DCE-MRI biomarkers for monitoring an anti-angiogenic triple combination therapy in experimental hypopharynx carcinoma xenografts with immunohistochemical validation

Abstract

Background

Novel anti-angiogenic treatments are increasingly complementing established cancer therapy strategies in head and neck tumors. Contrast-enhanced magnetic resonance imaging (MRI) can be applied for early and non-invasive therapy monitoring by non-invasive quantitative assessment of tumor microcirculation as in vivo imaging biomarkers of therapy response.

Purpose

To monitor the anti-angiogenic effects of a novel combination therapy on experimental head and neck squamous cell carcinomas (HNSCC) with dynamic contrast-enhanced (DCE)-MRI.

Material and Methods

Athymic rats (n = 18) with subcutaneous HNSCC xenografts were investigated by DCE-MRI before and after 7 days of a daily triple therapy regimen combining the COX-II-inhibitor celecoxib, the matrix-metalloproteinase-inhibitor GM6001, and the uPA-inhibitor upamostat. Quantitative measurements of tumor blood flow (tBF), tumor blood volume (tBV), and permeability-surface area product (PS) were calculated and validated by immunohistochemistry.

Results

Mean tBF and tBV in triple-therapy animals decreased significantly from day 0 to day 7 (tBF, 41.0 ± 14.2 to 20.4 ± 5.7 mL/100 mL/min; P < 0.01; tBV, 17.7 ± 3.9 to 7.5 ± 3.3%; P < 0.01). No significant effects on PS were observed in either group (P > 0.05). Immunohistochemical analysis showed a significantly lower tumor vascularity in the therapy group than in the control group (CD31), significantly fewer Ki-67+ proliferating tumor cells and significantly more Capase-3+ apoptotic tumor cells (P < 0.05). Significant (P < 0.05) correlations were observed between tBF/tBV and CD31 (tBF, r = 0.84; tBV, r = 0.70), tBV and Ki-67 (r = 0.62), as well as tBF and caspase-3 (r = −0.64).

Conclusion

DCE-MRI may be a suitable tool for the non-invasive monitoring of the anti-vascular effects of this innovative triple therapy regimen with potential for clinical translation.

Keywords

Dynamic contrast-enhanced MRI therapy monitoring upamostat celecoxic GM6001 hypopharyngeal squamous cell carcinoma

Introduction

Despite recent advances in surgery and radiochemotherapy, survival rates in patients with head and neck squamous cell cancer (HNSCC) have only marginally improved over the past four decades. Therefore, novel therapeutic strategies with greater efficacy and improved tolerability are being investigated, particularly for those patients not eligible for surgery or radiochemotherapy. Preclinical and clinical studies have highlighted the central role of the vascular endothelial growth factor (VEGF)-cascade and other pro-angiogenic pathways in mediating tumor growth, vasculature, invasiveness, and radioresistance of HNSCC (1,2). An array of agents with interference in the angiogenic cascade is currently being tested for therapeutic efficiency in HNSCC patients in phase II–III studies – with variable clinical outcomes (3,4).

Due to multiple links between tumor angiogenesis, tumor growth, and COX-(Cyclooxygenase)-2 expression, selective pharmacological inhibition of COX-2 represents a promising target for the treatment of malignant solid tumors and has already demonstrated potent anti-angiogenic and anti-proliferative effects in different tumor models (5). COX-2 is an inducible enzyme that plays an important role in tumorigenesis and has been shown to be upregulated in many carcinomas (5). The degradation of the surrounding extracellular matrix (ECM) via matrix metalloproteinases (MMP) as well as via the urokinase-type plasminogen activator system (uPA) is a critical prerequisite for both intra- and extravasation of tumor cells and their systemic dissemination as well as angiogenesis (6,7). Combined inhibitory efforts targeting the MMPs and uPA-system have yielded anti-angiogenic and anti-invasive synergies in preclinical HNSCC tumor studies (8,9) underscoring their potential as combined molecular anticancer strategies in the treatment of HNSCC. Among these is a promising novel triple therapy regime consisting of the combined administration of the COX-II-inhibitor celecoxib, the matrix metalloproteinase-inhibitor GM6001 and the novel uPA-inhibitor upamostat with demonstrated, profound anti-angiogenic and anti-invasive effects on head and neck squamous cell carcinoma and cervical carcinoma cells in vitro (9).

Several preclinical and clinical studies highlight the potential role of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) with gadolinium-based contrast medium for the non-invasive assessment of anti-angiogenic treatment strategies by means of functional parameters of tumor microcirculation serving as non-invasive functional imaging biomarkers of therapy response (10,11). Although DCE-MRI is already under investigation in clinical oncologic trials for the non-invasive assessment of therapy response (11), controversy still exists with particular regard to the pathophysiologic correlate of the perfusion parameters derived by DCE-MRI (12). Therefore, validation of non-invasive DCE-MRI findings by immunohistochemistry as gold standard is key in the assessment of the therapeutic efficacy of innovative anti-angiogenic therapeutic approaches.

We hypothesized that DCE-MRI would allow for a non-invasive, in vivo monitoring of possible anti-angiogenic, anti-proliferative, and pro-apoptotic effects of the in vitro developed triple combination therapy. Therefore, the purpose of our study was the in vivo translation of this newly established triple combination therapy (TT) in a subcutaneous model of human hypopharynx cancer in rats with concurrent evaluation of DCE-MRI imaging biomarkers for non-invasive therapy monitoring with immunohistochemical validation.

Material and Methods

Animal model and experimental protocol

The study was performed with the approval of the Institutional Committee for Animal Research in accordance with the guidelines of the National Institute of Health for the care and use of laboratory animals. A total of 7–8-week-old nude athymic rats (n = 18, Charles River, Sulzfeld, Germany) were subcutaneously injected with 6 × 10⁶ human hypopharyngeal squamous cell carcinoma cells (FaDu) into the left abdominal flank. Daily animal inspection was conducted for the assessment of general appearance and tumor growth. According to established standards, tumor size was assessed by caliper measurements along the longest two dimensions of the tumor x/y plane only (13). When tumors reached a size of approximately 100 mm², animals were randomly assigned to either the therapy group (n = 9) or the control group (n = 9).

The therapy group received a combined, weight-adapted administration of the COX-II-inhibitor celecoxib (Celebrex®, 25 mg/kg, Pfizer, Berlin, Germany), the matrix metalloproteinase-inhibitor GM6001 (Galardin®, 50 mg/kg; US Biological, Salem, MA, USA), and the uPA-inhibitor upamostat (Mesupron®, 0.03 mg/kg; Wilex Inc., Munich, Germany) suspended in a total volume of 0.3 mL (0.2 ethanol + 0.1 mL H₂O) daily via gastric gavage using a dedicated curved buttoned cannula. The control group received daily volume-equivalent applications of the solvent solution only. DCE-MRI was performed at baseline and on day 7 following the 1-week daily therapy protocol. For MRI examinations, animals were anesthetized by intraperitoneal injections of ketamine (Inresa Arzneimittel, Freiburg, Germany) 100 mg/kg bodyweight) and xylazine (Bayer, Leverkusen, Germany) 10 mg/kg bodyweight) and a 25-gauge butterfly tail vein catheter was placed for subsequent contrast media administration. Following the day 7 MRI scan, animals were sacrificed by intravenous injections of 1 mL of the Ketamine/Xylazin combination. Tumors were excised for subsequent immunohistochemical analysis.

DCE-MRI

DCE-MRI was performed on a clinical 3 Tesla system (Magnetom Verio, Siemens Healthcare, Erlangen, Germany) with animals in supine position, using a dedicated 4-channel small animal coil (RAPID MR International, Columbus, OH, USA). MRI acquisition was performed according to an established imaging protocol (14). In brief, a fast view-sharing 3D gradient-recalled echo time-resolved angiography with stochastic trajectories sequence (15) (TWIST, Siemens Healthcare) was optimized for fast bolus tracking and acquisition of the baseline, bolus passage, and contrast agent uptake and washout with high temporal resolution. A total of 300 3D data-sets were acquired, resulting in a total time of acquisition of 10 min 13 s. Sequence details were: TR, 6.34 ms; TE, 2.11 ms; α, 40°; matrix size, 128 × 128; field of view, 50 × 50 mm²; spatial resolution, 0.39 × 0.39 × 3.0 mm³; 72 mm coverage along the z-axis. The high-resolution 3D GRE sequence was continued during and after the standardized manual bolus injection of 0.1 mmol/kg bodyweight of gadobutrol (Gadovist®, Bayer Healthcare, Leverkusen, Germany) followed by a 0.5 mL saline bolus, in order to follow contrast enhancement kinetics with high sampling rate.

MRI data processing and kinetic analysis

Data were postprocessed on an external workstation using the software PMI 0.4 (16), written in-house in IDL 6.4 (ITT Visual Information Solutions, Boulder, CO, USA). An arterial region of interest (ROI) was drawn within the lumen of the abdominal aorta and a tissue ROI was placed in four adjacent slices over tumor periphery (Fig. 1). The tumor periphery was defined as the outer 3 mm rim of the tumor, a region representative of viable tumor tissue, less affected by elevated interstitial pressure and necrosis that may occur in the tumor center (17). Signal intensity (SI) vs. time curves were extracted for both ROIs, and tracer concentrations were approximated with the relative enhancement (S/S₀-1, where S₀ is the signal intensity before arrival of the contrast agent). A two-compartment exchange model was applied for fitting the DCE-MRI data-sets. Four independent model parameters were calculated by fitting the tissue curve to a two-compartment exchange model (18,19): tumor blood flow (tBF) (mL/100 mL/min) as a parameter of tumor perfusion, tumor blood volume (tBV) (%) as a parameter of tumor vascularity, the endothelial permeability surface-area product (PS) (mL/min/100 mL) as a measure of endothelial permeability and the extracellular, extravascular volume.

Fig. 1.

Representative set of DCE-MRI images of a rat bearing a subcutaneous FaDu-xenograft (arrow) at baseline. The upper panel shows a representative set of T1-weighted images before contrast agent administration as well as 2, 4, and 8 min after intravenous contrast media injection (from left to right). Note the intravascular enhancement of the inferior vena cava, the abdominal aorta (arrowhead), and the left kidney. Images in the lower panel illustrate a time-resolved signal-enhancement curve (picture in the middle) of ROI placed over the tumor rim (left picture). Representative arterial input function (right picture in the lower panel) for calculation of DCE-MRI tumor perfusion parameters was derived by placing a ROI in the abdominal aorta.

Immunohistochemistry

Consecutive cryosections (4 µm) of each tumor were fixed in acetone (10 min) and incubated in H₂O₂ (10 min, 0.03%) to block endogenous peroxidase activity. Subsequently, slides were incubated with either mouse anti-rat CD31 (1:100, 2 h, Becton Dickinson, Heidelberg, Germany), rabbit anti-human Caspase-3 (1:50; 2 h, Cell Signaling, Boston, MA, USA), mouse anti-human EpCAM VU1D9 (1:2000; 1 h, Cell Signaling, Boston, MA, USA) or mouse anti-human Ki67 (1:800; 1 h, Dako, Hamburg, Germany). Thereafter, sequential incubations with biotinylated anti-mouse rat adsorbed for CD31, EpCAM, and Ki-67, anti-rabbit secondary antibody for caspase-3 and peroxidase-labeled avidin-biotin-peroxidase complex stainings (Vector Lab Inc., Burlingame, CA, USA) were conducted. Amino-ethyl-carbazole (AEC) peroxidase substrate (Sigma, St. Louis, MO, USA) was used for the detection of antigen/antibody complexes indicated by red-brown staining. Counter-staining was achieved with hematoxylin (Sigma). Negative controls were conducted simultaneously using respective mouse/rabbit isotype control antibody (Cell Signaling, Boston, MA, USA). Finally, sections were mounted in Kaiser’s glycerol gelatine for subsequent analysis. Proliferation and apoptosis rate were measured as the percentage of Ki67- and caspase-3-positive cells among all tumor cells in a total of 40 fields (0.04 mm²) in the tumor periphery at a magnification of 200 taken from 10 randomly picked slices. Likewise, tumor vascularization was measured as the amount of CD31-positive vessels in the tumor in total of 40 fields (0.04 mm²) in the tumor periphery at a magnification of 200 taken from 10 randomly picked slices.

Statistical analysis

Continuous variables are presented as mean and standard deviation. DCE-MRI and immunohistochemistry values as well as tumor size between the treatment and the control group were compared using the unpaired, non-parametric Mann-Whitney U test. Analyses were carried out in SPSS for windows (version 11.5, SPSS Inc. Chicago, IL, USA). Relationships between MRI and immunohistochemistry were evaluated using Pearson’s correlation coefficients with a Bonferroni correction for multiple comparisons. P values <0.05 were considered statistically significant.

Results

Animals tolerated all procedures well with no adverse effects noted. DCE-MRI measurements with concomitant histological assessment were successfully performed without any technical difficulties in all 18 rats.

Tumor size

In the triple-therapy-treated group, tumor growth was significantly delayed (P < 0.05) compared to the control group. In the treatment group mean tumor size increased slightly but non-significantly (P > 0.05) from baseline 1.2 ± 0.1 cm² to 1.6 ± 0.2 cm² on day 7, whereas in the control group tumor size increased significantly (P < 0.05) from 1.1 ± 0.1 cm² at baseline to 2.7 ± 0.4 cm² (Fig. 2).

Fig. 2.

A 1-week course of daily administration of celecoxib, GM6001, and upamostat significantly inhibited tumor growth of FaDu xenografts in athymic female nude rats. Therapy regimen (n = 9) or solvent treatment (n = 9) was initiated after tumors reached a size of 100 mm². Data are presented as the mean ± SEM for the number of animals indicated for each group.

DCE-MRI

In the treatment group, mean tBF decreased significantly (P < 0.001) between day 0 and day 7 (41.0 ± 14.2 to 20.4 ± 5.7 mL/100 mL/min; Fig. 3) with a uni-directional decline of all values. In the control group mean tBF between day 0 and day 7 (40.8 ± 12.7 to 39.7 ± 12.2 mL/100 mL/min, Fig. 3) virtually remained unchanged (P = 0.82). Differences of mean tBF quantified by DCE-MRI were highly significant (P < 0.001) on day 7 between the therapy and the control group. With regard to tumor vascularity, a significant (P = 0.016) decline of mean tBV was observed in the therapy group between baseline and day 7 (17.7 ± 3.9 to 7.5 ± 3.3%). No significant changes (P = 0.31) of tumor vascularity were observed in the control group between baseline and day 7 (14.2 ± 6.4 to 12.8 ± 5.0%). No significant change (P > 0.05) of mean PS values, as a parameter of endothelial permeability, was observed in the treatment or the control group with omni-directional development of individual values between baseline and day 7 (Fig. 3). Individual values of MRI perfusion parameters are presented in Table 1.

Fig. 3.

Graphs in the upper row show significant and uni-directional decline of tBF in all tumors of the therapy group (P < 0.001) between baseline and follow-up MRI on day 7. Graphs in the lower row show no significant effect (P > 0.05) of triple therapy on tumor permeability-surface area-product (PS) with similar omni-directional development of individual values between baseline and day 7 as observed in control animals.

Table 1.

Individual values of MRI-derived parameters of tumor microcirculation (tumor blood flow, permeability surface-area product, tumor blood volume) are depicted for the therapy and the control group at baseline and after 7 days of therapy. Note the significant (P < 0.05) decline of tumor blood flow and tumor blood volume in therapy group. No significant changes (P > 0.05) were observed on permeability surface-area product in the therapy group, or on tumor microcirculation in the control group.

Therapy group	Blood flow (mL/100 mL/min)		Permeability-surface-area-product (mL/100 mL/min)		Blood volume (mL/100 mL)
Rat No.	Baseline	Day 7	Baseline	Day 7	Baseline	Day 7
1	66.0	20.1	3.2	2.4	16.3	5.6
2	48.9	28.6	2.8	2.3	10.2	7.3
3	25.0	14.7	3.1	2.4	15.3	4.3
4	35.2	22.0	4.2	4.3	22.1	6.8
5	22.1	16.2	4.5	3.9	22.0	8.4
6	38.0	33.3	5.5	11.6	19.5	14.0
7	32.7	15.3	12.9	7.7	20.4	9.0
Mean ± SD	38.3 ± 15	20.2 ± 7.5*	5.1 ± 3.5	7.9 ± 3.1	17.9 ± 4.3	7.9 ± 3.1*
Control group	Blood flow (mL/100 mL/min)		Permeability-surface-area-product (mL/100 mL/min)		Blood volume (mL/100 mL)
Rat No.	Baseline	Day 7	Baseline	Day 7	Baseline	Day 7
1	47.9	50.2	5.5	11.8	10.3	33.1
2	58.2	46.8	11.5	4.4	21.7	16.2
3	29.1	45.4	7.5	8.0	11.4	7.7
4	49.3	53.5	6.1	7.7	12.9	7.8
5	29.0	23.8	4.1	7.6	9.6	12.0
6	44.2	47.2	3.9	5.2	5.7	8.9
7	20.6	19.0	5.7	3.9	15.0	11.2
Mean ± SD	41.3 ± 13.4	40.8 ± 13.6	6.3 ± 2.6	7.0 ± 2.7	14.2 ± 6.9	15.7 ± 9.9

Significant (P < 0.05).

Immunohistochemical analysis

Immunohistochemical analysis revealed significantly lower tumor vascularity in the therapy group with significantly (P < 0.001) less CD31-positive endothelial cells on tumor sections than in the control group (CD31 644 ± 224 vs. 1265 ± 273). A moderate but significant anti-proliferative effect was observed in the therapy group with a significantly lower percentage of Ki-67-positive cells than in the control group (17.2 ± 3.7% vs. 24.1 ± 5.5%; P = 0.006). A moderate but significant pro-apoptotic effect was noted in the therapy group with a significantly (P = 0.008) higher percentage of caspase3-positive cells in tumor xenografts treated with the triple combination therapy than in the control group (2.2 ± 0.5% vs. 1.4 ± 0.6%). Complete immunohistochemical data on tumor vascularity, tumor cell proliferation, and apoptosis is presented in Table 2. Representative microscopic images of tumor sections stained for CD31, Ki-67 or Caspase-3 are shown in Fig. 4.

Fig. 4.

Immunohistochemical analysis revealed significant (P < 0.05) anti-angiogenic effects of the 1-week triple combination therapy on subcutaneous FaDu xenografts with significantly less CD31-positive cells (stained in brown) in the therapy than in the control group. Inhibition of tumor cell proliferation is demonstrated by significantly less Ki-67-positive cells in the therapy group than in the control group (stained in brown). Significant pro-apoptotic effects of the triple therapy with significantly more Caspase-3-positive tumor cells (stained in red) can be observed in the therapy than in the control group.

Table 2.

Individual immunohistochemical results for tumor microvascular density (MVD), proliferating tumor cells (Ki-67), and apoptotic tumor cells (Caspase-3) in the therapy and the control group. Note the significantly (P < 0.05) lower tumor vascularity in the therapy compared to the control group with correspondingly significantly less proliferating and significantly more apoptotic tumor cells on tumor sections.

Therapy group Rat No.	MVD	Ki-67 (%)	Caspase-3 (%)
1	1139	16.3	2.3
2	639	13.5	1.9
3	525	22.2	2.4
4	629	14.1	2.3
5	844	16.4	2.4
6	1037	25.0	2.5
7	513	2.0	20.6
Mean ± SD	760 ± 250	18.3 ± 4.3	2.3 ± 0.2
Control group Rat No.	MVD	Ki-67 (%)	Caspase-3 (%)
1	1348	19.8	1.3
2	1435	31.1	1.7
3	1162	25.8	1.8
4	1422	18.4	1.8
5	1356	24.3	0.8
6	1110	23.5	0.9
7	694	25.5	2.5
Mean ± SD	1218 ± 362*	24.1 ± 4.2*	1.5 ± 0.6*

Significant (P < 0.05).

Correlation between DCE-MRI parameters and immunohistochemistry

Significant correlations were noted between tBF and immunohistochemical tumor vascularity (r = 0.84; P < 0.001), between tBV and tumor vascularity (r = 0.70; P = 0.002), and between tBF and tumor cell apoptosis (r = –0.64; P = 0.005). A moderate but significant positive correlation was noted between tBV and immunohistochemical tumor cell proliferation (r = 0.62; P = 0.006). Complete Pearson’s correlations between MRI perfusion parameters and immunohistochemical results are listed in Table 3.

Table 3.

Correlation (Pearson’s r) of tumor blood flow (tBF), tumor blood volume (tBV), and tumor endothelial permeability-surface area product (PS) quantified by DCE-MRI with immunohistochemical measurements of tumor vascularity (CD31), tumor cell proliferation (Ki-67), and tumor cell apoptosis (Caspase-3). Statistically significant (P < 0.05) correlations between DCE-MRI parameters and immunohistochemical measurements are printed in bold. Statistically significant correlations were found between tBF/tBV and tumor vascularity (CD31), tBV and tumor cell proliferation (Ki-67) whereas a statistically significant inverse correlation was found between tBF and apoptosis (Caspase-3).

Correlation	Pearson’s r	P
tBF/Ki-67	0.47	0.075
tBF/Caspase-3	−0.62	0.014
tBF/CD31	0.78	0.001
tBV/Ki-67	0.63	0.012
tBV/Caspase-3	−0.31	0.25
tBV/CD31	0.55	0.031

Discussion

Phase-II clinical studies investigating different multityrosinekinase inhibitors as anti-angiogenic monotherapy regimens in patients with locally advanced/metastasized HNSCC reported only partial benefits with mainly poor response rates (4,20). Therefore, the development of combination therapies with different compounds targeting multiple aspects of the angiogenic and the pro-metastatic cascade in tumors may be a promising approach to take advantage of possible synergistic and additive effects. In the current study, the effects of an innovative triple combination therapy consisting of the COX-II-inhibitor celecoxib, the matrix metalloproteinase inhibitor GM6001, and the uPA-inhibitor upamostat were investigated by DCE-MRI in a subcutaneous model of human hypopharynx squamous cell carcinoma with correlations to an immunohistochemical gold standard.

Immunohistochemical analysis revealed significant anti-angiogenic, anti-proliferative, and pro-apoptotic effects of the triple combination therapy on the investigated subcutaneous hypopharynx carcinoma xenografts. The observed anti-angiogenic effects confirm the observations of preceding in vitro studies, in which synergistic therapeutic inhibitory effects were observed not only on tissue-invasiveness but also on the angiogenic activity of FaDu cells (9). However, in contrary to the pro-apoptotic and anti-proliferative effects observed in the investigated tumor xenografts, no direct inhibitory effects of the triple combination therapy were noted on tumor cell proliferation or tumor cell apoptosis in vitro. Of the investigated agents, the anti-angiogenic effects of celecoxib are best characterized, which are based on the suppression of COX-II-mediated secretion of pro-angiogenic prostaglandins (21). The broad-spectrum MMP Inhibitor GM6001 has been shown to have anti-angiogenic effects on sarcoma cells whereas the anti-angiogenic potential of the uPA-inhibitor upamostat – currently under investigation for its anti-proliferative and anti-metastatic effects in several clinical phase I–II studies – has not been investigated so far. Our in vitro studies suggested a dominant anti-angiogenic role of celecoxib in our tumor model. However, given the fact that both GM6001 and upamostat influence tumor angiogenesis also indirectly, e.g. via VEGF release from cleaved ECM components (6,7), anti-angiogenic therapy synergies may unfold in our in vivo tumor model, not properly reflected in in vitro studies.

DCE-MRI has been demonstrated to be a sensitive tool for the non-invasive monitoring of various anti-angiogenic therapies via surrogate parameters of tumor microcirculation (11). Our study showed a significant decline of tumor blood flow, as a parameter of tumor perfusion, and blood volume, as a measure of tumor vascularity, after a 1-week treatment with the triple combination therapy, whereas no significant changes were noted in control animals. These results are consistent with previous studies, describing the effects of the multityrosinekinase inhibitor sorafenib in an experimental model of prostate cancer in rats and found that tumor perfusion and tumor vascularity declined significantly and uni-directionally under therapy monitored by DCE-MRI (14). Raatschen et al. (22) investigated the effects of the anti-VEGF antibody bevacizumab on experimental breast carcinoma xenografts using DCE-MRI and described significant effects on tumor endothelial permeability and tumor vascularity. On the contrary, we were not able to show significant effects of the triple combination therapy on tumor endothelial permeability with a non-uniform development of the individual values between baseline and day 7. However, in contrast to our study which used a clinically available small molecular contrast agent for enhancement, Raatschen et al. applied the experimental, macromolecular contrast medium albumin-(Gd-DTPA) which follows a different biodistribution profile with a more selective extravasation in angiogenically active tissues (22,23), but is not available for clinical use due to concerns of delayed bioelimination and potential immunogenicity. Fournier et al. (24) investigated the anti-angiogenic effects of celecoxib in a human breast cancer model (MDA-MB-231) using DCE-MRI with macromolecular contrast media and found a significant decline of the endothelial transfer coefficient K^PS after 1 week of treatment. No significant effects, however, were noted on fractional plasma volume as a parameter of tumor vascularity. Interestingly, in the corresponding immunohistochemical work-up of the tumor xenografts significantly lower tumor vascularity was found in the celecoxib-treated therapy group than in the control group. This demonstrates that DCE-MRI is able to generate useful surrogate parameters of tumor microcirculation and has a potential as imaging biomarkers of therapy response. The method has limitations, however, and still requires validation by immunohistochemistry.

In the current study, correlation analysis between DCE-MRI parameters of microcirculation and immunohistochemistry was performed for validation purposes. Significant correlations were observed between MRI-derived tumor blood flow and tumor blood volume with immunohistochemical tumor microvascular density (MVD) in the examined tumor regions. These findings are in accordance with other published studies, which reported on significant correlations between various DCE-MRI parameters and tissue perfusion (25) as well as MVD (26). However, controversial results are also reported in the literature (27). One reason for these discrepancies may be tumor heterogeneity with substantial intravoxel heterogeneity in morphology and tissue composition between DCE-MRI images and histological sections (28). We addressed the issue of tumor heterogeneity by averaging out the immunohistochemical findings over several fields randomly picked from the tumor periphery and matched to the ROI used for determination of DCE-MRI perfusion parameters in order to obtain more valid agreements between DCE-MRI values and tumor MVD. Furthermore, discrepancies between DCE-MRI perfusion parameters and MVD may arise from the fact that not all histologically assessed intratumoral microvessels are perfused at time of imaging. This may also cause a mismatch between the number of functional vessels – assessed by DCE-MRI – and the absolute number of intratumoral microvessels, which also varies with the selected tumor model. Modest but significant inverse correlations between MRI-derived tumor blood flow and immunohistochemical tumor cell apoptosis as well as between tumor blood volume and tumor cell proliferation were observed. This correlation of impaired tumor cell proliferation in areas with reduced tumor vascularity – e.g. reduced blood supply – is in good agreement with studies by Tyninnen and Aronen who described associations between tumor vascularity and proliferation in human gliomas (29). Accordingly, the inverse correlation between tumor perfusion and tumor cell apoptosis may be attributed to increased apoptosis rate in pre-necrotic areas with reduced blood supply (30). These findings support the hypothesis that therapy-induced reductions of tumor vascularity and perfusion are associated with increased tumor cell apoptosis and reduced proliferation in this HNSCC tumor model. Also, DCE-MRI-derived parameters of microcirculation do not only allow for the detection of changes in tumor perfusion as early as 7 days after therapy initiation, but also support the concept that these surrogate markers are reflected by therapy-induced changes of tumor vascularity, tumor cell proliferation, and apoptosis, as confirmed by immunohistochemistry. However, DCE-MRI parameters of tumor microcirculation are not only influenced by their complex physiologic composition but also rely on technical factors such as MRI protocols, postprocessing techniques, and the details of kinetic analysis (12). Because of a lack in standardization, the transferability of results between different institutions remains limited (11).

The results of our study are limited in several aspects. The triple therapy regimen was tested in only one cell line of HNSCC. Given the wide heterogeneity of HNSCC tumors, the observed strong anti-vascular effects on the investigated FaDu xenografts may not be transferable to other sub-entities of HNSCC. Furthermore, this study did not evaluate the anti-vascular effects of each single compound of the triple regimen in separate studies. Therefore, it remains unclear if the observed anti-angiogenic, anti-proliferative, and pro-apoptotic effects are in fact attributable to synergies between the investigated therapeutic agents.

In conclusion, the combined administration of COX-II-inhibitor celecoxib, the MMP-inhibitor GM6001, and the uPA-inhibitor upamostat had significant anti-angiogenic, anti-proliferative, and pro-apoptotic effects on experimental hypopharynx carcinoma xenografts in vivo, which were monitored non-invasively by DCE-MRI with modest to good correlations to an immunohistochemical gold standard. DCE-MRI may be a suitable tool for the non-invasive monitoring of this anti-angiogenic triple combination therapy in patients suffering from HNSCC. The method has potential for clinical translation pending standardization of MRI data acquisition and postprocessing protocols.

Footnotes

Funding

This study was supported by the German Federal Ministry of Education and Research (BMBF) Excellence Cluster M⁴ (01EX1021X) and a research grant from Bayer Healthcare. CCC serves as a speaker for Bayer Healthcare and receives research grants from Bayer Healthcare and Novartis.

References

Seiwert

Cohen

. Targeting angiogenesis in head and neck cancer. Semin Oncol 2008; 35: 274–285.

Tong

Lloyd

Pei

. Human head and neck squamous cell carcinoma cells are both targets and effectors for the angiogenic cytokine. J Cell Biochem 2008; 105: 1202–1210.

Cohen

Davis

Karrison

. Erlotinib and bevacizumab in patients with recurrent or metastatic squamous-cell carcinoma of the head and neck: a phase I/II study. Lancet Oncol 2009; 10: 2472–2457.

Machiels

Henry

Zanetta

. Phase II study of sunitinib in recurrent or metastatic squamous cell carcinoma of the head and neck: GORTEC 2006-01. J Clin Oncol 2010; 28: 21–28.

Ghosh

Chaki

Mandal

. COX-2 as a target for cancer chemotherapy. Pharmacol Rep 2010; 62: 233–244.

Bourboulia

Stetler-Stevenson

. Matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs): Positive and negative regulators in tumor cell adhesion. Semin Cancer Biol 2010; 20: 161–168.

Dass

Ahmad

Azmi

. Evolving role of uPA/uPAR system in human cancers. Cancer Treat Rev 2008; 34: 122–136.

Gondi

Rao

. Therapeutic potential of siRNA-mediated targeting of urokinase plasminogen activator, its receptor, and matrix metalloproteinases. Methods Mol Biol 2009; 487: 267–281.

Zengel

Mack

Zahler

. Multimodal therapy for synergic inhibition of tumour cell invasion and tumour-induced angiogenesis. BMC Cancer 2010; 10: 92–92.

10.

Zhu

Niu

Fang

. Preclinical molecular imaging of tumor angiogenesis. Q J Nucl Med Mol Imaging 2010; 54: 291–308.

11.

O'Connor

Jackson

Parker

. Dynamic contrast-enhanced MRI in clinical trials of antivascular therapies. Nature reviews Clinical oncology 2012; 9: 167–177.

12.

Barnes

Whisenant

Loveless

. Practical dynamic contrast enhanced MRI in small animal models of cancer: data acquisition, data analysis, and interpretation. Pharmaceutics 2012; 4: 442–478.

13.

Tomayko

Reynolds

. Determination of subcutaneous tumor size in athymic (nude) mice. Cancer Chemother Pharmacol 1989; 24: 148–154.

14.

Cyran

Paprottka

Schwarz

. Perfusion MRI for monitoring the effect of sorafenib on experimental prostate carcinoma: a validation study. Am J Roentgenol 2011; 198: 384–391.

15.

Song

Laine

Chen

. Optimal k-space sampling for dynamic contrast-enhanced MRI with an application to MR renography. Mag Reson Med 2009; 61: 1242–1248.

16.

Sourbron

Biffar

Ingrisch

. PMI: platform for research in medical imaging. Magn Reson Mater Phy 2009; 22: 539–539.

17.

Preda

Turetschek

Daldrup

. The choice of region of interest measures in contrast-enhanced magnetic resonance image characterization of experimental breast tumors. Invest Radiol 2005; 40: 349–354.

18.

Ingrisch

Sourbron

. Tracer-kinetic modeling of dynamic contrast-enhanced MRI and CT: a primer. J Pharmacokinet Pharmacodyn 2013; 40: 281–300.

19.

Sourbron

Buckley

. Tracer kinetic modelling in MRI: estimating perfusion and capillary permeability. Phys Med Biol 2012; 57: R1–33.

20.

Williamson

Moon

Huang

. A phase II trial of sorafenib in patients with recurrent and/or metastatic head and neck squamous cell carcinoma (HNSCC):A Southwest Oncology Group (SWOG) trial. J Clin Oncol 2007; 25: 6044–6044.

21.

Rizzo

. Cyclooxygenase-2 in oncogenesis. Clin Chim Acta 2010; 412: 671–687.

22.

Raatschen

Shames

. Magnetic resonance imaging enhancement of normal tissues and tumors using macromolecular Gd-based cascade polymer contrast agents: preclinical evaluations. Invest Radiol 2006; 41: 860–867.

23.

Cyran

Raatschen

. New macromolecular polymeric MRI contrast agents for application in the differentiation of cancer from benign soft tissues. J Magn Reson Imaging 2008; 27: 581–589.

24.

Fournier

Novikov

Lucidi

. MR monitoring of cyclooxygenase-2 inhibition of angiogenesis in a human breast cancer model in rats. Radiology 2007; 243: 105–111.

25.

Parkka

Niemi

Saraste

. Comparison of MRI and positron emission tomography for measuring myocardial perfusion reserve in healthy humans. Magn Reson Med 2006; 55: 772–779.

26.

Giesel

Choyke

Mehndiratta

. Pharmacokinetic analysis of malignant pleural mesothelioma-initial results of tumor microcirculation and its correlation to microvessel density (CD-34). Acad Rad 2008; 15: 563–570.

27.

Cheung

Fruehauf

. Correlation of dynamic contrast enhancement MRI parameters with microvessel density and VEGF for assessment of angiogenesis in breast cancer. J Magn Reson Imaging 2003; 18: 467–477.

28.

Yang

Knopp

. Quantifying tumor vascular heterogeneity with dynamic contrast-enhanced magnetic resonance imaging: a review. J Biomed Biotechnol 2011; 2011: 732848–732848.

29.

Tynninen

Aronen

Ruhala

. MRI enhancement and microvascular density in gliomas. Correlation with tumor cell proliferation. Invest Radiol 1999; 34: 427–434.

30.

Atkin

Daley

Stirling

. Dynamic contrast-enhanced magnetic resonance imaging is a poor measure of rectal cancer angiogenesis. Br J Surg 2006; 93: 992–1000.