Tracheal Replacement Therapy with a Stem Cell-Seeded Graft: Lessons from Compassionate Use Application of a GMP-Compliant Tissue-Engineered Medicine

Tracheal replacement for the treatment of end‐stage airway disease remains an elusive goal. The use of tissue‐engineered tracheae in compassionate use cases suggests that such an approach is a viable option. Here, a stem cell‐seeded, decellularized tissue‐engineered tracheal graft was used on a compassionate basis for a girl with critical tracheal stenosis after conventional reconstructive techniques failed. The graft represents the first cell‐seeded tracheal graft manufactured to full good manufacturing practice (GMP) standards. We report important preclinical and clinical data from the case, which ended in the death of the recipient. Early results were encouraging, but an acute event, hypothesized to be an intrathoracic bleed, caused sudden airway obstruction 3 weeks post‐transplantation, resulting in her death. We detail the clinical events and identify areas of priority to improve future grafts. In particular, we advocate the use of stents during the first few months post‐implantation. The negative outcome of this case highlights the inherent difficulties in clinical translation where preclinical in vivo models cannot replicate complex clinical scenarios that are encountered. The practical difficulties in delivering GMP grafts underscore the need to refine protocols for phase I clinical trials. Stem Cells Translational Medicine2017;6:1458–1464

A comparison of tracheal scaffold strategies for pediatric transplantation in a rabbit model

OBJECTIVES/HYPOTHESIS

Despite surgical advances, childhood tracheal stenosis is associated with high morbidity and mortality. Various tracheal scaffold strategies have been developed as the basis for bioengineered substitutes, but there is no consensus on which may be superior in vivo. We hypothesized that there would be no difference in morbidity and mortality between three competing scaffold strategies in rabbits.

STUDY DESIGN

Pilot preclinical study.

METHODS

Tracheal scaffolds were prepared by three methods that have been applied clinically and reported: preserved cadaveric ("Herberhold") allografts, detergent-enzymatically decellularized allografts, and synthetic scaffolds (nanocomposite polymer [polyhedral oligomeric silsesquioxane poly(carbonate-urea) urethane (POSS-PCU)]). Scaffolds were implanted into cervical trachea of New Zealand White rabbits (n = 4 per group) without cell seeding. Control animals (n = 4) received autotransplanted tracheal segments using the same technique. Animals underwent bronchoscopic monitoring of the grafts for 30 days. Macroscopic evaluation of tissue integration, graft stenosis, and collapsibility and histological examinations were performed on explants at termination.

RESULTS

All surgical controls survived to termination without airway compromise. Mild to moderate anastomotic stenosis from granulation tissue was detected, but there was evidence suggestive of vascular reconnection with minimal fibrous encapsulation. In contrast, three of the four animals in the Herberhold and POSS-PCU groups, and all animals receiving decellularized allografts, required early termination due to respiratory distress. Herberhold grafts showed intense inflammatory reactions, anastomotic stenoses, and mucus plugging. Synthetic graft integration and vascularization were poor, whereas decellularized grafts demonstrated malacia and collapse but had features suggestive of vascular connection or revascularization.

CONCLUSIONS

There are mirror-image benefits and drawbacks to nonrecellularized, decellularized, and synthetic grafts, such that none emerged as the preferred option. Results from prevascularized and/or cell-seeded grafts (as applied clinically) may elucidate clearer advantages of one scaffold type over another.

LEVEL OF EVIDENCE

NA. Laryngoscope, 127:E449-E457, 2017.

Rapid Expansion of Human Epithelial Stem Cells Suitable for Airway Tissue Engineering

RATIONALE

Stem cell-based tracheal replacement represents an emerging therapeutic option for patients with otherwise untreatable airway diseases including long-segment congenital tracheal stenosis and upper airway tumors. Clinical experience demonstrates that restoration of mucociliary clearance in the lungs after transplantation of tissue-engineered grafts is critical, with preclinical studies showing that seeding scaffolds with autologous mucosa improves regeneration. High epithelial cell-seeding densities are required in regenerative medicine, and existing techniques are inadequate to achieve coverage of clinically suitable grafts.

OBJECTIVES

To define a scalable cell culture system to deliver airway epithelium to clinical grafts.

METHODS

Human respiratory epithelial cells derived from endobronchial biopsies were cultured using a combination of mitotically inactivated fibroblasts and Rho-associated protein kinase (ROCK) inhibition using Y-27632 (3T3+Y). Cells were analyzed by immunofluorescence, quantitative polymerase chain reaction, and flow cytometry to assess airway stem cell marker expression. Karyotyping and multiplex ligation-dependent probe amplification were performed to assess cell safety. Differentiation capacity was tested in three-dimensional tracheospheres, organotypic cultures, air-liquid interface cultures, and an in vivo tracheal xenograft model. Ciliary function was assessed in air-liquid interface cultures.

MEASUREMENTS AND MAIN RESULTS

3T3-J2 feeder cells and ROCK inhibition allowed rapid expansion of airway basal cells. These cells were capable of multipotent differentiation in vitro, generating both ciliated and goblet cell lineages. Cilia were functional with normal beat frequency and pattern. Cultured cells repopulated tracheal scaffolds in a heterotopic transplantation xenograft model.

CONCLUSIONS

Our method generates large numbers of functional airway basal epithelial cells with the efficiency demanded by clinical transplantation, suggesting its suitability for use in tracheal reconstruction.

Tracheal Replacement Therapy with a Stem Cell-Seeded Graft: Lessons from Compassionate Use Application of a GMP-Compliant Tissue-Engineered Medicine

Tracheal replacement for the treatment of end-stage airway disease remains an elusive goal. The use of tissue-engineered tracheae in compassionate use cases suggests that such an approach is a viable option. Here, a stem cell-seeded, decellularized tissue-engineered tracheal graft was used on a compassionate basis for a girl with critical tracheal stenosis after conventional reconstructive techniques failed. The graft represents the first cell-seeded tracheal graft manufactured to full good manufacturing practice (GMP) standards. We report important preclinical and clinical data from the case, which ended in the death of the recipient. Early results were encouraging, but an acute event, hypothesized to be an intrathoracic bleed, caused sudden airway obstruction 3 weeks post-transplantation, resulting in her death. We detail the clinical events and identify areas of priority to improve future grafts. In particular, we advocate the use of stents during the first few months post-implantation. The negative outcome of this case highlights the inherent difficulties in clinical translation where preclinical in vivo models cannot replicate complex clinical scenarios that are encountered. The practical difficulties in delivering GMP grafts underscore the need to refine protocols for phase I clinical trials. Stem Cells Translational Medicine 2017;6:1458-1464.

Vacuum-assisted decellularization: an accelerated protocol to generate tissue-engineered human tracheal scaffolds

Patients with large tracheal lesions unsuitable for conventional endoscopic or open operations may require a tracheal replacement but there is no present consensus of how this may be achieved. Tissue engineering using decellularized or synthetic tracheal scaffolds offers a new avenue for airway reconstruction. Decellularized human donor tracheal scaffolds have been applied in compassionate-use clinical cases but naturally derived extracellular matrix (ECM) scaffolds demand lengthy preparation times. Here, we compare a clinically applied detergent-enzymatic method (DEM) with an accelerated vacuum-assisted decellularization (VAD) protocol. We examined the histological appearance, DNA content and extracellular matrix composition of human donor tracheae decellularized using these techniques. Further, we performed scanning electron microscopy (SEM) and biomechanical testing to analyze decellularization performance. To assess the biocompatibility of scaffolds generated using VAD, we seeded scaffolds with primary human airway epithelial cells in vitro and performed in vivo chick chorioallantoic membrane (CAM) and subcutaneous implantation assays. Both DEM and VAD protocols produced well-decellularized tracheal scaffolds with no adverse mechanical effects and scaffolds retained the capacity for in vitro and in vivo cellular integration. We conclude that the substantial reduction in time required to produce scaffolds using VAD compared to DEM (approximately 9 days vs. 3–8 weeks) does not compromise the quality of human tracheal scaffold generated. These findings might inform clinical decellularization techniques as VAD offers accelerated scaffold production and reduces the associated costs.

Tissue-Engineered Tracheal Replacement in a Child: A 4-Year Follow-Up Study

In 2010, a tissue-engineered trachea was transplanted into a 10-year-old child using a decellularized deceased donor trachea repopulated with the recipient's respiratory epithelium and mesenchymal stromal cells. We report the child's clinical progress, tracheal epithelialization and costs over the 4 years. A chronology of events was derived from clinical notes and costs determined using reference costs per procedure. Serial tracheoscopy images, lung function tests and anti-HLA blood samples were compared. Epithelial morphology and T cell, Ki67 and cleaved caspase 3 activity were examined. Computational fluid dynamic simulations determined flow, velocity and airway pressure drops. After the first year following transplantation, the number of interventions fell and the child is currently clinically well and continues in education. Endoscopy demonstrated a complete mucosal lining at 15 months, despite retention of a stent. Histocytology indicates a differentiated respiratory layer and no abnormal immune activity. Computational fluid dynamic analysis demonstrated increased velocity and pressure drops around a distal tracheal narrowing. Cross-sectional area analysis showed restriction of growth within an area of in-stent stenosis. This report demonstrates the long-term viability of a decellularized tissue-engineered trachea within a child. Further research is needed to develop bioengineered pediatric tracheal replacements with lower morbidity, better biomechanics and lower costs.

Stimulatory effects of estrogen and progesterone on proliferation and differentiation of normal human osteoblast-like cells in vitro

Here we report that osteoblast-like cells derived from female and male adult human trabecular bone are able to directly respond to 17 beta-estradiol (E2) and progesterone (P). In short-term (1 day) cultures using serum-free and phenol red-free medium, both steroid hormones were found to stimulate DNA synthesis and growth of the human osteoblast-like cells. P was more potent in stimulating osteoblast growth compared to E2. On the other hand, E2 showed a stronger differentiation-inducing effect as determined by analysis of the number of cells displaying cytochemical alkaline phosphatase (AP) activity, a marker for the mature osteoblast phenotype. Combination of E2 and P resulted in a further increase in DNA synthesis, but did not further affect the number of cells expressing AP activity. In conclusion, female sex steroids may be involved in regulating bone mass in human adults via a direct anabolic action on the bone forming cells.

Effects of recombinant human insulin-like growth factor I and II (IGF-I/-II) and growth hormone (GH) on the growth of normal adult human osteoblast-like cells and human osteogenic sarcoma cells

Recombinant GH and IGF-I/-II were studied for their capacity to directly influence the growth of human bone cells maintained under defined serum-free conditions. Normal human osteoblast-like cell (HOB) cultures were established from trabecular bone explants obtained from adult human femoral head samples. IGF-I and IGF-II as well as GH stimulated the growth of the HOB cultures in a dose-dependent manner. Growth stimulatory effects were also found using the human osteogenic sarcoma cell line, SaOS-2. IGF-I and -II were powerful enhancers of the SaOS-2 cell growth and their effects greatly exceeded GH effects on these cells. The role of endogenously produced IGFs was studied using a specific monoclonal antibody to IGF-I having a partial cross-reactivity with IGF-II (sm1.2B). The IGF-I stimulated HOB growth was completely neutralised by sm1.2B to the level of control+antibody which in general showed a slight stimulation compared to controls without the antibody. Interestingly, sm1.2B was not able to interfere with the action of GH on the HOB suggesting that GH effects may be attributed to an action independent of endogenous IGF-I/-II. Unlike the HOB, SaOS-2 cells were strongly inhibited by sm1.2B in control medium indicating an autocrine role of IGF-I/-II in osteosarcoma cell growth. Sm1.2B completely neutralised the stimulatory effects of IGF-I and IGF-II on the SaOS-2 cells. Moreover, GH effects on the osteogenic sarcoma cells were abolished by the anti-IGF antibody showing that GH was acting via endogenously produced IGFs.(ABSTRACT TRUNCATED AT 250 WORDS)

Osteoclast growth factor activity in medium conditioned by fetal rat bones

The presence and biological activity of an Osteoclast Growth Factor (OGF) was investigated in serum-free medium conditioned by periostless fetal rat calvaria in culture. OGF activity was assessed using in vitro systems of fetal rat long bones and adult rat bone marrow cells. Rat calvaria conditioned medium (RCCM) increased the number of osteoclasts in the long bone cultures, partly due to stimulation of progenitor proliferation. RCCM did not exert a direct bone-resorbing activity (45Calcium release assay) on the pre-existing osteoclasts residing in the long bones, but stimulated bone resorption in long term cultures, apparently in an indirect manner by enhancing the number of osteoclasts. In cultures of bone marrow cells isolated from adult rats, RCCM markedly stimulated the formation of mononuclear cells which were positively stained for tartrate-resistant acid phosphatase (TRAP). The osteoclastic nature of the cells was confirmed by specific labeling with 125I-calcitonin. Formation of the TRAP-positive cells was significantly inhibited by salmon calcitonin. CM from fetal rat skin cultures did not display a significant OGF activity. Furthermore, unlike the bone marrow cells, peritoneal macrophages did not respond to RCCM and remained devoid of TRAP activity. Neutralization experiments with a specific antibody to GM-CSF indicated that OGF activity in the RCCM could not be ascribed to this hemopoietic growth factor. Secretion of OGF activity was mainly dependent on protein synthesis as addition of cycloheximide to the calvaria cultures significantly inhibited the secretion of OGF into the medium. G3000 HPLC fractionation of RCCM revealed two major OGF peaks with Mr 14,000 and 70,000. Two subsequent reverse-phase HPLC steps using the lower Mr OGF fraction led to a highly purified OGF fraction. The results of this study further provide evidence that bone tissue produces factor(s) which specifically govern the process of osteoclast development, thus providing information about one of the mechanisms controlling bone resorption.

Longitudinal bone growth in vitro: effects of insulin-like growth factor I and growth hormone

Longitudinal growth was studied using an in vitro model system of intact rat long bones. Metatarsal bones from 18- and 19-day-old rat fetuses, entirely (18 days) or mainly (19 days) composed of chondrocytes, showed a steady rate of growth and radiolabelled thymidine incorporation for at least 7 days in serum-free media. Addition of recombinant human insulin-like growth factor-I to the culture media resulted in a direct stimulation of the longitudinal growth. Recombinant human growth hormone was also able to stimulate bone growth, although this was generally accomplished after a time lag of more than 2 days. A monoclonal antibody to IGF-I abolished both the IGF-I and GH-stimulated growth. However, the antibody had no effect on the growth of the bone explants in control, serum-free medium. Unlike the fetal long bones, bones from 2-day-old neonatal rats were arrested in their growth after 1-2 days in vitro. The neonatal bones responded to IGF-I and GH in a similar fashion as the fetal bones. Thus in this study in vitro evidence of a direct effect of GH on long bone growth via stimulating local production of IGF by the growth plate chondrocytes is presented. Furthermore, endogenous growth factors, others than IGFs, appear to play a crucial role in the regulation of fetal long bone growth.

Stimulation of macrophage growth and multinucleated cell formation in rat bone marrow cultures by insulin-like growth factor I

In this study the effects of rhIGF-I on macrophage differentiation and growth have been studied using liquid suspension cultures of rat bone marrow cells. IGF-I stimulated macrophage growth in a dose-dependent manner, a maximum response was found at a concentration of 20 ng/ml. IGF-I effects could be ascribed to stimulation of both postmitotic and proliferating cells. A remarkable finding was that IGF-I induced formation of multinucleated cells (MNC). The MNC resembled macrophage-like cells (AcP, NSE positive). A monoclonal antibody to rhIGF-I significantly inhibited IGF-stimulated macrophage growth and MNC formation. A specific antibody to mouse CSF-1 reduced IGF-stimulated macrophage growth in mouse bone marrow cultures indicating that IGF-I effects could, at least in part, be ascribed to endogenous production of CSF-1. These findings indicate that IGF-I in concert with locally induced CSF-1 can influence the differentiation and growth of bone marrow-derived macrophages.

Retinoic acid and 1,25-dihydroxyvitamin D3 stimulate osteoclast formation by different mechanisms

The effects of retinoic acid (RA) and 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3 on osteoclast formation were examined in intact fetal long bones of different ages/developmental stages maintained in organ culture using a chemically defined medium with or without the presence of serum. Besides stimulating bone resorption, RA and 1,25-(OH)2D3 increased the number of osteoclasts in 19-day-old fetal rat tibiae. Likewise, these bone-resorbing agents induced and stimulated osteoclast formation in 19- and 18-day-old metatarsal bones which were osteoclast-free at the beginning of the culture. The response to 1,25-(OH)2D3 was greatly enhanced by 10% fetal bovine serum (FBS) irrespective of the developmental stage of the long bone. The response to RA was not. Light microscopic autoradiography after labeling of the cultures with tritiated thymidine showed that both RA and 1,25-(OH)2D3 induced osteoclast differentiation from proliferating and postmitotic precursors. However, neither agent was able to stimulate proliferation of osteoclast progenitor cells in the older bones (19 days). Studies on the formation of osteoclast-like (tartrate-resistant acid phosphatase positive) cells in bone marrow cultures indicated that FBS was a potent inducer of osteoclast-like cell formation. In the presence of FBS, 1,25-(OH)2D3 significantly stimulated this response, but RA did not. The results demonstrate that although both RA and 1,25-(OH)2D3 stimulate osteoclast formation from proliferating and postmitotic precursors in long bones in vitro, they do so by different mechanisms.

Immunohistochemical localization of native and denatured collagen types I and II in fetal and adult rat long bones

Collagen turnover during rat long bone development and growth was investigated using immunofluorescence methods with specific polyclonal antibodies against native (triple helix) and denatured (breakdown products) forms of type I and II collagen. Labeling of cryostat sections with anti-native and denatured collagen type II antibodies resulted in a positive staining throughout the cartilage matrix of fetal and adult long bones. Likewise, native and denatured type I collagen could be detected in mineralized and non-mineralized bone matrix. Moreover, labeling with anti-denatured type I antibody evoked a strong intracellular staining of osteoblasts, but not of osteocytes. Denatured type I was also localized intra-pericellularly in the small chondrocytes comprising the primitive cartilage cores and the epiphyses of older long bones. On the other hand, apart from its localization in the cartilage matrix, denatured type II collagen was found specifically within the chondrocytes. These observations indicate that a continuous turnover of the major collagen types takes place in fetal and adult rat long bone tissue. Degradation of collagen apparently occurs intra- and extracellularly, and is mainly independent of the presence and activity of osteoclasts. The presence of denatured type I collagen in cartilage suggests that chondrocytes synthesize small amounts of type I collagen, which is immediately degraded to a denatured form.