- Email: [email protected]
Enhancement of radiation therapy by the novel vascular targeting agent ZD6126
Int. J. Radiation Oncology Biol. Phys., Vol. 53, No. 1, pp. 164 –171, 2002 Copyright © 2002 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/02/$–see front matter
PII S0360-3016(02)02742-6
BIOLOGY CONTRIBUTION
ENHANCEMENT OF RADIATION THERAPY BY THE NOVEL VASCULAR TARGETING AGENT ZD6126 DIETMAR W. SIEMANN, PH.D.,*
AND
AMYN M. ROJIANI, M.D., PH.D.†
*Department of Radiation Oncology, Shands Cancer Center, University of Florida, Gainesville, FL; †Departments of Interdisciplinary Oncology and Pathology, H. Lee Moffitt Cancer Center and Research Institute, University of South Florida, Tampa, FL Purpose: The aim of this study was to evaluate the antitumor efficacy of the novel vascular targeting agent ZD6126 (N-acetylcochinol-O-phosphate) in the rodent KHT sarcoma model, either alone or in combination with single- or fractionated-dose radiation therapy. Methods: C3H/HeJ mice bearing i.m. KHT tumors were injected i.p. with ZD6126 doses ranging from 10 to 150 mg/kg. Tumors were irradiated locally in unanesthetized mice using a linear accelerator. Tumor response to ZD6126 administered alone or in combination with radiation was assessed by clonogenic cell survival assay or tumor growth delay. Results: Treatment with ZD6126 led to a rapid tumor vascular shutdown as determined by Hoechst 33342 diffusion. Histologic evaluation showed morphologic damage of tumor cells within a few hours after drug exposure, followed by extensive central tumor necrosis and neoplastic cell death as a result of prolonged ischemia. When combined with radiation, a 150 mg/kg dose of ZD6126 reduced tumor cell survival 10 –500-fold compared with radiation alone. These enhancements in tumor cell killing could be achieved for ZD6126 given both before and after radiation exposure. Further, the shape of the cell survival curve observed after the combination therapy suggested that including ZD6126 in the treatment had a major effect on the radiation-resistant hypoxic cell subpopulation associated with this tumor. Finally, when given on a once-weekly basis in conjunction with fractionated radiotherapy, ZD6126 treatment was found to significantly increase the tumor response to daily 2.5 Gy fractions. Conclusion: The present results demonstrated that in the KHT sarcoma, ZD6126 caused rapid tumor vascular shutdown, induction of central tumor necrosis, tumor cell death secondary to ischemia, and enhancement of the antitumor effects of radiation therapy. © 2002 Elsevier Science Inc. ZD6126, KHT sarcoma, Vascular targeting.
The initiation and maintenance of a supportive vasculature is critical to both the growth and survival of a solid tumor mass (1, 2). Indeed, in the absence of neovascularization, tumor cells remain dormant, and solid tumors fail to progress beyond a microscopic size (1, 2). Although the process of angiogenesis also plays a significant role in many normal and developmental as well as pathologic conditions, angiogenesis in adults is usually limited to specific reproductive organs (3), and the turnover of vascular endothelial cells typically is measured in months and years (4). Unlike the well-defined microvascular architecture of normal tissues, the abnormal nature of the tumor microcirculation is well documented (5, 6). A key characteristic of the tumor microcirculation is that tumor-associated blood vessels contain populations of endothelial cells that are actively divid-
ing in response to angiogenic factors (4, 7). Because the nutritional support of large numbers of neoplastic cells depends on small numbers of endothelial cells, damaging the tumor endothelium could have a marked impact on tumor cell survival and growth (8, 9). The expanding tumor endothelium therefore represents a key target for cancer therapy (2, 9). Unlike anti-angiogenic strategies that aim to inhibit new vessel formation (2), vascular targeting aims to cause selective and direct damage to existing tumor endothelium, resulting in a rapid and catastrophic shutdown of the tumor vascular network, leading to extensive secondary tumor cell death (8, 9). Several agents that elicit irreversible vascular shutdown selectively within solid tumors have been identified. These include the tubulin-binding agents, most notably colchicine, the vinca alkaloids, and the combretastatins
Reprint requests to: Dietmar W. Siemann, Ph.D., Department of Radiation Oncology, University of Florida Shands Cancer Center, 2000 SW Archer Road, Box 100385, Gainesville, Florida 32610. Tel: (352) 265-0287; Fax: (352) 265-0759; E-mail: siemadw@ ufl.edu Supported in part by a grant from AstraZeneca Inc. and the U.S.
National Cancer Institute (PHS grant CA 84408). Acknowledgments—The authors thank Sharon Lepler, Emma Mercer, Heather Newlin, Chris Pampo, and Jeffrey Lin for providing excellent technical assistance. Received Sep 6, 2001, and in revised form Dec 27, 2001. Accepted for publication Jan 7, 2002.
INTRODUCTION
164
Combinations of radiation and ZD6126
(10 –16). Although most tubulin-binding agents have demonstrated antivascular effects only when near-maximally tolerated doses are administered (11, 15), there are some exceptions. For example, combretastatin A4 disodium phosphate (CA4DP) and its derivative AVE8062 have been shown to lead to rapid vascular shutdown, reductions in tumor blood flow, and induction of necrosis in a variety of preclinical tumor models at minimally toxic doses in vivo (12, 13, 17–20). The principal mechanism of action responsible for these in situ antivascular effects observed after treatment with these agents is believed to be damage mediated through effects on proliferating endothelial cell microtubule polymerization leading to endothelial cell shape changes (14, 15, 17, 21). Recently, ZD6126 has been reported to have selective tumor vascular targeting activity (22, 23), to cause disruption of the cytoskeleton of proliferating endothelial cells, and to induce tumor necrosis in preclinical tumor models at well-tolerated doses (22, 23). After treatment with vascular targeting agents such as CA4DP and ZD6126, viable tumor cells are found at the tumor periphery, presumably because they are supplied by normal vessels not susceptible to damage by these agents (16). Because this residual tumor tissue is likely to be well oxygenated, and because the future application of vascular targeting strategies will probably include their utilization in conjunction with conventional anticancer therapies, the present investigations were undertaken to examine the efficacy of combining 2D6126 with ionizing radiation. METHODS AND MATERIALS Animal and tumor model KHT sarcoma cells (24) were injected i.m. (2 ⫻ 105 cells in a volume of 0.01 mL) into the hind limbs of 6 – 8-weekold female C3H/HeJ mice (Jackson Laboratories, Bar Harbor, ME). At the start of a given experiment, mice bearing tumors of a size equivalent to a weight of ⬃0.5 g (clonogenic cell survival assay) or ⬃0.2 g (regrowth delay assay) were selected from a large number of mice inoculated with the same cell numbers on the same day and treated with either ZD6126 alone or ZD6126 plus radiation, or kept as untreated controls. Drug treatment ZD6126 (AstraZeneca, Macclesfield, UK) was dissolved in 0.9% saline and injected i.p. into a volume of 0.01 mL/g body weight. Although a maximum tolerated dose was not established, single doses as large as 400 mg/kg ZD6126 have been given to C3H mice without overt toxicity. Irradiation Irradiations were performed on nonanesthetized mice using a 6-MeV linear accelerator operating at a dose rate of 4 Gy/min. Each mouse was confined to a plastic jig with its tumor-bearing leg extended through an opening on the side to allow the tumor to be irradiated locally. Five tumorbearing mice were irradiated simultaneously.
● D. W. SIEMANN AND A. M. ROJIANI
165
Morphologic and morphometric analysis The tumor was dissected from the hind limb, and adjacent connective tissue was removed. After fixation in 10% neutral buffered formalin, the tumor was inked around its entire surface to define borders and serially divided at 1-mm intervals. Alternate slices were submitted for routine histology processing. Sections were dehydrated through a series of graded alcohols, processed in xylene, and embedded in paraffin. Four m sections obtained from these paraffin blocks were stained with hematoxylin and eosin as well as Masson’s trichrome stain. Each hematoxylin and eosin– stained tumor section was divided into 4 – 8 grids using a fine permanent marker. Images were captured with a framegrabber, and sections were morphometrically assessed with the Image Pro Plus image analysis system (Media Cybernetics, Silver Spring, MD). Using an “irregular area of interest” tool, areas of necrosis were outlined and measured on each grid. Multiple grids for each tumor were combined and compared with the total area of the tumor, results being expressed as percent necrosis. Hoechst-33342 studies Hoechst-33342 (bisbenzimide, Sigma) solution was made up in 0.9% sterile saline immediately before use. KHT sarcoma– bearing mice were either untreated or treated with 150 mg/kg ZD6126. Hoechst-33342 then was administered at 40 mg/kg i.v. (volume 5 mL/kg) at various times after ZD6126 injection (25). One minute after Hoechst-33342 injection, the mice were killed, and the tumors were resected and immediately immersed in liquid nitrogen for subsequent frozen sectioning. For each tumor sample, 10 m cryostat sections were cut at three different levels between one pole and the equatorial plane. The sections were air dried and studied under ultraviolet illumination using a fluorescent microscope. Blood vessel outlines were identified by the surrounding halo of fluorescent H33342labeled cells. Vessel counts were performed using a Chalkley point array for random sample analysis (26). Briefly, each section was viewed at 10⫻ objective magnification. A 25-point Chalkley grid was positioned randomly over field of view. Any points falling within haloes of fluorescent cells were scored positive. Twenty random fields were counted per section, and a minimum of six sections per tumor were examined. Data from 3–5 tumors were pooled and presented. Clonogenic cell survival Clonogenic cell survival in treated or untreated tumors was determined using an in vivo to in vitro cell survival assay as previously described (24). Briefly, 24 h after treatment, KHT sarcomas were excised, and single-cell suspensions were prepared using a combined mechanical and enzymatic dissociation procedure. The cells were counted, and various dilutions were prepared. KHT cells were mixed with 0.2% agar containing ␣-minimum essential medium supplemented with 10% fetal bovine serum and plated into 24-well plates. Two weeks later, the plates were harvested,
166
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 1, 2002
Fig. 1. Vessel counts in KHT sarcoma– bearing mice as a function of time after treatment with a 150 mg/kg dose of ZD6126. At a given time after ZD6126 treatment, mice were injected with Hoechst 33342 (40 mg/kg) to identify patent vessels, and tumors were removed 1 min later. Counting was performed using a Chalkley point array for random sample analysis. Data are the mean ⫾ SE of 3–5 tumors.
and the resulting colonies were counted with a dissecting microscope. Tumor-surviving fractions were determined by multiplying the calculated fraction of surviving cells by the ratio of cells recovered in treated vs. untreated tumors. Tumor growth delay In the tumor regrowth assay, the tumor size was measured 3–5 times per week after treatment by passing the tumorbearing leg through a series of increasing-diameter holes in a plastic rod. The smallest-diameter hole the tumor-bearing leg would pass through was recorded and converted to a tumor weight using a calibration curve (27). The time required for the untreated or treated tumors to grow to five times the starting size was recorded. The median time for the tumors in each group of mice to reach this end point was calculated. Confidence intervals about the median were determined by nonparametric statistics (28). The median tumor responses were plotted, and significance between treatment groups was determined by a Wilcoxon rank-sum test (␣ ⫽ 0.025). RESULTS To examine the effects of ZD6126 treatment on the vasculature of KHT sarcomas, functional vascular volume measurements were carried out based on the use of the perivascular stain Hoechst 33342. The results showed that the vascular-damaging effects of this agent began to manifest themselves within 30 min after the administration of a single 150 mg/kg dose (Fig. 1). By 2 h after treatment, KHT
sarcomas showed functional vessels only near the periphery, suggesting that this dose of ZD6126 caused an almost complete vascular shutdown in the tumors. Histologic assessment reveals the tumors to be hypercellular, with significant nuclear atypia and mitotic activity. The tumor, although forming well-defined mass, could frequently be seen infiltrating and entrapping normal skeletal muscle (Fig. 2a). Tumors treated with ZD6126 displayed clear morphologic evidence of damage within a few hours after treatment. This was subsequently followed by the rapid onset of frank necrosis such that by 24 h after a 150 mg/kg dose of ZD6126, KHT sarcomas showed extensive central necrosis with viable tumor cells evident only at the periphery of the tumor adjacent to the surrounding normal tissues (Figs. 2b– d). The tissue necrosis involves both areas of tumor proliferation, as well as areas of tumor infiltrating skeletal muscle (Figs. 2b and 2c). This pattern of necrosis is well illustrated at higher magnification, where tissue destruction and nuclear fragmentation is readily evident in the entire central portion of the tumors, extending to within a few cell layers from the peripheral margin of the tumor (Fig. 2d). To quantify the extent of necrosis produced by ZD6126 treatment, sections from KHT sarcomas removed 24 h after treatment were assessed using an image analysis system. The results (Fig. 3, left-hand panel) showed that compared to the ⬃10% necrosis seen in untreated tumors, treatment with 150 mg/kg ZD6126 increased the extent of necrosis to ⬃90%. Antitumor effectiveness of this dose of ZD6126 was determined by measuring clonogenic cell survival. The data
Combinations of radiation and ZD6126
Fig. 2. Standard hematoxylin and eosin staining of 4-m sections from KHT sarcomas. Untreated KHT sarcomas show tumor cells infiltrating and entrapping host skeletal muscle fibers (arrows). [(a) Original magnification: 200⫻.] Only minute areas of necrosis are seen in these tumors. Sections from mice that had received a 150-mg/kg dose of ZD6126 24 h before assessment are shown in b– d. [(b– d) Original magnification: 40⫻, 200⫻, and 400⫻,] After ZD6126 treatment, KHT tumors show extensive central areas of necrosis with hemorrhage readily apparent at low power. A small rim of viable tumor can be readily identified at the periphery (arrowheads), and necrotic areas included regions infiltrating skeletal muscle (arrows), best seen in (b) and (c). This viable tumor rim consists of 6 to 8 layers of tumor cells resting against the unaffected host tissue. At high magnification (d), the areas of tumor necrosis are easily separated from the hyperchromatic, pleomorphic, mitotically active tumor cells.
showed that this treatment led to ⬃1–1.5 logs of tumor cell kill. Subsequent evaluations of KHT sarcomas treated with various doses of this agent demonstrated that increasing doses of ZD6126 resulted in a dose-dependent increase in tumor cell kill (Fig. 3, right-hand panel). In particular, doses ranging from 50 to 150 mg/kg resulted in a loss of viable cells from the tumors. Because the timing between agents is an important parameter in combined modality therapies, initial experiments were performed to evaluate the effect on tumor response of various time intervals between the vascular targeting agent and radiation. Figure 4 shows the effect of administering a 150 mg/kg dose of ZD6126 at various times before or after a 10 Gy dose of radiation on clonogenic cell survival in the KHT sarcoma. The data indicate that administering ZD6126
● D. W. SIEMANN AND A. M. ROJIANI
167
24 h before, immediately after, and up to 4 h after radiotherapy enhanced tumor cell killing compared to that achieved with radiation alone. However, giving ZD6126 shortly before (1 h) the 10 Gy dose failed to reduce tumor cell survival. The greatest increase in tumor cell killing resulted from a treatment strategy in which the vascular targeting agent was given 0.5–1 h after radiation. On the basis of the timing studies described above, a 1-h separation between the radiation and vascular targeting agent treatments was selected and applied in subsequent investigations. Using this treatment sequence, KHT sarcoma– bearing mice were irradiated with doses ranging from 0 to 25 Gy and then given a 150 mg/kg dose of ZD6126 (Fig. 5). When clonogenic cell survival was assessed 24 h later, the extent of tumor cell killing in mice treated with the combination of agents was increased significantly compared to that seen for radiation alone (Fig. 5, closed vs. open symbols). Not only did the addition of ZD6126 to the treatment protocol reduce tumor cell survival 10 –500-fold below that for radiation alone, but it also seemed to reduce the proportion of radiobiologic hypoxic tumor cells that dominate cell survival in this tumor at high doses (⬎10 Gy) and give rise to the characteristic “break” in the cell survival curve (24). This observation was similar to one previously made in our laboratories when combining single-dose radiotherapy with the vascular targeting agent CA4DP (13, 29). Because clinical radiotherapy typically is given using fractionated treatment protocols, additional experiments combining daily 2.5 Gy fractions with once-weekly ZD6126 (150 mg/kg) were performed. In these experiments, radiation was given for 2 weeks (Monday through Friday) with the vascular targeting agent being administered 1 h after each of the two Friday radiation fractions. Figure 6 shows the growth patterns of the median tumors of groups of mice treated with radiotherapy alone or radiotherapy plus ZD6126; Fig. 7 demonstrates the range of responses observed. In this protocol, ZD6126 alone failed to retard KHT sarcoma growth. However, it should be noted that only a single dose of ZD6126 (the first treatment) was given, because, as a result of rapid growth of the KHT sarcoma, the mice had to be killed before the second drug treatment could be given at the end of the second week. Although ZD6126 was not effective on its own as a single dose, a significant enhancement of the radiation response (p ⬍ 0.05) was achieved when two doses of ZD6126 were added to the radiotherapy treatment (Figs. 6 and 7). DISCUSSION There has recently been a great deal of rekindled interest in the application of agents that directly damage the tumor vasculature, a notion initially advanced nearly 20 years ago by Denekamp (8, 9). This resurgence has come about primarily because of the development of new agents showing a high degree of preferential toxicity to the growing tumor endothelium. Tubulin-binding agents have long been known
168
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 1, 2002
Fig. 3. (Left) Percent necrosis in KHT sarcomas after a 150 mg/kg dose of ZD6126. (Right) Tumor cell kill in KHT sarcomas treated with increasing doses of ZD6126. Data were determined 24 h after drug treatment and are the mean ⫾ SE of 3–5 (necrosis) or 3–12 (cell survival) tumors.
to possess some degree of antivascular action (11). However, the more recent identification of a select group of agents that, at doses well below the maximum tolerated dose, demonstrate a profound impact on tumor vasculature (12, 17, 22, 23) has generated substantial enthusiasm for the potential of such an anticancer treatment strategy (15, 16). Although the mechanisms governing these tumor-selective effects of tubulin-binding agents have not been unequivocally demonstrated, recent evidence supports the notion that the inhibition of tubulin polymerization by such agents affects endothelial cell shape, leading to thrombus formation and a consequent secondary cascade of ischemic tumor cell death (21). In the present study, fluorescent microscopy was used to visualize Hoechst 33342 distributions on tumor sections to identify tumor blood vessels actively involved in perfusion (25). The application of this vascular marker demonstrated a significant reduction in patent vessels that was detectable within 30 min after ZD6126 treatment and lasted for at least a 24-h period. This rapid and extensive shutdown in vascular function observed in the KHT sarcoma is typical of the previously reported reductions in blood perfusion associated with the vascular targeting agent CA4DP in this (13, 20) and other (12, 14, 15, 17, 18) tumor models. The suppression of functional vessel number by ZD6126 also is consistent with recent MRI evaluations of this agent’s action on tumor vasculature (30, 31). The consequences of the pathophysiologic effects of ZD6126 were clearly apparent in subsequent histologic analyses of KHT sarcomas (Fig. 2). Evidence of morphologic damage became detectable within 12 to 18 h after a
Fig. 4. Effect on KHT sarcoma cell survival of administering a 150 mg/kg dose of ZD6126 at various times before or after a 10 Gy dose of radiation. The solid square and the open circle represent the cell survival after ZD6126 and radiation alone, respectively. Data are the mean ⫾ SE of 6 –12 tumors.
Combinations of radiation and ZD6126
Fig. 5. Tumor cell survival in KHT sarcomas treated with a 150 mg/kg dose of ZD6126 1 h after a range of single doses of radiation. Tumors were evaluated 24 h after completing the treatment. Results are the mean ⫾ SE of 3– 6 experiments.
● D. W. SIEMANN AND A. M. ROJIANI
169
150 mg/kg treatment, and by 24 h, extensive hemorrhagic necrosis had occurred. The surviving viable rim of tumor cells (Figs. 2b– d) is a characteristic feature of tumors treated with antivascular pharmaceuticals (12–14, 18, 19, 22, 23) as well as vessel-targeting antibodies (32). Although no direct evidence supporting this currently exists, the failure to eliminate neoplastic cells at the periphery of the tumor is believed to reflect a lack of damage to vessels in the normal tissues adjacent to the tumor, which provide adequate nutrition and waste product removal and allow the tumor cells to survive and proliferate. Although the destruction of large areas, particularly in the central and typically most treatment-resistant regions of the tumors, is highly desirable, the survival of cells at the tumor periphery remains a concern. To attempt to address this issue, preclinical assessments of the efficacy of vascular targeting agents delivered in multidose regimens have been initiated. Results with CA4DP indicate that the response in some tumor models can be enhanced by repeated drug doses (15, 29). Similarly, significant growth delays have been observed when multiple doses of ZD6126 were administered (23). An alternate strategy to eliminate the cells surviving the vascular targeting agent and hence to ultimately control the tumor is to combine this treatment with a therapy directly targeting these cells (16). This approach has proven effective when combining CA4DP with conventional anticancer drugs (15, 29, 33, 34) and radiation (13, 20, 29). In the present study, we demonstrated that ZD6126 could effec-
Fig. 6. KHT sarcoma weight as a function of time after (open circle) no treatment, (filled circle) ZD6126 alone, (open square) 10 2.5-Gy fractions administered in 12 days, or (filled square) the combination of the fractionated radiotherapy protocol plus two 150 mg/kg doses of ZD6126, given 1 h after the 5th and 12th 2.5-Gy fraction, respectively. Data shown represent the responses of the median tumors of groups of 8 –10 mice.
170
I. J. Radiation Oncology
● Biology ● Physics
Fig. 7. Effect on tumor growth of including a 150 mg/kg dose of ZD6126 in fractionated radiotherapy treatment (drug and radiation protocol as described in Fig. 6). Data shown are the median times required to grow to 5⫻ the starting size ⫾ 96% confidence intervals in groups of 8 –10 mice (28).
tively enhance the radiation response of the KHT sarcoma to both single- and fractionated-dose radiotherapy (Figs. 4 –7). An important consideration in such investigations is the issue of sequencing and timing (Fig. 4). When the ZD6126 was injected 1 h before irradiating the tumors, the resultant cell survival was not significantly different from that observed for radiation alone. The lack of improvement in the treatment response may indicate that when the vascular targeting agent is given shortly before radiotherapy, some parts of the tumors experience transient reductions in blood flow sufficient to render cells hypoxic, but for a time insufficiently long for them to undergo ischemic death. However, all other sequence schedules investigated, including the administration of ZD6126 from 0 to 4 h after or 24 h before irradiation, led to significantly greater tumor cell killing with the combination treatment. Indeed, when a complete dose–response curve was established, administering a 150
Volume 53, Number 1, 2002
mg/kg dose of ZD6126 1 h after a range of radiation doses resulted in improved tumor cell killing at all doses investigated (Fig. 5). The displacement of the “tail” of the cell survival curve to lower survival values was of particular interest, because it suggests that the inclusion of ZD6126 in the treatment strategy had reduced the proportion of radiobiologically hypoxic cells in the tumor. A similar conclusion was reached in our previous evaluations of CA4DP (13, 29). It also is consistent with the present histologic evaluations demonstrating induction of extensive central necrosis in the tumors, leaving viable cells only at the periphery (Fig. 2). Taken together, these findings suggest that administering ZD6126 post-radiotherapy may provide the most appropriate scheme for combining these therapies. Because it is almost certain that any clinical combination therapy will have to involve fractionated radiotherapy, we also chose to examine whether including ZD6126 in a protocol of fractionated radiation exposures could improve the tumor response. Radiation was given Monday through Friday for 2 weeks at 2.5 Gy per fraction, and ZD6126 (150 mg/kg) was administered each Friday, 1 h after that day’s radiation dose. The results shown in Figs. 6 and 7 indicate that, while this radiation schedule induces on its own a tumor growth delay of 6 days, this delay is significantly increased (p ⬍ 0.05) to 12 days by the two ZD6126 doses added to the treatment regimen. In conclusion, ZD6126 was found in these studies to demonstrate striking antivascular effects in tumors, leading to the induction of necrosis and a consequential rapid loss of clonogenic neoplastic cells from the cell population. In practical terms, this agent may be acting to chemotherapeutically reduce the viable tumor mass. In light of histologic evidence that neoplastic cells adjacent to peripheral normal tissues may survive antivascular agent treatment because of their close proximity to normal blood vessels, it would seem that radiotherapy or chemotherapy may need to be administered to destroy the remaining rim of tumor cells. The present findings support this notion and clearly demonstrate that the addition of ZD6126 can significantly enhance the antitumor efficacy of both single- and fractionated-dose radiation therapy. While the response of tumors to the ZD6126-radiation combination would seem to be an additive effect of two therapies, complete dose–response analysis of both modalities would be required to prove this unequivocally. Although normal tissue toxicities were not measured in these experiments, the observation that such tumor response enhancements could occur at nontoxic doses (22, 23) is encouraging and suggests that a therapeutic benefit may be achievable. Further evaluations of the therapeutic potential of this agent used alone or in an adjuvant setting clearly seem warranted.
REFERENCES 1. Folkman J, Shing Y. Angiogenesis. J Biol Chem 1992;267: 10931–10934. 2. Folkman J. Seminars in medicine of the Beth Israel Hospital,
Boston. Clinical applications of research on angiogenesis. N Engl J Med 1995;333:1757–1763. 3. Winlaw DS. Angiogenesis in the pathobiology and treatment
Combinations of radiation and ZD6126
4. 5. 6. 7. 8. 9. 10.
11. 12. 13.
14. 15. 16. 17. 18.
19.
of vascular and malignant diseases. Ann Thorac Surg 1997; 64:1204 –1211. Hobson B, Denekamp J. Endothelial proliferation in tumors and normal tissue: Continuous labeling studies. Br J Cancer 1984;49:405– 413. Vaupel P, Kallinowski F, Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: A review. Cancer Res 1989;49:6449 – 6465. Konerding MA, Miodonski AJ, Lametschwandtner A. Microvascular corrosion casting in the study of tumor vascularity: A review. Scanning Microsc 1995;9:1233–1244. Gastl G, Hermann T, Steurer M, et al. Angiogenesis as a target for tumor treatment. Oncology 1997;54:177–184. Denekamp J. Vasculature as a target for tumour therapy. Prog Appl Microcirc 1984;4:28 –38. Denekamp J. Angiogenesis, neovascular proliferation and vascular pathophysiology as targets for cancer therapy. Br J Radiol 1993;66:181–196. Baguley BC, Holdaway KH, Thomsen LL, et al. Inhibition of growth of colon 38 adenocarcinoma by vinblastine and colchicine. Evidence for a vascular mechanism. Eur J Cancer 1991;27:482– 487. Hill SA, Sampson LE, Chaplin DJ. Anti-vascular approaches to solid tumor therapy: Evaluation of vinblastine and flavone acetic acid. Int J Cancer 1995;63:119 –123. Dark GG, Hill SA, Prise VE, et al. Combretastatin A-4, an agent that displays potent and selective toxicity toward tumor vasculature. Cancer Res 1997;57:1829 –1834. Li L, Rojiani A, Siemann DW. Targeting the tumor vasculature with combretastatin A-4 disodium phosphate: Effects on radiation therapy. Int J Radiat Oncol Biol Phys 1998;42:899 – 903. Grosios K, Holwell SE, McGown AT, et al. In vivo and in vitro evaluation of combretastatin A-4 and its sodium phosphate prodrug. Br J Cancer 1999;81:1318 –1327. Chaplin DJ, Pettit GR, Hill SA. Anti-vascular approaches to solid tumour therapy: Evaluation of combretastatin A4 phosphate. Anticancer Res 1999;19:189 –195. Siemann DW, Warrington KH, Horsman MR. Targeting tumor blood vessels: An adjuvant strategy for radiation therapy. Radiother Oncol 2000;57:5–12. Tozer GM, Prise VE, Wilson J, et al. Combretastatin A-4 phosphate as a tumor vascular-targeting agent: Early effects in tumors and normal tissues. Cancer Res 1999;59:1626 –1634. Horsman MR, Ehrnrooth E, Ladekarl M, et al. The effect of combretastatin A-4 disodium phosphate in a C3H mouse mammary carcinoma and a variety of murine spontaneous tumours. Int J Radiat Oncol Biol Phys 1998;42:895– 898. Hori K, Saito S, Nihei Y, et al. Antitumor effects due to irreversible stoppage of tumor tissue blood flow: Evaluation of a novel combretastatin A-4 derivative, AC7700. Jpn J Cancer Res 1999;90:1026 –1038.
● D. W. SIEMANN AND A. M. ROJIANI
171
20. Murata R, Siemann DW, Overgaard J, Horsman MR. Interaction between combretastatin A4 disodium phosphate and radiation in murine tumours. Radiother Oncol. 2001;60:155– 161. 21. Galbraith SM, Chaplin DJ, Lee F, et al. Effects of combretastatin A4 phosphate on endothelial cell morphology in vitro and relationship to tumour vascular targeting activity in vivo. Anticancer Res 2001;21:93–102. 22. Davis PD, Hill SA, Galbraith SM, et al. ZD6126: A new agent causing selective damage of tumor vasculature (Abstr). Proc Am Assoc Cancer Res 2000;41:329. 23. Blakey DC, Curry B, Westwood FR, Ashton SE. Vascular targeting activity of ZD6126 in two tumour models growing in athymic rats (Abstr). Proc Am Assoc Cancer Res 2001;42:823. 24. Siemann DW, Johansen IM, Horsman MR. Radiobiological hypoxia in the KHT sarcoma: Predictions using the Eppendorf histograph. Int J Radiat Oncol Biol Phys 1998;40:1171–1176. 25. Smith KA, Hill SA, Begg AC, et al. Validation of the fluorescent dye Hoechst 33342 as a vascular space marker in tumours. Br J Cancer 1988;57:247–253. 26. Curtis ACS. Area and volume measurements by random sampling methods. Med Biol Illustra 1960;10:261–266. 27. Siemann DW, Sutherland RM. In vivo tumor response to single and multiple exposures of adriamycin. Eur J Cancer 1980;16:1433–1440. 28. Noether J. Introduction to statistics: A fresh approach. Boston: Houghton Mifflin; 1971. p. 86 –112. 29. Li L, Rojiani AM, Siemann DW. Preclinical evaluations of therapies combining the vascular targeting agent combretastatin A-4 disodium phosphate and conventional anticancer therapies in the treatment of Kaposi’s sarcoma. Acta Oncol 2002, in press. 30. Evelhoch JL, He Z, Polin L, et al. MRI evaluation of the effects of ZD6126 on tumor vasculature (Abstr). Proc Am Assoc Cancer Res 2001;42:107. 31. Robinson SP, McIntyre DJ, Howe FA, Griffiths JR, Blakey DC, Waterton JC. Assessment of the novel tumor vascular targeting agent ZD6126 by non-invasive multi-gradient echo MRI (Abstr). Proc Am Assoc Cancer Res 2001;42:415. 32. Huang X, Molema G, King S, et al. Tumor infarction in mice by antibody-directed targeting of tissue factor to tumor vasculature. Science 1997;275:547–550. 33. Grosios K, Loadman PM, Swaine DJ, et al. Combination chemotherapy with combretastatin A-4 phosphate and 5-fluorouracil in an experimental murine colon adenocarcinoma. Anticancer Res 2000;20:229 –233. 34. Nelkin BD, Ball DW. Combretastatin A-4 and doxorubicin combination treatment is effective in a preclinical model of human medullary thyroid carcinoma. Oncology Reports 2001; 8:157–160.
PII S0360-3016(02)02742-6
BIOLOGY CONTRIBUTION
ENHANCEMENT OF RADIATION THERAPY BY THE NOVEL VASCULAR TARGETING AGENT ZD6126 DIETMAR W. SIEMANN, PH.D.,*
AND
AMYN M. ROJIANI, M.D., PH.D.†
*Department of Radiation Oncology, Shands Cancer Center, University of Florida, Gainesville, FL; †Departments of Interdisciplinary Oncology and Pathology, H. Lee Moffitt Cancer Center and Research Institute, University of South Florida, Tampa, FL Purpose: The aim of this study was to evaluate the antitumor efficacy of the novel vascular targeting agent ZD6126 (N-acetylcochinol-O-phosphate) in the rodent KHT sarcoma model, either alone or in combination with single- or fractionated-dose radiation therapy. Methods: C3H/HeJ mice bearing i.m. KHT tumors were injected i.p. with ZD6126 doses ranging from 10 to 150 mg/kg. Tumors were irradiated locally in unanesthetized mice using a linear accelerator. Tumor response to ZD6126 administered alone or in combination with radiation was assessed by clonogenic cell survival assay or tumor growth delay. Results: Treatment with ZD6126 led to a rapid tumor vascular shutdown as determined by Hoechst 33342 diffusion. Histologic evaluation showed morphologic damage of tumor cells within a few hours after drug exposure, followed by extensive central tumor necrosis and neoplastic cell death as a result of prolonged ischemia. When combined with radiation, a 150 mg/kg dose of ZD6126 reduced tumor cell survival 10 –500-fold compared with radiation alone. These enhancements in tumor cell killing could be achieved for ZD6126 given both before and after radiation exposure. Further, the shape of the cell survival curve observed after the combination therapy suggested that including ZD6126 in the treatment had a major effect on the radiation-resistant hypoxic cell subpopulation associated with this tumor. Finally, when given on a once-weekly basis in conjunction with fractionated radiotherapy, ZD6126 treatment was found to significantly increase the tumor response to daily 2.5 Gy fractions. Conclusion: The present results demonstrated that in the KHT sarcoma, ZD6126 caused rapid tumor vascular shutdown, induction of central tumor necrosis, tumor cell death secondary to ischemia, and enhancement of the antitumor effects of radiation therapy. © 2002 Elsevier Science Inc. ZD6126, KHT sarcoma, Vascular targeting.
The initiation and maintenance of a supportive vasculature is critical to both the growth and survival of a solid tumor mass (1, 2). Indeed, in the absence of neovascularization, tumor cells remain dormant, and solid tumors fail to progress beyond a microscopic size (1, 2). Although the process of angiogenesis also plays a significant role in many normal and developmental as well as pathologic conditions, angiogenesis in adults is usually limited to specific reproductive organs (3), and the turnover of vascular endothelial cells typically is measured in months and years (4). Unlike the well-defined microvascular architecture of normal tissues, the abnormal nature of the tumor microcirculation is well documented (5, 6). A key characteristic of the tumor microcirculation is that tumor-associated blood vessels contain populations of endothelial cells that are actively divid-
ing in response to angiogenic factors (4, 7). Because the nutritional support of large numbers of neoplastic cells depends on small numbers of endothelial cells, damaging the tumor endothelium could have a marked impact on tumor cell survival and growth (8, 9). The expanding tumor endothelium therefore represents a key target for cancer therapy (2, 9). Unlike anti-angiogenic strategies that aim to inhibit new vessel formation (2), vascular targeting aims to cause selective and direct damage to existing tumor endothelium, resulting in a rapid and catastrophic shutdown of the tumor vascular network, leading to extensive secondary tumor cell death (8, 9). Several agents that elicit irreversible vascular shutdown selectively within solid tumors have been identified. These include the tubulin-binding agents, most notably colchicine, the vinca alkaloids, and the combretastatins
Reprint requests to: Dietmar W. Siemann, Ph.D., Department of Radiation Oncology, University of Florida Shands Cancer Center, 2000 SW Archer Road, Box 100385, Gainesville, Florida 32610. Tel: (352) 265-0287; Fax: (352) 265-0759; E-mail: siemadw@ ufl.edu Supported in part by a grant from AstraZeneca Inc. and the U.S.
National Cancer Institute (PHS grant CA 84408). Acknowledgments—The authors thank Sharon Lepler, Emma Mercer, Heather Newlin, Chris Pampo, and Jeffrey Lin for providing excellent technical assistance. Received Sep 6, 2001, and in revised form Dec 27, 2001. Accepted for publication Jan 7, 2002.
INTRODUCTION
164
Combinations of radiation and ZD6126
(10 –16). Although most tubulin-binding agents have demonstrated antivascular effects only when near-maximally tolerated doses are administered (11, 15), there are some exceptions. For example, combretastatin A4 disodium phosphate (CA4DP) and its derivative AVE8062 have been shown to lead to rapid vascular shutdown, reductions in tumor blood flow, and induction of necrosis in a variety of preclinical tumor models at minimally toxic doses in vivo (12, 13, 17–20). The principal mechanism of action responsible for these in situ antivascular effects observed after treatment with these agents is believed to be damage mediated through effects on proliferating endothelial cell microtubule polymerization leading to endothelial cell shape changes (14, 15, 17, 21). Recently, ZD6126 has been reported to have selective tumor vascular targeting activity (22, 23), to cause disruption of the cytoskeleton of proliferating endothelial cells, and to induce tumor necrosis in preclinical tumor models at well-tolerated doses (22, 23). After treatment with vascular targeting agents such as CA4DP and ZD6126, viable tumor cells are found at the tumor periphery, presumably because they are supplied by normal vessels not susceptible to damage by these agents (16). Because this residual tumor tissue is likely to be well oxygenated, and because the future application of vascular targeting strategies will probably include their utilization in conjunction with conventional anticancer therapies, the present investigations were undertaken to examine the efficacy of combining 2D6126 with ionizing radiation. METHODS AND MATERIALS Animal and tumor model KHT sarcoma cells (24) were injected i.m. (2 ⫻ 105 cells in a volume of 0.01 mL) into the hind limbs of 6 – 8-weekold female C3H/HeJ mice (Jackson Laboratories, Bar Harbor, ME). At the start of a given experiment, mice bearing tumors of a size equivalent to a weight of ⬃0.5 g (clonogenic cell survival assay) or ⬃0.2 g (regrowth delay assay) were selected from a large number of mice inoculated with the same cell numbers on the same day and treated with either ZD6126 alone or ZD6126 plus radiation, or kept as untreated controls. Drug treatment ZD6126 (AstraZeneca, Macclesfield, UK) was dissolved in 0.9% saline and injected i.p. into a volume of 0.01 mL/g body weight. Although a maximum tolerated dose was not established, single doses as large as 400 mg/kg ZD6126 have been given to C3H mice without overt toxicity. Irradiation Irradiations were performed on nonanesthetized mice using a 6-MeV linear accelerator operating at a dose rate of 4 Gy/min. Each mouse was confined to a plastic jig with its tumor-bearing leg extended through an opening on the side to allow the tumor to be irradiated locally. Five tumorbearing mice were irradiated simultaneously.
● D. W. SIEMANN AND A. M. ROJIANI
165
Morphologic and morphometric analysis The tumor was dissected from the hind limb, and adjacent connective tissue was removed. After fixation in 10% neutral buffered formalin, the tumor was inked around its entire surface to define borders and serially divided at 1-mm intervals. Alternate slices were submitted for routine histology processing. Sections were dehydrated through a series of graded alcohols, processed in xylene, and embedded in paraffin. Four m sections obtained from these paraffin blocks were stained with hematoxylin and eosin as well as Masson’s trichrome stain. Each hematoxylin and eosin– stained tumor section was divided into 4 – 8 grids using a fine permanent marker. Images were captured with a framegrabber, and sections were morphometrically assessed with the Image Pro Plus image analysis system (Media Cybernetics, Silver Spring, MD). Using an “irregular area of interest” tool, areas of necrosis were outlined and measured on each grid. Multiple grids for each tumor were combined and compared with the total area of the tumor, results being expressed as percent necrosis. Hoechst-33342 studies Hoechst-33342 (bisbenzimide, Sigma) solution was made up in 0.9% sterile saline immediately before use. KHT sarcoma– bearing mice were either untreated or treated with 150 mg/kg ZD6126. Hoechst-33342 then was administered at 40 mg/kg i.v. (volume 5 mL/kg) at various times after ZD6126 injection (25). One minute after Hoechst-33342 injection, the mice were killed, and the tumors were resected and immediately immersed in liquid nitrogen for subsequent frozen sectioning. For each tumor sample, 10 m cryostat sections were cut at three different levels between one pole and the equatorial plane. The sections were air dried and studied under ultraviolet illumination using a fluorescent microscope. Blood vessel outlines were identified by the surrounding halo of fluorescent H33342labeled cells. Vessel counts were performed using a Chalkley point array for random sample analysis (26). Briefly, each section was viewed at 10⫻ objective magnification. A 25-point Chalkley grid was positioned randomly over field of view. Any points falling within haloes of fluorescent cells were scored positive. Twenty random fields were counted per section, and a minimum of six sections per tumor were examined. Data from 3–5 tumors were pooled and presented. Clonogenic cell survival Clonogenic cell survival in treated or untreated tumors was determined using an in vivo to in vitro cell survival assay as previously described (24). Briefly, 24 h after treatment, KHT sarcomas were excised, and single-cell suspensions were prepared using a combined mechanical and enzymatic dissociation procedure. The cells were counted, and various dilutions were prepared. KHT cells were mixed with 0.2% agar containing ␣-minimum essential medium supplemented with 10% fetal bovine serum and plated into 24-well plates. Two weeks later, the plates were harvested,
166
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 1, 2002
Fig. 1. Vessel counts in KHT sarcoma– bearing mice as a function of time after treatment with a 150 mg/kg dose of ZD6126. At a given time after ZD6126 treatment, mice were injected with Hoechst 33342 (40 mg/kg) to identify patent vessels, and tumors were removed 1 min later. Counting was performed using a Chalkley point array for random sample analysis. Data are the mean ⫾ SE of 3–5 tumors.
and the resulting colonies were counted with a dissecting microscope. Tumor-surviving fractions were determined by multiplying the calculated fraction of surviving cells by the ratio of cells recovered in treated vs. untreated tumors. Tumor growth delay In the tumor regrowth assay, the tumor size was measured 3–5 times per week after treatment by passing the tumorbearing leg through a series of increasing-diameter holes in a plastic rod. The smallest-diameter hole the tumor-bearing leg would pass through was recorded and converted to a tumor weight using a calibration curve (27). The time required for the untreated or treated tumors to grow to five times the starting size was recorded. The median time for the tumors in each group of mice to reach this end point was calculated. Confidence intervals about the median were determined by nonparametric statistics (28). The median tumor responses were plotted, and significance between treatment groups was determined by a Wilcoxon rank-sum test (␣ ⫽ 0.025). RESULTS To examine the effects of ZD6126 treatment on the vasculature of KHT sarcomas, functional vascular volume measurements were carried out based on the use of the perivascular stain Hoechst 33342. The results showed that the vascular-damaging effects of this agent began to manifest themselves within 30 min after the administration of a single 150 mg/kg dose (Fig. 1). By 2 h after treatment, KHT
sarcomas showed functional vessels only near the periphery, suggesting that this dose of ZD6126 caused an almost complete vascular shutdown in the tumors. Histologic assessment reveals the tumors to be hypercellular, with significant nuclear atypia and mitotic activity. The tumor, although forming well-defined mass, could frequently be seen infiltrating and entrapping normal skeletal muscle (Fig. 2a). Tumors treated with ZD6126 displayed clear morphologic evidence of damage within a few hours after treatment. This was subsequently followed by the rapid onset of frank necrosis such that by 24 h after a 150 mg/kg dose of ZD6126, KHT sarcomas showed extensive central necrosis with viable tumor cells evident only at the periphery of the tumor adjacent to the surrounding normal tissues (Figs. 2b– d). The tissue necrosis involves both areas of tumor proliferation, as well as areas of tumor infiltrating skeletal muscle (Figs. 2b and 2c). This pattern of necrosis is well illustrated at higher magnification, where tissue destruction and nuclear fragmentation is readily evident in the entire central portion of the tumors, extending to within a few cell layers from the peripheral margin of the tumor (Fig. 2d). To quantify the extent of necrosis produced by ZD6126 treatment, sections from KHT sarcomas removed 24 h after treatment were assessed using an image analysis system. The results (Fig. 3, left-hand panel) showed that compared to the ⬃10% necrosis seen in untreated tumors, treatment with 150 mg/kg ZD6126 increased the extent of necrosis to ⬃90%. Antitumor effectiveness of this dose of ZD6126 was determined by measuring clonogenic cell survival. The data
Combinations of radiation and ZD6126
Fig. 2. Standard hematoxylin and eosin staining of 4-m sections from KHT sarcomas. Untreated KHT sarcomas show tumor cells infiltrating and entrapping host skeletal muscle fibers (arrows). [(a) Original magnification: 200⫻.] Only minute areas of necrosis are seen in these tumors. Sections from mice that had received a 150-mg/kg dose of ZD6126 24 h before assessment are shown in b– d. [(b– d) Original magnification: 40⫻, 200⫻, and 400⫻,] After ZD6126 treatment, KHT tumors show extensive central areas of necrosis with hemorrhage readily apparent at low power. A small rim of viable tumor can be readily identified at the periphery (arrowheads), and necrotic areas included regions infiltrating skeletal muscle (arrows), best seen in (b) and (c). This viable tumor rim consists of 6 to 8 layers of tumor cells resting against the unaffected host tissue. At high magnification (d), the areas of tumor necrosis are easily separated from the hyperchromatic, pleomorphic, mitotically active tumor cells.
showed that this treatment led to ⬃1–1.5 logs of tumor cell kill. Subsequent evaluations of KHT sarcomas treated with various doses of this agent demonstrated that increasing doses of ZD6126 resulted in a dose-dependent increase in tumor cell kill (Fig. 3, right-hand panel). In particular, doses ranging from 50 to 150 mg/kg resulted in a loss of viable cells from the tumors. Because the timing between agents is an important parameter in combined modality therapies, initial experiments were performed to evaluate the effect on tumor response of various time intervals between the vascular targeting agent and radiation. Figure 4 shows the effect of administering a 150 mg/kg dose of ZD6126 at various times before or after a 10 Gy dose of radiation on clonogenic cell survival in the KHT sarcoma. The data indicate that administering ZD6126
● D. W. SIEMANN AND A. M. ROJIANI
167
24 h before, immediately after, and up to 4 h after radiotherapy enhanced tumor cell killing compared to that achieved with radiation alone. However, giving ZD6126 shortly before (1 h) the 10 Gy dose failed to reduce tumor cell survival. The greatest increase in tumor cell killing resulted from a treatment strategy in which the vascular targeting agent was given 0.5–1 h after radiation. On the basis of the timing studies described above, a 1-h separation between the radiation and vascular targeting agent treatments was selected and applied in subsequent investigations. Using this treatment sequence, KHT sarcoma– bearing mice were irradiated with doses ranging from 0 to 25 Gy and then given a 150 mg/kg dose of ZD6126 (Fig. 5). When clonogenic cell survival was assessed 24 h later, the extent of tumor cell killing in mice treated with the combination of agents was increased significantly compared to that seen for radiation alone (Fig. 5, closed vs. open symbols). Not only did the addition of ZD6126 to the treatment protocol reduce tumor cell survival 10 –500-fold below that for radiation alone, but it also seemed to reduce the proportion of radiobiologic hypoxic tumor cells that dominate cell survival in this tumor at high doses (⬎10 Gy) and give rise to the characteristic “break” in the cell survival curve (24). This observation was similar to one previously made in our laboratories when combining single-dose radiotherapy with the vascular targeting agent CA4DP (13, 29). Because clinical radiotherapy typically is given using fractionated treatment protocols, additional experiments combining daily 2.5 Gy fractions with once-weekly ZD6126 (150 mg/kg) were performed. In these experiments, radiation was given for 2 weeks (Monday through Friday) with the vascular targeting agent being administered 1 h after each of the two Friday radiation fractions. Figure 6 shows the growth patterns of the median tumors of groups of mice treated with radiotherapy alone or radiotherapy plus ZD6126; Fig. 7 demonstrates the range of responses observed. In this protocol, ZD6126 alone failed to retard KHT sarcoma growth. However, it should be noted that only a single dose of ZD6126 (the first treatment) was given, because, as a result of rapid growth of the KHT sarcoma, the mice had to be killed before the second drug treatment could be given at the end of the second week. Although ZD6126 was not effective on its own as a single dose, a significant enhancement of the radiation response (p ⬍ 0.05) was achieved when two doses of ZD6126 were added to the radiotherapy treatment (Figs. 6 and 7). DISCUSSION There has recently been a great deal of rekindled interest in the application of agents that directly damage the tumor vasculature, a notion initially advanced nearly 20 years ago by Denekamp (8, 9). This resurgence has come about primarily because of the development of new agents showing a high degree of preferential toxicity to the growing tumor endothelium. Tubulin-binding agents have long been known
168
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 1, 2002
Fig. 3. (Left) Percent necrosis in KHT sarcomas after a 150 mg/kg dose of ZD6126. (Right) Tumor cell kill in KHT sarcomas treated with increasing doses of ZD6126. Data were determined 24 h after drug treatment and are the mean ⫾ SE of 3–5 (necrosis) or 3–12 (cell survival) tumors.
to possess some degree of antivascular action (11). However, the more recent identification of a select group of agents that, at doses well below the maximum tolerated dose, demonstrate a profound impact on tumor vasculature (12, 17, 22, 23) has generated substantial enthusiasm for the potential of such an anticancer treatment strategy (15, 16). Although the mechanisms governing these tumor-selective effects of tubulin-binding agents have not been unequivocally demonstrated, recent evidence supports the notion that the inhibition of tubulin polymerization by such agents affects endothelial cell shape, leading to thrombus formation and a consequent secondary cascade of ischemic tumor cell death (21). In the present study, fluorescent microscopy was used to visualize Hoechst 33342 distributions on tumor sections to identify tumor blood vessels actively involved in perfusion (25). The application of this vascular marker demonstrated a significant reduction in patent vessels that was detectable within 30 min after ZD6126 treatment and lasted for at least a 24-h period. This rapid and extensive shutdown in vascular function observed in the KHT sarcoma is typical of the previously reported reductions in blood perfusion associated with the vascular targeting agent CA4DP in this (13, 20) and other (12, 14, 15, 17, 18) tumor models. The suppression of functional vessel number by ZD6126 also is consistent with recent MRI evaluations of this agent’s action on tumor vasculature (30, 31). The consequences of the pathophysiologic effects of ZD6126 were clearly apparent in subsequent histologic analyses of KHT sarcomas (Fig. 2). Evidence of morphologic damage became detectable within 12 to 18 h after a
Fig. 4. Effect on KHT sarcoma cell survival of administering a 150 mg/kg dose of ZD6126 at various times before or after a 10 Gy dose of radiation. The solid square and the open circle represent the cell survival after ZD6126 and radiation alone, respectively. Data are the mean ⫾ SE of 6 –12 tumors.
Combinations of radiation and ZD6126
Fig. 5. Tumor cell survival in KHT sarcomas treated with a 150 mg/kg dose of ZD6126 1 h after a range of single doses of radiation. Tumors were evaluated 24 h after completing the treatment. Results are the mean ⫾ SE of 3– 6 experiments.
● D. W. SIEMANN AND A. M. ROJIANI
169
150 mg/kg treatment, and by 24 h, extensive hemorrhagic necrosis had occurred. The surviving viable rim of tumor cells (Figs. 2b– d) is a characteristic feature of tumors treated with antivascular pharmaceuticals (12–14, 18, 19, 22, 23) as well as vessel-targeting antibodies (32). Although no direct evidence supporting this currently exists, the failure to eliminate neoplastic cells at the periphery of the tumor is believed to reflect a lack of damage to vessels in the normal tissues adjacent to the tumor, which provide adequate nutrition and waste product removal and allow the tumor cells to survive and proliferate. Although the destruction of large areas, particularly in the central and typically most treatment-resistant regions of the tumors, is highly desirable, the survival of cells at the tumor periphery remains a concern. To attempt to address this issue, preclinical assessments of the efficacy of vascular targeting agents delivered in multidose regimens have been initiated. Results with CA4DP indicate that the response in some tumor models can be enhanced by repeated drug doses (15, 29). Similarly, significant growth delays have been observed when multiple doses of ZD6126 were administered (23). An alternate strategy to eliminate the cells surviving the vascular targeting agent and hence to ultimately control the tumor is to combine this treatment with a therapy directly targeting these cells (16). This approach has proven effective when combining CA4DP with conventional anticancer drugs (15, 29, 33, 34) and radiation (13, 20, 29). In the present study, we demonstrated that ZD6126 could effec-
Fig. 6. KHT sarcoma weight as a function of time after (open circle) no treatment, (filled circle) ZD6126 alone, (open square) 10 2.5-Gy fractions administered in 12 days, or (filled square) the combination of the fractionated radiotherapy protocol plus two 150 mg/kg doses of ZD6126, given 1 h after the 5th and 12th 2.5-Gy fraction, respectively. Data shown represent the responses of the median tumors of groups of 8 –10 mice.
170
I. J. Radiation Oncology
● Biology ● Physics
Fig. 7. Effect on tumor growth of including a 150 mg/kg dose of ZD6126 in fractionated radiotherapy treatment (drug and radiation protocol as described in Fig. 6). Data shown are the median times required to grow to 5⫻ the starting size ⫾ 96% confidence intervals in groups of 8 –10 mice (28).
tively enhance the radiation response of the KHT sarcoma to both single- and fractionated-dose radiotherapy (Figs. 4 –7). An important consideration in such investigations is the issue of sequencing and timing (Fig. 4). When the ZD6126 was injected 1 h before irradiating the tumors, the resultant cell survival was not significantly different from that observed for radiation alone. The lack of improvement in the treatment response may indicate that when the vascular targeting agent is given shortly before radiotherapy, some parts of the tumors experience transient reductions in blood flow sufficient to render cells hypoxic, but for a time insufficiently long for them to undergo ischemic death. However, all other sequence schedules investigated, including the administration of ZD6126 from 0 to 4 h after or 24 h before irradiation, led to significantly greater tumor cell killing with the combination treatment. Indeed, when a complete dose–response curve was established, administering a 150
Volume 53, Number 1, 2002
mg/kg dose of ZD6126 1 h after a range of radiation doses resulted in improved tumor cell killing at all doses investigated (Fig. 5). The displacement of the “tail” of the cell survival curve to lower survival values was of particular interest, because it suggests that the inclusion of ZD6126 in the treatment strategy had reduced the proportion of radiobiologically hypoxic cells in the tumor. A similar conclusion was reached in our previous evaluations of CA4DP (13, 29). It also is consistent with the present histologic evaluations demonstrating induction of extensive central necrosis in the tumors, leaving viable cells only at the periphery (Fig. 2). Taken together, these findings suggest that administering ZD6126 post-radiotherapy may provide the most appropriate scheme for combining these therapies. Because it is almost certain that any clinical combination therapy will have to involve fractionated radiotherapy, we also chose to examine whether including ZD6126 in a protocol of fractionated radiation exposures could improve the tumor response. Radiation was given Monday through Friday for 2 weeks at 2.5 Gy per fraction, and ZD6126 (150 mg/kg) was administered each Friday, 1 h after that day’s radiation dose. The results shown in Figs. 6 and 7 indicate that, while this radiation schedule induces on its own a tumor growth delay of 6 days, this delay is significantly increased (p ⬍ 0.05) to 12 days by the two ZD6126 doses added to the treatment regimen. In conclusion, ZD6126 was found in these studies to demonstrate striking antivascular effects in tumors, leading to the induction of necrosis and a consequential rapid loss of clonogenic neoplastic cells from the cell population. In practical terms, this agent may be acting to chemotherapeutically reduce the viable tumor mass. In light of histologic evidence that neoplastic cells adjacent to peripheral normal tissues may survive antivascular agent treatment because of their close proximity to normal blood vessels, it would seem that radiotherapy or chemotherapy may need to be administered to destroy the remaining rim of tumor cells. The present findings support this notion and clearly demonstrate that the addition of ZD6126 can significantly enhance the antitumor efficacy of both single- and fractionated-dose radiation therapy. While the response of tumors to the ZD6126-radiation combination would seem to be an additive effect of two therapies, complete dose–response analysis of both modalities would be required to prove this unequivocally. Although normal tissue toxicities were not measured in these experiments, the observation that such tumor response enhancements could occur at nontoxic doses (22, 23) is encouraging and suggests that a therapeutic benefit may be achievable. Further evaluations of the therapeutic potential of this agent used alone or in an adjuvant setting clearly seem warranted.
REFERENCES 1. Folkman J, Shing Y. Angiogenesis. J Biol Chem 1992;267: 10931–10934. 2. Folkman J. Seminars in medicine of the Beth Israel Hospital,
Boston. Clinical applications of research on angiogenesis. N Engl J Med 1995;333:1757–1763. 3. Winlaw DS. Angiogenesis in the pathobiology and treatment
Combinations of radiation and ZD6126
4. 5. 6. 7. 8. 9. 10.
11. 12. 13.
14. 15. 16. 17. 18.
19.
of vascular and malignant diseases. Ann Thorac Surg 1997; 64:1204 –1211. Hobson B, Denekamp J. Endothelial proliferation in tumors and normal tissue: Continuous labeling studies. Br J Cancer 1984;49:405– 413. Vaupel P, Kallinowski F, Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: A review. Cancer Res 1989;49:6449 – 6465. Konerding MA, Miodonski AJ, Lametschwandtner A. Microvascular corrosion casting in the study of tumor vascularity: A review. Scanning Microsc 1995;9:1233–1244. Gastl G, Hermann T, Steurer M, et al. Angiogenesis as a target for tumor treatment. Oncology 1997;54:177–184. Denekamp J. Vasculature as a target for tumour therapy. Prog Appl Microcirc 1984;4:28 –38. Denekamp J. Angiogenesis, neovascular proliferation and vascular pathophysiology as targets for cancer therapy. Br J Radiol 1993;66:181–196. Baguley BC, Holdaway KH, Thomsen LL, et al. Inhibition of growth of colon 38 adenocarcinoma by vinblastine and colchicine. Evidence for a vascular mechanism. Eur J Cancer 1991;27:482– 487. Hill SA, Sampson LE, Chaplin DJ. Anti-vascular approaches to solid tumor therapy: Evaluation of vinblastine and flavone acetic acid. Int J Cancer 1995;63:119 –123. Dark GG, Hill SA, Prise VE, et al. Combretastatin A-4, an agent that displays potent and selective toxicity toward tumor vasculature. Cancer Res 1997;57:1829 –1834. Li L, Rojiani A, Siemann DW. Targeting the tumor vasculature with combretastatin A-4 disodium phosphate: Effects on radiation therapy. Int J Radiat Oncol Biol Phys 1998;42:899 – 903. Grosios K, Holwell SE, McGown AT, et al. In vivo and in vitro evaluation of combretastatin A-4 and its sodium phosphate prodrug. Br J Cancer 1999;81:1318 –1327. Chaplin DJ, Pettit GR, Hill SA. Anti-vascular approaches to solid tumour therapy: Evaluation of combretastatin A4 phosphate. Anticancer Res 1999;19:189 –195. Siemann DW, Warrington KH, Horsman MR. Targeting tumor blood vessels: An adjuvant strategy for radiation therapy. Radiother Oncol 2000;57:5–12. Tozer GM, Prise VE, Wilson J, et al. Combretastatin A-4 phosphate as a tumor vascular-targeting agent: Early effects in tumors and normal tissues. Cancer Res 1999;59:1626 –1634. Horsman MR, Ehrnrooth E, Ladekarl M, et al. The effect of combretastatin A-4 disodium phosphate in a C3H mouse mammary carcinoma and a variety of murine spontaneous tumours. Int J Radiat Oncol Biol Phys 1998;42:895– 898. Hori K, Saito S, Nihei Y, et al. Antitumor effects due to irreversible stoppage of tumor tissue blood flow: Evaluation of a novel combretastatin A-4 derivative, AC7700. Jpn J Cancer Res 1999;90:1026 –1038.
● D. W. SIEMANN AND A. M. ROJIANI
171
20. Murata R, Siemann DW, Overgaard J, Horsman MR. Interaction between combretastatin A4 disodium phosphate and radiation in murine tumours. Radiother Oncol. 2001;60:155– 161. 21. Galbraith SM, Chaplin DJ, Lee F, et al. Effects of combretastatin A4 phosphate on endothelial cell morphology in vitro and relationship to tumour vascular targeting activity in vivo. Anticancer Res 2001;21:93–102. 22. Davis PD, Hill SA, Galbraith SM, et al. ZD6126: A new agent causing selective damage of tumor vasculature (Abstr). Proc Am Assoc Cancer Res 2000;41:329. 23. Blakey DC, Curry B, Westwood FR, Ashton SE. Vascular targeting activity of ZD6126 in two tumour models growing in athymic rats (Abstr). Proc Am Assoc Cancer Res 2001;42:823. 24. Siemann DW, Johansen IM, Horsman MR. Radiobiological hypoxia in the KHT sarcoma: Predictions using the Eppendorf histograph. Int J Radiat Oncol Biol Phys 1998;40:1171–1176. 25. Smith KA, Hill SA, Begg AC, et al. Validation of the fluorescent dye Hoechst 33342 as a vascular space marker in tumours. Br J Cancer 1988;57:247–253. 26. Curtis ACS. Area and volume measurements by random sampling methods. Med Biol Illustra 1960;10:261–266. 27. Siemann DW, Sutherland RM. In vivo tumor response to single and multiple exposures of adriamycin. Eur J Cancer 1980;16:1433–1440. 28. Noether J. Introduction to statistics: A fresh approach. Boston: Houghton Mifflin; 1971. p. 86 –112. 29. Li L, Rojiani AM, Siemann DW. Preclinical evaluations of therapies combining the vascular targeting agent combretastatin A-4 disodium phosphate and conventional anticancer therapies in the treatment of Kaposi’s sarcoma. Acta Oncol 2002, in press. 30. Evelhoch JL, He Z, Polin L, et al. MRI evaluation of the effects of ZD6126 on tumor vasculature (Abstr). Proc Am Assoc Cancer Res 2001;42:107. 31. Robinson SP, McIntyre DJ, Howe FA, Griffiths JR, Blakey DC, Waterton JC. Assessment of the novel tumor vascular targeting agent ZD6126 by non-invasive multi-gradient echo MRI (Abstr). Proc Am Assoc Cancer Res 2001;42:415. 32. Huang X, Molema G, King S, et al. Tumor infarction in mice by antibody-directed targeting of tissue factor to tumor vasculature. Science 1997;275:547–550. 33. Grosios K, Loadman PM, Swaine DJ, et al. Combination chemotherapy with combretastatin A-4 phosphate and 5-fluorouracil in an experimental murine colon adenocarcinoma. Anticancer Res 2000;20:229 –233. 34. Nelkin BD, Ball DW. Combretastatin A-4 and doxorubicin combination treatment is effective in a preclinical model of human medullary thyroid carcinoma. Oncology Reports 2001; 8:157–160.