2. Materials and Methods
2.1. Materials
Titanium Oxide nanopowder (TiO2, Anatase, 99.5%, 5nm), Stock#: US3838, TiO2 CAS#: 13463-67-7, Net weight: 100g) was acquired from Sigma-Aldrich (St. Louis, MO, USA). Omniscan gadodiamide gadolinium-based nanoparticles were obtained from GE Healthcare (USA). All cell culture products (Dimethyl sulfoxide (DMSO), DMEM, RPMI, Trypsin, Fetal Bovine Serum, MEM non-essential amino acids, sodium pyruvate, β-mercaptoethanol, penicillin/streptomycin, and PBS pH 7.4) were purchased from either the American Type Culture Collection (USA), ThermoFisher Scientific (USA), or Life Tech-nologies (Waltham, MA, USA). The monoclonal antibody anti-mouse CD40 (FGK4.5/FGK45) was bought from BioXcell (New Hampshire, USA).
2.2. Liquid Immunogenic Fiducial Eluter (LIFE) Biomaterial Assembly
LIFE Biomaterials were fabricated following a previously published protocol [
23,
25]. In this study, LIFE biomaterials consisted of a mixture of two natural polymers: 4% (w/v) sodium alginate and 2% (w/v) chitosan, mixed in a 1:1 ratio. The polymer mixture was then mixed in a 1:1/2 ratio with a CT contrast nanoparticle, titanium dioxide (TiO2), combined with an MRI contrast gadolinium-based nanoparticle (Omniscan) in a 1:1 ratio. The LIFE Biomaterial was kept at 4⁰C until treatment. On treatment day, LIFE biomaterial was aliquoted based on the number of mice being treated, and 100µg per mouse of anti-CD40 antibody was added to the LIFE Biomaterial.
2.3. Cell Culture Preparations for Mice Inoculation
The pancreatic cancer (KPC) cell line was derived from an LS-Kras; p53+/floxed, Pdx-cre mouse. KPC (C57BL6 genetic background) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) medium supplemented with 10% Fetal Clone II FBS and 1% penicillin/streptomycin. All cells were cultured at 37°C in a humidified incubator with 5% CO2. Immunocompetent wild-type C57BL/6 strain male and female mice (with average body weight = 22 g) were acquired from Taconic Biosciences, Inc., or Charles River at 6 - 8 weeks old and were inoculated subcutaneously with 1.5 x 105 KPC cells procured in a 100μL volume of cells per flank. Following cancer cell injection, the tumors were allowed to grow for over 2 weeks singly to at least 3.0mm or greater in diameter before the start of treatment. Mice's body weights were not measured throughout the duration of the studies. Animal experiments followed the guidelines and regulations set by the Johns Hopkins University Animal Care and Use Committee (ACUC) under protocols# MO21M281 and MO24M298. Mice maintenance in the Johns Hopkins University animal facility was according to the Institutional Animal Care and Use Committee-approved guidelines, consisting of stable environmental conditions, including a 12–14-hour light/dark cycle and temperatures between 65-75°F (18-23°C) with 40-60% humidity. Mice tumor volume was determined using the formula: V = 0.5 x length x (width2). A Vernier caliper was used to measure the longitudinal and latitudinal outliers of the tumor, designated as the tumor length and the latitude outlier as the tumor width.
2.4. Data Collection and Image Analysis
To assess the feasibility of LIFE Biomaterial in providing image guidance during radiotherapy, contrast imaging was performed on a human cadaver. The human cadaver study was conducted in accordance with the Johns Hopkins Institutional Review Board (IRB), and the protocol was approved by the Institutional Review Board with Approval Code: IRB-X in January 2018. Frozen, unfixed cadaveric specimens were thawed and injected with LIFE Biomaterial, resulting in a CT simulation (TOSHIBA Helical CT scan with 2 mm slice thickness, 120 kVp, and X-ray tube current of 100 mA) with the cadaveric specimen in the supine position. A repeated scan of the specimens was taken. For MRI, a Philips Achieva 3.0 T MRI System with BODY Transmit Coil was utilized with a repetition time of 5.31 ms, a flip angle of 100, a percent phase field of view of 70.833, and a slice thickness of 0.9 mm.
2.5. Small Animal FLASH Radiation Research Platform to Deliver Conventional and Flash Radiotherapy
2.5.1. FLASH-SARRP
The FLASH-SARRP system was employed to deliver both CONV and ultrahigh dose-rate irradiation using kilovoltage (kV) x-rays. Details of the system’s components and dosimetry performance have been described previously [
26]. A brief overview is provided here: The system incorporates two high-capacity rotating-anode x-ray tubes (RAD44, Varex Imaging Inc., Salt Lake City, UT), each powered by a 100-kW generator (CPI International Inc., Palo Alto, CA), to deliver ultrahigh dose rate x-rays at short source-to-surface distances (SSDs). Each x-ray source can operate at tube potentials up to 150 kVp, with a broad range of tube currents (5 - 630 mA) and exposure times (1 to 6300 ms), allowing for precise selection of dose and dose-rate in the range of below 1 Gy/s to over 100 Gy/s, depending on the field size and SSD. The system is integrated with a robotic animal stage capable of high-precision motion in the x, y, and z directions, allowing accurate and reproducible positioning of the mouse bed for the tumor irradiation.
2.5.2. Animal Setup and Irradiation
A 2.1 mm thick tungsten collimator with a 10 mm diameter circular aperture was used to shape the radiation field. The collimator was mounted to the x-ray tube flange using a custom 3D-printed holder. To ensure high-precision and reproducible mouse positioning, a custom immobilization and docking device was developed and referenced to the mounting clamps on the robotic stage. The mouse was anesthetized using a mobile anesthesia system (Fisher anesthesia machine) delivering 2% isoflurane in the medical air via housing connected to the nose cone. The mouse's front and hind legs were immobilized in fixed locations on the bed, using specially designed clamps, to maintain consistent positioning. The mice were positioned at an SSD of 63 mm. The tumor was aligned under the 10 mm collimator aperture visually and using positioning lasers. Tumor positioning at the center of the field size was verified using beam’s eye view images on radiographic films. A thin metal wire was placed around the tumor, and the immobilized mouse on the robotic bed was adjusted to bring the tumor with wire outline into the center of the irradiation field. A piece of LD-V1 radiographic film was placed underneath the mouse, and a low-energy imaging exposure (42 kVp, 2 mAs) was delivered. The resulting image was used to verify that the wire ring completely fit within the irradiation field, confirming field localization. An illustration of the FLASH-SARRP system is displayed in
Figure 1.
Radiation dosimetry was performed using EBT3 Gafchromic film (Ashland Inc., Wayne, NJ, USA) within a solid water phantom composed of 1 mm and 5 mm thick slabs. EBT3 film sheets were laser-cut and placed from 1 to 5 mm depths within the phantom, which was positioned on the robotic stage to mimic the geometry used during mouse irradiation experiments. Percentage depth dose-rate (PDDR), dose-rate profiles, and 2D dose-rate distribution were measured to characterize the dose delivered to the tumor and surrounding tissues. Film calibration was performed using a NIST-traceable 120 kVp x-ray source at the University of Wisconsin Accredited Dosimetry Calibration. A dose rate of 61.5 ±3.5 Gy/s was delivered for FLASH-RT and 0.95±0.05 Gy/s for CONV RT in this setup.
2.5.3. Experimental Study Design
Mice were used pre-clinically due to their high degree of genetic homology with humans, with nearly 95% of genes in common [
27]. Here, three individual studies were conducted to: 1) identify the optimal RT dose for either Flash or CONV RT; 2) leverage the immune response using high dose Flash or CONV RT between 10Gy and 20Gy combined with LIFE biomaterial_anti-CD40 treatment; 3) influence the immune response using low dose Flash or CONV RT between 5Gy and 8Gy combined with LIFE biomaterial_anti-CD40 treatment. In total, 106 C57BL6 male and female mice with an average body weight of 22 g were utilized throughout the course of this study. The sample size for each experiment performed was chosen based on the availability of mice that were housed in our animal facility. There were no explicit criteria set for including experimental units unless the animals reached study endpoints such as tumor size reaching 2cm, bleeding tumors, or concave ulcerated tumors, hunched behavior, or low body score <2.5, as described in the Johns Hopkins University Animal Care and Use Committee (ACUC) under protocol# MO24M298. Mice ulcers were treated daily with a Veterinus Derma Gel to help mitigate the ulcerated wound. No analgesics were used throughout the duration of these studies. Mice were simply randomized according to tumor size to ensure similar tumor sizes are represented in all the cohorts. Potential confounders were minimized by using mice of the same strain and age, housed in the same mouse holding room and animal facility.
The anti-tumor effect of conventional versus flash photon RT was assessed in C57BL6 mice bearing bilateral subcutaneous tumors (size ~ 3mm), targeting only one tumor for treatment while observing both tumors to note any change in tumor growth post-treatment and prolonged mouse survival post-treatment, as illustrated in
Figure 2a. The experimental unit consists of n = 4 - 5 mice per cage. The cohorts (n = 8/group) were allocated as follows: no treatment, Flash_5Gy, CONV_5Gy, Flash_10Gy, CONV_10Gy, Flash_15Gy, and CONV_15Gy. The mouse was set up in the FLASH-SARRP system as illustrated in
Figure 2b, FLASH and CONV irradiation of the targeted subcutaneous tumor. 56 C57BL6 male mice bearing bilateral flank tumors were utilized for this analysis.
Next, a multiplex immunofluorescence analysis was performed to assess the immune cells infiltration (CD45, CD3, CD4 and CD8) within the tumor microenvironment at 34 days post-treatment for the following cohorts: No Treatment (n = 5); LIFE Biomaterial_20µg-Anti-CD40 (n = 4); Flash_10Gy_LIFE Biomaterial_20µg-Anti-CD40 (n = 4); Conv_10Gy_LIFE Biomaterial_20µg-Anti-CD40 (n = 4); Flash_20Gy_LIFE Biomaterial_20µg-Anti-CD40 (n = 4); Conv_20Gy_LIFE Biomaterial_20µg-Anti-CD40 (n = 4). 25 C57BL6 female mice bearing bilateral flank tumors were utilized for this analysis. The experimental unit consists of n = 4 - 5 mice per cage.
Subsequently, 25 C57BL6 female mice bearing a subcutaneous pancreatic right flank tumor were utilized to perform immunohistochemistry analysis on tumor tissues collected 10 days post-treatment to assess immune cell infiltration in the tumor microenvironment. The experimental unit consists of n = 2 - 3 mice per cage. The assessed cohorts were: a) No treatment (n = 3); b) LIFE Biomaterial_100µg-Anti-CD40 (n = 2); c) Flash_5Gy (n = 3); d) Conv_5Gy (n = 3); e) Flash_5Gy_LIFE Biomaterial_100µg-Anti-CD40 (n = 2); f) Conv_5Gy_LIFE Biomaterial_100µg-Anti-CD40 (n = 2); g) Flash_8Gy (n = 3); h) Conv_8Gy (n = 3); i) Flash_8Gy_LIFE Biomaterial_100µg-Anti-CD40 (n = 2); j) Conv_8Gy_LIFE Biomaterial_100µg-Anti-CD40 (n = 2).
2.6. Histology Staining
The Johns Hopkins Oncology Tissue and Imaging Service (OTIS) Core's histology technicians procure H&E and Unstained sections using a microtome. This high-precision instrument enables technicians to cut 4-micrometer-thick FFPE (formalin-fixed paraffin-embedded) tissue sections that are then suspended on a 41-degree Celsius warm water bath. Using specialized microscope slides, the technician retrieves the sections from the water bath subsequently mounting them onto the microscope sides for further processing, where the designated H&E sectioned slides are baked at 60-degrees Celsius for 30 minutes before auto staining for hematoxylin and eosin, and the designated Unstained sectioned slides are set aside—upright—on a specialized drip trays to air dry at room temperature for 48-72 hours.”
2.7. Immunohistochemistry Staining
Immunostaining was performed at the Oncology Tissue and Imaging Services Core of Johns Hopkins University. Immunolabeling for CD3-CD11b dual detection was performed on formalin-fixed, paraffin-embedded sections on a Ventana Discovery Ultra Auto Stainer (Roche Diagnostics). Briefly, following dewaxing and rehydration on board, epitope retrieval was performed using Ventana Ultra CC1 buffer (catalog #6414575001, Roche Diagnostics) at 96°C for 64 minutes. Primary antibody, anti-CD3 (1:200 dilution; catalog #ab16669, Abcam) was applied at 36°C for 60 minutes. CD3 primary antibodies were detected using an anti-rabbit NP and anti-NP AP detection system (catalog numbers #07425317001 and #07425325001, Roche Diagnostics), followed by the Discovery Yellow Detection Kit (Catalog #07698445001, Roche Diagnostics). Following CD3 detection, primary and secondary antibodies from the first round of staining were stripped from the board using Ventana Ultra CC1 buffer at 93°C for 8 minutes. Primary antibody, anti-CD11b (1:10,000 dilution; catalog #ab133357, Abcam) was applied at 36 °C for 60 minutes. CD11b primary antibodies were detected using an anti-rabbit HQ detection system (catalog #7017936001 and #7017812001, Roche Diagnostics), followed by the Discovery Teal Detection kit (catalog #8254338001, Roche Diagnostics), counterstaining with Mayer’s hematoxylin, dehydration, and mounting. Immunolabeling for CD8-CD4 dual detection was per-formed on formalin-fixed, paraffin-embedded sections on a Ventana Discovery Ultra Auto Stainer (Roche Diagnostics). Briefly, following dewaxing and rehydration on board, epitope retrieval was performed using Ventana Ultra CC1 buffer (catalog #6414575001, Roche Diagnostics) at 96°C for 64 minutes. Primary antibody, anti-CD8 (1:125 dilution; catalog #14-0195-82, ThermoFisher Scientific) was applied at 36 °C for 60 minutes, followed by rabbit anti-rat linker antibody (1:500 dilution; catalog #AI4001, Vector Labs) at 36°C for 32 minutes. Linker antibodies were detected using an anti-rabbit NP and anti-NP AP detection system (catalog numbers #07425317001 and #07425325001, Roche Diagnostics) followed by the Discovery Yellow Detection Kit (Catalog #07698445001, Roche Diagnostics). Following CD8 detection, primary and secondary antibodies from the first round of staining were stripped on the board using Ventana Ultra CC1 buffer at 93°C for 8 minutes. Primary antibody, anti-CD4 (1:1000 dilution; catalog# ab183685, Abcam) was applied at 36°C for 60 minutes. CD4 primary antibodies were detected using an anti-rabbit HQ detection system (catalog #7017936001 and #7017812001, Roche Diagnostics) followed by the Discovery Teal Detection kit (catalog #8254338001, Roche Diagnostics), counterstaining with Mayer’s hematoxylin, dehydration and mounting.
2.8. Multiplex Immunofluorescence
Immunostaining was performed at the Oncology Tissue Services Core of Johns Hopkins University School of Medicine. Quadruple immunolabeling for CD8+CD45+CD4+CD3 was performed on formalin-fixed, paraffin-embedded sections on a Ventana Discovery Ultra auto stainer (Roche Diagnostics). Following dewaxing and rehydration on board, epitope retrieval was performed using Ventana Ultra CC1 buffer (catalog #6414575001, Roche Diagnostics) at 96°C for 64 minutes. Primary antibody, anti-CD8 (1:125 dilution; catalog# 14-0808, eBioscience, Inc., San Diego, CA, USA) was applied at 36 °C for 40 minutes, followed by rabbit anti-rat linker antibody (1:500 dilution; catalog# AI4001, Vector Labs, Newark, CA, USA) at 36°C for 32 minutes. Linker antibodies were detected using an anti-rabbit HQ detection system (catalog# 7017936001 and 7017812001, Roche Diagnostics), followed by OPAL 520 (NEL871001KT, Akoya Biosciences) diluted 1:150 in 1X Plus Amplification Diluent (catalog # FP1498, Akoya Biosciences). Following CD8 detection, primary and secondary antibodies from the first staining round were stripped on board using Ventana Ultra CC1 buffer at 95°C for 12 minutes and neutralization using Discovery Inhibitor (catalog #7017944001, Roche Diagnostics). Primary antibody, anti-CD45 (1:200 dilution; catalog #702575S, Cell Signaling Technology) was applied at 36 °C for 40 minutes. CD45 primary antibodies were detected using an anti-rabbit HQ detection system (catalog# 7017936001 and 7017812001, Roche Diagnostics, Basel, Switzerland) followed by OPAL 570 (NEL871001KT, Akoya Biosciences, Marlborough, MA, USA) diluted 1:150 in 1X Plus Amplification Diluent (catalog # FP1498, Akoya Biosciences). Immunofluorescence stains were analyzed using Zeiss ZEN Lite software version 3.10. Images were first processed by performing a background subtraction and then an enhanced contour. The resulting images were then analyzed by evaluating the histograms of the images and their corresponding wavelength values. Arithmetic Mean Intensity Values and Sum of all pixel intensities from each channel were extracted. Averages were then calculated for the CD8 marker.
2.9. Statistical Analysis
GraphPad Prism v9 was used to generate Kaplan–Meier statistics (Madsen, 1986; Statistical Concepts, Prentice Hall, Englewood Cliffs, NJ, USA) to determine the p-value at *p < 0.05 for the survival curves and/or the sum fluorescence intensity comparing the immune cells infiltration within the treated cohorts compared to the controls. Non-significant data was denoted as n.s.