migration may account for observed systemic and effects in the contralateral kidneys of the MSC-treated cohort, such as the increase in kidney perfusion. The CLK were not evidently hypoxic and showed no consistent changes after MSC infusion. Overall changes in GFR, SBP and urine protein were more evident in the high dose group (5 × 10⁵ cells/kg). Functional characteristics of adipose-derived MSC vary between individuals. Previous studies indicate variation of migratory, proliferative and RNA expression profiles as a function of age and hypoxia ^{28, 29}. Studies of MSC properties suggest that renal failure itself may not adversely affect functional characteristics ³⁰. Our studies included measurements of supernatant cytokines during cell expansion, including hepatocyte growth factor (HGF). HGF is an angiogenic growth factor that, previous studies indicate may have a potential role in the treatment of peripheral and coronary vascular diseases^31–33. Levels of HGF per cell as measured from MSC in our study participants correlated with the attenuation of the kidney hypoxia three months after MSC administration. These data support defining and potentially modifying functional properties of MSC as a means to maximize the potential therapeutic efficacy of administered MSC.

Study limitations include the features of Phase 1a clinical trial: Numbers in each dose group were limited. Diabetic subjects were specifically excluded. Subjects were required to have reasonably preserved overall GFR (creatinine < 2.5 mg/dL) out of concerns for contrast and procedural exposure. Doses were administered in sequence by groups rather than in a randomized fashion. The comparison group with medical therapy only did not undergo arterial angiography or renal artery instrumentation. Follow-up was limited to the three month study period for this report. We cannot determine the long-term effects or potential need for repeated treatment with MSC infusion in these subjects.

Taken together, these data establish the potential for adipose-derived MSC to modulate microvascular perfusion of the kidney in human ARVD independent of large vessel revascularization. They indicate that tissue oxygenation in the kidney can increase with reductions in hypoxia, angiogenic biomarkers, and inflammatory biomarkers and can be associated with modest increments in GFR over a three-month interval. The changes in GFR and blood pressure were largest in the high dose MSC treated subjects with ARVD, which may be recommended as an effective dose. Our results support the potential role for MSCs to restore kidney function through immunomodulatory and proangiogenic properties ³⁴. Further studies related to the durability and potential role for adipose-derived MSC administration in clinical care are warranted.

Methods:

Study population:

Thirty-nine subjects with ARVD were studied in these protocols. Twenty-one subjects with ARVD were assigned sequentially to three different dose groups of seven subjects each (1, 2.5 and 5.0 × 10⁵ cells/kg) between 2013–2017 as part of a study registered with ClinicalTrials.gov (NCT02266394).

These results were compared with data from 18 subjects with ARVD matched for age, kidney function and blood pressure previously studied under an identical study protocol treated with medical therapy alone ^{14, 35} Except for renal vein biomarkers study. MSC group were tested for more biomarkers. ARVD was defined according to criteria analogous to the CORAL entry criteria for subjects with more than 60% stenosis, identified by CT angiography, renal artery duplex ultrasound or arterial angiography as previously described ^{7, 36}.

Subjects were admitted to the Clinical Research and Trials Unit (CRTU) of St. Mary’s Hospital, Rochester, MN for three days on two occasions three months apart. Subjects treated with MSC were enrolled into three consecutive groups with increasing doses (Fig 1a). Evaluations of kidney function and patient visits are shown (Fig 1b). Inclusion criteria included age between 40 and 80 years, hypertension (systolic BP>155 mmHg), and/or requirement for two or more antihypertensive medications. There were no restrictions on antihypertensive agents, although loop diuretics were changed to diluting site agents (e.g. hydrochlorothiazide, indapamide, or metolazone) prior to the study. Serum creatinine was ≤2.5mg/dL. Angiotensin-Converting Enzyme (ACE inhibitor) or Angiotensin Receptor Blocker (ARB) therapy was maintained during these studies. Subjects with the following criteria were excluded: diabetes requiring insulin or oral hypoglycemic medications, known allergy to furosemide, pregnancy, or a cardiovascular event within three months: myocardial infarction, stroke, congestive heart failure, or cardiac ejection fraction less than 30%. Two subjects (from lowest and highest doses) were excluded for technical issues related to inaccessible vessels at angiography. The analysis was completed for 19 MSC-treated subjects. subjects

MSC isolation, preparation, and safety evaluation:

Six weeks before the scheduled MSC administration, a subcutaneous abdominal fat biopsy sample (approximately 2 g) was obtained under sterile conditions. Fat digestion was followed by selection and expansion of plastic-adherent cells. Cells were cultured in Advanced MEM (Gibco Cat#12492021) supplemented with human platelet lysate (PLTMax®, Mill Creek Life Sciences) over two weeks using good manufacturing practices as previously described until a sufficient number of cells was obtained for each treatment protocol ³⁷. Cells were cryopreserved and samples used for release testing. Release criteria for administration after final passage included sterility testing using anaerobic and aerobic culture, endotoxin, mycoplasma, karyotype, and FACS for surface markers characteristic of MSC as previously described ^{13, 37, 38}. Four to six days before infusion, an aliquot was thawed and re-plated in growth media. After final release studies, cells were transferred for an intra-arterial administration diluted in sterile Lactated Ringer’s solution.

In-vitro MSC secretion of cytokines/biomarkers:

Additional Mesenchymal stem cells from the subjects were grown in T-75 cm² flasks at a density of (1.0–2.5 × 10³) cells/cm² in Advanced MEM* with 5 % PLTmax (Mill Creek Life Sciences, Rochester, MN, USA) and 2 mM L-glutamine (Invitrogen, Carlsbad, CA, USA) in a 37 °C and 5 % CO2 incubator for 3–4 days. When cells were 60–80 % confluent, they were passaged using TrypLE (Trypsin-like Enzyme, Invitrogen) ³⁷. Cell count was quantified with a Countess hemocytometer (Thermo Fisher Scientific, Inc., NY, USA). Aliquots of the conditioned media were collected before harvesting the cells and stored at −80 °C until they were assayed.

Hepatocyte growth factor (HGF), vascular endothelial growth factor (VEGF), epidermal growth factor (EGF) and Inter-Cellular Adhesion Molecule-1 (ICAM-1) were measured by enzyme-linked immunosorbent assays (ELISA) according to manufacturer protocols as previously described ²⁸.

Inpatient Study Protocol:

On the first study day, subjects were admitted to the Clinical Research and Trials Unit (CRTU) to maintain a specified sodium intake (150 mEq daily). GFR was measured using iothalamate clearance with three consecutive, timed periods. On the second day, Blood Oxygen Level Dependent (BOLD) MRI was performed before and after administration of furosemide (20 mg i.v.). On the third day, renal vein blood samples were obtained from each kidney for measurement of neutrophil gelatinase-associated lipocalin (NGAL), tissue inhibitor of metalloproteinases 2 (TIMP2), interferon-gamma (IFNγ), tumor necrosis factor alpha (TNFα), vascular endothelial growth factor (VEGF-A), vascular endothelial growth factor (VEGF-C) and angiopoietin-2 as previously described ^{36, 39}. Results were expressed as ng or pg/ml/cc of the associated kidney volume as previously described ¹⁶. Renal blood flow for each kidney was measured by examination of transit times of centrally administered contrast using a Multidetector CT scan (MDCT) as previously described ^40–42. The medical treatment alone group did not receive further treatment while the MSC-treated group underwent infusion of MSC into the renal artery of the stenotic kidneys without further intervention regarding the stenosis. (Fig. 1B) illustrates the study protocol.

Measurements of kidney function in contralateral kidneys (CLK) not infused with cells in those subjects undergoing treatment served as an internal control and measure of the potential systemic effect or adverse events associated with MSC treatment.

Kidney GFR:

Iothalamate clearance (iothalamate meglumine, Conray, Mallinckrodt) was determined over three 30-minute timed collection periods after oral hydration (20 ml/kg). Additionally, estimated glomerular filtration rate (eGFR) was determined using the MDRD eGFR equation at Baseline and three months. Single kidney–GFR (sGFR) was determined by apportioning the measured iothalamate clearance by percentage of blood flow for each kidney ⁴³.

Blood pressure measurements:

Blood pressures were obtained utilizing an automated (Omron) oscillometric device during day one of both baseline and the three months follow up visit. The average of three blood pressure measurements was used.

Blood Oxygen Level Dependent (BOLD) MRI imaging for tissue oxygenation:

On the second day, Blood Oxygen Level Dependent (BOLD) MR scans were performed on a 3.0-T MRI (GE Discovery MR750W; GE Medical Systems, Waukesha, WI) with a twelve-channel torso-phased array coil. A fast 2-dimensional gradient echo BOLD sequence with twelve echo times was used as previously described ^{13, 35}. BOLD MRI utilizes the paramagnetic properties of deoxyhemoglobin, which alter the T2* relaxation time of neighboring water molecules and in turn influence the MRI signal of T2*-weighted (gradient echo) images.

BOLD data were quantified using a Matlab Graphical User Interface (The MathWorks, Natick, MA) program developed in our laboratory by drawing ROIs on T2* images and then transferring that ROI to R2* BOLD maps. For each slice of the kidney where BOLD images were acquired, one ROI was drawn to include the total-kidney (cortex and medulla) and another to be limited to the cortex alone. Cortical oxygenation was assessed by the average cortical R2* value and whole kidney levels of hypoxia by measuring the fractional tissue hypoxia (percentage of the whole slice >20 and 30 sec⁻¹), which sample primarily the medulla as previously described ⁴⁴.

Renal vein inflammatory markers:

On the third day, the femoral vein was cannulated and a catheter was advanced to obtain individual renal vein blood samples for NGAL, IFNγ, TIMP2, TNFα, VEGF-A, VEGF-C, and angiopoietin-2.

NGAL and TIMP-2 were tested by ELISA according to the manufacturer’s protocol (BioPorto Diagnostics, catalog no. KIT 036), TNFα, IFNγ, VEGF-A and VEGF-C and angiopoietin-2 by Luminex (Millipore, Billerica, MA, USA). Samples were frozen and analyzed as paired samples for each patient after completion of the follow-up visit. Elevated” STK levels were further defined as those above the range of values obtained from the CLK (Mean ± STD). Renal vein samples in the medical therapy only group were limited to NGAL and TNFα.

MDCT imaging technique for tissue perfusion and volume:

After obtaining renal venous samples, a 5-F pigtail Cobra catheter (Cook, Bloomington, IN.) was advanced to the right atrium for injection of iodinated contrast. Functional and volume studies were performed using a 64-slice dual-source (SOMATOM Definition, Siemens Medical Solutions, Forchheim, Germany) shortly after an injected bolus of iopamidol 370 (0.5mL/kg not exceeding 40mL). Images of four adjacent 7.2-mm slices were obtained at each of the successive 45 time-points over approximately 2.5 minutes. Fifteen minutes later, a second bolus of contrast was injected. A helical multidetector-CT study was then performed in the axial plane with a slice thickness of 5-mm, which covered the entirety of both kidneys and was used to determine cortical and medullary volumes ⁴⁰. MDCT images were reconstructed and quantified using the Analyze software package (Biomedical Imaging Resource; Mayo Clinic Rochester, Mn). In brief, regions of interest (ROI) were traced on the helical scans in the cortex and medulla on every slice of both kidneys to measure the volume. Additionally, ROIs were drawn in the cortex, medulla, and aorta during the contrast transit images for hemodynamic measures. ROIs drawn on the functional MDCT images were used to plot time-attenuation curves which were then quantified using in an in-house Matlab (The MathWorks, Natick, Mass). A graphical user interface where a gamma-variate curve fitting model was used to determine perfusion.

MSC administration:

After the MDCT studies were completed, an angiographic catheter was advanced through the femoral artery into the single most-severely stenotic renal artery. MSC which had been suspended in 10 ml of Lactated Ringer’s solution were infused slowly into a single renal artery distal to the stenosis over 5 minutes as previously described ¹³.

Statistical Analyses:

Statistical analysis was performed using JMP software, version 13.0.0 (SAS Institute, Cary, NC). Results were expressed using mean values and STD or median values [interquartile range(IQR)] as appropriate. Qualitative variables were expressed as number (percentage). Comparisons between the two groups of subjects with RVD were performed using two-sample t-tests with unequal variance or Wilcoxon rank-sum test for skewed data for continuous variables as appropriate, and a chi-squared test for categorical variables. Comparisons between two kidneys within the same individuals and repeated measurements for specific kidneys within individuals before and after treatment were performed using paired t-tests or Wilcoxon signed-rank test as appropriate. Pearson correlation method was used to determine the correlations.

Supplementary Material

1

Figure 1 S: depicts individual line plots for the changes from baseline RBF for the MSC treated participant (A) and Med Rx participants (B)