Regenerative medicine as it applies to spina bifida is a multi-pronged endeavor involving spinal cord repair, tissue engineering and fetal regeneration, all of which can mutually overlap to variable extents. The efforts involving spinal cord repair, whether they be cell-based or not, are virtually indistinguishable from the enormous body of work related to spinal cord recovery after traumatic injury. Tissue engineering, on the other hand, can involve a variety of structures besides constructs used for covering the spina bifida defect, for example the urinary bladder, bone, muscle and skin. This brief review will not delve into any of these two main areas, which actually can also involve fetal interventions within their respective realms, but rather be devoted to a very recent development making use of the uniquely enhanced ability of the fetus to repair, or regenerate areas of tissue damage, coined transamniotic stem cell therapy, or TRASCET. TRASCET is a still experimental therapeutic paradigm for the treatment of not only spina bifida, but also other birth defects, based on the principle of harnessing/enhancing the normal biological role of a select population of stem cells that naturally occur in the amniotic fluid, specifically amniotic fluid-derived mesenchymal stem cells (afMSCs), for therapeutic benefit.
It was not until only less than a decade ago that a biological role for any cell present in the amniotic fluid was first described, specifically the activity of afMSCs in fetal wound healing [1]. In consecutive experiments in fetal lambs, our group has shown that, although not absolutely essential to the healing process, afMSCs populate fetal wounds, expedite wound closure and enhance its extracellular matrix profile. When compared with healing at any stage of postnatal life, wound healing in the fetus is a process more akin to tissue regeneration than to tissue repair. It involves significantly less inflammation and can be almost scarless, particularly early in gestation. The mechanisms behind the fetus’ greatly enhanced capacity to heal wounds have yet to be fully understood. It is known to encompass peculiarities of molecular pathways and gene expression patterns, yet that work uncovered a thitherto overlooked cellular component as well. Further, that finding was in accordance with the well-established knowledge that mesenchymal stem cells from other sources, most notably bone marrow, are known to home in to a variety of sites of tissue damage and help promote local repair.
An extensive discussion on the mechanisms by which afMSCs contribute to fetal tissue repair does not belong in this summarized review and in fact it is likely that many still remain to be revealed. It suffices to state that such a finding has added a new dimension to fetal and even postnatal wound healing, while lending biological support to the notion of using afMSCs in regenerative strategies. From a translational perspective, afMSCs are more applicable, at least in the prenatal setting, than any other stem cell, also because these can easily be autologous fetal cells procured from minute samples obtained by one of the least invasive of methods – a plain amniocentesis. Disease-associated stem cells, i.e. other than afMSCs, such as neural stem cells known to be present in the setting of spina bifida, can also be components of the TRASCET approach. However, when compared with afMSCs, these cells follow different engraftment routes and lead to different effects on the host and will not be discussed here.
afMSC-based TRASCET in experimental spina bifida
In three studies published as of this writing, TRASCET based on afMSCs has shown to have beneficial impact in a rodent model of spina bifida (2–4). In the first experiment, time-dated pregnant Sprague-Dawley dams ( 24) exposed to retinoic acid for the induction of fetal neural tube defects were divided into three groups. Group I had no further manipulations. Groups II and III received volume-matched intra-amniotic injections of either saline (group II) or a suspension of 2 10 cells/mL of afMSCs (group III) blindly in all fetuses ( 202) on gestational day 17 (term 21–22 days). Infused afMSCs consisted of syngeneic Lewis rat cells with identity confirmed by flow cytometry for CD29, CD44, CD45, CD73, and CD90 expressions, labeled with green fluorescent protein (GFP; 77–89% positivity by FACS analysis). Animals were killed before term. Statistical comparisons were performed by generalized estimating equations, ANOVA, the Wald test, and Bonferroni comparisons, as appropriate ( 0.05). A total of 165 fetuses were viable at euthanasia. Among fetuses with spina bifida (96/165; 58%), there were no significant differences in the overall dimensions of the discernible defect across the groups ( 0.19). However, there was a statistically significant increase in the proportion of fetuses with variable degrees of coverage (some complete) of the defect by a thin, rudimentary skin, confirmed histologically, in group III ( 0.001), with no differences between groups I and II ( 0.98). Donor afMSCs were identified in 83% (33/40) of the fetuses in that group via immunohistochemistry for GFP, interestingly preferably engrafting in bone. It could be concluded that afMSCs could induce partial or complete coverage of experimental spina bifida in the retinoic acid rodent model after concentrated intra-amniotic injection, seemingly via a paracrine effect. This study introduced the TRASCET concept as a potential option for the prenatal management of spina bifida [2].
Another experiment, subsequent to that initial report, was devoted to determining whether this therapeutic strategy could also have any impact on the Chiari-II malformation almost universally associated with spina bifida, also using the retinoic acid rodent model. Time-dated pregnant Sprague-Dawley dams ( 11) exposed to retinoic acid for the induction of fetal neural tube defects were divided into two groups: one ( 6) had no further manipulations and another ( 5) received volume-matched intra-amniotic injections of a suspension of 2 10 cells/mL of afMSCs blindly in all fetuses on gestational day 17 (term 21–22 days). Again, infused afMSCs again consisted of syngeneic Lewis rat cells with mesenchymal identity confirmed by flow cytometry, labeled with GFP. Animals were killed before term, when fetuses were divided into three groups: untreated controls with isolated spina bifida ( 21); isolated spina bifida treated with transamniotic delivery of afMSCs ( 28); and normal controls ( 13). Analyses included magnetic resonance imaging (MRI) with a high resolution (sub-millimeter) scanner and histology. The Chiari-II malformation was assessed on MRI by computer-generated specific angular and linear measurements of brainstem and cerebellar placement in relation to the basooccipital bone and the base of the skull, respectively. Statistical analyses were by Pearson chi-square, Fisher’s exact test, and ANOVA with Bonferroni comparisons (2-tailed 0.05). As expected, there was a statistically significant increase in the proportion of fetuses with variable degrees of coverage of the spina bifida by a rudimentary skin confirmed histologically in the afMSC-treated group ( 0.001). Overall, there were statistically significant differences across the groups in linear and angular measurements of brainstem placement ( 0.001), with the untreated group displaying the highest degree of caudal displacement. All pairwise comparisons of these parameters were statistically significant, with 0.014 between treated and normal controls in angular brainstem (caudal) displacement and 0.001 for all other angular and linear pairwise comparisons. Differences in cerebellar placement were also noted, albeit less pronounced, with 0.001 overall and significance in most pairwise comparisons, except between treated and untreated groups ( 0.10). Donor afMSCs were identified in 71% (20/28) of fetuses in the treated group via immunohistochemistry for GFP. It could be concluded that induced coverage of spina bifida by concentrated transamniotic delivery of amniotic mesenchymal stem cells does not completely reverse, however does minimize Chiari-II malformation in the retinoic acid rodent model. This further suggested that afMSC-based TRASCET could become a potential alternative or adjuvant for the prenatal management of spina bifida.
In addition to being a minimally invasive option using autologous fetal cells, another benefit of TRASCET is that it can be offered much earlier in gestation than surgical interventions. Different fetal-derived mesenchymal stem cells could be an option for TRASCET, for which the optimal cell type perhaps remains to be determined. In a subsequent study, we compared the effects of placenta-derived mesenchymal stem cells (pMSCs) with afMSCs for TRASCET in that same rodent model of spina bifida. To that end, time-dated pregnant Sprague-Dawley dams exposed to retinoic acid for the induction of fetal neural tube defects ( 29) were divided into four groups. The untreated group had no further manipulations. The other three groups received volume-matched intra-amniotic injections into all fetuses of either saline ( 38 fetuses) or a suspension of 2 10 cells/mL of afMSCs ( 73), or pMSCs ( 115) on gestational day 17. Infused afMSCs and pMSCs both consisted of syngeneic Lewis rat cells with identity confirmed by flow cytometry for CD29, CD44, CD45, CD73, and CD90 expressions, labeled with either fluorescent cytoplasmic nanocrystals or green fluorescent protein. Animals were killed before term. Defect coverage was categorized only if the presence of a rudimentary skin in the area was confirmed histologically. Statistical comparisons were by ANOVA and the Fisher’s exact test (2-tailed 0.05). Overall survival was not significantly different across the treatment groups (78–89%; 0.08–1.3 in pairwise comparisons). Among survivors with isolated spina bifida ( 100), there were statistically significant higher rates of any degree of defect coverage (whether partial or complete) in both the afMSC and pMSC groups when compared with the saline and untreated groups ( 0.001–0.03 in pairwise comparisons). There were no significant differences in the coverage rates between the afMSC and pMSC groups ( 1.00) and between the saline and untreated groups ( 0.35). There were no significant differences in the proportion of fetuses with complete defect coverage between the afMSC and pMSC groups ( 0.69). Interestingly, in that study, for the first time we noticed some percentage of coverage, albeit quite limited and none complete, in the untreated group. That finding further supported the notion that the inherent biological activity of afMSCs constitute a central underlying mechanism of defect coverage triggered by TRASCET, which may be simply augmenting such activity via the large number of cells delivered. Unfortunately, in that work the fluorescent cytoplasmic nanocrystal label did not allow for conclusive comparisons of engraftment patterns between the cell-based groups. We could conclude that both placental and amniotic mesenchymal stem cells can induce comparable rates of partial or complete coverage of experimental spina bifida after concentrated intra-amniotic injection in the rodent model. This finding broadens the options for timing and cell source for TRASCET as a potentially practicable alternative in the prenatal management of spina bifida.
Clinical translation of TRASCET
To date, cell-based therapies based on controlling and/or enhancing the biological role that the applicable donor cells normally play in nature have been the only ones to lead to major widespread impact in health care (e.g. different forms of blood transfusions and bone marrow transplantation). Therein lays much of the appeal of TRASCET, in that it is also based on the augmentation of the normal biological activities of select cells, in this case in the prenatal period and in a unique environment. Such equivalence supports the notion that TRASCET could be highly transformative to patient care, not only for spina bifida but also other diseases. Another facet of this pioneering strategy is its practicality as an outpatient procedure.
Novel cell-based therapies such as TRASCET require approval from the FDA prior to any form of clinical application, be it a level I, II, or III clinical trial, be it even anecdotal compassionate use. A long-standing concern of that agency in trials of cell-based therapies is the potential for uncontrolled cellular growth after cell delivery in vivo, following cell isolation and expansion in vitro. While we have documented survival of donor afMSCs for extended periods of time in vivo in numerous previous studies with no neoplastic transformation whatsoever, nor any evidence of dysfunction in any organ system, the FDA demands additional documentation of mid to long term safety prior to an eventual clinical trial, as well as data from large animal models, all of which are currently being pursued. Importantly, the TRASCET perspective is to be considered against the backdrop of a human afMSC manufacturing process previously described by our group, which has already been scrutinized by the FDA, providing support to its eventual regulatory viability, hopefully in a not too distant future [5, 6].
