| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Radiology, The University of Texas M. D. Anderson
Cancer Center, Houston, TX.
2 Present address: Department of Radiology, The University of California at San
Diego, 200 W Arbor Dr., San Diego, CA 92103-8755.
3 Department of Biostatistics and Applied Mathematics, The University of Texas
M. D. Anderson Cancer Center, Houston, TX.
4 Present address: Department of Radiology, Faculty of Medicine, Ramathibodi
Hospital, Bangkok, Thailand.
5 Department of Diagnostic Radiology, The Federal University of Sao Paulo, Sao
Paulo, Brazil.
6 Department of Genitourinary Medical Oncology, The University of Texas M. D.
Anderson Cancer Center, Houston, TX.
Received December 9, 2006;
accepted after revision March 16, 2007.
Address correspondence to S. C. Faria
(fariasil{at}hotmail.com).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Patients with proven metastatic renal cell carcinoma were examined prospectively with functional CT. Functional CT studies (cine mode, 4 x 5 mm) were performed through the tumor after IV administration of a bolus of contrast material before and every 12 weeks after treatment with thalidomide. Quantitative values for blood flow, blood volume, mean transit time, and permeability-surface area product were calculated with commercial software. The average difference in percentage change in functional CT parameters from pretreatment to 12 and 24 weeks after treatment and the median difference in percentage change in functional CT parameters between response groups were assessed. We also tested whether percentage changes in functional CT parameters 12 weeks after treatment correlated with time to progression of disease and size of the perfused lesion.
RESULTS. Sixteen patients with a total of 23 tumors underwent at least one follow-up functional CT examination. Blood flow, blood volume, and permeability-surface area product decreased significantly 12 weeks (-18%, p = 0.0039; -15%, p = 0.0350; -24%, p = 0.0010) and 24 weeks (-28%, p = 0.017; -19%, p = 0.0300; -25%, p = 0.0031) after treatment with thalidomide. Time to progression correlated significantly with percentage change in blood flow (r = -0.34; p = 0.040) and permeability-surface area product (r = -0.36, p = 0.023) at 12 weeks. Responders had a significantly larger decrease in blood flow 12 weeks after treatment than did nonresponders (-29% vs -6%; p = 0.032). We also found a significant correlation between decrease in size of the perfused lesion and percentage decrease in blood flow 12 weeks after treatment (r = 0.50; p = 0.019).
CONCLUSION. Changes in functional CT parameters 12 weeks after treatment may be useful for monitoring the effects of thalidomide and predicting treatment outcome among patients with metastatic renal cell carcinoma. Further study with a larger clinical trial is needed.
Keywords: angiogenesis • CT • functional CT • renal cell carcinoma • thalidomide • tumor blood flow
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Angiogenesis, the formation of capillary blood vessels, is essential to tumor growth and metastasis [3]. Therefore, therapeutic approaches that block angiogenesis may be useful in cancer therapy. During the past 10 years, many agents have been under investigation for antiangiogenic effects. Thalidomide is among those that have been shown to inhibit angiogenesis [4]. Thalidomide has been evaluated in many phase II clinical trials of the management of various solid and hematologic neoplasms, particularly RCC [5-7].
Monitoring of changes in tumor size is the standard method for evaluating the effectiveness of most anticancer therapies. This morphologic information, however, may not directly reflect biologic changes in tumors. Because most antiangiogenic agents are not cytotoxic, treatment with them may not lead to reduced tumor volume. Therefore, monitoring of changes in tumor size alone may not be a good indicator of treatment efficacy, because changes in size can occur later in patients undergoing antiangiogenic therapy than in those receiving conventional treatment [8].
Imaging techniques for both qualitative and quantitative analyses of angiogenesis have been the subject of extensive investigation [9, 10]. Functional CT is an imaging technique whereby CT data are continuously acquired in a cine mode. Changes in contrast enhancement over time are measured after IV bolus injection of a contrast agent. Physiologic parameters, including blood flow (BF), blood volume (BV), mean transit time (MTT), and permeability-surface area product, can be derived from mathematic or tracer kinetic analysis of time-attenuation data with CT perfusion software [11]. Functional CT has been used to measure BF, BV, and MTT in several animal models [12-14], in patients who have had strokes [15, 16], and in cancer patients [17, 18]. These measurements are in vivo biomarkers of tumor angiogenesis and have been proven to show changes in perfusion in patients with rectal cancer managed with bevacizumab (Avastin, Genentech), a vascular endothelial growth factor-specific antibody agent [18]. To our knowledge, no data are available on evaluation of changes in tumor perfusion after antiangiogenic treatment with thalidomide in patients with metastatic RCC. Because of its antiangiogenic effect, we hypothesized that thalidomide may reduce the perfusion parameters of tumors. Therefore, we evaluated the use of functional CT to quantify the effects of thalidomide in a phase II clinical trial of patients with metastatic RCC.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To screen participants for the study, an experienced abdominal radiologist
reviewed previous imaging studies to select the target lesions for functional
CT. The selection criteria for the target lesions included the following:
solid lesions larger than 2 cm in greatest diameter that had not been
subjected to irradiation or biopsy and, preferably, located where the largest
number of lesions could be scanned in the same slice during a single
breath-hold with the fewest involuntary motion artifacts. In addition,
patients had to have a creatinine level 
2.0 mg/dL at least 30 days before
functional CT and no history of allergy to iodine.
Among the total of 42 patients enrolled in the clinical trial, 10 did not participate in the functional CT study for various reasons, including a history of allergy (n = 2), elevated creatinine level (n =3), and a lack of appropriate lesions for scanning (n =5). The other 32 patients underwent pretreatment functional CT, but 15 of them were unable to undergo follow-up functional CT 12 weeks after treatment because of rapid disease progression (n = 9), poor performance status (n = 3), elevated creatinine level (n = 2), and small lesion size (n = 1). The other 17 patients underwent at least one follow-up functional CT study 12 weeks after treatment. However, one of the patients was excluded from analysis because of the presence of breathing artifacts on the functional CT images. Thus, the final study group was composed of 16 patients (14 men, two women; age range, 42-68 years; mean age, 53 years) who had one to three metastatic lesions for a total of 25 tumors. After 12 weeks of treatment, among the 25 tumors, one lesion responded completely, and another lesion, in a different patient, could not be scanned, leaving 23 evaluable lesions in the statistical analysis. Ten of the 16 patients, with 14 tumors, underwent a second follow-up functional CT examination 24 weeks after treatment. The anatomic locations of the 23 metastatic tumors were liver (n = 4), retroperitoneal lymph nodes (n = 4), lung (n = 3), lung hilar lymph nodes (n = 3), adrenal gland (n = 3), renal fossa recurrent disease (n = 2), chest wall (n =1), sacrum (n = 1), abdominal wall (n = 1), and pancreas (n = 1). The mean pretreatment size (measured as the longest diameter of each lesion in centimeters) of the 23 metastatic tumors (perfused lesions) was 4.2 cm (range, 2.0-7.2 cm).
CT Technique
Functional CT scans were acquired 0-6 days before thalidomide treatment and
then every 12 weeks (at the end of three consecutive 28-day cycles) while the
patients remained enrolled in the study. All scanning was performed with an
MDCT scanner (LightSpeed CT, GE Healthcare). After unenhanced CT was performed
to localize the target lesions, a 40-mL bolus of nonionic contrast material
(Optiray [ioversol], 320 mg I/100 mL, Mallinckrodt) was administered with an
automatic injector at a rate of 7 mL/s through an 18-gauge needle placed in an
antecubital vein. Continuous CT data were acquired at a single location 2 cm
wide (4 x 5 mm) at the solid portion of the tumor with a 1-second
scanning speed per rotation in cine mode for a 30- to 40-second scanning time
(30 seconds for the chest or upper abdomen, 40 seconds for the pelvis). The
acquisition delay time varied depending on the location of the lesion (5
seconds for chest or upper abdominal scans, 10 seconds for pelvic scans) to
allow acquisition of unenhanced images. The technical parameters included
80-120 kVp and 190-300 mA, depending on the location of the tumor and the
patient's habitus. When more than one lesion was scanned in the same patient,
the second scan was obtained 10 minutes after the first.
The image data were reconstructed by segmentation every one-half second and used for image processing. On the follow-up scans, we used reference images from the pretreatment study to scan the same location using similar technical parameters.
Image Processing Technique
All functional CT studies were transferred to a single workstation
(Advantage, GE Healthcare), and CT perfusion software (CT Perfusion 3, GE
Healthcare) was used to calculate the functional parameters. Image analysis
was performed by an abdominal radiologist with 4 years of experience in body
functional CT measurement. We selected image data from similar anatomic
locations and used the same number of images to calculate the pretreatment and
follow-up functional parameters for each patient. All functional CT studies
for the same patient were processed at the same time, and we used the previous
study as a reference to ensure that the same lesions were being measured and
that they were being measured in the same way.
To generate parametric maps of BF, BV, MTT, and permeability-surface area product, we always used the aorta or iliac artery (when the aorta was not present in the same slice of the target lesion) as the site of arterial input by placing a region of interest (ROI) over it. A time-attenuation curve was automatically generated for the arterial input along with four perfusion maps for all tissues within the scanning plane.
For calculation of BF, BV, MTT, and permeability-surface area product of the lesions, an ROI was drawn freehand around the peripheral margins of the tumors with an electronic cursor and a mouse on each of four axial images where the tumor was identified. The whole tumor, including solid and necrotic areas when present, was used to generate the functional CT parameters. A time-attenuation curve was derived for the selected tumor and the four functional parameters of the tumor tissue within the ROI. The values of the functional CT parameters for each tumor represented the average values of the available sections. For comparison, we recorded pretreatment and posttreatment values on a spreadsheet (Excel, Microsoft).
Response Evaluation
Thalidomide was administered orally at an initial dose of 200 mg/d that was
escalated in increments of 100-200 mg/d at weekly intervals to a maximum dose
of 1,200 mg/d. Twelve weeks and 24 weeks after the start of the treatment,
most of the patients were receiving the maximum dose of 1,200 mg/d.
Tumor response was assessed every 12 weeks after therapy while the patients remained enrolled in the study. Assessment of response included acquisition of a complete clinical history, performance of a physical examination, evaluation for the presence of toxicity, determination of performance status, blood tests, and measurement of tumors on the basis of findings on restaging imaging studies. Each patient's best overall response was categorized according to World Health Organization criteria [19]. Measurable lesions were quantified with the sum of the products of the longest tumor diameter and its perpendicular dimension. According to the World Health Organization criteria, a complete response was defined as the disappearance of all clinical and radiographic evidence of tumor persisting for at least 4 weeks. A partial response was defined as a 50% or greater decrease in the sum of the products of the diameters of all measurable lesions that persisted for at least 4 weeks without development of new lesions. Progressive disease was defined as a 25% or greater increase in the sum of the products of the diameters of all measurable lesions or the appearance of new lesions. All other responses were defined as stable disease. In our study, we placed patients who had a partial response and those who had stable disease in the same group and considered them responders. Patients who had progression of disease were considered nonresponders.
Time to Progression
Time to progression (TTP) was defined as the number of days from the start
of thalidomide therapy to the date disease progressed or the patient left the
study. Patients who left the study before disease progression occurred were
right-censored in the statistical analysis.
Statistical Analysis
For statistical analysis, the functional CT parameter values from different
tumors in the same patient were treated as independent measurements. For each
lesion, we calculated the percentage change between pretreatment value and
values 12 and 24 weeks after treatment and then averaged and compared the
differences. Percentage change was calculated as 12-week and 24-week values
minus pretreatment value, divided by the pretreatment value for each lesion.
We repeated the calculations for all four functional parameters.
The average difference in percentage change between pretreatment value for BF, BV, MTT, and permeability-surface area product and the values obtained 12 and 24 weeks after treatment along with the corresponding 95% CI were assessed for all lesions by the Hodges-Lehmann method calculated with the StatXact-4 program (Cytel). We used Wilcoxon's signed rank test to compute p values. For all lesions, Spearman's rank-order correlation coefficient was used to calculate the correlation between pretreatment absolute value of BF, BV, MTT, and permeability-surface area product and percentage change in the same functional CT parameters 12 weeks after treatment. We used Somers's Dxy rank correlation test for censored data to calculate the correlation between percentage change in functional CT parameter values 12 weeks after treatment and TTP for all lesions. We used randomization tests with 10,000 iterations to compute the p values for the Somers's Dxy correlation coefficients.
The presence of a correlation between percentage change in functional CT parameter values 12 weeks after treatment and best response on protocol was tested. The median difference in percentage change between response groups and the corresponding 95% CIs were assessed with the Hodges-Lehmann method calculated with the StatXact-4 program. Wilcoxon's rank sum test was used to compare baseline values and fold changes at 12 weeks for responders with those for nonresponders. Assessment of the probability of response as a function of percentage change in BF from pretreatment to 12 weeks after treatment was quantified with the logistic regression model.
Spearman's rank-order correlation coefficient test was used to assess for a correlation between percentage change in the size of all 23 perfused lesions and percentage change in the four functional CT parameters from pretreatment to 12 weeks after treatment. Statistical significance was considered p < 0.05 for all measurements.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
Comparison with Time to Progression
Among the 16 patients, 13 were withdrawn from the study because of
progressive disease, and three were withdrawn before disease progression. Two
of the three underwent resection of stable metastatic RCC lesions, and the
other experienced excessive toxic effects of thalidomide therapy.
We found a significant negative correlation between TTP and percentage change in BF from pretreatment to 12 weeks after treatment so that those who had a decrease in BF had longer TTP (Fig. 2). We also observed a significant negative correlation between TTP and percentage change in permeability-surface area product from pretreatment to 12 weeks after treatment (Fig. 3). There was also a negative correlation between TTP and percentage change in BV from pretreatment to 12 weeks after treatment, but it was not statistically significant (data not shown). TTP had a positive correlation with percentage change in MTT from pretreatment to 12 weeks after treatment so that patients who had an increase in MTT had longer TTP, but this difference also was not statistically significant (data not shown).
|
|
Comparison with Tumor Response
Of the 16 patients studied, nine were considered responders (two with a
partial response and seven with stable disease). The seven patients who had
progression of disease were considered nonresponders. We compared percentage
change in BF from pretreatment to 12 weeks after treatment between the
responders (12 tumors in nine patients) and nonresponders (11 tumors in seven
patients) and observed that the responders had a significantly larger decrease
in BF (Table 2). We estimated
that for each 10% increase in percentage change in BF from pretreatment to 12
weeks after treatment, the odds of response decreased 47% (p =
0.0033) (Fig. 4). We excluded
from this analysis a patient who had an extremely high BF value with a 439.15%
increase in BF value from the pretreatment value. The other 22 lesions had a
range of percentage change in BF between -58% and +28%. There was a trend
toward a greater percentage decrease in BV and permeability-surface area
product among the responders, but the changes were not statistically
significant (Table 2). There
was a greater percentage increase in MTT among the responders than among the
nonresponders, but the difference was not statistically significant
(Table 2).
|
|
|
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In this study, we explored the potential of using functional CT to measure tumor BF, BV, MTT, and permeability-surface area product for monitoring the effects of thalidomide in patients with metastatic RCC. Our overall results showed a significant decrease in BF, BV, and permeability-surface area product from pretreatment to 12 and 24 weeks after treatment with thalidomide. In addition, TTP correlated significantly with decreases in BF and permeability-surface area product values 12 weeks after treatment. We believe that these results may be due to the antiangiogenic effects of thalidomide.
The use of thalidomide, a sedative and antiemetic drug introduced in the late 1950s in Germany, was discontinued because of teratogenic effects. More than three decades later, after its immunomodulatory and antiin-flammatory properties were reported, thalidomide emerged as a new drug for the management of various diseases, including erythema nodosum leprosum, graft-versus-host disease, and AIDS-related Kaposi sarcoma [20, 21]. In 1994, D'Amato et al. [4] observed that thalidomide is a potent inhibitor of angiogenesis induced by basic fibroblast growth factor. This discovery led to the development of several clinical trials in which thalidomide was used as an anticancer agent in the management of various hematologic malignant diseases and solid tumors, including RCC [5-7].
The exact mechanism by which thalidomide inhibits angiogenesis remains
unknown [4]. One possible
pathway is the reduction of tumor necrosis factor 
levels by
enhancement of tumor necrosis factor messenger RNA degradation and
suppression of its production by human monocytes
[22,
23]. It has also been
suggested that the antiangiogenic effect of thalidomide is related to the
action of the drug in reducing levels of cytokines, particularly basic
fibroblast growth factor and interleukin-6
[24]. A more recent suggestion
is that thalidomide exerts an inhibitory effect on nitric oxide-mediated
angiogenesis, which leads to inhibition of endothelial cell migration
[25].
In our study, a 29% decrease in BF was significantly associated with clinical benefit, defined as partial response or stable disease rather than progressive disease, after 12 weeks of treatment. Moreover, the odds of response decreased 47% for each 10% increase in BF value 12 weeks after treatment. We believe that functional CT can be used to assess hemodynamic changes, especially in tumor perfusion, after therapy with thalidomide. However, it is still unknown exactly how thalidomide changes BF, BV, and permeability-surface area product. Kawasuji et al. [26] found that basic fibroblast growth factor increased regional BF in an induced myocardial infarction model in dogs. Therefore, it is reasonable to postulate that thalidomide, by blocking basic fibroblast growth factor, decreases BF.
Although our results suggest that functional CT parameters can show changes in tumor perfusion after treatment with thalidomide in patients with metastatic RCC, our study had limitations. We did not compare our functional CT values with values obtained with other correlative functional imaging techniques, such as the use of radionuclide tracers, 15O-water PET, and color Doppler sonography for BF and MRI for permeability-surface area product [27, 28]. However, there is currently no single reference standard imaging method for all four of the functional parameters analyzed in our study. In addition, the intrinsic mathematic models and the varied temporal resolution present in the different functional imaging methods must to be taken into consideration when they are compared, as observed by Xiong et al. [29] in a study of six patients who underwent functional CT and functional MRI.
In regard to the four functional parameters obtained with functional CT, the total scanning time of 30-40 seconds used in this study might have been insufficient for accurate measurement of permeability-surface area product. Future developments in CT technology may help to rectify this limitation.
Our results lacked histologic correlation. Biopsy, however, might have interfered with the follow-up functional CT parameters, and a small specimen for biopsy might not have been representative of the whole lesion. In an animal study of functional CT, Phongkitkarun et al. [30] found no significant correlation between the histologic vascular parameter microvascular density and functional CT parameters measured with CT perfusion software.
In functional CT studies, tumor coverage is restricted by the detector configuration of CT scanners. In this study, we used a 4-MDCT scanner, which gives a maximum composite coverage of 2.0 cm. Functional CT parameters of lesions measuring more than 2 cm may not reflect changes in the whole lesion because of tumor heterogeneity [9]. In the near future, however, improvements in CT scanners and the development of volumetric CT may allow study of whole tumors. In addition, functional CT is a technique that can be associated with a variety of technical failures such as contrast extravasation, motion and breathing artifacts, movement of lesions out of the field of view, failure to image a lesion in the same portion of the tumor as on previous functional CT scans, and lesion size too small for imaging.
We observed different responses in the same patient. Among the six patients who had more than one lesion scanned, one had three lesions that behaved differently from one another. One metastatic left adrenal lesion had a 53% decrease in BF value at 12 weeks, and of the two liver metastatic lesions evaluated, one had an 18% decrease in BF and the other a 4% increase in BF. These apparently discordant findings may be related in part to limitations in the measurement technique and associated reproducibility or related in part to tumor heterogeneity, which is a well-known phenomenon.
The process by which we obtained the data required the user to select multiple input variables, which can lead to questions about how the quantitative values should be obtained and whether minor variations of these user-defined inputs would affect the final quantitative values of the functional parameters. Some of these variables included selection of the input artery, size of the arterial ROI, and cutoff values for unenhanced and enhanced images. Hoeffner et al. [16] reported different initial quantitative results depending on the input artery chosen. Conversely, Scanelli et al. [31] observed no significant effect on mean BF, BV, and MTT, even with major variations in arterial ROI placement and arterial ROI size. In the same study, however, Scanelli et al. found that the choice of venous ROI placement or unenhanced and contrast-enhanced cutoff values can significantly influence quantitative deconvolution-based functional CT map values in patients with acute stroke. To minimize the effects of those variables in our calculations in the evaluation of body tumors, we needed to select only the artery input vessel, and we took special care always to use the same vessel (the aorta or iliac artery) and the same ROI size in all analyses of the same patient.
Although the quantitative functional CT parameters obtained with CT perfusion software have proven to be reproducible in body tumors [32] and to have good intraobserver and interobserver agreement [33], the presence of measurement variability in quantification of pretreatment and posttreatment functional CT parameters may interfere with the distinction between measurement error and a true therapeutic effect. This possibility should be taken into consideration in the use of functional CT measurements to assess response to antiangiogenic therapy.
The optimal time to perform follow-up functional CT after the start of antiangiogenic treatment has not been established. In our study, of the 32 patients who underwent pretreatment functional CT, nine were withdrawn from the study because of rapid disease progression and did not undergo the first follow-up functional CT examination 12 weeks after treatment. This factor might have introduced selection bias in our results, because most of the patients who underwent the first follow-up functional CT study 12 weeks after treatment were those who did not yet have progression of disease. It is possible that a shorter follow-up period may produce better correlation between clinical outcome and functional CT parameters. Morgan et al. [34] and Yao et al. [35] suggested that changes in the functional perfusion parameters of tumors can be detected as early as 48 hours after treatment with agents that block vascular endothelial growth factor signaling pathways. This observation may be important in designing clinical trials of antiangiogenic therapies.
In conclusion, our results show that changes in functional CT parameters 12 weeks after treatment with thalidomide may be used to monitor the effect of the drug in patients with metastatic RCC. However, the usefulness of functional CT findings as a predictor of treatment outcome after thalidomide therapy must be confirmed in a larger clinical trial.
Acknowledgments
We thank Delise H. Herron for performing the functional CT studies.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
by
enhancing mRNA degradation. J Exp Med1993; 177:1675
-1680 production by stimulated
human monocytes. J Exp Med 1991;173
: 699-703
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Schor-Bardach, D. C. Alsop, I. Pedrosa, S. A. Solazzo, X. Wang, R. P. Marquis, M. B. Atkins, M. Regan, S. Signoretti, R. E. Lenkinski, et al. Does Arterial Spin-labeling MR Imaging-measured Tumor Perfusion Correlate with Renal Cell Cancer Response to Antiangiogenic Therapy in a Mouse Model? Radiology, June 1, 2009; 251(3): 731 - 742. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
