| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Medical Imaging, Institut Gustave Roussy, 39 rue Camille
Desmoulins, 94805 Villejuif, France.
2 Section of Internal Medicine, Department of Internal Medicine and Public
Medicine, University of Bari, Bari, Italy.
3 Department of Surgery, Institut Gustave Roussy, Villejuif, France.
4 Department of Pathology, Institut Gustave Roussy, Villejuif, France.
5 Department of Medicine, Institut Gustave Roussy, Villejuif, France.
Received July 11, 2005;
accepted after revision September 18, 2005.
Presented at the 2003 meeting of the Radiological Society of North America,
Chicago, IL, and at the 2004 meeting of the American Society of Clinical
Oncology, New Orleans, LA. 39 rue Camille Desmoulins, 94805 Villejuif, France.
Address correspondence to N. Lassau
(lassau{at}igr.fr).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Thirty patients (59 tumors) with metastases or a recurrence from a GIST were prospectively included in a single-center imaging trial. Contrast-enhanced Doppler sonography was performed with an Aplio scanner the day before (day-1) starting oral treatment (400 mg) and at days 1, 7, 14, 60, 90, and 6 months, 9 months, and 1 year. The percentage of contrast uptake (Levovist or Sonovue) before treatment and at the different stages of follow-up was evaluated by two radiologists. Digitized quantification was performed using Photoshop software. To define the benchmark standard, all patients were rated as responders or nonresponders at 2 and 6 months by a board consisting of oncologists and radiologists who had all clinical and imaging data at their disposal. Changes in the percentage of contrast uptake at each sonographic examination were compared statistically.
RESULTS. A total of 185 examinations were performed. Forty-four lesions in 24 patients were completely evaluated at 2 months, and 29 lesions in 15 patients were completely evaluated at 6 months. Initial contrast uptake at day 1 was predictive of the future response. A strong correlation was found between the decline in tumor contrast uptake at days 7 and 14 and tumor response (p < 10-4).
CONCLUSION. Contrast-enhanced Doppler sonography is a noninvasive imaging technique that allows the early prediction of tumor response in c-kit-positive GIST treated with Glivec.
Keywords: angiogenesis • contrast media • Doppler sonography • gastrointestinal stromal tumors • gastrointestinal treatment • sonography
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Imatinib (STI 571, Glivec, Gleevec; Novartis Pharmaceuticals) is a newly developed tyrosine kinase inhibitor recently tested in clinical trials on patients with unresectable GISTs [7-10]. Imatinib has dramatically transformed the prognosis of locally advanced inoperable or metastatic GISTs. The objective response rate observed with conventional imaging techniques ranges between 50% and 60%. Patients with stable disease (response attaining 30-40% according the World Health Organization [WHO] criteria) have a similar outcome and are also considered to be responders to imatinib mesylate therapy [11].
This molecularly targeted treatment [12] induces changes in the tumor structure, such as decreased tumor vascularity, hemorrhage, or necrosis, that are consistent with a therapeutic response with or without a change in tumor volume [13]. Classic WHO criteria and the Response Evaluation Criteria in Solid Tumors (RECIST) morphologic criteria based on tumor size measurements often fail to accurately appraise tumor response to imatinib. Morphologic and functional imaging techniques such as contrast-enhanced MRI, CT, or PET should preferably be combined to assess tumor response. Both morphologic and perfusion data are provided by Doppler sonography with contrast agent injection [14]. The size of abdominal tumors can be accurately measured and the percentage of tumor volume taking up contrast agent, a yardstick of tumor vascularity, can be evaluated with this technique [15]. Treatment efficacy is based on an early decline in contrast uptake on successive examinations or on the absence of residual vascularization (Lassau et al., presented at the 2004 annual meeting of the American Society of Clinical Oncology).
The objective of this study was to evaluate contrast-enhanced Doppler sonography as an early predictor of tumor response or primary resistance to imatinib in patients with GISTs that were treated with imatinib (Glivec) according to the French BFR14 trial [16].
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Doppler sonography was performed on the day (day-1) before starting treatment and on days 1, 7, and 14; 2, 3, 6, and 9 months; and 1 year. The mean follow-up period was 145 ± 23 days.
The protocol was approved by our institutional review board, and all patients signed an informed consent form.
Materials
Sonography was performed by two radiologists with an Aplio scanner
(Toshiba) with a 4.4-MHz C37 convex array or a 12-MHz linear transducer
equipped with perfusion software DF (Dynamic Flow, Toshiba), which, because of
broadband Doppler technology, provided imaging of flow with excellent spatial
resolution, a rapid imaging rate, and suppression of the blooming effect. We
now use new software, VRI (Vascular Recognition Imaging, Toshiba), which
couples harmonic imaging with pulse subtraction and dynamic flow.
Methods
Contrast-enhanced Doppler sonography was performed in four steps:
First, the morphologic study was performed in B-mode sonography. This study allowed us to identify the target lesion and to select the best acoustic window for its assessment. The largest diameter of each lesion were measured with calipers, and the tumor volume was measured using the three perpendicular diameters.
Second, the sonographic contrast agent was injected. Before June 2003, we used Levovist (SHU 508 A, Schering) (58 examinations). Levovist is a suspension of micrometer-sized microparticles of galactose and microscopic gaseous bubbles combined with a weak concentration of palmitic acid prepared by shaking 4 g of microparticles in sterilized water. This yields a suspension of 10 mL at a concentration of 400 mg/mL that must be administered via an IV bolus injection. After June 2003, we used Sonovue (BR1, Bracco) (127 examinations) with a more stable encapsulated gas (perfluorocarbon) and Vascular Recognition Imaging. We injected a suspension of 4.8 mL at a concentration of 8 µL/mL that must be administered as an IV bolus injection. When multiple tumors were present in the same patient, target lesions were selected so that several lesions could be evaluated on the same sonographic slice. Patients whose target lesions could not be evaluated on a single slice had a second injection with the same amount of contrast agent but long enough after the first injection (10-15 minutes) to allow the effects of the first injection to disappear.
Third, a dynamic study was performed. The Doppler parameters used remained constant among studies of the same patient and among examinations of different patients. They were set by a program (Vascular Recognition Imaging). The total gain was the only parameter that the operator could modify. This gain was regulated according to the depth of the target lesion. This parameter had no impact on the data used for quantification. The slice passing through the largest tumor diameter was selected by the operator, and then the probe remained fixed in this plane for the dynamic study of contrast uptake. After injection of the contrast agent, signal enhancement of intratumor neovessels was evaluated visually in real time and the dynamic sequence was recorded on a digital tape.
Fourth, Doppler sonograms were reviewed by two radiologists. On the selected single 2D image, the percentage of the tumor surface taking up contrast agent was visually evaluated, based on the consensus of two reviewers, when contrast uptake was maximal. Digitized quantification of contrast uptake was performed on the same selected single 2D sonogram. After this image was selected, the lesion was outlined to determine a region of interest (ROI). Photoshop software (Adobe Systems) was used to quantify the percentage of pixels exhibiting signal enhancement by the contrast agent (Figs. 1A, 1B, 1C, 1D, and 1E) using the following formula: contrast-enhanced pixels / total number of pixels in the ROI. At each examination, this percentage of contrast uptake was compared with the value observed the day before treatment.
|
|
|
|
|
Tumor Response Criteria
All patients were rated as good responders, poor responders, or
nonresponders at 2 and 6 months by a board composed of oncologists and
radiologists according to the following criteria:
Nonresponders—Patients met RECIST criteria for progressive disease on CT (an increase of at least 20% in the sum of the greatest target tumor diameters).
Poor responders—Patients had stable disease according to RECIST criteria but with weak or no intratumor necrosis on CT.
Good responders—Patients met RECIST criteria for objective response on CT (a decrease of at least 30% in the sum of the greatest target tumor diameters). For targets with no significant modification of tumor volume, complete or almost complete (> 90%) intratumor necrosis was evaluated on CT.
Statistical Analysis
Modifications in the percentage of contrast uptake at each
contrast-enhanced Doppler sonography examination among good responders and
poor or nonresponders were compared statistically at 2 and 6 months using the
Period t test (NCSS software version 2004, Number Cruncher
Statistical Systems). All differences with a p value of less than
0.05 were considered significant.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In the morphologic study, the mean largest tumor diameter was 51.6 mm (range, 10-200 mm). The mean tumor volume was 188 cm3 (range, 0.21-3,230 cm3).
A total of 44 target lesions in 24 patients were completely evaluated at 2 months on the basis of clinical and imaging assessment (Table 1): 19 patients (37 lesions) were considered good responders, and five patients (seven lesions) were considered poor responders or nonresponders.
|
A total of 29 target lesions in 15 patients were completely evaluated at 6 months on the basis of clinical and imaging assessment (Table 2): 13 patients (27 tumors) were rated good responders, and two patients (two tumors) were rated poor responders or nonresponders.
|
Initial tumor contrast uptake evaluated with contrast-enhanced Doppler sonography before treatment was predictive of the future response: 71% (mean percentage of contrast uptake) of tumors in the group of good responders versus 42% in the group of poor responders and nonresponders. Abundantly vascularized tumors (> 70%) had a better response to imatinib than did poorly vascularized tumors (Table 3). A strong correlation (p < 10-4) was found between the decrease in tumor contrast uptake at day 7, day 14, and 2 months after the beginning of treatment and tumor response (Figs. 2A, 2B, 2C, 2D, and 2E). At day 14, the mean percentage of tumor contrast uptake decreased from 71% to 31% in tumors with a good response. This mean percentage remained at 43% in lesions exhibiting a poor response at 2 months.
|
|
|
|
|
|
At 9 and 12 months, the number of patients undergoing treatment in the study (5 and 6, respectively, at each time) was not enough for statistical analysis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Until now, contrast-enhanced CT has been the technique used to systematically follow up and evaluate tumor response, assessing tumor size, lesion density, and contrast enhancement [19-22]. Dynamic MRI can be used to differentiate active tumor from necrotic tissue. However, to our knowledge, no published series has evaluated the accuracy of this imaging technique in the assessment of tumor response in GIST metastases.
Stroobants et al. [23] tested 18F-FDG PET in 21 patients and concluded that the method could be considered sensitive for the evaluation of early response to imatinib treatment. Gayed et al. [24] compared the role of 18F-FDG PET and CT in 54 patients with recurrent or metastatic GISTs. Their results suggest that the performances of 18F-FDG PET and CT are comparable in terms of staging GISTs before treatment initiation. However, 18F-FDG PET was superior to CT in predicting early response (8 weeks) to therapy. PET is clearly the best imaging technique for the evaluation of early activity of these new noncytotoxic targeted therapies according to the findings of Antoch et al. [25]. Indeed, 80-90% of patients in their study achieved a response within 2 months of treatment with such functional assessment. Some patients had no glucose radiotracer uptake 24 hours after the first intake of imatinib, whereas the tumor volume remained unaltered or shrinkage was moderate and progressive over time, as depicted on conventional imaging. In addition, PET can detect which patients are initially refractory to imatinib early and thus avoid expensive and pointless treatment in 10-15% of patients.
High-frequency Doppler sonography can detect neovascularization in animal tumor models [26] and in human tumors [27, 28]. This technique has benefited from major technologic improvements such as digitization and electronic processing of the ultrasound signal, and the multiplicity of transducer channels allows enhanced resolution and greater sensitivity for microvessel detection [29]. Imaging of slow flow has also improved through innovations such as B-mode Flow Doppler software (GE Healthcare) or Dynamic Flow software, among others. Since 1999, contrast-enhanced Doppler sonography has been used to optimize the detection of tumor neovessels.
The use of contrast agents allows vessel signal enhancement and even the detection of neovessels as tiny as 40 µm in diameter [30]. The study of tumor neovascularization is currently an important objective in imaging research because of its role in the evaluation of the efficacy of new treatments and in assessing the metastatic potential of tumors [31, 32]. The injection of contrast material allowed us to optimize the detection of microvascularization and to confirm the absence of residual neovessels. This "parenchymography" (parenchymal vascularization) allows the detection of any viable residual tumor inside the induced necrosis [33]. New agents such as Sonovue are effective at a low mechanical index and prolong the duration of enhancement. Repeated passages are possible with this agent, as well as more effective parenchymal vascularization studies [34]. Several research groups are developing new contrast media that specifically target tumor neovascularization.
The major advantage of contrast-enhanced Doppler sonography for the evaluation of new antitumor treatments is that the cost of the examination is low and it can be repeated without adverse effects. In addition, we previously showed that this technique is not operator-dependent [35]. The availability of new software packages for objective signal quantification as a function of time will reduce limits such as operator dependence and will enable the differentiation of tumor tissue kinetics from those of inflammatory areas.
Our study shows that contrast-enhanced Doppler sonography allows the evaluation of tumor vascularization before treatment and immediately after the beginning of treatment. The abundance of tumor vascularity before treatment was shown to be a predictive factor of tumor response. However, more patients must be studied to confirm the ability of contrast-enhanced Doppler sonography to predict treatment efficacy on the basis of this initial evaluation. Contrast-enhanced Doppler sonography can rapidly evaluate a major decrease in tumor vascularity, thereby indicating the efficacy of imatinib treatment. The impact of this early prediction of tumor response on treatment planning remains to be determined.
Bleeding from the tumor site is not rare during the treatment of a GIST [36]. It occurred in two patients in our series and led to treatment discontinuation. The median time to treatment failure in pivotal studies testing the activity of imatinib is 18 months, with 15% of patients experiencing secondary resistance to imatinib per year of treatment. Clinical patterns of resistance are heterogeneous, but it seems that this secondary resistance to imatinib treatment is observed in some selected nodules that probably require aggressive locoregional therapeutic options.
The effect of imatinib on tumor morphology and function needs to be defined as soon as possible. The early detection of tumor response confirms treatment efficacy, whereas stable disease or even tumor progression may require higher doses.
Our study had some limitations. Because the patients did not undergo surgery, we have no histologic proof of tumor response or of the exact percentage of tumor necrosis. Because the duration of follow-up is relatively short (6 months), the tumor response predicted with contrast-enhanced Doppler sonography is a relatively short-term response and does not prejudge the long-term tumor response. In order to limit operator dependency of simple visual evaluation of tumor vascularity, digitized quantification of contrast uptake was performed measuring the percentage of pixels that exhibited signal enhancement relative to the total number of pixels in the outlined lesion. The accuracy of this method must be evaluated by histologic comparisons that are in progress on experimental tumors in our research laboratory.
In conclusion, contrast-enhanced Doppler sonography is a noninvasive imaging technique that allows early and accurate evaluation of the efficacy of imatinib. Decreased contrast uptake, assessed by contrast-enhanced Doppler sonography 7 and 14 days after the beginning of treatment, was correlated with a good response at 2 months. The objectives of our ongoing study are to correlate the percentage of the early decrease in contrast uptake on Doppler sonography with progression-free survival and to determine whether a further rise in contrast uptake after an initial response is predictive of future resistance to imatinib.
We plan to extend this study to evaluate the detection of active residual GIST cells in selected necrotic masses. Biopsies of tissue exhibiting neovascularization could be performed to characterize secondary resistance to imatinib in order to propose other treatment options (surgical resection, radiofrequency ablation, or different drugs) for patients with refractory disease (Demetri GD et al., presented at the 2004 annual meeting of the American Society Clinical Oncology).
Acknowledgments
We thank Lorna Saint Ange for editing our manuscript.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Murphy-Lavallee, H.-J. Jang, T. K. Kim, P. N. Burns, and S. R. Wilson Are Metastases Really Hypovascular in the Arterial Phase?: The Perspective Based on Contrast-Enhanced Ultrasonography J. Ultrasound Med., November 1, 2007; 26(11): 1545 - 1556. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Giraudet, D. Vanel, S. Leboulleux, A. Auperin, C. Dromain, L. Chami, N. Ny Tovo, J. Lumbroso, N. Lassau, G. Bonniaud, et al. Imaging Medullary Thyroid Carcinoma with Persistent Elevated Calcitonin Levels J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4185 - 4190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Escudier, N. Lassau, E. Angevin, J. C. Soria, L. Chami, M. Lamuraglia, E. Zafarana, V. Landreau, B. Schwartz, E. Brendel, et al. Phase I Trial of Sorafenib in Combination with IFN {alpha}-2a in Patients with Unresectable and/or Metastatic Renal Cell Carcinoma or Malignant Melanoma Clin. Cancer Res., March 15, 2007; 13(6): 1801 - 1809. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
