Mol Cancer Ther. 2005;4:751-760
© 2005 American Association for Cancer Research
Matrix metalloproteinase–activated doxorubicin prodrugs inhibit HT1080 xenograft growth better than doxorubicin with less toxicity
Charles F. Albright,
Nilsa Graciani,
Wei Han,
Eddy Yue,
Ross Stein,
Zhihong Lai,
Melody Diamond,
Randine Dowling,
Lisa Grimminger,
Shu-Yun Zhang,
Davette Behrens,
Amy Musselman,
Robert Bruckner,
Mingzhu Zhang,
Xiang Jiang,
Daniel Hu,
Anne Higley,
Susan DiMeo,
Maria Rafalski,
Sandya Mandlekar,
Bruce Car,
Swamy Yeleswaram,
Andrew Stern,
Robert A. Copeland,
Andrew Combs,
Steve P. Seitz,
George L. Trainor,
Rebecca Taub,
Pearl Huang and
Allen Oliff
Bristol-Myers Squibb Pharmaceutical Research Institute, Wallingford, Connecticut
Requests for reprints: Charles F. Albright, Bristol-Myers Squibb Pharmaceutical Research Institute, 5 Research Parkway, Wallingford, CT 06492. Phone: 203-677-5937; Fax: 203-677-7569. E-mail: Charlie.Albright{at}BMS.com
 |
Abstract
|
|---|
Matrix metalloproteinase (MMP)–activated prodrugs were formed by coupling MMP-cleavable peptides to doxorubicin. The resulting conjugates were excellent in vitro substrates for MMP-2, -9, and -14. HT1080, a fibrosarcoma cell line, was used as a model system to test these prodrugs because these cells, like tumor stromal fibroblasts, expressed several MMPs. In cultured HT1080 cells, simple MMP-cleavable peptides were primarily metabolized by neprilysin, a membrane-bound metalloproteinase. MMP-selective metabolism in cultured HT1080 cells was obtained by designing conjugates that were good MMP substrates but poor neprilysin substrates. To determine how conjugates were metabolized in animals, MMP-selective conjugates were given to mice with HT1080 xenografts and the distribution of doxorubicin was determined. These studies showed that MMP-selective conjugates were preferentially metabolized in HT1080 xenografts, relative to heart and plasma, leading to 10-fold increases in the tumor/heart ratio of doxorubicin. The doxorubicin deposited by a MMP-selective prodrug, compound 6, was more effective than doxorubicin at reducing HT1080 xenograft growth. In particular, compound 6 cured 8 of 10 mice with HT1080 xenografts at doses below the maximum tolerated dose, whereas doxorubicin cured 2 of 20 mice at its maximum tolerated dose. Compound 6 was less toxic than doxorubicin at this efficacious dose because mice treated with compound 6 had no detectable changes in body weight or reticulocytes, a marker for marrow toxicity. Hence, MMP-activated doxorubicin prodrugs have a much higher therapeutic index than doxorubicin using HT1080 xenografts as a preclinical model.
Key Words: Doxorubicin • prodrugs • matrix metalloproteinases
|
Introduction
|
|---|
Doxorubicin is an anthracycline natural product that is frequently used to treat breast cancer, liver cancer, soft-tissue sarcomas, and non-Hodgkin's lymphoma (reviewed in ref. 1). Doxorubicin kills tumor cells by causing topoisomerase II–stimulated DNA strand breaks and potentially other mechanisms (reviewed in ref. 2). Like other cytotoxic drugs, doxorubicin doses are limited by unwanted toxicity to nontumor tissues. Myelosuppression is the dose-limiting toxicity for doxorubicin, although stomatitis, mucositis, and alopecia are also frequently observed. In addition to these typical chemotherapeutic toxicities, doxorubicin causes cardiomyopathy that depends on the cumulative dose of doxorubicin.
Matrix metalloproteinases (MMP) are a family of >20 enzymes (reviewed in ref. 3). MMPs are synthesized as inactive proenzymes that become activated by proteolysis. For instance, the secreted enzyme pro-MMP-2 is proteolyzed and activated by the membrane-anchored MMP-14 (4). Once activated, MMPs can cleave a variety of extracellular matrix proteins, such as collagen, laminin, fibronectin, and elastin, which are potentially important for tumor angiogenesis, invasion, and metastasis. Additionally, MMPs cleave a variety of other proteins, such as growth factor receptors and cell adhesion molecules, which may be important for tumor growth and survival (reviewed in refs. 5, 6). MMPs can be inactivated by tissue inhibitors of MMPs, a family of four proteins, which bind tightly to MMPs.
Increasing the therapeutic index of doxorubicin could benefit cancer treatment by allowing greater concentrations of doxorubicin in tumors and thereby potentially greater tumor growth inhibition. Alternatively, increasing the therapeutic index of doxorubicin could allow reduced toxicities and thereby permit the use of other cytotoxics that would be prevented because of overlapping toxicities. One way to improve the therapeutic index of doxorubicin is to create prodrugs that are selectively activated in the tumor environment. MMPs are attractive enzymes to activate prodrugs in the tumor environment because MMPs are intimately connected with tumorigenesis. In particular, preclinical studies implicate MMPs in tumor progression (reviewed in ref. 6). For instance, overexpression of MMP-14 in mouse mammary tissue causes mammary tumors (7), whereas loss of MMP-2 or -9 reduced the lung colonization by B16-LBL6 melanoma cells and Lewis lung carcinoma cells (8, 9). Additionally, MMP expression in human tumors frequently correlates with disease progression (reviewed in ref. 6). For instance, one study found MMP-2 and -14 expression in 90% of breast carcinomas and a strong correlation between tumor cell membrane staining and lymph node metastases (10). Although there are strong connections between MMPs and tumor progression, published studies do not determine the absolute or relative MMP activity in the tumor environment. Such information is clearly critical to the design of MMP-activated prodrugs because the MMP enzyme activity must be sufficient to activate efficacious levels of prodrug.
Preclinical testing of MMP-activated prodrugs is complicated by differences between mouse and human tumors. Many human tumors, including breast cancers, are composed of roughly equal numbers of tumor cells and stromal cells, including stromal fibroblasts, inflammatory cells, and endothelial cells (reviewed in ref. 11). The interplay of these cell types is thought to be crucial for MMP activation in that different cell types express and activate specific MMPs (reviewed in ref. 6). For instance, stromal fibroblasts express MMP-2, yet tumor cells are the source of MMP-14 that convert pro-MMP-2 to active MMP-2. Unfortunately, mouse tumor models do not accurately reflect the cell types observed in human tumors because both xenografts and transgenic tumor models are composed primarily of rapidly growing tumor cells with relatively few stromal cells.
This study discovered MMP-activated doxorubicin prodrugs that had a greater therapeutic index than doxorubicin using HT1080 cells as a model system. In the absence of information about enzyme activity levels, prodrugs were identified that were efficiently cleaved by MMP-2, -9, and -14 because these enzymes were frequently overexpressed in human tumors, affected tumor progression in preclinical models, and were very active in vitro. Because these MMPs may have insufficient activity for prodrug activation in vivo, prodrug design was further constrained so that prodrugs were likely be activated by other MMPs, with the exception of MMP-11 (12), based on the known MMP substrate specificities (13). HT1080 cells were used as a model system because these cells grow as xenografts, are sensitive to doxorubicin, and express several different MMPs (14, 15). Furthermore, HT1080 cells are derived from a fibrosarcoma and were, therefore, hoped to mimic stromal fibroblasts, a major constituent of many human tumors. Using this strategy, prodrugs were identified that were excellent in vitro substrates for MMP-2, -9, and -14. Optimization of conjugate structure led to prodrugs that were selectively metabolized by MMPs in HT1080 cultures and preferentially deposited doxorubicin in HT1080 xenografts relative to heart tissue. The preferential doxorubicin deposition led to an improved therapeutic index, relative to doxorubicin, as shown by improved HT1080 xenograft growth inhibition with reduced toxicity.
|
Materials and Methods
|
|---|
Compound Synthesis
All peptides were synthesized following the same protocol. An example synthesis of compound 1 is provided. The peptide acid was synthesized on the solid phase starting with commercially available Fmoc-Leu-Wang resin (0.40 g, 0.25 mmol, Advance Chemtech, Louisville, KY). The synthesis was done on an ABI 433A peptide synthesizer (Applied Biosystems, Foster City, CA) using standard Fmoc protocols. The completed peptide on resin was N-acetylated with acetic anhydride. The desired peptide was cleaved from the resin with 90% trifluoroacetic acid in water for 2 hours. After solvent removal, the peptide was dissolved in H2O:CH3CN and freeze dried. Product was confirmed by mass spectrometry (ES m/e calculated for C21H36N4O6 [M-H]–, 439.54; found, 439.3). Analytic high-performance liquid chromatography on a Metachem Monochrome C18 reverse-phase column (50 x 4.6 mm, MetaChem Technologies, Torrance, CA) showed crude peptide to be 85% pure. To this intermediate (19.9 mg, 40 µmol) dissolved in dimethylformamide (0.2 mL) in a small amber vial was added benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate (20.8 mg, 40 µmol). Doxorubicin hydrochloride (18.6 mg, 32 µmol, Aldrich, Milwaukee, WI) was added as a suspension in dimethylformamide (0.1 mL) followed by diisopropylethylamine (0.0139 mL, 80 µmol). The reaction was stirred for 2 hours. Solvent was removed under vacuo. The sample was dissolved in H2O:CH3CN and purified using a Dynamax C18 reverse-phase column (41.4 x 250 mm, Varian, Palo Alto, CA) with a linear gradient from 30% to 50% acetonitrile, 0.05% ammonium acetate over 20 minutes with a flow rate of 45 mL/min. Fractions were pooled and freeze dried to afford the purified peptide-doxorubicin conjugate (ES m/e calculated for C48H63N5O16 [M-H]–, 965.06; found, 964.6). Caution: Doxorubicin is a highly toxic compound. Appropriate measures should be taken to avoid exposure to the compound.
Enzyme Assays
Doxorubicin prodrugs were incubated with MMP-2, -9, or -14 or neprilysin to compare the competency of these conjugates as substrates for these enzymes. Recombinant MMP-2, -9, and -14 were purified from HT1080 fibrosarcoma cells, insect cells, and Escherichia coli, respectively, as described previously (12, 16, 17). MMP-2 and -9 were activated with 4-aminophenylmercuric acetate for 2 hours at 37°C. Unreacted 4-aminophenylmercuric acetate was removed using Bio-Spin 6 columns (Bio-Rad, Hercules, CA). Neprilysin was purified from rat kidney as described (18). Conjugates (1 µmol/L) were incubated with either 10 nmol/L MMP-2, 2 nmol/L MMP-9, or 5 nmol/L MMP-14 in 50 mmol/L HEPES (pH 7.5), 10 mmol/L CaCl2, 0.1% Brij-35 at 37°C. At various times up to 60 minutes, an aliquot of the reaction was removed and injected into a Novapak C18 reverse-phase high-performance liquid chromatography column (Waters, Milford, MA). The reaction mixture was eluted with a 27% to 63% acetonitrile gradient containing 0.1% trifluoroacetic acid. Doxorubicin-containing compounds were detected and quantified using the intrinsic fluorescence of doxorubicin (excitation wavelength, 480 nm; emission wavelength, 580 nm). Alternately, conjugates (1 µmol/L) were incubated with 10 nmol/L neprilysin in 50 mmol/L Tris-HCl (pH 7.5), 10 mmol/L CaCl2, 0.1% Brij-35 and processed as above. The substrate peak area was used to determine the kcat/Km for each conjugate under first-order conditions (substrate << Km):
 |
where S0 is the initial substrate concentration, St is the substrate concentration at various times, t is time, and E is enzyme concentration (19).
Prodrug Metabolism in Cultured HT1080 Cells
To measure the rate of prodrug metabolism in cell culture, actively growing HT1080 fibrosarcoma cells (American Type Culture Collection, Manassas, VA) were plated at 60% confluence. Two hours after plating, the cell medium (DMEM, 10% fetal bovine serum) was removed and replaced with DMEM containing 0.1% bovine serum albumin and 40 nmol/L phorbol 12-myristate 13-acetate. Phorbol 12-myristate 13-acetate was added to increase the expression of MMP-9 and conversion of pro-MMP-2 to active MMP-2 (20). After 16 hours, doxorubicin prodrug (500 nmol/L) was added to the medium. Aliquots of medium (0.1 mL) were removed at the indicated times, 0.3 mL acetonitrile was added to each aliquot, and the sample was mixed and then centrifuged for 5 minutes. Solvent was evaporated from clarified supernatant (0.3 mL) using a nitrogen stream. Samples were then vortexed with 0.06 mL acetonitrile. Buffer A [0.06 mL; 14 mmol/L sodium phosphate, 0.5 mmol/L triethylamine (pH 4.2)] was then added, vortexed, and centrifuged. Supernatant (0.06 mL) was mixed with 0.06 mL buffer A and 0.1 mL was injected onto a Novapak C18 column at a flow rate of 1 mL/min and eluted with a linear gradient from 75% buffer A, 25% acetonitrile to 50% buffer A, 50% acetonitrile over 12 minutes. Control experiments showed that doxorubicin and doxorubicin conjugates were recovered with 
80% yield using this method. Doxorubicin-containing compounds were detected based on their intrinsic fluorescence as above. To determine the amount of metabolism that was due to MMPs, cultures were incubated with 0.15 mmol/L marimastat, a broad-spectrum MMP inhibitor (21). This concentration of marimastat was sufficient to inhibit MMPs based on in vitro assays and the conversion of pro-MMP-2 to MMP-2 in HT1080 cultures (data not shown). To determine the fraction of metabolism due to neprilysin, cultures were incubated 100 nmol/L phosphoramidon.
Prodrug Metabolism in Mice
HT1080 tumors were transplanted into naive mice using trocar needles. Dry conjugate powders were dissolved in N,N-dimethylacetamide and diluted with 5% dextrose to yield a final dosing solution of 10% N,N-dimethylacetamide, 4.5% dextrose. Mice were then given the desired dose using 0.1 mL dosing solution. Drugs were given i.v. into the tail vein. At the indicated times, three mice per time point were anesthetized with CO2. Blood (1 mL) was collected by cardiac puncture in a syringe containing 0.1 mL sodium citrate to prevent coagulation and cells were removed by centrifugation. The resulting plasma was transferred to another tube and frozen in liquid nitrogen. Tumor and heart tissue were removed, frozen in liquid nitrogen, and stored at –80°C. For analysis, tissues were thawed on ice, weighed, and minced with scissors, and cold, citrated mouse plasma was added (Cocalico Biological, Reamstown, PA). Plasma was added at 5 mL/g for tumor tissue and 10 mL/g for heart tissue. Slurries were homogenized for 1 minute and 0.5 mL was transferred to an Eppendorf tube. Silver nitrate (0.1 mL, 33%) was added to the homogenate to improve the recovery of doxorubicin and the mixture was vortexed briefly. Acetonitrile (0.05 mL) was then added, vortexed briefly, and mixed for 15 minutes. This mixture was centrifuged for 5 minutes. The supernatant was transferred to another tube, dried under a nitrogen stream, and stored at –80°C. Doxorubicin-containing compounds were separated by reverse-phase high-performance liquid chromatography as above and quantified using a doxorubicin standard curve. Control experiments showed that this method extracted 80% doxorubicin-containing compounds and that metabolism of doxorubicin-containing compounds during analysis was negligible.
Inhibition of HT1080 Xenograft Growth
HT1080 tumors were transplanted into naive mice. After 5 to 7 days, tumors had reached a median size of 75 to 100 mm3. Mice were then grouped into cohorts of 10 mice such that each cohort had the same median tumor size and a similar tumor size distribution. Treatment was then initiated as described in Results and tumors were measured every 3 days with calipers.
Reticulocyte Measurements
Three mice per condition were anesthetized with CO2 3 days after the last treatment because control experiments showed that this time corresponded to the nadir of the reticulocyte decrease. Blood (0.5 mL) was collected by cardiac puncture in a syringe containing EDTA to prevent coagulation. RBC and reticulocyte concentrations were determined the same day using an Advia 120 Hematology Analyzer.
Lethal Dose Determination
Increasing doses of drugs were given i.v. to five mice per dose. Animals were observed for 30 days for signs of severe toxicity and euthanized if required according to Dupont Pharmaceuticals Animal Care and Use Committee guidelines.
|
Results
|
|---|
Design of MMP-Activated Doxorubicin Prodrugs
Doxorubicin prodrugs were formed by linking the COOH terminus of a peptide to the amino group of doxorubicin (Fig. 1). The resulting prodrug is not cytotoxic because the peptide prevents the prodrug from entering cells (22). MMP cleavage occurs in the middle of the illustrated hexapeptide generating a tripeptide doxorubicin conjugate (Fig. 1). The tripeptide doxorubicin conjugate must be further cleaved by extracellular proteases to form Leu-Dox before doxorubicin-containing compounds can efficiently penetrate cells. Leu-Dox may be converted to doxorubicin either inside or outside cells. Leucine was chosen as the COOH-terminal residue because Leu-Dox was more efficiently converted to doxorubicin than other amino acid doxorubicin conjugates (23) and leucine allowed efficient MMP-2, -9, and -14 cleavage (data not shown). The peptide was capped to prevent aminopeptidase degradation and increase solubility. Sufficient aqueous solubility was required to allow the i.v. administration of compounds to humans. Although the necessary solubility would depend on clinical usage, a target solubility of 1 mg/mL was chosen to guide compound design.
View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Structure of doxorubicin (A) and compound 6, a representative prodrug, with expected enzymatic cleavage products (B). The MMP cleavage site is indicated by . Residues on the NH2-terminal side of the cleavage site are designated P1, P2, etc., whereas residues on the COOH-terminal side of the cleavage site are designated P1', P2', etc. See Results for further description.
|
|
Prodrug Cleavage by MMP-2, -9, and -14 In vitro
To determine the effect of prime-side length on MMP cleavage rate, peptides of various lengths were attached to doxorubicin and the catalytic efficiencies for cleavage by MMP-2, -9, and -14 were determined (Table 1). For this analysis, simple collagen-like peptide sequences (PLGL) were used because most MMPs cleave peptides containing this sequence (reviewed in ref. 13). A conjugate with the one prime-side residue, compound 1, was not detectably cleaved by MMP-2 or -9. A conjugate with two prime-side residues, compound 2, was efficiently cleaved by MMP-9, modestly cleaved by MMP-2, and poorly cleaved by MMP-14. Conjugates with three prime-side residues, compound 3, or four prime-side residues, compound 4, were cleaved efficiently by MMP-2, -9, and -14. The absolute kcat/Km values for efficiently cleaved conjugates, such as compounds 3 and 4, were close to values for the most efficient MMP substrates. In particular, kcat/Km values for the best reported fluorogenic peptide substrates for MMP-2, -9, and -14 are 629, 87, and 1,590 mmol/L–1 s–1, respectively (24–26).
Metabolism of Conjugates in Cultured HT1080 Cells
The metabolism rate of conjugates with two or more prime-side residues in cultured HT1080 cells was determined because in vitro enzymatic digestions showed that these conjugates were substrates for MMP-2, -9, or -14. For this analysis, conjugates were incubated with cultured HT1080 cells, aliquots of culture supernatant were removed after various times, and the composition of doxorubicin-containing compounds was determined. Using compound 2 as an example, the parental conjugates disappeared linearly with time for several hours and the major metabolite was L-Dox (Fig. 2A). Other metabolites, such as LL-Dox, which would be formed by MMP cleavage of compound 2, were only detected in trace amounts. To determine why these expected metabolites were not detected at higher levels, compounds were cleaved by MMP-2 in vitro and then added to cultured HT1080 cells. Analysis of these culture supernatants showed that post-MMP cleavage metabolites were rapidly converted to L-Dox (data not shown). The kinetics of conversion of these post-MMP cleavage metabolites to L-Dox were consistent with the trace amounts of these metabolites detected following incubation of the parental compounds. To allow comparisons between compounds, the fraction processed was defined as the fraction of compounds present in the culture supernatant that could result from MMP cleavage. The fraction processed per hour was calculated by fitting a straight line to the fraction processed values measured during the first 6 hours after incubation of the compound with cultured HT1080 cells. Although this method is considered the most rigorous, very similar values would be obtained by simply considering the formation of L-Dox. Analysis of compounds 2 to 4 showed a dramatic effect of prime-side length on processing in HT1080 cultures; the fraction processed per hour was 0.02, 0.065, and 0.3 for these compounds that contained 2, 3, or 4 prime-side residues, respectively (Fig. 2B). None of the compounds in Fig. 2B were detectably degraded in the absence of cultured HT1080 cells (data not shown).
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Metabolism rates of selected prodrugs by cultured HT1080 cells. A, amount of L-Dox and doxorubicin formed from compound 2 in cultured HT1080 cells. Amounts are represented as the fraction of all doxorubicin-containing compounds detected. The total amount of doxorubicin-containing compounds differed by <5% at different times. B, non-MMP and MMP processing of several conjugates in cultured HT1080 cells. The fraction of metabolism due to MMPs was determined in reactions with 150 nmol/L marimastat. C, effect of phosphoramidon (Phos.) and marimastat on compound 4 processing in cultured HT1080 cells.
|
|
To determine the amount of the metabolism that resulted from MMP cleavage, reactions were conducted with marimastat, a broad-spectrum MMP inhibitor. This analysis showed that most of the cleavage of compounds 2 and 4 were not due to MMPs (Fig. 2B). In the case of compound 3, approximately one half of the metabolism resulted from MMPs. Similar results were obtained with other small-molecule MMP inhibitors (data not shown).
To identify the putative non-MMP-metabolizing enzyme, the effect of other protease inhibitors on compound 4 metabolism was tested. These studies revealed that 100 nmol/L phosphoramidon greatly reduced the metabolism of compound 4 (Fig. 2C). At this concentration, phosphoramidon is specific for neprilysin, a cell surface protease (27). Other data supported the hypothesis that neprilysin was the major non-MMP-metabolizing enzyme for conjugates in cultured HT1080 cells. First, HT1080 cells expressed neprilysin based on Western blotting (data not shown). Second, the non-MMP metabolism rate of conjugates correlated well with the in vitro catalytic efficiency of neprilysin cleavage (Table 1; Fig. 2B; data not shown). Third, other cell lines that expressed neprilysin had non-MMP metabolism rates that correlated with their levels of neprilysin expression (data not shown).
Because neprilysin is expressed outside tumor tissue with high levels in kidneys (27) and would, therefore, lead to nontumor activation of prodrugs, compounds were identified that were good substrates for MMP-2, -9, and -14 but were poor substrates for neprilysin. The data for two such compounds, compounds 5 and 6, are summarized in Table 1. In the case of compound 5, MMP selectivity over neprilysin was first achieved by substituting ornithine for tyrosine in the P2' position. The ornithine substitution had the added benefit of increasing solubility to 0.3 mg/mL. MMP selectivity over neprilysin with good solubility was later obtained in compounds, such as compound 6, which did not contain any positively charged residues. Consistent with the neprilysin hypothesis, the MMP-selective compounds, compounds 5 and 6, were primarily metabolized by MMPs in cultured HT1080 cells (Fig. 2B).
Analysis of conjugate metabolism in cultured HT1080 cells also showed that there was little correlation between the enzymatic cleavage of conjugates by MMP-2, -9, or -14 and the cleavage of conjugates by cultured HT1080 cells (Table 1; Fig. 2B; data not shown). Potential reasons for this lack of correlation will be discussed later.
Compound 5 Preferentially Accumulates Doxorubicin in Tumor Xenografts Relative to Heart Tissue
To determine if the metabolism of conjugates by cultured HT1080 cells was sufficient to lead to the preferential accumulation of doxorubicin in tumor tissue, mice with HT1080 xenografts were injected with 1.4 µmol/kg compound 5, L-Dox, or doxorubicin. At various times after injection, tumor and heart tissue were analyzed for doxorubicin levels. Heart tissue was chosen as a reference tissue because there were published data comparing tumor and heart levels of doxorubicin and L-Dox. Both doxorubicin and L-Dox administration resulted in more doxorubicin in heart tissue than HT1080 xenografts (Fig. 3A and B), consistent with published results (28–31). In contrast, compound 5 administration led to more doxorubicin in HT080 tissue than heart (Fig. 3C). One way to quantify these differences is to integrate the area under the doxorubicin time-concentration curves. This integration showed that compound 5 had a 15-fold better tumor/heart doxorubicin ratio than doxorubicin (Table 2). This analysis also showed that the tumor doxorubicin levels from compound 5 were approximately one third those from doxorubicin at equimolar doses.
To gain further insight into compound 5 metabolism, the concentration-time profiles for compound 5 and its metabolites in plasma, tumor, and heart were determined. As in cultured HT1080 cells, the major doxorubicin-containing compounds in the tissues following administration of compound 5 were compound 5, L-Dox, and doxorubicin (Fig. 4). Compound 5 disappeared rapidly from the plasma compartment. Whereas L-Dox levels were measurable in plasma, doxorubicin levels in plasma were below detectable limits. In heart tissue, L-Dox levels were also less than the levels of compound 5 (Fig. 4B). Unlike plasma, heart tissue accumulated doxorubicin presumably by metabolism of L-Dox. Unlike plasma and heart, L-Dox levels in HT1080 tissue exceeded compound 5 levels. Because neither compound 5 nor L-Dox preferentially accumulated in HT1080 tissue (Fig. 4; data not shown), these findings provide strong evidence that compound 5 was preferentially metabolized in HT1080 tissue relative to heart and plasma. As a final test for tumor-specific metabolism, conjugates that could not be cleaved by MMP-2, -9, or -14 were injected into mice. These conjugates did not lead to detectable levels of doxorubicin in HT1080 xenografts (data not shown).
Some Conjugates Cause Acute Toxicity with Administration at High Doses
When compound 5 was given at higher doses that were compatible with xenograft efficacy studies, such as 14 mmol/kg, an unexpected acute toxicity was observed. In particular, mice became flushed due to redistribution of blood to the extremities. This redistribution also led to a rapid decrease in blood pressure and death within minutes after compound injection. Administration of an equivalent amount of doxorubicin did not cause detectable flushing or acute toxicity.
Several factors supported the hypothesis that conjugates caused mast cell degranulation leading to the release of histamine and other vasoactive peptides. First, the flushing and reduced blood pressure were consistent with mast cell degranulation. Second, preadministration of terfenadine, an antihistamine, partially prevented acute toxicity. Third, acute toxicity was prevented by administration of a nonlethal dose of compound 5 showing desensitization consistent with mast cell degranulation. Fourth, doxorubicin was reported to cause mast cell degranulation in mice, rats, and dogs when given at very high doses (32–34). Based on these data, we hypothesize that some non-doxorubicin structures cooperate with doxorubicin to cause mast cell degranulation.
Identification of Soluble, Selective Conjugates without Acute Toxicity
Based on these findings, compounds were screened for those that were specific for MMP cleavage in vitro and cultured cells yet lacked the acute toxicity observed with compound 5. This screening effort revealed that all compounds with positively charged residues caused acute toxicity at 14 µmol/kg. In contrast, compounds, such as compound 6, did not cause acute toxicity at doses up to 112 µmol/kg, the maximum dose that could be given because of compound solubility limitations. Like compound 5, compound 6 was selectively cleaved by MMP-2, -9, and -14, relative to neprilysin (Table 1), was primarily metabolized by MMPs in cultured HT1080 cells (Fig. 2B), and had good aqueous solubility (Table 1). Furthermore, compound 6 led to the selective deposition of doxorubicin in HT1080 xenografts, relative to heart tissue, with a tumor/heart doxorubicin ratio of 12 and doxorubicin HT1080 levels approximately one half those from an equimolar dose of doxorubicin at 14 µmol/kg.
Compound 6 Is More Efficacious Than Doxorubicin with Less Toxicity
To determine if the doxorubicin deposited by conjugates was effective at reducing tumor growth, mice were implanted with HT1080 xenografts, grouped into cohorts of 10 mice with a similar median tumor size and tumor size distribution, and treated with doxorubicin or compound 6. These studies showed that 14 µmol/kg compound 6 q4dx3 was slightly less effective than 7 µmol/kg doxorubicin q4dx3, whereas 14 µmol/kg compound 6 qdx10 was more effective than the maximum tolerated levels of doxorubicin (Fig. 5). In particular, 8 of 10 mice treated with the higher dose of compound 6 were cured of their HT1080 xenografts as defined by lack of detectable tumors 75 days after implantation. In contrast, only 2 of 10 and 0 of 10 mice treated with doxorubicin at 14 µmol/kg q4dx3 and 6 µmol/kg qdx10, respectively, were cured. At these doxorubicin doses, mice lost an average of 10% body weight, which was operationally defined as the maximum tolerated dose. Mice treated with compound 6 had no detectable weight loss.
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Compound 6 inhibits the growth of HT1080 xenografts better than doxorubicin. See legend for doses and schedules. Day 0, time of tumor implantation. Arrows, beginning and end of treatment. A tumor volume of 1 mm3 was assigned to tumors that were undetectable.
|
|
The observed tumor growth inhibition of compound 6 was in reasonable agreement with that expected from the levels of doxorubicin deposited in tumors. In particular, 14 µmol/kg compound 6 deposited approximately one half the doxorubicin concentration in tumors as 14 µmol/kg doxorubicin. If the doxorubicin deposited by compound 6 was equipotent with that deposited by doxorubicin, then 14 µmol/kg compound 6 q4dx3 should cause a similar growth inhibition to 7 µmol/kg doxorubicin q4dx3 and 14 µmol/kg compound 6 qdx10 should be more effective than 14 µmol/kg doxorubicin q4dx3. As shown in Fig. 5, tumor growth inhibition by compound 6 is consistent with these predictions.
Mice treated with compound 6 did not show any overt signs of toxicity, including weight loss or changes in coat texture. To investigate compound 6 toxicity in detail, mice were given increasing doses of either compound 6 or doxorubicin. This experiment showed that the LD50 for doxorubicin was 42 µmol/kg, whereas administration of compound 6 at 112 µmol/kg did not cause any detectable adverse effects. As a measure of marrow toxicity, reticulocytes were quantified. Reticulocytes are short-lived precursors of RBC in blood that are easily quantified (35). Doxorubicin-treated mice showed a dramatic decrease in reticulocytes with an ED50 of 2.5 µmol/kg (Fig. 6). In contrast, mice given 14 µmol/kg compound 6 qdx10 had reticulocyte levels (average, 5.6% RBC; SD, 3.3% RBC) that were not statistically different from vehicle-treated mice (average, 3.0% RBC; SD, 1.2% RBC). Taken together, these results show that compound 6 is more effective than doxorubicin at reducing the growth of HT1080 xenografts with significantly less toxicity than doxorubicin.
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Effect of doxorubicin dose on reticulocyte counts. Mice were given the indicated dose of doxorubicin daily for 10 d. Three days after the last dose, reticulocyte counts were determined. Bars, SD.
|
|
|
Discussion
|
|---|
In this study, MMP-activated doxorubicin prodrugs were discovered that had a higher therapeutic index than doxorubicin when HT1080 cells were used as a preclinical model. To achieve this increased therapeutic index, prodrugs were identified that were efficiently cleaved in vitro by MMP-2, -9, and -14, poorly cleaved by neprilysin, an off-target metalloprotease, and lacked positively charged residues to avoid acute toxicity. Such conjugates were specifically metabolized by MMPs in cultured HT1080 cells and preferentially deposited doxorubicin in HT1080 xenografts relative to heart. The doxorubicin deposited by compound 6 was effective at reducing the growth of HT1080 xenografts, whereas the reduced doxorubicin deposition in nontumor tissues led to an increased maximum tolerated dose and reduced marrow toxicity.
The results of this study with MMP-activated doxorubicin prodrugs are similar to those from previous studies with extracellular enzyme-activated doxorubicin prodrugs. In particular, doxorubicin prodrugs activated by prostate-specific antigen (36, 37), neprilysin (38), and plasmin (39, 40) have been studied in mice. For the prostate-specific antigen–activated prodrug L-377,202, the doxorubicin concentration in a prostate cell xenograft was increased 2-fold and the heart doxorubicin concentration was reduced 2-fold leading to a 4-fold increased tumor/heart doxorubicin ratio (41). For the neprilysin-activated prodrug CPI-0004Na, tumor doxorubicin concentrations were increased 2-fold and heart doxorubicin concentrations were decreased 10-fold, leading to a 20-fold increase in the doxorubicin tumor/heart ratio (38). In both of these cases, the prodrugs had a higher maximum tolerated dose than doxorubicin and inhibited xenograft growth better than doxorubicin (37–39). Very similar results with respect to improvements in the maximum tolerated dose and inhibition of xenograft growth were obtained with a different prostate-specific antigen–activated prodrug (36) and two plasmin-activated prodrugs (40).
Although preclinical studies with tumor-activated prodrugs show that it is possible to increase the therapeutic index of doxorubicin in mice, it is unclear whether the benefits observed in mice will translate to humans. The first human trial with L-377,202 were recently completed and suggest that some of the improvements observed in mice will be translated to humans (42). In the case of MMP-activated prodrugs, this issue is particularly difficult to address because mouse tumors are poor models for human tumors with respect to MMP expression and activity. In particular, MMPs are frequently expressed and activated by different cell types within human tumors. For instance, MMP-2 is primarily expressed by stromal fibroblasts, yet pro-MMP-2 is activated by MMP-14 that is present on tumor cells. Because xenografts and mouse transgenic tumor models contain relatively few stromal fibroblasts, it is difficult to mimic the MMP activity of human tumors in mice. To deal with this complexity, our studies used HT1080 cells, a fibrosarcoma cell line. By choosing a fibroblast cell tumor line that expressed MMP-1, -2, -3, -7, -9, -14, -15, -16, and -18 (14, 15), we hoped to better represent the MMP composition in human tumors than with more conventional epithelial tumor cell lines. Although HT1080 cells are a rational choice, inhibition of HT1080 growth does not remove many of the unknowns in translating these preclinical findings into the clinic.
If MMP-activated doxorubicin prodrugs have an increased therapeutic index relative to doxorubicin in humans, then there are two major ways to exploit this increased therapeutic index. Ideally, one would like to increase the efficacy of doxorubicin by increasing the tumor concentration of doxorubicin with prodrug administration. Although such a strategy clearly works in HT1080 xenografts, it is unclear whether this strategy will be successful in human tumors. In particular, several studies have investigated the use of high-dose chemotherapy where autologous hematopoietic progenitor cell transplantation is used to allow chemotherapeutic doses that would otherwise be myeloablative. Whereas overall survival was similar between high-dose chemotherapy and conventional chemotherapy in four trials for patients with metastatic breast cancer, six of seven trials found increased disease-free survival using high-dose chemotherapy (reviewed in ref. 43). Although the apparent lack of increase survival is discouraging, these studies may not have had sufficient patients to detect meaningful differences. For this and other reasons, high-dose chemotherapy trials are ongoing. Alternatively, an increased therapeutic index of doxorubicin prodrugs could be used to decrease the toxicity from doxorubicin therapy. Such decreased toxicity could allow the use of other therapies that would not have been possible due to toxicities from doxorubicin.
The reason for the lack of correlation between the in vitro enzymatic efficiencies for conjugate cleavage and the metabolism rates in cultured HT1080 cells is unclear. Possibly, the major metabolizing MMP in cultured HT1080 cells is not MMP-2, -9, or -14. Consistent with this hypothesis, HT1080 cells express several MMPs other than MMP-2, -9, and -14, including MMP-1, -3, -7, -15, -16, and -18 (14, 15). Alternately, the conditions used for in vitro assays may not accurately reflect the in vivo activities of the MMPs. The enzymatic assays use soluble MMPs in detergent-containing solutions with substrate concentrations well below the Km and the cell-based assays use membrane-bound MMPs with conjugates dissolved in 0.1% bovine serum albumin, DMEM. Although conjugate concentrations were not saturating for metabolism in HT1080 cultures (data not shown), differences in the assay conditions between the in vitro assays and the cell-based assays could yield divergent results. In this case, the differences would be specific for MMPs because the neprilysin in vitro and in vivo rates were well correlated. Clearly, additional work will be needed to resolve this issue.
An acute toxicity similar to that with compound 5 was reported with ß-Ala-Leu-Ala-Leu-Dox (44). Interestingly, capping of the positively charged NH2 terminus in this conjugate with a succininc group also eliminated the acute toxicity of this conjugate. These investigators, however, attributed this improvement to decreased aggregation of the prodrugs.
The use of MMPs to activate prodrugs should be applicable to other cytotoxics. For instance, other investigators have made paclitaxel and vinblastine prodrugs. Although the peptide sequence identified with the doxorubicin prodrug is a good starting point for these other cytotoxics, it is likely that additional optimization will be required because the cytotoxic moiety affects the enzymatic utilization of the resulting conjugate. It is also likely that other cytotoxic prodrugs will not have the acute toxicity due to mast cell degranulation that was observed with the doxorubicin prodrugs. In particular, the positively charged peptides apparently increase the amount of histamine release that can result from doxorubicin alone under some circumstances.
In a prodrug that requires multiple steps for activation, like the MMP-activated doxorubicin prodrugs, the initial activation step should have the slowest rate and subsequent steps should be rapid compared with the initial step for optimal prodrug performance. The MMP-activated prodrugs described here do not possess this optimal property. In particular, the removal of the prime-side residues following MMP cleavage appears rapid, relative to MMP cleavage, except for the conversion of L-Dox to doxorubicin. This conclusion is supported by the low concentrations of intermediates with multiple prime-side residues and the high concentration of L-Dox observed in cultured cells and tumor tissue. Additionally, L-Dox concentrations in HT1080 xenografts at times soon after injection exceeded doxorubicin concentrations at later times. Taken together, these data show that a significant fraction of the L-Dox that is formed in the HT1080 xenografts diffuses outside the tumor before conversion to doxorubicin. If this loss could be reduced, then the potency, and potentially therapeutic index, of prodrugs would be increased. Because the enzymatic conversion of L-Dox to doxorubicin appears relatively slow, it would be interesting to investigate nonenzymatic mechanisms to eliminate the final residue, such as the self-immolating linkers described previously (45).
|
Acknowledgments
|
|---|
We thank Bob Grafstrom, Kurt Auger, Simon Robinson, and Peter Ho for advice and encouragement and Jim Duan for the gift of marimastat.
|
Footnotes
|
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: N. Graciani is currently at Wyeth Pharmaceutical, 500 Arcola Road, Collegeville, PA 19426. E. Yue and S. Yeleswaram are currently at Incyte Pharmaceuticals, Route 141 and Henry Clay Drive, Wilmington, DE 19880-0400. R. Stein is currently at the Department of Neurology, Harvard Medical School, Cambridge, MA 02139. Z. Lai, M. Diamond, S-Y. Zhang, R.A. Copeland, P. Huang, and A. Oliff are currently at GlaxoSmithKline, Collegeville, PA 19426. L. Grimminger is currently at Johnson & Johnson, Malvern, PA 19355. A. Musselman is currently at Merck Research Laboratories, Merck & Co., West Point, PA 19486. M. Zhang is currently at Neurocrine Biosciences, Inc., 1055 Science Center Drive, San Diego, CA 92130. A. Stern is currently at Ensemble Discovery Corp., 99 Erie Street, Cambridge, MA 02139. R. Taub is currently at Roche Pharmaceuticals, Nutley, NJ 07110.
Received 1/ 7/05;
revised 2/ 9/05;
accepted 3/ 2/05.
|
References
|
|---|
Hardman J, Limbird L, Gilman A. The pharmacologic basis of therapeutics. 10th ed. New York: McGraw-Hill; 2001.
Gerwitz D. A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics Adriamycin and daunorubicin. Biochem Pharmacol 1999;57:727–41.[CrossRef][Medline]
Woessner J, Nagase H. Matrix metalloproteinases and TIMPs. 1st ed. New York: Oxford University Press; 2000.
Strongin A, Collier I, Bannikov G, et al. Mechanism of cell surface activation of 72-kDa type IV collagenase. Isolation of the activated form of the membrane metalloprotease. J Biol Chem 1995;270:5331–8.[Abstract/Free Full Text]
McCawley L, Matrisian L. Matrix metalloproteinases: they're not just for matrix anymore. Curr Opin Cell Biol 2001;13:534–40.[CrossRef][Medline]
Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2002;2:161–74.[Medline]
H-YHa, Moon H-B, Nam M-S, et al. Overexpression of membrane-type matrix metalloproteinase-1 gene induces mammary gland abnormalities and adenocarcinoma in transgenic mice. Cancer Res 2001;61:984–90.[Abstract/Free Full Text]
Itoh T, Tanioka M, Yoshida H, et al. Reduced angiogenesis and tumor progression in gelatinase A-deficient mice. Cancer Res 1998;58:1048–51.[Abstract/Free Full Text]
Itoh T. Experimental metastasis is suppressed in MMP-9-deficient mice. Clin Exp Metastasis 1999;17:177–81.[CrossRef][Medline]
Jones J, Glynn P, Walker R. Expression of MMP-2 and MMP-9, their inhibitors, and the activators MT1-MMP in primary breast carcinomas. J Pathol 1998;189:161–8.
Elenbaas B, Weinberg R. Heterotypic signaling between epithelial tumor cells and fibroblasts in carcinoma formation. Exp Cell Res 2001;264:169–84.[CrossRef][Medline]
Pan W, Arnone M, Kendall M, et al. Identification of peptide substrates for human MMP-11 (stromelysin-3) using phage display. J Biol Chem 2003;278:27820–7.[Abstract/Free Full Text]
Nagase H, Fields G. Human matrix metalloproteinase specificity studies using collagen sequence-based synthetic peptides. Biopolymers 1996;40:399–416.[CrossRef][Medline]
Giambernardi T, Grant G, Taylor G, et al. Overview of matrix metalloproteinase expression in cultured human cells. Matrix Biol 1997;16:483–96.
Grant G, Giambernardi T, Grant A, Klebe R. Overview of expression of matrix metalloproteinases (MMP-17, MMP-18, MMP-20) in cultured human cells. Matrix Biol 1999;18:145–8.[CrossRef][Medline]
Sang A, Bodden M, Windsor L. Activation of human progelatinase A by collagenase and matrilysin: activation by procollagenase by matrilysin. J Protein Chem 1996;15:243–53.[Medline]
George H, Marchand P, Murphy K, et al. Recombinant human 92-kilodalton type V collagenase/gelatinase from baculovirus infected insect cells: expression, purification, and characterization. Protein Expr Purif 1997;10:154–61.[CrossRef][Medline]
Sonnenberg J, Sakane Y, Jeng A, et al. Identification of protease 3.4.24.11 as the major atrial natriuretic factor degrading enzyme in the rat kidney. Peptides 1988;9:173–80.[CrossRef][Medline]
Copeland R. Enzymes: a practical introduction to structure, mechanism, and data analysis. 2nd ed. New York: John Wiley & Sons, Inc.; 2000.
Stanton H, Gavrilovic J, Atkinson S, et al. The activation of proMMP-2 (gelatinase A) by HT1080 fibrosarcoma cells is promoted by culture on a fibronectin substrate and is concomitant with an increase in processing of MT1-MMP (MMP-14) to a 45 kDa form. J Cell Sci 1998;111:2789–98.[Abstract]
Rasmussen H, McCann P. Matrix metalloproteinase inhibition as a novel anticancer strategy: a review with special focus on batimastat and marimastat. Pharmacol Ther 1997;75:69–75.[CrossRef][Medline]
Trouet A, Passioukov A, Van derpoorten K, et al. Extracellularly tumor-activated prodrugs for the selective chemotherapy of cancer: application to doxorubicin and preliminary in vitro and in vivo studies. Cancer Res 2001;61:2843–6.[Abstract/Free Full Text]
Baurain R, Masquelier M, Deprez-de Campeneere D, Trouet A. Amino acid and dipeptide derivatives of daunorubicin. 2. Cellular pharmacology and antitumor activity on L1210 leukemic cells in vitro and in vivo. J Med Chem 1980;23:1171–4.[CrossRef][Medline]
Knight C, Willenbrock F, Murphy G. A novel coumarin-labelled peptide for sensitive continuous assays of the matrix metalloproteinases. FEBS Lett 1992;296:263–6.[CrossRef][Medline]
Bickett D, Green M, Berman J, et al. A high throughput fluorogenic substrate for interstitial collagenase (MMP-1) and gelatinase (MMP-9). Anal Biochem 1993;212:58–64.[CrossRef][Medline]
Mucha A, Cuniasse P, Kannan R, et al. Membrane type-1 matrix metalloproteinase and stromelysin-3 cleave more efficiently synthetic substrates containing unusual amino acids in their P1' positions. J Biol Chem 1998;273:2763–8.[Abstract/Free Full Text]
Turner A. Neprilysin. In: Barrett A, Rawlings N, Woessner J, editors. Handbook of proteolytic enzymes. San Diego: Academic Press; 1998. p. 1080–5.
Broggini M, Colombo T, Martini A, Donnelli M. Studies on the comparative distribution and biliary excretion of doxorubicin and 4'-epi-doxorubicin in mice and rats. Cancer Treat Rep 1980;64:897–904.[Medline]
Formelli F, Carsana R, Pollini C. Pharmacokinetics of 4'-deoxy-4'-iodo-doxorubicin in plasma and tissues of tumor-bearing mice compared to doxorubicin. Cancer Res 1987;47:5401–6.[Abstract/Free Full Text]
van der Vijgh W, Maessen P, Pinedo H. Comparative metabolism and pharmacokinetics of doxorubicin and 4'-epidoxorubicin in plasma, heart, and tumor of tumor-bearing mice. Cancer Chemother Pharmacol 1990;26:9–12.[Medline]
de Jong J, Klein I, Bast A, van der Vijgh W. Analysis and pharmacokinetics of N-L-leucyldoxorubicin and metabolites in tissues of tumor-bearing BALB/c mice. Cancer Chemother Pharmacol 1992;31:156–60.[CrossRef][Medline]
Decorti G, Bartoli Klugman F, Candusio L, Baldini L. Characterization of histamine secretion induced by anthracyclines in rat peritoneal mast cells. Biochem Pharmacol 1986;35:1939–42.[CrossRef][Medline]
Decorti G, Bartoli Klugman F, Candusio L, et al. Uptake of Adriamycin by rat and mouse mast cells and correlation with histamine release. Cancer Res 1989;49:1921–6.[Abstract/Free Full Text]
Bristow M, Sageman W, Scott R, et al. Acute and chronic cardiovascular effects of doxorubicin in the dog: the cardiovascular pharmacology of drug-induced histamine release. J Cardiovasc Pharmacol 1980;2:487–515.[Medline]
Burkitt H, Young B, Heath J. Wheater's functional histology. In: Wheater's functional histology. 3rd ed. Churchill Livingstone; 1993. p. 44.
Khan S, Denmeade S. In vivo activity of a PSA-activated doxorubicin prodrug against PSA-producing human prostate cancer xenografts. Prostate 2000;45:80–3.[CrossRef][Medline]
DeFeo-Jones D, Garsky V, Wong B, et al. A peptide-doxorubicin "prodrug" activated by prostate-specific antigen selectively kills prostate tumor cells positive for prostate-specific antigen in vivo. Nat Med 2000;6:1248–52.[CrossRef][Medline]
Dubois V, Dasnois L, Lebtahi K, et al. CPI-0004NA, a new extracellularly tumor-activated prodrug of doxorubicin: in vivo toxicity, activity, and tissue distribution confirm tumor cell selectivity. Cancer Res 2002;62:2327–31.[Abstract/Free Full Text]
Pan C, Cardarelli P, Nieder M, et al. CD10 is a key enzyme involved in the activation of tumor-activated prodrug CPI-0004Na and novel analogues: implications for the design of novel peptide prodrugs for the therapy of CD10+ tumors. Cancer Res 2003;63:5526–31.[Abstract/Free Full Text]
Devy L, de Groot F, Balcher S, et al. Plasmin-activated doxorubicin prodrugs containing a spacer reduce tumor growth and angiogenesis without systemic toxicity. FASEB J 2004;18:565–7.[Abstract/Free Full Text]
Wong B, DeFeo-Jones D, Jones R, et al. PSA-specific and non-PSA-specific conversion of a PSA-targeted peptide conjugate of doxorubicin to its active metabolites. Drug Metab Dispos 2001;29:313–8.[Abstract/Free Full Text]
DiPaola R, Rinehart J, Nemunatis J, et al. Characterization of a novel prostate-specific antigen-activated peptide-doxorubicin conjugate in patients with prostate cancer. J Clin Oncol 2002;20:1874–9.[Abstract/Free Full Text]
Tartarone A, Romano G, Galasso R, et al. Should we continue to study high-dose chemotherapy in metastatic breast cancer patients? A critical review of the published data. Bone Marrow Transplant 2003;31:525–30.[CrossRef][Medline]
Fernandez A, Van derpoorten K, Dasnois L, et al. N-succinyl-(ß-alanyl-L-leucyl-L-alanyl-L-leucyl)doxorubicin: an extracellularly tumor-activated prodrug devoid of intravenous acute toxicity. J Med Chem 2001;44:3750–3.[CrossRef][Medline]
Niculescu-Duvaz I, Niculescu-Duvaz D, Friedlos F, et al. Self-immolative anthracycline prodrugs for suicide gene therapy. J Med Chem 1999;42:2485–9.[CrossRef][Medline]
Top
Abstract
Introduction
