Improved Effect of an Antiangiogenic Tyrosine Kinase Inhibitor (SU5416) by Combinations with Fractionated Radiotherapy or Low Molecular Weight Heparin

Abstract

The effect of combining SU5416 with fractionated radiotherapy or with low molecular weight (LMW) heparin (dalteparin) was studied in U87 human glioblastoma xenografts in nude mice. SU5416 is antiangiogenic by a specific inhibition of the vascular endothelial growth factor receptor 2 (VEGFR-2), and heparins are assumed to bind VEGF. Both SU5416 (100 mg/kg every second day in 5 days) and 3 Gy x 5 produced moderate, yet significant, growth inhibition. Tumors treated with concomitant irradiation and short-term SU5416 maintained a lower growth rate during regrowth than the other treatment groups (P=.007). Dalteparin (1000 IE/kg subcutaneously once a day) had no growth-inhibitory effect on its own, but when this LMW heparin was added to the SU5416 schedule, a significantly enhanced growth inhibition was obtained. VEGF protein content in tumors was not significantly altered by SU5416, but a significant decrease in VEGF levels was found in tumors treated with concomitant dalteparin and SU5416 compared with controls (P=.03). We conclude that: 1) an additive growth-inhibitory effect is obtained by combining SU5416 and fractionated radiotherapy; and 2) LMW heparin (dalteparin), in combination with SU5416, decreases the level of VEGF in tumors and increases the growth-inhibitory effect of SU5416.

Introduction

SU5416 is a small molecule inhibitor of tyrosine kinase receptors, including the vascular endothelial growth factor receptors (VEGFR-2) and the stem cell factor receptor c-kit [1,2]. The compound can slow tumor growth in a number of experimental tumors [3], and reduces vascular density in glioblastomas [4].

Meta-analyses comparing cancer patients treated with heparin or low molecular weight (LMW) heparin for venous thromboembolism have shown that LMW heparin improves the cancer outcome [5,6]. Heparins can affect tumors in several ways, by interactions with growth factors, enzymes, and structural proteins in the extracellular matrix, as well as by direct interaction with cells [7]. Antiangiogenic effects of LMW heparins have been shown in different studies [7–9]. The antiangiogenic mechanisms of LMW heparins may be complex, including the different processes mentioned above and the binding of angiogenic growth factors like VEGF and bFGF to the LMW heparin [8].

In the present study, we combined SU5416 with ionizing radiation (IR) or with dalteparin, a LMW heparin, and demonstrate that these combinations increase the growth-inhibitory effect of SU5416 on glioblastoma growth.

Materials and Methods

U87 Cells In Vitro

The human glioblastoma cell line U87 MG was grown at 37°C in a humidified atmosphere of 5% CO₂ in 175-cm² (650 ml) CELLSTAR culture flasks. The medium used was Eagle's MEM containing 10% FCS and no antibiotics. Cells were harvested in an exponential growth phase with a cell scraper. Controls and irradiated cells were harvested and kept on ice for 1 hour at the time of irradiation and subsequently the cells were added to new culture flasks with fresh medium. The cell medium was collected for analysis at 8, 24, and 48 hours after irradiation. Four independent experiments were performed.

Tumor Xenografts

Male 7-week-old athymic nude mice (NMRI-nu/nu) obtained from M&B (Ry, Denmark) were used. The mice were kept in laminar airflow benches. They received sterile food pellets and water ad libitum. Institutional guidelines for animal welfare and experimental conduct were followed. Xenografts were established by subcutaneous injection of tumor cells and maintained by serial transplantation. Prior to transplantation, the mice were anesthesized by a subcutaneous injection of ketamine (10 mg/kg) and xylazine (1 mg/kg) in 0.9% NaCl solution. Through a 1-cm incision in the dorsal skin, 1-mm³ tumor blocks were subcutaneously implanted into the flank. Tumors were measured daily during tumor growth, by two orthogonal diameters d₁ and d₂. Tumor volume was calculated according to the equation:

$$V=\text{π}/6{(d_1{d}_2)}^{3/2}k,$$

where k is an empirical constant equal to 0.67.

The tumor volume doubling time T_d was calculated for each individual tumor in the exponential growth phase after the discontinuation of therapy with IR and/or SU5416. The best fit exponential growth curve, V=aexp(bt), was computed using a SigmaPlot software program (SPSS, Holte, Denmark). The tumor volume doubling time for each tumor T_d is ln2/b.

Irradiation of Xenograft Tumors and Cell Cultures

A single dose of 10 or 15 Gy of IR, given as five daily fractions of 3 Gy, was applied using a Stabilipan (Siemens, Ballerup, DK) therapeutic unit that yields 4.58 Gy/min at 300 kV. Mice were anesthesized before each irradiation, as described above.

Drugs

SU5416 was delivered subcutaneously in a centyl-methyl-cellulose (CMC) suspension 100 mg/kg every second day from the day tumors reached a volume threshold of 120 mm³.

When combined with IR, the drug was given for a total of three times (days 1 to 5 of radiation therapy). The drug was delivered 1 to 2 hours before the IR. In another experiment, SU5416 was given every second day until tumors reached a volume of 600 mm³. In that experiment, the drug was used as monotherapy or in combination with dalteparin.

Dalteparin (Fragmin; Pharmacia, DK) was diluted in isotonic saline and delivered subcutaneously 1 IU/g once a day from the day tumors reached a volume of 120 mm³. Dalteparin was used as monotherapy or in combination with SU5416. Controls received no treatment.

Statistics

Student's t-test was used to compare different treatment groups.

VEGF ELISA Measurements

In the cell culture experiments, the cell number was counted at the time of harvesting. The medium was briefly spun, a proteinase inhibitor cocktail (Complete; Roche, Mannheim, Germany) was added, and medium was frozen in liquid nitrogen.

Tumor blocks were obtained at a tumor volume of 600 mm³ and homogenized with five short bursts on a Vibra Cell 50 (Sonics & Materials, Danbury, CT). The protein concentration was determined with a BCA Protein Assay (Pierce, Rockford, IL). The VEGF protein content in tumors and medium was determined using VEGF Quantikine immunoassay kits (R&D Systems, Abingdon, UK).

Immunohistochemistry

Tissue slices from tumors were frozen in cooled isopenthane. Frozen sections were fixed in acetone and CD31 immunostaining was performed as follows. Sections were washed in PBS and TBS and incubated with 10% rabbit serum for 30 minutes. They were then incubated with a mixture of two monoclonal rat antimouse CD31 antibodies at a dilution of 20 µg/ml overnight at 4°C. The antibodies used were clone 390 (Serotec, Oxford, UK) and MEC 13.3 (Pharmingen, San Diego, CA). Rat IgG2a (Serotec) was used as a negative control. Sections were incubated with biotin-conjugated rabbit antirat immunoglobulin (DAKO, Glostrup, DK) at a dilution of 1:600 (2.3 µg/ml) for 30 minutes, washed, and incubated with alkaline phosphatase-conjugated streptavidin (DAKO) at a dilution of 1:200 (1.5 µg/ml) for 30 minutes. As substrate for the alkaline phosphatase reaction, we used freshly prepared Fast Red Substrate System (DAKO), followed by a 10-minutes wash in tap water. Finally, the sections were counterstained with hematoxylin and mounted with aqueous mounting media.

Vessel Density

Vessel density was recorded as the number of point counts of CD31-positive vessels per field, at x200 magnification, viewed through an ocular Chalkley Point Array (Graticules, Tonbridge, UK). Ten fields per section, randomly selected from non-necrotic areas of tumors, were examined with a Leica DMRB microscope (Herlev, DK). The examination was blinded.

Interstitial Fluid Pressure (IFP)

Tumor IFP measurements were performed by the wickin-needle technique, as described previously elsewhere [10]. The IFP was recorded at tumor volumes of 600 mm³, just before the mice were sacrificed and the tumors excised for histological examination and VEGF measurements.

Results

Effects of Treatment on Tumor Growth

SU5416 had a moderate, but significant, growthinhibitory effect as short-term monotherapy (Figure 1). In contrast, fractionated irradiation 3 Gyx5 had a longerlasting effect on tumor growth. There was an additive effect of combining irradiation with SU5416 on tumor growth (Figure 1). In parallel, we tested the effect of combining SU5416 with dalteparin. Here the tumors were excised at a tumor size of 600 mm³ and processed for VEGF measurements. Dalteparin, in combination with SU5416, resulted in a significant inhibition of tumor growth, compared with controls (Figure 2, Table 1). There was no growth-retarding effect of dalteparin alone, rather a tendency towards a faster and a more uniform tumor growth appeared, compared with untreated controls (Figure 2, Table 1).

Figure 1.

SU5416 treatment in combination with irradiation. Effects on tumor growth. Mean values of tumor volume in different treatment groups. Bars: interquartile range. Treatment was initiated when tumors reached a volume of 120 mm³ and continued for 5 days. The tumor volume doubling time T_d for each tumor was determined in the exponential growth phase after discontinuation of therapy. The mean T_d values in tumors treated with both SU5416 and IR were significantly higher than in tumors treated with IR (P=.007, t-test). The time for tumors to reach 800 mm³ was significantly longer in tumors treated with SU5416 than in controls (P=.04, t-test) and also longer in tumors treated with IR+SU5416 compared with IR alone (P=.03, t-test).

Figure 2.

SU5416 treatment in combination with dalteparin. Effects on individual U87 tumor growth. Treatment was initiated when tumors reached a volume of 120 mm³, and continued until excision of tumors. Doses: 100 mg/kg SU5416 subcutaneously every second day; 1 IU/g dalteparin subcutaneously per day.

Table 1.

Growth Characteristics of U87 Tumors After SU5416 and Dalteparin.

	N	T(early)			T(late)			T(growth)
		Mean	SD	t-test Versus Controls	Mean	SD	t-test Versus Controls	Mean	SD	t-test Versus Controls
Controls	15	3.5	1.6		2.6	0.7		6.1	1.7
SU5416	13	5.2	4.5	P=.20	3.8	2.2	P=.06	8.9	5.7	P=.11
Dalteparin	15	3.1	1.1	P=.43	2.1	0.6	P=.08	5.3	1.3	P=.13
SU5416+dalteparin	15	4.7	2.1	P=.09	8.3	10.9	P=.05	13.2	11.8	P=.04

For each individual tumor, the time of first measurement of a volume of 200, 400, and 600 mm³ was recorded.

T_(early): The number of days from a volume of 200 mm³ to a volume of 400 mm³.

T_(late): The number of days from a volume of 400 mm³ to a volume of 600 mm³.

T_(growth): The number of days from a volume of 200 mm³ to a volume of 600 mm³.

VEGF Expression After Therapy

The level of secreted VEGF protein in the growth medium of U87 cells in vitro was higher than in controls 48 hours after 10 Gy irradiation (Figure 3).

Figure 3.

Relative VEGF levels in medium from U87 cell cultures after 10 Gy of irradiation. The VEGF level following 0 Gy (sham-irradiated) was 1.0. At 48 hours, the VEGF levels were increased in four of four experiments to a value of 1.6 to 2.6 (P=.025, t-test). No significant upregulation was seen at 8 and 24 hours. Bars: standard deviation. Four independent experiments were performed.

In contrast, the VEGF level was significantly reduced in U87 tumors treated with SU5416 in combination with dalteparin, compared with control tumor levels (Table 2). When 100 mg/kg SU5416 was administered in a CMC suspension subcutaneously every second day as monotherapy, the reduction in VEGF was not significant. Dalteparin monotherapy did not significantly change VEGF levels either.

Table 2.

VEGF Protein Levels in U87 Tumors After SU5416 and Dalteparin.

Treatment Group	Number of Tumors	VEGF Mean pg/10 µg Total Protein	SD	t-test Compared with Controls
Controls	15	236	88
SU5416	13	176	108	NS
Dalteparin	15	205	103	NS
SU5416+dalteparin	15	159	94	P=.03

Dalteparin, in combination with SU5416, significantly reduced the VEGF content in tumors compared with controls. The reduction after SU5416 or dalteparin as monotherapy was not statistically significant. There was no statistical difference in VEGF levels between the different treatment groups. The VEGF levels were measured at a tumor volume of 600 mm³.

NS=no significant difference.

Vessel Density After Treatment

Chalkley counting was performed in four to five U87 tumors from each of four groups treated with SU5416, dalteparin or dalteparin+SU5416, or no treatment. All tumors were evaluated at a tumor size of 600 mm³. No difference in vessel density was found between these treatment groups (Table 3).

Table 3.

Vessel Density in U87 Tumors After SU5416 and Dalteparin.

Treatment Group	N Fields Counted	Chalkley Counts		t-test Compared with Controls
		Mean Counts/Field	SD
Controls	50	5.1	1.6
SU5416	40	5.1	1.9	NS
Dalteparin	50	4.8	1.7	NS
SU5416+dalteparin	40	5.5	2.1	NS

NS= no significant difference.

IFP

Because VEGF is known to be a permeability factor, differences in VEGF levels in tumors might lead to differences in vessel permeability, resulting in differences in the IFP in tumors. We measured the IFP in untreated tumors and in tumors treated with SU5416, dalteparin, or both. IFP was measured at a tumor size of 600 mm³, immediately before excision of tumors and preparation of tumor tissues for VEGF measurements. There was no difference in tumor IFP between these different treatment groups (Figure 4).

Figure 4.

IFP after SU5416 and dalteparin treatment in U87 tumors. No significant difference between treatment groups was found. Each box shows the median, quartiles, and extreme values within a category.

Discussion

We report that combinations of the VEGFR-2 inhibitor SU5416 with the LMW heparin dalteparin or IR can additively increase the effect on glioblastoma growth.

The role of endothelial cell kill in the biological effect of ionizing irradiation effects was recently discussed by Folkman and Camphausen [11]. It is known that the apoptosis of endothelial cells after irradiation can be avoided or decreased by the addition of the angiogenic growth factors bFGF [12] or VEGF [13], and that the killing of endothelial cells and the regression of vessels are increased by blocking the VEGF response by VEGF antibodies, soluble VEGFR, or SU5416 [13,14].

Here we found an additive effect on the growth inhibition of SU5416 and fractionated radiotherapy with 3 Gy on five concomitant days in U87 glioblastoma tumors. In another glioblastoma line GL261, a similar additive effect of combining IR and SU5416 has been shown [14]. Doses of 30 and 70 Gy of IR, in combination with a monoclonal antibody against VEGFR-2, also showed an additive effect on U87 xenograft growth [15]. So the present findings add to the accumulating evidence that concomitant blockage of the VEGFR-2 signaling increases the effect of IR on glioblastoma growth.

Antiangiogenic treatment strategies aiming at other targets than survival signaling through the VEGFR-2 have shown a comparable potentiation of the effect of IR on tumor growth in several studies (e.g., [16,17]). In accordance, adding the growth factors VEGF or bFGF before irradiation protects against radiation-induced apoptosis in normal gut epithelial cells [18].

Secreted VEGF was upregulated in U87 cells in vitro after 10 Gy. This induction of VEGF secretion by irradiation is in line with previous findings both in vitro and in tumor extracts [13].

The observation that dalteparin had no growth-inhibitory effect alone, whereas combination treatment with SU5416 increased the growth inhibition of SU5416, along with a reduction of the VEGF content in tumors, holds significant clinical promise. The antiangiogenic mechanism of LMW heparin is not clear, but it is known that LMW heparins bind VEGF, and it is likely that the combination of a competitive binding of VEGF by dalteparin increases the effectiveness of a concomitant blockage of the VEGF receptor. Another explanation could be that dalteparin treatment rendered a more uniform dose distribution of SU5416 in solid tumors by a reduced microthrombus formation and a correspondingly improved perfusion rate. To the extent that thrombosis of tumor vessels plays a role in the normal variation in growth of untreated tumors, the antithrombotic effect of dalteparin might similarly explain the uniformly exponential growth in the dalteparin-treated tumors compared with controls.

VEGF is a permeability factor, and VEGF overexpression leads to a formation of leaky vessels [19]. VEGF overexpression leads to a formation of interendothelial gaps or transendothelial pores accounting for the hyperpermeability to plasma proteins and other circulating agents [20]. Because VEGF is overexpressed in many tumors, including U87 glioblastoma, we hypothesized that a reduction in VEGF levels in tumors might also reduce the IFP.

The IFP showed a great intertumor variation, similar to the experiences with other tumors [21]. IFP levels in U87 tumors were not changed by treatment with dalteparin and SU5416, despite a decrease in VEGF protein levels in the same tumors compared with controls. One explanation for this may be that some tumor vessels have defective cellular lining composed of disorganized, loosely connected endothelial cells with defects where tumor cells are directly in contact with the lumen [22,23]. The permeability of such vessels could be largely refractory to changes in VEGF, if they have large openings between endothelial cells. In that case, the fine-tuning of permeability through regression of transendothelial or interendothelial pores due to VEGF would make no differences to the overall permeability.

Vessel density, estimated by Chalkley counting, was unchanged in U87 tumors treated with SU5416 and dalteparin compared to controls. The vessel density was measured at the same tumor size (600 mm³) in controls and treated tumors (i.e., the treated tumors were growing at the time of sampling). In a study by Vajkoczy et al. [4], SU5416 was found to reduce the fractional area of microvasculature in C6 glioma. There the SU5416 treatment was initiated from day 0 after implantation of tumor cells. In the present study, treatment was initiated when tumors were established and growing. SU5416 has been shown to delay the angiogenic switch [24], and in breast cancer, VEGF is essential for the initial, but not continued, growth of tumors [25]. The difference in vessel density response between the C6 study and the present work may be due to a selection pressure in the early critical phase of tumor establishment towards tumor cell subclones with lower oxygen and nutrient requirements.

In summary, we find that SU5416 has an additive effect on tumor growth in combination with fractionated radiotherapy, and that the LMW heparin, dalteparin, in combination with SU5416, increases the growth-inhibitory effect of SU5416 and decreases the level of VEGF in U87 glioblastomas. This may be of particular therapeutic relevance, and a clinical evaluation of combinations of tyrosine kinase inhibitors with IR and LMW heparin is warranted.