Skip to main content
NCBI home page
Journal of Cancer Research and Clinical Oncology logoLink to Journal of Cancer Research and Clinical Oncology
. 2006 Jun 8;132(10):643–652. doi: 10.1007/s00432-006-0112-x

Metronomic trofosfamide inhibits progression of human lung cancer xenografts by exerting anti-angiogenic effects

T Klink 1, C Bela 1, S Stoelting 1, S O Peters 1, R Broll 2, T Wagner 1,3,✉
PMCID: PMC12161021  PMID: 16761121

Abstract

Background

Recent studies of conventional chemotherapeutic drugs administered in metronomic therapy schedules showed remarkable inhibitory effects on tumor angiogenesis. Subsequent and prolonged tumor regression was achieved moreover by circumventing acquired drug resistance. In this study, metronomic and conventional trofosfamide were compared on human NSCLC xenograft “LX-1.”

Materials and methods

In vitro cytotoxicity of trofosfamide on tumor and human umbilical cord endothelial cells was determined under normoxic and hypoxic conditions. Additionally fractions and duration of cell cycles were analyzed by flow cytometry. In vivo LX-1 xenotransplanted nude mice were treated with trofosfamide in conventional and metronomic schedules (i.p./p.o.). Tumor sections were evaluated for microvessel density (MVD), relative growth fraction and apoptosis.

Results

In contrast to the rapid growth of conventionally treated lung cancer, long lasting tumor growth retardation over the total treatment period was achieved with metronomic treatment. While growth fraction and apoptotic rate of LX-1 cells remained unchanged, the MVD was significantly reduced (50%).

Conclusion

Our results show advantages of a metronomic trofosfamide schedule compared to a conventional bolus therapy mainly due to inhibition of angiogenesis. In vitro data show that this mechanism works under normoxic and hypoxic conditions and suggest that this is in part a direct cytotoxic effect on endothelial cells.

Keywords: Anti-angiogenesis, Human, Lung cancer, Metronomic, Trofosfamide

Introduction

Anti-angiogenic treatment strategies were first described in 1971 by Judah Folkman as a promising approach to treat hypervascularized malignancies (Folkman 1971). The idea to primarily focus on endothelial cells was to target on genetic stable cells that lack mutations and thus show no susceptibility to acquire drug resistance (Yu et al. 2002). Several conventional cytotoxic drugs such as vinca alkaloids and folic acid antagonists as well as antibodies against VEGF and its receptors were shown to possess anti-angiogenic activity (Baguley et al. 1991; Boos et al. 1993; Joussen et al. 1999).

Cyclophosphamide is one of the drugs shown to have significant anti-angiogenic activity when administered in a low dose, metronomic schedule to non-human tumor bearing mice (Browder et al. 2000). This finding gave the rationale to investigate for effects of a higher maximum tolerable dose (MTD) administration to prevent tumor and vascular regrowth, even of drug-resistant tumors. Continuous administration of cyclophosphamide chemotherapy at low doses resulted in marked endothelial cell apoptosis, subsequently leading to inhibition of tumor cell growth. Further studies revealed that metronomic administration of cyclophosphamide reduces both the risk of secondary drug resistance of tumor cells, and toxicities on normal host cells (Kerbel et al. 2002; Klement et al. 2002; Man et al. 2002).

Trofosfamide is another oxazaphosphorine derivative like cyclophosphamide and is approved in Germany. Clinical trials showed its efficacy in hematological malignancies (Hodgkin lymphoma, NHL, CLL, hemoblastosis), and solid tumors like breast cancer, ovarian cancer, soft tissue sarcoma and non-small cell lung cancer (Horvath et al. 2000). Trofosfamide was shown to possess less toxic side effects than cyclophosphamide and is therefore preferable for long-term therapy (Falkson and Falkson 1978). Oral daily administrations for out-patients in low doses of 150 mg are well tolerated and lead to remarkable treatment responses (Wagner et al. 1997).

Although one clinical trial including trofosfamide along with a COX-2 inhibitor and pioglitazone has been published recently (Vogt et al. 2003), the efficacy of trofosfamide in a metronomic schedule and its mechanism of action have not been shown so far. Existing experimental studies on metronomic drug schedules with oxazaphosphorines as cyclophosphamide have rarely used tumors of human origin which would be necessary to reflect differences in proliferation depending on tumor-generating species (Man et al. 2002).

Here we report data about the effects of trofosfamide administered at metronomic schedules on human NSCLC “LX-1” in nude mice and describe its effects on microvessel density (MVD), growth fraction and apoptosis. Additionally IC50 values of trofosfamide and cell cycle analysis of LX-1 and human umbilical cord endothelial cells (HUVEC) were compared under normoxic and hypoxic conditions in vitro. Our data provide the experimental basis for our hypothesis that a metronomic low-dose therapy with trofosfamide inhibits tumor growth, predominantly by exerting cytotoxic effects on vascular endothelial cells.

Materials and methods

In vitro

Cell culture

Human non-small cell lung carcinoma “LX-1” cells were obtained from “Cell Lines Services” (Heidelberg, Germany) and endothelial cells were isolated from human umbilical cord veins (HUVEC) (Jaffe et al. 1973). The purity of HUVECs was at least 95% as assessed by flow-cytometry for CD31 positive cells from five different umbilical cord HUVEC harvests. LX-1 cells were cultured in RPMI1640 (Cambrex, Verviers, Belgium) supplemented with 10% fetal bovine serum (FBS). HUVEC were cultured in “Endothelial Cell Growth Medium” (ECGM; Promo Cell, Heidelberg, Germany) supplemented with 10% FBS. Both cell-lines grew as monolayer cultures to subconfluence. The cells were detached from culture layer surfaces and suspended into single cell solutions using 0.05% trypsin–EDTA (Gibco Invitrogen, Palsley, UK). Cells were washed with FBS containing medium and adjusted to the appropriate concentration.

Cytotoxicity

4-Hydroperoxytrofosfamide (4OOH-trofosfamide), which spontaneously decomposes in aqueous solution to the biologically active metabolite 4-hydroxytrofosfamide (4-OH-trofosfamide) (Brinker et al. 2002), was a kind gift from U. Niemeyer, Baxter Oncology (Frankfurt, Germany). Cytotoxic profiles of 4OOH-trofosfamide were determined for LX-1 and HUVEC using the crystal-violet assay, as described by Kueng et al. (1989).

Cells were sown into oxygen-permeable 96-well microtiter plates (3,000/well) and incubated under normoxic (20% O2) or hypoxic (1% O2) conditions. Twenty-four hours later 4OOH-trofosfamide was added in increasing concentrations of 2.5, 5, 7.5, 10, 15, 20, 30 and 40 μmol/l. The drug was dissolved in RPMI 1640 or ECGM both containing 10% FBS. Media were saturated with nitrogen when used for experiments in hypoxia. After 48 h of chemotherapy, cells were fixed by adding glutaraldehyde (11% v/v) to the wells and then washed. Vitality was assessed by staining cells with crystal-violet (0.1% w/v). The optical density (OD) at a wavelength of 450 nm was measured using the “Multiskan” microtiter plate reader (Dynex Technologies, Chantilly, VA, USA) after adding a blue dye solubilized by acetic acid (10% v/v). Then OD values were compared to untreated control cells using the formula OD(sample)/OD(control).

Cell cycle analysis

Cell cycles of LX-1 cells and HUVEC were analyzed by flow-cytometry using the “FACScan” cytometer (BD, Franklin Lakes, NJ, USA). For preparation of the samples we used the method of Hammers et al. (2000) by staining DNA with 7-aminoactinomycin D (7-AAD; Sigma, Deisenhofen, Germany). Cells were denaturized using hydrochloric acid treatment as described in Sasaki et al. (1988).

In a first step, 106 cells were sown into oxygen-permeable culture plates and cultured in normoxic and hypoxic atmospheres. After 72 h, culture cells were detached and separated into singlets by addition of 0.05% trypsin–EDTA and fixed with dimethylsuberimidate (Sigma, Deisenhofen, Germany). Next, the cell suspensions were treated with 2 M HCl for 30 min at room temperature for denaturation. When HCl was completely removed, 0.1 mol/l borax solution (Sigma, Deisenhofen, Germany) at pH 9.0 was added for 20 min at RT. After two washes, cells were suspended in 1 ml PBS and 20 μl 7-AAD was added. Samples were kept in dark until flow-cytometry. At least 104 cells were analyzed per sample.

Fractions of cell-cycle phases were calculated automatically with “Cylchred,” version 1.0.2 (CytonetUK, England) software, which is based on algorithms of Ormerod et al. (1987).

In vivo

Animal experiments

Six-week-old thymus-aplastic female nude mice were obtained from Taconic, Denmark. Animals were kept in the sterile unit of the animal facility at the University of Luebeck. Principles of laboratory animal care were followed as well as regulations of the ethics committee. LX-1 tumor tissue of 1–2 mm3 was implanted into the flank of experimental animals. During the treatment period, mice were weighed and tumor volumes calculated with a caliper three times a week using the formula (x 2 y)/2 for an ellipsoid. Trofosfamide chemotherapy was initiated when tumors reached a volume of 200 mm3.

Table 1 shows the treatment groups, doses and application schedules. Conventional therapy consisted of two applications of i.p. trofosfamide whereas metronomic therapy consisted of more frequent drug administration at lower doses. Both groups—metronomic i.p. therapy (Metro-ip) and metronomic p.o. therapy, in which trofosfamide was administered daily in the drinking water (Metro-o)—were divided into two subgroups. One subgroup received therapy for 29 days, the other for 59 days. Each mouse receiving trofosfamide through the drinking water was estimated to drink 1.5 ml of water per 10 g of body weight per day (Man et al. 2002). When tumors reached a volume greater than 1,000 mm3, mice were sacrificed and the tumor was explanted.

Table 1.

Trofosfamide schedules for conventional and metronomic therapy groups. Group I served as untreated control

Single dose (mg/kg BW) Application Total dose over 4 weeks (mg/kg BW) Toxic deaths over 4 weeks
Group II (Conv-266.4) 266.4 Day 1+15 i.p. 532.8 5 of 8 mice
Group III (Conv-176.4) 176.4 Day 1+15 i.p. 322.8 0 of 8 mice
Group IV (Metro-ip) 44.4 3×/ week (4-week therapy) i.p. 532.8 0 of 8 mice
Group V (Metro-o) 19 Daily (in the drinking water, 4-week therapy) p.o. 532.8 0 of 8 mice
Open in a new tab

Non-tumor bearing mice (three animals each group) were treated with a single trofosfamide dose of 266 mg/kg BW i.p. or the metronomic p.o. therapy of 19 mg/kg BW per day continuously. Blood was obtained by puncture of the retrobulbar plexus and levels of 4-OH-metabolites were determined at 0, 10, 20, 30, 45, 60, 80, 120 min when conventionally treated and at 72 and 96 h in mice receiving oral metronomic therapy. Blood levels of 4-OH-metabolites were measured by high performance liquid chromatography as described by Brinker et al. (2002). The lower limit of detection was 2.5 μmol/l.

Immunostaining of tissues

Immunohistochemical staining of tumor sections was done on day 15 for the conventional therapy groups and on days 15 and 29 for the metronomic therapy groups. These sections were analyzed for MVD, relative growth fraction and apoptotic rate. All tumors were fixed with 1% formalin and embedded in paraffin. Tissue slices of 5 μm were cut.

The relative growth fraction was determined by immunohistochemical staining of fixed samples on slides using a primary, monoclonal, mouse-anti-human Ki-67 antibody (DakoCytomation, Carpinteria, CA, USA) for streptavidin–biotin labeling. For detection the “ChemMate Detection Kit 5001” (DakoCytomation, Carpinteria, CA, USA) was used. MVD was measured on fixed samples stained with a primary, monoclonal, rat-anti-mouse CD31 antibody (DPC Biermann, Bad Nauheim, Germany) applying the catalyzed signal amplification (CSA) method and using the “CSA-System Peroxidase Kit 1500” (DakoCytomation, Carpinteria, CA, USA). Substrate was activated by applying diaminobenzidine-chromogene and sections were counterstained with Mayer’s hematoxylin. Apoptotic cells were determined using the “Terminal-Deoxynucleotidyl-Transferase-mediated dUTP-Nick End Labeling” method as described by Gavrieli et al. (1992) with the “ApopDETEK kit” (DakoCytomation, Carpinteria, CA, USA). In a final step slices were counterstained with Nuclear Fast Red.

Computerized image analysis

Five highly vascularized tissue areas (hot spots) were randomly selected of each tumor section at low magnification and documented with a photomicroscope. For the determination of the relative growth fraction, positively brown-stained nuclei were related to the total cell number. CD31-positively marked vessels were counted in each neovascular hot spot, as well as apoptotic cells. Sections were quantified in a blinded fashion by three independent viewers using the computer imaging software Scion Image 4.0.2 (Scion Corporation, Frederick, MD, USA) and Adobe Photoshop 7.0 (Adobe Systems Inc., San Jose, CA, USA).

Statistical analysis

For statistical analysis, “Prism 4.0” (GraphPad Software, San Diego, CA, USA) and “Biostatistics 4.04” (S. Glantz, McGraw Hill, OH, USA) software were used. Results were tested with Kruskal–Wallis test and Mann–Whitney’s non-parametric test. IC50 were determined by sigmoidal dose–response curves, then Wilcoxon matched pairs test was used for assessing significance. Results were considered significant, if P-values were <0.05.

Results

In vitro

Cytotoxicity

For in vitro cytotoxicity assays LX-1 and HUVEC were exposed to 4OOH-trofosfamide at increasing concentrations between 2.5 and 40 μmol/l. Both tumor and endothelial cells were comparably sensitive to administered trofosfamide (Fig. 1). After 72 h of normoxic culture, the IC50 of HUVEC was significantly lower (P<0.05) than the IC50 of LX-1. Under normoxic conditions, IC50 values were 11.1 μmol/l for LX-1 cells and 2.3 μmol/l for HUVEC. Under hypoxic conditions, LX-1 and HUVEC show no significant differences. IC50 values were 16.9 μmol/l for LX and 15.4 μmol/l for HUVEC.

Fig. 1.

Fig. 1

Open in a new tab

In vitro cytotoxicity of 4OOH-trofosfamide as determined with the crystal-violet assay (mean ± SEM, n=8). a Normoxic conditions (20% O2) and b hypoxic conditions (1% O2)

Cell-cycle analysis

Cell-cycle analysis gave similar results for both cell-lines showing that LX-1 cells and HUVEC go through active phases of the cell cycle (Fig. 2). DNA synthesis slowed when oxygenization was reduced. The fraction of LX-1 cells in S-phase was larger than the fraction of HUVEC, while the fraction of LX-1 cells in S-phase showed a larger difference between norm and hypoxia than the fraction of HUVEC. After 72 h incubation 40.1% (normoxia) and 24.8% (hypoxia) of the lung cancer cells LX-1 were in S-phase, while the fraction of HUVEC was 21.3% (normoxia) and 16.3% (hypoxia).

Fig. 2.

Fig. 2

Open in a new tab

Cell cycle analysis showing the fraction of cells in synthesis phase under normoxic (20% O2) and hypoxic (1% O2) conditions (open square S-phase; filled square other cell-cycle phases; mean ± SD, n=6). a Lung carcinoma LX-1, b endothelial cells HUVEC

In vivo

Comparison of conventional vs. metronomic trofosfamide scheduling

Treatment groups, administered doses and therapy toxicity are illustrated in Table 1.

In the conventional therapy group (II) two of eight animals died shortly after the first application and another three animals after the second application. Although this dose was well tolerated in both metronomic schedules as a cumulative dose within 2 weeks, the dose was apparently too toxic given in a conventional drug schedule.

When trofosfamide was administered in a metronomic therapy regimen, LX-1 tumor-bearing nude mice showed a retardation of tumor growth as early as 10 days after initiation of treatment (Fig. 3b, c). Oral administration of trofosfamide was as effective as intra-peritoneal application. A continuous, low-dose administration of trofosfamide kept tumors at constant sizes for the total treatment period of 29 or 59 days. Metronomic trofosfamide proved to be well tolerated by mice in our experiments; none of the animals showed signs of toxic side effects or weight loss. When chemotherapy was finished, mice were kept alive for an observation period of 30 days. All tumors grew rapidly during the post-therapeutic period.

Fig. 3.

Fig. 3

Open in a new tab

Growth retardation of established tumors in thymus-aplastic nude (nu/nu) mice by conventional or metronomic administration of trofosfamide (n=8 per group). Data on all graphs were taken from cohesive experiments. a Conventional therapy was injected i.p. on day 1 and 15 (arrows). Control (no therapy), Conv-266 (266.4 mg/kg BW trofosfamide), Conv-176 (176.4 mg/kg BW trofosfamide). b Metronomic therapy by i.p.-injection of 44.4 mg/kg BW trofosfamide three times a week. Control (no therapy), Metro-ip (therapy over 29 or 59 days). c Metronomic therapy by daily oral administration of 19.0 mg/kg BW trofosfamide. Control (no therapy), Metro-o (therapy over 29 or 59 days)

In contrast to the mice receiving metronomic chemotherapy, conventional treatment with trofosfamide resulted in a rapid tumor progress immediately after the second administration of trofosfamide (Fig. 3a). A moderate progression of tumor size was already visible after the first injection. These mice had to be sacrificed due to rapidly increasing tumor volumes by day 25 after initiation of therapy at the latest. When animals were treated with the conventional bolus of 266.4 mg/kg—which equals the cumulative dose of the metronomically applied trofosfamide within the same period of 14 days—all mice suffered severe weight loss or died.

To confirm our hypothesis that the metronomic therapy schedule is attributable to an anti-angiogenic effect, we additionally compared the IC50 concentration of 4-OH-trofosfamide and the measured blood levels of the 4-OH-metabolites in vivo. High dose conventional treatment of 266 mg/kg BW i.p. in vivo leads to blood levels of 4-OH-metabolites of more than 15 times higher than the IC50 of 4-OH-trofosfamide on normoxic cultured LX-1 cells. A mean 4-OH-metabolite peak-concentration of 173.4 μmol/l at 20 min was found. When mice received metronomic doses of 19 mg/kg BW in the drinking water, 4-OH-metabolite levels achieved were at the limit of detection only of 2.5 μmol/l, which is equivalent to the IC50 of HUVECs in vitro but which is less than 25% of the IC50 of LX-1 cells in vitro.

Analysis of MVD, relative growth fraction and apoptosis

In all treatment groups on day 15 neither the relative growth fraction and the number of apoptotic tumor cells (Fig. 4) nor the MVD (Fig. 5) showed significant changes compared to pretherapeutic tumors. Also no significant difference in tumor MVD between pretherapeutic tumors sizing 200 mm3 and the large tumors of approximately 1,000 mm3 of the untreated control could be observed. In contrast, there was a significant reduction of up to 50% of vessels in the tumor sections analyzed after 29 days of metronomic trofosfamide treatment (Fig. 5). Unfortunately analysis of MVD at day 29 was not possible in the conventional treated animals since the mice had to be sacrificed due to rapid tumor growth.

Fig. 4.

Fig. 4

Open in a new tab

Immunohistochemical analysis of growth fraction (a) and apoptotic rate (b)

Fig. 5.

Fig. 5

Open in a new tab

Quantification of microvessel density in different therapy regimens. Immunostaining of tumor vasculature with CD31 antibody in treated and control mice as described below (400× magnification). Pretherapeutic tumors: explanted on day 0 with volumes of ∼200 mm3. Control (no treatment) tumor sections at a maximum volume of 1,000 mm3. Conv-176: tumor sections of conventional therapy group (176.4 mg/kg BW trofosfamide) on day 15. Metro-ip: tumor sections of metronomic i.p. treatment (44.4 mg/kg BW trofosfamide three times a week) on day 15 and day 29. Metro-o: tumor sections of oral metronomic treatment (19.0 mg/kg BW trofosfamide/daily) on day 15 and day 29

Discussion

The goal of our study was to investigate the efficacy of metronomic scheduling of trofosfamide on human lung carcinoma LX-1 in thymus-aplastic nude mice. The results show a remarkable anti-tumor effect of metronomically applied trofosfamide. Animals tolerated this therapy well and showed a suppression of tumor growth over the total treatment period. Immunohistochemical analysis confirmed that this effect is achieved in part by anti-angiogenic mechanisms. Significant reductions of microvessel densities occurred when animals were treated metronomically. Additionally we did not observe a reduction of treatment efficacy in the metronomic therapy groups over the treatment period of 59 days. In conventionally treated tumors, a prolonged growth inhibition could not be observed possibly due to acquired drug resistance of tumor cells.

Our in vitro experiments were conducted at normoxic conditions with 20% O2 and hypoxic conditions with 1% O2, taking typical extreme differences of tumor tissue oxygenization into account. Of course, this remains a static model reflecting only two extremes of oxygenization that are not transferable to the complex biology of a growing tumor. But the important role of hypoxia with its effects on cell proliferation and drug sensitivity is essential for understanding the effects of cytotoxic drugs in anti-angiogenic therapy.

While endothelial cells (HUVEC) were more sensitive to trofosfamide than lung carcinoma cells (LX-1) at norm-atmospheric conditions, the IC50 of HUVEC and LX-1 were similar at hypoxia. Analysis of cell-cycle phases showed a decreased fraction of S-phase cells under hypoxic conditions for tumor and endothelial cell-line, but both cell-lines were able to adapt and continuously proliferate. Shizukuda et al. (1999) reported that among several factors, the down-regulation of protein kinase C (PKC) seems to play a key role in the adaptation of HUVEC to hypoxia. PKC enhance endothelial cell survival by decreasing non-apoptotic cell death, which might explain the continuing proliferation of HUVEC in low-oxygenized atmospheres. It has been proposed by Miller et al. (2001) that anti-angiogenic treatment requires lower chemotherapeutic doses than those needed for an additional toxic effect on tumor cells. Bocci et al. (2002) found that even lowest doses of various conventional chemotherapeutic drugs caused a selective inhibition of endothelial cells when administered over long culture periods as long as 144 h. In our experiments the finding that endothelial cells are more sensitive than tumor cells could only be confirmed at normoxic conditions. At hypoxic conditions sensitivity to 4OOH-trofosfamide of endothelial and tumor cells was almost equal. In our experiments cells were incubated for only 72 h therefore we could not observe protraced effects on HUVEC, that were described by Bocci et al. at normoxic conditions only.

In accordance to the seminal results of Browder et al. (2000) who studied anti-angiogenic effects of metronomic cyclophosphamide on rodent tumors, we found similar effectiveness in therapy responses, retardation of tumor growth and toxicity for trofosfamide in a human tumor. We decided to choose a tumor of human origin considering the fact that rodent or murine tumors—as used in most other preclinical studies—exhibit different proliferation and sensitivity to chemotherapy which can partially limit clinical relevance. Trofosfamide which is approved in Germany and frequently prescribed in out-patient care setting, is less known than cyclophosphamide. Its lipophilic molecule structure provides high bioavailability, therefore it is well suitable for oral applications and it allows very long treatment periods even over several years (Wagner et al. 1997). For out-patient purposes oral application formulations are highly preferable over parenteral routes allowing frequent drug administration (Jahnke et al. 2004). Therefore trofosfamide was administered continuously in the drinking water in this study, although the exact drug dose ingested by the mice may be variable. The amount of the drug ingested was estimated by taking into account the amount of water a mouse drinks per day (3 ml for a 20-g mouse) (Man et al. 2002). This was compared to mice receiving a metronomic i.p. schedule in which exact dosing was possible. Previous date and clinical experience showed that the cumulative toxicity of trofosfamide is substantially lower than that of cyclophosphamide. Trofosfamide has also shown to provide good treatment responses in a variety of malignant diseases such as lymphomas, soft tissue sarcomas and other malignant solid tumors which include breast and ovarian cancer, as well as non-small cell lung carcinoma (Al-Batran et al. 2004; Gunsilius et al. 2001; Hartmann et al. 2003; Jahnke et al. 2004; Schmidt-Sandte et al. 1999; Wagner et al. 1997).

The results of our study provide evidence for trofosfamide to act as a potent anti-angiogenic agent when used in a metronomic schedule. While the conventional administration of trofosfamide resulted in a good response only after the first injection, tumors showed an almost exponential growth progression when it was administered for the second time. This suggests that conventional dosed chemotherapy predominantly bases on cytotoxic effects on tumor cells, favoring the selection of mutated tumor cells that acquire drug resistance while genomically stable endothelial cells are not affected (Yu et al. 2002). We show that metronomic trofosfamide therapies administered intra-peritoneally and orally showed excellent treatment responses with tumor growth retardation over the total treatment periods of either 29 or 59 days. During treatment mice were observed for chemotherapeutic side effects. When trofosfamide was administered conventionally with 266.4 mg/kg BW, which apparently exceeds the MTD of i.p. push injection but equals the cumulative dose of the metronomic treatment of 14 days (refer to Table 1), mice suffered from weight loss and were increasingly prone to infections. Two animals died after the first and three after the second application. When the same amount of trofosfamide was given as a metronomic schedule continuously over 2 weeks, no drug toxicity assessed by observing mice for body weight reduction was apparent.

Furthermore we demonstrate that by comparing vitro and in vivo data the metronomic trofosfamide dosages generate very low levels of 4-hydroxy-metabolites that do not cause relevant toxic effects on LX-1 tumor cells. From these findings we conclude that the responses on tumor growth that were observed were attributable to an angiogenic effect. Interestingly we observed a diminished sensitivity of endothelial cells under hypoxic conditions in vitro. This may in part explain that hypoxic tumor areas may be a restricting factor for the cytotoxicity on angiogenic cells in large tumors. Our hypothesis of a predominant anti-angiogenic effect of trofosfamide is furthermore supported by our finding that a significant reduction of MVD of up to 50% can be observed in the metronomically treated tumors after 29 days of treatments. This reduction was not observed after 15 days of metronomic therapy, when there were no significant changes in MVD. For conventional therapy arms we could not demonstrate day 29 MVD’s as a consequence of reaching maximum tolerable tumor volumes before that day. This unavoidably limits the comparison of both therapeutic concepts, which can be avoided in future by use of earlier angiogenesis markers like endothelial progenitor cells. The diminution of circulating endothelial progenitors by metronomic chemotherapy—as described by Bertolini—shows that there are not only direct cytotoxic effects on vascular endothelial cells that mediate anti-angiogenic activity but also on their progenitors (Bertolini et al. 2003). Bocci et al. (2003) showed furthermore that metronomic therapy causes an up-regulation of thrombospondin-1, a potent endogenous angiogenesis inhibitor that has anti-proliferative and proapoptotic effects. Its anti-angiogenic activity is mediated by CD36 transmembrane receptor binding. Our tumor sections were further analyzed for relative tumor growth fractions as well as apoptosis induction. However, neither a change in the relative growth fraction nor an increase of apoptotic cells could be noticed, regardless of therapy schedules.

We conclude that metronomic scheduling of trofosfamide chemotherapy offers remarkable and obvious advantages regarding tumor progression and drug sensitivity compared to conventional scheduling. Trofosfamide proved to be a potent anti-angiogenic agent with high activity on LX-1 lung carcinoma in our experiments.

Acknowledgments

We thank Monica Vollmert for her invaluable assistance and Dr Clemens Engels for supporting our immunohistochemical studies.

References

  1. Al-Batran SE, Atmaca A, Bert F, Jager D, Frisch C, Neumann A, Orth J, Knuth A, Jager E (2004) Dose escalation study for defining the maximum tolerated dose of continuous oral trofosfamide in pretreated patients with metastatic lung cancer. Onkologie 27:534–538 [DOI] [PubMed] [Google Scholar]
  2. Baguley BC, Holdaway KM, Thomsen LL, Zhuang L, Zwi LJ (1991) Inhibition of growth of colon 38 adenocarcinoma by vinblastine and colchicine: evidence for a vascular mechanism. Eur J Cancer 27:482–487 [DOI] [PubMed] [Google Scholar]
  3. Bertolini F, Paul S, Mancuso P, Monestiroli S, Gobbi A, Shaked Y, Kerbel RS (2003) Maximum tolerable dose and low-dose metronomic chemotherapy have opposite effects on the mobilization and viability of circulating endothelial progenitor cells. Cancer Res 63:4342–4346 [PubMed] [Google Scholar]
  4. Bocci G, Nicolaou KC, Kerbel RS (2002) Protracted low-dose effects on human endothelial cell proliferation and survival in vitro reveal a selective antiangiogenic window for various chemotherapeutic drugs. Cancer Res 62:6938–6943 [PubMed] [Google Scholar]
  5. Bocci G, Francia G, Man S, Lawler J, Kerbel RS (2003) Thrombospondin 1, a mediator of the antiangiogenic effects of low-dose metronomic chemotherapy. Proc Natl Acad Sci USA 100:12917–12922 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Boos J, Kupker F, Blaschke G, Jurgens H (1993) Trofosfamide metabolism in different species—ifosfamide is the predominant metabolite. Cancer Chemother Pharmacol 33:71–76 [DOI] [PubMed] [Google Scholar]
  7. Brinker A, Kisro J, Letsch C, Bruggemann SK, Wagner T (2002) New insights into the clinical pharmacokinetics of trofosfamide. Int J Clin Pharmacol Ther 40:376–381 [DOI] [PubMed] [Google Scholar]
  8. Browder T, Butterfield CE, Kraling BM, Shi B, Marshall B, O’Reilly MS, Folkman J (2000) Antiangiogenic scheduling of chemotherapy improves efficacy against experimental drug-resistant cancer. Cancer Res 60:1878–1886 [PubMed] [Google Scholar]
  9. Falkson G, Falkson HC (1978) Trofosfamide in the treatment of patients with cancer. A pilot trial. S Afr Med J 53:886–888 [PubMed] [Google Scholar]
  10. Folkman J (1971) Tumor angiogenesis: therapeutic implications. N Engl J Med 285:1182–1186 [DOI] [PubMed] [Google Scholar]
  11. Gavrieli Y, Sherman Y, Ben-Sasson SA (1992) Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 119:493–501 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gunsilius E, Gierlich T, Mross K, Gastl G, Unger C (2001) Palliative chemotherapy in pretreated patients with advanced cancer: oral trofosfamide is effective in ovarian carcinoma. Cancer Invest 19:808–811 [DOI] [PubMed] [Google Scholar]
  13. Hammers HJ, Kirchner H, Schlenke P (2000) Ultraviolet-induced detection of halogenated pyrimidines: simultaneous analysis of DNA replication and cellular markers. Cytometry 40:327–335 [DOI] [PubMed] [Google Scholar]
  14. Hartmann JT, Oechsle K, Mayer F, Kanz L, Bokemeyer C (2003) Phase II trial of trofosfamide in patients with advanced pretreated soft tissue sarcomas. Anticancer Res 23:1899–1901 [PubMed] [Google Scholar]
  15. Horvath V, Hartlapp JH, Hegewisch-Becker S, Feyerabend T, Fischer v. Weikersthal L, Illiger HJ, Jäger E, Peters SO, Reichardt P, Uthgenannt D, Weber K, Wagner T, Wiedemann G, Zschaber R (2000) Oral trofosfamide in elderly and/or heavily pretreated patients with metastatic lung cancer. Onkologie 23:341–344 [Google Scholar]
  16. Jaffe EA, Nachman RL, Becker CG, Minick CR (1973) Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 52:2745–2756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jahnke K, Bechrakis NE, Coupland SE, Schmittel A, Foerster MH, Fischer L, Thiel E, Korfel A (2004) Treatment of primary intraocular lymphoma with oral trofosfamide: report of two cases and review of the literature. Graefes Arch Clin Exp Ophthalmol 242(9):771–776 [DOI] [PubMed] [Google Scholar]
  18. Joussen AM, Kruse FE, Volcker HE, Kirchhof B (1999) Topical application of methotrexate for inhibition of corneal angiogenesis. Graefes Arch Clin Exp Ophthalmol 237:920–927 [DOI] [PubMed] [Google Scholar]
  19. Kerbel RS, Klement G, Pritchard KI, Kamen B (2002) Continuous low-dose anti-angiogenic/metronomic chemotherapy: from the research laboratory into the oncology clinic. Ann Oncol 13:12–15 [DOI] [PubMed] [Google Scholar]
  20. Klement G, Huang P, Mayer B, Green SK, Man S, Bohlen P, Hicklin D, Kerbel RS (2002) Differences in therapeutic indexes of combination metronomic chemotherapy and an anti-VEGFR-2 antibody in multidrug-resistant human breast cancer xenografts. Clin Cancer Res 8:221–232 [PubMed] [Google Scholar]
  21. Kueng W, Silber E, Eppenberger U (1989) Quantification of cells cultured on 96-well plates. Anal Biochem 182:16–19 [DOI] [PubMed] [Google Scholar]
  22. Man S, Bocci G, Francia G, Green SK, Jothy S, Hanahan D, Bohlen P, Hicklin DJ, Bergers G, Kerbel RS (2002) Antitumor effects in mice of low-dose (metronomic) cyclophosphamide administered continuously through the drinking water. Cancer Res 62:2731–2735 [PubMed] [Google Scholar]
  23. Miller KD, Sweeney CJ, Sledge GW Jr (2001) Redefining the target: chemotherapeutics as antiangiogenics. J Clin Oncol 19:1195–1206 [DOI] [PubMed] [Google Scholar]
  24. Ormerod MG, Payne AW, Watson JV (1987) Improved program for the analysis of DNA histograms. Cytometry 8:637–641 [DOI] [PubMed] [Google Scholar]
  25. Sasaki K, Adachi S, Yamamoto T, Murakami T, Tanaka K, Takahashi M (1988) Effects of denaturation with HCl on the immunological staining of bromodeoxyuridine incorporated into DNA. Cytometry 9:93–96 [DOI] [PubMed] [Google Scholar]
  26. Schmidt-Sandte W, Dageforde J, Klapdor R, Wagner T, Wiedemann GJ (1999) Trofosfamide in patients with pancreatic cancer. Anticancer Res 19:2485–2487 [PubMed] [Google Scholar]
  27. Shizukuda Y, Helisch A, Yokota R, Ware JA (1999) Downregulation of protein kinase cdelta activity enhances endothelial cell adaptation to hypoxia. Circulation 100:1909–1916 [DOI] [PubMed] [Google Scholar]
  28. Vogt T, Hafner C, Bross K, Bataille F, Jauch KW, Berand A, Landthaler M, Andreesen R, Reichle A (2003) Antiangiogenetic therapy with pioglitazone, rofecoxib, and metronomic trofosfamide in patients with advanced malignant vascular tumors. Cancer 98:2251–2256 [DOI] [PubMed] [Google Scholar]
  29. Wagner A, Hempel G, Boos J (1997) Trofosfamide: a review of its pharmacodynamic and pharmacokinetic properties and therapeutic potential in the oral treatment of cancer. Anticancer Drugs 8:419–431 [PubMed] [Google Scholar]
  30. Yu JL, Coomber BL, Kerbel RS (2002) A paradigm for therapy-induced microenvironmental changes in solid tumors leading to drug resistance. Differentiation 70:599–609 [DOI] [PubMed] [Google Scholar]

Articles from Journal of Cancer Research and Clinical Oncology are provided here courtesy of Springer

RESOURCES