| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroradiology |
1 From the Dept of Radiology, Ohio State Univ Hospitals, 657 Means Hall, 1654 Upham Dr, Columbus, OH 43210-1228 (M.V.K.); Scott and White Clinic and Hospital, Texas A&M Univ Health Science Ctr, Temple (V.M.R.); Dept of Radiology, German Cancer Research Ctr, Heidelberg (M.E.); Dept of Neuroradiology, Univ of Heidelberg, Germany (M.H.); Dept of Neuroradiology, Univ of Kiel, Germany (O.J.); Worldwide Medical Affairs, Bracco Imaging, Milan, Italy (M.A.K.); Medidata, Konstanz, Germany (A.M.); Bracco Altana Pharma, Konstanz, Germany (A.H.S., K.P.L.). Received Aug 30, 2002; revision requested Oct 28; final revision received May 13, 2003; accepted June 9. Address correspondence to M.V.K. (e-mail: knopp-1@medctr.osu.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Twenty-seven patients with either high-grade glioma or metastases were enrolled in a bicentric intraindividual crossover study to compare lesion enhancement with doses of 0.1 mmol per kilogram of body weight of 0.5 mol/L gadopentetate dimeglumine and 0.5 mol/L gadobenate dimeglumine. MR imaging was performed before injection (T1-weighted spin-echo [SE] and T2-weighted fast SE acquisitions) and at 1, 3, 5, 7, 9, and 16 minutes after injection (T1-weighted SE acquisitions). Qualitative assessment was performed by blinded off-site readers (for 22 patients) and on-site investigators (for 24 patients) in terms of global contrast enhancement, lesion-to-brain contrast, lesion delineation, internal lesion morphology and structure, tumor vascularization, and global image preference. Additional quantitative assessment with region-of-interest analysis was performed by off-site readers alone. Statistical analysis of qualitative data was performed with the Wilcoxon signed rank test, whereas a nonparametric approach was adopted for analysis of quantitative data.
RESULTS: Significant (P < .05) preference for gadobenate dimeglumine over gadopentetate dimeglumine was noted both off-site and on-site for the global assessment of contrast enhancement. For off-site readers 1 and 2 and the on-site investigators, respectively, gadobenate dimeglumine was preferred in 13, 17, and 16 patients; gadopentetate dimeglumine was preferred in four, four, and four patients; and equality was found in five, one, and four patients). Similar preference for gadobenate dimeglumine was noted by off-site readers and on-site investigators for lesion-to-brain contrast and all other qualitative parameters. Off-site quantitative evaluation revealed significantly (P < .05) superior enhancement for gadobenate dimeglumine compared with that for gadopentetate dimeglumine at all time points from 3 minutes after injection.
CONCLUSION: Significantly superior contrast enhancement of intraaxial enhancing brain tumors was achieved with 0.1 mmol/kg gadobenate dimeglumine compared with that with 0.1 mmol/kg gadopentetate dimeglumine.
© RSNA, 2004
Index terms: Brain neoplasms, MR, 10.12143, 10.30 • Gadolinium • Magnetic resonance (MR), contrast media, 10.12143
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recent developments in MR contrast agent design have led to gadolinium-based compounds that differ from currently available agents by possessing increased in vivo T1 relaxivity that derives from a capacity for interaction with serum albumin. One such agent is gadobenate dimeglumine (MultiHance; Bracco Imaging, Milan, Italy), which is currently approved in Europe for MR imaging of the central nervous system and liver. This agent has a capacity for weak and transient interaction with serum albumin, which leads to an in vivo T1 relaxation rate that is approximately twice that of the standard gadolinium chelates: r1 = 9.7 (mmol/ L)-1 · sec-1 for gadobenate dimeglumine compared with between 4.3 and 5.6 (mmol/L)-1 · sec-1 for purely extracellular agents (15–18). Results in evaluations of vascular enhancement properties in animals and in phase I trials indicate that this agent may have the potential to improve contrast (15,19).
The purpose of our study was to evaluate the safety of and compare the enhancement characteristics of gadobenate dimeglumine with those of a standard gadolinium chelate (gadopentetate dimeglumine, Magnevist; Schering, Berlin, Germany) in primary and secondary brain tumors on the basis of qualitative and quantitative parameters, on an intraindividual basis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study Patients
To ascertain the number of patients to include in the study, preliminary power analysis was performed. Results showed that 24 patients would be required for a relevant study power of 80%. Assumptions were that no difference would be apparent for 50% of the assessments and that a proportion of 1:4 in favor of either contrast agent would be a clinically relevant proportion.
To allow for possible patient dropouts, 27 consecutive patients known to have an intraaxial glioma (n = 15) or intracerebral metastasis at least 10 mm in diameter (n = 12), as depicted at previous computed tomography (CT) or MR imaging, were enrolled in the study. Before being included, each patient was adequately informed of the aims, methods, anticipated benefits, and potential risks of the study, as well as the discomfort it might entail. Patients were excluded if they had undergone radiation therapy in the 12 months preceding this study, if they had undergone an intravascular contrast material–enhanced MR imaging procedure within 48 hours preceding this study, or if less than 90% of the initially detected tumor mass remained after tumor biopsy.
After inclusion in the study, patients were randomly assigned to one of two study groups. Of the 27 patients enrolled, 12 were allocated to group A and the remaining 15 patients to group B. Patients in group A were to receive gadobenate dimeglumine as the first contrast agent and gadopentetate dimeglumine as the second agent, while patients in group B were to receive the agents in the reverse order. All 27 patients underwent at least one contrast-enhanced MR examination, and all were included in the analysis of safety.
Three patients in group B withdrew prematurely from the study after administration of the first contrast agent (gadopentetate dimeglumine); two of these patients were not willing to undergo a second MR examination for personal reasons, and the third patient withdrew because of bleeding into the metastasis before the second MR examination. Therefore, on-site effectiveness evaluation was performed for 24 patients. Two other patients were withdrawn from the off-site assessment because of protocol violations: one patient in group A because of differences in section thickness between the first and second MR examinations and one patient in group B because of technical failure of the precontrast T1-weighted sequence in the first MR examination. Therefore, the final off-site effectiveness population comprised 22 patients: 11 patients in group A (four women [mean age, 58.0 years ± 17.3 [SD]; range, 33–73 years] and seven men [mean age, 53.4 years ± 10.6; range, 35–68 years]) and 11 patients in group B (five women [mean age, 58.6 years ± 16.5; range, 36–74 years] and six men [mean age, 50.0 years ± 12.4; range, 27–63 years]). There were no significant age or sex differences between the two groups.
The off-site effectiveness population comprised eight patients with cerebral metastases (deriving in two patients from primary adenocarcinoma of the lung; in one patient each from primary carcinoma of the breast, cervix, colon, and kidney; in one patient from malignant melanoma; and in the remaining patient from a tumor of unknown origin) and 14 patients with anaplastic, contrast-enhancing cerebral gliomas (World Health Organization stages III and IV). Three patients with gliomas had undergone radiation therapy 3–6 years before enrollment in the study. Diseases in the study population were evenly distributed between groups A and B, with four patients with metastases and seven with gliomas in each group.
MR Imaging
Two 1.5-T MR imaging systems were used in the study: Magnetom Vision (Siemens Medical Systems, Erlangen, Germany) for 15 patients, and Edge (Marconi, Cleveland, Ohio) for 12 patients. A series of phantom tests and comparisons of patient images were performed prior to patient enrollment to verify identical imaging characteristics of the two systems. The pre- and postcontrast examinations of individual patients in the study were performed with the same system.
Imaging was performed in the transverse plane with standardized parameters. Care was taken to ensure that patients were positioned identically for the two examinations and that identical brain sections were imaged. Precontrast imaging sequences in both MR examinations comprised T1-weighted spin echo (SE) (repetition time msec/echo time msec = 500/15) and T2-weighted fast SE (3,000/90). Postcontrast image acquisition began with the injection of contrast medium, and each image set was acquired in 112 seconds with conventional k-space readout (linear sequential traversal of k space with k = 0 in the middle of the acquisition). Therefore, the relevant reference time after injection was considered to be at 1 minute after start of the sequence. Postcontrast images acquired in both MR examinations comprised a series of five T1-weighted SE images ("T1-weighted SE loop" images) obtained at 1, 3, 5, 7, and 9 minutes after injection and late T1-weighted SE images acquired at 16 minutes after injection.
In all patients, the postcontrast imaging parameters were identical to those used for acquisition of the precontrast T1-weighted SE images. Section orientation was parallel to the anterior-posterior commisure line in all patients. Eighteen sections with a section thickness of 5 mm and an intersection gap of 25% (1.25 mm) were acquired. Given the comparatively large intersection gap, care was taken to ensure that the section alignment in the two examinations was as accurate and comparable as possible. A field of view of 250 mm and a matrix size of 256 x 256 were used in all patients. The overall times of acquisition were 112 seconds for the T1-weighted SE sequences (one signal acquired) and 121 seconds for the T2-weighted fast SE sequence (one signal acquired). Presaturation was used only for the T2-weighted fast SE sequence.
The contrast agent in each MR examination was administered intravenously at an injection rate of 2 mL/sec by using a power injector (Tomoset, Brukea, Ettlingen, Germany, or Spectris, Medrad, Indianola, Pa) and a 20-gauge needle. Each contrast agent was administered at a dose of 0.1 mmol/kg (corresponding to 0.2 mL/kg of a 0.5 mol/L formulation) according to the randomization list. The interval between the two MR imaging examinations was greater than 24 hours in all patients. For 16 of 22 patients, the mean interval between examinations was 59.6 hours (range, 45–96 hours). For the remaining six patients, the intervals were 24–25 hours (n = 3) or 6–13 days (n = 3). The longest interval between examinations (13 days) occurred for a patient with primary intraaxial glioma.
Image Evaluation
All available images were evaluated on-site by the principal investigator at each center (O.J., M.E.) and off-site by two independent radiologists with 15 (reader 1) and 12 (reader 2) years of experience. The evaluation criteria for the on-site investigators were the same as those for the off-site readers. The off-site readers were blinded to the results of the on-site evaluations and to all patient and contrast agent information. All images were stored digitally by using reference identification. Prior to the off-site assessment, all images were transferred to a central off-site processing center for further randomization. This procedure ensured that no information regarding imaging equipment, contrast agent, or patient was available to the off-site readers. Off-site quantitative and qualitative assessment was performed in a fully electronic monitored fashion with large-screen high-resolution cathode-ray tube displays, which enabled review of 16 images simultaneously. The two readers performed their independent blinded assessments in isolation in separate rooms on the same day. No communication of any kind of associated information was permitted.
All images from each patient examination were made available to the off-site readers for evaluation. Analysis was performed only for patients whose images from the two MR imaging examinations were, in the independent opinion of the readers, comparable in terms of orientation, region of imaging, and section positioning and for whom the detected lesions were comparable in terms of size, extension, and morphology.
The technical adequacy of the images was evaluated by the on-site principal investigators and both off-site readers. Evaluation was performed separately for the precontrast images, the T1-weighted SE loop images, and the late T1-weighted SE images. Images were rated as 1 = excellent, 2 = adequate (with artifacts but tolerable for assessment), or 3 = inadequate (not tolerable for further evaluation). Lesions were also evaluated in terms of size (in millimeters) and diagnosis (glioma or metastasis or, if the reader considered a lesion to be neither, then other or unknown). Quantitative assessment by the off-site readers was performed only for lesions considered suitable for the placement of regions of interest (ROIs) (ie, larger lesions with as homogeneous an appearance as possible).
Qualitative Assessment
Qualitative analysis comprised a dedicated simultaneous matched-pairs assessment from both MR imaging examinations of all five postcontrast T1-weighted SE loop images and the late T1-weighted SE images together.
An initial global assessment of contrast enhancement (defined as the increased difference in contrast between lesion and surrounding normal tissue after contrast agent administration) was performed by both off-site readers and the on-site investigators for all MR images combined by using a continuous linear scale: "MR study 1 better than MR study 2," "both MR studies equal", or "MR study 2 better than MR study 1." The impression of each reader was quantified on a scale ranging from -90 mm to +90 mm by measuring the marked position in millimeters from the center. Subsequent to this evaluation, assessments were performed separately for all T1-weighted SE loop images together and for the late T1-weighted SE images. Images from the two MR imaging examinations were thereafter compared for (a) degree of lesion-to-brain contrast, (b) degree of lesion delineation from surrounding tissue and edema, (c) degree of information about internal morphology and structure, (d) degree of information for tumor vascularization, and (e) time dependence of contrast enhancement. The same continuous scale was used in each patient, and, again, images were evaluated by both off-site readers and on-site investigators.
Finally, an evaluation of the global preference of the on-site investigators and off-site readers was conducted. For the off-site readers, this was performed both in an independent manner and in consensus after completion of all other assessments. Again, a continuous scale was used "MR study 1 preferred over MR study 2," "both MR studies preferred equally," or "MR study 2 preferred over MR study 1." Consensus evaluation of global preference was performed with the images displayed in a newly randomized order.
Quantitative Assessment
Quantitative analysis by the off-site readers was similarly based on a simultaneous matched-pairs assessment. Signal intensity measurements were recorded in ROIs positioned on (a) the enhancing tumor tissue and (b) the normal white matter (typical areas of lesion-free brain parenchyma without strong enhancing structures or artifacts). ROIs were placed on no more than three lesions per patient, and, if feasible, larger lesions were chosen with as homogeneous an appearance as possible. Additional ROIs were placed on selected areas external to the skull to determine the background noise. In all but one patient, the external ROIs were placed on the left side of the skull in areas deemed to be free of artifacts. In the remaining patient, the external ROI was placed on the right side of the skull.
For the placement of ROIs on images, the corresponding sections of all eight patient images (precontrast T1-weighted SE and T2-weighted fast SE images; five T1-weighted SE loop images obtained at 1, 3, 5, 7, and 9 minutes after injection; and late [16 minutes after injection] T1-weighted SE images) were presented for both examinations simultaneously in the same section on three adjacent cathode-ray tube screens. The reader was asked to select the most suitable image for ROI placement, and then circular or elliptic ROIs were positioned on the whole, or the largest possible extent, of the enhancing lesion(s). The ROI then appeared simultaneously on each of the other 15 images in an exactly corresponding position and size.
To correct for any movement of the patient, the section position of each ROI could then be adjusted manually as was deemed appropriate by the reader. Similarly, each reader was able to adjust the size of the ROI and the window and level settings as appropriate. Since the primary objective of the study was to compare the enhancement achieved with the two contrast agents, each ROI was positioned with utmost care on the region of maximum enhancement in each lesion. Inclusion of vessels or other strongly contrasting structures or artifacts in the ROI was strictly avoided. For extensive lesions with an irregular circumference, the ROI was positioned to include the largest possible extent of the lesion.
The ROI data were used to calculate lesion-to-brain contrast and differences from pre- to postcontrast lesion signal intensity.
Statistical Analysis
Overall.—Data analysis was performed (SAS, version 6.12; SAS Institute, Cary, NC). The primary effectiveness end point (global assessment of contrast enhancement) was evaluated by means of the Wilcoxon signed rank test at a level of significance of P < .05.
Signal intensity measurements were analyzed according to the crossover approach of Koch (20). Since the postcontrast values were not normally distributed, analysis was performed nonparametrically. Evaluations were performed with the Wilcoxon-Mann-Whitney U test to determine differences in carry-over effects (ie, interindividual comparison of the sum of values from examination 1 plus that for examination 2 for group A with the corresponding sum for group B) and period effects (ie, interindividual comparison of the absolute difference of values from examination 1 minus that for examination 2 for group A with the corresponding difference for group B). Differences in contrast enhancement properties (ie, intraindividual comparison of values from all gadobenate dimeglumine measurements with those from all gadopentetate dimeglumine measurements) were determined with the Wilcoxon signed rank test.
Assessment of reader agreement for global contrast enhancement was presented as a 3 x 3 cross table. Interreader variability was determined by means of the weighted Cohen 
statistic (21).
ROI signal intensities.—A different approach was implemented for the evaluation of ROI signal intensities. In many publications to date, quantitative assessments are based on the arithmetic mean values of ROI pixel intensities (1,22,23). However, local lesion inhomogeneity may lead to substantial distortion of these quantitative evaluations. For example, distortions may occur if ROIs include either nonenhancing necrotic tissues, which falsely lower the mean signal intensity values, or strongly enhancing structures such as blood vessels, which falsely exaggerate them. These disturbances misleadingly increase the SD of mean values and may explain why statistically significant differences are demonstrated less frequently in comparative contrast agent studies than results of qualitative assessments might lead one to expect (24–26).
To better evaluate the true enhancement characteristics, a nonparametric approach was implemented in which different decile levels were assessed. The 90% decile value was selected for statistical evaluation rather than the mean of ROI pixel signal intensities, since this value can be considered to represent signal intensity enhancement equally well for both contrast agents while excluding possible signal intensity distortions caused by blood vessels, necrotic areas, or normal tissue.
The quantitative data in the present study are patient related rather than lesion related. To maintain this relationship in patients with more than one lesion, all pixel values in all ROIs were pooled for each sequence before the pixels were sorted into decile groups. ROI pixel intensity values were normalized against those of normal white matter. Owing to the presence of an intact blood-brain barrier, normal white matter shows no contrast enhancement, and thus its signal intensity can be expected to be reproducible. When measuring the signal intensity of the white matter of healthy brain parenchyma, the readers took particular care to ensure that each ROI was free of enhancing structures, including possible lesion-related hemorrhage.
Safety
Safety was evaluated by the on-site principal investigators on the basis of the incidence of adverse events according to the established guidelines for clinical trials. Adverse events were classified as either serious (ie, death, life-threatening event, or event that necessitates or prolongs hospitalization) or nonserious (event rated as mild, moderate, or severe). The relationship of each adverse event to the study agent was classified as probable, possible, not related, or unknown.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
26 mm) on MR images obtained in both examinations, in each patient, the largest size (4–8 mm bigger) was noted following gadobenate dimeglumine administration. The lesions were gliomas in all but one patient. No evidence of any residual enhancement from the first MR examination was noted by either blinded reader for any of the 19 patients for whom the interval between contrast agent injections was more than 48 hours. Similarly, no evidence of any residual enhancement was noted by either reader for two of the three patients for whom the interval between examinations was approximately 24 hours. For the remaining patient, minimal visual residual enhancement was reported by one of the two readers. However, this finding was not considered to be sufficient to unduly influence any of the subsequent evaluations.
Qualitative Evaluation
Global assessment of contrast enhancement.—Results of the global assessment of contrast enhancement are presented in Table 1. The on-site investigators and both off-site readers considered the global contrast enhancement to be better with gadobenate dimeglumine in the majority of patients. For each reader, results of the evaluation were statistically significant (P < .05). On the basis of this assessment parameter, the power of the results was determined to be 82% for the on-site investigators and 62% and 87% for off-site readers 1 and 2, respectively.
|
= 0.51). Complete agreement with the on-site investigators was also achieved for 14 patients (reader 1, = 0.41) and 16 patients (reader 2, = 0.48).
|
|
Lesion delineation.—In general, trends similar to those for the degree of lesion-to-brain contrast were observed for the degree of lesion delineation (Fig 2). However, whereas both the off-site readers and the on-site investigators noted better lesion delineation for more patients with gadobenate dimeglumine than with gadopentetate dimeglumine on both the T1-weighted SE loop images and the late T1-weighted SE images, however, statistically significant superiority was noted by off-site reader 2 only (P = .019 for both the T1-weighted SE loop images and the late T1-weighted SE images, Wilcoxon signed rank test).
|
Tumor vascularization.—Findings of the off-site readers and on-site investigators for the degree of information available on tumor vascularization were similar to those for the level of information available on internal morphology and structure. For both the T1-weighted SE loop images and late T1-weighted SE images, reader 2 reported a statistically significant (P < .05) preference for gadobenate dimeglumine, which was preferred in 17 of 22 patients, while gadopentetate dimeglumine was preferred in four of 22 patients. Less clear differences between the agents were noted by reader 1 and the on-site investigators, although in both cases, gadobenate dimeglumine was preferred for more patients than was gadopentetate dimeglumine. On the T1-weighted SE loop images and late T1-weighted SE images, respectively, reader 1 preferred gadobenate dimeglumine in two and one of 22 patients and preferred gadopentetate dimeglumine in none and none of 22 patients, while the on-site investigators preferred gadobenate dimeglumine in six and five of 22 patients and preferred gadopentetate dimeglumine in three and three of 22 patients.
Time dependence of contrast enhancement.—No significant differences between the two contrast agents were noted by the off-site readers or the on-site investigators in terms of the time dependence for contrast enhancement.
Global preference.—Reader assessments of global preference are presented in Table 3. The global preference of both off-site readers was clearly in favor of gadobenate dimeglumine, although a significant preference was noted only for reader 2 (P < .001). Subsequent blinded consensus reading by the two off-site readers resulted in a significant preference for gadobenate dimeglumine (P < .001). The preference of the on-site investigators was also significantly in favor of gadobenate dimeglumine (P = .007). Representative patients are shown in Figures 3–6.
|
|
|
|
|
The median pre- to postinjection differences of the 90% decile of lesion signal intensity are presented in Table 4. From 3 minutes after injection, both blinded readers noted significantly superior enhancement with gadobenate dimeglumine in comparison to that with gadopentetate dimeglumine (P < .05). Significantly (P < .05) greater enhancement was also noted by both readers at all other deciles. Similar results were obtained in terms of the mean and median signal intensity increase at the different postinjection time points.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Numerous studies have been performed to investigate possible improvements in lesion detection and conspicuity on contrast-enhanced MR images. With standard gadolinium-based contrast agents (ie, gadopentetate dimeglumine, gadoteridol [Gd-HP-DO3A, Prohance; Bracco Imaging, Milan, Italy], gadoteric acid [Gd-DOTA, Dotarem; Laboratoire Guerbet, Aulnay-sous-Bois, France], gadodiamide [Gd-DTPA-BMA, Omniscan; Amersham Health, Amersham, UK], and gadoversetamide [OptiMARK: Mallinckrodt, St Louis, Mo]), benefits in lesion detection and delineation have frequently been achieved with higher doses of contrast material (1,7–14,22,23). However, these benefits need to be balanced against the increased costs incurred with the use of double and sometimes triple doses of contrast material (9,10).
Gadobenate dimeglumine is a gadolinium-based contrast agent with potentially advantageous properties for MR imaging of the central nervous system. The molecular structure of the contrast-effective moiety resembles that of the gadopentetate dimeglumine molecule apart from the presence of a protruding benzyloxymethyl group (17). Whereas this hydrophobic substituent only minimally affects the molecular weight (1,058 g/mol for gadobenate dimeglumine compared with 938 g/mol for gadopentetate dimeglumine) and physicochemical properties of this agent compared with standard gadolinium chelates, it is sufficient to confer a capacity for weak and transient interaction with serum albumin and a resulting markedly increased T1 relaxivity in blood (15–17).
Findings in a number of recent studies (24–26,30,31) suggest that this increased T1 relaxivity may prove beneficial for the detection and delineation of central nervous system lesions, particularly central nervous system metastases. As yet, however, there have been no reports in which gadobenate dimeglumine is compared rigorously and objectively with a standard gadolinium chelate in terms of contrast effectiveness in the clinical setting. The present intraindividual crossover comparison was designed to validate as definitively as possible the contrast enhancement properties of gadobenate dimeglumine relative to a standard gadolinium chelate for MR imaging of the central nervous system. The approach permitted high validity and excellent comparability even in a relatively small patient population. In this regard, the internal review boards were fully supportive of the intraindividual approach, since it was considered a potentially highly relevant comparison of a different imaging approach with current clinical practice.
Results in the current study demonstrate unequivocally that both qualitative and quantitative lesion enhancement parameters are superior on gadobenate dimeglumine–enhanced images compared with gadopentetate dimeglumine–enhanced images. Of note, for a relatively small patient population of 22 patients, substantial agreement was noted not only between the two blinded off-site readers but also between each off-site reader and the on-site investigators for the global evaluation of contrast enhancement. The fact that only moderate values were determined for the level of interreader agreement can be considered to be a result of the limited number of patients included in the study.
For the imaging of cerebral gliomas, the main goal is not merely detection of the tumor alone. These lesions are virtually always identifiable on the basis of mass effect and signal abnormalities. Rather, the goal with these lesions is to identify as accurately as possible the enhancing tumor margins to guide potential surgical resection or to delineate the radiosurgical target volumes (32–34). Enhancing regions of gliomas correspond histologically to the hypervascular tissue of viable tumor. This hypervascular tissue has high cellular density and is the principal target tissue for therapeutic interventions, although it is well known that glial tumors extend beyond the contrast enhancement and signal intensity margins on T2-weighted MR images and that improved therapy is associated with aggressive surgical approaches, which often involve resection of normal-appearing and functioning brain tissue (35,36).
Results in previous studies by Yuh and Maley (7) and Yuh et al (11) revealed a proportional relationship between the dose of contrast agent and the degree of enhancement observed in primary intraaxial tumors. In the high-dose groups in particular, extension of enhancement beyond the margins of the T2-weighted signal intensity changes was observed. The enhancement seen at tumor margins may be indicative of a subtle loss of blood-brain barrier integrity associated with tumor infiltration. Moreover, if the zone delineated by enhancement on high-dose contrast-enhanced MR images extends beyond the margins on nonenhanced MR images, results may correspond more precisely to the microscopic extent of the tumor.
In the patients with glioma in the present study, improved contrast enhancement and superior tumor delineation were noted more frequently on gadobenate dimeglumine–enhanced images than on gadopentetate dimeglumine–enhanced images. Unlike the latter, the former possesses a capacity for weak and transient interaction with serum albumin, which results in an approximately twofold greater T1 relaxivity compared with that of the latter and other standard gadolinium-based contrast agents (15–17,37). It is likely that the improved tumor delineation with gadobenate dimeglumine in the present study is due to weak protein interaction, which is more pronounced in smaller more distal vessels and tumor microvasculature. Such a conclusion is supported by the significantly improved lesion-to-brain contrast with gadobenate dimeglumine at all time points from 3 minutes after injection. An ability to improve enhancement of the tumor microvasculature, particularly in distal areas beyond the contrast-enhanced margins with gadopentetate dimeglumine, may help explain the slightly larger size of certain large gliomas following gadobenate dimeglumine injection.
Intracranial metastases occur in approximately 25% of all patients with cancer and account for 40% of all adult brain neoplasms (38). Up to 50% of patients with intracerebral metastases have only one lesion (39). Approximately 20% of cerebral metastases are diagnosed prior to or concurrently with the detection of the primary tumor, with a further 50% diagnosed within the 1st year after detection of the primary tumor. It is therefore apparent that some metastases are present at the time of the initial diagnosis but are not detected on concurrent images. In such patients, the improved detection of occult lesions is essential for optimal treatment, cost-effectiveness, and quality of life. Large brain metastases are readily detected on MR images because of associated vasogenic edema and a marked mass effect, but small metastases, particularly those located near the corticomedullary junction, are often not associated with either vasogenic edema or a marked mass effect and thus are frequently not detected (13,27).
Findings in numerous studies (7,8,11–14,23,40) show that, in addition to improving the delineation of larger lesions, higher doses of contrast material can substantially improve the detection of small lesions. In patients who are suspected of having intracranial metastases in particular, it is generally accepted that lesion detection is improved with triple doses of standard gadolinium-based contrast agents (8,12–14). Recently, findings in a number of studies indicate that gadobenate dimeglumine may have advantageous properties over standard gadolinium-based agents for the detection and diagnosis of central nervous system neoplasms in general (25,26) and intracranial metastases in particular (24,30,31).
The present study, with a comparatively small patient population with known intraaxial tumors, was designed principally to compare gadobenate dimeglumine with gadopentetate dimeglumine for contrast enhancement rather than lesion detection. Nevertheless, the improvement in lesion delineation with gadobenate dimeglumine lends support to the conclusions in earlier studies aimed at establishing its potential for improving lesion detection (24,30,31). The exact definition of tumor margins is important for any treatment planning and is especially important in patients with solitary lesions who require highly focused radiation therapy. It is likely that the improved lesion delineation with gadobenate dimeglumine is due to the higher T1 relaxivity of this agent, which derives from its capacity for weak and transient protein interaction, particularly in the microvasculature of smaller lesions.
Limitations of the current study include the lack of any direct evidence of the improved capacity of gadobenate dimeglumine for lesion detection. In addition, we did not attempt to correlate the differences between the two agents for enhancement of glioma margins with pathologic results. Further work might also be directed toward a comparison of the two agents in patients who had previously undergone radiation therapy. In the present study, three patients had undergone radiation therapy 3–6 years before enrollment in this study. Although no obvious differences were noted between these patients and those who had not undergone radiation therapy in terms of the enhancement after injection of the two contrast agents, it is possible that different effects on enhancement might be observed in a larger purposefully designed study.
In conclusion, findings in the present study demonstrate that gadobenate dimeglumine provides significantly better contrast enhancement in patients with primary and secondary brain tumors than does gadopentetate dimeglumine. The physiologic basis of this observation can possibly be ascribed to the capacity of the gadobenate dimeglumine molecule for weak protein interaction. It is likely that the effect of improved contrast enhancement with gadobenate dimeglumine will be seen clinically in the better visualization of small or poorly enhancing lesions and in the improved delineation of larger lesions indicated for surgical intervention; however, further work will be required to verify this.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, K.P.L., A.H.S.; study concepts and design, M.V.K., K.P.L.; literature research, M.V.K., M.A.K.; clinical studies, M.E., M.H., O.J.; data acquisition, all authors; data analysis/interpretation, M.V.K., A.M., K.P.L., M.A.K.; statistical analysis, A.M.; manuscript preparation and definition of intellectual content, M.V.K., M.A.K., K.P.L.; manuscript editing, M.A.K., K.P.L.; manuscript revision/review, M.A.K., M.V.K., K.P.L., V.M.R., M.E., A.H.S., A.M.; manuscript final version approval, M.A.K., M.V.K., K.P.L.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Pediconi, C. Catalano, S. Padula, A. Roselli, V. Dominelli, S. Cagioli, M. A. Kirchin, G. Pirovano, and R. Passariello Contrast-Enhanced MR Mammography: Improved Lesion Detection and Differentiation with Gadobenate Dimeglumine Am. J. Roentgenol., November 1, 2008; 191(5): 1339 - 1346. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.A. Rowley, G. Scialfa, P.-y. Gao, J.A. Maldjian, D. Hassell, M.J. Kuhn, F.J. Wippold II, M. Gallucci, B.C. Bowen, I.M. Schmalfuss, et al. Contrast-Enhanced MR Imaging of Brain Lesions: A Large-Scale Intraindividual Crossover Comparison of Gadobenate Dimeglumine versus Gadodiamide AJNR Am. J. Neuroradiol., October 1, 2008; 29(9): 1684 - 1691. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Sardanelli, M. Quarenghi, T. Schlosser, and J. Barkhausen Delayed Enhancement of Subendocardial Infarcted Myocardium with Gadobenate Dimeglumine: A Paradoxical Effect--Is a Double Dose Too Much? Radiology, September 1, 2006; 240(3): 914 - 916. [Full Text] [PDF]
|
|
|
|
|
|
K. R. Maravilla, J. A. Maldjian, I. M. Schmalfuss, M. J. Kuhn, B. C. Bowen, F. J. Wippold II, V. M. Runge, M. V. Knopp, S. Kremer, L. J. Wolansky, et al. Contrast Enhancement of Central Nervous System Lesions: Multicenter Intraindividual Crossover Comparative Study of Two MR Contrast Agents Radiology, August 1, 2006; 240(2): 389 - 400. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Pediconi, C. Catalano, R. Occhiato, F. Venditti, F. Fraioli, A. Napoli, M. A. Kirchin, and R. Passariello Breast Lesion Detection and Characterization at Contrast-enhanced MR Mammography: Gadobenate Dimeglumine versus Gadopentetate Dimeglumine Radiology, October 1, 2005; 237(1): 45 - 56. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Sardanelli, A. Iozzelli, A. Fausto, A. Carriero, and M. A. Kirchin Gadobenate Dimeglumine-enhanced MR Imaging Breast Vascular Maps: Association between Invasive Cancer and Ipsilateral Increased Vascularity Radiology, June 1, 2005; 235(3): 791 - 797. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. D. Doolittle, L. E. Abrey, W. A. Bleyer, S. Brem, T. P. Davis, P. Dore-Duffy, L. R. Drewes, W. A. Hall, J. M. Hoffman, A. Korfel, et al. New Frontiers in Translational Research in Neuro-oncology and the Blood-Brain Barrier: Report of the Tenth Annual Blood-Brain Barrier Disruption Consortium Meeting Clin. Cancer Res., January 15, 2005; 11(2): 421 - 428. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
