**4.3 PG as the carrier of PG-PTX**

Compared with other synthetic polymers that have been tested in clinical studies, PG is unique because it is composed of natural L-glutamic acid linked together through amide bonds rather than the nondegradable C-C backbone. The free γ-carboxyl group in each repeating unit of L-glutamic acid is negatively charged under a neutral pH condition, which

et al., 1998a). Because the conjugation site is the 2 hydroxyl group of PTX, which is a crucial site for tubulin binding, the conjugate does not interact with β-tubulin and is inactive (Li, 2002; Rogers, 1993). The median MW of PG-PTX is 38.5 kDa, with a PTX content of approximately 36% on a w/w basis, equivalent to about one PTX ester linkage per 11 PG

Morphological analysis and biochemical characterizations demonstrate that both PTX and PG-PTX are able to induce apoptosis in cells expressing wild-type p53 or mutant p53, to arrest cells in the G2/M phase of the cell cycle, and to down-regulate HER-2/neu expression. Furthermore, when PG-PTX is compared with other water-soluble derivatives of PTX, including small-molecular-weight sodium pentetic acid-PTX and polyethylene glycol-PTX conjugate (MW 5 kDa), they all show the same effects on telomeric association, mitotic index chromatin condensation, and formation of apoptotic bodies (Multani et al., 1999).

These results indicate that PG-PTX has the same mechanisms of action as PTX.

Fig. 2. Illustration of EPR effect and endocytosis. Different from blood vessels in normal tissue (A), those in tumor tissue (B) have porous openings, through which large-size

enzymes inside the lysosome of the tumor cell.

**4.3 PG as the carrier of PG-PTX** 

conjugate leaks and is preferentially trapped and distributed to the tumor tissue. Once in the tumor tissue, the conjugate is taken up by the tumor cells through a cellular process called endocytosis (C). The conjugate releases active agent (D) via metabolism by lysosomal

Compared with other synthetic polymers that have been tested in clinical studies, PG is unique because it is composed of natural L-glutamic acid linked together through amide bonds rather than the nondegradable C-C backbone. The free γ-carboxyl group in each repeating unit of L-glutamic acid is negatively charged under a neutral pH condition, which

units of the polymer (Fig. 3) (Bonomi, 2007; Rogers, 1993).

makes the polymer water-soluble. The carboxyl groups can also provide functionality for drug attachment. PG is not only water-soluble and biodegradable, it is also nontoxic. All these characteristics make PG a unique candidate as the carrier of polymer-drug conjugates for selective delivery of chemotherapeutic agents, especially for PTX (Parveen & Sahoo, 2008).

Fig. 3. Schematic representation of PG-PTX structure. The structure shown is illustration of a fragment of the molecule. On average there are approximately 10 non-conjugated monomer glutamic acid units (a + b) for every molecule conjugated to a PTX molecule (y).

### **4.4 The metabolism of PG-PTX**

PG-PTX conjugate is stable in the systemic circulation but can be broken down by intracellular lysosomal enzymes to release the drug after entering cells by endocytosis. The proposed mechanism by which PG-PTX is metabolized includes endocytosis of the polymerdrug conjugate followed by intracellular release of active PTX by proteolytic activity of the lysosomal enzyme cathepsin B, an exocarboxydipeptidase, and diffusion of PTX into the nucleus (Turk et al., 2001). The process is presented in Fig. 4. This finding may have biological relevance as expression of cathepsin B is upregulated in malignant cells, particularly during tumor progression period (Podgorski & Sloane, 2003). These data support a model in which PG-PTX accumulates in tumor tissues through the EPR effect, followed by the cathepsin B-mediated release of PTX. The kinetics of intracellular formation of several PG-PTX metabolites have been quantified in vitro and have been found to be largely dependent on cathepsin B. Metabolites that have been detected in vivo include diglutamyl-PTX and monoglutamyl-PTX. Monoglutamyl-PTX is an unstable compound that can be nonenzymatically degraded to release free PTX. Specific enzyme inhibitors such as CA-074 methyl ester, a cell-permeable irreversible inhibitor of cathepsin B, and EST, a cellpermeable irreversible inhibitor of cysteine proteases, dramatically decrease the formation of monoglutamate PTX and unconjugated PTX in tumor cells that have been incubated with PG-PTX (Rogers, 1993).

Fig. 4. Metabolism of PG-PTX by endocytosis and enzymatic degradation.

### **4.5 The advantages of PG-PTX**

### **4.5.1 PG-PTX reduces the side effects of PTX in clinical application**

Compared with PTX, the solubility, uptake, tumor retention, and anticancer efficacy of PG-PTX are increased. Many clinical studies so far have conrmed several advantages of PG-PTX in the treatment of cancer patients. This macromolecular conjugate PG-PTX eliminates the need for Cremophor® EL and, therefore, decreases infusion time and the risk of hypersensitivity. Compared to standard taxanes, PG-PTX in phase I and phase II studies shows encouraging outcomes with reduced neutropenia and alopecia, and allows a more convenient administration schedule without the need for routine premedications (Rogers, 1993). PG-PTX induces hypersensitivity reactions in less than 1% of patients, without premedication, and only rare severe reactions have been reported. Furthermore, patients undergoing PG-PTX therapy have a better quality of life, because there is no significant hair loss, nerve damage, or neutropenia at the current dose. PG-PTX is watersoluble and can be administrated rapidly as a 10 -20 min infusion rather than hours when administrating PTX. The recommended phase II dose of PG-PTX is 235 mg/m2 every 3 week administered over a 10 min infusion without premedication (Sabbatini et al., 2004). Twenty-six patients were treated with PG-PTX in the Phase I study of PG-PTX administered weekly for patients with advanced solid malignancies. The recommended dose of PG-PTX for subsequent disease-directed studies is 70 mg/m2 weekly. Most patients experienced at least one drug-related adverse event during the study (Table 1) (Mita et al., 2009). Ninety-nine patients were treated with PG-PTX in a multi-center phase II study of PG-PTX as an intravenous (i.v.) infusion (approximately 10 min) at a dose of 175 mg/m2 on day 1 of each 3-week cycle. And the treatment-related adverse events including non-laboratory-based and laboratory-based maximum CTC toxicities in all cycles are listed in Table 2.


Fig. 4. Metabolism of PG-PTX by endocytosis and enzymatic degradation.

**4.5.1 PG-PTX reduces the side effects of PTX in clinical application** 

Compared with PTX, the solubility, uptake, tumor retention, and anticancer efficacy of PG-PTX are increased. Many clinical studies so far have conrmed several advantages of PG-PTX in the treatment of cancer patients. This macromolecular conjugate PG-PTX eliminates the need for Cremophor® EL and, therefore, decreases infusion time and the risk of hypersensitivity. Compared to standard taxanes, PG-PTX in phase I and phase II studies shows encouraging outcomes with reduced neutropenia and alopecia, and allows a more convenient administration schedule without the need for routine premedications (Rogers, 1993). PG-PTX induces hypersensitivity reactions in less than 1% of patients, without premedication, and only rare severe reactions have been reported. Furthermore, patients undergoing PG-PTX therapy have a better quality of life, because there is no significant hair loss, nerve damage, or neutropenia at the current dose. PG-PTX is watersoluble and can be administrated rapidly as a 10 -20 min infusion rather than hours when administrating PTX. The recommended phase II dose of PG-PTX is 235 mg/m2 every 3 week administered over a 10 min infusion without premedication (Sabbatini et al., 2004). Twenty-six patients were treated with PG-PTX in the Phase I study of PG-PTX administered weekly for patients with advanced solid malignancies. The recommended dose of PG-PTX for subsequent disease-directed studies is 70 mg/m2 weekly. Most patients experienced at least one drug-related adverse event during the study (Table 1) (Mita et al., 2009). Ninety-nine patients were treated with PG-PTX in a multi-center phase II study of PG-PTX as an intravenous (i.v.) infusion (approximately 10 min) at a dose of 175 mg/m2 on day 1 of each 3-week cycle. And the treatment-related adverse events including non-laboratory-based and laboratory-based maximum CTC toxicities in all

**4.5 The advantages of PG-PTX** 

cycles are listed in Table 2.


Table 1. Adverse events related to PG-PTX administration (*n* = 26).


Abbreviation: CTC, National Cancer Institute Common Toxicity Criteria.

Table 2. Treatment-related adverse events (*n* = 99).

### **4.5.2 PG-PTX promotes anticancer efficacy and reduces toxicity by prolonging tumor exposure and minimizing systemic exposure to active drug**

A single i.v. injection of PG-PTX at its maximum tolerated dose (MTD) equivalent to 60 mg of PTX/kg and at a lower dose equivalent to 40 mg of PTX/kg results in the disappearance of an established implanted 13762F mammary adenocarcinoma (mean size, 2000 mm3) in rats. Similarly, mice bearing syngeneic OCa-1 ovarian carcinoma (mean size, 500 mm3) are tumorfree within 2 weeks after a single i.v. injection of PG-PTX at a dose equivalent to 160 mg of PTX/kg (Li et al., 1998b). MTD of PTX in rats and mice are 20 mg/kg and 60 mg/kg, respectively. In contrast, MTD of a single i.v. injection of PG-PTX in rats and mice are 60 mg/kg and 160 mg/kg, respectively. MTD of PG-PTX was approximately 160 mg/kg to 200 mg/kg in immunocompetent mice and 120 mg/kg to 150 mg/kg in immunodeficient animals. At their respective MTDs, single-dose PG-PTX is more efficacious than PTX in Cremophor® EL/ethanol (Li et al., 1999). PG-PTX has shown anticancer activity in preclinical studies with human tumor xenografts and in early phase I trials. MTD of PG-PTX as a single agent, based on the first cycle toxicity of patients, is 235 mg/m2 (Verschraegen et al., 2009). Biodistribution in mice bearing OCa-1 tumor treated with i.v. injections of tritium-labeled PG-[3H] PTX shows a five times greater distribution of PTX to tumor tissues than those treated with PTX (Li et al., 2000c), which was demonstrated by whole-body autoradiograph (Fig. 5).

Fig. 5. Whole-body autoradiographs of mice killed 1 day (A) and 6 days (B) after tail vein injection of PG-[3H]PTX. Most radioactivity was localized to tumor periphery at 1 day after injection, but by day 6, radioactivity had diffused into the center of the tumor. L: liver; M: muscle; Arrow head: tumor.

Preclinical studies in animal tumor models demonstrate that PG-PTX is more effective than PTX and it is associated with prolonged tumor exposure but minimized systemic exposure to the active drug. The slow release of the active drug from a well-designed polymer carrier results in sustained high intratumoral drug levels and lower plasma concentrations of the

established implanted 13762F mammary adenocarcinoma (mean size, 2000 mm3) in rats. Similarly, mice bearing syngeneic OCa-1 ovarian carcinoma (mean size, 500 mm3) are tumorfree within 2 weeks after a single i.v. injection of PG-PTX at a dose equivalent to 160 mg of PTX/kg (Li et al., 1998b). MTD of PTX in rats and mice are 20 mg/kg and 60 mg/kg, respectively. In contrast, MTD of a single i.v. injection of PG-PTX in rats and mice are 60 mg/kg and 160 mg/kg, respectively. MTD of PG-PTX was approximately 160 mg/kg to 200 mg/kg in immunocompetent mice and 120 mg/kg to 150 mg/kg in immunodeficient animals. At their respective MTDs, single-dose PG-PTX is more efficacious than PTX in Cremophor® EL/ethanol (Li et al., 1999). PG-PTX has shown anticancer activity in preclinical studies with human tumor xenografts and in early phase I trials. MTD of PG-PTX as a single agent, based on the first cycle toxicity of patients, is 235 mg/m2 (Verschraegen et al., 2009). Biodistribution in mice bearing OCa-1 tumor treated with i.v. injections of tritium-labeled PG-[3H] PTX shows a five times greater distribution of PTX to tumor tissues than those treated with PTX (Li et al.,

2000c), which was demonstrated by whole-body autoradiograph (Fig. 5).

Fig. 5. Whole-body autoradiographs of mice killed 1 day (A) and 6 days (B) after tail vein injection of PG-[3H]PTX. Most radioactivity was localized to tumor periphery at 1 day after injection, but by day 6, radioactivity had diffused into the center of the tumor. L: liver; M:

Preclinical studies in animal tumor models demonstrate that PG-PTX is more effective than PTX and it is associated with prolonged tumor exposure but minimized systemic exposure to the active drug. The slow release of the active drug from a well-designed polymer carrier results in sustained high intratumoral drug levels and lower plasma concentrations of the

muscle; Arrow head: tumor.

active drug. To accomplish this, the polymer conjugate should release the active drug in tumor tissues rather than in the plasma during circulation. As a result, exposure of normal tissues will be limited, which is potentially associated with a more favorable toxicity profile (Li et al., 2000c). Thus, enhanced tumor uptake and sustained release of PTX from PG-PTX in tumor tissues are major factors contributing to its markedly improved in vivo anticancer activities.

### **4.5.3 The favorable pharmacokinetic properties of PG-PTX**

The superior anticancer activity of PG-PTX in preclinical studies suggests that PG-PTX might have favorable pharmacokinetic properties. Many studies suggest that PG-PTX exerts the anticancer activity by the continuous release of free PTX, and that the favorable pharmacokinetics of PG-PTX conjugate in vivo is likely the main cause contributing to its advanced anticancer activity (Oldham et al., 2000). Female mice with subcutaneous B16 murine melanomas are given PG-[3H] PTX at the equivalent dose 40 mg/kg of [3H] PTX i.v. infusion. Tumor samples are collected at regular intervals up to 144 h after infusion, and the concentrations of PG-PTX and PTX are determined by LC/MS analysis. Tumor exposure to total taxanes is increased by a factor of 3 (Cmax) or a factor of 12 (AUC) in mice treated with PG-[3H] PTX compared with the mice treated with [3H] PTX (Table 3) (Chipman et al., 2006). PG-[3H] PTX has a much longer half-life in plasma than [3H] PTX (Fig. 6).

Whereas PTX has an extremely short half-life in the plasma of mice (t1/2 = 29 min), the apparent half-life of PG-PTX is prolonged (t1/2 = 317 min) (Li et al., 1998b). In clinical trials, PG-PTX is given as a 30-min infusion every 3 weeks. Patients were treated at dose levels ranging from 30 mg/m2 to 720 mg/m2. PG-PTX is detectable in plasma of all patients and has a long plasma half-life of up to 185 h, and the results are consistent with preclinical ndings. Furthermore, concentrations of free PTX released from PG-[3H] PTX remain relatively constant up to 6 d after infusing. Moreover, peak plasma concentrations of free PTX are less than 0.1 μM 24 h after PG-PTX administration at doses up to 480 mg/m2 (176 mg/m2 PTX equivalents) (Boddy et al., 2001).

Fig. 6. Tumor concentration of [3H]paclitaxel after treatment with PG-[3H] PTX and [3H] paclitaxel in female mice with s.c. B16 melanomas at a dose of 40 mg paclitaxel.

In another phase I study, PG-PTX is administrated at 70 mg/m2. The mean maximal concentration (Cmax) is 41.2 ± 8.60 μg/mL and the Cmax is reached right after the end of the infusion. The plasma concentration declines with a mean terminal half-life of 15.7 ± 3.17 h. The mean AUC at the MTD is 455 ± 112 μg/h/mL and the mean average systemic plasma clearance is 0.16 ± 0.04 L/h/m2 (Sabbatini et al., 2004). At the MTD, the mean volume of distribution at steady state and during the terminal phase are 1.41 ± 0.28 L/m2 and 3.62 ± 1.13 L/m2, the mean Cmax of unconjugated PTX is 0.21 ± 0.07 μg/mL, the mean Tmax is 0.56 ± 0.18 h, the mean terminal half life is 16.6 ± 7.85 h, and the mean AUC is 3.15 ± 1.16 μg/h/mL. The ratio of the free PTX AUC to the conjugated PTX AUC is 0.7%. The plasma concentration of PTX released from PG-PTX increases largely in proportion to the dose and remains similar after repeated administration (Sabbatini et al., 2004).


Table 3. Preclinical tumor pharmacokinetics.
