| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Preclinical |
Authors' Affiliations: 1 Developmental Therapeutics Program, Case Comprehensive Cancer Center and Division of Hematology/Oncology, Department of Pathology and Radiology, Case Western Reserve University and University Hospitals of Cleveland, Cleveland, Ohio and 2 National Cancer Institute, Bethesda, Maryland
Requests for reprints: Afshin Dowlati, Division of Hematology/Oncology, Case Western Reserve University, University Hospitals of Cleveland, 11100 Euclid Avenue, Cleveland, OH 44106. Phone: 216-844-1228; Fax: 216-844-5234; E-mail: afshin.dowlati{at}CASE.edu.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Experimental Design: Ten patients with advanced solid tumors were enrolled at each dose level. One of the following pharmacodynamic effects was considered significant: (a) a 35% decrease in microvessel density in sequential tumor biopsies and (b) a 35% decrease in blood flow within tumor as assessed by dynamic contrast-enhanced magnetic resonance imaging. In addition, soluble E-selectin, soluble intercellular adhesion molecule, soluble vascular cell adhesion molecule, and plasma vascular endothelial growth factor were measured sequentially.
Results: Nineteen patients were enrolled. Sequential tumor biopsies in all evaluable patients showed an increase in microvessel density. Only one patient met the intended pharmacodynamic end point of >35% reduction in blood flow. There was a significant increase in both soluble E-selectin and soluble intercellular adhesion molecule levels pretreatment versus levels at the time of removal of patients from study (P = 0.04 and P = 0.0007, respectively). Levels of serum fibrinogen rose with therapy. There was a trend toward increase in plasma vascular endothelial growth factor levels.
Conclusion: SU5416 does not result in decreased blood flow in tumors or a decrease in microvessel density. This corresponds to the lack of clinical activity seen with this agent. Our clinical trial design termed dose de-escalation is a novel approach to determine the in vivo biological effects of targeted therapies in cancer patients.
SU5416 inhibits tumor growth in vivo in a dose-dependent manner, whereas it has no effect in vitro (3). SU5416 inhibits in vitro VEGF-induced mitogenesis of endothelial cells with an IC50 of 0.04 µmol/L. It has shown significant activity in tumor xenografts in nude mice (3, 4). SU5416 inhibited tumor metastases, microvessel formation, and cellular proliferation in a colon cancer xenograft model (4). In this experiment, an increase in tumor cell apoptotic rate was seen based on the terminal deoxynucleotidyl transferase–mediated nick-end labeling assay (4). Assessment of blood flow using an intravital multifluorescence videomicroscopy in a glioma xenograft model showed a significant reduction in blood flow by day 22 of treatment as well as a reduction in microvessel density (5).
Initial in vitro data suggested a biological half-life of >24 hours. A phase I dose escalation study has been completed with SU5416 on a twice-weekly schedule. Patients were treated with twice-weekly administration of SU5416 for 4 weeks followed by 2 weeks of rest (6, 7). Dose-limiting toxicity was achieved at 190 mg/m2 and consisted of projectile vomiting and headache. The maximum tolerated dose (MTD) was determined to be 145 mg/m2. There was evidence of clinical activity during this phase I trial.
The goal of the current study was to determine the biological effects of SU5416 at the tumor level during a phase I clinical trial. We employed a novel clinical trial design for this endeavor, where dose de-escalation was based on demonstrable biological changes observed at the MTD. The primary end point was to show a (a) 35% decline in tumor blood flow as determined by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) or (b) 35% decrease in microvessel density on sequential tumor biopsies. The biological correlates and timing of anticipated drug effect seen in in vivo animal models were replicated in our phase I trial design. Indeed, Shaheen et al. (4) showed at least a 35% decrease in tumor microvessel density in a colon cancer xenograft model, and Vajkoczy et al. (5) showed at least a 35% decrease in blood flow and microvessel density by day 22.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Ten evaluable patients were to be enrolled at the first dose level (MTD) at 145 mg/m2 given twice weekly. We required that at least five patients show the intended activity (i.e., that there be at least five biological responses) at the first dose level (the MTD) to continue this trial. If the intended effect is shown (i.e., at least five biological responses at the MTD), the trial would proceed to compare the MTD and the de-escalated dose (110 mg/m2) in terms of biological activity. The strategy is summarized in Table 1. The rationale is that if the lower dose seems similar in biological response rate to the MTD, it qualifies as a reasonable candidate for the BMD. Operationally, this means that the number of biological responses at the lower dose must not be less than the number of MTD responses minus 2 (e.g., if there are 9 of 10 responding at the MTD, there must be seven or more responding at the lower dose; at the minimum of five responses at the MTD, there must be three or more responding at the lower dose). Using a Fisher's exact test (one sided), this rule represents a P > 0.20, showing lack of evidence that the two dose levels differed in terms of biological activity.
|
Eligibility criteria
Patients with solid tumors not amenable to curative or otherwise effective surgical, radiation therapy, or chemotherapy treatment programs were eligible. Other eligibility included Eastern Cooperative Oncology Group performance status of 0, 1 or 2; life expectancy of at least 12 weeks; at least 4 weeks must have elapsed since prior large-field radiation therapy, and patients must have been off previous anticancer therapy for at least 3 weeks and recovered from all treatment-related toxicity; adequate hematologic variables (WBC >3,500/L, platelets >100,000/L, hemoglobin >9.0 g/dL); normal coagulation screen; and adequate renal and hepatic variables (creatinine <1.5 mg/dL, bilirubin <1.6 mg/dL, aspartate aminotransferase and alanine aminotransferase more than twice the normal). Patients must have at least two distinct lesions of metastatic or primary tumor >1 cm, which will permit biopsy of a single lesion for the laboratory correlative studies and another lesion for DCE-MRI tumor perfusion studies. It was important that two distinct lesions were identified so that correlative computed tomography/MRI–guided visceral tumor biopsy would not interfere with the correlative DCE-MRI tumor perfusion (blood flow) studies. All patients gave written informed consent.
Drug administration
SU5416 was given over 1 hour via a central venous access device. Treatment was given, without interruption, twice a week on Mondays and Thursdays. Cycles were repeated every 4 weeks. Premedications included diphenhydramine (25-50 mg) and ranitidine (50 mg i.v.). Dexamethasone was given at a dose of 10 mg i.v. for one dose before each dose of SU5416 or 10 mg orally 12 and 6 hours before each dose of SU5416.
Cell adhesion molecules and serum fibrinogen
Plasma samples collected from patients before SU5416 and at scheduled intervals according to treatment schema. Collections correlated to day 0 (pretreatment), day 1 (cycle1/day 1, C1D1), day 15 (C1D15), day 29 (C2D1), day 58 (C3D1), and day 140 (C5D1). Samples were analyzed for levels of soluble intercellular adhesion molecule-1 (ICAM-1), soluble vascular cell adhesion molecule-1 (VCAM-1), and soluble E-selectin using commercially available ELISA kits (R&D Systems, Minneapolis, MN). Samples and normals were assayed in triplicate. The limits of detection for each assay were 0.35 ng/mL for soluble ICAM-1, 2.0 ng/mL for soluble VCAM-1, and 0.1 ng/mL for soluble E-selectin. The mean intra-assay and interassay variability for the kits are 4.8% and 10.1%, respectively. Thirteen normal plasma samples considered as negative controls were collected and tested along with patient samples. Analysis of adhesion molecule data was done using GraphPad Prism version 3.00 statistical software (GraphPad Software, San Diego, CA). Two-tailed, paired t tests were done on intragroup comparison; Ps < 0.05 were considered significant.
Dynamic contrast-enhanced magnetic resonance imaging assessment of blood flow
Data acquisition. Patients were studied using a 1.5-T scanner (Siemens Magnetom Vision). Perfusion studies were obtained with a single slice Flash sequence (repetition time = 10, echo time = 4, flip angle = 30 degree, slice thickness = 10 mm, matrix = 128 x 256, field of view = 16 x 25 to 35 x 35 cm) repeated 128 times, once every 2.6 seconds. The first 10 images were obtained without contrast enhancement, after which an 
10-second bolus injection of gadolinium-diethylenetriaminepentaacetic acid contrast agent (Magnevist, Berlex Laboratories, Montville, NJ) was administered i.v. Gadolinium is a paramagnetic agent that, proportional to its concentration, decreases both the longitudinal magnetization recovery time (T1) and the transverse magnetization decay time (T2*) in tissue. In MRI, a shorter T1 increases signal intensity, whereas a shorter T2* decreases it. However, for the FLASH sequence described above, the signal decrease caused by T2* shortening is negligible (<2% signal change over the range of obtainable gadolinium-diethylenetriaminepentaacetic acid tissue concentrations at the study dosage). It can also be shown that over the same range of tissue concentration, the relative signal increase (postcontrast/baseline) has an essentially linear dependence on tissue concentration. This allows us to interpret proportional change in signal intensity over time as equivalent to proportional change in contrast agent tissue concentration. DCE-MRI was done twice on each patient: once to obtain a pretreatment baseline and once after two cycles of therapy (before cycle 3). The slice position and orientation between days within each patient was carefully matched to obtain data from the same region of interest in the tumor.
Post-processing and analysis. Image processing was done offline using a custom software package that permitted a trained rater to review all 128 images in a study and to interactively identify the tumor region of interest. Mean tumor signal intensity versus time curve for each subject was generated. These curves were then converted to proportional enhancement versus time curves by dividing by the precontrast mean signal intensity. Four measures were obtained from each set of curves. The uptake of contrast into the extracellular extravasation space from blood plasma is typically described by the two-compartment Kety equation (8). Using this model, mathematical simulations were done to assess the sensitivity of each of the variables to blood flow, variations in the arterial plasma concentration input function, and the extracellular extravasation space volume fraction. Three of the variables, gradient peak (Gpeak), enhancement (E), and time until gradient drops to 10% of Gpeak (tgp10), were sensitive to flow; however, only Gpeak and E were approximately linear over the range of expected variation, and only Gpeak had the lowest variation with respect to changes in the extracellular extravasation space volume fraction (extracellular extravasation space volume – total tissue volume). All of the measures were sensitive to differences in the time-to-peak of the plasma input function, but only time-to-gradient peak (tgp) was essentially insensitive to changes in flow and therefore served as an index of the consistency of our bolus injections. Because relative signal change with respect to tissue concentration was different with each patient because of T1 differences between tumor types, pretreatment and posttreatment values of each variable were compared across patients nonparametrically using the Wilcoxon signed rank test, with two-tailed evaluation of significance levels.
Sequential tumor biopsies and microvessel density
Sequential tumor biopsies, using a core needle biopsy, were obtained before starting therapy and before cycle 3. Tumor biopsies were fixed in 10% buffered formalin and paraffin embedded. Three-micrometer tissue sections on a slide are deparaffinized by treatment with xylene followed by a 3-minute wash with absolute, 95%, and 70% ethanol. Sections were then treated by high-heat epitope enhancement for 10 minutes (70% of 850-W heat source) before antibody identification to allow for antibody penetration and enhance antigen identification. Slides were then rinsed twice in a buffered saline (TBS) and then incubated for 10 minutes with 10% normal goat serum in 0.05 mol/L TBS to eliminate nonspecific background staining of the secondary antibody. Primary antibodies (CD34 and factor VIII) at the appropriate dilution were then applied to the sections, and the slides were placed in the humidity chamber overnight at room temperature. Slides were then rinsed, the biotinylated secondary antibody (Kirkegaard and Perry Laboratories, Gaithersburg, MD) was applied for 30 minutes, rinsed again, and incubated with horseradish peroxidase–tagged streptavidin antibody (Kirkegaard & Perry Laboratories) for 30 minutes. The samples were rinsed and incubated with freshly prepared 3,3'-diaminobenzidine (Sigma Chemical, St. Louis, MO) at a concentration of 0.75 mg/mL in Tris buffer containing 5 µL of 30% H2O2. Slides are observed for a brown reaction product and stopped by immersion in water. The slides are counterstained with hematoxylin, cleared, and mounted for microscopy. Immunoreactivity in the tissue biopsies was identified in small and medium-sized vessels under low power to identify areas with microvessel density, angiogenic "hotspots." Microvessel density per area (mean microvascular density/mm2) SD of three randomly selected hotspots were then determined under higher magnification (x400, 0.01885 mm2/field) as previously described (9–11). A single microvessel was considered as an endothelial cell or cell cluster with brown reaction product that is distinct from adjacent microvessels, tumor cells, or other tissue elements. Slides were independently read by three observers, and microvessel density count was averaged.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Microvessel density. From the 19 patients enrolled on this study, only five patients had tumor tissue evaluable in both the pretreatment biopsy and the biopsy before cycle 3. Sequential tumor biopsies in all five evaluable patients showed an increase in microvessel density over the 2-month period (Table 2). Figure 1 is a patient with recurrent breast cancer showing an increase in microvessel density count on pretreatment and posttreatment biopsy.
|
|
|
|
Fibrinogen. Levels of serum fibrinogen rose with therapy (Fig. 3). Mean pretreatment fibrinogen was 487 ± 109 mg/dL. This rose to 669 ± 160 mg/dL by C1D15. This difference was very highly significant P < 0.0001 (paired analysis). However, the levels then remained stable with day C1D28 levels at 657 ± 195 mg/dL. There was no difference between C1D15 and C1D28 levels, thus suggesting that the increase in fibrinogen was related to investigational drug rather than disease progression.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, our success in obtaining evaluable sequential tumor biopsies on patients was significantly less than our previous reports (12) on such procedures (26% versus 88% success rate). This can be attributed to several factors. Eight of 19 patients on this study had progressed before reaching the time point necessary for the second biopsy (before cycle 3). In our published experience of sequential tumor biopsies, the follow-up repeat biopsy occurred within 24 to 48 hours of initiation of the experimental drug (12). In this setting, progression of disease would less likely a factor in not performing the second biopsy. The other six failed to show tumor presence on one of the two biopsies.
Explanations for the lack of observation of a pharmacodynamic effect may be related to different factors, although the use of microvessel density as a marker of angiogenesis has been criticized (13). However, our use of microvessel density was based on preclinical data and many published studies on the methodology of such an approach (9–11, 14). The use of DCE-MRI has been validated within our institution (15) and by others (16). Another explanation for the lack of pharmacodynamic effect may pertain to the agent itself or issues related to dosing. It is also possible that the current schedule of administration was not optimal. Indeed, after completion of this study, reports emerged that this agent failed to improve survival in patients with metastatic colorectal cancer when combined with 5-fluourouracil (17).
Because the primary target of SU5416 is the endothelial cell, we hypothesized that damage and/or apoptosis of vascular endothelial cells will be associated with the release of endothelial cell specific markers in plasma. The rationale for this hypothesis is supported by the detection of elevated levels of certain of these markers in thrombotic thrombocytopenic purpura (18, 19), a disorder characterized by endothelial cell apoptosis, as well as in hypercholesterolemia (20), another disorder characterized by endothelial cell dysfunction. Similarly, elevated levels of the endothelial cell–specific adhesion molecules E-selectin and VCAM-1 reflect early endothelial dysfunction in patients with glucose intolerance (21). Levels of VCAM- 1 and E-selectin are also elevated in the plasma of patients receiving the antiangiogenic, vascular targeting agent CM-101 (22). Our trial showed no early change in any of these markers by C1D15. Comparison of pretreatment values to values obtained at the time patient was taken off study, however, showed a significant increase in E-selectin and soluble ICAM but not in soluble VCAM. It is, however, unclear if this reflects drug effect or just progressive disease. Another study of chemotherapy plus SU5416 showed a similar increase in soluble E-selectin (23).
Although no change in basic fibroblast growth factor was seen, an increase in plasma VEGF level was observed. Although levels of VEGF have correlated with advanced stage disease, poor prognosis, and chemotherapy refractoriness in many cancers (24), no data exists on changes of this marker during VEGF-directed therapies. This has not been evaluated with bevacizumab, a humanized monoclonal antibody targeting VEGF, as the use of antibodies makes plasma measurements useless. Therefore, our trial also constitutes the first trial demonstrating an increase in plasma VEGF level with VEGF receptor–targeted therapy and may constitute a sign of in vivo biological activity. When SU5416 has been combined with IFN-
in renal cell carcinoma, a decrease in plasma VEGF levels has been observed (25).
An increase in thrombotic rates has been reported in trials of SU5416 when combined with chemotherapy (26). In this study, the authors hypothesized that endothelial cells deprived of VEGF after exposure to SU5416 become activated and more susceptible to damage with chemotherapy. Our trial shows a significant increase in serum fibrinogen with SU5416. This may provide another theoretical explanation to the increase thrombosis seen with this agent.
Although our hypothesis-driven study design dose de-escalation strategy to isolate a biological modulatory dose did not come to fruition, we believe this concept is valid. Perhaps the simplest explanation is that the investigational agent in this study is not clinically active. In addition, as experience with antiangiogenic agents in the clinic has evolved, the utility of biomarkers continues to be explored.
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 12/ 9/04; revised 2/24/05; accepted 3/24/05.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
in advanced renal cell carcinoma: a phase II California Cancer Consortium Study with biological and imaging correlates of angiogenesis inhibition. Clin Cancer Res 2003;9:4772–81.This article has been cited by other articles:
|
|
 |
|
  A. Dowlati, S. Manda, J. Gibbons, S. C. Remick, L. Patrick, and P. Fu Multi-Institutional Phase I Trials of Anticancer Agents J. Clin. Oncol., April 20, 2008; 26(12): 1926 - 1931. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Dowlati, R. Gray, A. B. Sandler, J. H. Schiller, and D. H. Johnson Cell Adhesion Molecules, Vascular Endothelial Growth Factor, and Basic Fibroblast Growth Factor in Patients with Non-Small Cell Lung Cancer Treated with Chemotherapy with or without Bevacizumab--an Eastern Cooperative Oncology Group Study Clin. Cancer Res., March 1, 2008; 14(5): 1407 - 1412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Overmoyer, P. Fu, C. Hoppel, T. Radivoyevitch, R. Shenk, M. Persons, P. Silverman, K. Robertson, N. P. Ziats, J. K. Wasman, et al. Inflammatory Breast Cancer as a Model Disease to Study Tumor Angiogenesis: Results of a Phase IB Trial of Combination SU5416 and Doxorubicin Clin. Cancer Res., October 1, 2007; 13(19): 5862 - 5868. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Emmenegger and R. S. Kerbel A Dynamic De-Escalating Dosing Strategy to Determine the Optimal Biological Dose for Antiangiogenic Drugs Clin. Cancer Res., November 1, 2005; 11(21): 7589 - 7592. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
