1 Tumor Biology Center, Freiburg; 2 ProQuinase GmbH, Freiburg, Germany; 3 Translational and Clinical Development, Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA; 4 Preclinical Research, Novartis Pharma, Basel, Switzerland; 5 Leceister Royal Infirmary Hospital, Leicester, UK; 6 Schering AG, Berlin, Germany
* Correspondence to: Dr J. Drevs, Tumor Biology Center, Department of Medical Oncology, Breisacherstrasse 117, 79106 Freiburg, Germany. Tel: +49-761-206-2178; Fax: +49-761-206-2180; Email: drevs{at}tumorbio.uni-freiburg.de
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods: Patients with colorectal cancer (CRC) (n=63) were enrolled into two phase I/II dose escalation trials of PTK/ZK in 28-day cycles until discontinuation. Patients with stable disease for 2 months were categorized as non-progressors. Plasma markers of angiogenesis, VEGF-A and basic fibroblast growth factor (bFGF), and the serum markers of activated endothelial cells, sTIE-2 and sE-Selectin, were assessed at baseline, and pre-dose on days 1, 8, 15, 22 and 28 of every cycle, with additional assessments 10 h post-dose on days 1 and 15. The percentage change from baseline was subsequently correlated with AUC and Cmax of PTK/ZK on day 1, cycle 1 and clinical outcome.
Results: A dose-dependent increase in plasma VEGF-A and bFGF was observed in the first cycle of PTK/ZK treatment. The correlation of change in plasma VEGF-A with AUC and Cmax was characterized by an Emax model, suggesting that a change of 150% from baseline VEGF-A correlated with non-progressive disease. Change from baseline plasma VEGF-A within the first cycle of treatment was significantly correlated with clinical outcome by logistic regression analysis (P=0.027).
Conclusions: In patients with CRC treated with PTK/ZK, changes in plasma VEGF-A and bFGF demonstrate biological activity of PTK/ZK, may help to establish optimal dose and correlate with outcome.
Key words: angiogenesis, biomarker, VEGF, VEGF receptor inhibitor
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Tumor regression is usually measured by standard non-invasive imaging techniques such as computed tomography (CT) and magnetic resonance imaging (MRI). Antiangiogenic agents may not induce immediate tumor regression; therefore, traditional imaging techniques provide a limited assessment of biological activity. Biomarkers, however, have been used successfully to assess biological activity in response to antiangiogenic agents. Dynamic contrast-enhanced MRI (DCE-MRI), while its availability is limited, appears to be a promising biomarker for assessing early changes in tumor-associated vasculature in response to treatment with antiangiogenic agents [4]. Soluble plasma and serum markers of angiogenesis and of activated endothelial cells can also be used to assess antiangiogenic activity. We have investigated several soluble biomarkers, including a member of the vascular endothelial growth factor family (VEGF-A), basic fibroblast growth factor (bFGF) and several endothelial cell-associated molecules. Members of the VEGF family (VEGF-A, -B, -C and -D) play a critical role in angiogenesis [5
]. In addition to increasing vascular permeability, the family of VEGF ligands stimulate endothelial cell proliferation, migration and tube formation [6
]. The effects of each VEGF ligand are mediated through receptors, VEGFR-1, -2 and -3, located on endothelial cells. bFGF is also an endothelial cell-specific angiogenic factor that plays a key role in tumor angiogenesis [7
]. The family of VEGF proteins and bFGF are produced by tumor cells in response to tumor hypoxia. Therefore, the concentration of these angiogenic factors in the plasma may reflect the hypoxic status of a tumor and may increase in response to effective blockade of VEGF receptors on endothelial cells.
Endothelial cell-associated molecules that are shed into the plasma can provide an indirect measure of the amount of newly forming vessels in a patient's tumor. E-Selectin is an endothelial cell adhesion molecule, and the soluble form of E-Selectin (sE-SEL) may increase tumor angiogenesis and the adhesion of tumor cells to endothelial cells at distant sites [8]. The transmembrane tyrosine kinase TIE-2, the receptor for the angiopoietin-1 and -2, is also involved in angiogenic processes and is ubiquitously expressed on endothelial cells throughout the vasculature. The soluble form of TIE-2 (sTIE-2) can be detected in serum from healthy individuals [9
, 10
].
We have used these biomarkers to assess the biological activity of PTK787/ZK 222584 (PTK/ZK), a novel oral angiogenesis inhibitor (co-developed by Novartis and Schering AG, Berlin) that selectively inhibits the phosphorylation of all known VEGF receptor tyrosine kinases, as well as the platelet-derived growth factor receptor and c-kit, to a lesser degree. Here we report the effects of PTK/ZK on plasma/serum concentrations of soluble biomarkers, including VEGF-A and bFGF, sE-SEL and sTIE-2, in patients with colorectal cancer (CRC).
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients with histologically confirmed advanced solid malignancies with no standard curative therapy were eligible. All patients were required to have at least one site of measurable or evaluable disease as determined by the Southwest Oncology Group (SWOG) criteria. Inclusion was irrespective of stage of disease or extent of prior therapy. Patient entry criteria included: age 18 years; WHO performance status of 02; adequate hematological (absolute neutrophil count
1.5x109/l, hemoglobin
9 g/dl, platelets
100x109/l), renal [serum creatinine
1.5xupper limit of normal (ULN), serum bilirubin
1.5xULN, 24-h creatinine clearance
50 ml/min] and hepatic (aspartate aminotransferase and alanine aminotransferase
2.5xULN) function; no known brain metastases; no recent prior chemotherapy or biological therapies, radiotherapy or surgery; and a life expectancy of at least 12 weeks.
All patients were informed about the investigational nature of the study according to institutional and regional guidelines, and provided written consent before beginning therapy. Permission of local ethics regulatory bodies was obtained at each center.
Response criteria
Patients were evaluated for tumor response at the end of every 28-day cycle using the SWOG Solid Tumor Response Criteria. All measurable evaluable and non-evaluable lesions were accounted for in the tumor assessment. Measurable lesions were quantified using the product of perpendicular diameters. Methods for assessment varied between patients (but were consistent for each patient throughout the course of study) and included physical examination, laboratory values, ultrasound and MRI. Minor response (MR) was defined as a 25% but
50% decrease from baseline in measurable lesions. Progressive disease (PD) was defined as a
50% increase or an increase of 10 cm2 (whichever was smaller) in measurable lesions, clear worsening from previous assessment of any evaluable disease, reappearance of any lesion which had disappeared or appearance of any new lesion/site. Stable disease (SD) was defined as not meeting the criteria for either MR or PD.
Best response criteria were used to categorize all evaluable patients as either non-progressors or progressors for the surrogate marker analysis to identify the differences in biological effects in response to PTK/ZK efficacy. The best response was determined retrospectively from two consecutive evaluations 28 days apart. Patients with either MR or SD on two consecutive evaluations were prospectively defined as non-progressors; no patient achieved a complete or partial response.
Study design, drug administration and biomarkers sampling schema
PTK/ZK was administered orally once daily in 28-day cycles until discontinuation secondary to adverse events or tumor progression. Dose levels are shown in Table 1. Three patients were enrolled per dose cohort, and an additional three patients were enrolled at the same dose level in the event that a dose-limiting toxicity was observed. An additional six to 25 patients were enrolled at the dose level defined as optimal (with respect to toxicity, pharmacokinetic and biomarkers) for further assessment of biological activity and safety (dose expansion cohort).
|
|
PTK/ZK pharmacokinetic assessment
Full pharmacokinetic samples were obtained on days 1, 15 and 28 in the cycle 1 at the following time points: pre-dose (0), and 0.25, 0.5, 1, 1.5, 2, 4, 6, 10 and 24 h post-dose. The pharmacokinetic parameters, area under the plasma concentration curve (AUC), maximum plasma concentration (Cmax), minimum plasma concentration (Cmin) and elimination half-life (t) were determined for each individual plasma concentrationtime data on day 1 and end of cycle 1 (EC1). The Cmax and Cmin were determined on day 1 and at EC1 by visual inspection of each patient's plasma concentration. Using non-compartmental methods (WinNonlin Pro 3.2; Pharsight Corporation, Mountain View, CA, USA), AUC0
was calculated on day 1 using the linear trapezoidal rule up to the last measurable data point, and then extrapolated to infinity. AUC024 was calculated on EC1 using the linear trapezoidal rule up to 24 h post-dose. The t
on day 1 and EC1 was calculated by dividing 0.693 by ke (elimination rate constant), which was calculated by the linear regression of the WinNonlin selected data points on the terminal log-linear portion of the plasma concentrationtime curve.
Biomarker bioanalytical methodology
Plasma samples were assayed for VEGF-A, bFGF and sE-SEL using the relevant quantitative sandwich enzyme-linked immunosorbent assay (ELISA) (R&D Systems Europe, Oxford, UK) according to the manufacturer's instructions. All samples and standards were assayed in duplicate. sTIE-2 serum levels were determined by ELISA as described by Reusch et al. [8].
Biomarker assessment
The units for the biomarker concentration were as follows: VEGF-A [limit of quantitation (LOQ) 31.25] and bFGF (LOQ 1.00) were expressed in pg/ml; sTIE-2 and sE-SEL (LOQ 0.20) were expressed in ng/ml. All values below the limit of quantitation were set to the LOQ for the biomarker analysis.
Changes in plasma and serum markers in response to PTK/ZK treatment were evaluated in the first cycle of treatment as indicators of biological activity. Because of high baseline variability, data were expressed as percentage of baseline and subsequently averaged across the first cycle of treatment. Patients who met the following inclusion criteria for biomarker analysis were included in the analysis: evaluable disease status of non-progressive or progressive disease; enrolled on either the dose escalation or expansion groups of the trial; treated with PTK/ZK on the day of biomarker sampling; and must have at least five of the seven data points in the first 28 days of PTK/ZK treatment.
Pharmacokinetic biomarker data: tumor response analysis and statistical analysis
To characterize the relationship between mean change from baseline plasma/serum marker versus PTK/ZK exposure (AUC) and maximum PTK/ZK concentration (Cmax) on day 1, pharmacodynamic modeling was performed by fitting the data to an Emax model. The pharmacokinetic data from day 1, instead of EC1, was used in the modeling to allow for a wider range of exposure and concentration to be characterized in the exposure/concentration versus effect curve.
![]() |
A uniform weighting scheme was used. The final model and parameters were selected based on visual inspection, statistical estimation of the goodness of fit and an understanding of the biology with antiangiogenic agents. Furthermore, to assess the correlation between clinical outcome and change in plasma or serum markers, the data were fit to a logistic regression model.
![]() |
The relationship between change from baseline for biomarkers and PTK/ZK dose and clinical outcome were analyzed using regression analysis (S-PLUS, version 2.0; Mathsoft Engineering & Education, Cambridge, MA, USA). The degree of correlation between pharmacodynamic parameters and change from baseline biomarkers was assessed using Spearman Rank correlations. Significance was assigned at P <0.05; P <0.10 was described as a trend.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
Changes in plasma VEGF-A and bFGF levels were observed with PTK/ZK treatment in the first cycle of treatment, and were prominent for patients who received 1000 mg dose (Figure 3). Plasma levels of VEGF-A and bFGF showed an exposure- and concentration-dependent increase in the first cycle of treatment, followed by a decline during the second cycle of treatment. The mean change from baseline VEGF-A was significantly correlated with PTK/ZK AUC at EC1 (P=0.049), with a positive trend on day 1 (P=0.063). The correlation between mean percentage of baseline VEGF-A and maximum PTK/ZK concentration also showed a positive trend on day 1 (P=0.086) and EC1 (P=0.094). The correlations between PTK/ZK exposure and concentration versus mean percentage of baseline bFGF did not reach statistical significance at any time point (data not shown). No changes in sTIE-2 and sE-SEL were observed at any dose level with PTK/ZK treatment, and therefore no subsequent analyses were performed.
|
|
Correspondingly, at 1000 mg/day PTK/ZK, the lower limit of the standard deviation at day 1 lies close to an AUC of 100 hxµM and Cmax of 15 µM. At this AUC and Cmax, the mean change from baseline bFGF is 400%.
A logistic model revealed a statistically significant relationship between change in plasma VEGF-A in the first cycle of PTK/ZK treatment and disease status (P=0.027). This suggests that the change in plasma VEGF-A in the first cycle of PTK/ZK treatment is a good indicator of clinical outcome; i.e. a mean percentage of baseline VEGF-A of 150% correlates with a 50% probability of achieving non-progressive disease. In Figure 5, a plot of the best change in tumor size in the first two cycles of PTK/ZK treatment versus mean percentage of baseline VEGF-A in the first cycle of treatment supports this and the previous findings that most non-progressors achieved a 150% change from baseline plasma VEGF-A in the first cycle of treatment. Non-progressors achieved up to 40% tumor regression in the first or second cycle of PTK/ZK treatment. The change in plasma bFGF did not show statistical significance with disease status (P=0.633) and is therefore not a good indicator of non-progressive disease.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Several studies have implicated the VEGF pathway in human colon cancer angiogenesis [11, 12
]. Further studies have assessed the role of VEGF signaling in the angiogenesis, metastasis and proliferation of human colon cancer [13
, 14
]. These studies support the hypothesis that VEGF signaling is an important angiogenic factor in colon cancer, and indicate that vessel count and the expression of VEGF may be useful in predicting metastasis from colon cancer [13
]. Other studies have examined the regulation of VEGF secretion by colon cancer cells in vitro and showed that alteration of VEGF expression by colon cancer cells may affect the proliferative activity of the target endothelial cells [14
].
Intuitively, one would expect that if PTK/ZK treatment is effective, a decrease in VEGF-A expression would be observed. These results, however, at early time points show a positive dose relationship with percentage change from baseline VEGF-A and bFGF, with a subsequent decline in VEGF levels as treatment continues. Similar findings have been observed in a mouse tumor model. In tumor-bearing mice, plasma VEGF-A levels were correlated with tumor burden. However, in comparison with vehicle-treated mice, plasma concentrations of VEGF-A in mice treated with 50 mg/kg PTK/ZK were found to increase after 1 day of treatment, with a subsequent decrease of plasma VEGF after 8 days of treatment. This decline in VEGF-A levels correlated with a significant decrease in tumor burden in the PTK/ZK-treated mice. The acute rise in plasma VEGF-A and bFGF concentrations observed in the animal studies and in the clinical trials described herein would be consistent with an initial increased expression of VEGF-A and bFGF by tumor cells in response to hypoxia induced by the reduction in tumor vascular permeability and vascularization induced by PTK/ZK treatment, with concomitant decreases in VEGF-A expression as the size of the tumor stabilizes or regresses. Previously reported results of DCE-MRI analysis in patients with CRC [15], as well as in animal studies [16
, 17
], have clearly shown an early reduction in tumor vascular permeability and vascularization with PTK/ZK treatment that significantly correlates with subsequent clinical response.
In these studies, non-progressive disease was associated with PTK/ZK doses 1000 mg/day. At 1000 mg/day PTK/ZK, a mean 150% increase from baseline VEGF-A levels was observed, and this may be considered a threshold that is associated with non-progressive disease. The percentage change from baseline VEGF-A in cycle 1 of treatment was significantly correlated with clinical outcome at the end of cycle 2, demonstrating the early clinical activity of PTK/ZK and the utility of plasma VEGF-A levels as a biomarker in helping to establish the optimal biological dose for PTK/ZK in this patient population. It can be hypothesized that the decline in plasma VEGF-A seen in the higher doses (
1000 mg) after the first cycle may be due to the regressing tumor size with PTK/ZK treatment, but other factors may be involved.
This may indicate that an early rise in plasma VEGF-A is desirable and is suggestive of biological activity with PTK/ZK treatment; however, a decline in plasma VEGF-A is generally the real indicator of tumor regression. Changes from baseline VEGF-A appear to be a better indicator of clinical outcome than bFGF, suggesting that blockage of VEGF receptor activity by PTK/ZK has more impact in the mechanism by which angiogenesis is promoted. Changes in the endothelial cell-associated molecules sTIE-2 and sE-SEL were not observed over 2 months of treatment with PTK/ZK, suggesting that these molecules have limited application as biomarkers for assessment of therapy.
These studies have shown that the plasma concentrations of VEGF-A and bFGF may be useful as biomarkers to detect biological activity in patients with CRC under treatment with PTK/ZK, and may therefore be of value in defining the optimal biologically active dose for phase II/III studies.
These results correlate strongly with those reported by Morgan et al. [15] in patients with CRC treated with oral, once-daily PTK/ZK at doses ranging from 50 to 2000 mg/day. In that study, early changes in tumor vascularity and vascular permeability (Ki) were assessed by DCE-MRI as biomarkers of clinical activity and correlated with PTK/ZK pharmacokinetics and subsequent clinical outcome after 56 days of treatment. A 60% decrease in Ki was significantly correlated with non-progressive disease after two cycles of PTK/ZK treatment, which required a dose
1000 mg/day PTK/ZK to achieve [15
]. Taken together, these biomarker studies support the conclusion that the optimal biological dose of PTK/ZK in patients with CRC will be
1000 mg/day. However, the biomarker data alone are not sufficient to determine the optimal biological dose for further clinical testing. That determination must incorporate changes in tumor size, safety and pharmacokinetic data from studies that are currently underway.
![]() |
Acknowledgements |
---|
Received for publication August 27, 2004. Revision received November 16, 2004. Accepted for publication November 17, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Drevs J, Muller-Driver R, Wittig C et al. PTK787/ZK 222584, a specific vascular endothelial growth factor-receptor tyrosine kinase inhibitor, affects the anatomy of the tumor vascular bed and the functional vascular properties as detected by dynamic enhanced magnetic resonance imaging. Cancer Res 2002; 62: 40154022.
3. Marx GM, Steer CB, Harper P et al. Unexpected serious toxicity with chemotherapy and antiangiogenic combinations: time to take stock! J Clin Oncol 2002; 20: 14461448.
4. Libutti SK, Choyke P, Carrasquillo JA et al. Monitoring responses to antiangiogenic agents using noninvasive imaging tests. Cancer J Sci Am 1999; 5: 252256.[ISI][Medline]
5. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature 2000; 407: 249257.[CrossRef][ISI][Medline]
6. Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 1995; 146: 10291039.[Abstract]
7. Wittig BM, Kaulen H, Thees R et al. Elevated serum E-selectin in patients with liver metastases of colorectal cancer. Eur J Cancer 1996; 32A: 12151218.[CrossRef][ISI][Medline]
8. Reusch P, Barleon B, Weindel K et al. Identification of a soluble form of the angiopoietin receptor TIE-2 released from endothelial cells and present in human blood. Angiogenesis 2001; 4: 123131.[CrossRef][Medline]
9. Werther K, Christensen IJ, Brunner N, Nielsen HJ. Soluble vascular endothelial growth factor levels in patients with primary colorectal carcinoma. The Danish RANX05 Colorectal Cancer Study Group. Eur J Surg Oncol 2000; 26: 657662.[CrossRef][ISI][Medline]
10. Kumar H, Heer K, Greenman J et al. Soluble FLT-1 is detectable in the sera of colorectal and breast cancer patients. Anticancer Res 2002; 22: 18771880.[ISI][Medline]
11. Brown LF, Berse B, Jackman RW et al. Expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in adenocarcinomas of the gastrointestinal tract. Cancer Res 1993; 53: 47274735.[Abstract]
12. Takahashi Y, Tucker SL, Kitadai Y et al. Vessel counts and expression of vascular endothelial growth factor as prognostic factors in node-negative colon cancer. Arch Surg 1997; 132: 541546.[Abstract]
13. Ellis LM, Liu W, Wilson M. Down-regulation of vascular endothelial growth factor in human colon carcinoma cell lines by antisense transfection decreases endothelial cell proliferation. Surgery 1996; 120: 871878.[ISI][Medline]
14. Shaheen RM, Davis DW, Liu W et al. Antiangiogenic therapy targeting the tyrosine kinase receptor for vascular endothelial growth factor receptor inhibits the growth of colon cancer liver metastasis and induces tumor and endothelial cell apoptosis. Cancer Res 1999; 59: 54125416.
15. Morgan B, Thomas AL, Drevs J et al. Dynamic contrast-enhanced magnetic resonance imaging as a biomarker for the pharmacological response of PTK787/ZK 222584, an inhibitor of the vascular endothelial growth factor receptor tyrosine kinases, in patients with advanced colorectal cancer and liver metastases: results from two phase I studies. J Clin Oncol 2003; 21: 39553964.
16. Anderson H, Price P, Blomley M et al. Measuring changes in human tumour vasculature in response to therapy using functional imaging techniques. Br J Cancer 2001; 85: 10851093.[CrossRef][ISI][Medline]
17. Drevs J, Muller-Driver R, Wittig C et al. PTK787/ZK 222584, a specific vascular endothelial growth factor-receptor tyrosine kinase inhibitor, affects the anatomy of the tumor vascular bed and the functional vascular properties as detected by dynamic enhanced magnetic resonance imaging. Cancer Res 2002; 62: 40154022.