Laryngotracheal stenosis results from pathologic scar deposition that critically narrows the tracheal airway and lacks effective medical therapies. Using a PLLA-PCL (70% poly-L-lactide and 30% polycaprolactone) stent as a local drug delivery system, potential therapies aimed at decreasing scar proliferation in the trachea can be studied.
Laryngotracheal stenosis (LTS) is a pathologic narrowing of the subglottis and trachea leading to extrathoracic obstruction and significant shortness of breath. LTS results from mucosal injury from a foreign body in the trachea, leading to tissue damage and a local inflammatory response that goes awry, leading to the deposition of pathologic scar tissue. Treatment for LTS is surgical due to the lack of effective medical therapies. The purpose of this method is to construct a biocompatible stent that can be miniaturized to place into mice with LTS. We demonstrated that a PLLA-PCL (70% poly-L-lactide and 30% polycaprolactone) construct had optimal biomechanical strength, was biocompatible, practicable for an in vivo placement stent, and capable of eluting drug. This method provides a drug delivery system for testing various immunomodulatory agents to locally inhibit inflammation and reduce airway fibrosis. Manufacturing the stents takes 28−30 h and can be reproduced easily, allowing for experiments with large cohorts. Here we incorporated the drug rapamycin within the stent to test its effectiveness in reducing fibrosis and collagen deposition. Results revealed that PLLA-PCL tents showed reliable rapamycin release, were mechanically stable in physiological conditions, and were biocompatible, inducing little inflammatory response in the trachea. Further, the rapamycin-eluting PLLA-PCL stents reduced scar formation in the trachea in vivo.
Laryngotracheal stenosis (LTS) is a pathologic narrowing of the trachea most often due to iatrogenic post-intubation injury. The combination of bacterial colonization, foreign body response to a tracheostomy or endotracheal tube, and patient-specific factors lead to an aberrant inflammatory response. This maladaptive immune response leads to the deposition of collagen in the trachea, resulting in luminal narrowing of the trachea and subsequent stenosis1,2. As current treatment for this disease is primarily surgical, developing an alternative medically-based treatment paradigm targeting the aberrant inflammatory and profibrotic pathways that lead to excessive collagen deposition has been studied. Rapamycin, which inhibits the mTOR signaling complex, has been shown to have immunosuppressive effects as well as a robust antifibroblast effect. However, when rapamycin is systemically administered, common side effects (e.g., hyperlipidemia, anemia, thrombocytopenia) can be pronounced3. The purpose of our methodology is to develop a vehicle for local drug delivery practicable for use in the airway that would lessen these systemic effects. Our assessments focus on investigating the local immune response to the drug delivery construct as well as its capacity to inhibit fibroblast function and alter the local immune microenvironment. Disease-specific outcomes include in vivo testing that evaluate markers of fibrosis.
Biodegradable drug-eluting stents have been used in animal models of disease in multiple organ systems, including the airway4. For the management of airway stenosis or collapse, previous investigations have used drug-coated silicone and nickel-based stents5. A PLLA-PCL construct was chosen for this particular method because of its drug elution profile and mechanical strength in physiological conditions over a period of 3 weeks, which has been demonstrated in previous published studies6. PLLA-PCL is also a biocompatible and biodegradable material already approved by the FDA4. Biocompatible stents eluting cisplatin and MMC have been studied in large animal models such as rabbits and dogs. However, in these animal models, stents were not placed in an animal model of disease and were implanted transcervically. This study provides a unique method for assessing a biocompatible drug-eluting stent placed transorally in a mouse model of airway injury and laryngotracheal stenosis. A biocompatible stent that elutes an immunomodulatory drug locally and can be miniaturized for study in a murine model is valuable for translational preclinical research. Previous attempts at stent utilization with other material constructs generated robust foreign body responses worsening the underlying inflammation that distinguishes LTS7. This methodology, to our knowledge, is the first of its kind to study the immunomodulatory and antifibrotic effects of a stent-based drug delivery system in a murine model of LTS. The murine model itself offers several advantages for studying the effects of an immunomodulatory drug on the trachea. Genetically modified mice and experimental cohorts of healthy and diseased mice can be studied, which can lead to experimental reproducibility and improve cost-effectiveness. Moreover, the delivery of the stent transorally into the mouse trachea mimics clinical delivery of such a stent in humans, which further highlights the translational advantage of this method. Finally, the relative ease with which the PLLA-PCL stent with the drug can be produced allows for modifications to deliver alternate drug therapies aimed at reducing scar formation in the trachea.
NOTE: All methods described here were approved by the Johns Hopkins University Animal Care and Use Committee (MO12M354).
1. Preparation of rapamycin in PLLA-PCL
2. Rapamycin elution testing
3. Creation of rapamycin-eluting PLLA-PCL murine airway stents
NOTE: Perform steps 3.2−3.9 using sterile materials and sterile technique to avoid contamination that would influence in vivo and in vitro applications.
4. Laryngotracheal stenosis induction in mice
5. Transoral PLLA-PCL stent placement in mice
6. Histologic preparation of samples
7. Stent biocompatibility in vivo
8. Mouse trachea quantitative gene expression analysis
The biodegradable PLLA-PCL stent construct loaded with rapamycin used in this study was capable of eluting rapamycin in a consistent and predictable fashion in physiological conditions (Figure 1). Figure 2 shows the PLLA-PCL stent casted around a 22 G angiocatheter for use in a murine model of LTS. To determine if the effects of rapamycin elution in the trachea is efficacious in attenuating fibrosis, measured changes in fibrosis-related gene expression and markers of acute inflammation can be assessed through gene expression analysis, flow cytometry, immunofluorescence, and ELISA. Successful placement of a miniaturized stent into the trachea of the mouse using the method described above has been demonstrated. A schematic of the method is shown in Figure 3. Figure 4 shows the miniaturized biocompatible stent in situ in the trachea as indicated by the black marker on the stent, which is visualized through the translucent mouse trachea. In initial experiments, using a 0.8 mm sialendoscope to confirm placement of the stent in the trachea was helpful. After 21 days, to confirm that the transoral placement of the stent was efficacious and the stent did not migrate from its originally placed position, the neck incision was reopened to determine the placement of the stent. As shown in Figure 4B, the black dye marker of the stent showed the stent maintained its position in the trachea. Resection of the trachea after 21 days of treatment with the stent is shown in Figure 4C.
Representative images of biocompatibility testing using immunofluorescent staining for markers of acute and chronic inflammation are shown in Figure 5. This demonstrated that the PLLA-PCL stent construct (without rapamycin) was not immunoreactive as determined by the minimal number of immune cells present after placement. It is important to note that in this method, normal uninjured tracheas were used and a PLLA-PCL stent without rapamycin was placed to determine the inflammatory response to the construct itself.
Next, to determine whether the rapamycin-eluting PLLA-PCL stent was effective in mitigating scars, we previously demonstrated gene expression changes in markers of acute inflammation and fibrosis6. Specifically, there is a 90.3 fold reduction (SEM ± 26.0; n = 4; p < 0.01) reduction in col1a1 at day 4, as well as the acute inflammatory markers INF-γ, CD11b, Arg-1, and IL-1B6. Though the differences in fold change for some genes were not significant, it is possible that with a greater cohort of mice, significance could be achieved. To determine whether there were changes to the trachea due to drug elution histologically, or whether there were changes to the trachea due to radial force exertion by the stent, we demonstrated that there was a decrease in the width of the lamina propria in those tracheas treated with rapamycin-eluting stents compared to those without rapamycin-eluting stents6.
Figure 1: Rapamycin PLLA-PCL elution. The PLLA-PCL construct containing 1% rapamycin demonstrated a consistent and predictable release of rapamycin over a 14 day period. Data points represent the mean ± SEM of sampled elution (n = 3). Please click here to view a larger version of this figure.
Figure 2: Stent casting. (A) The PLLA-PCL solution was allowed to dry around a 22 G angiocatheter. (B) The cast was then removed from the angiocatheter. (C) Stents were cut to 3 mm lengths for use in the mouse model6. Please click here to view a larger version of this figure.
Figure 3: Transoral stent placement in mice. (A) The stent was loaded onto an empty angiocatheter and placed transorally into the trachea. (B) The black dye marking on the stent may be seen through a transcervical incision to confirm its position in the mouse trachea. (C) Representative drawing of the stent in situ in the diseased mouse trachea6. Please click here to view a larger version of this figure.
Figure 4: In situ images of stent. (A–B) The stent with black dye markings may be seen in situ in the murine trachea. (C) An image of the stent and the murine laryngotracheal complex after harvest at 21 days6. Please click here to view a larger version of this figure.
Figure 5: Stent biocompatibility. Immunofluorescent staining for F4/80 (macrophages, red chromophore) and CD3 (green chromophore, T-lymphocytes) at day 4 revealed minimal inflammatory cells in the (A) uninjured trachea and (B) trachea with PLLA-PCL stent. This contrasts with (C) an injured trachea, with a thickened lamina propria and the presence of numerous cells with positive F4/80 and CD3 staining6. Please click here to view a larger version of this figure.
The most critical steps for successfully constructing and using a drug-eluting stent in vivo are 1) determining the optimal PLLA-PCL ratio for the desirable drug elution rate, 2) determining the appropriate concentration of drug to be eluted, 3) molding the stents around the angiocatheter for in vivo use, and 4) transorally delivering the stent into the mice after LTS induction without causing fatal airway obstruction.
While there are several methods for drug delivery using stents in animal models of airway disease, development of a biocompatible stent capable of drug delivery in a diseased murine model is a first. In developing this method for drug delivery, several modifications were made to the method. It was important to determine the appropriate PLLA-PCL composition for making stents that were mechanically rigid enough to be placed into the trachea and to remain in the trachea despite physiological secretions. A 70:30 composition of PLLA-PCL was decided upon for stent construction with an angiocatheter molding because it could successfully elute rapamycin in a predicable nature, be placed in a safe and reliable fashion in our mouse model, and not degrade in physiological conditions. A difficult and potentially manufacturer-dependent portion of this method is molding the stent around the angiocatheters. Initially, in constructing the in vivo stents, 22 G angiocatheters were placed inside the tip of a glass pipette and the polymer solution was poured into the space between the angiocatheter and the glass pipette, forming a cast of the space between. However, the wall of the stents resulting from this method often were too thin and were not able to be reliably placed into the trachea. A limitation of the current method for molding stents on the angiocatheter is the dependence on the manufacturer for consistency, and the need for meticulous attention to detail to ensure homogeneity in stent production. The potential for variance in thickness between casted stents needs to be addressed in future studies. With further studies, we hope to design a mold for a 22 G angiocatheter surrounded by another glass or noncorrosive material such that the space between the angiocatheter and the glass encasing is 50 µm, which we determined to be an adequate wall thickness of the stent for easy placement into the trachea.
There are several benefits to using a biocompatible stent to study drug delivery to the affected trachea. Overall, stents composed of metal or silicone that have been coated with a polymer containing the drug have been shown to produce granulation tissue and further inflammatory response to an already scarred portion of the trachea. This method, which interrogates the use of a stent made entirely of biomaterial that is biocompatible and also releases an immunomodulatory drug in a reliable manner is advantageous. The PLLA-PCL stent is also shown to be biocompatible in the murine model and does not elicit an acute inflammatory response. In studying drugs that can combat fibrosis, using a stent entirely composed of biocompatible material such as PLLA-PCL as described in this method is beneficial.
The advantage and ease of using this method to construct usable stents is that the composition of the PLLA-PCL can be varied, allowing for differences in release profiles for the drug mixed in the composition. Previous studies of the PLLA-PCL material show that variation in PLLA-PCL blends can lead to greater degradation of the material, allowing for faster drug release9. Moreover, using this method to construct stents that can be placed in mice is advantageous, as most stent studies for airway disease were done in larger animals10,11,12. The use of a mouse model that can be modified for different disease paradigms and allow for testing of the drug-eluting stent in a diseased state is ideal. Testing the drug-eluting stent in a diseased state and being able to compare its efficacy to those in animals without disease allows for greater experimental rigor. This method also shows how a drug-eluting stent for the airway can be placed transorally, as opposed to previous studies where stents were implanted surgically.
This study demonstrates a very usable platform for stent development and testing in a small animal model. However, other factors for use in human subjects must be considered. Given the firm and rigid nature of the stent it likely cannot be placed through a channeled flexible bronchoscope but will need to be placed transorally with direct laryngoscopy and rigid bronchoscopy. Ideally, the stent would be placed after balloon dilation of the airway to help mitigate restenosis. From the prospective of patient safety, the potential for the stent to migrate is a large concern as it could potentially lead to life threatening airway compromise. Secondarily, the potential for stent obstruction secondary to the accumulation of mucus or blood must also be considered. Further testing and development are needed to minimize these risks.
Future studies can utilize this method to test different drugs in the PLLA-PCL blend to understand further how different immunosuppressive treatments could mitigate scar formation in the trachea. Because such stents can be placed in mice, using a larger cohort of mice or mice with genetic modifications to scar-forming genes can also be tested. Future experiments can also include understanding the changes to the trachea that can occur after chronic implantation of the stent (3–6 months) and the changes to the lumen of the trachea as well as the gene expression profiles of inflammatory markers and fibrosis markers.
The authors have nothing to disclose.
National Institute on Deafness and Other Communication Disorders of the National Institutes of Health under award numbers 1K23DC014082 and 1R21DC017225 (Alexander Hillel). This study was also financially supported by the Triological Society and American College of Surgeons (Alexander Hillel), the American Medical Association Foundation, Chicago, IL (Madhavi Duvvuri) and a T32 NIDCD training grant (Kevin Motz).
1. For stent | |||
22-gauge angiocatheter | Jelco | 4050 | |
Dichloromethane | Sigma Aldrich | 270997-100ML | |
Glycerol | Fisher Scientific | 56-81-5 | Available from other vendors as well. |
PDLGA | Sigma Aldrich | 739955-5G | |
PLLA-PCL (70 : 30) | Evonik Industries AG | 65053 | |
Rapamycin | LC Laboratories | R-5000 | |
2. Animal surgery | |||
Wire brush | Mill-Rose Company | 320101 | |
3. For immunohistochemistry staining | |||
Antigen retrival buffer | Abcam | ab93678 | Available from other vendors as well; acidic pH needed |
DAPI | Cell Signaling | 8961S | |
DMEM | ThermoFisher Scientific | 11965-092 | Available from other vendors as well. |
FBS (Fetal Bovine Serum) | MilliporeSigma | F4135-500ML | |
Goat anti-rabbit-488 antibody | Lif technology | a11008 | |
Goat anti-rat-633 antibody | Lif technology | a21094 | |
Hydrophilic plus slide | BSB7028 | ||
PBS | ThermoFisher Scientific | 100-10023 | Available from other vendors as well. |
Rabbit anti-CD3 antibody | Abcam | ab5690 | |
Rat antiF4/80 antibody | Biolengend | 123101 | |
Zeiss LSM 510 Meta Confocal Microscope | Zeiss | ||
4. For quantative PCR | |||
0.5mm glass beads | OMNI International | 19-645 | |
Bead Mill Homoginizer | OMNI International | ||
Gene Specific Forward/Reverse Primers | Genomic Resources Core Facility | ||
Nanodrop 2000 spectrophotometer | Thermo Scientific | ||
Power SYBR Green Mastermix | Life Technologies | 4367659 | |
RNeasy mini kit | Qiagen | 80404 | |
StepOnePlus Real Time PCR system | Life Technologies |