**Development, Optimization and Absorption Mechanism of DHP107, Oral Paclitaxel Formulation for Single-Agent Anticancer Therapy**

In-Hyun Lee, Jung Wan Hong, Yura Jang, Yeong Taek Park and Hesson Chung *Korea Institute of Science and Technology and Daehwa Pharmaceutical Korea* 

#### **1. Introduction**

356 New Advances in the Basic and Clinical Gastroenterology

Wolfe, F. & Michaud, K. 2004. Lymphoma in rheumatoid arthritis: the effect of methotrexate

Xue, F. B., Xu, Y. Y., Wan, Y., Pan, B. R., Ren, J. & Fan, D. M. 2001. Association of H. pylori infection with gastric carcinoma: a Meta analysis. *World J Gastroenterol,* 7, 801-4. Zachos, M., Tondeur, M. & Griffiths, A. M. 2007. Enteral nutritional therapy for induction of remission in Crohn's disease. *Cochrane Database Syst Rev*, CD000542. Zhang, S. M., Hunter, D. J., Rosner, B. A., Giovannucci, E. L., Colditz, G. A., Speizer, F. E. &

Wong, M. T. & Eu, K. W. 2006. Primary colorectal lymphomas. *Colorectal Dis,* 8, 586-91. Wong, N. A., Herbst, H., Herrmann, K., Kirchner, T., Krajewski, A. S., Moorghen, M.,

51.

201, 312-8.

477-85.

and anti-tumor necrosis factor therapy in 18,572 patients. *Arthritis Rheum,* 50, 1740-

Niedobitek, F., Rooney, N., Shepherd, N. A. & Niedobitek, G. 2003. Epstein-Barr virus infection in colorectal neoplasms associated with inflammatory bowel disease: detection of the virus in lymphomas but not in adenocarcinomas. *J Pathol,*

Willett, W. C. 2000. Intakes of fruits, vegetables, and related nutrients and the risk of non-Hodgkin's lymphoma among women. *Cancer Epidemiol Biomarkers Prev,* 9,

> Paclitaxel, administered as intravenous infusion currently, is an effective anticancer drug belonging to the taxane family (Figure 1) and used in the treatment of a wide variety of cancers including breast and ovarian cancers (Rowinsky et al., 1995). Researchers have been looking for more effective and convenient ways to administer paclitaxel with less formulation-related toxicities than the commercially available formulations like Taxol® by Bristol-Meyers Squibb. It is well known that some of the limitations of the current formulation come from Cremophor EL, a polyoxyethylated castor oil. This particular component is known to cause hypersensitivity (Weiss et al., 1990), to be responsible for nonlinear pharmacokinetic behavior (Kearns et al., 1995; Gianni, 1995) and to cause paclitaxel precipitation oftentimes when diluted during the infusion process (Pfeifer, 1993). Formulations for paclitaxel and its analogs have been prepared in many different ways for various administration procedures (Kim et al., 2006; Xie et al., 2007; Sugahara et al., 2007; Singla et al., 2002; Hennenfent & Govindan, 2006). Several years ago, Abraxane®, a suspension of albumin nanoparticles containing paclitaxel, obtained approval to treat metastatic breast cancer patients (Ibrahim et al., 2002; Garber, 2004). Abraxane®, a solvent-

Fig. 1. Structure of paclitaxel

Development, Optimization and Absorption Mechanism of

Peltier et al., 2006).

DHP107, Oral Paclitaxel Formulation for Single-Agent Anticancer Therapy 359

formulations. Paclitaxel, however, is known to be absorbed poorly in the gastrointestinal tract when administered orally (Sparreboom et al., 1997; Bardelmeijer et al., 2004; Choi & Li, 2005; Bardelmeijer et al., 2000). The reasons for poor absorption were identified to be drug metabolism by cytochrome P450 (CYP) and the existence of the efflux system, such as Pglycoproteins in intestinal epithelial cells and liver (Schellens et al., 2000). Tellingen and coworkers identified that the epithelial efflux system composed of P-glycoproteins lowered the drug absorption by demonstrating that orally administered paclitaxel can be absorbed well in mice with a homozygously disrupted mdr1a gene (mdr1a(-/-) mice) (Sparreboom et al., 1997). Proof-of-concept experiments in mice and man have also shown that oral absorption of paclitaxel could be increased dramatically by concomitant administration of a P-glycoprotein inhibitor, cyclosporine A (Schellens et al., 2000; Asperen et al., 1998; Meerum Terwogt et al., 1999). Other P-glycoprotein blockers that could also serve as CYP 450 inhibitors also boosted the oral absorption of paclitaxel (Sparreboom et al., 1997; Bardelmeijer et al., 2004; Choi & Li, 2005; Kim et al., 2004; Asperen et al., 1997). Formulations that do not contain Cremophor EL, that could lower the *in vivo* toxicity and increase solubilization of paclitaxel, were shown to deliver paclitaxel efficiently via oral administration especially when P-glycoprotein inhibitors were given simultaneously (Kim et al., 2004; Gao et al., 2003; Yang et al., 2004; Woo et al., 2003; Choi et al., 2004a, 2004b;

In Table 1, we reviewed the literature on oral paclitaxel formulations and summarized the pharmacokinetic data of paclitaxel obtained from the small-animal experiments (FVB mice and Sprague-Dawley rats). In the Table, we have listed Area under the plasma concentration vs. time curve (AUC) value, maximum drug concentration in the blood (Cmax) and oral dose. One has to keep in mind that the data from different papers may not be compared directly since the AUC values in the literature were estimated for different time intervals and drug doses while the pharmacokinetics of paclitaxel is known to be non-linear (Kearns et al., 1995; Gianni, 1995). Different time intervals could not be normalized due to lack of information in the pharmacokinetics data to make such estimations and thus must be viewed with caution. Another important point to note is that Cremophor EL in Taxol®, used for intravenous administration in some cases (Sparreboom et al., 1997; Bardelmeijer et al., 2004; Asperen et al., 1998; Gao et al., 2003; Yang et al., 2004; Peltier et al., 2006), causes nonlinear pharmacokinetic behavior (Sparreboom et al., 1996). Also, paclitaxel dissolved in Tween 80, also used for intravenous controls in others (Choi & Li, 2005; Bardelmeijer et al., 2000; Woo et al., 2003; Choi et al., 2004a, 2004b), does show linear pharmacokinetic behavior, but has *ca*. 5~10 times lower AUC values when compared to diluted Taxol® (Sparreboom et al., 1996).

For these reasons, it was impossible to list the bioavailability values in Table 1.

In the past several years, we have been developing paclitaxel formulations with biocompatible oils, lipids and emulsifiers (Lee et al. Hong et al., 2004; I. H. Lee et al., 2005; S. J. Lee et al., 2005). Paclitaxel dissolved in a stable Lipiodol formulation was shown to retard the growth of hepatocellular carcinoma efficiently when transcatheter arterial chemoembolization was performed in rabbits (Yoon et al., 2003). Also, a paclitaxel formulation prepared with monoolein, tricaprylin and Tween 80 was mucoadhesive when given intravesically (S. J. Lee et al., 2005). Paclitaxel in this formulation penetrated to lamina propria and close to the muscle layer of the bladder while the paclitaxel concentration was low throughout the depth of the bladder tissue when Taxol® was used. We also have shown


\* Time-interval used for the calibration of AUC,

\*\* Dose of the inhibitor in mg/kg, and

\*\*\* Estimated concentration

Table 1. Mini-review: Selective pharmacokinetic data of paclitaxel after oral administration to rodents.

free formulation has a number of improved features such as shorter infusion time and no hypersensitivity without premedication (Garber, 2004).

Since an oral route would clearly be an attractive alternative to injection for patients as well as for medical personnel, there have been many attempts to prepare effective oral paclitaxel

Cmax (μg/ml)

Taxol 0.37 (0-4) 0.13\*\*\* 10 mice Bardelmeijer et al., 2004

Taxol 0.41 (0-6) 0.12 10 mice Bardelmeijer et al., 2000 Cremophor/ethanol 0.51 (0-24) 0.10 5 mice Asperen et al., 1998 Supersaturable-SEDDS 0.44 (0-∞) 0.28 10 rat Gao et al., 2003 SMEDDS 0.81 (0-24) 0.05 2 rat Yang et al., 2004 SMEDDS 0.97 (0-24) 0.05 10 rat Yang et al., 2004 microemulsion 0.45 (0-24) 0.05 25 rat Woo et al., 2003

Suspension in Tween 80 1.65 (0-24) 0.11 40 rat Choi et al., 2004a, 2004b Lipid nanocapsule 1.05 (0-∞) 0.37 10 rat Peltier et al., 2006

SMEDDS+cyclosporine A (40) 1.14 (0-24) 0.16 2 rat Yang et al., 2004 Taxol+verapamil (15) 4.27 (0-∞) 0.19 50 rat Choi & Li, 2005 Microemulsion+KR-30031 (20) 3.37 (0-24) 0.22 25 rat Woo et al., 2003 Surfactant mix + KR30031 (5) 0.39 (0-12) 0.06 5 rat Kim et al., 2004

Taxol+cyclosporine A (50) 3.61 (0-4) 0.82\*\*\* 10 mice Bardelmeijer et al., 2004 Taxol+PSC 833 (50) 2.71 (0-4) 1.47\*\*\* 10 mice Bardelmeijer et al., 2004 Taxol+GF120918 (25) 1.39 (0-4) 0.55\*\*\* 10 mice Bardelmeijer et al., 2004 Taxol+LY335979 (80) 1.41 (0-4) 0.42\*\*\* 10 mice Bardelmeijer et al., 2004 Taxol+R101933 (80) 0.93 (0-4) 0.32\*\*\* 10 mice Bardelmeijer et al., 2004 Taxol+GF120918 (25) 2.65 (0-10) 0.77 10 mice Bardelmeijer et al., 2000 Taxol+PSC 833 (50) mice Asperen et al., 1997

Table 1. Mini-review: Selective pharmacokinetic data of paclitaxel after oral administration

free formulation has a number of improved features such as shorter infusion time and no

Since an oral route would clearly be an attractive alternative to injection for patients as well as for medical personnel, there have been many attempts to prepare effective oral paclitaxel

Taxol 1.89 (0-∞) 0.11 50 rat Choi & Li, 2005

Dose (mg/kg)

6.68 (0-24) 0.73 10 mice Asperen et al., 1998

1.05 (0-∞) 0.31 10 rat Gao et al., 2003

5.00 (0-24) 0.28 40 rat Choi et al., 2004b

Rodent species

) 0.14 10 mice Sparreboom et al., 1997

Ref

(μgh/ml)

Formulation AUC

Taxol 0.50 (0-8\*

Without inhibitors

With inhibitors Cremophor/Ethanol +Cyclosporine A (50\*\*)

Supersaturable-SEDDS +cyclosporine A (30)

Suspension in Tween 80 +

\* Time-interval used for the calibration of AUC,

hypersensitivity without premedication (Garber, 2004).

\*\* Dose of the inhibitor in mg/kg, and

\*\*\* Estimated concentration

to rodents.

quercetin (20)

formulations. Paclitaxel, however, is known to be absorbed poorly in the gastrointestinal tract when administered orally (Sparreboom et al., 1997; Bardelmeijer et al., 2004; Choi & Li, 2005; Bardelmeijer et al., 2000). The reasons for poor absorption were identified to be drug metabolism by cytochrome P450 (CYP) and the existence of the efflux system, such as Pglycoproteins in intestinal epithelial cells and liver (Schellens et al., 2000). Tellingen and coworkers identified that the epithelial efflux system composed of P-glycoproteins lowered the drug absorption by demonstrating that orally administered paclitaxel can be absorbed well in mice with a homozygously disrupted mdr1a gene (mdr1a(-/-) mice) (Sparreboom et

al., 1997). Proof-of-concept experiments in mice and man have also shown that oral absorption of paclitaxel could be increased dramatically by concomitant administration of a P-glycoprotein inhibitor, cyclosporine A (Schellens et al., 2000; Asperen et al., 1998; Meerum Terwogt et al., 1999). Other P-glycoprotein blockers that could also serve as CYP 450 inhibitors also boosted the oral absorption of paclitaxel (Sparreboom et al., 1997; Bardelmeijer et al., 2004; Choi & Li, 2005; Kim et al., 2004; Asperen et al., 1997). Formulations that do not contain Cremophor EL, that could lower the *in vivo* toxicity and increase solubilization of paclitaxel, were shown to deliver paclitaxel efficiently via oral administration especially when P-glycoprotein inhibitors were given simultaneously (Kim et al., 2004; Gao et al., 2003; Yang et al., 2004; Woo et al., 2003; Choi et al., 2004a, 2004b; Peltier et al., 2006).

In Table 1, we reviewed the literature on oral paclitaxel formulations and summarized the pharmacokinetic data of paclitaxel obtained from the small-animal experiments (FVB mice and Sprague-Dawley rats). In the Table, we have listed Area under the plasma concentration vs. time curve (AUC) value, maximum drug concentration in the blood (Cmax) and oral dose. One has to keep in mind that the data from different papers may not be compared directly since the AUC values in the literature were estimated for different time intervals and drug doses while the pharmacokinetics of paclitaxel is known to be non-linear (Kearns et al., 1995; Gianni, 1995). Different time intervals could not be normalized due to lack of information in the pharmacokinetics data to make such estimations and thus must be viewed with caution. Another important point to note is that Cremophor EL in Taxol®, used for intravenous administration in some cases (Sparreboom et al., 1997; Bardelmeijer et al., 2004; Asperen et al., 1998; Gao et al., 2003; Yang et al., 2004; Peltier et al., 2006), causes nonlinear pharmacokinetic behavior (Sparreboom et al., 1996). Also, paclitaxel dissolved in Tween 80, also used for intravenous controls in others (Choi & Li, 2005; Bardelmeijer et al., 2000; Woo et al., 2003; Choi et al., 2004a, 2004b), does show linear pharmacokinetic behavior, but has *ca*. 5~10 times lower AUC values when compared to diluted Taxol® (Sparreboom et al., 1996). For these reasons, it was impossible to list the bioavailability values in Table 1.

In the past several years, we have been developing paclitaxel formulations with biocompatible oils, lipids and emulsifiers (Lee et al. Hong et al., 2004; I. H. Lee et al., 2005; S. J. Lee et al., 2005). Paclitaxel dissolved in a stable Lipiodol formulation was shown to retard the growth of hepatocellular carcinoma efficiently when transcatheter arterial chemoembolization was performed in rabbits (Yoon et al., 2003). Also, a paclitaxel formulation prepared with monoolein, tricaprylin and Tween 80 was mucoadhesive when given intravesically (S. J. Lee et al., 2005). Paclitaxel in this formulation penetrated to lamina propria and close to the muscle layer of the bladder while the paclitaxel concentration was low throughout the depth of the bladder tissue when Taxol® was used. We also have shown

Development, Optimization and Absorption Mechanism of

**2.4 Differential scanning calorimetry** 

light/dark cycle.

et al., 2001)

stored at -70 C until analysis.

DHP107, Oral Paclitaxel Formulation for Single-Agent Anticancer Therapy 361

and handling followed institutional guidelines (Korea Institute of Science and Technology). Mice were maintained with free access to food and water under a 12-h

Differential scanning calorimetry (DSC) was performed to obtain the heating thermograms of paclitaxel and the formulations in Table 2 (DSC 821e, Mettler Toledo, Columbus, OH, equipped with Intracooler, Haake EK90/MT, Haake, Denmark) at a heating scan rate of 5 C/min. Scans were made with samples contained in hermetically sealed aluminum crucibles (ME-27331, Mettler Toledo). Initial heating scans were reported for paclitaxel and the formulations in Table 2. In case of the formulations, thermal history did not alter the

The average particle size in the dispersions G8, G9 and G10 in Table 3 was determined by quasielastic laser light scattering with a Malvern Zetasizer (Malvern Instruments Limited, England). The dispersions were diluted by 300 times in water before the measurement. The size determination was repeated 3 times/sample. The average size and the size distribution were estimated from the log-normal size distribution function as shown previously (Chung

Paclitaxel formulations in Tables 2 and 3 were liquefied at 37 ~ 50 C and administered orally at doses of 25, 50, 75, and 100 mg/kg with a blunt needle via the esophagus into the stomach. The male ICR mice were fasted for 8 h prior to oral administration except for the group G3. Taxol® prepared by dissolving paclitaxel in an equivolume mixture of Cremophor EL and ethanol at 6 mg/ml (Taxol®) was diluted by 6 times with the saline solution, and was administered via bolus tail-vein injection at a dose of 10 mg/kg as a positive control. Blood samples were collected at various time points (n=6) after drug administration, and were

Whole blood (200 μl) was spiked with irbesartan (0.5 μg/ml; internal standard), mixed and added to acetonitrile (400 μl) to precipitate proteins. After centrifugation at 14,000 RCF for 20 min, the supernatant was collected and mixed with the mobile phase to adjust the volume to 0.6 ml. Ten microliters of the blood was injected into the LC/MS/MS system. Analyses were performed with a Thermo-Finnigan Discovery Max LC/MS/MS (San Jose, CA, USA). The LC system was performed at 35 C on a Capcellpak C18 column (150 X 2.0 mm i.d., 5 μm particle size, Shisheido, Japan) equipped with a Zorbax SB-Aq (12.5 X 2.1 mm i.d.) guard column. The mobile phase consisted of 55 % acetonitrile, 0.08 % formic acid and 44.92 % water, and the flow rate was 0.4 ml/min. The instrument was operated in SRM mode (positive ion), monitoring the ion transitions from m/z 854 285 (paclitaxel) and m/z 429 195 (internal standard). The paclitaxel LC/MS/MS assay was linear over the range of 2 ~ 1000 ng/ml with a lower quantitation limit of 2 ng/ml in blood. Paclitaxel

heating thermograms if the samples were cooled to – 20 C before reheating.

**2.5 Determination of the particle size in the dispersion** 

**2.6 Oral administration of paclitaxel formulations** 

**2.7 Analysis of paclitaxel concentrations in blood** 

that this mucoadhesive formulation can deliver paclitaxel effectively when given by oral route without additional active pharmaceutical ingredient as an absorption enhancer (Lee et al., 2004; Hong et al., 2004; I. H. Lee et al., 2005, Shin et al., 2009). As an endeavor to formulate more efficient oral paclitaxel formulations, we have prepared oil-based paclitaxel formulations with monoolein, saturated triglycerides and emulsifiers. Monoolein was included in the formulation as a main ingredient due to its high bioadhesiveness (Nielsen et al., 1998) and its ability to release the encapsulated drug in a controlled fashion (Clogston et al., 2005a, 2005b). Triglycerides were added since they can solubilize paclitaxel efficiently (Kan et al., 1999). To access the effectiveness of the formulations, pharmacokinetic and antitumor efficacy studies were performed in mice models.
