Cell-Based Therapy To Boost Right Ventricular Function And Cardiovascular Performance In Hypoplastic Left Heart Syndrome: CURRENT APPROACHES AND FUTURE DIRECTIONS

Abstract

Congenital heart disease remains one of the most frequently diagnosed congenital diseases of the newborn, with hypoplastic left heart syndrome (HLHS) being considered one of the most severe. This univentricular defect was uniformly fatal until the introduction, 40 years ago, of a complex surgical palliation consisting of multiple staged procedures spanning the first 4 years of the child’s life. While survival has improved substantially, particularly in experienced centers, ventricular failure requiring heart transplant and a number of associated morbidities remain ongoing clinical challenges for these patients. Cell-based therapies aimed at boosting ventricular performance are under clinical evaluation as a novel intervention to decrease morbidity associated with surgical palliation. In this review, we will examine the current burden of HLHS and current modalities for treatment, discuss various cells therapies as an intervention while delineating challenges and future directions for this therapy for HLHS and other congenital heart diseases.

1. Introduction

1.1. Diagnosis

Congenital heart disease (CHD) remains one of the most frequently diagnosed congenital diseases of the newborn, with Hypoplastic Left Heart Syndrome (HLHS) one of the most severe. HLHS is characterized by dysgenesis of the left ventricle resulting in profound distortion of the cardiac anatomy and failure of systemic circulation. HLHS can be detected by the second-trimester (18 to 24 week) ultrasound (Figure 1)[1–4], which facilitates appropriate preparation for the post-natal course, coordination of care for a medically complex infant and adequate time for transfer to a specialized center. While early detection is now commonplace, HLHS can infrequently remain undetected until after birth.

Figure 1:

Hypoplastic Left Heart Syndrome. Legend: Diagram showing A) Normal cardiac anatomy B) Hypoplastic left heart syndrome. Four-chamber view fetal echocardiography demonstrating C) Normal US findings and D) Hypoplastic left heart syndrome. RA= Right Atrium RV= Right Ventricle, LA= Left Ventricle LV= Left Ventricle

Prenatal HLHS babies with an intact atrial septum may be candidates for surgical atrial septectomy or balloon atrial septostomy where the timing depends on center preference and patient condition.[5–8] Selective prenatal HLHS babies with severe aortic stenosis can undergo aortic valvuloplasty in order to improve blood flow to the left side of the fetal heart which may reduce progression of HLHS.[5] Selective prenatal HLHS babies with an absent or small opening in the atrial septum are candidates for catheter-based atrial septal interventions where a large enough opening is created to allow blood to flow through the fetal heart.[9]

1.2. Physiologic Transition

In-utero, pulmonary blood flow is limited and oxygenated blood from the placenta enters the right side of the heart, transitioning to the systemic circulation via the patent ductus arteriosus (PDA) that connects the pulmonary artery to the aorta. There is also limited flow to the left atrium and ventricle across the patent foramen ovale. The natural transition following birth involves a number of physiologic changes that should lead to a normal transition from the fetal to the neonatal circulation.

Immediately after birth, as the lungs expand there is a gradual decrease in pulmonary vascular resistance (PVR). This facilitates, over a 48–72 hour period, an increase in blood flow into the pulmonary vascular bed. Simultaneously, this decrease in PVR allows blood flow across the closing PDA to transition to a primarily left to right flow until it eventually closes functionally and anatomically.

In HLHS, this transition heralds the onset of clinical symptoms. As blood blow starts moving towards the pulmonary arteries, there is reduced systemic flow from the diminutive, hypoplastic left ventricle. This leads to poor systemic perfusion, relative hypotension and metabolic acidosis. Additionally, excessive blood flow to the pulmonary circulation produces increased hydrostatic pressure and pulmonary edema.

Without intervention, closure of the PDA in the neonate with HLHS leads to persistent and worsening systemic hypotension, hypoxemia, acidosis, and eventually circulatory failure and death.

To address this, a series of surgical interventions have been devised to reconfigure the right ventricle to ultimately become the systemic ventricle in the HLHS patient. For postnatal HLHS babies, three or more surgeries are required depending on the severity of the defect.[10]

1.3. Epidemiology

The epidemic of congestive heart failure (CHF) is a growing worldwide concern and is expected to become worse. In adults, there are approximately 15 million people worldwide and 6.2 million patients in the United States who have CHF, with 400,000 new cases per year in the U.S. alone. In children, there is an equal trend of CHF with an increase in prevalence in the last decade.[11] There were over 7 thousand Emergency Department (ED) visits and 12 thousand CHF-related hospitalizations in children in 2016. Children with CHF-related hospitalizations have a 24-fold increased cost vs other childhood diagnoses at a median cost of almost two hundred thousand dollars per admission.[12]The cost for children with CHF is almost 1 billion dollars[11], on the same scale as that spent for all cancer-related diagnoses ($2.24 billion) and myocardial infarction ($3.18 billion). These numbers in children are increasing due in large part to that fact that patients with CHD are now living longer than the number of those being born.

HLHS is a prime example of a disorder that was once uniformly fatal, but now with heroic surgical intervention, patients survive into their teenage and young adult years. With development of staged surgical treatments, outcomes of surgical palliation for HLHS have steadily improved. Approximately 1025 infants are born each year with HLHS[13], and 318 die during the neonatal period for a death rate of approximately 33%. The cost of managing these patients is substantial with typical hospitalizations for the neonatal first-stage surgery of $280, 909.[14] Despite surgical improvements the overall 5-year survival rate remains limited with reported survivals over 50%−60%, with cardiac transplantation remaining as the only alternative for patients with failing circulations. [15, 16]

As an increasing number of these patients survive the neonatal period there is a growing population of patients who present later in life with failure of the single ventricle. Maintenance of ventricular function for these patients is therefore very important for long term survival. All patients who require cardiac transplantation have a waiting list mortality of over 20%. In addition, even after cardiac transplantation, HLHS patients have a decreased survival rate when compared to patients transplanted for other causes. [17, 18] Although the expenditure for heart transplantation exceeds that for other operations that are required for these HLHS patients, patients who fail these other operations for HLHS (see the following) due to ventricular dysfunction prove to be ultimately even more costly and have increased postoperative morbidity than heart transplantation by itself.[19]

Before consideration for transplant, patients with CHF are optimized with medical treatment. CHF in infants with HLHS is not related to typical coronary ischemia and thus revascularization is not applicable. Treatment such as extracorporeal membrane oxygenation (ECMO) or left ventricular assist device (LVAD) therapy are limited by the size of the patient and the duration of support on these devices. Furthermore, support of HLHS patients with these devices is technically challenging because they have only one ventricle.

2. Surgical Palliation

2.1. Surgical Steps

Since the early 1980s, a series of surgical interventions has been used to create a viable circulation in these infants, allowing the right ventricle to serve as the sole pump for systemic circulation. Currently, the strategy consists of three palliative operations.

Stage I is the Norwood operation, Stage II is the bi-directional cardiopulmonary anastomosis BDCPA / Glenn or Hemi-Fontan Procedure, and Stage III is the Fontan operation.[16] The overall objective of the staged surgeries is to allow the right ventricle to serve as the systemic ventricle, ejecting blood into a surgically expanded aortic root, and for the systemic venous blood to return passively to the pulmonary circulation in the absence of a right ventricular (RV) pumping chamber.

2.1.1. Norwood

The goals of the Stage I Norwood operation are to reconstruct the aortic arch, create an atrial septal defect, and establish reliable pulmonary blood flow. With these anatomic modifications, the single RV becomes responsible for pumping blood to both the body and lungs. Typically, pulmonary blood flow is achieved by either a modified Blalock-Taussig shunt (mBTS) using a polytetrafluoroethylene conduit to connect the subclavian artery to the pulmonary artery, or a Sano shunt which connects the right ventricle to the pulmonary artery using a similar polytetrafluoroethylene conduit (Figure 2). A prospective randomized study compared the outcomes of the two shunt types [20], and reported that the Sano shunt had better survival in the first year, but intermediate-term outcomes were not different for the two shunt types (Figure 3).

Figure 2:

The Norwood Procedure with a Modified Blalock – Taussig Shunt and a Right Ventricle – Pulmonary Artery Shunt. Legend: The Norwood procedure: a ‘neoaorta’ and the isolated pulmonary artery. Pulmonary blood flow is directed to the lungs by either a modified Blalock–Taussig shunt (top) or a right ventricle–pulmonary artery shunt (bottom).

Figure 3:

Primary Outcome and the Hazard of Death or Transplantation at 12 Months, According to Shunt Assignment. Legend: Panel A shows the Kaplan–Meier curves for transplantation-free survival among all infants who underwent the Norwood procedure, according to the intention-to-treat analysis. Panel B shows the estimated hazard of death or transplantation among all infants who underwent the Norwood procedure, according to the assigned shunt. P = 0.02 for the difference in the treatment effect for the period before and the period after 12 months. MBT= modified Blalock–Taussig RVPA = right ventricle– pulmonary artery.

Cardiovascular surgeons continue to use both shunt variations pending long-term findings on whether one method is superior to the other. Sometimes in high-risk HLHS patients who may have severe tricuspid regurgitation or severe right ventricular dysfunction, the Stage I Norwood operation is not recommended. Instead, the Stage I Hybrid operation is preferred which uses a median sternotomy approach for pulmonary artery banding, patent ductus arteriosus stenting, and an atrial septostomy. With these Hybrid patients, the Stage II operation occurs at 4–6 months of age and involves the BDCPA operation and the Norwood operation.

2.1.2. The Glenn procedure

The Stage II operation, the BDCPA operation, is typically performed at 4 to 6 months of age. The procedure connects the superior vena cava to the right pulmonary artery to bypass the right ventricle and allow systemic blood to flow directly to the pulmonary circulation. The second part of the BDCPA operation includes ligation of the pulmonary shunt placed during the Norwood procedure (either the mBTS or the Sano).

2.1.3. The Fontan procedure

Finally, the Stage III Fontan operation, performed when the child is between 18 months to 3 years of age, directs inferior vena caval flow into the pulmonary artery, thus bypassing the heart. At this stage, all of the venous return flows passively to the pulmonary circulation, with oxygenated blood returning from the lungs to the RV from where it is pumped systemically, finalizing the reconfiguration of the RV to the systemic ventricle. This procedure, which has allowed HLHS patients to survive, has created the long-term sequela of RV failure given that the ventricle must endure substantially higher loading conditions throughout life.

2.2. Morbidity After Palliation

Following these staged procedures, patients are left with a functional albeit abnormal circulation. As mentioned above, the RV must now serve as the systemic ventricle ejecting blood via the neoaorta to the body. In the absence of a subpulmonic ventricle, desaturated blood passively drains from the upper and lower extremities into the pulmonary bed to be oxygenated. When successful, this circulation allows for improved systemic saturations, resolution of volume overload and facilitates survival of many patients into adulthood. However, the Fontan physiology is also associated with increased central venous pressure and decreased cardiac output. [21]

Fontan physiology has improved survival in patients with single ventricles, however morbidity-free survival is far less likely. The Fontan circulation, due to the non-pulsatile pulmonary blood flow, leads to endothelial dysfunction and reduced nitric oxide (NO) production. As a result, there is a gradual and persistent rise in pulmonary vascular resistance (PVR) and eventually Fontan circulatory failure.[22] Patients with right-sided univentricular physiology, like HLHS, are at the highest risk of death, failure and subsequent need for transplant.[23]

Heart failure is a feature of almost half of these patients, with early age of correction seeming to be a protective factor.[24] Patients present with typical features including fatigue and exercise intolerance. The right ventricle is morphologically not designed to be systemic, and the hemodynamic changes produced by a Fontan circulation often result in compromise of the atrio-ventricular valve leading to right ventricular dilation with decreased function, further worsening the already reduced cardiac output.[25] Further, molecular analysis has shown that Fontan hearts have an abnormal myocardial gene expression pattern, similar to that seen in adult patients with heart failure. [26]

The largest and most comprehensive outcome analysis of patients undergoing a Stage I Norwood operation was reported by the Congenital Heart Surgeons’ Society (CHSS) in 2003. [27] Between 1994 and 2000 twenty-nine institutional members of CHSS enrolled 985 neonates who had either critical aortic stenosis or atresia. A total of 710 of the 985 patients underwent a Stage I Norwood operation. The survival was 76% at 1 month, 60% at 1 year, and 54% at 5 years. Numerous studies have determined the physiologic risk factors for poor perioperative outcome after Stage I, with conflicting results[28–32]. Surprisingly, RV function was not reported as a predictor of survival. This is an unexpected finding since in other congenital heart surgeries for other malformations poor ventricular function has been associated with adverse outcome after operative procedures. [33] There are two possible explanations for these results: 1) previous analyses of RV function in the setting of HLHS have been limited to qualitative assessment of the right ventricle, and 2) other perioperative factors in the initial unstable hemodynamic state may overshadow the impact of RV function in the early post-stage I Norwood period.

More accurate and quantitative evaluations of the right ventricle using the biplane pyramidal approximation method by echocardiography (ECHO) and phase contrast magnetic resonance imaging (PC MRI) [34–36] do demonstrate that the systemic RV, working against the high-pressure systemic circulation, becomes dysfunctional, which may correlate with decreased survival. In fact, Altman and colleagues reported that a new qualitative assessment of the right ventricle using biplane pyramidal approximation method by ECHO was an operative predicator of survival after Norwood stage I. [34, 35] Their study revealed that patients with decreased RV function, which they defined as a qualitative functional index 2.5 or less had a 35% survival rate compared to a 70% survival rate in patients with normal RV function during 18-month follow-up.

Measurements of RV function:

Several measures of RV performance are shown to correlate with survival, including RV strain and TR. Change in longitudinal strain (ΔLS), which measures percentage change in the length of the RV during systole, is another measure of RV function. In a cohort of 27 HLHS patients, a ΔLS greater than 8.7% (p=0.0004) predicted the composite outcome of death or heart transplant, 1 month after Norwood and persisting through the stage II BCDPA procedure.[37] Other studies using PC MRI have reported similar findings of decreased function of the RV in HLHS. Vincenti and colleagues showed that patients with HLHS after BCDPA have a reduced baseline EF compared to other RV dominant SV physiologies (55 +/− 7% vs 58 +/− 3% p=0.001) and also had a greater decline in EF following BDCPA which did not recover prior to Fontan.[38] Patients with decreased RV function and/or TR have a four-fold increased odds of interstage death between BCDPA and Fontan operations.[39] These studies emphasize the importance of RV function as a potential surrogate for survival rates and suggest a possible clinical value of strategies that boost RV function in HLHS patients.

Many HLHS survivors are now living longer and if they have RV dysfunction, they may eventually require heart transplantation, the only remaining treatment option.[40] Other HLHS patients die suddenly due to complications that rely on a systemic right ventricle.[40] This has led to the hypothesis that using cell-based therapy for HLHS patients may boost RV function or at least prevent RV dysfunction, and in so doing improve clinical outcomes, reducing both mortality and the need for transplantation.

Neurodevelopmental outcomes are an increasingly relevant part of the outcome measures for patients with HLHS as survival as continues to increase. Multiple studies have shown that there is a clear decrease in neurodevelopmental metrics among patients with single ventricle physiology. The surgery type does not seem to have an impact based on analyses of various cohorts to date. Deficits are noted across multiple metrics including gross motor, language, and intelligence quotient.[41] It is likely that the cause of these results is multifactorial including repeated hospitalizations, periods of illness and additional factors yet to be defined. Additional complications from the Fontan physiology include liver fibrosis. In one cohort, all patients had some degree of fibrosis as determined by biopsy, with time since Fontan having a direct correlation with degree of fibrosis.[42]Thromboembolic events are common among patients as a result of the persistent stasis, deranged coagulation factors, conduit stenosis and other individual patient factors.[43, 44]Morbidities related to bone density[45] are also well documented.

3. Cell-based Therapy

3.1. Animal Models

CHD is complex and multi-factorial resulting from genetic factors, cell-cell interactions and intrauterine mechanical factors. This has led to great difficultly creating animal models which are critical to gaining a better understanding of the disease and testing potential therapeutic strategies.

Hemodynamic models produce right ventricular strain, a key feature of HLHS, using primarily pulmonary artery banding in small animals. In a rat model, Hoashi and colleagues performed intramyocardial injection of skeletal myoblasts and demonstrated improved diastolic function and decreased ventricular fibrosis.[46] Similarly, pig models demonstrated preserved RV function after mesenchymal stem cell (MSC) and umbilical cord blood mononuclear cell intramyocardial injections (Figure 4).[47, 48] In a similar sheep model with epicardial delivery of cord blood stem cells investigators showed improved end-systolic elastance and stroke work in the increased RV load groups compared to placebos after 1 month. [49] While these models do not fully recapitulate all of the features of HLHS, they model RV overload which is the key lesion in ventricular failure in HLHS and as such provide valuable animal data for the use of cell-based therapy as a possible therapeutic modality in CHD.

Figure 4:

Echocardiographic assessment of neonatal Yorkshire swine following pulmonary arterial banding and RV injection with placebo or MSCs. (adapted from Wehman[48]. Legend: A–C: RV FAC, EDA, and ESA are equivalent pre- and postbanding. At 4 wk, the RV FAC in MSC-treated swines was preserved relative to baseline and postbanding values, whereas the FAC of placebo-treated swines was significantly reduced. The ESA and EDA in the placebo group were significantly enlarged at 4 wk relative to the MSC-treated swines, which had relatively preserved size. D–G: representative echocardiographic images of the RV immediately and at 4 wk postbanding in the placebo-treated and MSC-treated swines. EDA =end-diastolic area ESA= end-systolic area FAC =fractional area of change; MSCs= human mesenchymal stem cells RV= right ventricle

Investigators continue to attempt to produce animal models that more fully represent the spectrum of HLHS pathology. Reducing left ventricular preload produces LV hypoplasia and current models utilize lambs[50] and chicks[51] to create phenotype similar to HLHS emphasizing the LV hypoplastic component of the pathology. Additionally, a di-genetic model of knockout of Sap130 and protocadherin 9 (Pcdha9) has recently been developed that produces rat pups with ventricular hypoplasia and aortic arch abnormalities but no specific HLHS phenotype.[52] Most recently, Rahman and co-workers reported a novel murine model with complete recapitulation of the HLHS phenotype. [50] This is surgically induced by using an embolizing agent that partially impedes blood flow to the left atrium.[53] None of these models have been tested in the context of stem cell therapies.

3.2. Cell Therapy Sources, Timing, and Route of Administration

Cell therapy delivers cells to the region of injury to promote repair and or regeneration[54] and has shown some promise in adult patients with ischemic and nonischemic dilated cardiomyopathy.[55–62] A small collection of studies have shown that cell based therapy may play a role in ameliorating right ventricular failure in HLHS patients.[61, 63–67] The choice of cells for therapy included MSCs derived from bone marrow or umbilical cord, umbilical cord derived mononuclear cells, cardiosphere-derived cells, and C-kit+ cardiac progenitor cells. MSCs are particularly attractive because of their potential to be used as an off the shelf allograft. There are a limited number of studies performed in pediatric heart disease, which are summarized in Table 1.

Table 1.

Summary of pre-clinical studies in pediatric heart disease.

Dose	Animal model	Infusion Method	Time of delivery post injury	Results	Ref
6.6×106 of human pediatric CPC	Athymic Crl:NIH-Foxn1rnu rat	Intramyocardial	2 weeks	- ↑ RV ejection fraction - ↑ tricuspid annular plane systolic excursion - ↓ fibrosis	Agarwal[77]
5×105 of TMSC seeded on graft and cultured for 2 weeks	Landrace pig	Cellularized graft implantation	No injury	- ↑ myocardial strain - ↑ cardiac remodeling - ↑ endothelialization - ↓ fibrosis	Albertario[70]
4×106 of human neonatal TMSC	RNU nude rats	Cell sheets transplant	2 weeks	- ↑ survival rate - ↑ RV capillary density - ↓ fibrosis	Chery[71]
10 to 20×106 autologous skeletal muscle cells	Ile de France sheep with dexamethasone	Epicardial injection	No injury	Showed engraftment of cells in myocardium without the need of expanding cells in culture	Borenstein[8 2]
8×106 of hCBMNCs/kg	Border-Leicester lamb	Intracoronary	During cardiopul monary bypass surgery	Hemodynamically stable and shown highly abundant myocardial distribution without adverse effect	Brizard[81]
Unknown amount of hCBMNCs	Leicester sheep with cyclospor in	Epicardial injection	30 min	- ↑ end-systolic elastance - ↑ stroke work slope	Davies[106]
5×105 rat cardiosphere derived cells	Lewis rats	Intracoronary	10 days	- ↑ LVEF - ↓ LV end-systolic volume - ↓ myocardial inflammation and T-cell infiltration - ↓ fibrosis	Nana-Leventaki[74]
7.5×105 hEMSC on CF-ECM	Sprague Dawley rats	CF-ECM implant into epicardial surface of the MI area	24 hours	EMSCs were distributed from the epicardium to the endocardium.	Schmuck[107]
1×106 hBM-MSC	Yorkshir e swine with cyclospor ine and methylpr ednisolone	intramyocar dial	30 minutes	- ↑ neovessel formation - ↓ recruitment of endogenous c-kit+ CSC - ↑ proliferation of cardiomyocyte and endothelial cells - ↑ RV function	Wehman[48]
5×105 in monolayer or spheres of human pediatric C-kit+ CPC	Athymic Crl:NIH-Foxn1rnu rat	intramyocar dial	2 weeks	Only spheres and not in monolayers: - ↑ RV function - ↑ angiogenesis - ↓ fibrosis	Trac[78]
1×106 hC-kit+ CPC	Yorkshire swine with cyclospor ine and methylpr ednisolone	intramyocardial	30 minutes	- ↑ arteriole formation - ↓ myocardial fibrosis	Wehman[79]

TMSC- thymus derived mesenchymal stem cell; hCBMNC-human cord blood mononuclear cell; CPC – cardiac progenitor cells; BM-MSC – bone marrow derived mesenchymal stem cell; hEMSC – human embryonic cells derived mesenchymal stem cells; CF-ECM – cardiac fibroblast derived extracellular matrix; hiPS-CMs – human induced pluripotent stem cell derived cardiomyocytes; PGA – polyglycolic acid; PLCL – poly(l-lactic-ε-caprolactone); hBM-MSC – human bone marrow derived mesenchymal stem cell

3.2.1. MSCs

MSCs are a group of multi-potent cells that can be isolated from bone marrow, umbilical cord blood, adipose tissue and peripheral blood. They are identified based on cell surface expression of CD105, CD 73, CD 90, CD26 and CD166 and lack expression of CD45, CD34, CD14 or CD11b, CD79alpha or CD19 and HLA-DR surface molecules.[68] They differentiate into mesenchymal tissue including muscle and cartilage in vitro, but only to a very limited extent after transplantation.[69] Thymus derived MSCs were utilized to reconstruct the right ventricular outflow tract landrace pigs.[70]With regard to actual improvement of right ventricular function, studies conducted using human bone marrow derived MSCs (hBMMSC) in pigs[48] and rats[71] showed improved RV function in a pulmonary artery banding model of right ventricular strain(figure 4), which has been used to model the RV overload that occurs in HLHS. In addition to effects on right ventricular strain, umbilical cord derived MSCs (UC-MSCs) have been shown to promote angiogenesis.[47, 72]

3.2.2. Cardiosphere Derived Cells (CDCs)

CDCs are endogenous cardiac-committed stem cells comprised of a multi-cellular cluster of undifferentiated cells expressing stem cell antigens along with differentiating cardiac cells. They are obtained by endomyocardial biopsy and subsequent culture.[73] These CDCs have been shown to attenuate cardiac fibrosis and preserve systolic function in a rat model of myocardial injury.[74]

3.2.3. C-kit+ Cardiac Progenitor Cells (CPCs)

C-kit+ CPCs represent a pool of stem cells capable of differentiating into all 3 mesodermal cardiac cell types i.e. cardiac, endothelial and smooth muscle. These cells are ubiquitous in the neonatal heart but almost absent in the adult.[75] Mishra and colleagues reported in 2011 a large and systematic series, which characterized and tested the functionality of cardiac stem cells in congenital heart patients.[76] In this study, 140 myocardial samples were collected from congenital heart patients, which included 10 HLHS patients. The major findings of that study were that resident C-kit+ cells are present in the maturing postnatal human heart and are most abundant during the neonatal period with a steady decline with advancing age (neonates=9% versus children=3%). Finally, they showed that intramyocardial delivery of these cells into a nude rat triggered myocardial regeneration by differentiating into mature cardiomyocytes, smooth muscle cells, and endothelial cells. This histological evidence correlated with functional improvement by ECHO in the treated group with these cells by 15 percent in comparison with controls. More importantly, the neonatal derived cardiac stem cells from HLHS patients had the highest ability to regenerate the ischemic myocardium when compared to infant or adult derived cardiac stem cells. These data provide the important preclinical evidence of the existence and strong regenerative abilities of C-kit+ cells within the hearts of HLHS patients. This study shows the feasibility of harvesting and expanding from HLHS patient’s right atrial appendage tissue and the regenerative efficacy of these cells.

The transplantation of human pediatric CPCs in the rat,[77] and human pediatric C-kit⁺ CPC into rat[78] and pig[79] have all been shown to reduce myocardial fibrosis and promote angiogenesis.

The timing of cell transplantation is critical and partially dependent on the type of cells transplanted and their potential mechanism such as reduction in fibrosis, decreased inflammatory and/or increase cardiac remodeling and protection. [80] The timing of transplantation is also dependent on the source of cells. Intracoronary administration of cells is often performed using a stop-flow technique. This technique relies on an angioplasty balloon to occlude the targeted coronary artery while cells are infused distal to the balloon. It is maintained less than 2 minutes before reperfusion; both the Transcoronary Infusion of CPCs in patient with Single Ventricle Physiology (TICAP) [61] and Cardiac Progenitor Cell Infusion to Treat Univentricular Heart Disease (PERSEUS)[65] trials used this technique. It was shown to be safe but has transient periprocedural troponin elevation without clinical evidence of myocardial infarction. This technique is suitable for cells of smaller size such as human cord blood mononuclear cell (hCBMNC)[81] or cardiosphere derived cells[74] as shown in Table 1. Larger cells such as mesenchymal stem cells,[48, 82] CPC,[77] or C-Kit+ CPC[78, 79] should be transplanted by direct intramyocardial injection. Intramyocardial injection can be performed during Glenn or Fontan surgery in patients to avoid additional surgical procedures. There are current trials, outlined below, utilizing this method but the technique appears safe based on preclinical studies. There are more limitations when using autologous cells compared to allogenic cells where the former needs time for in vitro expansion and qualify the product before transplantation back to the patient.

3.3. Clinical Trials

Cell-based therapies are under evaluation in clinical trial programs as adjuncts to surgical treatment for patients with HLHS aimed at attenuating RV dysfunction and offsetting maladaptive RV remodeling. These trials, summarized below, target different patient populations, utilize different stem cell types, and varying routes of administration for delivery of the stem cells.

3.3.1. CDCs

Two clinical trials have examined the use of CDCs for the adjunctive treatment of HLHS. These were the TICAP trial [61] and the PERSEUS trial. [65] Both trials used intracoronary infusion as a delivery methods for stem cells with patients undergoing stage 2 or stage 3 surgical palliation. TICAP was a prospective, controlled study of 14 consecutive patients with HLHS who received CDC infusion 1 month after cardiac surgery (n = 7), followed by 7 patients allocated to a control group with standard care alone. The primary end point was to assess procedural feasibility and safety which was verified; the secondary end point was to evaluate cardiac function and heart failure status through 18-month follow-up. Patients at baseline had significant RV dysfunction prior to treatment (Average EF was 46%). Similarly, PERSEUS enrolled 41 patients, 14 with HLHS, with single ventricle physiology who were randomized to receive intracoronary infusion of CDCs 4 to 9 weeks after surgery vs standard of care. Both studies showed improved RVEF, in TICAP 54.0%±2.8% vs 46.9%±4.6% (p= 0.004) at 18 months, which persisted for 36 months[83] and in PERSEUS+6.4% vs +1.3% (p= 0.003) after 3 months, maintained up to 1 year follow-up. Both studies also showed improved somatic growth and heart failure status.

3.3.2. MSCs

Allogenic Human MSC Injection In Patients with Hypoplastic Left Hearts (ELPIS)[84, 85] marks the first clinical trial to investigate the use of MSCs in HLHS. This was a phase 1 open-label multi-center United States based study to examine the safety of intramyocardial injection of Lomecel-B^™, a formulation of allogenic human MSCs, in treating patients with HLHS undergoing stage II palliation (Figure 5). The trial included 10 patients with HLHS and normal RV function at the time of surgery (EF = 62.62±5.99), an additional 4 patients were included as run-in patients who also received allogenic MSCs. Lomecel-B proved to be safe with 100% transplant-free survival and no major adverse cardiac events (MACE) at 1 year assessment. Only 1 patient experienced a cardiovascular morbidity, which was ascending aorta obstruction requiring angioplasty.

Figure 5:

ELPIS phase 1 Trial. Legend: Graphical abstract of the ELPIS phase I trial

In addition to the safety endpoint, RV function was assessed using cardiac MRI (CMR) at 6-and 12-months post administration. In this analysis, there was a preservation of RVEF with a trend to improvement at the assessment intervals. This is critical, as there is classically a fall in RVEF after stage II palliation due to increased RV load. Further, similar findings were found when examining tricuspid regurgitation, global longitudinal strain, and brain natriuretic peptide (BNP) levels. There were also no significant changes in length-for-age or weight-for-age in at assessment intervals. It is important however to note that the study was not powered for these outcomes but supports the conduct of further controlled studies on the potential benefit of MSCs in these parameters. In this regard, a study of 38 patients is currently enrolling patients (NCT04925024).

3.3.3. C-Kit+

Autologous Cardiac Stem Cell Injection in Patients with Hypoplastic Left Heart Syndrome (CHILD Study) [86] is a pilot study to evaluate the feasibility and safety of intramyocardial injection of autologous C-kit+ cells during the Stage II BDCPA operation. The primary outcome will assess incidence of MACE, observe effects on clinical outcome including RV function, severity of tricuspid regurgitation and secondary outcomes include incidence of serious adverse events, re-hospitalizations and the need for transplantation or mortality.

3.3.4. UC Stem Cells

There are 3 clinical trials underway examining different aspects of the use of cord blood derived stem cells. The first, Umbilical Cord Blood Collection and Processing for Severe Congenital Heart Disease (NCT01856049), aims to validate the use of UCB as a source of autologous cells for the purpose of cardiac repair of CHD. Cells will be isolated from the cord blood to determine the feasibility of collection, processing, and storage of these samples at the time of birth of infants with prenatal diagnosis of HLHS. This study may be useful for the development of pre-clinical and clinical studies aimed at the long-term goal of repairing damaged heart muscle. There are two phase 1 trials. Safety Study of Autologous Cord Blood Stem Cell Treatment in Hypoplastic Left Heart Syndrome Patients Undergoing the Norwood Heart Operation (NCT03431480) used autologous cord blood mononuclear cells (stem cell containing, cord blood buffy coat fraction) delivered by coronary infusion during Norwood procedure in 10 antenatally diagnosed patients with HLHS within 2–3 days of birth. Primary outcome was MACE at 1 month. Safety and clinical status monitoring was performed to Stage II surgical intervention for HLHS. Treated patients will be assessed for RV function at 1, 3 and 12 months between Stage II and III surgical interventions. Safety Study of Autologous Umbilical Cord Blood Derived Mononuclear Cells During Surgical Stage II Palliation of Hypoplastic Left Heart Syndrome (NCT01883076) aims to determine the safety and feasibility of intramyocardial injections of autologous umbilical cord blood (UCB) cells into the RV of children with HLHS undergoing a Glenn procedure. The primary outcome will examine mortality at up to 2 years, MACE and secondary outcomes focused on RV function at 1, 3 and 6 months.

3.4. Challenges with Stem Cell Therapies

Stem cell therapy is theoretically based on the concept of delivery of isolated and expanded cells to the region of injury in an attempt to promote repair and/or regeneration [54]. The overall clinical cardiac regeneration experience suggests that stem cell therapy can be safely performed.

Important challenges to improve the effectiveness of stem cell therapy for CVD include: (1) identification of the most effective route to administer cells; (2) The C-kit+ cells are traditionally autologous and the quality of these cells can be affected by factors such as age, comorbidities such as obesity, and genetic signature inherent to the host. [87] Allogeneic C-kit+ cells provide an off-the-shelf product that can be transplanted at any time; (3) identification of mobilizing and homing agents that increase recruitment; (4) despite promising preclinical results, clinical trials of MSCs are not as impressive. Different interventions as currently under investigation to improve the efficacy of transplanted MSCs[88] and (5) development of large-scale cell manufacturing techniques.

4. Future Directions and Applications

While knowledge and interest in stem cell therapies has grown exponentially in the past few years, there is still much to know and it is clear there is a lot of additional work is needed. Important future directions will include work looking at the mechanisms of stem cell cardiac regeneration and adjunctive strategies that may also utilize stem cell therapies. Models have shown that injected stem cells do not lead to direct engraftment and regeneration as was initially hoped but rather seem to transiently engraft and function to through apparent paracrine mechanism to groom host cells to regenerate.[89]

4.1. Extracellular Vesicles

This begs the question if administration of stem cell extracellular vesicles (EVs), and their contained exosomes, may be a future direction of regenerative therapies. Evs are produced by all cell types, varying in size from 30–150nm and their diverse package contains variable components including proteins and RNA/DNA fragments. These vesicles are thought to exert paracrine and autocrine signaling mechanisms primarily by microRNA contents and may form the primary means of cell-cell communication.[90, 91]

CDCs have been shown to reduce infarct size, ventricular remodeling and grow new cardiac tissue.[92] It has also previously been shown that blocking the activity of EVs causes failure in the regenerative abilities of CDCs. A porcine animal model showed that CDC-derived EVs preserved ventricular function while decreasing remodeling associated with myocardial infarction, this model utilized IM injection which has been associated with significant arrythmias in other studies, intracoronary(IC) infusion was less impressive.[93, 94]The suggestion being the relative low retention of the tiny EVs when administered IC. Some of the potential benefits of EVs as intervention include their relative stability even at room temperature and ability to freeze and thawing without significant degradation.[95] EVs have the potential to be well tolerated by the immune system in animal models[96], and the microRNA contents of these EVs could serve as targets for bioengineering delivering specific miRNAs as has been investigated in oncologic studies.[97, 98]

4.2. Tissue Engineering

Combining stem cells, growth factors and scaffolding is described as tissue engineering. The aim of this process is to not only generate viable, functional cardiac tissue to replace damaged tissues but also generate graphs that can grow with the patient after repair, this is particularly appealing for childhood procedures. This is however particularly challenging for cardiac cell engineering because of the organ’s complexity involving multiple cell types and unique and varied functionality. Cells for engineering must be able to generate all these cardiac cell types. Current sources so far being primarily embryonic stem cell, adult stem cells and adult somatic cells.[99]

4.3. Induced pluripotential stem cells (iPSCs)

iPSCs are human somatic cells for example fibroblasts engineered to behave like stem cells.[100] These cell types are particularly attractive since they are derived from the host and overcome some of the immune-related concerns associated with other cell types. They also provide the potential of ready access to stem cells but safety concerns remain. Despite cardiac cells being among the first functional cell types formed from human pluripotent stem cells,[101] the complex interplay among cardiac cell types and the need for a 3-dimensional scaffold continues to limit progress towards large scale grafts. Questions regarding electromechanical coupling, immune tolerance and large-scale vascularization remain to be resolved.

4.4. Scaffolds

Scaffolds work akin to an extracellular matrix providing a framework on which stem cells can proliferate and produce the targeted tissue. The ideal scaffold requires a number of structural and biological characteristics to facilitate successful engraftment and function. These include being biodegradable, flexible, durable and allow neovascularization, all while being cost effective.[102] So far success with scaffolds have been impressive but limited small segments and shunts. The first reported was a pulmonary artery graft shown to be patent 7 months post engraftment[103], since then a number of researchers have published reports of use of scaffolds for conduits and other vascular replacements. They seem well tolerated without reports of any catastrophic complications.[104] No large scale trials are currently underway and no reports of any large component replacements.

4.5. Other Congenital Heart Disease

While the vast majority of trials of stem cell therapies have been used to HLHS, they do show potential for other congenital heart diseases. Rupp et al were used intracoronary infusion of stem cells for patients with congenital heart disease and cardiomyopathy who showed improvement in heart failure status or survived to transplant.[105] More recently, researchers at the Mayo clinic have demonstrated improved RV area and LV ejection fraction after intramyocardial injection of bone-marrow derived stem cells during reconstruction in patients with Ebstein’s Anomaly.

5. Conclusion

HLHS and other congenital heart diseases remain one of the most clinically and scientifically relevant topics requiring further research. While surgical palliation has offered opportunities for those affected to live into adulthood, morbidity and mortality continue to pose major health challenges. Cell-based therapies provide an novel opportunity to improve the outcome of this patient population. In recent years, improvements in pre-clinical modelling, continued research into modes of administration and stem cell sources and early-stage clinical trials are showing promise. The use of bio-engineering techniques provides a further opportunity to continue to take stem cell therapies from the bench to the bedside.

Table 2.

Clinical trials for stem cell therapies in congenital heart disease

Cell Type	Dose	Infusion Method	Time of delivery	n	Results/Status	Identifier
Autologous CPC	0.3×106/kg	Intracoronary infusion	One month aftereither Norwood-Glenn, Glenn, or Fontan	14	- ↑ RVEF - ↓brain natriuretic peptide - ↓ incidence of unplanned catheter interventions - ↑ weight-for-age z score[83] - ↓ myocardial fibrosis[108]	NCT01273857 (TICAP)[61]
Autologous CPC	0.3×106/k g	Intracoronary infusion	Glenn or Fontan	34	- ↑ RV function - ↓ mortality and late complications in patients with heart failure with reduced EF - ↓ myocardial fibrosis[108]	NCT01829750 (PERSEUS)[65]
Autologous UCB MNC	30×106/kg	Intramyocardial injections	--	30	Completed	NCT01883076
Autologous cord blood	Unknown dose	Intracoronary infusion	Norwood	10	Completed	NCT03431480
Lomecel-B medicinal signaling cells (BMMSCs)	2.5×105 cells/kg	Intramyocardial injections	Glenn or Fontan	38	Recruiting	NCT04925024 (ELPISII)[84]
c-kit cells	12,500 cells/kg	Intramyocardial injections	Glenn	32	Recruiting	NCT03406884 (CHILD)[86]
Autologous cord blood	Unknown dose	unknown	Unknown	600	Recruiting	NCT01856049
Autologous UCB MNC	0.1 mL/kg; 1–3×106 TNC/kg	Intramyocardial injections	Glenn or Fontan	95	Active, not recruiting	NCT03779711