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


     

Published online before print October 23, 2003, 10.1148/radiol.2293021045
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2293021045v1
229/3/885    most recent
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 Daldrup-Link, H. E.
Right arrow Articles by Rummeny, E. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Daldrup-Link, H. E.
Right arrow Articles by Rummeny, E. J.
(Radiology 2003;229:885-892.)
© RSNA, 2003

Breast Imaging

Quantification of Breast Tumor Microvascular Permeability with Feruglose-enhanced MR Imaging: Initial Phase II Multicenter Trial1

Heike E. Daldrup-Link, MD, Jana Rydland, MD, Thomas H. Helbich, MD, Atle Bjørnerud, PhD, Karl Turetschek, MD, Kjell Arne Kvistad, MD, PhD, Elisabeth Kaindl, MD, Thomas M. Link, MD, Karin Staudacher, MS, David Shames, MD, Robert C. Brasch, MD, Olav Haraldseth, MD and Ernst J. Rummeny, MD

1 From the Dept of Radiology, Univ Hosp, Technical Univ of Munich, Ismaningerstr 22, 81675 Munich, Germany (H.E.D.L., T.M.L., E.J.R.); MR-Ctr and Dept of Anaesthesia and Medical Imaging, Norwegian Univ of Science and Technology, Trondheim, Norway (J.R., K.A.K., O.H.); Dept of Radiology, Univ Hosp of Vienna, Austria (T.H.H., K.T., E.K.); Dept of Radiology, Univ Hosp of Oslo, Norway (A.B.); Div of Statistics, Amersham Health, Oslo, Norway (K.S.); and Ctr of Pharmaceutical and Molecular Imaging, Univ of California, San Francisco (D.S., R.C.B.). Received Aug 24, 2002; revision requested Oct 18; final revision received May 23, 2003; accepted May 27. Supported in part by NIH grant RO1 CA82923. Address correspondence to H.E.D.L. (e-mail: daldrup@roe.med.tum.de).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To investigate use of the macromolecular contrast agent feruglose for differentiating and grading of human benign and malignant breast tumors on the basis of their microvascular characteristics.

MATERIALS AND METHODS: Sixty-three women with 63 primary breast lesions were examined with dynamic T1-weighted gradient-echo magnetic resonance (MR) imaging after intravenous injection of feruglose. A two-compartment unidirectional kinetic model applied to the dynamic data yielded estimates of the endothelial transfer coefficient, KPS, and the fractional plasma volume of the tumors. These MR imaging–derived parameters were correlated with the histologic tumor grade and quantified according to the Scarff-Bloom-Richardson (SBR) score by means of Pearson product moment correlation analyses. Differences between malignant and nonmalignant breast lesions with respect to KPS for feruglose were evaluated by means of the {chi}2 test and by calculating the sensitivity, specificity, and positive and negative predictive values.

RESULTS: Histologic analysis revealed 26 benign and 37 malignant tumors. A moderate yet statistically significant correlation between KPS and SBR score was found (R = 0.496, P < .001). No significant correlation was observed between fractional plasma volume and SBR score (R = 0.085, P = .507). The KPS values were zero for 19 (73%) of the 26 benign tumors and were greater than zero for 27 (73%) of the 37 carcinomas. This distribution was significantly different ({chi}2 = 13.035, P = .001). With the criterion KPS > 0 in carcinomas, sensitivity was 0.73, specificity was 0.73, and the positive predictive value was 0.79.

CONCLUSION: Quantitative measures of tumor microvascular permeability can be used for breast tumor characterization. The probability of breast tumor microvascular hyperpermeability to be associated with malignancy is 79%.

© RSNA, 2003

Index terms: Breast neoplasms, MR, 00.12143 • Iron • Magnetic resonance (MR), contrast media, 00.12143


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Magnetic resonance (MR) imaging of the breast with standard small molecular gadolinium chelates has a high sensitivity of up to 98% for breast tumor detection (1–3) but a limited specificity of 30%–71% for breast tumor characterization (4–8). While the high sensitivity makes MR imaging of the breast appropriate for staging applications, it is unlikely to develop as a time- and cost-effective diagnostic method if it does not provide precise and definite characterization of the lesions detected. Improved tumor specificity would improve the treatment of women with breast masses, decrease the number of biopsies and surgeries performed, and optimize therapy regimens for malignancies.

New macromolecular contrast agents may improve the specificity of MR imaging of the breast because they extravasate across the discontinuous microvessel endothelium ofmalignant tumors but not across the intact microvessel wall of benign lesions (9). Findings in experimental studies show that this pathologic transendothelial leakage of macromolecular contrast agents is highly characteristic for carcinomas and can be quantified by means of kinetic analyses of dynamic MR data with a two-compartment mathematic model (9–11). Furthermore, the microvascular hyperpermeability seen in carcinomas increased with increasing tumor grade, which provides a noninvasive estimate of tumor biology (9). Feruglose (Clariscan; Amersham Health, Oslo, Norway) is one of the first macromolecular contrast agents to become available clinically (12). In a rodent breast tumor model, feruglose-enhanced MR imaging provided accurate quantitative measures of tumor microvascular permeability that correlated with tumor grade (13,14).

The purpose of our study was to investigate use of the macromolecular contrast agent feruglose for differentiating and grading of human benign and malignant breast tumors on the basis of their microvascular characteristics.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Patients
In a phase II open-label proof-of-concept multicenter clinical trial, 63 consecutive women (mean age, 52 years; age range, 25–83 years) underwent feruglose-enhanced MR imaging of the breast. Two additional patients who were initially enrolled in the study withdrew their informed consent before the start of the MR examination as a result of claustrophobia in one and chest pain in the other. The study was conducted between October 2000 and April 2001 at three university hospitals in three countries. The study was approved by the local ethical committees of the three institutions and was performed in accordance with the current version of the Declaration of Helsinki and the International Conference on Harmonisation Good Clinical Practice guidelines. Informed consent was obtained from all patients after the nature of the examinations was explained fully.

Inclusion criteria included planned surgery for a suspect lesion in the right (34 [54%] of 63 tumors) or left (29 [46%] tumors) breast that was diagnosed at mammography as a Breast Imaging Reporting and Data System, or BI-RADS, grade IV or V lesion. Forty-eight (76%) tumors appeared as masses at mammography, 13 (21%) had an architectural distortion, and 19 (30%) had microcalcifications. The majority of tumors (54 [86%]) were palpable. The following exclusion criteria were defined: age younger than 18 years, patients under guardianship, standard MR contraindications (eg, intracorporal metal devices such as pacemakers or intracranial clips), administration of iron oxide nanoparticles within 7 days before the study, participation in another investigational trial, pregnancy and breast-feeding, as well as history of serious adverse reactions to contrast agents, iron compounds, or other drugs. Patients with hemosiderosis, liver cirrhosis, or other diseases with serious liver dysfunction (Child-Pugh class C) were also excluded.

Safety
Safety evaluations were performed before and up to 72 hours after feruglose administration. Blood pressure and heart rate were measured immediately before administration and 5, 15, and 30 minutes and 1, 2–4, 24, and 72 hours after administration. At these time points, the patients were also questioned about any side effects by using a nonleading question, such as, "How do you feel?" A physical examination, including assessment of general appearance, lungs, cardiovascular system, and the abdomen, was performed before and 1, 2–4, 24, and 72 hours after administration. In addition, 12-lead electrocardiograms, venous blood samples, and urine samples were obtained before and 2–4, 24, and 72 hours after administration.

Blood samples were taken from a vein contralateral to the injection site and were analyzed hematologically (erythrocytes, reticulocytes, hemoglobin, hematocrit, leukocytes, thrombocytes) and serum chemistry (liver enzymes, bilirubin, kidney function parameters, protein, lactate dehydrogenase, creatin kinase, amylase, electrolytes, and parameters of iron metabolism). Urinalysis at these time points included evaluations of pH, specific gravity, erythrocytes, ketones, protein, glucose, bilirubin, and urobilinogen. Any change in a patient’s subjective feelings, physical status, vital signs, or laboratory test findings was recorded and determined by the investigators to be related to the study drug, not related to the study drug, or of unknown cause.

Contrast Agent
Feruglose is a colloid-based ultrasmall superparamagnetic iron oxide with a mean particle diameter of 11–15 nm (15). These particles consist of nonstoichiometric magnetic crystalline cores with a diameter of approximately 6 nm and a 5–9-nm-thick oxidized starch coating that consists of carbohydrate-polyethylene glycol (14). The R1 relaxivity at 37°C and 0.47 T is approximately 20 mmol/L/sec, and the R2 relaxivity at 37°C and 0.5 T is approximately 35 mmol/L/sec. The initial distribution in the intravascular space approximates the blood volume (16). Blood-pool half-life was reported to be more than 2 hours in humans (15). The iron is subsequently phagocytosed by macrophages in lymph nodes, liver, spleen, and bone marrow, incorporated into the iron stores of the body, and metabolized.

For this study, feruglose was supplied as a low-viscosity (approximately 1.5 mPa/sec) isotonic solution that contains 30 mg of iron per milliliter. A dose of 2 mg of iron per kilogram of body weight was withdrawn through a 5-µm filter for intravenous injection. This dose of iron (140 mg for a 70-kg patient) is smaller than that present in one unit of blood (approximately 200 mg). Feruglose was initially used in phase I–II studies as a blood-pool agent for MR angiography with good tolerability and safety profile in a dose range of 1–10 mg Fe/kg (12,16,17). The value of feruglose in tumor MR imaging has not yet been investigated in humans.

MR Imaging
MR imaging was performed with 1.5-T systems (Magnetom Vision, Siemens, Erlangen, Germany; Symphony, Philips, Best, the Netherlands). The patients were placed prone on a dedicated breast coil. Dynamic MR imaging was performed with three-dimensional gradient-echo sequences with repetition times of 12–14 msec, echo times of less than 2 msec (range, 0.9–1.8 msec), flip angles of 25°–30°, acquisition matrixes of 100–256 x 256 pixels, and fields of view of 200–300 mm2. After acquisition of three to five transverse precontrast MR images, feruglose (2 mg of iron per kilogram of body weight) was injected as a bolus into an antecubital or hand vein, and the catheter was subsequently flushed with 20 mL of 0.9% saline. Thereafter, five to 20 postcontrast MR images were obtained every minute up to 5 minutes after injection, then every 2–5 minutes up to 30 minutes, and then every 5 minutes up to 60 minutes. Five of the 63 patients were able to finish only a 30-minute dynamic MR imaging period because they were too uncomfortable in the prone position in the magnet.

Histologic Findings
All 63 primary tumors were surgically removed with lumpectomy (n = 35), mastectomy (n = 3), or excisional biopsy (n = 25). In addition, five secondary lesions that were depicted at MR imaging in five patients were also excised (three lesions) or sampled at core biopsy (two lesions). However, these lesions were excluded from further analyses to exclude dependencies of the results. After tumor resection, the orientation of the tumor was marked; then, the tumor was fixed in 10% formalin, embedded in paraffin, and sectioned in the same plane as was used at MR imaging. Hematoxylin-eosin staining was performed for standard histologic analyses. At each center, one experienced pathologist determined the maximal tumor size, categorized the tumors as benign or malignant, specified the histologic subtype, and performed subsequent tumor grading according to the Scarff-Bloom-Richardson (SBR) method (18,19).

Ten fields in the region of highest mitotic activity were analyzed with a phase-constrast microscope with high magnification and field diameter of 0.44 mm. Each tumor was assigned from one to three points in each of three categories: gland formation, anaplasia, and mitoses: 1 = highly differentiated tumor, 2 = intermediate, 3 = undifferentiated tumor. The sum of the points describes the tumor grade. The lowest score possible is 3, which indicates a highly differentiated lesion. The highest score is 9, which indicates a poorly differentiated malignancy (18). In accordance with the original definition, the SBR score can only be applied to malignant tumors. To enable correlation with MR imaging findings for all malignant and nonmalignant lesions, the nonmalignant lesions were assigned an SBR score of 3 (ie, the lowest possible tumor grade). Efforts were made to minimize a potential for bias in correlations of histologic findings with MR imaging findings of feruglose-enhanced data. Imaging groups, pathologists, and analyzers acquired their data independently, and pathologists and analyzers were blinded to clinical data and results of conventional mammography and ultrasonography.

Data Analysis
Mean signal intensity (SI) for the tumor and the blood in the vena cava or internal mammary vein were measured by one observer at each institution (H.E.D., J.R., K.T.) in the central section of the imaging volume by using two to four operator-defined regions of interest. The size of the regions depended on the total lesion diameter, with a minimum of 30 pixels and a maximum of 2,375 pixels per region. Change in signal intensity ({Delta}SI) was calculated according to the following formula: {Delta}SI = [(SIpost - SIpre)/SIpre] x 100, where "post" means postcontrast and "pre" means precontrast. The {Delta}SI data in blood ({Delta}SIB[t]) and tumor ({Delta}SIT[t]) were assumed to be proportional to the concentration of feruglose in blood and tumor, respectively (20). The former was divided by the quantity (1 minus hematocrit) to provide a measure proportional to the concentration of feruglose in plasma ({Delta}SIP[t]).

Kinetic Analysis
The {Delta}SIP(t) and {Delta}SIT(t) data from blood and tumor, respectively, were used for kinetic analysis to estimate the coefficient of endothelial permeability KPS (in milliliters per minute per 100 mm of tissue) and the fractional plasma volume of the tumor tissue (in milliliters per millimeter of tissue). A two-compartment unidirectional model of tumor tissue was used that was composed of plasma and interstitial fluid, as previously described (9,11). In this model, a monoexponential function fitted to the {Delta}SIP(t) data from blood was used as a forcing function to represent the plasma response of the contrast agent in the tumor after scaling for fractional plasma volume. The KPS values (product of fractional plasma volume and the fractional leak rate of contrast agent from plasma to the interstitial water of the tumor) determined from the model were scaled upward by 100 to correspond to 100 g of tumor tissue. All data fitting was performed by one expert in kinetic analyses (D.S.) by using a commercially available computer program (SAAM II; SAAM Institute, Seattle, Wash) that makes use of standard variance-weighted nonlinear regression, the fractional SD of the data that are assumed to be known within a proportional constant (21). The precision of the estimates of the model parameters was determined from the covariance matrix at the least squares fit. Small KPS values with large uncertainties were set equal to zero since these values could not be statistically differentiated from zero at the P < .05 level.

In addition, KPS measurements of the data obtained on a pixel-by-pixel basis by using the technique of Patlak et al (22) and Patlak and Blasberg (23) were transferred by another expert in kinetic analyses (A.B.) to "permeability maps." These maps relate each pixel in this functional MR image to a gray scale proportional to the magnitude of the KPS values.

Statistical Analysis
All quantitative data for the 63 breast lesions were displayed as means ± SDs. The histologic tumor diameter was described with mean and median values. Mean fractional plasma volume and KPS values from the feruglose model were compared between malignant and nonmalignant breast tumors by means of an unpaired Student t test. Pearson product moment correlation analyses were used to compare the estimated MR imaging–derived parameters, fractional plasma volume and KPS, with the SBR score. These analyses were performed by including and excluding the five patients with secondary tumors. In addition, group differences between malignant and nonmalignant breast lesions with respect to KPS for feruglose were evaluated by means of a {chi}2 test. Differences were considered significant with P < .05. In addition, sensitivity, specificity, and positive and negative predictive values were calculated to differentiate between malignant and nonmalignant breast lesions.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Safety of Feruglose
The 63 patients who underwent feruglose-enhanced MR imaging were evaluated with respect to safety of the study drug. Up to 72 hours after administration of feruglose, no serious adverse events were observed. In general, no substantial changes in physical status, mean vital signs, or 12-lead electrocardiograms were seen. In 14 (22%) of the patients, mild or moderate adverse events were observed. The investigators judged that the events were related to the study drug in six patients (10%), were not related in four (6%), and were of unknown origin in four (6%). These adverse events included minor transient hypotension directly after iron oxide injection (n = 4), headache (n = 1), nausea (n = 2), diarrhea (n = 1), epistaxis (n = 1), insomnia (n = 1), coughing (n = 1), pharyngitis (n = 1), transient increase in lactate dehydrogenase (n = 1), and increased white blood cell counts due to clinically evident mastitis (n = 1). As expected, slight increases in serum iron and ferritin levels were seen (Table 1). No significant changes in the serum chemistry parameters that reflected liver function, renal function, or coagulation were observed (P > .05). Urinalysis findings remained in the normal range in all patients.


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

  		TABLE 1. Serum Iron Laboratory Variables before and after Injection of Feruglose

 
Tumor Histologic Findings
Histologic examination revealed 26 nonmalignant breast lesions and 37 malignant breast tumors. Diameters of the tumors ranged from 5 to 55 mm, with a mean diameter of 21 mm ± 11 (SD) and a median diameter of 16 mm. Two tumors had a diameter of less than 1 cm. Nonmalignant lesions included 11 cases of fibroadenoma, five cases of focal fibrosis, three cases of proliferative mastopathy with ductal hyperplasia, two cases of proliferative mastopathy with atypical hyperplasia, two cases of sclerosing adenosis, one case of papillomatosis, one case of chronic inflammation, and one case of atypical mycobacteriosis. Malignant breast tumors included 31 cases of invasive ductal carcinoma, five cases of invasive lobular carcinoma, and one case of medullary carcinoma. In the two cases of invasive ductal carcinoma, ductal carcinoma in situ was found at the edge of one tumor. Thus, the SBR tumor grades ranged from 3 to 9.

MR Imaging
After bolus injection of feruglose, blood in the aorta and inferior vena cava showed pronounced enhancement with maximal mean {Delta}SI of 16.3% ± 3.9 within 2 minutes after contrast medium administration. After this peak enhancement, only minor gradual decline in blood {Delta}SI was noted during the 60-minute imaging period, which reflected the prolonged retention of the macromolecular blood-pool agent in the intravascular space (Fig 1). Normal breast parenchyma showed a minor or no enhancement response, with maximal mean {Delta}SI of 0.08% ± 0.05 in the first 5 minutes after administration of feruglose (Fig 1).



View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Representative subtracted (postcontrast minus precontrast) T1-weighted gradient-echo MR images (repetition time msec/echo time msec of 14/1.4 with flip angle of 25°) of breast tumors after injection of feruglose. Top row: Benign fibroadenoma (arrow, {blacktriangleup}) in left breast demonstrates enhancement that parallels the blood enhancement curve ({blacksquare}). Middle row: Low-grade carcinoma in right breast. Bottom row: High-grade carcinoma in right breast. In the middle and bottom rows, carcinomas (arrows, {blacktriangleup}) show a steadily increasing contrast enhancement while blood enhancement remains nearly constant. This is indicative of microvascular hyperpermeability to macromolecular feruglose. High-grade cancer shows an even higher enhancement compared with that of low-grade cancer. Normal breast tissue ({bullet}) shows only minor enhancement, which parallels the blood enhancement curve.

 
Accumulation of feruglose in the tumors over time differed between nonmalignant and malignant breast tumors: 19 of the 26 benign lesions (73%) exhibited no significant enhancement (n = 7) or a minor degree of enhancement (n = 12), with {Delta}SI values that paralleled the blood enhancement response (Fig 1). Therefore, transendothelial leakage of the macromolecular agent was excluded. In contrast, 27 (73%) of the 37 carcinomas demonstrated enhancement that increased steadily over time as blood enhancement decreased (Fig 1). This tumor-blood relationship is indicative of endothelial hyperpermeability to the macromolecular feruglose.

Morphologic enhancement patterns did not show any differences between nonmalignant and malignant lesions in either the early or late perfusion phases. The two cases of ductal carcinoma in situ could not be reliably distinguished from adjacent invasive carcinomas on the basis of their enhancement. Enhancement was homogeneous in six (23%) benign tumors and 15 (40%) carcinomas and was inhomogeneous in nine (35%) benign tumors and 16 (43%) malignant tumors. Rim enhancement was found in four (15%) benign lesions (two inflammatory lesions, one fibroadenoma with central fibrosis, one sclerosing adenosis) and six (16%) carcinomas.

Kinetic Analysis
The permeability model for feruglose fit the data well. Mean values for tumor fractional plasma volume did not differ significantly between nonmalignant lesions (0.043 mL/mm of tissue ± 0.059) and malignant tumors (0.049 mL/mm ± 0.030, P = .568). Fractional plasma volume values for benign lesions were highly variable, as reflected in the large SDs from the group mean, and also demonstrated broad overlap with the carcinomas. No significant correlation was found between fractional plasma volume and tumor grade (R = 0.085, P = .507). Thus, the variance in SBR tumor grade is not reflected in the magnitude of fractional plasma volume.

The greatest difference between nonmalignant and malignant tumors with respect to the kinetic parameters of our model was exhibited by KPS for feruglose. Findings at kinetic analysis revealed no leakage of the macromolecular contrast agent into the interstitium in 19 of 26 benign lesions but positive transendothelial leak in 27 of 37 carcinomas (Fig 2). In addition, there appeared to be an increasing degree of leakage with increasing tumor grade (Fig 2).



View larger version (73K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Representative tumor microvascular permeability (KPS) maps obtained from feruglose-enhanced dynamic gradient-echo MR images (14/1.4 with flip angle of 25°). Top row: Fibroadenoma (arrow) in left breast. Middle row: Low-grade carcinoma (arrow) in left breast. Bottom row: High-grade carcinoma (arrow) in right breast. Benign fibroadenoma demonstrates zero microvascular permeability to macromolecular feruglose. Carcinomas show increased KPS values. Microvascular hyperpermeability to feruglose appears to be characteristic for malignancy.

 
Accordingly, 19 (73%) of the 26 nonmalignant tumors had a KPS of zero, and 27 (73%) of the 37 malignant tumors had a KPS greater than zero. The mean KPS value of the benign tumors (0.011 mL/min/100 mm of tissue ± 0.020) was significantly lower than that of the carcinomas (0.045 mL/min/100 mm ± 0.047, P < .001). The distribution of zero and positive KPS values differed significantly between nonmalignant and malignant breast tumors, with a {chi}2 of 13.035 and a probability of P = .001 (Table 2). With the diagnostic criterion that carcinomas are characterized by abnormal microvascular permeability to macromolecular feruglose (ie, KPS > 0 in carcinomas), sensitivity of 0.73, specificity of 0.73, positive predictive value of 0.79, and negative predictive value of 0.66 were obtained. These data did not change when the five patients with secondary tumors were included or excluded. The positive predictive value indicates that 79% of all breast tumors with positive transendothelial leakage of feruglose were carcinomas. False-negative malignancies included invasive ductal carcinomas of low (n = 4), intermediate (n = 2), and high (n = 2) grades; one scirrhous carcinoma; and one high-grade invasive lobular carcinoma. All false-negative intermediate- and high-grade cancers were relatively large, with diameters between 20 and 55 mm. False-positive benign lesions included inflammatory lesions (n = 2), benign focal fibrosis (n = 2), fibroadenoma (n = 2), and proliferative mastopathy with ductal hyperplasia (n = 1).


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

  		TABLE 2. MR Imaging-derived Quantitative Estimates of Tumor Microvascular Permeability to Feruglose

 
Findings at analysis of the 63 histologically proved primary breast lesions showed a moderate but statistically significant correlation between KPS value and SBR tumor grade, with a correlation coefficient of 0.496 and a probability of P < .001. Since the statistical test assumed no correlation between KPS and SBR score, the moderate correlation coefficient can be significant. With inclusion of the five secondary tumors, the correlation between KPS and SBR score increased to 0.520 (P < .001). When secondary tumors and the two inflammatory lesions were excluded from the analysis, the correlation between KPS value and SBR tumor grade increased to 0.533 (P < .001).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
With a noninvasive MR imaging technique and a macromolecular contrast agent, we were able to quantitatively assess microvascular characteristics in human breast tumors. Seventy-nine percent of all breast tumors that showed positive transendothelial leakage of the blood-pool agent feruglose were carcinomas. In addition, the tumor microvascular permeability of feruglose showed a moderate but significant correlation with tumor grade. To our knowledge, this is the first clinical application of a macromolecular contrast agent for MR imaging of the breast.

Results of this clinical study are in accordance with those obtained in experimental studies that demonstrate macromolecular hyperpermeability of microvessels in malignancies with invasive techniques, such as histologic examination and techniques with vital dyes and radiotracers (24–29). Furthermore, hyperpermeability of tumor microvessels to macromolecular contrast agents was also observed in experimentally induced mammary adenocarcinomas but was not seen in benign fibroadenomas (9,13,30). The underlying mechanisms that lead to microvascular hyperpermeability in malignant tumors consist of initial formation of structurally defective tumor capillaries, secondary defects in the microvascular endothelium due to ischemic or immunologic damage, and formation of venular endothelial gaps in response to vasoactive mediators (31).

Findings in a recent experimental study with feruglose in a rodent breast tumor model reveal a correlation coefficient of 0.82 between tumor microvascular permeability and SBR tumor grade (13). This correlation was substantially higher than that observed in the present study (R = 0.47). Furthermore, in the animal model, all carcinomas showed positive microvascular permeability of the macromolecular probe feruglose (13), whereas in the current study, 10 of 37 carcinomas did not show any detectable transendothelial leakage of feruglose. There are potential explanations for these differences. (a) In the animal model, the spectrum of tumor subtypes, fibroadenomas or adenocarcinomas, is limited. In humans, the histologic tumor spectrum is more variable, with a higher percentage of benign tumors and low-grade malignancies. (b) In the animal model, inflammatory lesions, which regularly exhibit marked microvascular hyperpermeability, do not develop (10). Since patients were not preselected in the current study, two patients with chronic or atypical benign inflammatory lesions were included; they demonstrated false-positive hyperpermeability to the macromolecular contrast agent. (c) In the animal model, tumors with uniform sizes of about 1–2 cm were investigated, whereas in the present study tumors exhibited a wide range in size, from 0.5 to 5.5 cm. On the basis of findings in histologic studies, it is well known that tumor vessel density, diameter, and surface area first increase and then decrease with increasing tumor size (25,32,33). In large carcinomas, elevated interstitial pressure may lead to a reduction in the extravasation of blood-borne macromolecules (34). (d) The signal-to-noise ratio of MR images obtained in anesthetized rats was substantially higher than that seen in the present study in humans, who always experienced some motion during the 1-hour study period. This may have decreased the accuracy of differentiation of a small KPS value from zero and determination of the absolute KPS magnitude. (e) Differentiation by a pathologist between benign tumors and low-grade carcinomas may be difficult (9,35). Since microvascular hyperpermeability is one of the first characteristics of the "angiogenesis switch" and subsequent progression to an accelerated growth phase (35), characterization of malignancy with MR imaging before it becomes apparent at standard histologic examination should also be considered.

Our data show high variability of breast cancer microvessel permeability. This diversity in presence and degree of microvascular hyperpermeability in carcinomas, as quantified with our MR imaging procedure, may have important implications for tumor growth, seeding of metastases, and accumulation of therapeutic drugs. Microvascular hyperpermeability allows albumin, fibrinogen, and other plasma proteins to gain access to the tumor interstitium, which forms a favorable matrix for subsequent ingrowth of new microvessel buds, angiogenesis, and tumor cell proliferation (36). Thus, microvascular hyperpermeability increases the potential for tumor cells to escape into the circulation and to metastasize (35). Therefore, estimates of tumor microvascular permeability at MR imaging may be adjunctive prognostic parameters in patients with unknown or suspect nodal status, as in cases of preadjuvant therapy, incomplete axillary dissection, or negative axillary nodes (18). Furthermore, since the degree of tumor microvascular permeability influences the interstitial accumulation of macromolecular tumoricidal drugs, as well as monoclonal antibodies, estimates of tumor microvascular permeability at MR imaging may be important for treatment planning (37,38).

Endothelial permeability to standard small molecular gadolinium chelates that are currently used in routine clinical practice reflects different biology than the endothelial permeability shown for macromolecular probes. Small (500-Da) molecular gadolinium chelates, which are about 200 times smaller than the molecules of feruglose (100 kDa), extravasate rapidly and completely through the intact microvessel endothelium of normal organs. The transendothelial permeability seen with these agents in benign lesions cannot increase significantly in malignant tumors (9). Thus, small molecular permeability appears to be less useful for characterizing cancers than it is for characterizing macromolecular permeability.

Both micromolecular gadolinium chelates and macromolecular contrast agents may be used to estimate blood volume or "vascularity" of breast lesions (eg, on the basis of the initial tumor enhancement or kinetic analysis findings) (1–8). Some investigators found mean blood volume in malignant tumors to be higher than that in benign tumors when taken as groups (39,40). Other investigators could not find a significant difference in blood volume estimates as a result of the high standard variations of the mean values (4,9,41). This finding was also reflected in the high SDs of fractional plasma volume values determined for benign and malignant breast lesions in the current study. Since breast lesions have wide histologic and biologic diversities, individual blood volume estimates exhibit a wide overlap between benign and malignant conditions and thus limit the potential value of this measure as a diagnostic tool (8,41).

Limitations of the current study include the following: The contrast agent dose of 2 mg Fe/kg provided considerably lower breast tumor enhancement than that obtained with standard small molecular gadolinium chelates. This dose was selected on the basis of results of previous experimental studies (13,30). A dose of 5 mg of iron per kilogram of body weight proved to be too high because of potential confounding T2* effects, and a dose of 1 mg of iron proved to be too small (13,30). Thus, further studies may evaluate intermediate doses. We evaluated additional patients with a dose of 4 mg of iron but did not achieve better T1 enhancement of breast tumors (unpublished data, 2001). Another approach to better accommodate the macromolecular probe may be implementation of optimized MR pulse sequences, such as T1-weighted sequences with further decreased echo times (eg, by using fractional echoes) or T2*-weighted sequences. These sequences might be advantageous as a result of the considerably higher T2* relaxivity compared with T1 relaxivity of the iron oxide–based contrast agent (15–17).

Considering the smaller T1 effect of feruglose compared with that of standard small molecular gadolinium chelates, we applied dynamic pulse sequences with a reduced matrix of 100–256 x 256 pixels to optimize the signal-to-noise ratio and, thus, the delineation of transendothelial leak of contrast agent. However, this limited spatial resolution may have restricted perception of morphologic tumor features.

Safety data generated in this study are in accordance with those found in previous phase I and II studies (12,16,17) and demonstrate feruglose to be a safe and generally well-tolerated MR contrast agent when administered at a dose of 2 mg of iron per kilogram of body weight. Additional investigations are necessary if higher doses are to be applied.

Pathologic characteristics of the breast tumors evaluated in the current study cover a broad spectrum of tumor grades, from benign to highly undifferentiated malignant. However, we did not evaluate the feruglose enhancement pattern of ductal carcinoma in situ. The exact anatomic location of two cases of ductal carcinoma in situ found at the edge of excised invasive breast cancers could not be exactly associated with the findings on MR images. Although ductal carcinoma in situ was not explicitly excluded, the investigators tended to include mainly lesions that appeared as "mass lesions" at mammography (see Materials and Methods) or those that necessitated routine preoperative MR imaging. Thus, patients with microcalcifications only, which are not indications for gadolinium-enhanced MR imaging of the breast, were not included.

The mean (21-mm) diameter and median (16-mm) diameter of the lesions evaluated in the current study are in good agreement with findings in a previous study (8). However, only two lesions with a diameter of less than 1 cm were evaluated. On one hand, this prevented problems with quantitative analyses due to partial volume effects. On the other hand, results cannot be directly transferred to very small lesions but have to be proved for a larger number of small lesions.

Microvascular hyperpermeability is not unique to neoplasms but has also been demonstrated in inflammation, ischemia, and wound healing (10,42). In the current study, two inflammatory lesions showed "false-positive" hyperpermeability to feruglose. Thus, our MR imaging method is not suited for characterization of breast tumors when clinical signs of inflammation are apparent or after core biopsy, when granulation tissue is present. However, our MR imaging technique may be useful to grade the activity and severity of these other pathologic conditions (10,42).

In this study, care was taken to acquire the enhancement profile of feruglose in human breast tissue as accurately as possible from the early arterial phase up to the equilibrium phase. However, the length of this procedure, as well as the subsequent kinetic analyses, might create practical problems for routine applications. A feasible clinical application could be acquisition of dynamic data up to 15–20 minutes after feruglose injection. If these data do not provide sufficient information concerning the presence or absence of tumor microvascular hyperpermeability, delayed postcontrast MR imaging (eg, 1 hour after injection) could be added. Kinetic analyses can be generated by means of dedicated software programs that automatically transform the acquired dynamic data into fractional plasma volume and KPS maps.

In summary, to our knowledge this is the first clinical study on in vivo quantification of breast tumor microvascular characteristics with macromolecular contrast medium–enhanced MR imaging. Quantitative measurements of tumor microvascular permeability, which can be depicted as KPS maps, may be used as an adjunctive diagnostic tool for breast tumor characterization. Preoperatively, macromolecular contrast-enhanced MR imaging may help predict the malignant potential and biologic behavior of breast cancers, depict the most dedifferentiated tumor areas, and help guide selective biopsies by using puncture devices compatible with MR imaging. During chemotherapy or antiangiogenesis therapy, this MR imaging method may also permit noninvasive follow-up studies of tumor angiogenesis. Further studies are needed to optimize the imaging technique, contrast agent molecular size and dose, and data postprocessing to further increase the specificity of this method.


   FOOTNOTES
 
Abbreviations: SBR = Scarff-Bloom-Richardson, SI = signal intensity

Author contributions: Guarantors of integrity of entire study, H.E.D.L., J.R., T.H.H.; study concepts, H.E.D.L., J.R., T.H.H., A.B., R.C.B., O.H., E.J.R.; study design, all authors; literature research, H.E.D.L.; clinical studies, H.E.D.L., J.R., T.H.H., K.T., E.K., T.M.L., O.H., E.J.R.; data acquisition, H.E.D.L., J.R., T.H.H., K.T., E.K.; data analysis/interpretation, H.E.D.L., J.R., T.H.H., A.K., K.T., E.K., D.S.; statistical analysis, K.S.; manuscript preparation, H.E.D.L.; manuscript definition of intellectual content, H.E.D.L., J.R., T.H.H., R.C.B., O.H., E.J.R.; manuscript editing, R.C.B., E.J.R.; manuscript revision/review and final version approval, all authors


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Kaiser W, Zeitler E. MR imaging of the breast: fast imaging sequences with and without Gd-DTPA. Radiology 1989; 170:681-686.[Abstract/Free Full Text]
  2. Heywang-Köbrunner S, Beck R, Lommatsch B. Contrast enhanced MR imaging of the breast: survey of 1200 patient examinations. Radiology 1992; 185:246-252.
  3. Heywang-Köbrunner S. Contrast enhanced MR imaging of the breast. Invest Radiol 1994; 29:94-109.[CrossRef][Medline]
  4. Helbich T, Roberts T, Gossmann A, et al. Quantitative gadopentetate-enhanced MRI of breast tumors: testing of different analytic methods. Magn Reson Med 2000; 44:915-924.[CrossRef][Medline]
  5. Greenstein Orel S, Schnall M, LiVolsi V, Troupin R. Suspicious breast lesions: MR imaging with pathologic correlation. Radiology 1994; 190:485-493.[Abstract/Free Full Text]
  6. Kelcz F, Santyr GE, Mongin SJ, Fairbanks EJ. Reducing false positive gadolinium enhanced breast MRI results through parameter analysis of the enhancement profile (abstr) In: Proceedings of the 12th Annual Scientific Meeting of the Society of Magnetic Resonance in Medicine. Berkeley, Calif: Society of Magnetic Resonance in Medicine, 1993; 121.
  7. Kaiser W, Reiser M. False positive cases in dynamic MR mammography (abstr). Radiology 1992; 185(P):934.
  8. Heywang-Köbrunner S, Bick U, Bradley W, Jr, et al. International investigation of breast MRI: results of a multicentre study (11 sites) concerning diagnostic parameters for contrast-enhanced MR imaging based on 519 histopathologically correlated lesions. Eur Radiol 2001; 11:531-546.[CrossRef][Medline]
  9. Daldrup H, Shames D, Wendland M, et al. Correlation of dynamic contrast-enhanced magnetic resonance imaging with histologic tumor grade: comparison of macromolecular and small molecular contrast media. AJR Am J Roentgenol 1998; 171:941-949.[Abstract/Free Full Text]
  10. Brasch R, Shames D, Cohen F, et al. Quantification of capillary permeability to macromolecular magnetic resonance imaging contrast media in experimental mammary adenocarcinomas. Invest Radiol 1994; 29(suppl 2):8-11.
  11. Shames D, Kuwatsuru R, Vexler V, Muehler A, Brasch R. Measurement of capillary permeability to macromolecules by dynamic magnetic resonance imaging: a quantitative non-invasive technique. Magn Reson Med 1993; 29:616-622.[Medline]
  12. Taylor A, Panting J, Keegan J, et al. Safety and preliminary findings with the intravascular contrast agent NC100150 injection for MR coronary angiography. J Magn Reson Imaging 1999; 9:220-227.[CrossRef][Medline]
  13. Turetschek K, Huber S, Floyd E, et al. MR imaging characterization of microvessels in experimental breast tumors by using a particulate contrast agent with histopathologic correlation. Radiology 2001; 218:562-569.[Abstract/Free Full Text]
  14. Weidner N, Semple J, Welch W, Folkmann J. Tumor angiogenesis and metastasis: correlation in invasive breast carcinoma. N Engl J Med 1991; 324:1-8.[Abstract]
  15. Kellar K, Jujii D, Guther W, et al. "NC100150 injection," a preparation of optimized iron oxide nanoparticles for positive contrast MR angiography. J Magn Reson Imaging 2000; 11:488-494.[CrossRef][Medline]
  16. Saeed M, Wendland M, Engelbrecht M, Sakuma H, Higgins C. Value of blood-pool contrast agents in magnetic resonance angiography of the pelvis and lower extremities. Eur Radiol 1998; 8:1047-1053.[CrossRef][Medline]
  17. Johansson L, Akeson P, Ragnarsson A, Ahlström H. Myocardial perfusion using a new superparamagnetic iron oxide (abstr) In: Proceedings of the Sixth Scientific Meeting of the International Society for Magnetic Resonance in Medicine 1998. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1998; 903.
  18. Doussal VL, Tubiana-Hulin M, Friedman S, Hacene K, Spyratos F, Brunet M. Prognostic value of histologic grade nuclear components of Scarff-Bloom-Richardson (SBR). Cancer 1989; 64:1914-1921.[CrossRef][Medline]
  19. Howel L, Gandour-Edwards R, O’Sullivan D. Application of the Scarff-Bloom-Richardson tumor grading system to fine needle aspirates of the breast. Am J Clin Pathol 1994; 101:262-265.[Medline]
  20. Roberts TPL. Physiologic measurements by contrast-emhanced MR imaging: expectations and limitations. J Magn Reson Imaging 1997; 7:82-90.[Medline]
  21. Cobelli C, Foster D, Toffolo G. Tracer kinetics in biomedical research: from data to model Dordrecht, the Netherlands: Kluwer Academic/Plenum, 2000.
  22. Patlak C, Blasberg R, Fenstermacher J. Graphical evaluation of blood-to-brain transfer constants from multiple time uptake data. J Cereb Blood Flow Metab 1983; 3:1-7.[Medline]
  23. Patlak C, Blasberg R. Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data: generalizations. J Cereb Blood Flow Metab 1985; 5:584-590.[Medline]
  24. Groscurth P, Kristler G. Human renal cell carcinoma in the nude mouse: long term observations. Beitr Pathol 1977; 160:337-360. (German).[Medline]
  25. Jain R. Transport of molecules across tumor vasculature. Cancer Metastasis Rev 1987; 6:559-593.[CrossRef][Medline]
  26. Jain R. Determinants of tumor blood flow: a review. Cancer Res 1988; 48:2641-2658.[Abstract/Free Full Text]
  27. Warren B. The ultrastructure of the advancing edge of Walker 256 carcinoma. Microvasc Res 1970; 2:443-453.[CrossRef][Medline]
  28. Warren B, Chauvin W. Transmission and scanning electron microscopy of renal adenocarcinomas. Ann R Coll Phys Surg Can 1977; 10:74.
  29. Vogel A. Intratumoral vascular changes with increased size of a mammary tumor: new methods and results. J Natl Cancer Inst 1965; 34:571-578.
  30. Turetschek K, Roberts T, Floyd E, et al. Tumor microvascular characterization using ultra small superparamagnetic iron oxide particles (USPIO) in an experimental breast cancer model. J Magn Reson Imaging 2001; 13:882-888.[CrossRef][Medline]
  31. Dvorak H, Nagy J, Dvorak J, Dvorak A. Identification and characterization of the blood vessels of solid tumors that are leaky to circulating macromolecules. Am J Pathol 1988; 133:95-109.[Abstract]
  32. Gullino P, Grantham F. The vascular space of growth tumors. Cancer Res 1964; 24:1727-1732.
  33. Hilmas D, Gillette E. Morphometric analysis of the microvasculature of tumors during growth and after irradiation. Cancer 1974; 33:103-110.[CrossRef][Medline]
  34. Boucher Y, Jain R. Microvascular pressure is the principal driving force for interstitial hypertension in solid tumors: implications for vascular collapse. Cancer Res 1992; 52:5110-5114.[Abstract/Free Full Text]
  35. Folkman J. Angiogenesis and breast cancer. J Clin Oncol 1994; 12:441-443.[Medline]
  36. Brasch R, Li K, Husband J, et al. In vivo monitoring of tumor angiogenesis with MR Imaging. Acad Radiol 2000; 7:812-823.[CrossRef][Medline]
  37. Jain R. Vascular and interstitial barriers to delivery of therapeutic agents in tumors. Cancer Metastasis Rev 1990; 9:253-266.[CrossRef][Medline]
  38. Jain R. Barriers to drug delivery in solid tumors. Sci Am 1994; 271:58-65.[Medline]
  39. Buadu LD, Murakami J, Murayama S, et al. Breast lesions: correlation of contrast medium enhancement patterns on MR images with histopathologic findings and tumor angiogenesis. Radiology 1996; 200:639-649.[Abstract/Free Full Text]
  40. Delille JP, Slanetz PJ, Yeh ED, Kopans DB, Garrido L. Breast cancer: regional blood flow and blood volume measured with magnetic susceptibility-based MR imaging—initial results. Radiology 2002; 223:558-565.[Abstract/Free Full Text]
  41. Kinkel K, Hylton NM. Challenges to interpretation of breast MRI. J Magn Reson Imaging 2001; 13:821-829.[CrossRef][Medline]
  42. Brasch R, Berthezene Y, Vexler V, et al. Pulmonary oxygen toxicity: demonstration of abnormal capillary permeability using contrast enhanced MRI. Pediatr Radiol 1993; 23:495-500.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
H.-J. Raatschen, G. H. Simon, Y. Fu, B. Sennino, D. M. Shames, M. F. Wendland, D. M. McDonald, and R. C. Brasch
Vascular Permeability during Antiangiogenesis Treatment: MR Imaging Assay Results as Biomarker for Subsequent Tumor Growth in Rats
Radiology, May 1, 2008; 247(2): 391 - 399.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
N. H. G. M. Peters, I. H. M. Borel Rinkes, N. P. A. Zuithoff, W. P. T. M. Mali, K. G. M. Moons, and P. H. M. Peeters
Meta-Analysis of MR Imaging in the Diagnosis of Breast Lesions
Radiology, December 1, 2007; 246(1): 116 - 124.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
D. J. A. Margolis, J. M. Hoffman, R. J. Herfkens, R. B. Jeffrey, A. Quon, and S. S. Gambhir
Molecular Imaging Techniques in Body Imaging
Radiology, November 1, 2007; 245(2): 333 - 356.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
L. S. Fournier, V. Novikov, V. Lucidi, Y. Fu, T. Miller, E. Floyd, D. M. Shames, and R. C. Brasch
MR Monitoring of Cyclooxygenase-2 Inhibition of Angiogenesis in a Human Breast Cancer Model in Rats
Radiology, April 1, 2007; 243(1): 105 - 111.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
V. Goh, S. Halligan, and C. I. Bartram
Quantitative Tumor Perfusion Assessment with Multidetector CT: Are Measurements from Two Commercial Software Packages Interchangeable?
Radiology, March 1, 2007; 242(3): 777 - 782.
[Abstract] [Full Text] [PDF]

Home page
 J Ultrasound MedHome page
T. E. Yankeelov, K. J. Niermann, J. Huamani, D. W. Kim, C. C. Quarles, A. C. Fleischer, D. E. Hallahan, R. R. Price, and J. C. Gore
Correlation Between Estimates of Tumor Perfusion From Microbubble Contrast-Enhanced Sonography and Dynamic Contrast-Enhanced Magnetic Resonance Imaging
J. Ultrasound Med., April 1, 2006; 25(4): 487 - 497.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Radiol.Home page
A R Padhani
MRI for assessing antivascular cancer treatments
Br. J. Radiol., December 1, 2003; 76(suppl_1): S60 - S80.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2293021045v1
229/3/885    most recent
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 Daldrup-Link, H. E.
Right arrow Articles by Rummeny, E. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Daldrup-Link, H. E.
Right arrow Articles by Rummeny, E. J.

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