Blockade of Platelet-Derived Growth Factor Receptor-Beta by CDP860, a Humanized, PEGylated di-Fab', Leads to Fluid Accumulation and Is Assoc
http://www.100md.com
《临床肿瘤学》
the Cancer Research UK Departments of Medical Oncology and Radiology, Christie Hospital
Imaging Science and Biomedical Engineering, University of Manchester, Manchester
Oncology Group, Celltech R & D Ltd, Slough, UK
ABSTRACT
PATIENTS AND METHODS: Patients with advanced ovarian or colorectal cancer and good performance status received intravenous infusions of CDP860 on days 0 and 28. Patients had serial dynamic contrast-enhanced magnetic resonance imaging studies to measure changes in tumor vascular parameters.
RESULTS: Three of eight patients developed significant ascites, and seven of eight showed evidence of fluid retention. In some patients, the ratio of vascular volume to total tumor volume increased significantly (P < .001) within 24 hours following CDP860 administration, an effect suggestive of recruitment of previously non-functioning vessels.
CONCLUSION: These observations suggest that inhibition of PDGFR- might improve delivery of a concurrently administered therapy. However, in cancer patients, further exploration of the dosing regimen of CDP860 is required to dissociate adverse effects from beneficial effects. The findings challenge the view that inhibition of PDGF alone is beneficial, and confirm that effects of PDGFR kinase inhibition mediate, to some extent, the fluid retention observed in patients treated with mixed tyrosine kinase inhibitors.
INTRODUCTION
PDGFR- is expressed on many tumor types,5 including ovarian 5,6 and colon cancers.5,7,8 In addition, the receptors are expressed on pericytes,9 suggesting that PDGF signaling may be involved in the regulation of angiogenesis.10 Inhibition of pericyte and endothelial function is associated with enhanced tumor growth inhibition9 in animals.
PDGF probably plays a role in the control of tissue interstitial fluid pressure (IFP). Evidence suggests that PDGF is an important signaling molecule for the maintenance of the interaction of myofibroblasts with collagen in the extracellular matrix, via 1 integrins.11 It was previously shown that PDGF effects on IFP were specific to PDGF BB. In a rat model, administration of PDGF BB, but not PDGF AA, normalized dermal IFP that had been experimentally lowered.12 Studies using rodent tumor models showed that inhibition of PDGF using the tyrosine kinase inhibitor imatinib, or an anti-PDGF B DNA aptamer, was associated with a decrease in tumor IFP.13 The authors hypothesized that the reduction in IFP by PDGF inhibition would increase the blood flow–driven uptake of small molecules, and confirmed the hypothesis using 51Cr-EDTA,13 and subsequently, [3H]-paclitaxel14 and epothilone B.15 In the Pietras et al study,15 imatinib increased the tumor uptake of epothilone B without increasing uptake in liver, kidney, or intestinal tract, and without reduction in the tolerability of the cytotoxic agent.
In patients with colorectal and ovarian carcinoma, we report the phase II evaluation of a pegylated di-Fab' molecule,16 CDP860, which binds to and inhibits PDGFR-. CDP860 cross-reacts with PDGFR- from cynomolgus monkey, and was well tolerated in toxicology evaluation in this species at doses up to 400 mg/kg, given weekly for 4 weeks.
One hypothetical benefit of inhibiting PDGF is the potential to prevent coronary artery restenosis after stent insertion. Primate models of neointimal formation suggested a role for PDGFR- in intimal hyperplasia.17,18 A study was performed in which 145 patients undergoing intracoronary stent insertion received placebo or 25 mg/kg CDP860 (n = 76), the drug having been well tolerated at doses up to 30 mg/kg in a healthy volunteer study. Although there was no improvement in restenosis rate, 47 of the patients receiving CDP860 developed peripheral or facial edema between 2 hours and 3 weeks after dosing. These observations were attributed to the mode of action of CDP860.19
In total, 97 volunteers and patients had received CDP860 before this study. Following the reported effects of PDGF BB inhibition on IFP in animal models, we wanted to determine if changes in tumor vascular parameters could be detected in humans, and to assess whether CDP860 would be likely to increase the uptake of a concurrently administered small molecule in future studies.
PATIENTS AND METHODS
Phase II Evaluation of CDP860 in Advanced Ovarian and Colorectal Cancer
Aims of the study The study examined changes in tumor vascular parameters assessed using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in patients with advanced colorectal or ovarian cancer who had received CDP860. The primary end point was the change in tumor vascular permeability measured for up to 45 days after administration of the CDP860. Secondary end points were effects of CDP860 on other vascular parameters including tumor blood flow, safety and tolerability of CDP860, plasma pharmacokinetics, and tumor response.
Patients Patients with measurable primary or secondary colon or ovarian cancer over 18 years of age, an Eastern Cooperative Oncology Group score between 0 and 2, and life expectancy of at least 3 months were eligible. They were required to have adequate liver (normal bilirubin and transaminases < 2.5x the upper limit of normal), renal (creatinine < 1.5x the upper limit of normal), and hematologic (hemoglobin >10 g/dL, white cell count > 3 x 109/L, and platelet count >100 x 109/L) function, as well as normal coagulation (prothrombin time and activated partial thromboplastin time) and a normal ECG.
Patients were excluded if they had an additional chronic disease affecting a major organ, infection requiring antibiotics, clinically significant ascites or pleural effusion, major surgery within the previous 4 weeks, previous history of reaction to biologic agents or drugs containing PEG, a contra-indication to magnetic resonance imaging (MRI), history of infection with hepatitis B, C, HIV-1 or HTLV-1, alcohol or drug addiction, or treatment within the previous 4 weeks. Patients were not permitted to take drugs known to alter vascular flow and they had to have recovered from the effects of previous treatments. Patients with known or suspected brain or CNS disease were excluded.
Study design Patients were treated at the Christie Cancer Center (Manchester, UK) and received CDP860 25 mg/kg on days 0 and 28. A computed tomography (CT) scan was performed during the 2 weeks before treatment, and a post-treatment scan was performed on day 45. CT was also performed as part of the routine clinical monitoring of the patients, on average, 68 days before the first study-specific CT scan. MRI scans were performed twice at baseline to establish reproducibility,22,23 and on days 1, 7, 27, and 45. Blood samples were taken to measure hematologic and biochemical toxicity, CDP860 concentration, and anti-idiotype responses at screening before treatment, and on days 1, 7, 28, 35, and 50. Adverse events were assessed before treatment and on days 0, 1, 7, 27, 28, 35, 45, and 50. Toxicity was assessed using the National Cancer Institute Common Toxicity Criteria (version 2.0). Response was assessed using Response Evaluation Criteria in Solid Tumors.24 Any patient with stable disease or better at the end of treatment was eligible to continue treatment for a further 6 months.
All patients gave written informed consent and the study was approved by the South Manchester Local Research Ethics Committee and the Christie Hospital Research Committee.
MRI The location of the tumor to be studied using DCE-MRI was ascertained from the CT scans. All MRIs were performed on a Philips Gyroscan NT Intera system (Philips Medical Systems, Best, Netherlands) at 1.5 T at the University of Manchester. The MRI protocol consisted of localizer acquisitions, followed by axial T1-weighted fast field echo (gradient echo), and T2-weighted fast spin echo volumetric acquisitions covering the tumor region. The DCE-MRI acquisition series was followed with a final axial post-contrast T1-weighted acquisition. All images were acquired using the Philips Synergy Body phased array coil (Philips Medical Systems).
All image acquisitions for a given patient visit were designed to have the same field of view and slice location to facilitate definition of the region of interest (ROI) for DCE-MRI data analysis. The DCE-MRI protocol consisted of three-dimensional radiofrequency-spoiled fast field echo acquisitions with a temporal resolution of 2.32 sec. Native tissue T1 was determined using three separate acquisitions before the DCE-MRI time series, with four signal averages and flip angles of 2°, 10°, and 30°, and by fitting the standard relationship between T1 and spoiled gradient echo signal intensity,25 dynamic acquisition consisted of 100 single average volumes (flip angle, 30°). The image matrix for all scans was 128 x 128 in-plane, with 20 slices. An elliptical k-space window and over-contiguous slicing (ie, interpolation in the slice direction within the three-dimensional slab) were utilized to maintain a short total acquisition time. Repetition time was 2.5 msec and echo time was 0.85 msec. Omniscan 0.1 mmol/kg (Amersham Health, Amersham, UK) was administered after the fifth dynamic time point as a bolus using a Spectris MR (Medrad Inc, Indianola, PA) power injector at a rate of 3 mL/sec.
DCE-MRI data analysis Tumor volumes were defined by a radiologist as two-dimensional ROIs on each slice location containing the tumor of interest. ROIs were defined on the precontrast T2-weighted images, with the pre- and postcontrast T1-weighted images used to provide additional visual reference. ROIs were defined once all visits for a given patient had been completed to ensure the same tumor was always being outlined (on occasion, multiple secondary lesions were visible within the field of view). The radiologist was not blind to the order of patient visits and was aware of the CDP860 administration schedule. Tumor volumes were also determined in the pretreatment CT scans by the same radiologist.
The DCE-MRI time series was analyzed using a software package written in-house. We aimed to produce parameter values reflecting tumor microvasculature using methods that have commonly been applied in previous studies of antivascular agents (see methods 1 and 2), and more advanced methods that are designed to provide more specific physiologic information (see methods 3 and 4): (1) Initial area under the tissue contrast agent concentration time curve (IAUC), defined over 60 sec beginning at the point of bolus administration26; (2) standard Tofts and Kermode kinetic modeling to derive estimates of the volume transfer coefficient for contrast agent between the blood pool and the tissue extracellular extravascular space Ktrans (often approximated to the capillary wall permeability surface area product) and the volume of the extracellular extravascular space as a fraction of total tissue volume ve27; a standardized arterial input function (AIF) was assumed for all patients28; (3) kinetic modeling incorporating an estimate of vascular plasma volume vp29; for this method, as well as method 4, we defined the AIF in each patient at each visit using an automated AIF extraction method.30 This method also produces estimates of Ktrans and ve; and (4) tissue homogeneity model. This model (St Lawrence and Lee31) allows estimates of Ktrans, ve, and vp to be made, plus estimates of blood flow and capillary permeability-surface area product.
The use of methods 1 and 2 was intended to provide a directly comparable output to previous studies of antivascular agents. Methods 3 and 4 were additionally applied with the aim of providing a more accurate characterization of the vascular state of the tumors studied.
The change in tissue signal intensity due to contrast agent over time was converted into estimates of contrast agent concentration via estimation of T1.32 Each voxel within the tumor was identified as "enhancing" if the signal intensity rose above 3 standard deviations of the data noise level during the dynamic series; otherwise it was classified as "nonenhancing" and no modeling was performed for that voxel. The total volume occupied by enhancing and nonenhancing tumor was recorded. For each derived parameter (IAUC, three estimates of Ktrans and ve, and two estimates of vp), the median and mean values calculated within the enhancing tumor volume were recorded.
Pharmacokinetics Concentrations of CDP860 were assessed using a sandwich enzyme-linked immunosorbent assay (ELISA) consisting of a murine Fc-PDGFR fusion-coated plate and a goat anti-human kappa-horse radish peroxide conjugate as the detection layer. The limit of quantification for this assay (allowing for the minimum 1/100 dilution) was 1.20 μg/mL. Antibodies to CDP860 were assessed using a double antigen sandwich ELISA consisting of a CDP860 coated plate and a CDP860 biotin detection system. Samples were quantified against a rabbit anti-CDP860 high titer standard and the limit of quantification was 0.6 units/mL.
Statistical design We planned to recruit six patients with colon cancer and six patients with ovarian cancer. Accepting a standard deviation of 0.019 min–1, identified from previous reproducibility studies in glioma,22,23 and using a two-sided {alpha} = 0.05 significance level, this study had an 80% power to detect a 0.025 min–1 change in Ktrans.
RESULTS
Three patients received two drug administrations while the others received only one. The study was stopped after three patients developed dramatic and rapid onset ascites and/or pleural effusions. These data and the other toxicities are summarized in Tables 3 and 4. Largely, they show that besides the fluid accumulation toxicities, the antibody was well tolerated with only grade I and II toxicities that typify the complications seen in patients with advanced cancer (Table 3). All of the patients had progressive disease or developed such marked toxicity in the form of fluid accumulation that further treatment was not possible and they were withdrawn from the study.
Fluid Accumulation
One of the entry criteria for the study stated that the patients should not have significant ascites or pleural effusions. Despite this, seven of eight patients developed clinically significant fluid accumulation that included ascites, pleural effusions, and facial and/or peripheral edema (Table 4). In three patients (patients 3, 4, and 8), the onset and volume of ascites and pleural effusions were dramatic.
Patient 4 was a 58-year-old woman with advanced ovarian adenocarcinoma who had no clinical evidence of ascites or pleural effusion, although the pretreatment CT scan revealed loculated pelvic ascites. Eleven days after administration of CDP860, she presented with massive tense ascites, which had accumulated over 4 days. Sixteen liters of sterile exudative hemorrhagic ascitic fluid was drained over 1 week. A small left-sided pleural effusion was also noted; however, this did not require any intervention and resolved spontaneously. Cytologic examination of the ascitic fluid revealed malignant cells compatible with her ovarian cancer. Following withdrawal from the study, the patient did require a further paracentesis, but the pleural effusion did not recur.
Patient 3 was a 60-year-old woman with advanced granulosa theca cell ovarian tumor whose pretreatment CT scan showed minimal pelvic ascites. There was no clinical evidence of ascites or pleural effusion. Within 14 days of drug administration she was admitted for therapeutic paracentesis, wherein 3.5 L of hemorrhagic exudative fluid were drained, although her weight had increased by 11 kg, suggesting that she had developed 11 L ascites in total. After withdrawal from the trial, the ascites did not re-accumulate.
Patient 8 was a 54-year-old woman with advanced ovarian cancer who had no pretreatment evidence of a fluid collection. By day 13, she had developed gradually worsening abdominal distension, and 6 L of hemorrhagic ascitic fluid were drained and did not reaccumulate. Bilateral small pleural effusions were noted radiologically, but these resolved spontaneously.
Although these patients had ovarian cancer, the development of fluid accumulation was noted in seven of the eight patients on the trial, and so the phenomenon was not disease-specific. Following the admission of patient 8 for paracentesis, recruitment was stopped.
Magnetic Resonance Imaging
Tumor growth Tumor growth during the study for each patient was assessed using both MRI and pretreatment CT data. These data (not shown) demonstrate that a characteristically exponential volume increase pattern was not changed by the administration of CDP860. In patient 3, the rate of tumor growth was extremely slow both before and after drug treatment.
Tumor vascularized volume We examined the relationship between the tumor vascularized volume, defined as the tumor tissue that was found to enhance during the DCE-MRI series, and total tumor volume (Table 5). Figure 1 shows the relationship between these measures. In patients 4, 5, and 8, the vascularized volume of the tumor appeared to increase following administration of CDP860 to a greater extent than did the total tumor volume. An independent samples t test on the percentage change at each patient visit, in comparison with the distribution of percentage variability at baseline, showed visit-specific proportional vascular volume increases after treatment in these patients (Table 5). In each case, the increase in the ratio of vascular tumor volume to total tumor volume was evident on the scan taken 1 day after the dose of CDP860 and maintained thereafter in those patients for whom scans were available (patients 5 and 8). A significant increase was also noted in patient 1 at days 7 and 27, but not at day 1. An independent samples t test on the percentage change over all post-treatment time points in the proportional vascularized volume—when grouping the patients into numbers 2, 3, 6, 7 and 1, 4, 5, 8—is significant at the P < .001 level (two-tailed).
Vascular permeability and flow All approaches applied to investigate tumor microvasculature using the DCE-MRI data showed no detectable effect during the administration of CDP860, suggesting that the drug did not alter tumor IAUC (Table 6) or Ktrans, as determined using any of the available methods (Table 7). This suggests that the characteristics of vessels that were patent before administration of CDP860 were unchanged, and that any newly recruited or newly patent vessels had similar characteristics to the pre-dose viable vessels. Similarly, there was no observable effect on vascular flow in enhancing vessels following CDP860 administration (Table 8).
Pharmacokinetics The highest geometric mean plasma CDP860 concentration was measured at the end of the infusion on day 0. The geometric mean concentration had decreased on day 1 (261,957 μg/mL) and was further reduced on day 7 (93,220 μg/mL). Each subject who received two doses of CDP860 had similar plasma concentrations of CDP860 after each dose. The data are similar to those in patients without cancer (Celltech, data on file), and are consistent with a plasma half-life of approximately 5 days. In all subjects, plasma concentrations of anti-CDP860 antibodies were below the level of quantification at all times.
DISCUSSION
We examined changes in vascular parameters to assess whether CDP860 administration resulted in changes predictive of increased uptake of concomitantly administered drug. There was no change in mean tumor Ktrans or blood flow after CDP860 administration. This suggests that newly perfused vessels had similar perfusion and leakage characteristics to those originally present.
In some patients, the vascularized volume of the tumor appeared to increase proportionally to a greater extent than the total tumor volume following CDP860 administration. This appeared to be independent of both the absolute volume and location of the tumor (Table 5 and Fig 1). The case of patient 5 is striking, where the vascularized volume more than doubled following CDP860, and the ratio of tumor volume to vascularized volume is subsequently maintained as the tumor grows (Table 5). One explanation for the increase in vascularized volume might be recruitment of pre-existing blood vessels that were compressed before CDP860 administration, owing to high tumor IFP. This hypothesis is consistent with previously reported animal experiments. Firstly, treatment of xenograft-bearing rodents with a DNA aptamer against PDGF B is associated with increased tumor volumetric density and reduced IFP.1 Secondly, there are compressed blood vessels in rodent xenograft tissue, thought to result from increased IFP.33 Following paclitaxel administration to animals, which was associated with reduction in tumor IFP, the vessel density of the xenograft increased.33 There is growing evidence for the importance of PDGF in the establishment of vascular stability and vascular pattern formation in tumors.34 The rapid onset of increased vascularized tumor volume in a subset of CDP860-treated patients is, however, more consistent with the recruitment of previously poorly perfused unperfused vessels, than it is with the role of PDGF in tumor angiogenesis.
Seven of eight patients developed adverse events suggestive of fluid accumulation. Although two of the patients most severely affected also experienced the increase in vascular volume (patients 4 and 8), one patient with increased vascular volume had no apparent fluid accumulation (patient 5). This implies that toxic and potentially beneficial effects of CDP860 are not always associated.
The exact mechanism of fluid accumulation associated with CDP860 is unknown. Based on our results with this PDGFR- inhibitor, similar adverse events observed with other less specific inhibitors of PDGF (eg, imatinib,35,36 SU101,37,38 SU6668,39-43 and SU1124844) probably resulted from inhibition of PDGF signaling rather than other mechanisms. As some of these drugs (eg, SU11248) affect other receptor kinases, fluid accumulation may not occur to the same extent. This might be the case because either the inhibition of other kinases overcomes the effect of inhibiting PDGFR- or because the inhibition of other kinases dominates the dose-limiting toxicity. Although the mechanism of ascites generation is unclear, it is likely that the dynamic balance between intravascular and extravascular fluid is more precarious in patients with diffuse intra-abdominal disease. A perturbation such as a reduction in tumor interstitial pressure, mediated by PDGF antagonists, might tip the balance in favor of extravascular fluid thereby allowing ascites to collect. In fact, two patients had a trace of ascites present initially, and perhaps their dynamic equilibrium was the most vulnerable to reductions in interstitial pressure. However, it remains striking that in our analyses there was no change in the Ktrans per unit tissue volume, suggesting that changes in vascular permeability were not responsible for the ascites. It will be particularly interesting to see if other studies find similar changes in vascularized volume in patients treated with mixed kinase inhibitors that target the PDGF receptors.
It may be important to consider whether inhibition of PDGFR- should be avoided for mixed kinase inhibitors. Kinase inhibitors whose selectivity profiles exclude PDGFR may be more clinically useful through their avoidance of ascites. Alternatively, the effects observed when PDGFR- is specifically inhibited may not occur with mixed inhibitors if inhibition of other kinases can counteract this biologic effect. As the signaling pathways of PDGF are complex, it will be interesting to determine which downstream pathways mediate the fluid accumulation effect. Mice expressing PDGFR- mutated in the PI3K binding sites were unable to normalize IFP in response to PDGF following administration of a mast cell degranulating agent, indicating a role for the PI3K pathway.45 There is evidence to suggest that inhibition of PDGFR may be clinically useful through other mechanisms, such as inhibition of autocrine effects, or inhibition of angiogenesis. As we better understand the mechanisms involved in fluid accumulation, we may learn to manage this side effect. While this study was limited by the small number of patients, our observations support further study of controlled manipulation of PDGF-mediated IFP through biologic agents such as CDP860, or small molecule antagonists. Modulation of other pathways may also lead to a decrease in IFP. Willett et al46 have recently shown that inhibition of vascular endothelial growth factor causes a decrease in IFP in human rectal carcinomas.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Simon Tickle, Martyn Robinson, and Jeff Smith for input into the concept of CDP860 in the oncology setting. We also thank Hilary Done for operational support of the study, Fran Oates and Nalini Parmar for management of the data, Pete Jeffrey for statistical advice during preparation of the protocol, and Joby Jose for pharmacokinetic and antibody assays.
NOTES
Supported by Celltech R & D Ltd, Slough, UK.
Terms in blue are defined in the glossary, found at the end of this issue and online at www.jco.org.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Heldin C-H, Westermark B: Mechanism of action and in vivo role of platelet-derived growth factor. Physiol Rev 79:1283-1316, 1999
2. Li X, Ponten A, Aase K, et al: PDGF-C is a new protease-activated ligand for the PDGF alpha-receptor. Nat Cell Biol 2:302-309, 2000
3. Bergsten E, Uutela M, Li X, et al: PDGF-D is a specific, protease-activated ligand for the PDGF beta-receptor. Nat Cell Biol 3:512-516, 2001
4. LaRochelle WJ, Jeffers M, McDonald WF, et al: PDGF-D, a new protease-activated growth factor. Nat Cell Biol 3:517-521, 2001
5. Leu KM, Thomas DG, Baker LH: Platelet-derived growth factor receptor (PDGF-R) a and expression in human malignancy. Proc American Soc Clin Oncol 21:1774, 2002
6. Dabrow MB, Francesco MR, McBrearty FX, et al: The effects of platelet-derived growth factor and receptor on normal and neoplastic human ovarian surface epithelium. Gynecol Oncol 71:29-37, 1998
7. Lindmark G, Sundberg C, Glimelius B, et al: Stromal expression of platelet-derived growth factor beta-receptor and platelet-derived growth factor B-chain in colorectal cancer. Lab Invest 69:682-689, 1993
8. Craven RJ, Xu LH, Weiner TM, et al: Receptor tyrosine kinases expressed in metastatic colon cancer. Int J Cancer 60:791-797, 1995
9. Bergers G, Song S, Meyer-Morse N, et al: Benefits of targeting both pericytes and endothelial cells in the tumor vasculature with kinase inhibitors. J Clin Invest 111:1287-1295, 2003
10. Hellstrom M, Kalen M, Lindahl P, et al: Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development 126:3047-3055, 1999
11. Grinnell F, Ho CH, Lin YC, et al: Differences in the regulation of fibroblast contraction of floating versus stressed collagen matrices. J Biol Chem 274:918-923, 1999
12. Rodt SA, Ahlen K, Berg A, et al: A novel physiological function for platelet-derived growth factor-BB in rat dermis. J Physiol 495:193-200, 1996
13. Pietras K, Ostman A, Sjoquist M, et al: Inhibition of platelet-derived growth factor receptors reduces interstitial hypertension and increases transcapillary transport in tumors. Cancer Res 61:2929-2934, 2001
14. Pietras K, Rubin K, Sjoblom T, et al: Inhibition of PDGF receptor signaling in tumor stroma enhances antitumor effect of chemotherapy. Cancer Res 62:5476-5484, 2002
15. Pietras K, Stumm M, Hubert M, et al: STI571 enhances the therapeutic index of epothilone B by a tumor-selective increase of drug uptake. Clin Cancer Res 9:3779-3787, 2003
16. Chapman A and King D: Divalent antibody fragments. WO99/64460, 1999
17. Hart CE, Kraiss LW, Vergel S, et al: PDGFbeta receptor blockade inhibits intimal hyperplasia in the baboon. Circulation 99:564-569, 1999
18. Davies MG, Owens EL, Mason DP, et al: Effect of platelet-derived growth factor receptor-alpha and -beta blockade on flow-induced neointimal formation in endothelialized baboon vascular grafts. Circ Res 86:779-786, 2000
19. Serruys PW, Heyndrickx GR, Patel J, et al: Effect of an anti-PDGF b receptor blocking antibody on restenosis in patients undergoing elective stent placement. Int J Cardiovasc Intervent 5:214-22, 2003
20. Adair JR, Athwal DS, Emtage JS: Humanised antibodies. US Patent Application 20040076627, 1991
21. Chapman AP, Antoniw P, Spitali M, et al: Therapeutic antibody fragments with prolonged in vivo half-lives. Nat Biotechnol 17:780-783, 1999
22. Jackson A, Jayson GC, Li KL, et al: Reproducibility of quantitative dynamic contrast-enhanced MRI in newly presenting glioma. Br J Radiol 76:153-162, 2003
23. Jackson A, Haroon H, Zhu XP, et al: Breath hold perfusion and permeability mapping of hepatic malignancies using magnetic resonance imaging and a first pass leakage profile model. NMR Biomed 15:164-173, 2002
24. Therasse P, Arbuck SG, Eisenhauer EA, et al: New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92:205-216, 2000
25. Haase A: Snapshot FLASH MRI. Applications to T1, T2, and chemical-shift imaging. Magn Reson Med 13:77-89, 1990
26. Evelhoch JL: Key factors in the acquisition of contrast kinetic data for oncology. J Magn Reson Imaging 10:254-259, 1999
27. Tofts PS, Brix G, Buckley DL, et al: Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a diffusable tracer: Standardized quantities and symbols. J Magn Reson Imaging 10:223-232, 1999
28. Tofts PS, Kermode AG: Measurement of the blood-brain barrier permeability and leakage space using dynamic MR imaging. 1. Fundamental concepts. Magn Reson Med 17:357-367, 1991
29. Daldrup HE, Shames DM, Husseini W, et al: Quantification of the extraction fraction for gadopentetate across breast cancer capillaries. Magn Reson Med 40:537-543, 1998
30. Parker GJ, Jackson A, Waterton JC: Automated arterial input function extraction for T1-weighted DCE-MRI. Proceedings of the International Society for Magnetic Resonance in Medicine 11:2003 (abstr 1264)
31. St Lawrence KS, Lee TY: An adiabatic approximation to the tissue homogeneity model for water exchange in the brain: I. Theoretical derivation. J Cereb Blood Flow Metab 18:1365-1377, 1998
32. Parker GJM, Padhani AR: T1-weighted dynamic contrast enhanced MRI, in Tofts PS (ed): Quantitative MRI of the brain—Measuring changes caused by disease. Chichester, John Wiley and Sons Ltd, 2003, pp 341-364
33. Griffon-Etienne G, Boucher Y, Brekken C, et al: Taxane-induced apoptosis decompresses blood vessels and lowers interstitial fluid pressure in solid tumors: clinical implications. Cancer Res 59:3776-3782, 1999
34. Abramsson A, Lindblom P, Betsholtz C: Endothelial and nonendothelial sources of PDGF-B regulate pericyte recruitment and influence vascular pattern formation in tumors. J Clin Invest 112:1142-1151, 2003
35. Savage DG, Antman KH: Imatinib mesylate–A new oral targeted therapy. N Engl J Med 346:683-693, 2002
36. Hussain M, Kotz H, Minasian L: Occurrence of ascites secondary to STI571 in ovarian cancer patients. Proc Am Soc Clin Oncol 22:220, 2003 (abstr 880)
37. Ko YJ, Small EJ, Kabbinavar F, et al: A multi-institutional phase II study of SU101, a platelet-derived growth factor receptor inhibitor, for patients with hormone-refractory prostate cancer. Clin Cancer Res 7:800-805, 2001
38. Eckhardt SG, Rizzo J, Sweeney KR, et al: Phase I and pharmacologic study of the tyrosine kinase inhibitor SU101 in patients with advanced solid tumors. J Clin Oncol 17:1095-1104, 1999
39. Britten CD, Rosen LS, Kabbinavar F: Phase I trial of SU6668, a small molecule receptor tyrosine kinase inhibitor, given twice daily in patients with advanced cancers. Proc Am Soc Clin Oncol 21:28b, 2002 (abstr 1922)
40. Kuenen B, Ruijter R, Hoekman K: Dose finding study of SU6668 given thrice daily by oral route under fed conditions in patients with advanced malignancies. Proc Am Soc Clin Oncol 20:110a, 2001 (abstr 437)
41. Brahmer JR, Kelsey S, Scigalla P, et al: A phase I study of SU6668 in patients with refractory solid tumors. Proc Am Soc Clin Oncol 21:84a, 2002 (abstr 335)
42. Ueda Y, Shimoyama T, Murakami H, et al: Phase I study of TSU-68, VEGF receptor tyrosine kinase inhibitor, by twice daily oral administration between meals in patients with advanced solid tumors. Proc Am Soc Clin Oncol 21:111a, 2002 (abstr 443)
43. Murakami H, Yamamoto N, Shimoyama T, et al: Phase I, pharmacokinetic, and biological studies of TSU-68, the oral vascular endothelial growth factor receptor tyrosine kinase inhibitor, administered after meals in patients with advanced solid tumors. Proc Am Soc Clin Oncol 22:217, 2003 (abstr 870)
44. Raymond E, Faivre S, Vera K, et al: Final results of a phase I and pharmacokinetic study of SU11248, a novel multi-target tyrosine kinase inhibitor, in patients with advanced cancers. Proc Am Soc Clin Oncol 22:192, 2003 (abstr 769)
45. Heuchel R, Berg A, Tallquist M, et al: Platelet-derived growth factor beta receptor regulates interstitial fluid homeostasis through phosphatidylinositol-3' kinase signaling. Proc Natl Acad Sci U S A 96:11410-11415, 1999
46. Willett CG, Boucher Y, di Tomaso E, et al: Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 10:145-147, 2004(G.C. Jayson, G.J.M. Parke)
Imaging Science and Biomedical Engineering, University of Manchester, Manchester
Oncology Group, Celltech R & D Ltd, Slough, UK
ABSTRACT
PATIENTS AND METHODS: Patients with advanced ovarian or colorectal cancer and good performance status received intravenous infusions of CDP860 on days 0 and 28. Patients had serial dynamic contrast-enhanced magnetic resonance imaging studies to measure changes in tumor vascular parameters.
RESULTS: Three of eight patients developed significant ascites, and seven of eight showed evidence of fluid retention. In some patients, the ratio of vascular volume to total tumor volume increased significantly (P < .001) within 24 hours following CDP860 administration, an effect suggestive of recruitment of previously non-functioning vessels.
CONCLUSION: These observations suggest that inhibition of PDGFR- might improve delivery of a concurrently administered therapy. However, in cancer patients, further exploration of the dosing regimen of CDP860 is required to dissociate adverse effects from beneficial effects. The findings challenge the view that inhibition of PDGF alone is beneficial, and confirm that effects of PDGFR kinase inhibition mediate, to some extent, the fluid retention observed in patients treated with mixed tyrosine kinase inhibitors.
INTRODUCTION
PDGFR- is expressed on many tumor types,5 including ovarian 5,6 and colon cancers.5,7,8 In addition, the receptors are expressed on pericytes,9 suggesting that PDGF signaling may be involved in the regulation of angiogenesis.10 Inhibition of pericyte and endothelial function is associated with enhanced tumor growth inhibition9 in animals.
PDGF probably plays a role in the control of tissue interstitial fluid pressure (IFP). Evidence suggests that PDGF is an important signaling molecule for the maintenance of the interaction of myofibroblasts with collagen in the extracellular matrix, via 1 integrins.11 It was previously shown that PDGF effects on IFP were specific to PDGF BB. In a rat model, administration of PDGF BB, but not PDGF AA, normalized dermal IFP that had been experimentally lowered.12 Studies using rodent tumor models showed that inhibition of PDGF using the tyrosine kinase inhibitor imatinib, or an anti-PDGF B DNA aptamer, was associated with a decrease in tumor IFP.13 The authors hypothesized that the reduction in IFP by PDGF inhibition would increase the blood flow–driven uptake of small molecules, and confirmed the hypothesis using 51Cr-EDTA,13 and subsequently, [3H]-paclitaxel14 and epothilone B.15 In the Pietras et al study,15 imatinib increased the tumor uptake of epothilone B without increasing uptake in liver, kidney, or intestinal tract, and without reduction in the tolerability of the cytotoxic agent.
In patients with colorectal and ovarian carcinoma, we report the phase II evaluation of a pegylated di-Fab' molecule,16 CDP860, which binds to and inhibits PDGFR-. CDP860 cross-reacts with PDGFR- from cynomolgus monkey, and was well tolerated in toxicology evaluation in this species at doses up to 400 mg/kg, given weekly for 4 weeks.
One hypothetical benefit of inhibiting PDGF is the potential to prevent coronary artery restenosis after stent insertion. Primate models of neointimal formation suggested a role for PDGFR- in intimal hyperplasia.17,18 A study was performed in which 145 patients undergoing intracoronary stent insertion received placebo or 25 mg/kg CDP860 (n = 76), the drug having been well tolerated at doses up to 30 mg/kg in a healthy volunteer study. Although there was no improvement in restenosis rate, 47 of the patients receiving CDP860 developed peripheral or facial edema between 2 hours and 3 weeks after dosing. These observations were attributed to the mode of action of CDP860.19
In total, 97 volunteers and patients had received CDP860 before this study. Following the reported effects of PDGF BB inhibition on IFP in animal models, we wanted to determine if changes in tumor vascular parameters could be detected in humans, and to assess whether CDP860 would be likely to increase the uptake of a concurrently administered small molecule in future studies.
PATIENTS AND METHODS
Phase II Evaluation of CDP860 in Advanced Ovarian and Colorectal Cancer
Aims of the study The study examined changes in tumor vascular parameters assessed using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in patients with advanced colorectal or ovarian cancer who had received CDP860. The primary end point was the change in tumor vascular permeability measured for up to 45 days after administration of the CDP860. Secondary end points were effects of CDP860 on other vascular parameters including tumor blood flow, safety and tolerability of CDP860, plasma pharmacokinetics, and tumor response.
Patients Patients with measurable primary or secondary colon or ovarian cancer over 18 years of age, an Eastern Cooperative Oncology Group score between 0 and 2, and life expectancy of at least 3 months were eligible. They were required to have adequate liver (normal bilirubin and transaminases < 2.5x the upper limit of normal), renal (creatinine < 1.5x the upper limit of normal), and hematologic (hemoglobin >10 g/dL, white cell count > 3 x 109/L, and platelet count >100 x 109/L) function, as well as normal coagulation (prothrombin time and activated partial thromboplastin time) and a normal ECG.
Patients were excluded if they had an additional chronic disease affecting a major organ, infection requiring antibiotics, clinically significant ascites or pleural effusion, major surgery within the previous 4 weeks, previous history of reaction to biologic agents or drugs containing PEG, a contra-indication to magnetic resonance imaging (MRI), history of infection with hepatitis B, C, HIV-1 or HTLV-1, alcohol or drug addiction, or treatment within the previous 4 weeks. Patients were not permitted to take drugs known to alter vascular flow and they had to have recovered from the effects of previous treatments. Patients with known or suspected brain or CNS disease were excluded.
Study design Patients were treated at the Christie Cancer Center (Manchester, UK) and received CDP860 25 mg/kg on days 0 and 28. A computed tomography (CT) scan was performed during the 2 weeks before treatment, and a post-treatment scan was performed on day 45. CT was also performed as part of the routine clinical monitoring of the patients, on average, 68 days before the first study-specific CT scan. MRI scans were performed twice at baseline to establish reproducibility,22,23 and on days 1, 7, 27, and 45. Blood samples were taken to measure hematologic and biochemical toxicity, CDP860 concentration, and anti-idiotype responses at screening before treatment, and on days 1, 7, 28, 35, and 50. Adverse events were assessed before treatment and on days 0, 1, 7, 27, 28, 35, 45, and 50. Toxicity was assessed using the National Cancer Institute Common Toxicity Criteria (version 2.0). Response was assessed using Response Evaluation Criteria in Solid Tumors.24 Any patient with stable disease or better at the end of treatment was eligible to continue treatment for a further 6 months.
All patients gave written informed consent and the study was approved by the South Manchester Local Research Ethics Committee and the Christie Hospital Research Committee.
MRI The location of the tumor to be studied using DCE-MRI was ascertained from the CT scans. All MRIs were performed on a Philips Gyroscan NT Intera system (Philips Medical Systems, Best, Netherlands) at 1.5 T at the University of Manchester. The MRI protocol consisted of localizer acquisitions, followed by axial T1-weighted fast field echo (gradient echo), and T2-weighted fast spin echo volumetric acquisitions covering the tumor region. The DCE-MRI acquisition series was followed with a final axial post-contrast T1-weighted acquisition. All images were acquired using the Philips Synergy Body phased array coil (Philips Medical Systems).
All image acquisitions for a given patient visit were designed to have the same field of view and slice location to facilitate definition of the region of interest (ROI) for DCE-MRI data analysis. The DCE-MRI protocol consisted of three-dimensional radiofrequency-spoiled fast field echo acquisitions with a temporal resolution of 2.32 sec. Native tissue T1 was determined using three separate acquisitions before the DCE-MRI time series, with four signal averages and flip angles of 2°, 10°, and 30°, and by fitting the standard relationship between T1 and spoiled gradient echo signal intensity,25 dynamic acquisition consisted of 100 single average volumes (flip angle, 30°). The image matrix for all scans was 128 x 128 in-plane, with 20 slices. An elliptical k-space window and over-contiguous slicing (ie, interpolation in the slice direction within the three-dimensional slab) were utilized to maintain a short total acquisition time. Repetition time was 2.5 msec and echo time was 0.85 msec. Omniscan 0.1 mmol/kg (Amersham Health, Amersham, UK) was administered after the fifth dynamic time point as a bolus using a Spectris MR (Medrad Inc, Indianola, PA) power injector at a rate of 3 mL/sec.
DCE-MRI data analysis Tumor volumes were defined by a radiologist as two-dimensional ROIs on each slice location containing the tumor of interest. ROIs were defined on the precontrast T2-weighted images, with the pre- and postcontrast T1-weighted images used to provide additional visual reference. ROIs were defined once all visits for a given patient had been completed to ensure the same tumor was always being outlined (on occasion, multiple secondary lesions were visible within the field of view). The radiologist was not blind to the order of patient visits and was aware of the CDP860 administration schedule. Tumor volumes were also determined in the pretreatment CT scans by the same radiologist.
The DCE-MRI time series was analyzed using a software package written in-house. We aimed to produce parameter values reflecting tumor microvasculature using methods that have commonly been applied in previous studies of antivascular agents (see methods 1 and 2), and more advanced methods that are designed to provide more specific physiologic information (see methods 3 and 4): (1) Initial area under the tissue contrast agent concentration time curve (IAUC), defined over 60 sec beginning at the point of bolus administration26; (2) standard Tofts and Kermode kinetic modeling to derive estimates of the volume transfer coefficient for contrast agent between the blood pool and the tissue extracellular extravascular space Ktrans (often approximated to the capillary wall permeability surface area product) and the volume of the extracellular extravascular space as a fraction of total tissue volume ve27; a standardized arterial input function (AIF) was assumed for all patients28; (3) kinetic modeling incorporating an estimate of vascular plasma volume vp29; for this method, as well as method 4, we defined the AIF in each patient at each visit using an automated AIF extraction method.30 This method also produces estimates of Ktrans and ve; and (4) tissue homogeneity model. This model (St Lawrence and Lee31) allows estimates of Ktrans, ve, and vp to be made, plus estimates of blood flow and capillary permeability-surface area product.
The use of methods 1 and 2 was intended to provide a directly comparable output to previous studies of antivascular agents. Methods 3 and 4 were additionally applied with the aim of providing a more accurate characterization of the vascular state of the tumors studied.
The change in tissue signal intensity due to contrast agent over time was converted into estimates of contrast agent concentration via estimation of T1.32 Each voxel within the tumor was identified as "enhancing" if the signal intensity rose above 3 standard deviations of the data noise level during the dynamic series; otherwise it was classified as "nonenhancing" and no modeling was performed for that voxel. The total volume occupied by enhancing and nonenhancing tumor was recorded. For each derived parameter (IAUC, three estimates of Ktrans and ve, and two estimates of vp), the median and mean values calculated within the enhancing tumor volume were recorded.
Pharmacokinetics Concentrations of CDP860 were assessed using a sandwich enzyme-linked immunosorbent assay (ELISA) consisting of a murine Fc-PDGFR fusion-coated plate and a goat anti-human kappa-horse radish peroxide conjugate as the detection layer. The limit of quantification for this assay (allowing for the minimum 1/100 dilution) was 1.20 μg/mL. Antibodies to CDP860 were assessed using a double antigen sandwich ELISA consisting of a CDP860 coated plate and a CDP860 biotin detection system. Samples were quantified against a rabbit anti-CDP860 high titer standard and the limit of quantification was 0.6 units/mL.
Statistical design We planned to recruit six patients with colon cancer and six patients with ovarian cancer. Accepting a standard deviation of 0.019 min–1, identified from previous reproducibility studies in glioma,22,23 and using a two-sided {alpha} = 0.05 significance level, this study had an 80% power to detect a 0.025 min–1 change in Ktrans.
RESULTS
Three patients received two drug administrations while the others received only one. The study was stopped after three patients developed dramatic and rapid onset ascites and/or pleural effusions. These data and the other toxicities are summarized in Tables 3 and 4. Largely, they show that besides the fluid accumulation toxicities, the antibody was well tolerated with only grade I and II toxicities that typify the complications seen in patients with advanced cancer (Table 3). All of the patients had progressive disease or developed such marked toxicity in the form of fluid accumulation that further treatment was not possible and they were withdrawn from the study.
Fluid Accumulation
One of the entry criteria for the study stated that the patients should not have significant ascites or pleural effusions. Despite this, seven of eight patients developed clinically significant fluid accumulation that included ascites, pleural effusions, and facial and/or peripheral edema (Table 4). In three patients (patients 3, 4, and 8), the onset and volume of ascites and pleural effusions were dramatic.
Patient 4 was a 58-year-old woman with advanced ovarian adenocarcinoma who had no clinical evidence of ascites or pleural effusion, although the pretreatment CT scan revealed loculated pelvic ascites. Eleven days after administration of CDP860, she presented with massive tense ascites, which had accumulated over 4 days. Sixteen liters of sterile exudative hemorrhagic ascitic fluid was drained over 1 week. A small left-sided pleural effusion was also noted; however, this did not require any intervention and resolved spontaneously. Cytologic examination of the ascitic fluid revealed malignant cells compatible with her ovarian cancer. Following withdrawal from the study, the patient did require a further paracentesis, but the pleural effusion did not recur.
Patient 3 was a 60-year-old woman with advanced granulosa theca cell ovarian tumor whose pretreatment CT scan showed minimal pelvic ascites. There was no clinical evidence of ascites or pleural effusion. Within 14 days of drug administration she was admitted for therapeutic paracentesis, wherein 3.5 L of hemorrhagic exudative fluid were drained, although her weight had increased by 11 kg, suggesting that she had developed 11 L ascites in total. After withdrawal from the trial, the ascites did not re-accumulate.
Patient 8 was a 54-year-old woman with advanced ovarian cancer who had no pretreatment evidence of a fluid collection. By day 13, she had developed gradually worsening abdominal distension, and 6 L of hemorrhagic ascitic fluid were drained and did not reaccumulate. Bilateral small pleural effusions were noted radiologically, but these resolved spontaneously.
Although these patients had ovarian cancer, the development of fluid accumulation was noted in seven of the eight patients on the trial, and so the phenomenon was not disease-specific. Following the admission of patient 8 for paracentesis, recruitment was stopped.
Magnetic Resonance Imaging
Tumor growth Tumor growth during the study for each patient was assessed using both MRI and pretreatment CT data. These data (not shown) demonstrate that a characteristically exponential volume increase pattern was not changed by the administration of CDP860. In patient 3, the rate of tumor growth was extremely slow both before and after drug treatment.
Tumor vascularized volume We examined the relationship between the tumor vascularized volume, defined as the tumor tissue that was found to enhance during the DCE-MRI series, and total tumor volume (Table 5). Figure 1 shows the relationship between these measures. In patients 4, 5, and 8, the vascularized volume of the tumor appeared to increase following administration of CDP860 to a greater extent than did the total tumor volume. An independent samples t test on the percentage change at each patient visit, in comparison with the distribution of percentage variability at baseline, showed visit-specific proportional vascular volume increases after treatment in these patients (Table 5). In each case, the increase in the ratio of vascular tumor volume to total tumor volume was evident on the scan taken 1 day after the dose of CDP860 and maintained thereafter in those patients for whom scans were available (patients 5 and 8). A significant increase was also noted in patient 1 at days 7 and 27, but not at day 1. An independent samples t test on the percentage change over all post-treatment time points in the proportional vascularized volume—when grouping the patients into numbers 2, 3, 6, 7 and 1, 4, 5, 8—is significant at the P < .001 level (two-tailed).
Vascular permeability and flow All approaches applied to investigate tumor microvasculature using the DCE-MRI data showed no detectable effect during the administration of CDP860, suggesting that the drug did not alter tumor IAUC (Table 6) or Ktrans, as determined using any of the available methods (Table 7). This suggests that the characteristics of vessels that were patent before administration of CDP860 were unchanged, and that any newly recruited or newly patent vessels had similar characteristics to the pre-dose viable vessels. Similarly, there was no observable effect on vascular flow in enhancing vessels following CDP860 administration (Table 8).
Pharmacokinetics The highest geometric mean plasma CDP860 concentration was measured at the end of the infusion on day 0. The geometric mean concentration had decreased on day 1 (261,957 μg/mL) and was further reduced on day 7 (93,220 μg/mL). Each subject who received two doses of CDP860 had similar plasma concentrations of CDP860 after each dose. The data are similar to those in patients without cancer (Celltech, data on file), and are consistent with a plasma half-life of approximately 5 days. In all subjects, plasma concentrations of anti-CDP860 antibodies were below the level of quantification at all times.
DISCUSSION
We examined changes in vascular parameters to assess whether CDP860 administration resulted in changes predictive of increased uptake of concomitantly administered drug. There was no change in mean tumor Ktrans or blood flow after CDP860 administration. This suggests that newly perfused vessels had similar perfusion and leakage characteristics to those originally present.
In some patients, the vascularized volume of the tumor appeared to increase proportionally to a greater extent than the total tumor volume following CDP860 administration. This appeared to be independent of both the absolute volume and location of the tumor (Table 5 and Fig 1). The case of patient 5 is striking, where the vascularized volume more than doubled following CDP860, and the ratio of tumor volume to vascularized volume is subsequently maintained as the tumor grows (Table 5). One explanation for the increase in vascularized volume might be recruitment of pre-existing blood vessels that were compressed before CDP860 administration, owing to high tumor IFP. This hypothesis is consistent with previously reported animal experiments. Firstly, treatment of xenograft-bearing rodents with a DNA aptamer against PDGF B is associated with increased tumor volumetric density and reduced IFP.1 Secondly, there are compressed blood vessels in rodent xenograft tissue, thought to result from increased IFP.33 Following paclitaxel administration to animals, which was associated with reduction in tumor IFP, the vessel density of the xenograft increased.33 There is growing evidence for the importance of PDGF in the establishment of vascular stability and vascular pattern formation in tumors.34 The rapid onset of increased vascularized tumor volume in a subset of CDP860-treated patients is, however, more consistent with the recruitment of previously poorly perfused unperfused vessels, than it is with the role of PDGF in tumor angiogenesis.
Seven of eight patients developed adverse events suggestive of fluid accumulation. Although two of the patients most severely affected also experienced the increase in vascular volume (patients 4 and 8), one patient with increased vascular volume had no apparent fluid accumulation (patient 5). This implies that toxic and potentially beneficial effects of CDP860 are not always associated.
The exact mechanism of fluid accumulation associated with CDP860 is unknown. Based on our results with this PDGFR- inhibitor, similar adverse events observed with other less specific inhibitors of PDGF (eg, imatinib,35,36 SU101,37,38 SU6668,39-43 and SU1124844) probably resulted from inhibition of PDGF signaling rather than other mechanisms. As some of these drugs (eg, SU11248) affect other receptor kinases, fluid accumulation may not occur to the same extent. This might be the case because either the inhibition of other kinases overcomes the effect of inhibiting PDGFR- or because the inhibition of other kinases dominates the dose-limiting toxicity. Although the mechanism of ascites generation is unclear, it is likely that the dynamic balance between intravascular and extravascular fluid is more precarious in patients with diffuse intra-abdominal disease. A perturbation such as a reduction in tumor interstitial pressure, mediated by PDGF antagonists, might tip the balance in favor of extravascular fluid thereby allowing ascites to collect. In fact, two patients had a trace of ascites present initially, and perhaps their dynamic equilibrium was the most vulnerable to reductions in interstitial pressure. However, it remains striking that in our analyses there was no change in the Ktrans per unit tissue volume, suggesting that changes in vascular permeability were not responsible for the ascites. It will be particularly interesting to see if other studies find similar changes in vascularized volume in patients treated with mixed kinase inhibitors that target the PDGF receptors.
It may be important to consider whether inhibition of PDGFR- should be avoided for mixed kinase inhibitors. Kinase inhibitors whose selectivity profiles exclude PDGFR may be more clinically useful through their avoidance of ascites. Alternatively, the effects observed when PDGFR- is specifically inhibited may not occur with mixed inhibitors if inhibition of other kinases can counteract this biologic effect. As the signaling pathways of PDGF are complex, it will be interesting to determine which downstream pathways mediate the fluid accumulation effect. Mice expressing PDGFR- mutated in the PI3K binding sites were unable to normalize IFP in response to PDGF following administration of a mast cell degranulating agent, indicating a role for the PI3K pathway.45 There is evidence to suggest that inhibition of PDGFR may be clinically useful through other mechanisms, such as inhibition of autocrine effects, or inhibition of angiogenesis. As we better understand the mechanisms involved in fluid accumulation, we may learn to manage this side effect. While this study was limited by the small number of patients, our observations support further study of controlled manipulation of PDGF-mediated IFP through biologic agents such as CDP860, or small molecule antagonists. Modulation of other pathways may also lead to a decrease in IFP. Willett et al46 have recently shown that inhibition of vascular endothelial growth factor causes a decrease in IFP in human rectal carcinomas.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Simon Tickle, Martyn Robinson, and Jeff Smith for input into the concept of CDP860 in the oncology setting. We also thank Hilary Done for operational support of the study, Fran Oates and Nalini Parmar for management of the data, Pete Jeffrey for statistical advice during preparation of the protocol, and Joby Jose for pharmacokinetic and antibody assays.
NOTES
Supported by Celltech R & D Ltd, Slough, UK.
Terms in blue are defined in the glossary, found at the end of this issue and online at www.jco.org.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Heldin C-H, Westermark B: Mechanism of action and in vivo role of platelet-derived growth factor. Physiol Rev 79:1283-1316, 1999
2. Li X, Ponten A, Aase K, et al: PDGF-C is a new protease-activated ligand for the PDGF alpha-receptor. Nat Cell Biol 2:302-309, 2000
3. Bergsten E, Uutela M, Li X, et al: PDGF-D is a specific, protease-activated ligand for the PDGF beta-receptor. Nat Cell Biol 3:512-516, 2001
4. LaRochelle WJ, Jeffers M, McDonald WF, et al: PDGF-D, a new protease-activated growth factor. Nat Cell Biol 3:517-521, 2001
5. Leu KM, Thomas DG, Baker LH: Platelet-derived growth factor receptor (PDGF-R) a and expression in human malignancy. Proc American Soc Clin Oncol 21:1774, 2002
6. Dabrow MB, Francesco MR, McBrearty FX, et al: The effects of platelet-derived growth factor and receptor on normal and neoplastic human ovarian surface epithelium. Gynecol Oncol 71:29-37, 1998
7. Lindmark G, Sundberg C, Glimelius B, et al: Stromal expression of platelet-derived growth factor beta-receptor and platelet-derived growth factor B-chain in colorectal cancer. Lab Invest 69:682-689, 1993
8. Craven RJ, Xu LH, Weiner TM, et al: Receptor tyrosine kinases expressed in metastatic colon cancer. Int J Cancer 60:791-797, 1995
9. Bergers G, Song S, Meyer-Morse N, et al: Benefits of targeting both pericytes and endothelial cells in the tumor vasculature with kinase inhibitors. J Clin Invest 111:1287-1295, 2003
10. Hellstrom M, Kalen M, Lindahl P, et al: Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development 126:3047-3055, 1999
11. Grinnell F, Ho CH, Lin YC, et al: Differences in the regulation of fibroblast contraction of floating versus stressed collagen matrices. J Biol Chem 274:918-923, 1999
12. Rodt SA, Ahlen K, Berg A, et al: A novel physiological function for platelet-derived growth factor-BB in rat dermis. J Physiol 495:193-200, 1996
13. Pietras K, Ostman A, Sjoquist M, et al: Inhibition of platelet-derived growth factor receptors reduces interstitial hypertension and increases transcapillary transport in tumors. Cancer Res 61:2929-2934, 2001
14. Pietras K, Rubin K, Sjoblom T, et al: Inhibition of PDGF receptor signaling in tumor stroma enhances antitumor effect of chemotherapy. Cancer Res 62:5476-5484, 2002
15. Pietras K, Stumm M, Hubert M, et al: STI571 enhances the therapeutic index of epothilone B by a tumor-selective increase of drug uptake. Clin Cancer Res 9:3779-3787, 2003
16. Chapman A and King D: Divalent antibody fragments. WO99/64460, 1999
17. Hart CE, Kraiss LW, Vergel S, et al: PDGFbeta receptor blockade inhibits intimal hyperplasia in the baboon. Circulation 99:564-569, 1999
18. Davies MG, Owens EL, Mason DP, et al: Effect of platelet-derived growth factor receptor-alpha and -beta blockade on flow-induced neointimal formation in endothelialized baboon vascular grafts. Circ Res 86:779-786, 2000
19. Serruys PW, Heyndrickx GR, Patel J, et al: Effect of an anti-PDGF b receptor blocking antibody on restenosis in patients undergoing elective stent placement. Int J Cardiovasc Intervent 5:214-22, 2003
20. Adair JR, Athwal DS, Emtage JS: Humanised antibodies. US Patent Application 20040076627, 1991
21. Chapman AP, Antoniw P, Spitali M, et al: Therapeutic antibody fragments with prolonged in vivo half-lives. Nat Biotechnol 17:780-783, 1999
22. Jackson A, Jayson GC, Li KL, et al: Reproducibility of quantitative dynamic contrast-enhanced MRI in newly presenting glioma. Br J Radiol 76:153-162, 2003
23. Jackson A, Haroon H, Zhu XP, et al: Breath hold perfusion and permeability mapping of hepatic malignancies using magnetic resonance imaging and a first pass leakage profile model. NMR Biomed 15:164-173, 2002
24. Therasse P, Arbuck SG, Eisenhauer EA, et al: New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92:205-216, 2000
25. Haase A: Snapshot FLASH MRI. Applications to T1, T2, and chemical-shift imaging. Magn Reson Med 13:77-89, 1990
26. Evelhoch JL: Key factors in the acquisition of contrast kinetic data for oncology. J Magn Reson Imaging 10:254-259, 1999
27. Tofts PS, Brix G, Buckley DL, et al: Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a diffusable tracer: Standardized quantities and symbols. J Magn Reson Imaging 10:223-232, 1999
28. Tofts PS, Kermode AG: Measurement of the blood-brain barrier permeability and leakage space using dynamic MR imaging. 1. Fundamental concepts. Magn Reson Med 17:357-367, 1991
29. Daldrup HE, Shames DM, Husseini W, et al: Quantification of the extraction fraction for gadopentetate across breast cancer capillaries. Magn Reson Med 40:537-543, 1998
30. Parker GJ, Jackson A, Waterton JC: Automated arterial input function extraction for T1-weighted DCE-MRI. Proceedings of the International Society for Magnetic Resonance in Medicine 11:2003 (abstr 1264)
31. St Lawrence KS, Lee TY: An adiabatic approximation to the tissue homogeneity model for water exchange in the brain: I. Theoretical derivation. J Cereb Blood Flow Metab 18:1365-1377, 1998
32. Parker GJM, Padhani AR: T1-weighted dynamic contrast enhanced MRI, in Tofts PS (ed): Quantitative MRI of the brain—Measuring changes caused by disease. Chichester, John Wiley and Sons Ltd, 2003, pp 341-364
33. Griffon-Etienne G, Boucher Y, Brekken C, et al: Taxane-induced apoptosis decompresses blood vessels and lowers interstitial fluid pressure in solid tumors: clinical implications. Cancer Res 59:3776-3782, 1999
34. Abramsson A, Lindblom P, Betsholtz C: Endothelial and nonendothelial sources of PDGF-B regulate pericyte recruitment and influence vascular pattern formation in tumors. J Clin Invest 112:1142-1151, 2003
35. Savage DG, Antman KH: Imatinib mesylate–A new oral targeted therapy. N Engl J Med 346:683-693, 2002
36. Hussain M, Kotz H, Minasian L: Occurrence of ascites secondary to STI571 in ovarian cancer patients. Proc Am Soc Clin Oncol 22:220, 2003 (abstr 880)
37. Ko YJ, Small EJ, Kabbinavar F, et al: A multi-institutional phase II study of SU101, a platelet-derived growth factor receptor inhibitor, for patients with hormone-refractory prostate cancer. Clin Cancer Res 7:800-805, 2001
38. Eckhardt SG, Rizzo J, Sweeney KR, et al: Phase I and pharmacologic study of the tyrosine kinase inhibitor SU101 in patients with advanced solid tumors. J Clin Oncol 17:1095-1104, 1999
39. Britten CD, Rosen LS, Kabbinavar F: Phase I trial of SU6668, a small molecule receptor tyrosine kinase inhibitor, given twice daily in patients with advanced cancers. Proc Am Soc Clin Oncol 21:28b, 2002 (abstr 1922)
40. Kuenen B, Ruijter R, Hoekman K: Dose finding study of SU6668 given thrice daily by oral route under fed conditions in patients with advanced malignancies. Proc Am Soc Clin Oncol 20:110a, 2001 (abstr 437)
41. Brahmer JR, Kelsey S, Scigalla P, et al: A phase I study of SU6668 in patients with refractory solid tumors. Proc Am Soc Clin Oncol 21:84a, 2002 (abstr 335)
42. Ueda Y, Shimoyama T, Murakami H, et al: Phase I study of TSU-68, VEGF receptor tyrosine kinase inhibitor, by twice daily oral administration between meals in patients with advanced solid tumors. Proc Am Soc Clin Oncol 21:111a, 2002 (abstr 443)
43. Murakami H, Yamamoto N, Shimoyama T, et al: Phase I, pharmacokinetic, and biological studies of TSU-68, the oral vascular endothelial growth factor receptor tyrosine kinase inhibitor, administered after meals in patients with advanced solid tumors. Proc Am Soc Clin Oncol 22:217, 2003 (abstr 870)
44. Raymond E, Faivre S, Vera K, et al: Final results of a phase I and pharmacokinetic study of SU11248, a novel multi-target tyrosine kinase inhibitor, in patients with advanced cancers. Proc Am Soc Clin Oncol 22:192, 2003 (abstr 769)
45. Heuchel R, Berg A, Tallquist M, et al: Platelet-derived growth factor beta receptor regulates interstitial fluid homeostasis through phosphatidylinositol-3' kinase signaling. Proc Natl Acad Sci U S A 96:11410-11415, 1999
46. Willett CG, Boucher Y, di Tomaso E, et al: Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 10:145-147, 2004(G.C. Jayson, G.J.M. Parke)