Bronchopulmonary dysplasia (BPD), the chronic lung disease of prematurity, remains
the main complication of extreme prematurity.[1] Dramatic improvements in perinatal care have substantially increased neonatal survival
from 34 weeks' gestation since the original description of BPD in 1967[2] to 22 to 24 weeks' gestation nowadays. The ransom of success is the increasing challenge
of protecting an ever more immature and fragile lung from the multiple deleterious
effects of inflammation, oxidative stress, mechanical ventilation, and suboptimal
growth.[3] Even so the lung has robust repair potential, the life-long consequences of interfering
with early lung development remain unknown. Increasing evidence in the literature
suggests reduced lung function, airway hyperreactivity,[4] and impaired lung vascular growth,[5] as well as other organ impairment[6] justifying the need for exploring effective early interventions that are currently
lacking.
Recent insights into the regenerative potential of stem cells[7] have created excitement in neonatology for the treatment of complications of extreme
prematurity including BPD. Stem cells are defined as cells capable of self-renewal
and differentiation into at least one other cell type. However, it is not these fundamental
properties that seem to underlie their mechanism of action. Preclinical and some clinical
evidence do not support the local proliferation and differentiation of these cells
into lung cells for repair. The therapeutic benefits seem to emanate from a by-stander
effect: the secretion of repair modulating factors in the microenvironment,[8] cell-to-cell interactions via the exchange of micro-organelles, and release of extracellular
vesicles (EVs).[9] Given the excitement in the field, a panoply of cell-based therapies have emerged.
Brief Appraisal of Cell-Based Therapies Investigated in Preclinical and Clinical Studies
of Neonatal Lung Injury
Mesenchymal Stromal Cells—The Front Runner
Definition
Mesenchymal stromal cells (MSCs) are plastic adherent, multipotent cells (they can
be induced to become cartilage, bone, or fat) that express certain cell surface markers
(CD105, CD73 and CD90, and lack of CD45, CD34, CD14 or CD11b, CD79α or CD19 and HLA-DR
surface molecules) as defined by the International Society for Cellular Therapy.[10] Originally described in the bone marrow,[11] MSCs have now been shown to exist in all organs including the umbilical cord tissue
and cord blood.[12]
Rationale and Biological Plausibility
MSCs are potent immune-modulatory cells, but also exert pleiotropic effects that target
numerous other pathophysiological mechanisms that contribute to BPD: MSCs are antifibrotic,
antioxidant, and proangiogenic.[8] Other logistical aspects make them appealing as an off-the-shelf ready-to-use cell
therapy product as MSCs express HLA class I, but not HLA class II, and thus do not
elicit alloreactive lymphocyte proliferation, enabling allogeneic transplantation
of MSCs. Furthermore, evidence suggests that MSCs are more abundant in the tracheal
aspirate of preterm infants that go on to develop BPD versus those that do not.[13] These MSCs seem to acquire by default a profibrotic phenotype suggesting they may
lose their repair potential and/or even contribute to the disease pathogenesis. These
findings are corroborated by observations of human fetal lung mesenchymal cells exposed
to hyperoxia in vitro[14] and in resident lung MSCs isolated from neonatal rats exposed to 14 days of hyperoxia.[15]
Preclinical Evidence
Numerous investigators have demonstrated the lung protective effect of MSCs derived
from rat bone marrow[16]
[17] and human umbilical cord tissue[18] and cord blood[19] in the neonatal rodent model exposed to hyperoxia. Even so various endpoints were
assessed including the route of administration, dose–response, timing, long-term effect,
and toxicity, as well as some mechanisms of action, most of these studies were exploratory
in nature. While a systematic review and meta-analysis of all preclinical studies
with MSCs in experimental neonatal lung injury confirmed the therapeutic benefit of
MSCs on alveolarization, inflammation, and lung vascular growth, it also revealed
some limitations in experimental design, reporting, and risk of bias, as well as the
lack of studies in large animals.[20]
Clinical Trials
Nonetheless, the promising preclinical data and the apparent safety and lack of engraftment
of MSCs have encouraged early phase clinical trials in preterm infants at risk of
developing BPD. The first phase I clinical trial tested allogeneic cord blood-derived
MSCs in nine preterm neonates born between 23 and 29 weeks' gestational age at 5 to
14 days of life if still on mechanical ventilation.[21] A single intratracheal injection of 107 or 2 × 107 MSCs was shown to be feasible and well tolerated without serious adverse events or
dose-limiting toxicity. At 2 years follow-up, there were no adverse effects on growth,
respiratory, and neurodevelopmental outcomes.[22] Results of a U.S. trial with the same cell product and similar design are pending
(NCT02381366). A phase II randomized, double-blinded, multicenter, controlled trial
using a low dose is currently underway (NCT01828957). A recent case report of two
preterm infants with severe BPD treated at late stages of the disease showed that
the repeated intravenous (IV) injection of bone marrow-derived MSCs was feasible and
safe, but it did not influence disease outcome.[23] Postmortem lung analysis confirmed absence of lung engraftment of transplanted cells.
Well-designed early phase trials over the coming years will provide more information
regarding feasibility and safety of routes of administration, timing, and repeated
administration.[24]
Human Amniotic Epithelial Cells—The Mimic: Different Cell Type, Same Mechanism of
Action
Definition
A large number of human amniotic epithelial cells (hAECs) can be isolated from the
amniotic membrane of the placenta after birth. hAECs are stem cell-like cells with
self-renewal and multilineage potential.[25] Similar to MSCs, they have limited class IA and II HLA expression and thus present
a low risk of allogeneic rejection.
Rationale and Biological Plausibility
hAECs also share similar pleiotropic, mostly anti-inflammatory, but also proangiogenic,
properties with MSCs.[25] hAECs do not engraft and exert their therapeutic benefit through a paracrine activity.
Preclinical Evidence
Extensive experimental evidence has demonstrated the lung protective effects of hAECs
not only in neonatal rodents exposed to hyperoxia, but also in fetal sheep exposed
in utero to lipopolysaccharide or mechanical ventilation.[26]
[27]
[28] Similar to MSCs, the relative ease of cell isolation, expansion and the lack of
risk of rejection, and apparent safety profile are appealing characteristics for an
off-the-shelf cell product.
Clinical Trials
Given the promising preclinical data, a phase I trial was recently conducted in six
preterm infants born at less than 29 weeks' gestation with established severe BPD
receiving a single IV infusion of 1 million hAECs/kg.[29] An important lesson from this trial was the adjustment of the infusion protocol
after the first infant experienced transient bradycardia and hypoxia during the infusion.
This was likely related to a pulmonary microembolic event and did not reoccur in the
remaining five infants after changes were made to the cell infusion protocol. This
trial showed feasibility and lack of toxicity of a single IV dose of hAECs. The same
group is now launching a phase I trial to test the feasibility and safety of a dose
escalation (2, 10, and 30 million cells/kg) in 24 preterm infants at high risk of
severe BPD still requiring an FiO2 ≥25% while mechanically ventilated or an FiO2 ≥35% while on noninvasive respiratory support at day 14 of life.[30]
Endothelial Progenitor Cells—The Vascular Hypothesis of BPD
Definition
The term endothelial progenitor cell (EPC) is used to describe a cell that can regenerate
the endothelial lining of blood vessels. As for other stem cells, the nomenclature
remains ambiguous, leading to controversy in interpreting study results. A recent
consensus statement on nomenclature of EPCs suggests the use of precise terminology
based on defining cellular phenotype and function.[31] Two distinct and well-defined cell types have been considered as EPCs: endothelial
colony forming cells (ECFCs) and myeloid angiogenic cells (MACs). Both promote vascular
repair and may thus be considered for therapeutic purposes, although their mechanisms
of action appear to be different. ECFCs are characterized by (1) robust proliferative
potential, (2) secondary and tertiary colony formation upon replating, and (3) de
novo blood vessel formation in vivo when transplanted into immunodeficient mice. Of
clinical relevance for neonatology, these cells exist in cord blood and can be expanded
for therapeutic purposes.[32] MACs also participate in angiogenesis but do not directly form the endothelial monolayer
of new vessels and display various monocyte/macrophage phenotypes and function.
Rationale and Biological Plausibility
Increasing evidence suggests that lung vascular growth and angiogenic growth factors
are crucial for lung growth and repair.[33] Interestingly, resident ECFCs are present in the developing human and rat lung[34] and their function is impaired in experimental neonatal lung injury.[35] Combining the data from a recent systematic review indicating a link between impaired
EPC function in circulating/cord blood of preterm infants and preterm birth-associated
complications[36] suggests that exogenous supplementation of EPCs may be therapeutic. Unlike MSCs,
however, ECFCs may be immunogenic, thus requiring autologous transplantation.
Preclinical Evidence
The first observation with bone marrow-derived MACs showed restoration of alveolar
and lung vascular growth in hyperoxic neonatal mice.[37] In another study, short-term cultured bone marrow-derived cells with “EPC characteristics”
lead to partial recovery in alveolar septal number in this same animal model.[38] Short-term cultured EPCs impaired alveolar growth in normoxic pups and long-term
cultured cells caused some aberrant tissue growth raising the importance of safety
and long-term studies in the preclinical setting.
IV infusion of human cord blood-derived ECFCs into immune-deficient rats and mice
exposed to hyperoxia promotes alveolar and lung vascular growth, and attenuates pulmonary
hypertension.[35] These effects persist after 10 months and no tumor formation was noted. Surprisingly,
ECFCs do not engraft and seem to act through a paracrine effect since the cell-free
ECFC-derived conditioned media exert similar benefit to whole cell therapy.[35] In a bleomycin-induced neonatal lung injury model, ECFC-derived conditioned media
had no effect on lung growth but attenuated pH.[39]
Focused preclinical studies are required before clinical translation of this cell
type can be considered in neonates.
CD34+ Mononuclear Cells—The Old Fashioned, Straight from the Bone Marrow Transplant Literature
Definition
Hematopoietic stem cell (HSC) transplantation has been practiced for over 50 years
in cancer patients to reconstitute bone marrow function. The consideration of cell
therapy for regenerative indications is new.
Rationale and Biological Plausibility
Within HSCs, CD34+ represents a marker for engraftment. Earlier studies suggested some lung engraftment
and differentiation of bone marrow-derived cells.[40] These observations have now been dispelled[41] and the rationale for using these mononuclear cells (MNCs) should be re-examined.
Preclinical Evidence
Four studies have explored the lung protective effect of MNCs.[42]
[43]
[44]
[45] Intranasal inoculation of fresh human cord blood CD34+ MNCs into neonatal mice with apoptosis-induced lung injury resulted in sparse engraftment
and very limited alveolar epithelial differentiation yet improved lung growth 1 year
after transplantation. IV injection of MNCs from cryopreserved cord blood into hyperoxia-exposed
newborn mice significantly attenuated methacholine-induced airway hyperreactivity,
and mildly improved alveolarization, lung compliance, and elastance. Interestingly,
total nucleated cells had no effect and granular cells caused high mortality and an
emphysematous phenotype.[46] Intraperitoneal injection of human umbilical cord blood MNCs in a double-hit mouse
model combining antenatal hypoxia and postnatal hyperoxia improved septal thickness
and decreased Tgfβ3 mRNA expression and proinflammatory IL-1β. Finally, intratracheal
injection of cord blood MNCs in neonatal hyperoxic rats had no effect on alveolarization
and lower anti-inflammatory and no angiogenic effects compared with cord blood-derived
MSCs or human adipose tissue-derived MSCs.[47] This study highlights an interesting debate about therapeutic potency of various
cell therapies.
Human Amniotic Fluid Stem Cells—Don't Throw Away the Bath Water
Definition
The discovery of cells in the amniotic fluid displaying stem cell characteristics
provided yet another potential perinatal source for therapeutic cells. These human
amniotic fluid stem cells (hAFSCs) express embryonic and adult stem cell markers,
expand extensively without feeders, and are multipotent as they can differentiate
into cell types representing each embryonic germ layer, including cells of adipogenic,
osteogenic, myogenic, endothelial, neuronal, and hepatic lineages.
Rationale and Biological Plausibility
Previous literature suggests anti-inflammatory effects of hAFSCs[48] as well as the ability to integrate and differentiate into epithelial lung lineages.[49]
Preclinical Evidence
One study so far explored the lung protective effect of hAFSCs.[50] Intratracheal injection after established hyperoxic lung injury in neonatal rats
showed low hAFSC retention, improved alveolar and lung vascular structure as well
as increased vascular endothelial growth factor expression, suggesting yet again—despite
their multipotency—a paracrine mechanism of action. Given the ease of access to other
perinatal sources such as cord, cord blood, and placenta, it remains to be tested
if hAFSCs provide superior repair properties than MSCs and hAECs.
Induced Pluripotent Stem Cells—Personalized Cell Therapy
Definition
Induced pluripotent stem cells (iPSCs) are differentiated cells that have been genetically
reprogrammed to an embryonic stem cell (ESC)-like state by forced expression of a
combination of four transcription factors (Oct3/4, Sox2, Klf4, and c-Myc) important
for maintaining the defining properties of ESCs.[51]
[52]
iPSCs possess desirable characteristics for use as a cell therapy. Like ESCs, iPSCs
have the potential to become any cell type in the body. Increasingly robust methods
of differentiation into alveolar epithelial type 2 cells (AT2) are described.[53] Human iPSCs (hiPSCs) can be generated from each patient, thus representing an autologous
source of cells that overcome the likelihood of immune-mediated rejection. An unlimited
number of cells can be generated from hiPSCs as they can be maintained and expanded
in vitro.
Rationale and Biological Plausibility
AT2s, a subset of which represent distal lung progenitor cells, promote normal lung
growth and repair after injury.[42] AT2 depletion is postulated to contribute to persistent lung injury in BPD.[42]
[43]
[44]
[45]
Preclinical Evidence
One study so far explored the lung-protective effect of iPSC-derived AT2s. Proof-of-concept
airway delivery of primary murine AT2s prevent hyperoxia-induced impairment in lung
function and alveolar growth in neonatal mice.[54] Undifferentiated murine and human (h)iPSCs also preserve lung function and alveolar
growth but cause local teratoma formation and systemic cellular infiltration in various
organs. Conversely, airway delivery of hiPSC-derived AT2 improves lung function and
structure without evidence of tumor formation at 8 months and some degree of engraftment.
Further studies need to confirm the mechanism of action and importantly, the safety
of differentiated iPSCs. The pace of advances in the iPSC field is impressive—as indicated
by the recent description of fail-safe iPSCs[55]—and the ancillary knowledge generated from iPSC biology may lead to a better understanding
of organ regeneration.
Overall, these observations indicate stem cell-based therapies as a growing field
of investigation with MSCs and hAECs as the front runners since they have entered
early-phase clinical trials. A common denominator of all these putative cell products—independent
of the original rationale and biological plausibility for using these cells in the
first place—is the lack of engraftment and the absence of potency assays to predict
therapeutic efficacy in vivo before infusing the product.
Knowledge Gaps
Cell Manufacturing
Clinical-grade cell products have to be manufactured according to good manufacturing
practices to ensure a consistent and controlled product according to quality standards.
Since the “process is the product,” each step along the manufacturing process will
affect the quality and thus the therapeutic potency of the product.[56]
For example, preconditioning of bone marrow MSCs by ex vivo exposure to hyperoxia
prior to intratracheal injection enhances the lung protective effect in neonatal hyperoxia-induced
lung injury in rats.[57] Interestingly, female bone marrow-derived MSCs seem to have greater therapeutic
efficacy than male MSCs on lung inflammation and vascular remodeling in this same
animal model and this effect is more pronounced in male animals.[58] Conversely, exposure to surfactant reduces MSC viability and combined MSC + surfactant
administration does not exert additive lung protective effects on lung structure in
hyperoxia-induced lung injury.[59]
These findings indicate that much more needs to be learned about the biology and manufacturing
of cell-based therapies, starting with the cell source (adult bone marrow, cord blood,
cord tissue, placenta, etc.), isolation–culture–expansion techniques, cryopreservation
protocols, the choice of using a frozen versus a fresh cell product, an autologous
versus an allogeneic strategy, the cell type, or a cell versus a cell-free product.[60]
Indeed, one exciting new avenue is the recognition that MSCs exert their therapeutic
benefit through the release of EVs. Amongst these EVs, exosomes, membrane-derived
nano-sized particles that contain proteins and miRNA, have attracted particular attention.[61] Exosomes perform as well as if not better in protecting the lung from hyperoxic
injury.[9]
[62]
[63]
[64] Rigorous preclinical studies need to confirm the initial results of these exploratory
studies. However, similar to the actual cell product, manufacturing at the clinical
scale and prediction of the bioactivity remain a challenge for the implementation
of these “nanotherapies for micropreemies.”[65]
Potency Assays
An ideal scenario can be found in the HSC field where CD34 represents a marker for
engraftment and thus a potential indicator of successful replenishment of bone marrow
function. Unfortunately, cell therapies in regenerative medicine still lack reliable
and rapid disease- and mechanism-specific in vitro assays to predict their therapeutic
efficacy. In vivo tests are still lacking. Recent advances in 3D lung organoids that
replicate an organ-in-a-dish may yield more reliable results than traditional culture
techniques.[66] The most promising approach today is the CLinical Indications Prediction (CLIP)
scale to predict how donor-to-donor heterogeneity and culture conditions impact the
therapeutic efficacy of MSCs[67]: high Twist1 levels predict proangiogenic properties, while low Twist1 levels predict
a more anti-inflammatory activity. Combined with omic technology, a multipanel of
factors may achieve accurate prediction of the potency of a given cell product which
in turn would substantially benefit our understanding of the mechanism of action and
facilitate the manufacturing process.
