A parenteral docetaxel-loaded lipid microsphere with decreased 7-epidocetaxel conversion in vitro and in vivo

Accepted Manuscript A parenteral docetaxel-loaded lipid microsphere with decreased 7-epidocetaxel conversion in vitro and in vivo

Lifeng Luo, Xiuzhi Wang, Qiuyue Chen, Linlin Miao, Xuezhi Zhuo, Lu Liu, Jiawen Xu, Yu Zhang, Haibing He, Tian Yin, Xing Tang PII: DOI: Reference:

S0928-0987(17)30514-6 doi: 10.1016/j.ejps.2017.09.022 PHASCI 4216

To appear in:

European Journal of Pharmaceutical Sciences

Received date: Revised date: Accepted date:

20 July 2017 10 September 2017 13 September 2017

Please cite this article as: Lifeng Luo, Xiuzhi Wang, Qiuyue Chen, Linlin Miao, Xuezhi Zhuo, Lu Liu, Jiawen Xu, Yu Zhang, Haibing He, Tian Yin, Xing Tang , A parenteral docetaxel-loaded lipid microsphere with decreased 7-epidocetaxel conversion in vitro and in vivo, European Journal of Pharmaceutical Sciences (2017), doi: 10.1016/ j.ejps.2017.09.022

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT A parenteral docetaxel-loaded lipid microsphere with decreased 7-epidocetaxel conversion in vitro and in vivo

Lifeng Luoa, Xiuzhi Wanga, Qiuyue Chena, Linlin Miaoa, Xuezhi Zhuoa, Lu

Department of Pharmaceutics Science, Shenyang Pharmaceutical University,

Shenyang, 110016, People's Republic of China.

SC

a

RI

PT

Liua, Jiawen Xua, Yu Zhanga, Haibing Hea, Tian Yina, Xing Tanga,*

NU

The corresponding author: Xing Tang, Department of Pharmaceutics Science,

TEL: 86-24-23986343 ; Fax:86-24-23911736

AC

CE

PT E

D

Email address: [email protected] ;

MA

Shenyang Pharmaceutical University, Shenyang, 110016, People's Republic of China. ;

1

ACCEPTED MANUSCRIPT Abstract: The purpose of the study was to develop a parenteral docetaxel lipid microsphere to inhibit its 7-epidocetaxel conversion in vitro and in vivo. 7-epidocetaxel conversion as the main indicator was investigated to optimize the formulation and process. 10%

PT

medium-chain triglyceride/long-chain triglyceride(3:1) as the oil phase, egg lethcin

RI

E80 as the emulsifier and 0.02% NaHSO3 as the acidity regulator were selected to

SC

prepare docetaxel lipid microsphere. This study found that pH and temperature were dominant factors on the epimerization of docetaxel in lipid microsphere, and that

NU

optimum conditions were a pH of 5.3 and thermal sterilization conditions of 121℃

MA

autoclaving for 8 min. According to the degradation kinetics, docetaxel lipid microsphere had a wider pH range where 7-epidocetaxel(%) stayed at low levels than

D

Docetaxel for Injection, and might improve the docetaxel stability by loading drug in

PT E

lecithin layer instead of altering the degradation mechanism. Docetaxel lipid microsphere decreased epimerization in plasma in vitro obviously. Pharmacokinetics

CE

of docetaxel and 7-epidocetaxel were investigated to quantify the 7-epidocetaxel

AC

conversion in vivo. The resulrs indicated that there was less conversion of docetaxel in lipid microspheres than in Docetaxel for Injection. The convert ratios were 0.61% and 3.04% respectively. In conclusion, lipid microsphere is a promising delivery system for intravenous administration of docetaxel with decreased 7-epidocetaxel conversion.

Keywords:

chemical

stability;docetaxel;

7-epidocetaxel

2

epimerization;lipid

microsphere;

ACCEPTED MANUSCRIPT Introduction Docetaxel (DTX), a semi-synthetic analogue of paclitaxel, is a potent chemotherapeutic agent widely used for the treatment of various solid tumors, including ovarian cancer,non-small cell lung cancer,breast cancer and others(Clarke

PT

and Rivory, 1999). It is known to exert its therapeutic efficacy by promoting tubulin

RI

assembly, inhibiting depolymerization of the resulting microtubules and stabilizing

SC

microtubules, which leads to mitotic arrest and cell death (Horwitz, 1992; Lavelle et al., 1995). The primary marketable product (Taxotere®) used in the clinic are

NU

co-injected with polysorbate 80 and 13% ethanol in water as the solvent, followed by

MA

further dilution in 0.9% normal saline or 5% dextrose solution prior to administration. It has been shown that this composition is fairly unstable and generates a significant

D

amount of 7-epidocetaxel (EDTX) due to the degradation of DTX when exposed to

PT E

heat (AG, 2008). The EDTX, a main degradation impurity of DTX, is formed by epimerization of the baccatin moiety at the C7-position, and this degradation of DTX

CE

can produce substances with decreased activity or complete inactivity. It has been

AC

reported that DTX conversion results in a obvious reduction in both cytotoxic and myelotoxic properties compared with the parent compound (Sparreboom et al., 1996). Further, these impurities demonstrate entirely different pharmacological and toxicological characteristics from DTX (Machado et al., 2008). Previous studies (Bournique and Lemarié, 2002) confirm that EDTX is a potent activator of human cytochrome P450 CYP1B1 , which is overexpressed in a variety of human tumors. Additionally, the CYP1B1 is considered to be responsible for the development of

3

ACCEPTED MANUSCRIPT anticancer drug resistance, including resistance against DTX. Simultaneously, the results of the acute toxicity of Taxotere® in a B16F10 experimental metastasis model suggest that the presence of EDTX contributed to significant weight loss, and thereby a higher systemic toxicity at a single i.v. dose of 40mg/Kg(Manjappa et al., 2013).

PT

Epimerization of the baccatin moiety at the C7 position is a common in vitro

RI

degradation pathway for taxanes (Czejka et al., 2010). It is reported that the

SC

degradation of DTX was pH dependent, and one of the primary degradation paths was the epimerization of the hydroxyl at the C7-position by way of retro aldol reaction.

NU

DTX shows improved stability in acidic pH ranges, and usually citric acid is added as

MA

a stabilizer to commercially available products, which can adjust the pH to an appropriate range to inhibit the degradation of DTX. It has also been reported that

D

EDTX is generated in heat stress, base stress and acid stress conditions (Vasu et al.,

PT E

2006), and that EDTX forms in stability samples of DTX injection. The generation of EDTX is not only in vitro but also in vivo, such as the reported study of Czejka et al

CE

(Czejka et al., 2010), where the epimerization of DTX in vivo was described. Many

AC

experiments have been performed in order to isolate and characterize EDTX (Vasu et al., 2006). In one example, the ESI mass spectrum of EDTX showed an identical protonated molecular ion at m/z 808 with DTX but a different HPLC retention time from DTX,indicating that DTX and EDTX could be isomers. The 1H NMR spectrum of EDTX and the NOESY correlations in EDTX as compared to DTX were also studied. The structures and atom numberings of DTX and EDTX are shown in Fig. 1. Taking the above findings into account, it is apparent that a safe and stable

4

ACCEPTED MANUSCRIPT formulation for parenteral administration of DTX is required to be developed, which can inhibit the conversion of DTX into EDTX during not only manufacturing and storage, but also in vivo to improve stability. Its poor solubility in aqueous solution, as well as numerous side effects limits the

PT

clinical application of DTX. Taxotere® has been reported to have many adverse

RI

reactions due to either to the drug itself or to the solvent system, such as

SC

hypersensitivity,neurotoxicity, fluid retention, neutropenia and musculoskeletal toxicity (ten Tije et al., 2003). Considering the drawbacks of Taxotere® mentioned

NU

above, alternative delivery systems such as liposomes (Immordino et al., 2003),

MA

micelles (Liu et al., 2008), inclusion compounds (Grosse et al., 1998), polymeric nanoparticles (Hwang et al., 2008), solid lipid nanoparticles (Xu et al., 2009) and

D

nanostructured lipid carriers (Li et al., 2009) have been proposed. However, none of

PT E

these previous studies considered the effect of EDTX conversion on the safety and effectiveness of the formulations, where EDTX conversion may result in decreased

CE

activity, changed pharmacological and toxicological characteristics as well as drug

AC

resistance. For enhanced delivery of DTX,the control of EDTX conversion is an important factor that must be considered. EDTX is limited to less than 0.5% in USP 35. In the US patent 20090221688A1, an organic acid with a pKa between 2.5 and 4.5 was employed as a degradation inhibitor, and was able to reduce the EDTX to between 1 and 5%. The extent of epimerization of DTX into EDTX was determined in plasma in patients receiving conventional Taxotere®. In plasma, the apparent epimerization coefficient ranged from 0.06 to 0.28, which indicates a convert ratio of

5

ACCEPTED MANUSCRIPT 6%-28%. The mean convert ratio was 14.4±11.1% (Czejka et al., 2010). Lipid microsphere (LM) has been widely used over the last few years, and Japanese researchers first proposed this concept in the early 1990s(Manfredi and Chiodo, 1998), which is also known as lipid emulsions, submicron emulsions,

PT

nanoemulsions, miniemulsions, fine dispersed emulsions, etc. LM can also be

RI

classified as: (1) LM for nutrition (phospholipid based); (2) LM as matrix for APIs

SC

(drug delivery systems); (3) LM with cationic surface charge; (4) LM with PEGylation on droplet surface; (5) LM with PEGylation on droplet surface with

NU

targeting ligand(Hörmann and Zimmer, 2015). LM is suitable as parenteral delivery

MA

systems for lipophilic active substances(Chen et al., 2015) and drugs with low chemical stability(Yanjie et al., 2014) or strong irritation(Gong et al., 2016), which led

D

to the development of numerous formulations. Natural lecithin mixtures are usually

PT E

employed as a main emulsifer, which vests LM with highly biocompatible,and such lecithin-based LM is applicable even for sensitive administration routes, such as treatment(Klang

and

Valenta,

2011).

LM

embraces

higher

CE

intravenous

AC

physicochemical stability for long-term storage in the form of liquid and they could endure thermal terminal sterilization without significant change, which can guarantee sterility well. However, some other carriers, such as liposomes, nanoparticles and micelles show poor physicochemical stability in the form of liquid for long-term storage, and most of them could not tolerate thermal terminal sterilization, so they usually employ freeze-drying technique to improve storage stability and conduct strict aseptic operation to ensure sterility，which makes sterility assurance level (SAL) to be

6

ACCEPTED MANUSCRIPT three ordres higher than thermal terminal sterilization. All these excellent features make LM be an attractive carrier for i.v. administration of DTX. The superiority of LM in this case is that DTX can be incorporated into the oil phase and interfacial film,which isolates the drug from acid, alkali, and heat stimulation, and therefore LM

PT

can effectively improve the chemical stability and inhibit the generation of EDTX in

RI

vitro. In vivo, LM can prevent DTX from coming into direct contact with the blood,

SC

and protect the drug in oil from bioconversion while in the blood circulation (due to drug not being released into the blood) for an extended period.

NU

In this work, a novel docetaxel lipid microsphere(DTX-LM) was designed and

MA

prepared by high-pressure homogenization. The factors affecting EDTX conversion, such as pH value,sterilization conditions, oil phase, emulsifier compositions and

D

acidity regulators were investigated in detail to optimize the formulation. Following

PT E

this, in vitro drug release and in vitro stability studies, including degradation kinetics, storage stability and plasma stability were carried out to evaluate EDTX conversion of

CE

the optimum DTX-LM formulation.Finally, the conversion of DTX into EDTX in rats

AC

was determined to assess the safety and effectiveness of DTX-LM in comparison with Docetaxel for Injection (DI).

2. Materials and methods 2.1. Materials The following materials were provided by or purchased from the sources in parentheses: DTX and paclitaxel (Shanghai sanwei Pharma Co. Ltd.Shanghai, China). Egg lecithin(Lipoid E80, S75) and oleic acid (Lipoid GmbH ,Shanghai Toshisun

7

ACCEPTED MANUSCRIPT Biology & Technology Co., Ltd). Egg lecithin(PL-100M) (Shanghai Advanced Vehicle Technology Ltd. Co., Shanghai,China). Long-chain triglyceride (LCT) and medium-chain triglyceride (MCT) (TieLing BeiYa Pharmaceutical Corporation, TieLing,China). Glycerin for intravenous use (Zhejiang Suichang Huikang Pharma Co. Zhejiang,China).

Poloxamer

188(Pluronic

F68)(BASF

PT

Ltd.

Co.

Ltd.

Chengdu,China).

Citric

acid,

caproic

acid

and

SC

Excipients

RI

AG ,Ludwigshafen,Germany). Sodium bisulfite (Chengdu Huayi Pharmaceutical

tartaric(Hongshuncheng Co.Ltd. Tianjin,China). Docetaxel for Injection (DI) (Jiangsu

NU

Hengrui Medicine Co. Ltd. Lianyungang,China). Tert-butyl methyl ether (TBME)

MA

(Sinopharm Chemical Reagent Co.Ltd. Shenyang,China). Formic acid (Dima Technology Inc. Richmond Hill,USA). Dehydrated alcohol, acetonitrile and methanol

D

(Tianjin Concord Technology Co.Ltd. Tianjin,China). All other reagents and

PT E

chemicals were of chromatographic or analytical grade. 2.2. Preparation of DTX-LM

CE

DTX-LM with a concentration of 0.8mg/mL was prepared by high-pressure

AC

homogenization. 0.08% (w/v) DTX and 3.0% (w/v) E80 were dissolved in about 5% (v/v) of dehydrated alcohol. The oil phase consisted of 0.025% (w/v) oleic acid, 2.5%(w/v) LCT and 7.5%(w/v) MCT was heated at 70℃, and then added to the phospholipid complex to form a transparent solution under stirring. 0.02%(w/v) sodium bisulfite, 2.5%(w/v) glycerin and 0.2%(w/v) F68 were dissolved in water with agitation at 70℃ to acquire the aqueous phase. The oil phase was added slowly to the aqueous phase with continuous stirring of high speed shear mixing(ULTRA

8

ACCEPTED MANUSCRIPT TURRAX® IKA® T18 basic,Germany) at 16000 rpm for 3 min to obtain the coarse emulsion. The volume was added up to 100% with purified water. After that, the final emulsion was gained by subjecting the coarse emulsion to a high-pressure homogenizer (Pharmaceutical ultra-high-pressure homogenizer of AH100D,ATS

PT

Engineering Inc. Shanghai,China) at 800 bar for 8 cycles controlling the temperature

RI

under 40℃. The pH of the final emulsion was adjusted to 5.3 with 0.1 mol/L NaOH.

SC

Finally, the DTX-LM was transferred to vials and flushed with nitrogen gas followed by autoclaving at 121℃ for 8 min. It should be noted that the metric for successful

NU

formulation in this work was based on EDTX conversion as the main indicator to

2.3. Characterization of DTX-LM

MA

evaluate the formulation and process of LM.

D

The appearance of DTX-LM is a uniform milky liquid. The microscopic

PT E

observation was conducted by use of BS230 optical microscope(Chongqing COIC Industrial Co., Ltd. Chongqing,China). The pH within DTX-LM was measured with

CE

PB-10 digital pH meter (Sartorius of Germany). The particle size distribution（PSD）

AC

of DTX-LM was measured by dynamic light scattering (DLS) in a NicompTM 380 Particle Sizing system (Zeta Potential/Particle Sizer NICOMPTM 380ZLS, Santa Barbara, California, USA). The entrapment efficiency (EE) of DTX-LM was calculated using ultrafiltration, which was operated five times at 3000 rpm for 40 min using ultrafiltration centrifuge tubes with a molecular weight cut-off of approximately 100KD (Beijing Genosys Tech-Trading Co. Ltd., Beijing, China). This method was developed and validated,and specificity, linearity and range(R2=0.9996, 0.02632μ

9

ACCEPTED MANUSCRIPT g/mL～21.05μg/mL) , recovery(98.6%~99.5%，RSD≤2.0%) and precision(RSD 0.04%) all met the requirments. The EE (%) was calculated by the following Eq.(1) (Groves et al., 1985): CtotalVtotal  CwaterVwater ×100 CtotalVtotal

(1)

PT

EE (%) 

Ctotal= the concentration of DTX in DTX-LM,

Vtotal=the volume of DTX-LM ,

NU

Vwater=the volume of external water phase.

SC

RI

Cwater=the concentration of DTX in external water phase,

2.4. Analyses of DTX and EDTX by HPLC

MA

The analyses of DTX and EDTX referred to USP38 Monographs “Docetaxel Injection” and they were detected by a reverse phase HPLC (RP-HPLC) system

PT E

D

(Hitachi Company,Japan) which was interfaced with Chromaster software along with a Luna C18 column (250×4.6mm, 5um; Phenomenex Corporation, Torrance,

CE

America). The DTX and EDTX chromatographic conditions referred to the section of “Docetaxel Injection” in USP.

AC

The percentage of EDTX in the portion of injection was calculated according to the following Eq.(2): EDTX（%） (rU / rT ) × (1/F) ×100

rU = peak area of EDTX from the sample solution rT = sum of all of the peak areas from the sample solution F = relative response factor for EDTX (F=1) 2.5. Drug release in vitro 10

(2)

ACCEPTED MANUSCRIPT DTX release from LM and DI was studied using the dialysis method(Ma et al., 2014; Wang et al., 2006). Samples equivalent to approximately 0.13 mg DTX were added to dialysis membranes (MWCO14000, Beijing Ruida Henghui Technology Development Co. Ltd., Beijing, China), which were then submerged in 10 ml PBS

PT

(pH 7.4) containing 0.5% Tween 80 in a flask with continuous vibration at 100 rpm.

RI

The temperature was maintained at 37C using a water bath. At predetermined

SC

intervals (0.25, 1, 2, 4, 6, 8, 12, 24, 48, 72 hours), aliquots were withdrawn and 10 ml fresh medium were added. The collected sample solutions were filtered through a 0.22

NU

µm membrane for HPLC analysis in triplicate. The percentage of DTX released

MA

versus time was then calculated. 2.6. Stability study

PT E

2.6.1.1. Effect of pH

D

2.6.1. The degradation kinetics of DTX

A series of DTX-LM (0.8 mg/mL) with pH values ranging from 4.0-9.0 were

CE

prepared by adjusting the final emulsions with 0.1 mol/L NaOH or HCl solution.

AC

Similarly, DI with pH values between 4.0-9.0 were obtained. All samples were simultaneously placed in a temperature constant oven at 60C. At predetermined time points (1, 3, 5, 10d), samples were removed and cooled to room temperature instantly, and the content of DTX and EDTX was determined by HPLC to determine the optimal pH of the formulations. 2.6.1.2. Effect of temperature The effect of temperature on the degradation of DTX in LM and DI with a pH of

11

ACCEPTED MANUSCRIPT 5.3 and 4.0 respectively was investigated at three different temperatures (40, 60 and 80C). At time intervals of 2, 4, 6, 8, 10, 12, 24, 48, 72, 96 hours, samples were withdrawn and cooled immediately at once to terminate the reaction. The results of the content of DTX and EDTX from HPLC were used to calculate the degradation

PT

rate constant and activation energy that characterizes the degree of stability.

RI

2.6.2. Storage stability test

SC

The storage stability test at 6ºC and 25ºC is employed to simulate the storage condition at refrigerate and room temperature, respectively. Because of the

NU

thermodynamic instability of LM, DTX-LM is proposed to be stored at refrigerated

MA

condition. The temperature of 6ºC and 25ºC could be used to carry on the long-term testing and accelerated testing. Three batches of the optimal DTX-LM was newly

D

prepared and stored at 6±2 C for 24 months and 25±2C for 6 months. At

PT E

predetermined intervals, samples were removed, and the content of EDTX was monitored to evaluate the DTX-LM chemical stability.

CE

2.6.3 Plasma stability test in vitro

AC

In the human body, the CYP3A family plays a key role in the metabolism of DTX. CYP 3A4 and 3A5 isoenzymes mediate the primary and subsequent oxidation of DTX. It has been reported that EDTX is formed from DTX metabolism by human liver microsomes (Ringel and Horwitz, 1987; Ringel and Horwitz, 1991), but the epimerization is not inhibited with the presence of 3A4 inhibitor. Thus, epimerization appears to be independent of P450 catalysis (Shou et al., 1998). The specific enzymes for catalyzing the epimerization of DTX have been not reported in the literatuer. It is

12

ACCEPTED MANUSCRIPT proposed that degradation may occur due to the neutral environment and some epimerase of plasma. Based on this, plasma was used as the medium to investigate the chemical stability of DTX-LM. DI was prepared as follows: 10 mg citric acid and 100 mg DTX was dissolved in

PT

5 mL ethanol and the mixture was made up to 10 mL with Tween 80. 0.8 mL of the

RI

solution was diluted to 10 mL with saline to obtain a solution containing DTX 0.8

SC

mg/mL. DTX-LM was prepared according to section 2.2.

200 µl DTX-LM and DI were added to 2 mL blank plasma preheated at 37℃,

NU

shaken well and placed in a 37℃ shaking bath. 100 µl samples were collected at

MA

times 0, 1, 2, 4, 6, 8, 12, 24, 36, 48, 60 and 72 hours.

The plasma samples treatment and analysis were conducted as per the methods in

D

sections 2.7.3 and 2.4 respectively.

2.7.1. Animals

PT E

2.7. Pharmacokinetics study in rats

CE

Sixteen pathogen-free Sprague-Dawley rats (eight male and eight female)

AC

weighting 180-220 g were kindly provided by the Experimental Animal Center of Shenyang Pharmaceutical University. Animals were fasted for 24h before experiments. All animal studies were carried out in compliance with the Guideline for Animal Experimentation approved by the Animal Ethics Committee of the institution. 2.7.2. Analytical methods A previously described UPLC-MS/MS analytical method (Zhao et al., 2010) was further improved and validated for determining the plasma concentrations of DTX

13

ACCEPTED MANUSCRIPT and EDTX in DTX-LM as well as in DI. Liquid chromatographic separation was performed on an ACQUITY Ultra-Performance Liquid Chromatography system (Waters Corp., Milford, MA, USA) using a XBridgeTM C18 column (75 × 4.6 mm, 2.5µm, Waters Corp., Milford, MA, USA). Mobile phase were acetonitrile and 0.1%

PT

formic acid solution.The elution procedure are shown as follows: T(min)/

RI

acetonitrile %=0/30,0.6/70,7.5/70, 8.0/30, 8.5/30. The parameters of Xevo TQ mass

SC

spectrometer were the same as described in the literature. Quantitative analysis was performed in multiple reaction monitoring (MRM) mode with a transition at 830.22→

NU

549.14 for DTX [M+Na]+ and 876.20→591.13 for paclitaxel [M+Na]+. The mass

MA

spectrometric data was collected using MasslynxTM NT4.1 software (Waters Corp., Milford, MA, USA) and analyzed by the QuanLynxTM program ((Waters Corp.,

PT E

2.7.3. Sample preparation

D

Milford, MA, USA).

Plasma samples (100 µl) were mixed thoroughly with both 20 µl of internal

CE

standard (IS) solution and methanol on a vortex for one minute, and then extracted

AC

with 3 mL tert-butyl methyl ether (TBME) by vortexing for 10 min. After centrifugation at 4,000 rpm for 10 min, 2 mL of supernatant was evaporated to dryness at 35C with a gentle stream of nitrogen. 200 µl of methanol was then added followed by vortexing for 10 min and centrifugation at 12,000 rpm for 10 min. Finally, 60 µl aliquots were transferred for characterization.

14

ACCEPTED MANUSCRIPT 2.7.4. Drug administration and sampling Sixteen SD rats were randomly divided into two groups with eight (four male and four female) in each. The clinical recommended dose of Taxotere® is 75 mg/m2, and so the intravenous dose given to rats was 10mg/kg as a result of conversion from

PT

the skin surface area. One group was administered with DTX-LM via the tail vein,

RI

and the other group received the corresponding dose of DI. At predetermined time

SC

points (0, 5, 15, 30 and 60 min), blood samples of approximately 0.5mL were collected by suborbital puncture into heparinized centrifuge tubes, and immediately

NU

centrifuged at 8,000 rpm for 10 min. Harvested plasma samples were then stored at

MA

-20C until analysis. 2.7.5.Data analysis

D

Pharmacokinetic analysis was conducted by drug and statistics (DAS) version

Shanghai,China).

PT E

2.0 software (Mathematical Pharmacology Professional Committee of China,

CE

The convert ratios were calculated by the following Eq.(3): (3)

AC

Convert ratio（%） CEDTX / CDTX × 100

where CEDTX represents the concentration [μg/ml] of EDTX and CDTX represents the amount of the parent compound DTX [μg/ml]. The AUC ratios of EDTX to DTX were also compared between DTX-LM and DS at different intervals to investigate the EDTX conversion difference. 2.8. Statistical analysis Data were presented as mean ± standard derivations (SD) of independent

15

ACCEPTED MANUSCRIPT experiments. Statistical analysis was performed using one-way ANOVA and paired Student’s t-test by the statistical package for social science (SPSS, version 11.5,SPSS Inc., Chicago, IL, USA). Differences were considered significant at P＜0.05.

3. Results and discussion

PT

3.1 The appearence and optical micrograph of DTX-LM

RI

The physical appearance and optical micrograph of DTX-LM has been shown in

SC

Fig.2. DTX-LM is a uniform milky liquid without flocculation. As shown in the optical micrograph, DTX-LM is spherical or near-spherical and droplets aggregation

NU

does not appear. There is no droplets over 5 µm are detected, which can make oil

MA

droplets across the smallest blood vessels which exists in the lung with diameters less than 5μm, so that blood vessels embolisms can be avoided .

PT E

3.2.1. Oil phase composition

D

3.2. Investigation of DTX-LM formulation

An increase in EDTX conversion results from the loss of protection of LM'

CE

encapsulation, which is induced as a result of poor physicochemical stability. It has

AC

been reported that the composition of the oil phase plays an important role in the physicochemical stability of parenteral LM (Jumaa and Müller, 1998) and parenteral nutrition (Yang et al., 2016). Owing to a series of shortcomings of LCT and MCT alone, an LCT-MCT mixture was selected as the oil phase. The MCT/LCT mixture could decrease conversion to EDTX over LCT alone, as pure LCT contains more unsaturated bonds which are liable to be oxidized to produce free radicals and peroxides, which can in turn accelerate DTX epimerization. In addition, the poor

16

ACCEPTED MANUSCRIPT solubility of DTX in LCT (about 2.17±0.12 mg/g) and the larger in MCT (about 7.93±1.96 mg/g) was also used to determine the combination to achieve a high entrapment efficiency. LCT and MCT in combination could contribute to a lower energy barrier (Teng et al., 2014) and provide more stable all-in-one mixtures

PT

(Smyrniotis et al., 2001). According to the preliminary results, the ideal oil phase

RI

composition was determined to be MCT:LCT = 3:1.

SC

The oil phase ratios have an impact on the encapsulation of DTX, and potentially a change in the oil film resistance to thermal sterilization, which makes a difference in

NU

terms of EDTX conversion. The oil phase ratios of 10% and 15% in DTX-LM with

MA

LCT-MCT 1:3 were studied in order to investigate the effect of a higher proportion of oil phase on the formation of EDTX. The results showed that the content of EDTX of

D

the 15% oil phase was approximately 1.5 %, which is beyond the 0.5 % standard limit

PT E

of EDTX in USP 35. This indicates that the conversion of DTX cannot be reduced by increasing the ratio of the oil phase, but rather that excess oil may result in a larger

CE

particle size and standard deviation due to the presence of small free oil drops.

AC

Therefore, the 10 % oil phase with a 1:3 mixture of LCT and MCT in the DTX-LM formulation was the optimal formulation. 3.2.2. Emulsifier composition The emulsification process and the emulsifier greatly influences the stability of the emulsion (Yamano and Seike, 1983). In DTX-LM formulations, lecithin was chosen as the major emulsifier due to its known biocompatibility. As reported, lecithin is comprised of two main components, phosphatidylethanolamine (PE) and

17

ACCEPTED MANUSCRIPT phosphatidylcholine (PC) (Yanjie et al., 2014), and the emulsifying capacity of PE is higher than PC. Therefore, lecithins with a high PE ratio such as E80 (82% PC, 9.2% PE),S75 (72.7% PC, 9.4% PE) and PL-100M (78% PC, 18% PE) were chosen as emulsifiers to improve the physical stability of the formulations to reduce the

PT

formation of EDTX. The results suggest that DTX-LM prepared with PL-100M could

RI

not withstand autoclave sterilization. This is perhaps due to the weak emulsifying

SC

character of PL-100M at a low pH, resulting in physical instability. DTX-LM prepared with E80 had a lower EDTX content (0.45±0.04%) compared with LM

NU

using S75(0.79±0.07%). E80 is more acidic than S75 and PL-100M,and so the final

MA

emulsion had a lower pH. This contributed to a more stable environment for DTX and decreased the conversion in the DTX-LM formulation. As a result, E80 was chosen as

D

the main emulsifier for further studies.

PT E

3.2.3. Investigation of different types of acids It has been reported that the incorporation of an organic or inorganic acid can

CE

improve the stability of DTX even after heating at high temperatures(AG, 2008) .

AC

Based on this, four different acids were used to adjust the pH value of the formulations, and the content of EDTX was regarded as the metric for investigation. As shown in Table 1, NaHSO3 contributed to the lowest content of EDTX, indicating it could more successfully inhibit EDTX conversion. Additionally, NaHSO3 can be used as an antioxidant in order to prevent the drug from oxidizing during the sterilization process. Therefore, NaHSO3 was selected as both the acidity regulator and antioxidant.

18

ACCEPTED MANUSCRIPT After this, the inclusion of different amounts of NaHSO3 was investigated as compared to formulations without NaHSO3. As indicated in Table 2, the larger the amount of NaHSO3 added in the formulation, the lower the epimerization of DTX. It was also apparent that the higher pH values of the final emulsions resulted in a higher

PT

conversion of DTX. The content of EDTX was able to be controlled to below 0.5 %

RI

when at least 0.01% NaHSO3 was included. Despite the increased content of NaHSO3

SC

reducing the EDTX conversion, the PSD of the formulation simultaneously became larger. When 0.02% NaHSO3 was added,the PSD was 154.5±52.1 nm. But the PSD

NU

increased to 198.2±74.3 nm when 0.05% NaHSO3 was used. This is likely because

MA

NaHSO3 as an electrolyte can increase the zeta-potential followed by flocculation of the system, which is not favorable for the physical stability of DTX-LM. Thus, taking

D

both factors into account, the ideal amount of NaHSO3 was determined to be 0.02%.

PT E

3.2.4. The correlation between pH and DTX conversion in LM The pH value has a critical effect on the formation of EDTX, as DTX is most

CE

stable within an acidic pH range of 4.0-6.0. As shown in Table 2, the pH of all

AC

formulations tested had a significant impact on the conversion of DTX. The pH values of the final emulsions were adjusted to 4.0, 4.5, 5.0, 5.3, 6.0 and 6.5 with 0.1 mol/L NaOH, and the pH value before and after thermal sterilization and the content of DTX and EDTX were determined in order to investigate and confirm the function of pH in inhibiting generation of EDTX. Based on the results shown in Table 3,the content of EDTX increased as the pH also increased. Over a pH range from 4.0 to 6.0, the content of EDTX could be contained within the standard limit (0.5%). Considering

19

ACCEPTED MANUSCRIPT the changes in DTX content, the pH of the final emulsion (before sterilization) was adjusted to pH 5.3. It was noteworthy that the pH value was significantly decreased after the sterilization process, which could be attributed to NaHSO3 being oxidized to produce

PT

H+ (2HSO3- +O2→2H+ +2SO42-). It is known that the pH of LM for i.v. injection

RI

should be between 4 to 9 as the pH value of human blood is 7.4(Lu et al., 2008). For

SC

DTX-LM, sterilization dramatically affected the pH of formulation, and only when the pH of the final emulsion was adjusted to 6.5 or more could the pH value after

NU

sterilization remain greater than 4.0. However, this process meant that the content of

MA

EDTX was as high as up to 0.9 %. Despite this, it has been reported that the pH value of glucose injections in quality standard of Ch.P 2010 is 3.2-6.5, indicating that a pH

PT E

was defined as 3.5-5.5.

D

below 4.0 could be tolerable. Based on this, the pH of DTX-LM (after sterilization)

3.2.5. The effect of temperature on DTX conversion in LM

CE

To ensure that the formulation is appropriate for clinical application, LM intended

AC

for i.v. injection requires a terminal sterilization with F0 value greater than 8. LM is a thermodynamic instability dispersed system and thermal sterilization will compromise the physical and chemical stability of LM. Meanwhile the EDTX conversion will also be influenced by the sterilization process. So bioburden based process rather than the overkill method was employed for the steriliazation. Provided the sterilization effects are comparable, the sterilization method with the higher temperature and shorter time is favored, because it can better maintain the

20

ACCEPTED MANUSCRIPT stability of the drug due to the lower activation energy (Shi et al., 2009). Experimental results showed that DTX-LM could remain physicochemically stable during the autoclaving process, and therefore sterilization in a 121℃ autoclave was selected as the method for further studies. The different sterilizing time of 8,10,12,15 min were

PT

investigated and the results indicated that the amount of EDTX increased as the

RI

sterilization time increased, which is likely due to the long time under high

SC

temperatures destroying the phospholipid structure, thus reducing its emulsifying ability, followed by increase in DTX epimerization. The content of EDTX was within

NU

the ideal range limit (0.5%) when the sterilization time was controlled within 8 min.

MA

In total, the ideal DTX-LM thermal sterilization conditions were defined to be autoclaving at 121℃ for 8 min. Except for the 8 minutes at 121℃, the process of

D

heating and cooling also produce sterilization effect, so the F0 value will exceed 8

PT E

minutes. After calculation, the F0 was 10.05～10.35 minutes which made SAL to be not more than 10-6, and the sterility of the final preparation gets guarantee.

CE

3.3.Drug release in vitro

AC

DTX is almost completely insoluble in water, and so 0.5 % Tween80 was added to experiments to maintain sink conditions with a solubility of around 40.36 ug/mL. As shown in Fig. 3, DTX-LM showed decreased drug release obviously compared with DI (p<0.05).The cumulative drug release of DI was up to 70.08±4.77 % in 1 hour and quickly reached the drug release equilibrium, while that of DTX-LM was only 5.19±0.18 % in this same time frame. The percentage of released DTX in DTX-LM basically reached steady state until 72 hours. This is because the molecular

21

ACCEPTED MANUSCRIPT weight of free DTX is low, and can thus easily pass through the dialysis membrane; however more than 90% of the drugs in DTX-LM are entrapped in the lipid core or the interface layer and so their release is reliant on corrosion and degradation of the carrier. This indicated that DTX-LM can show a sustained release effect, which makes

PT

DTX go into blood slowly, and so more DTX is protected from the stimulating

RI

environment in the blood. It would better control the release of DTX in blood and

SC

maintains the maximum amount of active DTX instead of EDTX. This could inhibit

3.4.Stability study

MA

3.4.1.The degradation kinetics of DTX

NU

the EDTX conversion in vivo , which should be reflected in pharmacokinetic studies.

3.4.1.1.Effect of pH

D

The degradation of DTX in LM and DI at 60C over a pH range of 4.0-9.0 was

PT E

investigated. As shown in Fig.4, the degradation reactions of DTX in LM and DI could all be fitted with a pseudo first-order kinetics model. From the pH-rate profiles

CE

displayed in Fig.5, the pHm of DTX in DI and LM were 4.05 and 6.98, respectively.

AC

This result of the pH 6.98 formulation improved over the pH 5.3 formulation that has been identified as optimal above, and this can be explained by the analytical method and purpose. The research aims to inhibit the EDTX conversion on account of a series of side effects of EDTX. Although decreased DTX degradation is also expected, EDTX generation is the research focus, which is one of the greatest influence factors on efficacy and safety among DTX impurities. This result from pH 6.98 was only analyzed from the quantity of DTX degraded, whereas earlier the pH 5.3 formulation

22

ACCEPTED MANUSCRIPT was shown to be more favorable when taking into account both DTX degradation and EDTX generation shown in Fig.6. Statistical analysis was performed using one-way ANOVA, and there were significant differences at pH 5.3, pH 6, pH 7 and pH 8 (p<0.05) except pH 4 and pH 9 (p>0.05). As shown in Fig.6, the content of EDTX in

PT

LM is much lower than that in DI during the whole experiment, and DTX-LM has a

RI

wider pH range where EDTX (%) stays at low levels than DI. These results conclude

SC

that LM can greatly improve the chemical stability of DTX and decrease EDTX conversion.

NU

3.4.1.2.Effect of temperature

MA

The effect of temperature on the degradation of DTX in DI with pH 4.0 and in LM with pH 5.3 was investigated at 40, 60 and 80C. The DTX degradation process

D

followed pseudo first-order kinetics as shown in Fig.7，and the slope of the curve is

PT E

degradation rate constant k. Arrhenius plots based on k of DI and LM at different temperatures were shown in Fig. 8. Then the activation energies (Ea) of DI and LM

CE

were calculated from the Arrhenius equation (ln k = ln A – Ea/RT)，which gave a

AC

result of 18.89 and 19.58 kJ/mol respectively, while the frequency factors (A) were 6.28 and 7.44/h respectively. The close Ea values of DTX degradation reaction in DI and in LM indicate that LM may be improving the stability of DTX by loading the drug within its lecithin layer, rather than altering the degradation mechanism. 3.4.2.Storage stability test It is known that LM is thermodynamically unstable as characterized by coalescence, flocculation, creaming and degradation or precipitation of drugs during

23

ACCEPTED MANUSCRIPT storage , and thus the DTX-LM should be assessed in practical conditions over long times to analyze stability. The results showed that DTX-LM possessed good physical stability and could be stored for at least 24 months at 6±2℃ and 6 months at 25± 2℃. Particular attention was paid to the EDTX conversion in our study , and this data

PT

is summarized in Table 4. It can be seen that EDTX was effectively inhibited during

RI

storage. The EE was maintained at over 99% for several months, and so a lesser

SC

amount of drug went into the external phase and therefore there was less to be converted. It was observed that the pH of DTX-LM was gradually reduced and the

NU

EDTX content was increased with the extended periods of storage. The reason for the

MA

decrease of pH was considered to be the chemical stability of oil and lecithins in DTX-LM formulations. These two components belonged to glycerides and the

D

unsaturated fatty acids in oils and lecithins are more active than the inert saturated

PT E

fatty acids, which makes it more liable to be hydrolyzed and oxidized. The hydrolysis process involves breaking of the ester bond to generate free fatty acids, which will

CE

contribute to decreasing pH values. NaHSO3 which was oxidized to produce H+, also

AC

slightly contributed to the decreasing pH. The lowering pH resulted in further hydrolyzation of the oil phase and lecithins which resulted in decreased stability and EE of LM, and further caused EDTX conversion. The oxidation process was due to rupture of the double bonds in the unsaturated fatty acids. This process induced the release of peroxides and radicals(Chen et al., 2015; Gong et al., 2016; Klein, 1970) and these substances accelerated the conversion of DTX into EDTX.The DTX epimerization mechanism in vitro is shown in Fig.9 (A).

24

ACCEPTED MANUSCRIPT 3.4.3. Plasma stability test in vitro As described in Section 2.6.3，the apparent epimerization coefficients (Repi) of DTX-LM and DI were measured by HPLC to evaluate the chemical stability of DTX in plasma. The Repi was calculated by dividing the concentration of EDTX by the

PT

parent compound DTX. This could be simplified to be the ratio of peak area of EDTX

RI

and DTX bacause of the same relative response factors of EDTX and DTX. The

SC

results can be seen in Fig. 10, which show that the Repi is gradually increased with the incubation time, which represents an increased conversion of EDTX. The Repi of

NU

DTX-LM was significantly smaller than DI (p<0.05). DTX-LM showd an excellent

MA

effect on inhibiting EDTX generation and the Repi of DTX-LM and DI at 72 hours were 12.78% and 67.96% respectively. In DI, was rapidly over the first 24 hours and

D

showed a small increase over later times. The majority of DI is comprised of free drug,

PT E

and thus all DTX is in contact with plasma in a short time, which leads to a faster epimerization. DTX-LM possessed a lower epimerization rate and degree and

CE

achieved a relative balance at a later time , which may be attributed to the efficient

AC

entrapment of LM delivery system for DTX, meaning less of the drug would be exposed to plasma. In DTX-LM, the drug would release slowly from the oil phase to the water phase and plasma, where the free drug is then prone to being epimerized, and so the epimerization rate would be decreased in comparison. In brief, DTX-LM showed better plasma stability and decreased epimerization in vitro than DI, and this result could be employed to predict behavior in vivo.

25

ACCEPTED MANUSCRIPT 3.5.Pharmacokinetics study 3.5.1.Method validation A selectivity test showed that endogenous substances did not interfere with the determination of DTX (6.29min), IS (6.27min) and EDTX. The results of the matrix

PT

effect showed that all the ratios were in the range of 90%-110%, which suggested that

RI

there was no significant matrix effect for DTX, EDTX or IS. The calibration curves of

SC

DTX and EDTX were linear over the plasma concentration range of 10.0-400 µg/mL and 0.10-4.00 µg/mL respectively.The lower limit of quantitation(LLOD) of DTX was

NU

0.01 mg/mL, while that of EDTX was 0.1 ug/ml.The precision, accuracy and recovery

MA

of DTX or EDTX were within the acceptable limits (RE%:±15%;RSD%:15%). The extraction recoveries of DTX, EDTX and IS at different concentration levels were all

PT E

3.5.2.Pharmacokinetics study

D

greater than 60%.

DTX epimerization has been implicated in decrease of activity of the drug as

CE

well as in the development of anticancer drug resistance. From this point of view it is

AC

significant to know the amount of DTX epimerization in vivo. The pharmacokinetics study conducted here in rats was designed to evaluate the conversion of DTX in LM into EDTX in vivo in comparison with that in DI. The mean plasma concentration-time curves of DTX and EDTX were presented in Fig. 11, and the plasma concentrations of DTX and EDTX for each individual rat at the end of i.v. administration were shown in Table 5. At the same time, convertion ratio were also calculated and summarized in Table 5. The plasma concentration-time curves of

26

ACCEPTED MANUSCRIPT DTX-LM and DI were fitted with two-compartment model. The pharmacokinetics parameters of DTX-LM and DI were listed in Table 6, and they had significant difference (p<0.05) except T1/2 . The AUC、CL and Vss of DTX-LM were 2.7, 0.3 and 0.3 times DI, respectively. The results shows that following intravenous

PT

administration of DTX-LM and DI (10 mg/kg) in rats, the plasma drug concentration

RI

of DTX and EDTX reaches a peak rapidly, and then quickly decreases. Two blood

SC

samples were selected: pre dose for control purposes and the point at the end of i.v. administration for investigation of conversion ratios, which should avoid times when

NU

the amount of EDTX is too little to be detectable. DI contains no EDTX before dosing,

MA

whereas 3.04% EDTX conversion was detected at the end of i.v. administration of DI. This indicates that DTX can convert to EDTX in vivo, mainly in the blood, but

D

perhaps also in other organizations such as the liver (Ringel and Horwitz, 1987;

PT E

Ringel and Horwitz, 1991), which needs further study. For LM, there was approximately 0.61% of DTX in LM converting to EDTX, which is much lower than

CE

that of DI.

AC

To study the EDTX conversion pharmacokinetics further, the AUC ratios of EDTX to DTX were compared between the two formulations at different intervals.As shown in Table 7, DTX was evenly converted to EDTX in both formulations. However, the AUC ratios of DI were higher than that of DTX-LM at each interval and there was significant difference between two sets of data by use of paired t-test(p<0.01), indicating that DTX-LM has a lower conversion rate. DTX in the blood (free DTX) was more easily able to be epimerized than

27

ACCEPTED MANUSCRIPT encapsulated DTX in LM. Due to the instability of DTX in blood, the more the DTX was released into blood circulation, the more DTX converted into EDTX. More than 90% of the drugs in DTX-LM were entrapped in the lipid core or the interface layer and so DTX release is reliant upon two main mechanisms: (1) the diffusion process

PT

(Salmela and Washington, 2014). , where DTX-LM achieves sink conditions after

RI

dilution into the blood, and DTX establishes a partition equilibrium between oil phase

SC

and blood which causes DTX to release into blood slowly and (2) corrosion and degradation process of the LM skeletal structure. LM forms a barrier to prevent the

NU

DTX release, and the release is therefore reliant on liposis of lipoprotein lipase

MA

(Kawakami et al., 2000).

The different conversion mechanism in vivo of the two formulations may

D

contribute to the differences of EDTX conversion of DI and DTX-LM. Most of the

PT E

DTX in LM adsorbs on the interfacial film, and thus the diffusion from LM and distribution in blood is slow. As well, the barrier action of LM also means DTX needs

CE

some time to move out from the LM into blood, and then conversion into EDTX can

AC

occur. The half-life of Tween -80 in DI is extremely short, and the micelles formed will be destroyed quickly, which will accelerates DTX release. The released DTX will bind to plasma proteins, whereas the Tween-80 catabolite oleic acid has a higher plasma protein affinity than DTX, so DTX is quickly replaced and can go on be converted. The mechanism of decreased EDTX conversion in vivo in LM is shown in Fig.9 (B). Due to the correlation of the two, the mechanism of DTX release in vitro can

28

ACCEPTED MANUSCRIPT also be employed to explain the EDTX conversion in vivo. DTX released from LM was found to be significantly less than DI within 1 hour in vitro. This release pattern could be reflected in vivo. DTX-LM was able to prevent more DTX going into contact with the blood because of the depot effect, so the convert ratio of EDTX was only

PT

0.61%. However, DI had a high convert ratio of 3.04%, as most of the DTX was

RI

released into the blood within a short time.

SC

4. Conclusion

In summary, a novel intravenous DTX-LM was developed to overcome

NU

epimerization, which is the main difficulty associated with chemical instability of

MA

DTX. The results of drug release in vitro and stability studies suggest that pH and temperature are the two most important factors determining EDTX conversion and

D

that DTX-LM could inhibit epimerization effectively in vitro. Quantitative EDTX

PT E

conversion in vivo was also investigated in our study, and pharmacokinetics showed that DTX-LM had a lower EDTX conversion than DI in vivo. Therefore our

CE

formulation had less EDTX conversion in LM in vivo, decreased toxicity and tumor

AC

resistance induced by EDTX compared to Taxotere®. Our research results are consistent with existing in vitro and in vivo experiments: EDTX is a vital metabolite of DTX and it can convert to EDTX not only in vitro but also in vivo. It is significant to know the amount of DTX epimerization, particularly in the blood from pharmacokinetic perspectives, as it may have great bearing on the development of new taxane derivatives and new formulations of DTX. The control of epimerization is essential to ensure a safe and effective treatment of this class of

29

ACCEPTED MANUSCRIPT chemical compounds. This work will afford valuable information for the research and development of formulations of taxane derivatives.

Acknowledgments Amanda Pearce is gratefully thanked for correcting the manuscript. We would

PT

like to thank the reviewers for their critical comments and constructive suggestions

RI

which benifit the improvement of the study.

SC

Declaration of interest

The authors have no relevant affiliations or financial involvement with any

NU

organization or entity with a financial interest in or financial conflict with the subject

AC

CE

PT E

D

MA

matter or materials discussed in the manuscript.

30

ACCEPTED MANUSCRIPT

References AG, S., 2008. Pharmaceutical composition of improved stability containing taxane derivatives. EP.

PT

Bournique, B., Lemarié, A., 2002. Docetaxel (Taxotere) Is Not Metabolized by Recombinant Human CYP1B1 in Vitro, but Acts as an Effector of This Isozyme. Drug

RI

Metabolism and Disposition 30, 1149-1152.

SC

Chen, X., Zhang, L., Hu, X., Lin, X., Zhang, Y., Tang, X., 2015. Formulation and

NU

preparation of a stable intravenous disulfiram‐loaded lipid emulsion. European Journal of Lipid Science & Technology 117, 869-878.

MA

Clarke, S.J., Rivory, L.P., 1999. Clinical pharmacokinetics of docetaxel. Clinical Pharmacokinetics 36, 99-114.

PT E

D

Czejka, M., Greil, R., Ulsperger, E., Schnait, H., Kienesberger, K., Brumnik, T., Farkouh, A., Schierholz, J., 2010. Evidence for the conversion of docetaxel into

CE

7'-epidocetaxel in patients receiving Taxotere-based conventional chemotherapy. International Journal of Clinical Pharmacology & Therapeutics 48, 483.

AC

Gong, H., Geng, S., Zheng, Q., Wang, P., Luo, L., Wang, X., Zhang, Y., Zhang, Y., He, H., Tang, X., 2016. An intravenous clarithromycin lipid emulsion with a high drug loading, H-bonding and a hydrogen-bonded ion pair complex exhibiting excellent antibacterial activity. Asian Journal of Pharmaceutical Sciences 11, 618-630. Grosse, P.Y., Bressolle, F., Pinguet, F., 1998. In vitro modulation of doxorubicin and docetaxel antitumoral activity by methyl-beta-cyclodextrin. European Journal of Cancer 34, 168. 31

ACCEPTED MANUSCRIPT Groves, M.J., Wineberg, M., Brain, A.P.R., 1985. THE PRESENCE OF LIPOSOMAL MATERIAL IN PHOSPHATIDE STABILIZED EMULSIONS. Journal of Dispersion Science and Technology, 237-243. Hörmann, K., Zimmer, A., 2015. Drug delivery and drug targeting with parenteral

PT

lipid nanoemulsions - A review. Journal of Controlled Release 223, 85-98.

RI

Horwitz, S.B., 1992. Mechanism of action of taxol. Trends in Pharmacological

SC

Sciences 13, 134.

Hwang, H.Y., Kim, I.S., Kwon, I.C., Kim, Y.H., 2008. Tumor targetability and

NU

antitumor effect of docetaxel-loaded hydrophobically modified glycol chitosan

MA

nanoparticles. Journal of Controlled Release 128, 23-31. Immordino, M.L., Brusa, P., Arpicco, S., Stella, B., Dosio, F., Cattel, L., 2003.

D

Preparation, characterization, cytotoxicity and pharmacokinetics of liposomes

PT E

containing docetaxel. Journal of Controlled Release Official Journal of the Controlled Release Society 91, 417.

CE

Jumaa, M., Müller, B.W., 1998. The effect of oil components and homogenization

AC

conditions on the physicochemical properties and stability of parenteral fat emulsions. International Journal of Pharmaceutics 163, 81-89. Kawakami, S., Yamashita, F., Hashida, M., 2000. Disposition characteristics of emulsions and incorporated drugs after systemic or local injection. Advanced Drug Delivery Reviews 45, 77-88. Klang, V., Valenta, C., 2011. Lecithin-based nanoemulsions. Journal of Drug Delivery Science & Technology 21, 55-76.

32

ACCEPTED MANUSCRIPT Klein, R.A., 1970. The detection of oxidation in liposome preparations. Biochimica et Biophysica Acta (BBA) - Lipids and Lipid Metabolism 210, 486-489. Lavelle, F., Bissery, M.C., Combeau, C., Riou, J.F., Vrignaud, P., André, S., 1995. Preclinical evaluation of docetaxel (Taxotere), Seminars in oncology, pp. 3-16.

PT

Li, X., Wang, D., Zhang, J., Pan, W., 2009. Preparation and pharmacokinetics of

RI

docetaxel based on nanostructured lipid carriers. Journal of Pharmacy &

SC

Pharmacology 61, 1485.

Liu, B., Yang, M., Li, R., Ding, Y., Qian, X., Yu, L., Jiang, X., 2008. The antitumor

NU

effect of novel docetaxel-loaded thermosensitive micelles. European Journal of

MA

Pharmaceutics & Biopharmaceutics 69, 527-534.

Lu, Y., Wang, Y., Tang, X., 2008. Formulation and thermal sterile stability of a less

D

painful intravenous clarithromycin emulsion containing vitamin E. International

PT E

Journal of Pharmaceutics 346, 47.

Ma, J., Teng, H., Wang, J., Yu, Z., Ren, T., Xing, T., Cai, C., 2014. A highly stable

CE

norcantharidin loaded lipid microspheres: Preparation, biodistribution and targeting

AC

evaluation. International Journal of Pharmaceutics 473, 475. Machado, A., Maranduba, A., Guimarães, E., Machado, L., Santiago Junior, M., Silva, M., 2008. PHARMACEUTICAL COMPOSITIONS CONTAINING DOCETAXEL AND A DEGRADATION INHIBITOR AND A PROCESS FOR OBTAINING THE SAME. US. Manfredi, R., Chiodo, F., 1998. Case-control study of amphotericin B in a triglyceride fat emulsion versus conventional amphotericin B in patients with AIDS.

33

ACCEPTED MANUSCRIPT Pharmacotherapy 18, 1087-1092. Manjappa, A.S., Goel, P.N., Vekataraju, M.P., Rajesh, K.S., Makwana, K., Ukawala, M., Nikam, Y., Gude, R.P., Murthy, R.S., 2013. Is an alternative drug delivery system needed for docetaxel? The role of controlling epimerization in formulations and

PT

beyond. Pharmaceutical Research 30, 2675-2693.

RI

Ringel, I., Horwitz, S.B., 1987. Taxol is converted to 7-epitaxol, a biologically active

SC

isomer, in cell culture medium. Journal of Pharmacology & Experimental Therapeutics 242, 692.

NU

Ringel, I., Horwitz, S.B., 1991. Studies With RP 56976 (Taxotere): A Semisynthetic

MA

Analogue of Taxol. Journal of the National Cancer Institute 83, 288. Salmela, L., Washington, C., 2014. A continuous flow method for estimation of drug

D

release rates from emulsion formulations. International Journal of Pharmaceutics 472,

PT E

276-281.

Shi, S., Chen, H., Cui, Y., Tang, X., 2009. Formulation, stability and degradation

CE

kinetics of intravenous cinnarizine lipid emulsion. International Journal of

AC

Pharmaceutics 373, 147–155. Shou, M., Martinet, M., Korzekwa, K.R., Krausz, K.W., Gonzalez, F.J., Gelboin, H.V., 1998. Role of human cytochrome P450 3A4 and 3A5 in the metabolism of taxotere and its derivatives: enzyme specificity, interindividual distribution and metabolic contribution in human liver. Pharmacogenetics 8, 391. Smyrniotis, V.E., Kostopanagiotou, G.G., Arkadopoulos, N.F., Theodoraki, K.A., Kotsis, T.E., Lambrou, A.T., Vassiliou, J.G., 2001. Long-chain versus medium-chain

34

ACCEPTED MANUSCRIPT lipids in acute pancreatitis complicated by acute respiratory distress syndrome: effects on pulmonary hemodynamics and gas exchange. Clinical Nutrition 20, 139-143. Sparreboom, A., Van, T.O., Scherrenburg, E.J., Boesen, J.J., Huizing, M.T., Nooijen, W.J., Versluis, C., Beijnen, J.H., 1996. Isolation, purification and biological activity of

PT

major docetaxel metabolites from human feces. Drug Metabolism & Disposition the

RI

Biological Fate of Chemicals 24, 655.

SC

ten Tije, A.J., Verweij, J., Loos, W.J., Sparreboom, A., 2003. Pharmacological effects of formulation vehicles : implications for cancer chemotherapy. Clinical

NU

Pharmacokinetics 42, 665-685.

MA

Teng, F., Yang, H., Li, G., Lin, X., Zhang, Y., Tang, X., 2014. Parenteral formulation of larotaxel lipid microsphere tackling poor solubility and chemical instability.

D

International Journal of Pharmaceutics 460, 212-219.

PT E

Vasu, D.R., Moses, B.J.K., Sai, R.P., Ramachandra, P., Sekhar, N.M., Mohan Reddy, D.N., Srinivasa, R.N., 2006. Isolation and characterization of impurities in docetaxel.

CE

Journal of Pharmaceutical & Biomedical Analysis 40, 614-622.

AC

Wang, J.J., Sung, K.C., Hu, O.Y., Yeh, C.H., Fang, J.Y., 2006. Submicron lipid emulsion as a drug delivery system for nalbuphine and its prodrugs. Journal of Controlled Release 115, 140. Xu, Z., Chen, L., Gu, W., Gao, Y., Lin, L., Zhang, Z., Xi, Y., Li, Y., 2009. The performance of docetaxel-loaded solid lipid nanoparticles targeted to hepatocellular carcinoma. Biomaterials 30, 226. Yamano, Y., Seike, N., 1983. Stability of Emulsion Prepared with Soybean

35

ACCEPTED MANUSCRIPT Phospholipids and Transfer of Phospholipid to Separated Phases. NIPPON SHOKUHIN KOGYO GAKKAISHI 30, 618-623. Yang, Z., Ren, T., Lu, D., Guo, H., Li, W., Huang, C., He, H., Liu, D., Tang, X., 2016. Evaluating the safety of phytosterols removed perilla seed oil-based lipid emulsion.

PT

Expert Opinion on Drug Delivery 13, 1345-1356.

RI

Yanjie, S., Chungang, Z., Qing, Y., Yueqi, W., Bin, T., Xing, T., Yanjiao, W., 2014.

SC

Improving cabazitaxel chemical stability in parenteral lipid emulsions using cholesterol. European Journal of Pharmaceutical Sciences Official Journal of the

NU

European Federation for Pharmaceutical Sciences 52, 1-11.

MA

Zhao, M., Su, M.X., Luo, Y., He, H., Cai, C., Tang, X., 2010. Evaluation of docetaxel-loaded intravenous lipid emulsion: pharmacokinetics, tissue distribution,

AC

CE

PT E

D

antitumor activity, safety and toxicity. Pharmaceutical Research 27, 1687-1702.

36

ACCEPTED MANUSCRIPT

SC

RI

PT

(A)

PT E

D

MA

NU

(B)

AC

CE

Fig. 1. The structures of (A) DTX and (B) EDTX.

37

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

Fig.2. The physical appearance and optical micrograph of DTX-LM

38

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Fig.3. Cumulative release curves of DTX in DI and LM(n=3)

39

ACCEPTED MANUSCRIPT

pH 4.05 pH 5.28 pH 6.04 pH 7.04 pH 8.05 pH 9.01

(A) 3.0

2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8

PT

lnC (ug/ml)

2.8 2.6 2.4

2

4

6

Time (d) 3.0 2.8 2.6

MA

2.4 2.2 2.0

4.06 5.27 6.02 6.98 8.02 9.02

D

1.6 1.4 1.2

PT E

lnC (ug/ml)

1.8

10

pH pH pH pH pH pH

NU

(B)

8

SC

0

RI

0.6 0.4 0.2 0.0

1.0 0.8 0.6

0.2 0.0

AC

0

CE

0.4

2

4

6

8

10

Time (d)

Fig. 4. DTX degradation curves at 60℃ over a pH range of 4.0-9.0: (A) DTX in DI; (B) DTX in LM(n=3).

40

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Fig. 5. The pH-rate profiles for the degradation of DTX in DI and LM at 60℃.

41

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig.6. EDTX (%) in DI and LM at 60℃ over a pH range of 4.0-9.0(n=3).

42

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig.7. DTX degradation curves at different temperatures: (A) DTX in DI; (B) DTX in LM(n=3).

43

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

Fig. 8. Arrhenius plots of DTX.

44

ACCEPTED MANUSCRIPT

SC

RI

PT

A:

PT E

D

MA

NU

B:

CE

Fig.9. (A)Schematic diagram of DTX epimerization mechanism in vitro;(B)Schematic

AC

diagram of lower EDTX conversion of DTX-LM in vivo compared with DI. DTX,

:free DTX,

45

:EDTX.

:encapsulated

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig.10. Apparent epimerization coefficient curves at different time of DTX-LM and DI (37℃

AC

CE

PT E

D

MA

NU

plasma incubation,n=3)

46

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 11. Mean plasma concentration-time curves of DTX and EDTX after intravenous

AC

CE

administration of DTX-LM and DI at a dose of 10 mg/kg(n=8).

47

ACCEPTED MANUSCRIPT Table 1 Effect of different acids on the formation of EDTX (n=3). pH value of the final emulsion

Content of EDTX (%)

HCl

5.11±0.06

6.09±0.14

tartaric acid

4.52±0.23

1.65±0.10

caproic acid

5.06±0.07

2.07±0.10

NaHSO3

5.33±0.19

PT

Type of acid

AC

CE

PT E

D

MA

NU

SC

RI

0.16±0.07

48

ACCEPTED MANUSCRIPT Table 2 Effects of different amounts of NaHSO3 on the formation of EDTX(n=3). Amount of

Physical

Content of EDTX pH value

NaHSO3 (%, w/v)

(%) 3.95

0.61±0.11

4.50

1.03±0.10

Good

PT

0

appearance

4.00 Good

Good

D

0.01

0.47±0.05

4.56

0.68±0.03

3.99

0.32±0.07

4.51

0.41±0.10

4.01

0.16±0.08

4.51

0.20±0.04

Good

AC

CE

PT E

0.02

0.82±0.09

4.02

NU

Good

MA

0.005

SC

4.49

0.56±0.05

RI

0.0025

49

ACCEPTED MANUSCRIPT Table 3 Effects of different pH values of the final emulsions on the characteristics of DTX-LM (n=3). pH value pH value

Content of DTX(%)

Content of EDTX(%)

0.16±0.08

AS

4.0

4.01

3.46±0.13

-

4.5

4.51

3.60±0.07

-

5.0

4.93

3.66±0.10

93.6±2.82

5.3

5.29

3.72±0.13

100.0±1.85

6.0

5.96

3.88±0.21

6.5

6.50

4.10±0.13

PT

BS

NU

SC

RI

0.20±0.04

0.49±0.12

96.9±1.93

0.89±0.07

MA

D PT E CE AC

0.44±0.10

84.6±5.51

BS, before sterilization; AS, after sterilization.

50

0.41±0.14

ACCEPTED MANUSCRIPT Table 4 Stability of DTX-LM over 6 months storage at 25±2℃ and 24 months storage at 6±2℃(n=3). Time pH

EDTX content

(%)

(%)

99.73±0.26

0.22±0.06

EE (%)

(months)

3.88±0.08

-

98.23±1.74

0.23±0.04

2

3.86±0.15

-

3

3.84±0.17

-

6

RI

0.24±0.05

97.50±0.48

0.40±0.05

3.82±0.25 99.07±0.16

96.70±0.49

0.68±0.12

0

NU

PT

1

3.89±0.15 99.37±0.75

99.73±0.26

0.22±0.06

3

3.87±0.14

-

98.90±0.23

0.26±0.08

6

3.84±0.10

-

98.53±0.14

0.28±0.09

9

3.81±0.09

-

98.37±0.07

0.32±0.08

12

3.78±0.15

-

98.03±0.18

0.35±0.09

18

3.75±0.13

-

97.33±0.44

0.47±0.11

24

3.75±0.08 99.03±0.18

96.60±0.41

0.68±0.08

AC

CE

PT E

6±2C

3.89±0.15 99.37±0.75

MA

25±2C

0

D

temperature

DTX content

51

97.63±1.04

SC

Storage

ACCEPTED MANUSCRIPT Table 5 Plasma concentrations of DTX and EDTX as well as convert ratios of EDTX for each individual rat at the end of i.v. administration of DTX-LM and DI at a dose of 10mg/kg (n=8). DTX-LM 78.55

70.25

69.56

10.35

37.94

107.81

50.18

35.31

63.60

100.57

66.69

59.43

45.70

59.75

1.81

3.43

1.22

1.22

1.64

1.95

0.68

75.44

0.15

(ug/ml)

1.26

2.27

1.34

2.15

1.79

1.62

0.54

0.98

1.69

4.36

1.74

1.75

1.40

4.33

1.81

1.35

3.56

3.57

1.78

2.14

2.68

2.73

1.18

1.64

PT E

MA

EDTX

D

NU

SC

DTX (ug/ml)

PT

107.53

RI

DI

AC

CE

Convert ratios（%）

Mean（%）

3.04±1.11

1.67±0.29

Background values（%）

-

1.06

52

ACCEPTED MANUSCRIPT Table 6 Pharmacokinetics parameters of DTX-LM and DI fitted by non-compartment model (n=8). DTX-LM

DI

Dosage

(mg/kg)

10

10

AUC(0-t)a

(μg /L/ min)

555880.719 ±85042.775

208537.518±86062.167

AUC(0-∞)a

(μg /L/ min)

578744.063±96699.434

AUMC(0-t)a

(µg/L)·min2

6408729.84±1699471.5

AUMC(0-∞)a

(µg/L)·min2

8340957.65±3339314.7

1768025.87±1396282.6

T1/2

(min)

13.409±3.308

13.859±7.983

CLa

(L/min/kg)

0.018±0.0030

0.056±0.027

Vssa

(L/kg)

214681.105±89413.459

RI

SC

NU

MA

0.336±0.074

CE

PT E

D

p<0.05

AC

a

PT

Units

53

1125198.57±649218.35

0.971±0.51

ACCEPTED MANUSCRIPT Table 7 Comparison of the quantitative relationship between EDTX and DTX in different intervals after intravenous injection of DI and DTX-LM in rats (n = 8) (%)

0-5min 5-15min 0-15min 15-30min 0-30min 30-60min 0-60min 3.13

2.30

3.00

2.49

2.96

3.18

2.97

DTX-LM

0.555

0.025

0.357

0.161

0.315

0.454

0.336

AC

CE

PT E

D

MA

NU

SC

RI

PT

DI

54

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

Graphical abstract

55