- Email: [email protected]
Pulsed Radiation Therapy With Concurrent Cisplatin Results in Superior Tumor Growth Delay in a Head and Neck Squamous Cell Carcinoma Murine Model
Accepted Manuscript Pulsed Radiotherapy with Concurrent Cisplatin results in Superior Tumor Growth Delay in a Head and Neck Squamous Cell Carcinoma Murine Model Kurt Meyer, MD, Sarah. A. Krueger, PhD, Jonathan L. Kane, BS, Thomas G. Wilson, BS, Alaa Hanna, MD, Mohamad Dabjan, MD, Katie M. Hege, BS, George D. Wilson, PhD, Inga Grills, MD, Brian Marples, PhD PII:
S0360-3016(16)30181-X
DOI:
10.1016/j.ijrobp.2016.04.031
Reference:
ROB 23575
To appear in:
International Journal of Radiation Oncology • Biology • Physics
Received Date: 17 February 2016 Revised Date:
13 April 2016
Accepted Date: 30 April 2016
Please cite this article as: Meyer K, Krueger SA, Kane JL, Wilson TG, Hanna A, Dabjan M, Hege KM, Wilson GD, Grills I, Marples B, Pulsed Radiotherapy with Concurrent Cisplatin results in Superior Tumor Growth Delay in a Head and Neck Squamous Cell Carcinoma Murine Model, International Journal of Radiation Oncology • Biology • Physics (2016), doi: 10.1016/j.ijrobp.2016.04.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Pulsed Radiotherapy with Concurrent Cisplatin results in Superior Tumor Growth Delay
RI PT
in a Head and Neck Squamous Cell Carcinoma Murine Model
Kurt Meyer MD, Sarah. A. Krueger PhD, Jonathan L. Kane BS, Thomas G. Wilson, BS,
SC
Alaa Hanna MD, Mohamad Dabjan MD, Katie M. Hege BS,
M AN U
George D. Wilson PhD, Inga Grills MD, Brian Marples PhD
TE D
Department of Radiation Oncology, William Beaumont Hospital, 3811 W. Thirteen Mile Rd, 105-RI, Royal Oak, Michigan 48073 Running title: Pulsed Radiotherapy and Cisplatin
AC C
Pages: 24 Figures: 5 Tables: 0
EP
Keywords: Radiation, UT-SCC-14, Chemoradiation, animal imaging
Address correspondence to: Brian Marples Ph.D., Department of Radiation Oncology, William Beaumont Hospital, 3811 W. Thirteen Mile Rd, 105-RI, Royal Oak, MI 48073. Tel: 248-551-0213; Fax: 248-551-2443; E-Mail: [email protected]
ACCEPTED MANUSCRIPT
Pulsed Radiotherapy with Concurrent Cisplatin results in Superior Tumor Growth Delay
M AN U
Running title: Pulsed Radiotherapy and Cisplatin
SC
RI PT
in a Head and Neck Squamous Cell Carcinoma Murine Model
AC C
EP
TE D
Keywords: Radiation, UT-SCC-14, Chemoradiation, animal imaging
Page 1 of 25
ACCEPTED MANUSCRIPT
CONFLICTS OF INTEREST NOTIFICATION
AC C
EP
TE D
M AN U
SC
RI PT
No conflicts of interest exist.
Page 2 of 25
ACCEPTED MANUSCRIPT
ABSTRACT
Purpose: To assess the efficacy of 3-week schedules of low-dose pulsed radiation treatment
squamous cell carcinoma (HNSCC) xenograft model.
RI PT
(PRT) and standard radiotherapy (SRT), with concurrent cisplatin (CDDP) in a head and neck
SC
Methods and Materials: Subcutaneous UT-SCC-14 tumors were established in athymic NIH III HO female mice. 30 Gy was administered as 2 Gy/day (d), 5d/wk for 3 wks either by PRT
M AN U
(10x0.2 Gy/day, with a 3-min break between each 0.2 Gy dose) or SRT (2 Gy/day, noninterrupted delivery) in combination with concurrent 2 mg/kg CDDP 3 times/wk in the final 2 weeks of RT. Treatment-induced growth delays were defined from twice-weekly tumor volume measurements. Tumor hypoxia was assessed by 18F-FMISO PET imaging, and calculated
TE D
SUVmax values compared with tumor histology. Tumor vessel density and hypoxia were measured by quantitative immunohistochemistry. Normal tissues effects were evaluated in gut
EP
and skin.
Results: Non-treated tumors grew to 1000mm3 in 25.4d (±1.2), compared with delays of 62.3d
AC C
(±3.5) for SRT+CDDP and 80.2d (±5.0) for PRT+CDDP. Time to reach 2x pre-treatment volume ranged from 8.2d (±1.8) for non-treated tumors to 67.1d (±4.7) after PRT+CDDP. Significant differences in tumor growth delay were observed for SRT vs SRT+CDDP (p=0.04), PRT vs PRT+CDDP (p=0.035) and SRT+CDDP vs PRT+CDDP (p=0.033), and for survival between PRT vs PRT+CDDP (p=0.017) and SRT+CDDP vs PRT+CDDP (p = 0.008). Differences in tumor hypoxia were evident by 18F-MISO PET imaging between SRT vs PRT
Page 3 of 25
ACCEPTED MANUSCRIPT
(p=0.025), although not with concurrent CDDP. Tumor vessel density differed between SRT+CDDP vs PRT+CDDP (p=0.011). No differences in normal tissue parameters were seen.
RI PT
Conclusions: Concurrent CDDP was more effective in combination PRT than SRT at restricting tumor growth. Significant differences in tumor vascular density were evident between PRT and
AC C
EP
TE D
M AN U
SC
SRT, suggesting a preservation of vascular network with PRT.
Page 4 of 25
ACCEPTED MANUSCRIPT
SUMMARY Tumor-bearing mice were irradiated using a clinical schedule of 2 Gy/d for 3 wks, with concurrent cisplatin in weeks 2 and 3. Two distinct radiation schemes were compared, standard
RI PT
continuous radiation delivery (SRT) and pulsed radiotherapy (PRT). The aim was to test the hypothesis that the preservation of tumor vasculature seen in previous PRT studies would
facilitate improved drug delivery leading to an improved therapeutic response. PRT+cisplatin
AC C
EP
TE D
M AN U
SC
was more effective than SRT+cisplatin, supporting this hypothesis.
Page 5 of 25
ACCEPTED MANUSCRIPT
INTRODUCTION
Many patients with locally advanced head and neck squamous cell carcinoma (HNSCC) are
RI PT
treated with an organ preserving approach utilizing a combination of chemotherapy and radiation. This approach avoids the morbidity of radical surgical resection. Despite the
development of new targeted agents (1,2), as well as improved imaging (3) and advances in
SC
radiation treatment techniques, local control of HNSCC remains a significant issue in a subset of patients (4,5). Two year control of locally advanced p16 negative HNSCCs continues to be poor,
M AN U
and salvage from failure generally requires invasive surgery with resultant major complications (6). Even with aggressive salvage, the majority of local failures will result in cancer related death. Tumor hypoxia has been identified as a significant component of HNSCC radioresistance and plays an important role in local tumor response to radiotherapy (7). Increased levels of tumor
TE D
hypoxia have been correlated with poor tumor control in both tumor models and patients.
A potential mechanism for overcoming HNSCC radioresistance is exploiting low-dose hyper-
EP
radiosensitivity (HRS) (8) by delivering a standard dose of radiation (2 Gy) in ten 0.2 Gy pulses with a 3 minute break between each pulse; a concept named pulsed radiotherapy (PRT). By
AC C
limiting the radiation dose to less than 0.25 Gy per pulse, the amount of DNA damage is insufficient to trigger DNA damage repair pathways or induce cell cycle arrest via ATMdependent cell cycle checkpoint, resulting in increased cell killing per unit dose compared with standard 2 Gy doses of radiation (9,10). The University of Wisconsin has shown this technique to be safe and effective for clinical use in the treatment of recurrent glioblastoma (11), although
Page 6 of 25
ACCEPTED MANUSCRIPT
their clinical rationale to irradiate discontinuously was to reduce the apparent radiation dose rate (12).
RI PT
Previous pre-clinical xenograft studies in glioblastoma (13,14) and HNSCC (15) have
demonstrated that radiation fractionation given by PRT is not inferior to SRT despite the prolonged delivery time, and in some cases has been shown to be more effective.
SC
Mechanistically, in comparison to SRT, PRT appears to be less damaging to the vascular
network within the irradiated tumor (13,15). The preservation of vascular architecture after PRT
M AN U
could improve tumor oxygenation during a protracted course of radiotherapy and may explain the improved therapeutic effect, a hypothesis that has been supported by experimental data from 18
F-FMISO PET imaging (15).
TE D
Since PRT has been shown to preserve tumor vascularity and reduce tumor hypoxia compared to SRT, it was hypothesized that PRT might also allow for better concomitant drug delivery during weekly fractionation resulting in a superior chemo-radiosensitization. The aim of this study was
EP
to test this hypothesis using fractionated radiation and cisplatin in a UT-SCC-14 murine xenograft model. This chemoradiation regimen was chosen because radiation and concomitant
AC C
cisplatin is the clinical standard of care of HNSCC (16), and cisplatin is a recognized radiosensitizer (17).
Page 7 of 25
ACCEPTED MANUSCRIPT
MATERIALS AND METHODS
Animal Model: 8-9 week old athymic female mice (NIH-Lystbg-JFoxn1nuBtkxid) were purchased
RI PT
from CRL (Strain #201; Wilmington, MA). UT-SCC-14 (primary oral tongue, T3N1M0 and grade 2, p53 mutated and HPV negative) cells were obtained from Dr. Reidar Grénman
(University of Turku, Finland). Tumors were established by injecting 2x106 cells in 100 µL of
SC
Matrigel (BD, Franklin Lakes, NJ) into the right posterior flank. Xenografts were measured by a single observer, twice per week using digital calipers. Animals were sacrificed when tumor
M AN U
volume reached >2000mm3, or 100 days after the conclusion of treatment. All animal experiments were approved by the Institute Animal Care and Use Committee.
Chemotherapy and Radiation Treatments: After implantation, animals were monitored until
TE D
palpable tumors were seen and then volumes determined twice-weekly from digital calipers measurements. Treatments began once tumor volumes exceeded 200mm3, which occurred 3 weeks post-implant. Animals were randomized into six treatment cohorts (no-treatment controls,
EP
cisplatin [CDDP], PRT, SRT, PRT+CDDP and SRT+CDDP: n=7‒9 per group) and mean tumor
AC C
volumes for each group fell with the range of 250–320 mm3. Within each cohort, animals were randomly-assigned intraperitoneal (i.p.) CDDP or i.p. saline as sham-drug control.
Radiation was delivered using a cabinet X-ray irradiation system (160 kVp, 40 mA, HVL: 0.77mm Cu, cabinet temperature 30‒32°C) (Faxitron Bioptics, Tucson, AZ) at a dose rate of 0.25 Gy/min. Animals were positioned in a customized lead jig and anesthetized using 1‒3% isoflurane for the duration of treatment. For both radiation arms, 30 Gy was delivered in 2 Page 8 of 25
ACCEPTED MANUSCRIPT
Gy/day fractions for five days per week (Monday–Friday). Each fraction of SRT was a single continuous treatment of 8 minutes. PRT was 10 pulses of 0.2 Gy with a 3 minute break between pulses for total treatment time of 35 minutes. Cisplatin (2 mg/kg in 0.1 mL sterile saline) was
RI PT
given i.p. 30 minutes prior to RT three times per week (Mon, Wed, Fri) during the second and third weeks of radiotherapy. Sterile saline (i.p. 0.1 mL) was used as vehicle control using the
SC
same schedule.
PET Imaging: PET/CT images were acquired using a FLEX TriumphTM tri-modality
M AN U
MicroPET/SPECT/CT system (TriFoil Imaging, Northridge, CA). Briefly, 18.5 MBq of 18FMISO was injected via tail vein, followed by a 2 hour uptake period, after which animals were anesthetized with 1‒3% isoflurane and PET/CT imaging performed. Images were acquired before initiation of treatment and on completion. PET images were reconstructed using a 3D-
TE D
OSEM reconstruction algorithm and co-registered to CT images using a fixed-distance translation calculation in proprietary imaging analysis software (TriFoil Imaging, Northridge, CA). Tumors were identified on CT imaging and the corresponding volume in the PET dataset
EP
was analyzed to define SUVmax. 18F-FMISO-PET isotope was produced in-house using an automated synthesis procedure on the TRACERlab FXF-N (GE Medical, Milwaukee, WI)(18).
AC C
Precursors and reagents for synthesis were purchased from Advanced Biochemical Compounds (Radeburg, Germany).
Quantitative Immunohistochemistry: A second group of animals (n=3/gp) were treated as outlined above and sacrificed immediately after completion of treatment, following an i.p. injection of 60 mg/kg pimonidazole (in 100 µl of sterile PBS) (Hypoxyprobe Inc., Burlington,
Page 9 of 25
ACCEPTED MANUSCRIPT
MA). Tumor and in-field normal tissues (skin, gut) were collected for histology and immunohistochemical analysis.
RI PT
Immunohistochemistry Image Analysis: Tumor sections were stained with CD34 and ERCC1 (see supplementary materials), and counterstained with haematoxylin. Slides were scanned and digitized using the Aperio ScanScope (Aperio Technologies Inc. Buffalo Grove, IL). Analysis
SC
was performed using Definiens Tissue Studio® software (Definiens Inc. Carlsbad, CA) to detect and classify the stained marker of choice. The analysis algorithm was then applied to all the
M AN U
digital slides to ensure that the same standards were used throughout the study. Damaged or necrotic tissue was manually excluded from analysis; all subject information was blinded to the operator.
TE D
Normal tissues analysis: Skin samples (both in-field and lead-shielded out-of field) and in-field small intestine were collected. For skin, tissue sections were stained with H&E and digital scans were analyzed using algorithms to define length, width and length/width features to measure
EP
thickness of the epidermal layer. For gut, cellular density was calculated by dividing the number of cells in each ROI by the measured area in µm² and multiplying that solution by 1e6 to
AC C
measure cells/mm², villi length and basal width (crypt height) measurements, and images analyzed by the Definiens Tissue Studio® software (see supplementary materials).
Statistical analysis: Individual tumor volumes were normalized to their volume on the first day of treatment, and values for each treatment cohort combined as mean±SEM. Tumor growth delay was defined as time for the volume to double from the start of treatment volume. An
Page 10 of 25
ACCEPTED MANUSCRIPT
approximate curve was fitted to each tumor’s growth plot and solved for 2 to estimate time to double in volume. Time to reach 1000 mm3 was also calculated, using the absolute volume values. The Student’s t-test was then used to calculate significance between each treatment
AC C
EP
TE D
M AN U
SC
RI PT
group.
Page 11 of 25
ACCEPTED MANUSCRIPT
RESULTS
Individual tumor volumes were normalized against their pre-treatment starting volume, and then
RI PT
mean±SEM values calculated for each treatment cohort. Figure 1 shows the change in mean normalized tumor volumes for each group plotted as a function of time from the initiation of treatment on Day 0. The average time to reach twice the pre-treatment volume was 8.18 days
SC
(SEM±1.83d) for non-treated control tumors, 13.15 days (±3.63d) after CDDP only, 45.65 days (±2.53d) after SRT, 50.39 days (±4.16d) after PRT, 58.80 days (±2.42d) after SRT+CDDP and
M AN U
67.05 days (±4.65d) for PRT+ CDDP. Significant differences in tumor growth delay were found between SRT vs. SRT+CDDP (p=0.04), PRT vs PRT+CDDP (p=0.035), and also between SRT+CDDP vs PRT+CDDP (p=0.033).
TE D
Growth delay to reach an absolute tumor volume of 1000mm3 was also analyzed. These data are plotted in Figure 2. The average time from the start of treatment to reach 1000mm3 was 25.46 days (±1.20d) for non-treated controls, 31 days (±2.85d) for CDDP, 56.67 days (±3.58d) for
EP
SRT, 66.64 days (±5.18d) for PRT, 62.33 days (±3.50d) for SRT+CDDP, and 80.22 days (±4.97d) for PRT+CDDP. These data demonstrate that adding CDDP to RT increases overall
AC C
tumor growth delay to a volume of 1000mm3. Absolute tumor volumes for all cohorts as a function of time post treatment can be seen in supplementary data Figure S1.
Survival time, defined as time to reach sacrifice criteria or 100 days from the conclusion of treatment, was analyzed using the Kaplan-Meier Estimator (Figure 2). The average survival time from the start of treatment (day 0) was 43.5 days (SEM±4.66d) for non-treated controls, 53.5 days (±2.39d) for CDDP, 74.3 days (±6.67d) for SRT, 83.7 days (±5.84d) for PRT, 87.2 days Page 12 of 25
ACCEPTED MANUSCRIPT
(±3.80d) for SRT+CDDP and 103.7 days (±1.38) for PRT+CDDP. Statistical significance was found between PRT vs PRT+CDDP (p=0.017) and between SRT+CDDP vs PRT+CDDP
RI PT
(p=0.008).
Tumor hypoxia was measured non-invasively by 18F-FMISO PET imaging. Assessments were made before (Pre-Tx) and after treatment (Post-Tx), and the change in SUVmax value was
SC
calculated to determine the alteration in tumor hypoxia for each treatment cohort. Figure 3 shows mean (±SEM) FMISO SUVmax values normalized to Pre-Tx. A measure of normalized
M AN U
SUVmax=1 indicates no change in tumor hypoxia between the Pre-Tx and Post-Tx assessments, while values that exceed 1 indicates higher levels of tumor hypoxia at Post-Tx. For non-treated control tumors, the mean normalized SUVmax value increased from 1.0 to 1.64 (±0.05) as the tumor increased in size, indicating an increasing level of tumor hypoxia. SRT-treated
TE D
(1.51±0.23) and PRT-treated (1.26±0.27) tumors also exhibited increases, albeit to a lesser extent. When the two radiation schemes are directly compared, PRT-treated tumors exhibited
EP
lower levels of intratumoral hypoxia (p=0.025).
After CDDP only, the normalized SUVmax (0.88±0.05) values decreased indicating a reduction
AC C
in tumor hypoxia compared with pre-treatment scans. The addition of concomitant CDDP to both radiation regimens resulted in SUVmax values of approximately 1, indicating no overall change in the level of tumor hypoxia. The SRT±CDDP regimen significantly reduced SUVmax compared with SRT alone (p=0.012), although the commitment addition of CDDP to PRT did not significantly change tumor hypoxia.
Page 13 of 25
ACCEPTED MANUSCRIPT
Representative tumor sections from each of the four treatment cohorts; non-treated controls, CDDP only, SRT+CDDP and PRT+CDDP are shown (Figure 4). The bar graphs illustrate data generated from multiple sections of six distinct tumors from each treatment group. These four
RI PT
cohorts were selected and compared because the experimental aim was to evaluate the
commitment use of CDDP. Panel A shows the mean vessel density determined by staining with vascular marker CD34. Comparable numbers of CD34-positive vessels are detected in non-
SC
treated control tumors (Mean 261.6 (SD±95.9)) and after PRT+CDDP-treatment (275.7±60.6) (Figure 4, panel A). However, significantly lower vessel densities were seen after CDDP
tumors were not different from controls.
M AN U
(p=0.011) and SRT+CDDP (p=0.047) compared with untreated controls. PRT+CDDP treated-
The majority of cisplatin-induced DNA adducts are repaired by the NER pathway (19,20), and
TE D
the expression of the ERCC1 NER gene is associated with tumor resistance to CDDP cytotoxicity. Higher expression of ERRC1 was associated with inferior progression free survival for HNSCC after cisplatin radiotherapy (21,22), and ERCC1 expression has good prognostic
EP
value in patients with HPV-negative tumors (21). The percent of ERCC1-positive cell nuclei in viable regions of the tumor was measured after the CDDP containing regimens. A similar
AC C
number of ERCC1-positive nuclei were seen after SRT+CDDP and PRT+CDDP treatments but fewer positive nuclei were seen than after CDDP treatment alone (Figure 4, panel B).
Pimonidazole, an exogenous marker of hypoxia, has been shown to have prognostic value for loco-regional control in patients with head and neck squamous cell carcinoma tumors (23) and in pre-clinical experimental murine xenografts (24). Quantification of pimonidazole staining indicated higher levels of positively stained areas after CDDP and the two chemoradiation Page 14 of 25
ACCEPTED MANUSCRIPT
regimens compared with non-treated tumors, albeit no significant differences were detected between the three treatment groups (Figure 4, panel C).
RI PT
In past studies and the current investigation, PRT has been demonstrated to be more efficacious than SRT. To initially determine the effect of PRT on the therapeutic ratio, samples of normal skin (in-field and out-of-field) and sections of jejunal gut (in-field) were harvested at the
SC
completion of treatment (Figure 5). The aim of this preliminary examination was to determine if PRT resulted in more radiation-induced normal tissue injury than SRT when given concurrently
M AN U
with CDDP, following a three week schedule. Cellular density in crypts and villi in the jejunal cross-section were assessed using published methodology from studies examining damage after radiation or CDDP administration (25,26). No differences in crypt cell density around the circumference of the cross section (Figure 5, Panel A), villi length (not shown) or basal width
TE D
(not shown) were seen. Likewise, the mean epidermal thickness was calculated from nonirradiated (shielded, out-of-filed) and irradiated (in-field) skin sections (Figure 5, Panel B). While the out-of-field skin samples showed no differences in mean epidermal width between the
EP
three CDDP-treated cohorts, non-significant differences were seen between animals given CDDP
AC C
only and those treated with radiation + CDDP.
Page 15 of 25
ACCEPTED MANUSCRIPT
DISCUSSION
In this report, we have demonstrated that PRT with concomitant CDDP produced a longer tumor
RI PT
growth delay than SRT with concomitant CDDP. This response correlated with improved
survival, which in this study was defined as time to reach pre-determined sacrifice criteria based on tumor burden. We further showed differences in tumor hypoxia and tumor vessel density
SC
between PRT+CDDP and SRT+CDDP treated tumors. We hypothesize that the prolonged tumor growth delay and improvement in survival seen for the PRT+CDDP group, compared with the
M AN U
SRT+CDDP group, may be due to the improved delivery of CDDP to the tumor due to a higher tumor vessel density, although this hypothesis remains to be confirmed.
No differences in normal tissue response were seen in either skin or gut between PRT+CDDP
TE D
and SRT+CDDP schemes, albeit after an assessment at a single time point. Taken together, the tumor and normal tissue responses are consistent with the concept of an improved therapeutic
EP
ratio after PRT+CDDP in comparison to SRT+CDDP in this tumor model.
Solid tumors require angiogenesis for their growth and metastasis (27,28). Resistance to
AC C
radiation therapy is somewhat dependent on the degree of tumor hypoxia. Tumor xenografts can exhibit steady-state (diffusion-limited) hypoxia (29) and cycling (perfusion-modulated) hypoxia (30); it is unclear which of these hypoxia mechanisms is affected by PRT. Anti-angiogenic therapy has been extensively studied in both preclinical and clinical settings, with the goal of augmenting the direct killing mechanism of radiation. For example, interference with tumor– stromal interactions to reduce tumor vascular density and alter angiogenesis (31), along with
Page 16 of 25
ACCEPTED MANUSCRIPT
combinations of angiogenic inhibitors (32,33) have all shown promise. Directly inhibiting vascular endothelial growth factor signaling is an alternate approach (34). Xenograft studies have demonstrated that anti-angiogenic therapy may actually increase tumor oxygenation (35) as a
RI PT
result of tumor vascular normalization (36), and allow for a more efficient oxygen and drug delivery (37,38). PRT may be working by a similar vascular normalization mechanism, but more extensive validation studies are needed to confirm this. However, vascular preservation after
SC
PRT has been reported in other pre-clinical xenograft studies (13-15), the mechanism of this observation as linked with VEGF expression (15). Indirect support for this concept of improved
M AN U
drug delivery after PRT+CDDP compared with SRT+CDDP can be drawn from CDDP only data, since CDDP alone gave the largest reductions in vascular density (Figure 4) and tumor hypoxia (Figure 3). We hypothesize this treatment group reflects the highest degree of CDDP
TE D
delivery because of the absence of any radiation damage to vascular network.
In the current studies, a preservation of the vascular network as defined by a higher vessel density within the tumor was seen after PRT+CDDP delivery (Figure 4A). Although the
EP
difference between PRT+CDDP and SRT+CDDP did not reach statistical significance (p=0.08), only PRT+CDDP was statistically indistinguishable from control values. This observation is
AC C
consistent with the higher vessel densities recorded in tumors treated with PRT compared with SRT in prior studies with the same UT-SCC-14 tumor model (15) and other xenograft models (13,14). Other experimental approaches have been developed to enhance drug efficacy, such as phosphoinositide-3 kinase (PI3K) to alter vascular structure (39), or the use of liposomal (40) or other encapsulated drugs (41) to enhanced release within the tumors. If validated, the use of PRT
Page 17 of 25
ACCEPTED MANUSCRIPT
to improve drug delivery could be an additional strategy that could be more rapidly translated to clinical use, since no additional pharmaceutical interventions are required.
RI PT
The data presented in Figure 1 demonstrate that the combination of the concurrent cisplatin and PRT results in superior tumor growth delay compared to the concurrent cisplatin and SRT. No significant difference was seen between SRT and PRT alone in the absence of systemic
SC
chemotherapy. While the precise reason for the improved efficacy of PRT+CDDP compared with SRT+CDDP remains to be elucidated, several possibilities exist: (1) improved drug delivery
M AN U
due to maintenance of the vascular network within the tumor after PRT as described above (2), improved radiosensitization of CDDP-induced DNA adducts after PRT compared with SRT, or (3) improved oxygen delivery via preserved vascular network leading to more oxygen-dependent cell killing after PRT compared with SRT. The most likely explanation remains the improved
TE D
delivery of drug from vascular network. However, to address other possibilities, we also investigated improved radiosensitization by assessing the repair of DNA adducts by NER. The data showing ECRR1 staining in tumor sections (Figure 4B) were not conclusive, and future
EP
studies are being planned to determine the absolute delivery of CDDP to irradiated tumors using
AC C
atomic absorption spectrophotometry (39,40).
Previously published studies using PRT for recurrent glioblastoma have demonstrated that PRT is clinically feasible, and patients have tolerated PRT treatment well (11). Volumetric modulated arc radiotherapy (VMAT) reduces treatment delivery time significantly and a PRT plan can be delivered in approximately 50 minutes (42). However, this length of treatment may still represent a challenge for head and neck patients as they near the completion of their treatment when the
Page 18 of 25
ACCEPTED MANUSCRIPT
toxicities of treatment are reaching a maximum. There may still be a benefit for the use of PRT during the initial weeks of radiotherapy followed by SRT. Current preclinical studies are ongoing testing the efficacy of combined SRT and PRT delivery. A significant advantage of PRT is that
RI PT
the technique could be implemented in that majority of radiation oncology facilities without any new equipment and with minimal additional training of therapy staff. Furthermore, based on our
SC
normal tissue data, the risk of increased toxicity or other adverse events is low.
The discovery of the prognostic importance of HPV in head and neck cancers in conjunction
M AN U
with other prognostic factors, such as stage and tobacco use, has provided a rationale for deescalation of treatment. However, these factors also allow for the identification of patients with a poor prognosis who may benefit from intensification of treatment. As the current standard treatment of locally advanced head and neck cancer has significant acute and chronic toxicities,
TE D
we must investigate novel approaches to improving outcomes in these high risk patients that do not subject our patients to additional treatment related toxicity. The findings presented suggest that the use of concurrent cisplatin with PRT may represent a novel treatment options for patients
EP
with high risk locally advanced head and neck squamous cell carcinoma. A primary potential
AC C
barrier to the clinical implementation of PRT is the increased treatment time compared to SRT.
Page 19 of 25
ACCEPTED MANUSCRIPT
FIGURE LEGENDS
Figure 1: Normalized tumor volumes as a function of time from initiation on treatment
RI PT
(Mean±SEM, n=7-9 per group). Radiation treatments began on Day 0. Negative values on the abscissa indicate tumor measurements made prior to initiation of treatment. Data are censored
SC
when the group sizes reached n=5.
Figure 2A: Time to reach 1000mm3 tumor volume as a function of time (Mean±SEM) after
M AN U
implantation. 2B: Kaplan-Meier plot indicating survival after the 6 treatment regimens.
Figure 3: FMISO images from representative animal in each treatment group (left) and quantified as Normalized SUVmax (right, mean±SD, n=3 per group). SUVmax values are
TE D
normalized to pre-treatment levels.
Figure 4: Representative images for H&E, CD34, ERRC1 and pimonidazole staining for each
EP
treatment group containing CDDP, and quantified using automated blinded image analysis.
AC C
Figure 5: Quantification of normal tissue damage in jejunum and skin. Upper panels are representative sections illustrating the assay methodology, lower panels show comparison between each group.
Page 20 of 25
ACCEPTED MANUSCRIPT
ACKNOWLEDGEMENTS
This work was funded by the Department of Radiation Oncology, XXXX and XXXX seed grant
RI PT
awarded to Dr. XXXX. The authors thank XXXX for technical expertise and XXXX for
AC C
EP
TE D
M AN U
SC
assistance with pathology.
Page 21 of 25
ACCEPTED MANUSCRIPT
REFERENCES 1. Mockelmann N, Kriegs M, Lorincz BB, et al. Molecular targeting in combination with platinumbased chemoradiotherapy in head and neck cancer treatment. Head Neck 2015.
RI PT
2. Ferreira MB, Lima JP Cohen EE. Novel targeted therapies in head and neck cancer. Expert Opin Investig Drugs 2012;21:281-295.
3. Baschnagel AM, Wobb JL, Dilworth JT, et al. The association of (18)F-FDG PET and glucose
carcinomas. Radiother Oncol 2015;117:118-124.
SC
metabolism biomarkers GLUT1 and HK2 in p16 positive and negative head and neck squamous cell
M AN U
4. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature 2015;517:576-582.
5. Ang KK, Harris J, Wheeler R, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med 2010;363:24-35.
6. Rischin D, Young RJ, Fisher R, et al. Prognostic significance of p16INK4A and human
TE D
papillomavirus in patients with oropharyngeal cancer treated on TROG 02.02 phase III trial. J Clin Oncol 2010;28:4142-4148.
7. Nordsmark M, Bentzen SM, Rudat V, et al. Prognostic value of tumor oxygenation in 397 head and
2005;77:18-24.
AC C
8. XXX
EP
neck tumors after primary radiation therapy. An international multi-center study. Radiother Oncol
9. XXX 10. XXX
11. Adkison JB, Tome W, Seo S, et al. Reirradiation of large-volume recurrent glioma with pulsed reduced-dose-rate radiotherapy. Int J Radiat Oncol Biol Phys 2011;79:835-841. 12. Rasmussen KH, Hardcastle N, Howard SP, et al. Reirradiation of glioblastoma through the use of a reduced dose rate on a tomotherapy unit. Technol Cancer Res Treat 2010;9:399-406. 13. XXX Page 22 of 25
ACCEPTED MANUSCRIPT
14. XXX 15. XXX 16. Adelstein DJ, Li Y, Adams GL, et al. An intergroup phase III comparison of standard radiation
RI PT
therapy and two schedules of concurrent chemoradiotherapy in patients with unresectable squamous cell head and neck cancer. J Clin Oncol 2003;21:92-98.
17. Dewit L. Combined treatment of radiation and cisdiamminedichloroplatinum (II): a review of
SC
experimental and clinical data. Int J Radiat Oncol Biol Phys 1987;13:403-426.
18. Oh SJ, Chi DY, Mosdzianowski C, et al. Fully automated synthesis of [18F]fluoromisonidazole using
M AN U
a conventional [18F]FDG module. Nucl Med Biol 2005;32:899-905.
19. Dijt FJ, Fichtinger-Schepman AM, Berends F, et al. Formation and repair of cisplatin-induced adducts to DNA in cultured normal and repair-deficient human fibroblasts. Cancer Res 1988;48:6058-6062.
20. Larminat F Bohr VA. Role of the human ERCC-1 gene in gene-specific repair of cisplatin-induced
TE D
DNA damage. Nucleic Acids Res 1994;22:3005-3010.
21. Hao D, Lau HY, Eliasziw M, et al. Comparing ERCC1 protein expression, mRNA levels, and genotype in squamous cell carcinomas of the head and neck treated with concurrent chemoradiation
EP
stratified by HPV status. Head Neck 2012;34:785-791. 22. Vaezi A, Wang X, Buch S, et al. XPF expression correlates with clinical outcome in squamous cell
AC C
carcinoma of the head and neck. Clin Cancer Res 2011;17:5513-5522. 23. Kaanders JH, Wijffels KI, Marres HA, et al. Pimonidazole binding and tumor vascularity predict for treatment outcome in head and neck cancer. Cancer Res 2002;62:7066-7074. 24. Yaromina A, Zips D, Thames HD, et al. Pimonidazole labelling and response to fractionated irradiation of five human squamous cell carcinoma (hSCC) lines in nude mice: the need for a multivariate approach in biomarker studies. Radiother Oncol 2006;81:122-129.
Page 23 of 25
ACCEPTED MANUSCRIPT
25. Allan SG, Smyth JF, Hay FG, et al. Protective effect of sodium-2-mercaptoethanesulfonate on the gastrointestinal toxicity and lethality of cis-diamminedichloroplatinum. Cancer Res 1986;46:35693573.
in tumor bearing mice. BMC Complement Altern Med 2013;13:103.
RI PT
26. Son TG, Gong EJ, Bae MJ, et al. Protective effect of genistein on radiation-induced intestinal injury
27. Folkman J. The role of angiogenesis in tumor growth. Semin Cancer Biol 1992;3:65-71.
SC
28. Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971;285:1182-1186.
29. Thomlinson RH Gray LH. The histological structure of some human lung cancers and the possible
M AN U
implications for radiotherapy. Br J Cancer 1955;9:539-549.
30. Trotter MJ, Chaplin DJ, Durand RE, et al. The use of fluorescent probes to identify regions of transient perfusion in murine tumors. Int J Radiat Oncol Biol Phys 1989;16:931-934. 31. Li C, Huang S, Armstrong EA, et al. Antitumor Effects of MEHD7945A, a Dual-Specific Antibody against EGFR and HER3, in Combination with Radiation in Lung and Head and Neck Cancers. Mol
TE D
Cancer Ther 2015;14:2049-2059.
32. Ma J Waxman DJ. Combination of antiangiogenesis with chemotherapy for more effective cancer treatment. Mol Cancer Ther 2008;7:3670-3684.
EP
33. Fukumura D Jain RK. Tumor microvasculature and microenvironment: targets for anti-angiogenesis and normalization. Microvasc Res 2007;74:72-84.
AC C
34. Kozin SV, Boucher Y, Hicklin DJ, et al. Vascular endothelial growth factor receptor-2-blocking antibody potentiates radiation-induced long-term control of human tumor xenografts. Cancer Res 2001;61:39-44.
35. McGee MC, Hamner JB, Williams RF, et al. Improved intratumoral oxygenation through vascular normalization increases glioma sensitivity to ionizing radiation. Int J Radiat Oncol Biol Phys 2010;76:1537-1545. 36. Jain RK. Normalizing tumor vasculature with anti-angiogenic therapy: a new paradigm for combination therapy. Nat Med 2001;7:987-989. Page 24 of 25
ACCEPTED MANUSCRIPT
37. Dings RP, Loren M, Heun H, et al. Scheduling of radiation with angiogenesis inhibitors anginex and Avastin improves therapeutic outcome via vessel normalization. Clin Cancer Res 2007;13:33953402.
RI PT
38. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 2005;307:58-62.
39. Qayum N, Im J, Stratford MR, et al. Modulation of the tumor microvasculature by phosphoinositide-3
SC
kinase inhibition increases doxorubicin delivery in vivo. Clin Cancer Res 2012;18:161-169.
40. Hagtvet E, Roe K Olsen DR. Liposomal doxorubicin improves radiotherapy response in hypoxic
M AN U
prostate cancer xenografts. Radiat Oncol 2011;6:135.
41. Eggen S, Afadzi M, Nilssen EA, et al. Ultrasound improves the uptake and distribution of liposomal Doxorubicin in prostate cancer xenografts. Ultrasound Med Biol 2013;39:1255-1266. 42. Tyagi N, Yang K, Sandhu R, et al. External beam pulsed low dose radiotherapy using volumetric
AC C
EP
TE D
modulated arc therapy: planning and delivery. Med Phys 2013;40:011704.
Page 25 of 25
Figure 1
RI PT
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
p=0.429
p=0.033
AC C
EP
p=0.045
p=0.035
ACCEPTED MANUSCRIPT
RI PT
Figure 2
AC C
EP
TE D
M AN U
SC
p=0.013
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 3
p=0.025 p=0.003
p=0.012
H&E
CD34
ERRC1
PIMO
p=0.047 p=0.011
TE D
AC C
CD34
B
EP
A
ERCC1
#187
M AN U
PRT+CDDP SRT+CDDP
#193
SC
CDDP
#200
RI PT
#199
Control
Figure 4
ACCEPTED MANUSCRIPT
C
PIMO
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 5
S0360-3016(16)30181-X
DOI:
10.1016/j.ijrobp.2016.04.031
Reference:
ROB 23575
To appear in:
International Journal of Radiation Oncology • Biology • Physics
Received Date: 17 February 2016 Revised Date:
13 April 2016
Accepted Date: 30 April 2016
Please cite this article as: Meyer K, Krueger SA, Kane JL, Wilson TG, Hanna A, Dabjan M, Hege KM, Wilson GD, Grills I, Marples B, Pulsed Radiotherapy with Concurrent Cisplatin results in Superior Tumor Growth Delay in a Head and Neck Squamous Cell Carcinoma Murine Model, International Journal of Radiation Oncology • Biology • Physics (2016), doi: 10.1016/j.ijrobp.2016.04.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Pulsed Radiotherapy with Concurrent Cisplatin results in Superior Tumor Growth Delay
RI PT
in a Head and Neck Squamous Cell Carcinoma Murine Model
Kurt Meyer MD, Sarah. A. Krueger PhD, Jonathan L. Kane BS, Thomas G. Wilson, BS,
SC
Alaa Hanna MD, Mohamad Dabjan MD, Katie M. Hege BS,
M AN U
George D. Wilson PhD, Inga Grills MD, Brian Marples PhD
TE D
Department of Radiation Oncology, William Beaumont Hospital, 3811 W. Thirteen Mile Rd, 105-RI, Royal Oak, Michigan 48073 Running title: Pulsed Radiotherapy and Cisplatin
AC C
Pages: 24 Figures: 5 Tables: 0
EP
Keywords: Radiation, UT-SCC-14, Chemoradiation, animal imaging
Address correspondence to: Brian Marples Ph.D., Department of Radiation Oncology, William Beaumont Hospital, 3811 W. Thirteen Mile Rd, 105-RI, Royal Oak, MI 48073. Tel: 248-551-0213; Fax: 248-551-2443; E-Mail: [email protected]
ACCEPTED MANUSCRIPT
Pulsed Radiotherapy with Concurrent Cisplatin results in Superior Tumor Growth Delay
M AN U
Running title: Pulsed Radiotherapy and Cisplatin
SC
RI PT
in a Head and Neck Squamous Cell Carcinoma Murine Model
AC C
EP
TE D
Keywords: Radiation, UT-SCC-14, Chemoradiation, animal imaging
Page 1 of 25
ACCEPTED MANUSCRIPT
CONFLICTS OF INTEREST NOTIFICATION
AC C
EP
TE D
M AN U
SC
RI PT
No conflicts of interest exist.
Page 2 of 25
ACCEPTED MANUSCRIPT
ABSTRACT
Purpose: To assess the efficacy of 3-week schedules of low-dose pulsed radiation treatment
squamous cell carcinoma (HNSCC) xenograft model.
RI PT
(PRT) and standard radiotherapy (SRT), with concurrent cisplatin (CDDP) in a head and neck
SC
Methods and Materials: Subcutaneous UT-SCC-14 tumors were established in athymic NIH III HO female mice. 30 Gy was administered as 2 Gy/day (d), 5d/wk for 3 wks either by PRT
M AN U
(10x0.2 Gy/day, with a 3-min break between each 0.2 Gy dose) or SRT (2 Gy/day, noninterrupted delivery) in combination with concurrent 2 mg/kg CDDP 3 times/wk in the final 2 weeks of RT. Treatment-induced growth delays were defined from twice-weekly tumor volume measurements. Tumor hypoxia was assessed by 18F-FMISO PET imaging, and calculated
TE D
SUVmax values compared with tumor histology. Tumor vessel density and hypoxia were measured by quantitative immunohistochemistry. Normal tissues effects were evaluated in gut
EP
and skin.
Results: Non-treated tumors grew to 1000mm3 in 25.4d (±1.2), compared with delays of 62.3d
AC C
(±3.5) for SRT+CDDP and 80.2d (±5.0) for PRT+CDDP. Time to reach 2x pre-treatment volume ranged from 8.2d (±1.8) for non-treated tumors to 67.1d (±4.7) after PRT+CDDP. Significant differences in tumor growth delay were observed for SRT vs SRT+CDDP (p=0.04), PRT vs PRT+CDDP (p=0.035) and SRT+CDDP vs PRT+CDDP (p=0.033), and for survival between PRT vs PRT+CDDP (p=0.017) and SRT+CDDP vs PRT+CDDP (p = 0.008). Differences in tumor hypoxia were evident by 18F-MISO PET imaging between SRT vs PRT
Page 3 of 25
ACCEPTED MANUSCRIPT
(p=0.025), although not with concurrent CDDP. Tumor vessel density differed between SRT+CDDP vs PRT+CDDP (p=0.011). No differences in normal tissue parameters were seen.
RI PT
Conclusions: Concurrent CDDP was more effective in combination PRT than SRT at restricting tumor growth. Significant differences in tumor vascular density were evident between PRT and
AC C
EP
TE D
M AN U
SC
SRT, suggesting a preservation of vascular network with PRT.
Page 4 of 25
ACCEPTED MANUSCRIPT
SUMMARY Tumor-bearing mice were irradiated using a clinical schedule of 2 Gy/d for 3 wks, with concurrent cisplatin in weeks 2 and 3. Two distinct radiation schemes were compared, standard
RI PT
continuous radiation delivery (SRT) and pulsed radiotherapy (PRT). The aim was to test the hypothesis that the preservation of tumor vasculature seen in previous PRT studies would
facilitate improved drug delivery leading to an improved therapeutic response. PRT+cisplatin
AC C
EP
TE D
M AN U
SC
was more effective than SRT+cisplatin, supporting this hypothesis.
Page 5 of 25
ACCEPTED MANUSCRIPT
INTRODUCTION
Many patients with locally advanced head and neck squamous cell carcinoma (HNSCC) are
RI PT
treated with an organ preserving approach utilizing a combination of chemotherapy and radiation. This approach avoids the morbidity of radical surgical resection. Despite the
development of new targeted agents (1,2), as well as improved imaging (3) and advances in
SC
radiation treatment techniques, local control of HNSCC remains a significant issue in a subset of patients (4,5). Two year control of locally advanced p16 negative HNSCCs continues to be poor,
M AN U
and salvage from failure generally requires invasive surgery with resultant major complications (6). Even with aggressive salvage, the majority of local failures will result in cancer related death. Tumor hypoxia has been identified as a significant component of HNSCC radioresistance and plays an important role in local tumor response to radiotherapy (7). Increased levels of tumor
TE D
hypoxia have been correlated with poor tumor control in both tumor models and patients.
A potential mechanism for overcoming HNSCC radioresistance is exploiting low-dose hyper-
EP
radiosensitivity (HRS) (8) by delivering a standard dose of radiation (2 Gy) in ten 0.2 Gy pulses with a 3 minute break between each pulse; a concept named pulsed radiotherapy (PRT). By
AC C
limiting the radiation dose to less than 0.25 Gy per pulse, the amount of DNA damage is insufficient to trigger DNA damage repair pathways or induce cell cycle arrest via ATMdependent cell cycle checkpoint, resulting in increased cell killing per unit dose compared with standard 2 Gy doses of radiation (9,10). The University of Wisconsin has shown this technique to be safe and effective for clinical use in the treatment of recurrent glioblastoma (11), although
Page 6 of 25
ACCEPTED MANUSCRIPT
their clinical rationale to irradiate discontinuously was to reduce the apparent radiation dose rate (12).
RI PT
Previous pre-clinical xenograft studies in glioblastoma (13,14) and HNSCC (15) have
demonstrated that radiation fractionation given by PRT is not inferior to SRT despite the prolonged delivery time, and in some cases has been shown to be more effective.
SC
Mechanistically, in comparison to SRT, PRT appears to be less damaging to the vascular
network within the irradiated tumor (13,15). The preservation of vascular architecture after PRT
M AN U
could improve tumor oxygenation during a protracted course of radiotherapy and may explain the improved therapeutic effect, a hypothesis that has been supported by experimental data from 18
F-FMISO PET imaging (15).
TE D
Since PRT has been shown to preserve tumor vascularity and reduce tumor hypoxia compared to SRT, it was hypothesized that PRT might also allow for better concomitant drug delivery during weekly fractionation resulting in a superior chemo-radiosensitization. The aim of this study was
EP
to test this hypothesis using fractionated radiation and cisplatin in a UT-SCC-14 murine xenograft model. This chemoradiation regimen was chosen because radiation and concomitant
AC C
cisplatin is the clinical standard of care of HNSCC (16), and cisplatin is a recognized radiosensitizer (17).
Page 7 of 25
ACCEPTED MANUSCRIPT
MATERIALS AND METHODS
Animal Model: 8-9 week old athymic female mice (NIH-Lystbg-JFoxn1nuBtkxid) were purchased
RI PT
from CRL (Strain #201; Wilmington, MA). UT-SCC-14 (primary oral tongue, T3N1M0 and grade 2, p53 mutated and HPV negative) cells were obtained from Dr. Reidar Grénman
(University of Turku, Finland). Tumors were established by injecting 2x106 cells in 100 µL of
SC
Matrigel (BD, Franklin Lakes, NJ) into the right posterior flank. Xenografts were measured by a single observer, twice per week using digital calipers. Animals were sacrificed when tumor
M AN U
volume reached >2000mm3, or 100 days after the conclusion of treatment. All animal experiments were approved by the Institute Animal Care and Use Committee.
Chemotherapy and Radiation Treatments: After implantation, animals were monitored until
TE D
palpable tumors were seen and then volumes determined twice-weekly from digital calipers measurements. Treatments began once tumor volumes exceeded 200mm3, which occurred 3 weeks post-implant. Animals were randomized into six treatment cohorts (no-treatment controls,
EP
cisplatin [CDDP], PRT, SRT, PRT+CDDP and SRT+CDDP: n=7‒9 per group) and mean tumor
AC C
volumes for each group fell with the range of 250–320 mm3. Within each cohort, animals were randomly-assigned intraperitoneal (i.p.) CDDP or i.p. saline as sham-drug control.
Radiation was delivered using a cabinet X-ray irradiation system (160 kVp, 40 mA, HVL: 0.77mm Cu, cabinet temperature 30‒32°C) (Faxitron Bioptics, Tucson, AZ) at a dose rate of 0.25 Gy/min. Animals were positioned in a customized lead jig and anesthetized using 1‒3% isoflurane for the duration of treatment. For both radiation arms, 30 Gy was delivered in 2 Page 8 of 25
ACCEPTED MANUSCRIPT
Gy/day fractions for five days per week (Monday–Friday). Each fraction of SRT was a single continuous treatment of 8 minutes. PRT was 10 pulses of 0.2 Gy with a 3 minute break between pulses for total treatment time of 35 minutes. Cisplatin (2 mg/kg in 0.1 mL sterile saline) was
RI PT
given i.p. 30 minutes prior to RT three times per week (Mon, Wed, Fri) during the second and third weeks of radiotherapy. Sterile saline (i.p. 0.1 mL) was used as vehicle control using the
SC
same schedule.
PET Imaging: PET/CT images were acquired using a FLEX TriumphTM tri-modality
M AN U
MicroPET/SPECT/CT system (TriFoil Imaging, Northridge, CA). Briefly, 18.5 MBq of 18FMISO was injected via tail vein, followed by a 2 hour uptake period, after which animals were anesthetized with 1‒3% isoflurane and PET/CT imaging performed. Images were acquired before initiation of treatment and on completion. PET images were reconstructed using a 3D-
TE D
OSEM reconstruction algorithm and co-registered to CT images using a fixed-distance translation calculation in proprietary imaging analysis software (TriFoil Imaging, Northridge, CA). Tumors were identified on CT imaging and the corresponding volume in the PET dataset
EP
was analyzed to define SUVmax. 18F-FMISO-PET isotope was produced in-house using an automated synthesis procedure on the TRACERlab FXF-N (GE Medical, Milwaukee, WI)(18).
AC C
Precursors and reagents for synthesis were purchased from Advanced Biochemical Compounds (Radeburg, Germany).
Quantitative Immunohistochemistry: A second group of animals (n=3/gp) were treated as outlined above and sacrificed immediately after completion of treatment, following an i.p. injection of 60 mg/kg pimonidazole (in 100 µl of sterile PBS) (Hypoxyprobe Inc., Burlington,
Page 9 of 25
ACCEPTED MANUSCRIPT
MA). Tumor and in-field normal tissues (skin, gut) were collected for histology and immunohistochemical analysis.
RI PT
Immunohistochemistry Image Analysis: Tumor sections were stained with CD34 and ERCC1 (see supplementary materials), and counterstained with haematoxylin. Slides were scanned and digitized using the Aperio ScanScope (Aperio Technologies Inc. Buffalo Grove, IL). Analysis
SC
was performed using Definiens Tissue Studio® software (Definiens Inc. Carlsbad, CA) to detect and classify the stained marker of choice. The analysis algorithm was then applied to all the
M AN U
digital slides to ensure that the same standards were used throughout the study. Damaged or necrotic tissue was manually excluded from analysis; all subject information was blinded to the operator.
TE D
Normal tissues analysis: Skin samples (both in-field and lead-shielded out-of field) and in-field small intestine were collected. For skin, tissue sections were stained with H&E and digital scans were analyzed using algorithms to define length, width and length/width features to measure
EP
thickness of the epidermal layer. For gut, cellular density was calculated by dividing the number of cells in each ROI by the measured area in µm² and multiplying that solution by 1e6 to
AC C
measure cells/mm², villi length and basal width (crypt height) measurements, and images analyzed by the Definiens Tissue Studio® software (see supplementary materials).
Statistical analysis: Individual tumor volumes were normalized to their volume on the first day of treatment, and values for each treatment cohort combined as mean±SEM. Tumor growth delay was defined as time for the volume to double from the start of treatment volume. An
Page 10 of 25
ACCEPTED MANUSCRIPT
approximate curve was fitted to each tumor’s growth plot and solved for 2 to estimate time to double in volume. Time to reach 1000 mm3 was also calculated, using the absolute volume values. The Student’s t-test was then used to calculate significance between each treatment
AC C
EP
TE D
M AN U
SC
RI PT
group.
Page 11 of 25
ACCEPTED MANUSCRIPT
RESULTS
Individual tumor volumes were normalized against their pre-treatment starting volume, and then
RI PT
mean±SEM values calculated for each treatment cohort. Figure 1 shows the change in mean normalized tumor volumes for each group plotted as a function of time from the initiation of treatment on Day 0. The average time to reach twice the pre-treatment volume was 8.18 days
SC
(SEM±1.83d) for non-treated control tumors, 13.15 days (±3.63d) after CDDP only, 45.65 days (±2.53d) after SRT, 50.39 days (±4.16d) after PRT, 58.80 days (±2.42d) after SRT+CDDP and
M AN U
67.05 days (±4.65d) for PRT+ CDDP. Significant differences in tumor growth delay were found between SRT vs. SRT+CDDP (p=0.04), PRT vs PRT+CDDP (p=0.035), and also between SRT+CDDP vs PRT+CDDP (p=0.033).
TE D
Growth delay to reach an absolute tumor volume of 1000mm3 was also analyzed. These data are plotted in Figure 2. The average time from the start of treatment to reach 1000mm3 was 25.46 days (±1.20d) for non-treated controls, 31 days (±2.85d) for CDDP, 56.67 days (±3.58d) for
EP
SRT, 66.64 days (±5.18d) for PRT, 62.33 days (±3.50d) for SRT+CDDP, and 80.22 days (±4.97d) for PRT+CDDP. These data demonstrate that adding CDDP to RT increases overall
AC C
tumor growth delay to a volume of 1000mm3. Absolute tumor volumes for all cohorts as a function of time post treatment can be seen in supplementary data Figure S1.
Survival time, defined as time to reach sacrifice criteria or 100 days from the conclusion of treatment, was analyzed using the Kaplan-Meier Estimator (Figure 2). The average survival time from the start of treatment (day 0) was 43.5 days (SEM±4.66d) for non-treated controls, 53.5 days (±2.39d) for CDDP, 74.3 days (±6.67d) for SRT, 83.7 days (±5.84d) for PRT, 87.2 days Page 12 of 25
ACCEPTED MANUSCRIPT
(±3.80d) for SRT+CDDP and 103.7 days (±1.38) for PRT+CDDP. Statistical significance was found between PRT vs PRT+CDDP (p=0.017) and between SRT+CDDP vs PRT+CDDP
RI PT
(p=0.008).
Tumor hypoxia was measured non-invasively by 18F-FMISO PET imaging. Assessments were made before (Pre-Tx) and after treatment (Post-Tx), and the change in SUVmax value was
SC
calculated to determine the alteration in tumor hypoxia for each treatment cohort. Figure 3 shows mean (±SEM) FMISO SUVmax values normalized to Pre-Tx. A measure of normalized
M AN U
SUVmax=1 indicates no change in tumor hypoxia between the Pre-Tx and Post-Tx assessments, while values that exceed 1 indicates higher levels of tumor hypoxia at Post-Tx. For non-treated control tumors, the mean normalized SUVmax value increased from 1.0 to 1.64 (±0.05) as the tumor increased in size, indicating an increasing level of tumor hypoxia. SRT-treated
TE D
(1.51±0.23) and PRT-treated (1.26±0.27) tumors also exhibited increases, albeit to a lesser extent. When the two radiation schemes are directly compared, PRT-treated tumors exhibited
EP
lower levels of intratumoral hypoxia (p=0.025).
After CDDP only, the normalized SUVmax (0.88±0.05) values decreased indicating a reduction
AC C
in tumor hypoxia compared with pre-treatment scans. The addition of concomitant CDDP to both radiation regimens resulted in SUVmax values of approximately 1, indicating no overall change in the level of tumor hypoxia. The SRT±CDDP regimen significantly reduced SUVmax compared with SRT alone (p=0.012), although the commitment addition of CDDP to PRT did not significantly change tumor hypoxia.
Page 13 of 25
ACCEPTED MANUSCRIPT
Representative tumor sections from each of the four treatment cohorts; non-treated controls, CDDP only, SRT+CDDP and PRT+CDDP are shown (Figure 4). The bar graphs illustrate data generated from multiple sections of six distinct tumors from each treatment group. These four
RI PT
cohorts were selected and compared because the experimental aim was to evaluate the
commitment use of CDDP. Panel A shows the mean vessel density determined by staining with vascular marker CD34. Comparable numbers of CD34-positive vessels are detected in non-
SC
treated control tumors (Mean 261.6 (SD±95.9)) and after PRT+CDDP-treatment (275.7±60.6) (Figure 4, panel A). However, significantly lower vessel densities were seen after CDDP
tumors were not different from controls.
M AN U
(p=0.011) and SRT+CDDP (p=0.047) compared with untreated controls. PRT+CDDP treated-
The majority of cisplatin-induced DNA adducts are repaired by the NER pathway (19,20), and
TE D
the expression of the ERCC1 NER gene is associated with tumor resistance to CDDP cytotoxicity. Higher expression of ERRC1 was associated with inferior progression free survival for HNSCC after cisplatin radiotherapy (21,22), and ERCC1 expression has good prognostic
EP
value in patients with HPV-negative tumors (21). The percent of ERCC1-positive cell nuclei in viable regions of the tumor was measured after the CDDP containing regimens. A similar
AC C
number of ERCC1-positive nuclei were seen after SRT+CDDP and PRT+CDDP treatments but fewer positive nuclei were seen than after CDDP treatment alone (Figure 4, panel B).
Pimonidazole, an exogenous marker of hypoxia, has been shown to have prognostic value for loco-regional control in patients with head and neck squamous cell carcinoma tumors (23) and in pre-clinical experimental murine xenografts (24). Quantification of pimonidazole staining indicated higher levels of positively stained areas after CDDP and the two chemoradiation Page 14 of 25
ACCEPTED MANUSCRIPT
regimens compared with non-treated tumors, albeit no significant differences were detected between the three treatment groups (Figure 4, panel C).
RI PT
In past studies and the current investigation, PRT has been demonstrated to be more efficacious than SRT. To initially determine the effect of PRT on the therapeutic ratio, samples of normal skin (in-field and out-of-field) and sections of jejunal gut (in-field) were harvested at the
SC
completion of treatment (Figure 5). The aim of this preliminary examination was to determine if PRT resulted in more radiation-induced normal tissue injury than SRT when given concurrently
M AN U
with CDDP, following a three week schedule. Cellular density in crypts and villi in the jejunal cross-section were assessed using published methodology from studies examining damage after radiation or CDDP administration (25,26). No differences in crypt cell density around the circumference of the cross section (Figure 5, Panel A), villi length (not shown) or basal width
TE D
(not shown) were seen. Likewise, the mean epidermal thickness was calculated from nonirradiated (shielded, out-of-filed) and irradiated (in-field) skin sections (Figure 5, Panel B). While the out-of-field skin samples showed no differences in mean epidermal width between the
EP
three CDDP-treated cohorts, non-significant differences were seen between animals given CDDP
AC C
only and those treated with radiation + CDDP.
Page 15 of 25
ACCEPTED MANUSCRIPT
DISCUSSION
In this report, we have demonstrated that PRT with concomitant CDDP produced a longer tumor
RI PT
growth delay than SRT with concomitant CDDP. This response correlated with improved
survival, which in this study was defined as time to reach pre-determined sacrifice criteria based on tumor burden. We further showed differences in tumor hypoxia and tumor vessel density
SC
between PRT+CDDP and SRT+CDDP treated tumors. We hypothesize that the prolonged tumor growth delay and improvement in survival seen for the PRT+CDDP group, compared with the
M AN U
SRT+CDDP group, may be due to the improved delivery of CDDP to the tumor due to a higher tumor vessel density, although this hypothesis remains to be confirmed.
No differences in normal tissue response were seen in either skin or gut between PRT+CDDP
TE D
and SRT+CDDP schemes, albeit after an assessment at a single time point. Taken together, the tumor and normal tissue responses are consistent with the concept of an improved therapeutic
EP
ratio after PRT+CDDP in comparison to SRT+CDDP in this tumor model.
Solid tumors require angiogenesis for their growth and metastasis (27,28). Resistance to
AC C
radiation therapy is somewhat dependent on the degree of tumor hypoxia. Tumor xenografts can exhibit steady-state (diffusion-limited) hypoxia (29) and cycling (perfusion-modulated) hypoxia (30); it is unclear which of these hypoxia mechanisms is affected by PRT. Anti-angiogenic therapy has been extensively studied in both preclinical and clinical settings, with the goal of augmenting the direct killing mechanism of radiation. For example, interference with tumor– stromal interactions to reduce tumor vascular density and alter angiogenesis (31), along with
Page 16 of 25
ACCEPTED MANUSCRIPT
combinations of angiogenic inhibitors (32,33) have all shown promise. Directly inhibiting vascular endothelial growth factor signaling is an alternate approach (34). Xenograft studies have demonstrated that anti-angiogenic therapy may actually increase tumor oxygenation (35) as a
RI PT
result of tumor vascular normalization (36), and allow for a more efficient oxygen and drug delivery (37,38). PRT may be working by a similar vascular normalization mechanism, but more extensive validation studies are needed to confirm this. However, vascular preservation after
SC
PRT has been reported in other pre-clinical xenograft studies (13-15), the mechanism of this observation as linked with VEGF expression (15). Indirect support for this concept of improved
M AN U
drug delivery after PRT+CDDP compared with SRT+CDDP can be drawn from CDDP only data, since CDDP alone gave the largest reductions in vascular density (Figure 4) and tumor hypoxia (Figure 3). We hypothesize this treatment group reflects the highest degree of CDDP
TE D
delivery because of the absence of any radiation damage to vascular network.
In the current studies, a preservation of the vascular network as defined by a higher vessel density within the tumor was seen after PRT+CDDP delivery (Figure 4A). Although the
EP
difference between PRT+CDDP and SRT+CDDP did not reach statistical significance (p=0.08), only PRT+CDDP was statistically indistinguishable from control values. This observation is
AC C
consistent with the higher vessel densities recorded in tumors treated with PRT compared with SRT in prior studies with the same UT-SCC-14 tumor model (15) and other xenograft models (13,14). Other experimental approaches have been developed to enhance drug efficacy, such as phosphoinositide-3 kinase (PI3K) to alter vascular structure (39), or the use of liposomal (40) or other encapsulated drugs (41) to enhanced release within the tumors. If validated, the use of PRT
Page 17 of 25
ACCEPTED MANUSCRIPT
to improve drug delivery could be an additional strategy that could be more rapidly translated to clinical use, since no additional pharmaceutical interventions are required.
RI PT
The data presented in Figure 1 demonstrate that the combination of the concurrent cisplatin and PRT results in superior tumor growth delay compared to the concurrent cisplatin and SRT. No significant difference was seen between SRT and PRT alone in the absence of systemic
SC
chemotherapy. While the precise reason for the improved efficacy of PRT+CDDP compared with SRT+CDDP remains to be elucidated, several possibilities exist: (1) improved drug delivery
M AN U
due to maintenance of the vascular network within the tumor after PRT as described above (2), improved radiosensitization of CDDP-induced DNA adducts after PRT compared with SRT, or (3) improved oxygen delivery via preserved vascular network leading to more oxygen-dependent cell killing after PRT compared with SRT. The most likely explanation remains the improved
TE D
delivery of drug from vascular network. However, to address other possibilities, we also investigated improved radiosensitization by assessing the repair of DNA adducts by NER. The data showing ECRR1 staining in tumor sections (Figure 4B) were not conclusive, and future
EP
studies are being planned to determine the absolute delivery of CDDP to irradiated tumors using
AC C
atomic absorption spectrophotometry (39,40).
Previously published studies using PRT for recurrent glioblastoma have demonstrated that PRT is clinically feasible, and patients have tolerated PRT treatment well (11). Volumetric modulated arc radiotherapy (VMAT) reduces treatment delivery time significantly and a PRT plan can be delivered in approximately 50 minutes (42). However, this length of treatment may still represent a challenge for head and neck patients as they near the completion of their treatment when the
Page 18 of 25
ACCEPTED MANUSCRIPT
toxicities of treatment are reaching a maximum. There may still be a benefit for the use of PRT during the initial weeks of radiotherapy followed by SRT. Current preclinical studies are ongoing testing the efficacy of combined SRT and PRT delivery. A significant advantage of PRT is that
RI PT
the technique could be implemented in that majority of radiation oncology facilities without any new equipment and with minimal additional training of therapy staff. Furthermore, based on our
SC
normal tissue data, the risk of increased toxicity or other adverse events is low.
The discovery of the prognostic importance of HPV in head and neck cancers in conjunction
M AN U
with other prognostic factors, such as stage and tobacco use, has provided a rationale for deescalation of treatment. However, these factors also allow for the identification of patients with a poor prognosis who may benefit from intensification of treatment. As the current standard treatment of locally advanced head and neck cancer has significant acute and chronic toxicities,
TE D
we must investigate novel approaches to improving outcomes in these high risk patients that do not subject our patients to additional treatment related toxicity. The findings presented suggest that the use of concurrent cisplatin with PRT may represent a novel treatment options for patients
EP
with high risk locally advanced head and neck squamous cell carcinoma. A primary potential
AC C
barrier to the clinical implementation of PRT is the increased treatment time compared to SRT.
Page 19 of 25
ACCEPTED MANUSCRIPT
FIGURE LEGENDS
Figure 1: Normalized tumor volumes as a function of time from initiation on treatment
RI PT
(Mean±SEM, n=7-9 per group). Radiation treatments began on Day 0. Negative values on the abscissa indicate tumor measurements made prior to initiation of treatment. Data are censored
SC
when the group sizes reached n=5.
Figure 2A: Time to reach 1000mm3 tumor volume as a function of time (Mean±SEM) after
M AN U
implantation. 2B: Kaplan-Meier plot indicating survival after the 6 treatment regimens.
Figure 3: FMISO images from representative animal in each treatment group (left) and quantified as Normalized SUVmax (right, mean±SD, n=3 per group). SUVmax values are
TE D
normalized to pre-treatment levels.
Figure 4: Representative images for H&E, CD34, ERRC1 and pimonidazole staining for each
EP
treatment group containing CDDP, and quantified using automated blinded image analysis.
AC C
Figure 5: Quantification of normal tissue damage in jejunum and skin. Upper panels are representative sections illustrating the assay methodology, lower panels show comparison between each group.
Page 20 of 25
ACCEPTED MANUSCRIPT
ACKNOWLEDGEMENTS
This work was funded by the Department of Radiation Oncology, XXXX and XXXX seed grant
RI PT
awarded to Dr. XXXX. The authors thank XXXX for technical expertise and XXXX for
AC C
EP
TE D
M AN U
SC
assistance with pathology.
Page 21 of 25
ACCEPTED MANUSCRIPT
REFERENCES 1. Mockelmann N, Kriegs M, Lorincz BB, et al. Molecular targeting in combination with platinumbased chemoradiotherapy in head and neck cancer treatment. Head Neck 2015.
RI PT
2. Ferreira MB, Lima JP Cohen EE. Novel targeted therapies in head and neck cancer. Expert Opin Investig Drugs 2012;21:281-295.
3. Baschnagel AM, Wobb JL, Dilworth JT, et al. The association of (18)F-FDG PET and glucose
carcinomas. Radiother Oncol 2015;117:118-124.
SC
metabolism biomarkers GLUT1 and HK2 in p16 positive and negative head and neck squamous cell
M AN U
4. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature 2015;517:576-582.
5. Ang KK, Harris J, Wheeler R, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med 2010;363:24-35.
6. Rischin D, Young RJ, Fisher R, et al. Prognostic significance of p16INK4A and human
TE D
papillomavirus in patients with oropharyngeal cancer treated on TROG 02.02 phase III trial. J Clin Oncol 2010;28:4142-4148.
7. Nordsmark M, Bentzen SM, Rudat V, et al. Prognostic value of tumor oxygenation in 397 head and
2005;77:18-24.
AC C
8. XXX
EP
neck tumors after primary radiation therapy. An international multi-center study. Radiother Oncol
9. XXX 10. XXX
11. Adkison JB, Tome W, Seo S, et al. Reirradiation of large-volume recurrent glioma with pulsed reduced-dose-rate radiotherapy. Int J Radiat Oncol Biol Phys 2011;79:835-841. 12. Rasmussen KH, Hardcastle N, Howard SP, et al. Reirradiation of glioblastoma through the use of a reduced dose rate on a tomotherapy unit. Technol Cancer Res Treat 2010;9:399-406. 13. XXX Page 22 of 25
ACCEPTED MANUSCRIPT
14. XXX 15. XXX 16. Adelstein DJ, Li Y, Adams GL, et al. An intergroup phase III comparison of standard radiation
RI PT
therapy and two schedules of concurrent chemoradiotherapy in patients with unresectable squamous cell head and neck cancer. J Clin Oncol 2003;21:92-98.
17. Dewit L. Combined treatment of radiation and cisdiamminedichloroplatinum (II): a review of
SC
experimental and clinical data. Int J Radiat Oncol Biol Phys 1987;13:403-426.
18. Oh SJ, Chi DY, Mosdzianowski C, et al. Fully automated synthesis of [18F]fluoromisonidazole using
M AN U
a conventional [18F]FDG module. Nucl Med Biol 2005;32:899-905.
19. Dijt FJ, Fichtinger-Schepman AM, Berends F, et al. Formation and repair of cisplatin-induced adducts to DNA in cultured normal and repair-deficient human fibroblasts. Cancer Res 1988;48:6058-6062.
20. Larminat F Bohr VA. Role of the human ERCC-1 gene in gene-specific repair of cisplatin-induced
TE D
DNA damage. Nucleic Acids Res 1994;22:3005-3010.
21. Hao D, Lau HY, Eliasziw M, et al. Comparing ERCC1 protein expression, mRNA levels, and genotype in squamous cell carcinomas of the head and neck treated with concurrent chemoradiation
EP
stratified by HPV status. Head Neck 2012;34:785-791. 22. Vaezi A, Wang X, Buch S, et al. XPF expression correlates with clinical outcome in squamous cell
AC C
carcinoma of the head and neck. Clin Cancer Res 2011;17:5513-5522. 23. Kaanders JH, Wijffels KI, Marres HA, et al. Pimonidazole binding and tumor vascularity predict for treatment outcome in head and neck cancer. Cancer Res 2002;62:7066-7074. 24. Yaromina A, Zips D, Thames HD, et al. Pimonidazole labelling and response to fractionated irradiation of five human squamous cell carcinoma (hSCC) lines in nude mice: the need for a multivariate approach in biomarker studies. Radiother Oncol 2006;81:122-129.
Page 23 of 25
ACCEPTED MANUSCRIPT
25. Allan SG, Smyth JF, Hay FG, et al. Protective effect of sodium-2-mercaptoethanesulfonate on the gastrointestinal toxicity and lethality of cis-diamminedichloroplatinum. Cancer Res 1986;46:35693573.
in tumor bearing mice. BMC Complement Altern Med 2013;13:103.
RI PT
26. Son TG, Gong EJ, Bae MJ, et al. Protective effect of genistein on radiation-induced intestinal injury
27. Folkman J. The role of angiogenesis in tumor growth. Semin Cancer Biol 1992;3:65-71.
SC
28. Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971;285:1182-1186.
29. Thomlinson RH Gray LH. The histological structure of some human lung cancers and the possible
M AN U
implications for radiotherapy. Br J Cancer 1955;9:539-549.
30. Trotter MJ, Chaplin DJ, Durand RE, et al. The use of fluorescent probes to identify regions of transient perfusion in murine tumors. Int J Radiat Oncol Biol Phys 1989;16:931-934. 31. Li C, Huang S, Armstrong EA, et al. Antitumor Effects of MEHD7945A, a Dual-Specific Antibody against EGFR and HER3, in Combination with Radiation in Lung and Head and Neck Cancers. Mol
TE D
Cancer Ther 2015;14:2049-2059.
32. Ma J Waxman DJ. Combination of antiangiogenesis with chemotherapy for more effective cancer treatment. Mol Cancer Ther 2008;7:3670-3684.
EP
33. Fukumura D Jain RK. Tumor microvasculature and microenvironment: targets for anti-angiogenesis and normalization. Microvasc Res 2007;74:72-84.
AC C
34. Kozin SV, Boucher Y, Hicklin DJ, et al. Vascular endothelial growth factor receptor-2-blocking antibody potentiates radiation-induced long-term control of human tumor xenografts. Cancer Res 2001;61:39-44.
35. McGee MC, Hamner JB, Williams RF, et al. Improved intratumoral oxygenation through vascular normalization increases glioma sensitivity to ionizing radiation. Int J Radiat Oncol Biol Phys 2010;76:1537-1545. 36. Jain RK. Normalizing tumor vasculature with anti-angiogenic therapy: a new paradigm for combination therapy. Nat Med 2001;7:987-989. Page 24 of 25
ACCEPTED MANUSCRIPT
37. Dings RP, Loren M, Heun H, et al. Scheduling of radiation with angiogenesis inhibitors anginex and Avastin improves therapeutic outcome via vessel normalization. Clin Cancer Res 2007;13:33953402.
RI PT
38. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 2005;307:58-62.
39. Qayum N, Im J, Stratford MR, et al. Modulation of the tumor microvasculature by phosphoinositide-3
SC
kinase inhibition increases doxorubicin delivery in vivo. Clin Cancer Res 2012;18:161-169.
40. Hagtvet E, Roe K Olsen DR. Liposomal doxorubicin improves radiotherapy response in hypoxic
M AN U
prostate cancer xenografts. Radiat Oncol 2011;6:135.
41. Eggen S, Afadzi M, Nilssen EA, et al. Ultrasound improves the uptake and distribution of liposomal Doxorubicin in prostate cancer xenografts. Ultrasound Med Biol 2013;39:1255-1266. 42. Tyagi N, Yang K, Sandhu R, et al. External beam pulsed low dose radiotherapy using volumetric
AC C
EP
TE D
modulated arc therapy: planning and delivery. Med Phys 2013;40:011704.
Page 25 of 25
Figure 1
RI PT
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
p=0.429
p=0.033
AC C
EP
p=0.045
p=0.035
ACCEPTED MANUSCRIPT
RI PT
Figure 2
AC C
EP
TE D
M AN U
SC
p=0.013
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 3
p=0.025 p=0.003
p=0.012
H&E
CD34
ERRC1
PIMO
p=0.047 p=0.011
TE D
AC C
CD34
B
EP
A
ERCC1
#187
M AN U
PRT+CDDP SRT+CDDP
#193
SC
CDDP
#200
RI PT
#199
Control
Figure 4
ACCEPTED MANUSCRIPT
C
PIMO
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 5