Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2401050709
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McCarville, M. B.
Right arrow Articles by Davidoff, A. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McCarville, M. B.
Right arrow Articles by Davidoff, A. M.
Related Collections
Right arrowRelated Article
(Radiology 2006;240:73-81.)
© RSNA, 2006

Experimental Studies

Angiogenesis Inhibitors in a Murine Neuroblastoma Model: Quantitative Assessment of Intratumoral Blood Flow with Contrast-enhanced Gray-Scale US1

M. Beth McCarville, MD, Christian J. Streck, MD, Paxton V. Dickson, MD, Chin-Shang Li, PhD, Amit C. Nathwani, MD and Andrew M. Davidoff, MD

1 From the Departments of Radiological Sciences (M.B.M.), Surgery (C.J.S., P.V.D., A.M.D.), and Biostatistics (C.S.L.), St Jude Children's Research Hospital, 332 N Lauderdale St, Memphis, TN 38105-2794; Departments of Radiology (M.B.M.) and Surgery (C.J.S., P.V.D., A.M.D.), University of Tennessee–Memphis Health Science Center, Memphis, Tenn; and Department of Haematology, University College London, London, England (A.C.N.). Received April 27, 2005; revision requested June 22; revision received August 22; final version accepted September 22. Supported by grants from the Assisi Foundation of Memphis 94-000, American Cancer Society grant IRG-87-008-09, Cancer Center Support CORE grant P30 CA 21765, and American Lebanese Syrian Associated Charities. A.M.D. supported by a Young Investigator Research Award from the Alliance for Cancer Gene Therapy. Address correspondence to M.B.M. (e-mail: beth.mccarville{at}stjude.org).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To quantify intratumoral ultrasonographic (US) contrast agent flow at gray-scale imaging as a measure of functional tumor vascularity in an orthotopic murine neuroblastoma model treated with angiogenesis inhibitors.

Materials and Methods: After Institutional Animal Care and Use Committee approval, retroperitoneal neuroblastomas were established in mice with unmodified NXS2 cells (n = 13) or with cells engineered to overexpress an angiogenesis inhibitor—either tissue inhibitor of matrix metalloproteinase-3 (n = 22) or a truncated soluble form of the vascular endothelial growth factor receptor-2 (truncated soluble fetal liver kinase-1; n = 13). When tumors were approximately 600 mm3, contrast material–enhanced gray-scale US was performed, and the imaging was recorded on cine clips. Regions of interest within tumors were analyzed off-line to determine postcontrast change in signal intensity (SI) from baseline to initial peak ({Delta}SI), rate of SI increase from baseline to initial peak (RSI), and contrast material washout. The Mann-Whitney test was used to evaluate potential differences in these US parameters between treatment groups. The mean intratumoral endothelial cell (CD34) and pericyte (smooth muscle actin [SMA]) counts at immunohistochemical analysis were also evaluated. Spearman correlation test was used to investigate the relation between US parameters and these histologic markers.

Results: The {Delta}SI and RSI were lower in tumors overexpressing an angiogenesis inhibitor than in control tumors (all P < .03). Contrast material washout did not differ between groups. For the entire cohort, the RSI correlated with the immunohistochemical assessment of tumor vascularity (SMA and CD34 counts) (P < .003).

Conclusion: Quantification of intratumoral flow of a US contrast agent at gray-scale imaging shows promise for monitoring tumor vascular response to antiangiogenic therapy.

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Since 1971, when Judah Folkman first hypothesized that tumor growth depends on angiogenesis, considerable research has focused on validating this theory and applying it to preclinical models (1–3). However, the conventional measure of antitumor efficacy, tumor shrinkage, may not be the most appropriate endpoint for evaluating angiogenesis inhibitors, which typically prevent the development of new blood vessels needed to support tumor growth but may not destroy existing vessels. Angiogenesis inhibition, therefore, may restrict tumor microvasculature without a clinically apparent change in tumor size (4,5). The successful introduction of antiangiogenic therapies into cancer clinical trials will require the development of reliable methods of assessing angiogenesis and its modification or inhibition in vivo (6).

Currently, histologic determination of the mean intratumoral microvessel density is the most commonly used method for assessing angiogenesis (4). However, obtaining tissue for histologic evaluation may necessitate an invasive procedure that would normally not be required. Furthermore, determination of the microvessel density does not provide an accurate assessment of the functionality of tumor vessels because many poorly functioning or collapsed vessels have endothelial cells that are stained and counted. Therefore, although microvessel density is a potentially useful marker for assessing angiogenesis in tumors at diagnosis, determination of changes in microvessel density may not accurately reflect the effectiveness of antiangiogenic therapy (4).

Because there is a crucial need for noninvasive methods to evaluate angiogenesis in situ, several functional imaging modalities are being investigated, including dynamic contrast material–enhanced magnetic resonance (MR) imaging, computed tomography, positron emission tomography (PET), and ultrasonography (US) (5–7). Of these, US is a rapid, relatively inexpensive technique that, in experienced hands, can be easily used in the laboratory and clinic. It has the added advantages of not requiring sedation or exposing patients to ionizing radiation. Microbubble US contrast agents are composed of microspheres that contain a gas and are highly reflective on US images, even when very small doses are used. These agents offer a potential way to enhance visualization of tumor vasculature in small volumes of tissue and to monitor the early effects of antiangiogenic therapy on small tumor vessels, which are typically smaller than 50 µm in diameter (8).

To optimize visualization of US contrast agent flow into tumors, power Doppler US has been used. This method is hindered by blooming artifact and the high mechanical index necessary to perform Doppler US, which destroys some of the microbubble contrast agent. This technique generally relies either on the off-line analysis of the color pixel count from regions of interest within tumors on selected static images or on a subjective grading of intratumoral flow of contrast agent (9–12). The purpose of our study was to quantify intratumoral US contrast agent flow at gray-scale imaging as a measure of functional tumor vascularity in an orthotopic murine neuroblastoma model treated with angiogenesis inhibitors.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Industry Support
GE Healthcare (Waukesha, Wis) provided the contrast agent and Siemens Medical (Mountain View, Calif) provided software support for this project. The authors had control of the study data and information submitted for publication.

Cell Lines
The murine neuroblastoma cell line NXS2 was provided (R. Reisfeld, La Jolla, Calif) and was modified by using retroviral vectors encoding either tissue inhibitor of matrix metalloproteinase-3 (TIMP-3) or a truncated, soluble form of the vascular endothelial growth factor receptor-2 (truncated soluble fetal liver kinase-1 [tsFlk-1]), as previously described (13,14). The complementary DNA was provided (P. Lin, Durham, NC) for the truncated, soluble vascular endothelial growth factor receptor. The plasmid (pBLAST-mTIMP-3) containing complementary DNA–encoding murine TIMP-3 was purchased from InvivoGen (San Diego, Calif).

Murine Tumor Model Cohorts
Localized, retroperitoneal neuroblastoma was established by injecting 1.5 x 106 NXS2 cells posterior to the left adrenal gland of CB-17 severe combined immunodeficiency mice (Charles River Laboratories, Wilmington, Mass) via a left subcostal incision during administration of 2% isoflourane. On the basis of prior experience, we expected that the growth of tumors with enforced expression of TIMP-3 or tsFlk-1 would be significantly less than that of unmodified tumors (13,14). Therefore, to evaluate tumors of similar size, we performed US 2 weeks after implantation of tumor cells expressing TIMP-3 and tsFlk-1 and 1 week after implantation of unmodified cells.

The contrast-enhanced US imaging technique was initially tested in a cohort of five mice with TIMP-3–expressing tumors and five mice with unmodified NXS2 tumors. We then imaged an additional 47 mice that had unmodified tumors (n = 14) or tumors expressing TIMP-3 (n = 20) or tsFlk-1 (n = 13). This cohort included a subset of 15 mice (five from each treatment group) that were imaged separately and after the larger cohort to test the reproducibility of our analysis. To detect potential changes in tumor vascularity over time, we imaged all three treatment groups in this later subset at 1 and 2 weeks after tumor implantation by using both Doppler and contrast-enhanced region-of-interest analysis.

Of the 57 mice imaged, six NXS2 control mice and three TIMP-3 mice were excluded from analysis: two because no tumor was identified at US, four because contrast agent injections were inadequate, and three because slight movement of the tumor precluded an accurate region-of-interest analysis. The final cohort consisted of 48 mice (35 from the initial two cohorts and 13 from the subset when the study was repeated to assess reproducibility) and was composed of 13 mice with unmodified tumors, 25 with TIMP-3–expressing tumors, and 10 with tsFlk-1–expressing tumors. After US, all mice were sacrificed and tumors were excised. Caliper measurements of tumor volume in 25 mice were compared with US volume measurements. These studies were performed with a protocol approved by the Institutional Animal Care and Use Committee at St Jude Children's Research Hospital.

US Protocol, Contrast Agent, and Injection Technique
During imaging sessions, mice were kept under a heat lamp to avoid reductions in body temperature that could affect tumor blood flow. To facilitate imaging, they were anesthetized with 2% inhaled isoflurane (Baxter Healthcare, Deerfield, Calif). Mice were scanned in random order with regard to treatment group. The study radiologist (M.B.M.) was aware of the tumor treatment status for the initial study cohorts consisting of 35 mice (10 with NXS2 control tumors, 17 with TIMP-3–expressing tumors, and eight with tsFlk-1–expressing tumors) but was blinded to treatment status when the experiment was repeated to assess reproducibility in the subset of 13 mice (three with NXS2 control tumors, five with TIMP-3–expressing tumors, and five with tsFlk-1–expressing tumors).

US was performed with an Acuson Sequoia machine (Siemens Medical). The small size of our study subjects required the use of a 15L8-MHz high-frequency transducer. Therefore, we adjusted the contrast agent dose to optimize microbubble detection at a fundamental frequency of 14 MHz. The mechanical index on the US imager was lowered to 0.4 or less to reduce microbubble disruption. The mechanical index was adjusted by using the power output function and was selected after adjusting the imaging depth, focal zone location, and transmit frequency to optimize the balance between image resolution, contrast agent detection, and microbubble preservation. Before contrast agent injection, the greatest longitudinal, transverse, and anteroposterior dimensions of tumors were measured to determine tumor volume. We calculated tumor volumes by using the formula for an ellipsoid model: length x width x depth x .52.

In the 13 mice used to assess reproducibility (NXS2, n = 3; TIMP-3, n = 5; tsFlk-1, n = 5), we also performed color and power Doppler evaluation before contrast agent injection to assess the value of this subjective method of estimating vascularity. Doppler imaging was performed at 7 MHz, with a gain setting of 50 and a pulse repetition frequency of 0.092 for color and 0.046 for power Doppler imaging. The Doppler images were reviewed by the study radiologist (M.B.M.), who subjectively assessed the degree of vascularity within tumors and determined whether a quantitative analysis of the Doppler signal could be performed by using DataPro software (Noesis, Les Ulis, France). For evaluation of contrast agent flow, tumors were imaged in the transverse plane in the region with the largest dimensions, and the transducer was held in this position throughout the examination.

The US contrast agent Optison (GE Healthcare) is a suspension of human serum albumin microspheres encapsulating octafluoropropane gas. Mean microsphere size ranges from 2.0 to 4.5 µm, and particles remain in the intravascular space. The contrast agent was administered by one of three study investigators (C.J.S., P.V.D., or A.M.D.), who used a 27-gauge needle to administer one bolus injection of 0.1 mL of contrast material diluted in 0.1 mL of saline to each mouse via a tail vein. Imaging was recorded on cine clips starting several seconds before the injection and continuing for 2–3 minutes at a frame rate of 30–70 Hz. The clips were downloaded for off-line processing with the DataPro software (Noesis).

Contrast agent injections were considered adequate if a blush of contrast enhancement was visible within the inferior vena cava immediately after the injection. Because histologic evaluation was performed by assessing "hot spots" (see Tumor Immunohistochemistry) within tumors, we chose to evaluate contrast agent flow in individual regions of interest rather than in the entire transverse tumor area. The study radiologist (M.B.M.) chose three round regions of interest within each tumor. The regions of interest were made as large as possible and were positioned to include only areas that enhanced with contrast material and were thought to be viable (ie, not cystic, necrotic, or calcified) (Fig 1). The sizes of regions of interest varied depending on the size of the tumor.


Figure 1
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Transverse US image of a TIMP-3–expressing tumor (arrows) shows representative example of placement of three regions of interest ({circ}) within the tumor for analysis. A = aorta.

 
Each region of interest was evaluated for precontrast baseline signal intensity (SI) in decibels, change in SI from baseline to initial peak ({Delta}SI) in decibels, rate of SI increase from baseline to initial peak (RSI) in decibels per second, and contrast agent washout (both the time until 50% of the initial peak SI was reached and the percentage of peak SI at 30 seconds after injection). The mean of SIs for all pixels within each region of interest was used for analysis. The software calibrated the SI on the basis of a 0–70-dB range and allowed continuous acquisition of region-of-interest mean SI over a 1-minute imaging period.

Tumor Immunohistochemistry
By using previously described methods, formalin-fixed (10%) paraffin-embedded tumor specimens were analyzed by means of immunohistochemical evaluation for pericyte (smooth muscle actin [SMA]) and endothelial cell (CD34) density (14,15). Following heat-induced epitope retrieval in citrate buffer (pH of 6.0; Zymed, San Francisco, Calif), the following primary antibodies were used: biotinylated monoclonal mouse antihuman SMA antibody (clone 1A4; Dako, Carpinteria, Calif) and rat antimouse CD34 antibody (RAM34; PharMingen, San Diego, Calif). The slides for evaluating CD34 expression were then incubated with a secondary, biotinylated rabbit antirat (Vector Laboratories, Burlingame, Calif) antibody. Following incubation with streptavidin conjugated to horseradish peroxidase (Dako), a substrate containing the chromagen 3,3' diaminobenzidine tetrahydrochloride was added. All slides were counterstained with a 1:5 dilution of hematoxylin.

Endothelial cell and pericyte densities were determined by scanning stained tumor sections at low power, and areas of greatest CD34- or SMA-positive density ("hot spots") were chosen for analysis. After identification of the "hot spot," the number of individual brown staining cells were counted at a magnification of x400. CD34 and SMA counts for each tumor section were quantitated independently by at least two of the study investigators (C.J.S., A.M.D., P.V.D.), who were blinded to tumor treatment status and to the other reviewer's results. The average of the two reviewer's results was used for statistical analysis.

Statistical Analysis
The two-sided Mann-Whitney test was used to determine whether US volume measurements, CD34 and SMA counts, and the average baseline SI of the three regions of interest in the NXS2 control tumors were significantly different from those in TIMP-3– and tsFlk-1–expressing tumors. The one-sided Mann-Whitney test was used to determine whether the average {Delta}SI, RSI, and contrast agent washout of the three regions of interest in the NXS2 control tumors exceeded those in the TIMP-3– and tsFlk-1–expressing tumors. The Spearman correlation test was used to evaluate the relation between the average {Delta}SI or RSI and CD34 or SMA or the relation between US tumor volume measurements and caliper tumor volume measurements. P ≤ .05 was considered to indicate a statistically significant difference. All analyses were performed with StatXact software (version 5; Cytel Software, Cambridge, Mass) implemented by using SAS (version 9.1; SAS Institute, Cary, NC).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Region-of-Interest Analysis
Results of the region-of-interest analyses for the initial study cohort (NXS2 [control], n = 10; TIMP-3, n = 17; tsFlk-1, n = 8) and the repeat cohort (NXS2 [control], n = 3; TIMP-3, n = 5; tsFlk-1, n = 5) were similar and are summarized in Tables 1 and 2. These results were also similar to results obtained when the cohorts were combined (Table 3). For the combined cohort, there was no significant difference between the precontrast baseline SI in 22 TIMP-3–expressing tumors (median, 28.1 dB; range, 14.1–36.8 dB) and that in 13 NXS2 control tumors (median, 25.6 dB; range, 9.0–38.8 dB; P = .67). The baseline SI was significantly lower in the 13 tsFlk-1–expressing tumors (median, 18.8 dB; range, 9.6–27.4 dB) than in the control tumors (P = .03).


View this table:
[in this window]
[in a new window]

  		Table 1. Imaging Results for the Initial Study Cohort

 

View this table:
[in this window]
[in a new window]

  		Table 2. Imaging Results for the Repeat Study Cohort

 

View this table:
[in this window]
[in a new window]

  		Table 3. US Imaging and Immunohistochemical Results for the Combined Study Cohorts

 
Following contrast agent administration, the {Delta}SI was significantly lower in both the TIMP-3–expressing tumors (median, 1.4 dB; range, 0.2–5.9 dB; P = .003) and the tsFlk-1–expressing tumors (median, 1.5 dB; range, 0.4–4.8 dB; P = .004) than in the unmodified control tumors (median, 3.6 dB; range, 0.08–16.0 dB) (Figs 2–5). The RSI was also significantly lower in both the TIMP-3–expressing tumors (median, 2.9 dB/sec; range, 0.04–9.7 dB/sec; P = .03) and the tsFlk-1–expressing tumors (median, 1.9 dB/sec; range, 0.08–7.7 dB/sec; P = .01) than in the control tumors (median, 7.1 dB/sec; range, 0.28–21.1 dB/sec) (Figs 4, 5). There was no significant difference in the contrast agent washout (neither in the percentage of peak SI at 30 seconds after injection nor in the time to 50% of peak SI) between treated tumors and control tumors (for all, P ≥ .09; Tables 1–3).


Figure 2
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Transverse US images of an NXS2 untreated control tumor (straight arrows) obtained (a) immediately before contrast agent injection (curved arrow = unenhanced inferior vena cava), (b) immediately after contrast agent injection, showing contrast agent blush in inferior cava (curved arrow), and (c) at peak contrast enhancement. Note the increase in SI of the tumor relative to the precontrast image in a.

 

Figure 2
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Transverse US images of an NXS2 untreated control tumor (straight arrows) obtained (a) immediately before contrast agent injection (curved arrow = unenhanced inferior vena cava), (b) immediately after contrast agent injection, showing contrast agent blush in inferior cava (curved arrow), and (c) at peak contrast enhancement. Note the increase in SI of the tumor relative to the precontrast image in a.

 

Figure 2
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: Transverse US images of an NXS2 untreated control tumor (straight arrows) obtained (a) immediately before contrast agent injection (curved arrow = unenhanced inferior vena cava), (b) immediately after contrast agent injection, showing contrast agent blush in inferior cava (curved arrow), and (c) at peak contrast enhancement. Note the increase in SI of the tumor relative to the precontrast image in a.

 

Figure 3
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: Transverse US images of a tsFlk-1–treated tumor (straight arrows) obtained (a) immediately before contrast agent injection (curved arrow = unenhanced inferior vena cava), (b) immediately after contrast injection, showing contrast blush in inferior vena cava (curved arrow), and (c) at peak contrast enhancement. Note the increase in SI of the tumor relative to the precontrast image in a but diminished enhancement relative to the untreated tumor in Figure 2c.

 

Figure 3
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: Transverse US images of a tsFlk-1–treated tumor (straight arrows) obtained (a) immediately before contrast agent injection (curved arrow = unenhanced inferior vena cava), (b) immediately after contrast injection, showing contrast blush in inferior vena cava (curved arrow), and (c) at peak contrast enhancement. Note the increase in SI of the tumor relative to the precontrast image in a but diminished enhancement relative to the untreated tumor in Figure 2c.

 

Figure 3
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c: Transverse US images of a tsFlk-1–treated tumor (straight arrows) obtained (a) immediately before contrast agent injection (curved arrow = unenhanced inferior vena cava), (b) immediately after contrast injection, showing contrast blush in inferior vena cava (curved arrow), and (c) at peak contrast enhancement. Note the increase in SI of the tumor relative to the precontrast image in a but diminished enhancement relative to the untreated tumor in Figure 2c.

 

Figure 4
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4: Representative example of smoothed time-intensity curves shows enhancement in the inferior vena cava (IVC) and the average of three regions of interest obtained in an untreated NXS2 control tumor and a TIMP-3–expressing tumor. The differences in the {Delta}SI and RSI were statistically significant between treatment groups.

 

Figure 5
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: Box plot shows that (a) {Delta}SI and (b) RSI were significantly greater in untreated NXS2 control tumors than in tumors expressing either TIMP-3 or tsFlk-1.

 

Figure 5
View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: Box plot shows that (a) {Delta}SI and (b) RSI were significantly greater in untreated NXS2 control tumors than in tumors expressing either TIMP-3 or tsFlk-1.

 
Among the subset followed longitudinally, there was no significant difference in the {Delta}SI, RSI, or contrast material washout measured in weeks 1 and 2 in any of the three treatment groups (for all, P ≥ .13). Also, the value of color and power Doppler US was limited; although large feeding vessels were often identified at the periphery of tumors, minimal flow was detected centrally. Subjectively there was no visible difference between treatment groups or change in the Doppler appearance of tumors over time. Because of the lack of color and power Doppler flow within tumors, we were unable to quantitate the Doppler signal by using the DataPro software (Noesis).

Agreement between US and Caliper Measurements of Tumor Volume
For 25 tumor volumes obtained by means of both caliper and US measurement, there was a positive correlation between these methods ({rho} = 0.5405, P = .005); the median US volume was 270 mm3 (range, 56–1008 mm3), and the median caliper volume was 320 mm3 (range, 40–1183 mm3). Among the entire cohort of 48 mice, there was no significant difference in US volume measurements between the NXS2 control tumors (median, 600 mm3; range, 56–2677 mm3) and the tsFlk-1–expressing tumors (median, 294 mm3; range, 88–2059 mm3; P = .15) or between the control tumors and the TIMP-3–expressing tumors (median, 650 mm3; range, 270–2488 mm3; P = .74) (Table 3).

Histologic Findings in Tumors
Significant differences in immunohistochemical staining were seen in the three groups (Table 3, Fig 6). TIMP-3 expression was associated with a significantly lower amount of SMA staining (median, 8.75 counts per high-power field [HPF]; range, 0–30 counts per HPF) compared with control tumors (median, 30 counts per HPF; range, 9.75–50.0 counts per HPF; P < .001), reflecting a paucity of pericytes associated with tumor vessels. The CD34 count in the TIMP-3–expressing tumors (median, 30 counts per HPF; range, 5–45 counts per HPF) was not significantly different from that in control tumors (median, 35 counts per HPF; range, 9.5–45.0 counts per HPF; P = .08). Conversely, CD34 counts (median, 10 counts per HPF; range, 6.25–30.0 counts per HPF) were significantly lower in tumor hot spots in tsFlk-1–expressing tumors than in control tumors (P < .001), while there was no significant difference in SMA count between tsFlk-1–expressing tumors (median, 25 counts per HPF; range, 6.25–44.0 counts per HPF) and control tumors (P = .23) (Table 3, Fig 6).


Figure 6
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6: Representative immunohistochemical analysis of SMA content, assessed with an anti-SMA antibody (top row), and endothelial cell density, assessed with an anti-CD34 antibody (bottow row), at x40 in untreated NXS2 control specimens (left column), TIMP-3–expressing tumors (middle column), and tsFlk-1–expressing tumors (right column). Brown areas reflect positive staining of cells. Note diminished SMA staining in the TIMP-3 tumor and diminished CD34 staining in tsFlk-1 tumor, reflective of different mechanisms of action of these angiogenesis inhibitors.

 
Correlation of Contrast Agent Flow and Immunohistochemical Results
For the entire cohort (treated and untreated tumors), RSI was positively correlated with both CD34 (P = .003) and SMA (P < .001) counts (Fig 7). There was no correlation between {Delta}SI and either CD34 or SMA count.


Figure 7
View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a: Scatterplots show the positive relationship between RSI and (a) pericyte density measured by SMA staining and (b) endothelial cell density measured by CD34 staining.

 

Figure 7
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b: Scatterplots show the positive relationship between RSI and (a) pericyte density measured by SMA staining and (b) endothelial cell density measured by CD34 staining.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
In this study, quantitative measurements of the flow of the US microbubble contrast agent (Optison; GE Healthcare) into retroperitoneal tumors with enforced expression of the angiogenesis inhibitors tsFlk-1 or TIMP-3 revealed that the {Delta}SI and RSI were significantly decreased compared with those in untreated control tumors. Similar results were obtained when the experiment was repeated, which suggests that contrast-enhanced US may compare favorably with other modalities, such as PET and dynamic contrast-enhanced MR imaging, as a noninvasive method of evaluating angiogenesis in vivo. The rate of contrast material washout, however, was not a useful measure of assessing tumor vascularity, since no significant difference was noted in this measure between treated and control tumors. When we evaluated tumors over time, we found no significant change in any of the quantitative measurements of contrast agent flow into tumors. However, this is not surprising because the tumor cells had been engineered to express the angiogenesis inhibitors themselves; therefore, they were essentially receiving continuous treatment during the course of their growth. Conventional power color Doppler evaluation of tumors was not a useful method of assessing tumor vascularity because there was no substantial difference between the Doppler appearance of treated tumors and that of untreated tumors.

On the basis of prior experience with the genetically modified tumors used in this study, we anticipated a profound restriction in growth of the modified tumors relative to unmodified control tumors (13,14). The two inhibitors used cause unique changes in tumor vascularization, as demonstrated by means of immunohistochemical staining. tsFlk-1, a decoy receptor for vascular endothelial growth factor, inhibits endothelial cell activation and affects absolute endothelial cell counts (CD34 count). In contrast, TIMP-3 blocks vessel maturation by preventing pericyte covering of new vessels (reflected in decreased SMA count) and thus restricting the blood supply to an expanding tumor (13,14). To control for potential differences in US enhancement that might result from differences in tumor volume alone, we evaluated treated and untreated tumors when they were of similar size. Therefore, we were able to demonstrate that quantitative measurements of tumor contrast enhancement at gray-scale imaging were valuable in measuring the clinical effect of an angiogenesis inhibitor beyond assessment of tumor size.

It is not clear why some investigators have demonstrated a correlation between contrast-enhanced US estimates of blood flow and histologic measures of angiogenesis while other investigators have not. Nevertheless, contrast-enhanced US has been used to measure tumor response to therapy in several phase I clinical trials and, not surprisingly, has produced mixed results (15–18). Clearly, there remains a need for in vivo functional imaging, with better image resolution and specificity, that is consistently correlated with histologic assessment of tumors treated with the spectrum of different angiogenesis inhibitors (5,8). We found a significant correlation between one of the contrast-enhanced US measures of intratumoral blood flow, RSI, and mean intratumoral endothelial cell and pericyte density as assessed by means of CD34 and SMA staining, respectively. The value of this imaging method for evaluating the effects of angiogenesis inhibitors is highlighted by the fact that it successfully demonstrated the efficacy of inhibitors with distinct and different mechanisms of action.

A potential limitation of our technique is that the DataPro software (Noesis) calibrates pixel intensity on the basis of a 0–70-dB dynamic range, while probably assuming a linear relationship between pixel data and signal amplitude. Fortunately, this assumption should not have significantly distorted our results because changes in decibels after contrast enhancement, in both the control and treated tumors, were relatively small, and within this small range the relationship was probably linear or nearly linear. We do acknowledge that a more formal means of accounting for log compression when relating changes in pixel values to changes in signal amplitude could be achieved with the use of a reference phantom with targets having known differences in scattering (19). Another limitation was the lack of availability of microbubble-specific US technology when we embarked on this study. Contrast pulse sequencing technology is now generally available and will capitalize on nonlinear microbubble energies, which support more effective high-frequency imaging and colorized differentiation of micro- and macrovasculature.

Practical application: Our results support the hypothesis that gray-scale US measurements of microbubble contrast agent flow can be used to measure the functional consequences of antiangiogenic therapy in orthotopic tumors, and, at least to some degree, they reflect histologic changes in tumor vascularity. Additionally, we have shown that this technique can be used in the assessment of angiogenesis inhibitors that act through different mechanisms. This property is advantageous given the broad range of angiogenesis inhibitors that are being introduced in clinical trials. Recent advances in US technology will make this type of analysis easier and quicker to perform. Complementing this technology is autotracking contrast quantification, which allows online acquisition of multiple time-intensity curves for region-of-interest analysis (20).


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   ACKNOWLEDGMENTS
 
We thank Thaddeus A. Wilson, PhD, and Dennis Paul, RDMS, RDCS, for reviewing the manuscript and sonographer Stacey Glass, RT(R), RDMS, RVT, for excellent technical assistance.


   FOOTNOTES
 

Abbreviations: {Delta}SI = change in SI from baseline to initial peak • HPF = high-power field • RSI = rate of SI increase from baseline to initial peak • SI = signal intensity • SMA = smooth muscle actin • TIMP-3 = tissue inhibitor of matrix metalloproteinase-3 • tsFlk-1 = truncated soluble fetal liver kinase-1

See Materials and Methods for pertinent disclosures.

See also Science to Practice in this issue.

Author contributions: Guarantors of integrity of entire study, M.B.M., P.V.D., A.M.D.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, M.B.M., C.J.S., A.C.N., A.M.D.; experimental studies, M.B.M., C.J.S., P.V.D., A.C.N., A.M.D.; statistical analysis, C.J.S., C.S.L.; and manuscript editing, M.B.M., P.V.D., A.C.N., A.M.D.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971;285(21):1182–1186.[Medline]
  2. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature 2000;407:249–257.[CrossRef][Medline]
  3. Zetter BR. Angiogenesis and tumor metastasis. Annu Rev Med 1998;49:407–424.[CrossRef][Medline]
  4. Hlatky L, Hahnfeldt P, Folkman J. Clinical application of antiangiogenic therapy: microvessel density, what it does and doesn't tell us. J Natl Cancer Inst 2002;94:883–893.[Free Full Text]
  5. Padhani AR, Neeman M. Challenges for imaging angiogenesis. Br J Radiol 2001;74:886–890.[Free Full Text]
  6. Anderson H, Price P, Blomley M, Leach MO, Workman P. Measuring changes in human tumour vasculature in response to therapy using functional imaging techniques. Br J Cancer 2001;85(8):1085–1093.[CrossRef][Medline]
  7. McDonald DM, Choyke PL. Imaging of angiogenesis: from microscrope to clinic. Nat Med 2003;9(6):713–725.[CrossRef][Medline]
  8. Ferrara KW, Merritt CR, Burns PN, Foster FS, Mattrey RF, Wickline SA. Evaluation of tumor angiogenesis with US: imaging, Doppler, and contrast agents. Acad Radiol 2000;7:824–839.[CrossRef][Medline]
  9. Forsberg F, Dicker AP, Thakur ML, et al. Comparing contrast-enhanced ultrasound to immunohistochemical markers of angiogenesis in a human melanoma xenograft model: preliminary results. Ultrasound Med Biol 2002;28(4):445–451.[CrossRef][Medline]
  10. Iordanescu I, Becker C, Zetter B, Dunning P, Taylor GA. Tumor vascularity: evaluation in a murine model with contrast-enhanced color Doppler US—effect of angiogenesis inhibitors. Radiology 2002;222:460–467.[Abstract/Free Full Text]
  11. Denis F, Bougnoux P, De Poncheville L, Prat M, Catroux R, Tranquart F. In vivo quantitation of tumour vascularisation assessed by Doppler sonography in rat mammary tumours. Ultrasound Med Biol 2002;28(4):431–437.[CrossRef][Medline]
  12. Delorme S, Haberkorn U, Kinscherf R, Zuna I, Bahner ML, Van Kaick G. Changes of tumor vascularity during gene therapy monitored with color Doppler US. Ultrasound Med Biol 2001;27(12):1595–1603.[CrossRef][Medline]
  13. Davidoff AM, Leary MA, Ng CY, Vanin EF. Gene therapy–mediated expression by tumor cells of the angiogenesis inhibitor flk-1 results in inhibition of neuroblastoma growth in vivo. J Pediatr Surg 2001;36:30–36.[CrossRef][Medline]
  14. Spurbeck WW, Ng CY, Strom TS, Vanin EF, Davidoff AM. Enforced expression of tissue inhibitor of matrix metalloproteinase-3 affects functional capillary morphogenesis and inhibits tumor growth in a murine tumor model. Blood 2002;100:3361–3368.[Abstract/Free Full Text]
  15. Weidner N, Folkman J, Pozza F, et al. Tumor angiogenesis: a new significant and independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst 1992;84:1875–1887.[Abstract/Free Full Text]
  16. Yang WT, Tse GM, Lam PK, Metreweli C, Chang J. Correlation between color power Doppler sonographic measurement of breast tumor vasculature and immunohistochemical analysis of microvessel density for the quantitation of angiogenesis. J Ultrasound Med 2002;21:1227–1235.[Abstract/Free Full Text]
  17. Marson LP, Kurian KM, Miller WR, Dixon JM. The effect of tamoxifen on breast tumour vascularity. Breast Cancer Res Treat 2001;66:9–15.[CrossRef][Medline]
  18. Escudier B, Lassau N, Couanet D, et al. Phase II trial of thalidomide in renal-cell carcinoma. Ann Oncol 2002;13:1029–1035.[Abstract/Free Full Text]
  19. Knipp BS, Zagzebski JA, Wilson TA, Dong F, Madsen EL. Attenuation and backscatter estimation using video signal analysis applied to B-mode images. Ultrason Imaging 1997;19:221–233.[Medline]
  20. Phillips P, Gardner E. Contrast-agent detection and quantification. Eur Radiol 2004;14(suppl 8):P4–P10.

Related Article

Will Improved Assessment of Response to Antiangiogenic Therapies Be Achieved with Contrast-enhanced Gray-Scale US?
Jonathan B. Kruskal
Radiology 2006 240: 1-2. [Full Text] [PDF]



This article has been cited by other articles:


Home page
 The Journal of RheumatologyHome page
M.-J. KAISER, J.-P. HAUZEUR, S. BLACHER, J.-M. FOIDART, M. DEPREZ, A. ROSSKNECHT, and M. G. MALAISE
Contrast-enhanced Coded Phase-inversion Harmonic Sonography of Knee Synovitis Correlates with Histological Vessel Density: 2 Automated Digital Quantifications
J Rheumatol, July 1, 2009; 36(7): 1391 - 1400.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Surg. Oncol.Home page
T. L. Sims Jr., J. B. Hamner, R. A. Bush, R. F. Williams, J. Zhou, S. U. Kim, K. S. Aboody, M. K. Danks, and A. M. Davidoff
Neural Progenitor Cell-mediated Delivery of Interferon Beta Improves Neuroblastoma Response to Cyclophosphamide
Ann. Surg. Oncol., November 1, 2008; 15(11): 3259 - 3267.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
P. V. Dickson, J. B. Hamner, T. L. Sims, C. H. Fraga, C. Y.C. Ng, S. Rajasekeran, N. L. Hagedorn, M. B. McCarville, C. F. Stewart, and A. M. Davidoff
Bevacizumab-Induced Transient Remodeling of the Vasculature in Neuroblastoma Xenografts Results in Improved Delivery and Efficacy of Systemically Administered Chemotherapy
Clin. Cancer Res., July 1, 2007; 13(13): 3942 - 3950.
[Abstract] [Full Text] [PDF]

Home page
 Mol Cancer ResHome page
P. V. Dickson, J. B. Hamner, C. J. Streck, C. Y.C. Ng, M. B. McCarville, C. Calabrese, R. J. Gilbertson, C. F. Stewart, C. M. Wilson, M. W. Gaber, et al.
Continuous Delivery of IFN-{beta} Promotes Sustained Maturation of Intratumoral Vasculature
Mol. Cancer Res., June 1, 2007; 5(6): 531 - 542.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
J. B. Kruskal
Will Improved Assessment of Response to Antiangiogenic Therapies Be Achieved with Contrast-enhanced Gray-Scale US?
Radiology, July 1, 2006; 240(1): 1 - 2.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McCarville, M. B.
Right arrow Articles by Davidoff, A. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McCarville, M. B.
Right arrow Articles by Davidoff, A. M.
Related Collections
Right arrowRelated Article

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE