Infants born preterm face a number of challenges. Depending on the degree of prematurity, they are at a risk of developing several specific conditions and diseases related to organ immaturity and complications of long-term neonatal intensive care. Various organ systems are affected, such as the lung, resulting in bronchopulmonary dysplasia (BPD); the vascular system, resulting in pulmonary hypertension; the brain, with the risk of intracranial hemorrhage; the eye with retinopathy of prematurity; and the gut, manifesting in the severe complication of necrotizing enterocolitis. A common hallmark for all these prematurity-related conditions is that inflammation seems to be a major driving force in the pathogenesis, and that injury repair is essential for recovery and long-term health. In addition, the available treatment options are often only supportive, not curative. This chapter reviews the recent advances of stem cell therapy that have opened up new possibilities to restore organ function following prematurity.
Highlights
- •
Many diseases of prematurity involve inflammation and dysfunctional growth.
- •
Diseases of prematurity are suitable candidates for cell-based treatment strategies.
- •
Clinical data for cell-based treatments in neonates are still very limited.
- •
Experimental findings are promising, but many issues remain unsolved.
Introduction
Advances in perinatal medicine over the last decades have dramatically improved survival after preterm birth in developed countries . However, in a global perspective, prematurity is by far the leading cause of neonatal deaths, and it remains the second leading cause of death after pneumonia in children under the age of 5 years . Preterm birth rates are rising, and, although it does not appear to be at the cost of higher rates of severe disability, there is increasing concern that survivors of preterm birth are at a risk of learning difficulties, as well as visual and hearing impairments. Emerging evidence regarding long-term health also suggests that prematurity is a risk factor for later development of chronic lung disease , cardiovascular disease , and type 2 diabetes . The public health consequences of the increasing numbers of survivors after preterm birth are not yet fully understood, but it is clear that there is much to be gained by improving treatments in the neonatal period, for the individual infant with severe complications and for general future health.
Cellular therapies show great potential in a diversity of clinical conditions and diseases, including neurodegenerative disorders, cardiovascular disease, lung injury, diabetes, graft-versus-host reactions, sepsis, and hepatic and renal failure . In various experimental models, stem cells are demonstrated to attenuate organ injury , and they induce tissue repair processes, the multipotency of these cells leading to them being described as the cellular “Doctors Without Borders.” However, the questions of the role, function, and mechanisms of action still hold intriguing gaps of knowledge. In the therapeutic situation, the engraftment of cells seems to be limited, leading to the assumption that the cells may rather serve as bioactive mediators that interact with inflammatory cells and modulate their response . The identification of these paracrine effects has opened new options for treatment that may be of great importance in the neonatal period, as the potential side effects associated with cell-based therapies can be mitigated .
Infants born preterm, and particularly those with extreme prematurity defined as a gestational age below 28 weeks, face significant morbidity in the neonatal period and lengthy hospitalizations. Intensive care for the smallest and most immature infants has improved dramatically, and a number of life-saving interventions are often effective in the battle against severe complications. Nevertheless, treatments are most of the time nonspecific, and therefore cellular therapies directed at different organ systems at risk of injury following preterm birth are new and promising forms of regenerative medicine in the neonatal period. Most of the work so far has been focused toward the lung and the brain, but novel findings have ignited interest in other areas, such as the gut and the eye, as outlined subsequently.
Lung
Bronchopulmonary dysplasia (BPD) is the most common severe complication of preterm birth, affecting approximately 45% of infants with a birth weight below 1500 g . The pathophysiology of BPD is characterized by inflammation, fibrosis, simplification of lung structure, and alveolar arrest , and the development of BPD is associated with other inflammatory diseases of pregnancy, such as preeclampsia . Lung injury related to mechanical ventilation and oxidative stress still contributes to an increased risk of BPD, but presently the use of noninvasive ventilation, targeted surfactant treatment, and antenatal steroid treatment has shifted the etiology toward greater developmental immaturity of the alveolar and vascular systems of the lung, including developmentally impaired defense and repair mechanisms. Arrested lung growth, the histological hallmark of BPD today, may have an impact on lung function in childhood, which persists into adulthood and increases the risk of chronic lung disease and emphysema later in life . Preventive measures and treatment options for BPD in very preterm infants are limited to supportive therapies; hence, new treatments that facilitate and promote normal lung development are strongly warranted.
The demonstration that bone marrow-derived stem cells could differentiate into alveolar epithelial cells led to stem cell research in neonatal lung disease . A link between low levels of endothelial colony-forming cells (ECFCs) in cord blood and the risk of developing BPD further enhanced the rationale for adopting stem cell therapy in BPD. In nearly 100 preterm infants, with a gestational age <32 weeks, a decrease in ECFCs in cord blood was linked with more pulmonary vascular immaturity and an increased risk of BPD, whereas high levels of ECFCs in cord blood seemed to be protective for BPD development, even in extremely preterm infants . In tracheal aspirates from preterm infants, the isolation of mesenchymal stromal cells is reported to be strongly predictive of the development of BPD . Cells from infants who develop BPD are also reported to hold stable gene alterations favoring hypoalveolarization, suggesting involvement in the pathophysiology of the disease .
The effect
Although “repopulate, repair, and grow” have been said to be the goal of cellular therapy in neonatal lung disease, the actual cells do not appear to be crucial for the effect. Instead, recent findings indicate that paracrine effects, rather than regeneration, mediate positive effects. In vivo engraftment and differentiation into lung tissue is low , and the transdifferentiation of administered mesenchymal stem cells (MSCs) into alveolar type II cells rarely occurs in vivo . In murine experimental models of hyperoxia-induced BPD, MSC-conditioned media show equal or better ability to ameliorate parenchymal and vascular lung injury than whole cells . Furthermore, MSCs have recently been recognized for releasing membrane vesicles, exosomes, that contain bioactive material and microRNA, which may play a particularly important role in BPD by silencing specific genes aggravating lung injury . MSC-derived exosomes are reported to decrease cytokine levels in bronchoalveolar lavage and to attenuate lung macrophage influx and vascular remodeling in mice with hyperoxia-induced pulmonary hypertension (PH) . The results from various preclinical models are very promising; however, it is clear that MSCs from different sources secrete different populations of exosomes, varying in composition and capabilities, and that exact pathways of action are yet to be defined. The use of the MSC secretome rather than live cells may hold a unique possibility to modulate lung injury without the potential risk of tumorigenesis from engrafted cells, although a number of questions remain to be answered before specific microvesicle production for the treatment of BPD can become a reality. Furthermore, some argue that whole cells are indeed preferable, and it has been recently demonstrated that mitochondrial transfer from MSCs to resident lung cells provides protection against lung injury . In addition, MSC may release factors that protect resident lung cells from injury or promote the proliferation of epithelial progenitor cells in the lung . Although it is becoming evident that the effects of stem cell therapy in BPD are likely multifold, such as modulating immune responses, inducing repair processes, and promoting growth, the importance of each action remains to be determined in relation to the type of cell, route of administration, dose, and timing, being considered in conjunction with the clinical situation.
Experimental studies
Preclinical studies of stem cell therapy for BPD have used different experimental models and various cell types. MSCs are the most extensively examined. The source of the cells may be adult bone marrow, umbilical cord blood, Wharton’s jelly, fetal tissue, placenta, or adipose tissue, the first two being the most widely used in neonatal studies . Apart from MSCs, endothelial progenitor cells (EPCs) and amnion epithelial cells (AECs) are the focus of study. The observed depletion of both circulating and lung-resident EPCs and the crucial role of EPCs for lung growth and injury repair suggest a therapeutic potential of exogenous EPC administration . Amniotic fluid-derived multipotent stem cells seem to preferentially home to the injured lung, and a fraction of them appears to differentiate into alveolar cells in a mice model of hyperoxia-induced injury . Human AECs have been investigated in fetal sheep models of lung injury and shown to improve lung function and structure . In addition, amniotic fluid-derived MSCs augment fetal lung growth in an explant model, suggesting a role in the future treatment of lung hypoplasia . The described therapeutic effects in the preclinical studies of cell-based strategies to treat BPD include improved survival, reduced oxidative stress, attenuated inflammation, less impairment of alveolar growth and fibrosis, ameliorated vascular injuries, and a reduction in associated PH .
Route of administration, dose, and timing
For the translation of cellular therapies into clinical praxis, it is critical to determine the optimal route of delivery. MSCs have been shown to migrate toward an injured lung area, following both intratracheal and systemic administration . However, the only study to date directly comparing local (intratracheal) and systemic (intraperitoneal) transplantation of human umbilical cord-derived (hUC) MSCs in an experimental rodent model of BPD revealed that local administration produced more effective engraftment and greater reduction in hyperoxia-induced impaired alveolarization compared with systemic administration . This finding suggests that intratracheal administration, already used in neonatal medicine for surfactant treatment, might be the preferred route of delivery of stem cells in BPD, but further studies are needed.
The optimal dose of stem cells in BPD is yet to be established. Chang et al. tested the therapeutic efficacy of three different doses of intratracheally delivered hUC-derived MSCs in a newborn rat model (5 × 10 3 , 5 × 10 4 , and 5 × 10 5 to rat pups with a weight of approximately 10 g), and they could demonstrate a dose-dependent reduction in symptoms associated with hyperoxia-induced lung injury, such as decreased alveolarization . The highest dose provided the best protection, and at least a medium dose, of 5 × 10 4 cells, was necessary for a significant effect on the anti-inflammatory response and attenuation of oxidative stress. The only clinical trial in human infants, which until now has published results, had a dose-escalating design, testing 1 × 10 7 and 2 × 10 7 cells/kg . The higher dose was not associated with any advantages; instead, there was a trend toward longer duration of intubation and higher BPD severity score in the high-dose group, indicating that the search for optimal dosing of MSCs for the clinical treatment of BPD required further studies.
The timing of treatment is another crucial issue that still needs to be defined. During the early phases of BPD, usually characterized by profound inflammation, when structural changes are still reversible, there may be an attractive window for cellular treatments to prevent and protect lung architecture. This is supported by the findings in the experimental rodent model of BPD in which early (P3–4) versus late (P10–14) treatment of bone marrow-derived or hUC-derived MSCs revealed improved lung structure, decreased apoptosis, inflammation, and oxidative stress only after early treatment . Presently, we lack good biomarkers to predict later development of severe BPD. If early stem cell therapy is to be implemented into the clinical praxis, this emphasizes the need for better identification of at-risk infants in order to further optimize the timing of treatment .
Long-term outcome
Data on long-term outcome are important to ensure the safe translation of findings from experimental to clinical settings and to detect any potential adverse effects of cellular therapy, such as tumor formation. In neonatal experimental models, Ahn et al. evaluated outcome up to postnatal day 70 in rats, and they found that the beneficial effects following intratracheal transplantation of MSCs on P5 were sustained over time without any observed abnormalities in histological examinations of internal organs . Sutsko et al. reported outcome at postnatal day 100 in newborn rats with hyperoxic lung injury after intratracheal treatment with MSCs and cell-free MSC-conditioned medium on P9, and they showed persisting improvements in alveolar and vascular development, decreased inflammation, and upregulation of angiogenetic factors . The effect on P100 was more marked following MSC treatment compared with only the conditioned medium. Similarly, Pierro et al. followed up newborn rats up to 6 months after intratracheal administration of hUC-derived perivascular cells and MSCs on P4, showing long-term prevention of lung damage and improved lung function despite low cell engraftment, suggesting paracrine actions to be predominantly responsible for the effect . Moreover, no adverse effects were seen. This indicates the long-term safety of cellular treatment in experimental models of neonatal lung disease, but clinical data are still lacking, and potential added benefits from repeated doses over time remain to be studied.
Clinical data
One clinical study in preterm infants at risk of BPD has been published to date . In this phase 1 dose-escalation study, Chang et al. assessed the safety and feasibility in nine extremely preterm infants with a mean gestational age of 25 weeks and a mean birth weight of approximately 800 g. Infants required continuous ventilator support with no signs of imminent improvement. The transplantation of hUC-derived MSC by intratracheal administration was performed between 7 and 14 days of life (mean day 10). Six infants received low dose, 1 × 10 7 cells/kg, and three infants high dose, 2 × 10 7 cells/kg. Treatment was well tolerated without any serious adverse events. On day 7 following treatment, the cytokine levels in the tracheal aspirate were significantly reduced compared with baseline. The BPD severity score was lower in MSC-treated infants compared with a control group of infants matched for gestational age and respiratory severity. Although too early to draw any firm conclusion on efficacy, the study indicates that MSC treatment is safe and feasible in human infants. A long-term follow-up study and a phase 2 double-blind, randomized controlled trial to assess therapeutic efficacy are currently under way , and a total of seven studies of stem cell therapy for neonatal lung disease are currently registered in clinicaltrials.gov .
Lung
Bronchopulmonary dysplasia (BPD) is the most common severe complication of preterm birth, affecting approximately 45% of infants with a birth weight below 1500 g . The pathophysiology of BPD is characterized by inflammation, fibrosis, simplification of lung structure, and alveolar arrest , and the development of BPD is associated with other inflammatory diseases of pregnancy, such as preeclampsia . Lung injury related to mechanical ventilation and oxidative stress still contributes to an increased risk of BPD, but presently the use of noninvasive ventilation, targeted surfactant treatment, and antenatal steroid treatment has shifted the etiology toward greater developmental immaturity of the alveolar and vascular systems of the lung, including developmentally impaired defense and repair mechanisms. Arrested lung growth, the histological hallmark of BPD today, may have an impact on lung function in childhood, which persists into adulthood and increases the risk of chronic lung disease and emphysema later in life . Preventive measures and treatment options for BPD in very preterm infants are limited to supportive therapies; hence, new treatments that facilitate and promote normal lung development are strongly warranted.
The demonstration that bone marrow-derived stem cells could differentiate into alveolar epithelial cells led to stem cell research in neonatal lung disease . A link between low levels of endothelial colony-forming cells (ECFCs) in cord blood and the risk of developing BPD further enhanced the rationale for adopting stem cell therapy in BPD. In nearly 100 preterm infants, with a gestational age <32 weeks, a decrease in ECFCs in cord blood was linked with more pulmonary vascular immaturity and an increased risk of BPD, whereas high levels of ECFCs in cord blood seemed to be protective for BPD development, even in extremely preterm infants . In tracheal aspirates from preterm infants, the isolation of mesenchymal stromal cells is reported to be strongly predictive of the development of BPD . Cells from infants who develop BPD are also reported to hold stable gene alterations favoring hypoalveolarization, suggesting involvement in the pathophysiology of the disease .
The effect
Although “repopulate, repair, and grow” have been said to be the goal of cellular therapy in neonatal lung disease, the actual cells do not appear to be crucial for the effect. Instead, recent findings indicate that paracrine effects, rather than regeneration, mediate positive effects. In vivo engraftment and differentiation into lung tissue is low , and the transdifferentiation of administered mesenchymal stem cells (MSCs) into alveolar type II cells rarely occurs in vivo . In murine experimental models of hyperoxia-induced BPD, MSC-conditioned media show equal or better ability to ameliorate parenchymal and vascular lung injury than whole cells . Furthermore, MSCs have recently been recognized for releasing membrane vesicles, exosomes, that contain bioactive material and microRNA, which may play a particularly important role in BPD by silencing specific genes aggravating lung injury . MSC-derived exosomes are reported to decrease cytokine levels in bronchoalveolar lavage and to attenuate lung macrophage influx and vascular remodeling in mice with hyperoxia-induced pulmonary hypertension (PH) . The results from various preclinical models are very promising; however, it is clear that MSCs from different sources secrete different populations of exosomes, varying in composition and capabilities, and that exact pathways of action are yet to be defined. The use of the MSC secretome rather than live cells may hold a unique possibility to modulate lung injury without the potential risk of tumorigenesis from engrafted cells, although a number of questions remain to be answered before specific microvesicle production for the treatment of BPD can become a reality. Furthermore, some argue that whole cells are indeed preferable, and it has been recently demonstrated that mitochondrial transfer from MSCs to resident lung cells provides protection against lung injury . In addition, MSC may release factors that protect resident lung cells from injury or promote the proliferation of epithelial progenitor cells in the lung . Although it is becoming evident that the effects of stem cell therapy in BPD are likely multifold, such as modulating immune responses, inducing repair processes, and promoting growth, the importance of each action remains to be determined in relation to the type of cell, route of administration, dose, and timing, being considered in conjunction with the clinical situation.
Experimental studies
Preclinical studies of stem cell therapy for BPD have used different experimental models and various cell types. MSCs are the most extensively examined. The source of the cells may be adult bone marrow, umbilical cord blood, Wharton’s jelly, fetal tissue, placenta, or adipose tissue, the first two being the most widely used in neonatal studies . Apart from MSCs, endothelial progenitor cells (EPCs) and amnion epithelial cells (AECs) are the focus of study. The observed depletion of both circulating and lung-resident EPCs and the crucial role of EPCs for lung growth and injury repair suggest a therapeutic potential of exogenous EPC administration . Amniotic fluid-derived multipotent stem cells seem to preferentially home to the injured lung, and a fraction of them appears to differentiate into alveolar cells in a mice model of hyperoxia-induced injury . Human AECs have been investigated in fetal sheep models of lung injury and shown to improve lung function and structure . In addition, amniotic fluid-derived MSCs augment fetal lung growth in an explant model, suggesting a role in the future treatment of lung hypoplasia . The described therapeutic effects in the preclinical studies of cell-based strategies to treat BPD include improved survival, reduced oxidative stress, attenuated inflammation, less impairment of alveolar growth and fibrosis, ameliorated vascular injuries, and a reduction in associated PH .
Route of administration, dose, and timing
For the translation of cellular therapies into clinical praxis, it is critical to determine the optimal route of delivery. MSCs have been shown to migrate toward an injured lung area, following both intratracheal and systemic administration . However, the only study to date directly comparing local (intratracheal) and systemic (intraperitoneal) transplantation of human umbilical cord-derived (hUC) MSCs in an experimental rodent model of BPD revealed that local administration produced more effective engraftment and greater reduction in hyperoxia-induced impaired alveolarization compared with systemic administration . This finding suggests that intratracheal administration, already used in neonatal medicine for surfactant treatment, might be the preferred route of delivery of stem cells in BPD, but further studies are needed.
The optimal dose of stem cells in BPD is yet to be established. Chang et al. tested the therapeutic efficacy of three different doses of intratracheally delivered hUC-derived MSCs in a newborn rat model (5 × 10 3 , 5 × 10 4 , and 5 × 10 5 to rat pups with a weight of approximately 10 g), and they could demonstrate a dose-dependent reduction in symptoms associated with hyperoxia-induced lung injury, such as decreased alveolarization . The highest dose provided the best protection, and at least a medium dose, of 5 × 10 4 cells, was necessary for a significant effect on the anti-inflammatory response and attenuation of oxidative stress. The only clinical trial in human infants, which until now has published results, had a dose-escalating design, testing 1 × 10 7 and 2 × 10 7 cells/kg . The higher dose was not associated with any advantages; instead, there was a trend toward longer duration of intubation and higher BPD severity score in the high-dose group, indicating that the search for optimal dosing of MSCs for the clinical treatment of BPD required further studies.
The timing of treatment is another crucial issue that still needs to be defined. During the early phases of BPD, usually characterized by profound inflammation, when structural changes are still reversible, there may be an attractive window for cellular treatments to prevent and protect lung architecture. This is supported by the findings in the experimental rodent model of BPD in which early (P3–4) versus late (P10–14) treatment of bone marrow-derived or hUC-derived MSCs revealed improved lung structure, decreased apoptosis, inflammation, and oxidative stress only after early treatment . Presently, we lack good biomarkers to predict later development of severe BPD. If early stem cell therapy is to be implemented into the clinical praxis, this emphasizes the need for better identification of at-risk infants in order to further optimize the timing of treatment .
Long-term outcome
Data on long-term outcome are important to ensure the safe translation of findings from experimental to clinical settings and to detect any potential adverse effects of cellular therapy, such as tumor formation. In neonatal experimental models, Ahn et al. evaluated outcome up to postnatal day 70 in rats, and they found that the beneficial effects following intratracheal transplantation of MSCs on P5 were sustained over time without any observed abnormalities in histological examinations of internal organs . Sutsko et al. reported outcome at postnatal day 100 in newborn rats with hyperoxic lung injury after intratracheal treatment with MSCs and cell-free MSC-conditioned medium on P9, and they showed persisting improvements in alveolar and vascular development, decreased inflammation, and upregulation of angiogenetic factors . The effect on P100 was more marked following MSC treatment compared with only the conditioned medium. Similarly, Pierro et al. followed up newborn rats up to 6 months after intratracheal administration of hUC-derived perivascular cells and MSCs on P4, showing long-term prevention of lung damage and improved lung function despite low cell engraftment, suggesting paracrine actions to be predominantly responsible for the effect . Moreover, no adverse effects were seen. This indicates the long-term safety of cellular treatment in experimental models of neonatal lung disease, but clinical data are still lacking, and potential added benefits from repeated doses over time remain to be studied.
Clinical data
One clinical study in preterm infants at risk of BPD has been published to date . In this phase 1 dose-escalation study, Chang et al. assessed the safety and feasibility in nine extremely preterm infants with a mean gestational age of 25 weeks and a mean birth weight of approximately 800 g. Infants required continuous ventilator support with no signs of imminent improvement. The transplantation of hUC-derived MSC by intratracheal administration was performed between 7 and 14 days of life (mean day 10). Six infants received low dose, 1 × 10 7 cells/kg, and three infants high dose, 2 × 10 7 cells/kg. Treatment was well tolerated without any serious adverse events. On day 7 following treatment, the cytokine levels in the tracheal aspirate were significantly reduced compared with baseline. The BPD severity score was lower in MSC-treated infants compared with a control group of infants matched for gestational age and respiratory severity. Although too early to draw any firm conclusion on efficacy, the study indicates that MSC treatment is safe and feasible in human infants. A long-term follow-up study and a phase 2 double-blind, randomized controlled trial to assess therapeutic efficacy are currently under way , and a total of seven studies of stem cell therapy for neonatal lung disease are currently registered in clinicaltrials.gov .
Heart and vessels
Stem cells have been successfully tested for the heart in the SCIPIO trial of myocardial ischemia in adults . PH triggered by inflammation and vascular cell dysfunction develops in up to 25% of infants with moderate to severe BPD . It is an added complication that, in its most severe form, may lead to cor pulmonale and greatly increase mortality in these children . In older children with idiopathic PH, the safety and efficacy of autologous EPC transplantation have been tested in an open-label pilot study . In neonatal PH, however, the results are only available from preclinical experimental studies so far. EPCs have a crucial role in normal vascular growth and, as impaired lung structure plays an important part in the development of neonatal PH, especially in preterm infants . Therefore, stem cell treatment is an appealing approach targeting both alveolar and vascular structure. In the rodent model of BPD, MSC-treated lungs demonstrated larger and rounder vessels, resembling a more normal pulmonary vasculature organization, as illustrated in Fig. 1 . In a mouse model of BPD, conditioned medium from UC-derived ECFCs was shown to reduce right ventricular hypertrophy . Hansmann et al. also demonstrated in their mice model of hyperoxia-induced BPD and PH that a single dose of MSC-conditioned medium administered intravenously at 4 weeks of age completely reversed PH and right ventricular hypertrophy and attenuated peripheral smooth muscle hypertrophy .
