Continuous Liquid Interface Production of 3D Printed Drug-Loaded Spacers to Improve Prostate Cancer Brachytherapy Treatment

Abstract

Brachytherapy, which is the placement of radioactive seeds directly into tissue such as the prostate, is an important curative treatment for prostate cancer. By delivering a high dose of radiation from within the prostate gland, brachytherapy is an effective method of killing prostate cancer cells while limiting radiation dose to normal tissue. The main shortcomings of this treatment are: less efficacy against high grade tumor cells, acute urinary retention, and sub-acute urinary frequency and urgency. One strategy to improve brachytherapy is to incorporate therapeutics into brachytherapy. Drugs, such as docetaxel, can improve therapeutic efficacy, and dexamethasone is known to decrease urinary side effects. However, both therapeutics have high systemic side effects. To overcome this challenge, we hypothesized that we can incorporate therapeutics into the inert polymer spacers that are used to correctly space brachytherapy seeds during brachytherapy to enable local drug delivery. To accomplish this, we engineered 3D printed drug-loaded brachytherapy spacers using continuous liquid interface production (CLIP) with different surface patterns to control drug release. These devices have the same physical size as existing spacers, allowing them to easily replace commercial spacers. We examined these drug-loaded spacers using docetaxel and dexamethasone as model drugs in a murine model of prostate cancer. We found that drug-loaded spacers led to higher therapeutic efficacy for brachytherapy, and there was no discernable systemic toxicity from the drug-loaded spacers.

1. Introduction

Prostate cancer is the most diagnosed cancer in men in the United States, with an estimated 249,000 new cases in 2021 [1]. With over 34,000 deaths, it is the second leading cause of cancer death in men. The most common curative treatments for localized prostate cancer involve radical prostatectomy or radiation therapy (RT). Radiotherapy treatments generally involve either external beam RT, interstitial brachytherapy, or both [2]. In one study examining the treatments received by patients with early-stage prostate cancer in the United States, 36% received brachytherapy as a component of treatment [3]. In brachytherapy, spacers are utilized to facilitate the spatial placement of seeds but serve no other function. The most common side effects of prostate brachytherapy include dysuria, urinary retention, frequency and urgency, as well as rectal irritation [4]. Dexamethasone (DM) is known to decrease acute side effects associated with prostate brachytherapy but long-term use can lead to severe side effects [5,6]. There are also a number of systemic agents, such as docetaxel (DTXL), that can improve therapeutic efficacy but can lead to hematologic toxicity and shows interpatient clearance variability of more than 10-fold [7–13]. We hypothesized that we can improve prostate brachytherapy’s efficacy and decrease its toxicity by incorporating local drug delivery in spacers.

Current brachytherapy spacers are made from biodegradable polymers such as poly(lactic-co-glycolic acid) (PLGA), and can be utilized for drug delivery. The challenge in incorporating therapeutics into spacers has been to incorporate drugs without changing spacer integrity and properties, and to control drug release so that the therapeutics can improve efficacy and reduce toxicity. With advances in drug delivery science and 3D printing, we theorized that we can address these challenges using 3D printing. We hypothesized that we can control drug release by engineering spacers with unique surfaces/areas. Herein, we report the formulation of drug-loaded brachytherapy spacers using continuous liquid interface production (CLIP). CLIP is employed to generate complex spacer surfaces. Using dexamethasone palmitate (DMP) and DTXL as model drugs, we printed brachytherapy spacers and examined the relationship between drug release and spacer surface geometry. Using a murine xenograft model of prostate cancer, we also studied the efficacy of drug-loaded spacers in brachytherapy.

2. Materials and methods

2.1. Materials

HEMA, PEG₅₅₀-DMA, BLS, and LTPO were purchased from Sigma Aldrich. DTXL was purchased from LC Laboratories. DM was purchased from Fisher Scientific. DMP was purchased from Toronto Research Chemicals. Microcentrifuge filter units were purchased from Corning. Gibco brand cell culture reagents were purchased from Fisher Scientific. Calcein AM was purchased from Fisher Scientific, #C1430. Propidium iodide was purchased from Fisher Scientific, #ICN19545825. Standard Imaging (Middleton, WI, USA) Seed Slider, Theragenics Corp. (Buford, GA, USA) Bio Spacer 910 spacers, and Bard (Covington, GA, USA) BrachySource I-125 seeds were obtained via intra-institutional transfer from UNC Hospitals. All other chemicals were acquired from Sigma Aldrich and used without further purification.

2.2. Cell culture

PC3 human squamous cell carcinoma cells were obtained from the University of North Carolina Lineberger Cancer Center Tissue Culture Facility, which had previously obtained the cells from the American Type Culture Collection (ATCC^® CRL-1435^™). Cells were cultured in a complete growth medium made of F-12K supplemented with 10% (v v⁻¹) fetal bovine serum and 1% (v v⁻¹) penicillin/streptomycin.

2.3. Animal maintenance

6–8 week old male athymic nude mice weighing 20–30 g were supplied by the University of North Carolina animal facility and maintained under pathogen-free conditions in the Center for Experimental Animals (an AAALAC accredited experimental animal facility). The animal use protocol (20–028.0) was approved by the University of North Carolina Institutional Animal Care and Use Committee and conformed to the Guide for the Care and Use of Laboratory Animals [14].

2.4. Device fabrication

Brachytherapy spacers were designed in Autodesk Fusion 360 (Autodesk, San Rafael, CA). We chose to use a resin made from hydroxyethylmethacrylate (HEMA) and polyethylene glycol dimethacrylate M_n 550 (PEG₅₅₀-DMA) as both components are commonly used, biocompatible, and crosslinkable; and this polymer combination has shown previous success with drug loading and good CLIP 3D printing resolution. The resin was also formulated with 0.5 wt% 2-tert-Butyl-6-(5-chloro-2H-benzotriazol-2-yl)-4-methylphenol (BLS) added as a UV absorber and 2.5 wt% ethyl (2,4,6-trimethylbenzoyl) phenylphosphinate (LTPO) as a photoinitiator/crosslinker [15]. Drug-loaded spacers included 10 wt% DTXL in the resin, along with 1.4 wt% DMP for dual-loaded spacers. DMP was used as a DM prodrug as it is more hydrophobic, allowing it to more readily dissolve into the resin. Devices were printed on a Carbon S1 printer at a speed of 10 mm hr⁻¹ with a light intensity of 10 mW cm⁻¹. Spacers were printed vertically along their long axis in a 4 × 6 array layout on each print. After printing, devices were first rinsed in deionized water for 5 s, followed by a brief 2 s rinse with acetone, and finally dried under a gentle stream of air until all acetone had evaporated. Spacers were then placed in an LED UV oven (365 nm, 90 mW cm⁻¹) for post-curing set at 50% for 30 s. Longer exposure times were not used as DM is UV labile [16,17]. This cured and crosslinked a majority of the remaining polymer throughout the printed network.

2.5. In vitro release

Device drug release studies were performed with spacers first being weighed and then placed into a microcentrifuge filter unit with no filter which was then placed in a floating rack in 4 L of pH 7.4 PBS which was replaced every 2 to 3 days, and kept under constant stirring at 37 °C. Spacers (n=3) were removed at each time point with the drug extracted in a 1 mL solution of 1:1 acetonitrile:2-propanol (ACN:IPA) for 3 days in a shaker at 150 rpm. Spacers also underwent a second extraction to confirm complete initial extraction. The concentration of drug in extracted solutions was determined using a Shimadzu SPD-M20A high performance liquid chromatography (HPLC) equipped with a diode array detector and a Chromolith Fast Gradient RP-18e 50 × 2mm column (EMD Millipore, Burlington, MA). Samples (10 μL) were injected into the HPLC and eluted using a binary solvent system (phase A and B, A for Water, B for ACN) at a flow rate of 0.25 mL min⁻¹. The linear gradient program was set as follows: 0 to 20 min, 0% to 100% B; 20 to 25 min, 100% B; 25 to 30 min, 50% B; 30 to 35 min, 0% B. The column and sample temperature were maintained at 30 °C and 4 °C respectively. DTXL (retention time 11.2 min) and DMP (retention time 20.3 min) were monitored at 200 and 239 nm, respectively. Concentrations of the extracted drugs were calculated by comparing peak integrations to standard curves made with known concentrations for each run.

2.6. In vitro cytotoxicity

3D tumor models were created and imaged as previously described [18]. To assess in vitro toxicity of DTXL loaded spacer, PC3 cells were cultured as previously described and suspended in fresh medium at a concentration of 5×10⁵ cells mL⁻¹. Cell suspensions were mixed 1:1 with PureCol Ez Col collagen solution (Advanced Biomatrix, Carlsbad, CA) to achieve a final concentration of 2.5×10⁵ cells mL⁻¹. 1.5 mL of this cell/collagen mixture was added to a 35 mm dish with a 14 mm coverslip optical glass bottom (MatTek, Ashland, MA). A spacer was placed into the center of the dish, submerged in the solution. Dishes were then placed in an incubator at 37 °C. Gels were washed every 24 hours with 300 μL PBS which was aspirated and then 300 μL of fresh media was added to prevent gels from drying out. Cells were stained after 4 days by adding 3 μL of Calcein Am (4 μM, Invitrogen, Carlsbad, CA) and 200 μL of propidium iodide (1 mg ml⁻¹, MP Biomedicals, Santa Ana, CA) and incubated for 30 min. Gels were then washed 2x with 300 μL of PBS and imaged on an LSM 700 confocal microscope (Zeiss, White Plains, NY). Tiled z-stack images were obtained, vertically covering the diameter of the spacer, with tiles stitched in Zen imaging software before exporting. Z-stacks were trimmed in ImageJ (NIH, Bethesda, MA) to remove slices above or below the spacer, then maximum intensity projections were obtained. Particle analysis was run using 370×175 pixel windows corresponding to one third the length and the full diameter of the spacer. Sampling windows were moved away from the spacer edge to obtain live vs dead cell counts in both non-drug-loaded and drug-loaded spacers to determine cell death at varying distances with a half size window used for the closest 0.4 mm measurement.

2.7. In vivo efficacy

Mice were injected subcutaneously with 1×10⁶ PC3 cells in 100 μL of 50:50 matrigel:plain medium on the left flank (Fig. 4a). Tumors were allowed to grow until they were 80–150 mm³ (21 days) and then mice were divided into six treatment groups: PBS IV only, PBS IV + seed, DTXL IV + seed, plain spacer + seed, DTXL spacer + seed, DTXL/DMP spacer + seed. All mice except those in the PBS IV only group received a Bard I-125 brachytherapy seed implanted subcutaneously, adjacent to the tumor using an 18 gauge Brachystar Seed Implant Needle (Bard Medical Division, Covington, GA) with plunger. Groups with spacers had them implanted intratumorally using the same type Brachystar needle with plunger. IV DTXL dosing was based on a human clinical equivalent dose of 75 mg m⁻¹ [12], converted using a normal man equivalency of 70 kg and 1.8 m² [19]. This resulted in a 2 mg/kg dose in mice given via tail vein injections of 100 μL drug solution in PBS. Mice were monitored every two to three days with tumors having their length and width measured to calculate the tumor volume as $Volume=\frac{length\times {width}^2}{2}$.

Fig. 4.

(a) Treatment scheme for the evaluation of drug-loaded brachytherapy spacers. Day 0 – Tumor inoculation through subcutaneous injection of 1×10⁶ PC3 cells in 1:1 matrigel:medium. Day 21 – Mice with tumors of 80–150 mm³ have a radioactive I-125 brachytherapy seed implanted adjacent to the tumor, except for one PBS control group. Mice are also given either an IV injection (PBS or DTXL) or receive a spacer implanted intratumorally (non-drug-loaded, DTXL-loaded, or DTXL/DMP-loaded). > Day 21 – Drug locally diffuses out from spacers to radiosensitize and treat tumors in conjunction with radioactive I-125 brachytherapy seeds. (made in ©BioRender – biorender.com) (b) Average tumor volume post inoculation with each group ending once any mouse from that group reaches a study endpoint. Both drug-loaded spacer groups showed significant tumor growth inhibition compared to IV treatment and non-drug-loaded spacers. (c) Survival curve. Both drug-loaded spacer groups showed significant survival benefits compared to IV treatment and non-drug-loaded spacers, with 78% and 88% cure rates in the DTXL and DTXL/DM spacer groups, respectively. (* p < .05, ** p < .01, *** p < .001, **** p < .0001)

2.8. In vivo hematological toxicity, hepatotoxicity, and nephrotoxicity assay

In vivo toxicity was determined using blood collected via submandibular bleed. Blood was collected 1 day after treatment (n=3). For hematological toxicity, whole-blood (>35 μL) was stored in a heparin pre-treated stopper covered tube at 4 °C and analyzed for white and red blood cell counts. For hepatotoxicity and nephrotoxicity, whole-blood (> 50 μL) was centrifuged at 2,500 g for 10 min with plasma then pipetted off the top into an Eppendorf tube. The whole-blood and the isolated plasma were analyzed by the Department of Pathology & Laboratory Medicine, Animal Histopathology & Laboratory Medicine Core, University of North Carolina, for blood cell counts, AST, BUN, and Crea levels.

2.9. Statistical analysis

All experiments were performed at least three times and expressed as mean±SD for in vitro or mean±SEM for in vivo studies. Statistical significance for in vitro drug release was determined using two-way ANOVA with variables of time and drug released, in vivo growth comparisons were determined using area under the curve and unpaired t-tests [20], and Kaplan-Meier survival curves used log-rank tests. All tests were performed in Graphpad Prism 8 (GraphPad Software, San Diego, CA). Release studies fit lines were modeled as non-linear regressions using two phase exponential association and the formula $y=y_{{max}_1}*\left(1-e^{-k_{1}t}\right)+y_{{max}_2}*(1-e^{-k_{2}t})$. Spacer cytotoxicity fit lines were modeled as non-linear regressions using an inhibitor vs response four parameter model with the formula $y=y_{min}+\frac{(y_{max}-y_{min})}{(1+{\left(I{C}_{50}÷x\right)}^{Hillslope})}$. Differences were considered significant when p < .05 and significance level in all figures was represented with * p < .05, ** p < .01, *** p < .001, **** p < .0001.

3. Results

To ensure compatibility with existing equipment, spacers were designed with the same dimensions as existing spacers: length (5 mm) and diameter (0.8 mm) (Fig. 1b). To enable different rates of drug release, we engineered spacers with different surface geometries/areas. To incorporate DTXL and DMP, both therapeutics were dissolved in a resin that is comprised of HEMA and PEG₅₅₀-DMA [15]. The resin was supplemented with a UV absorber BLS and photoinitiator LTPO to encourage cross linking during printing. As the selected resin was hydrophobic, the hydrophobic DM prodrug DMP was used instead of DM. DMP can be hydrolyzed to the active form of DM by esterases in vivo [21], which are often overexpressed in cancer cells [22], including in prostate cancer [23]. Both DTXL and DMP readily dissolved into the resin, with DTXL having a maximum solubility of 20 wt%. We chose a final DTXL concentration of 10 wt% to avoid saturation concerns and allow a window for future loading modifications as dictated by therapeutic results. DMP loading was selected at 1.4 wt% to maintain a molar ratio comparable to other DTXL/DM combination treatments [7,12].

Fig. 1.

(a) Bio Spacer 910 standard commercial spacer. (b) 3D computer renderings of brachytherapy spacer designs – cylinder, star, and twist left to right. (c) SEM images (50x) of CLIP printed brachytherapy spacer side profiles. (d) SEM images (250x) of CLIP printed brachytherapy spacer cross section profiles.

3.1. Drug release profile can be controlled using spacer design

To engineer drug-loaded spacers with the same dimensions but different surface areas and volumes, we utilized circular and star shaped cross-section profiles for the spacers. The star shaped profile resulted in a 47% reduction in computed model volume vs the circular profile (Fig. 1). This design, however, provided a 22% larger surface area, increasing the surface area to volume ratio 2.3-fold. To further vary the surface area without changing the cross-section profile, we designed a third spacer utilizing the star profile but undergoing two full rotations over its 5 mm length (termed twist spacer). The twist spacer has a similar volume to the star spacer, but its surface area is 40% more than that of the standard cylinder spacer, resulting in a 2.6-fold increase in the surface area to volume ratio. After printing, relative volumes of spacers were verified by weight (Fig. S1), with the star and twist spacers yielding 51±3% and 46±4% reductions respectively, confirming fabrication accuracy. Spacers were also test loaded into brachytherapy implant needles using a commercial seed slider (Fig. S2).

To characterize drug release from each type of spacer, in vitro release studies were performed in PBS at 37 °C. Using DTXL as the model drug, we found the twist spacer had a t_1/2 = 26 days, the star spacer had a t_1/2 = 29 days, and the cylinder spacer had only released 42% by day 41 (Fig. 2). Since early release is more important for mitigating side effects and improving efficacy (radiation dose is higher in early days), the twist design was chosen for further in vitro and in vivo studies. Drug elution time variability on HPLC was less than 1% between pure drugs and drugs extracted from 3D printed spacers, indicating no significant degredation of the drugs due to the CLIP printing process.

Fig. 2.

In vitro drug release in PBS at 37 °C over 41 days. Star and twist spacers show t_1/2 = 29 and t_1/2 = 26 days respectively. Cylinder spacers only released 42% of loaded drug over 41 days. All fit lines were calculated by non-linear regression as two phase exponential associations. (* p < .05)

To further study the release characteristics of 3D printed spacers and how design translates to tumor cytotoxicity, we examined the cylinder and twist spacers using a bioengineered tumor mimic [18,24]. This 3D tumor model provides an environment where tumor cells fully surround the spacers to better simulate the 3D effects of local drug release. Control arms included bioengineered tumors with a non-drug-loaded cylinder spacer and tumors with no spacer at all (Fig. 3b & c). Both DTXL-loaded cylinder spacers and DTXL-loaded twist spacers were studied in this bioengineered tumor model (Fig. 3d & e). We found a baseline level of cell death from control arms which slightly increased adjacent to the non-drug-loaded spacer, indicative of mild cytotoxicity from resin (Fig. 3b). Both of the DTXL-loaded spacers exhibited significant cytotoxicity to tumor cells around the spacers (Fig. 3d & e). These were analyzed by live/dead cell counts using ImageJ. Cell death percentages were plotted against distance from the spacer, yielding a distance response curve, analogous to a dose response curve. The DTXL-loaded twist spacers had a right-shifted response curve, indicative of increased drug release and cell death, with a 50% increased cell death radius 2.3 mm from the spacer, compared to 1.5 mm for the DTXL-cylinder spacer. These data demonstrate that we are able to control drug release and treatment effects by varying the surface to volume ratio of the spacer.

Fig. 3.

Comparison of non-drug-loaded spacers and DTXL-loaded spacers in 3D bioengineered tumor mimics. Diffusion tests demonstrated significantly increased cell death surrounding the drug-loaded spacers compared to the non-drug-loaded spacers. Live cells were stained with Calcein AM and dead cells with propidium iodide. Image stacks were taken with an LSM 700 confocal microscope. (a) Spacer cytotoxicity graph showing cell death as the y-axis and distance from spacer (center of dish in case of no spacer control) as the x-axis. The DTXL-loaded spacer curves indicate a dose response relationship based on distance from DTXL-loaded spacers, with minimal effect from the non-drug-loaded spacers. The DTXL-loaded twist spacer also shows an improved distance-response relationship over the drug-loaded cylinder spacer, indicating an increased drug release rate. (b-e) Live/dead cell overlay images of (b) no spacer, (c) non-drug-loaded cylinder spacer, (d) DTXL-loaded cylinder spacer, and (e) DTXL-loaded twist spacer, with autofluoresecent spacers in the center. The DTXL-loaded cylinder spacer has a recognizable increase in cell death in the region surrounding it, with a similar but larger area surrounding the DTXL-loaded twist, indicative of its increased drug release rate.

3.2. Drug-loaded spacers have minimal systemic toxicity

While our brachytherapy spacers are designed for local drug delivery, some of the therapeutics may enter systemic circulation. To evaluate potential systemic toxicity from these brachytherapy spacers and common dose-limiting toxicities of chemotherapeutics, we examined hepatotoxicity, nephrotoxicity, and hematotoxicity through serum enzyme levels of aspartate aminotransferase (AST), blood urea nitrogen (BUN), and creatinine (Crea) along with CBC counts (Table S1). Enzyme levels and CBC counts from both IV and spacer treated mice remained within similar ranges, with no significant deviations from normal values, indicating no additional toxicities from drug-loaded spacers.

3.3. Drug-loaded spacers improve in vivo efficacy

To examine the potential of using drug-loaded spacers to improve brachytherapy efficacy in vivo, we used a PC3 prostate cancer xenograft murine model. To best simulate the clinical use of spacers in brachytherapy, all mice, except for the PBS only control group, were implanted with an I-125 brachytherapy seed (activity of 0.31–0.34 mCi on day of implant) directly adjacent to the tumor. Mice were given either PBS IV only, PBS IV + seed, DTXL IV + seed, non-drug-loaded spacer + seed, DTXL spacer + seed, or DTXL/DMP spacer + seed. Mice that received spacers had a twist spacer inserted intratumorally (Fig. 4a). Both seeds and spacers were inserted using Bard Brachystar commercial brachytherapy 18-gauge implant needles, demonstrating spacer compatibility with current equipment. At 23 days post implant, mice with drug-loaded spacers were noted to have mild discoloration and loss of hair in an area extending radially from the tumor and implant (Fig. S3). This was only present in mice with drug-loaded spacers. All groups with brachytherapy seeds showed significantly delayed tumor growth over the control group without brachytherapy seeds, showing the significant benefits of brachytherapy radiation alone (Fig. 4b, S4). The non-drug-loaded spacer + seed and the DTXL IV + seed groups also outperformed the PBS + seed group, but showed no significant difference versus each other (p = .0003 and p = .0004 respectively). This revealed that IV administered DTXL further inhibited growth and that the intratumoral non-drug-loaded twist spacers also exhibited a growth inhibition likely due to a combination of the physical insertion of the intratumoral spacer and the mild cytotoxicity of the spacer as previously noted in vitro. Groups with seeds and drug-loaded spacers had the greatest tumor inhibition and showed significantly improved efficacy over all other groups (p < .0001). There was no significant difference in growth delay between the DTXL only spacer and the DTXL/DMP spacer. Survival rates varied from 11–30% in groups that received a radioactive seed without a drug-loaded spacer but was significantly higher with DTXL-loaded spacers, demonstrating a 78% cure rate, and DTXL/DMP-loaded spacers, demonstrating an 88% cure rate (Fig. 4c). Median survival time of the PBS only group was the shortest at 82 days, followed by DTXL IV + seed, PBS IV + seed, and non-drug-loaded spacer + seed at 117, 138, and 166 days, respectively. Median survival of both drug-loaded spacer + seed groups was not reached after 315 days. These results demonstrate that drug-loaded spacers are able to enhance standard brachytherapy to inhibit tumor growth and improve survival.

4. Conclusions

Advances in 3D printing offer exciting new opportunities in medical device development. However, existing efforts have been largely focused on tissue engineering. One advantage of 3D printing over traditional manufacturing technologies is the ability to engineer intricate shapes with complex chemistry. Our work on drug-loaded spacers demonstrates this potential. We have shown that CLIP can be utilized to formulate drug-loaded spacers with complex geometries and varying surface areas. We further demonstrated that complex geometry is important in controlling drug release. Lastly, we demonstrated that drug-loaded spacers can improve prostate cancer brachytherapy treatment. We believe this work will facilitate further investigations in applying 3D printing to medical applications.

Abbreviations

AST

aminotransferase

BLS

2-tert-Butyl-6-(5-chloro-2H-benzotriazol-2-yl)-4-methylphenol

BUN

blood urea nitrogen

Crea

creatinine

DTXL

docetaxel

DM

dexamethasone

DMP

dexamethasone palmitate

HEMA

hydroxyethlmethacrylate

I-125

iodine-125

LTPO

ethyl (2,4,6-trimethylbenzoyl) phenylphosphinate

PEG₅₅₀-DMA

polyethylene glycol dimethacrylate M_n 550

RT

radiation therapy

Data Availability

The main data supporting the results in this study are available within the paper and its supplementary Information. The raw and analyzed datasets generated during the study are available for research purposes from the corresponding authors on reasonable request.