Physicochemical evaluation and non-isothermal kinetic study of the drug–excipient interaction between doxepin and lactose

    Physicochemical evaluation and non-isothermal kinetic study of the drugexcipient interaction between doxepin and lactose Faranak Ghaderi, Mahboob Nemati, Mohammad R. Siahi-Shadbad, Hadi Valizadeh, Farnaz Monajjemzadeh PII: DOI: Reference:

S0032-5910(15)30053-X doi: 10.1016/j.powtec.2015.09.007 PTEC 11226

To appear in:

Powder Technology

Received date: Revised date: Accepted date:

25 April 2015 2 August 2015 2 September 2015

Please cite this article as: Faranak Ghaderi, Mahboob Nemati, Mohammad R. SiahiShadbad, Hadi Valizadeh, Farnaz Monajjemzadeh, Physicochemical evaluation and nonisothermal kinetic study of the drug-excipient interaction between doxepin and lactose, Powder Technology (2015), doi: 10.1016/j.powtec.2015.09.007

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 Physicochemical evaluation and non-isothermal kinetic study of the drugexcipient interaction between doxepin and lactose

SC R

IP

T

Faranak Ghaderi a,b, Mahboob Nemati a, Mohammad R. Siahi-Shadbada,c , Hadi Valizadeh d, Farnaz Monajjemzadeh a,c*, a

NU

Department of pharmaceutical and Food Control, Tabriz University of Medical Sciences, Tabriz, Iran b Student Research Committee, Tabriz University of Medical Sciences, Tabriz, Iran c Drug Applied Research Center, Tabriz University of Medical Sciences, Tabriz,Iran d Department of Pharmaceutics , Tabriz University of Medical Sciences, Tabriz, Iran

MA

*Corresponding author: Farnaz Monajjemzadeh, Department of Pharmaceutical and Food Control, Tabriz University of Medical Sciences, Tabriz, Zip code: 5166414766, Iran, Tel: +9841133392606; Fax: +9841133344798; E-mail: [email protected], [email protected]

D

Abstract

TE

In this study, the incompatibility of doxepin in solid physical mixtures with lactose (monohydrate and

CE P

anhydrous) was investigated. The compatibility testing was made using various physicochemical techniques, such as differential scanning calorimetry (DSC), Fourier-transform infrared (FTIR) spectroscopy, and Mass spectrometry. Non-Isothermally stressed physical mixtures were used to

AC

analyze the solid–state kinetic parameters. Data was fitted to different thermal models, such as Friedman, Flynn–Wall–Ozawa (FWO) and Kissinger–Akahira–Sunose (KAS) for different drugexcipient mixtures separately. Overall, the incompatibility of doxepin as a tertiary amine, with lactose as a reducing carbohydrate was successfully evaluated. DSC based kinetic analysis provided a simple and fast comparative data in different drug-excipient mixtures. It can be recommended to exclude lactose from doxepin’s solid dosage formulations, and also to use the described procedure in the kinetic evaluation of drug-excipient incompatibility studies. Keywords: Doxepin; Lactose; Kinetic; DSC; Mass spectrometry; incompatibility

ACCEPTED MANUSCRIPT *Corresponding author: Farnaz Monajjemzadeh Tel: +9841133392606; Fax: +9841133344798; E-mail: [email protected], [email protected]

IP

T

1. Introduction An assessment of the incompatibilities between the drug substance and the excipients is important

SC R

during preformulation studies, and the quality control of pharmaceutical dosage forms. Drug and excipients may change their chemical and physical stability of pharmaceutical formulations, which

NU

may consequently lead to altered bioavailability, as well as the efficacy. It also increases the safety of the medicinal product [1, 2]. Common drug-excipient interactions which have been reported till now

MA

are acid-base and Maillard interactions. Maillard type of reaction is defined as a probable incompatibility reactions in amine-containing drugs, such as doxepin and reducing excipients, which leads to formation of a Schiff`s base as an unstable intermediate reaction product. It then undergoes

D

Amadori rearrangement and generates a ketoamine compound [1, 3]. This type of incompatibility

TE

leads to a color abrasion, and results in new chemical entities with unknown efficacy and suspected

CE P

safety. From the safety issues, it is previously reported that Maillard reaction products, which are generated from aldose and ketose-lysine reactants have a cytotoxic effect on both rat and human cells[4].

AC

Various analytical methods have been employed to prove this type of interaction, such as thermal analysis, UV and infra-red spectroscopy, mass spectrometry and NMR studies. Doxepin, a dibenzoxepin-derivative tricyclic antidepressant is used to treat depression disorders, anxiety , insomnia, and as a second line treatment for chronic idiopathic urticaria [5, 6]. This drug was first approved in 1969. Its mechanism of action is inhibition of serotonin and norepinephrine reuptake. Thus, it acts as an antagonist at various serotonergic, adrenergic, muscarinic, dopaminergic and histaminergic receptors [7]. Its chemical formula is C19H21NO, and its structure is shown in Fig.1.

Fig. 1. Structure of doxepin

ACCEPTED MANUSCRIPT Although it has been previously shown that different amine containing drugs (primary and secondary amines), such as baclofen and acyclovir, can interact with reducing carbohydrates via the Maillard reaction, there is only one report about a type 3 amine containing drug (promethazine), with no

IP

T

detailed analytical evaluation which makes this hypothesis as a simple assumption [8]. Although Maillard reaction is a well-known chemical interaction, there is still a great tendency to

SC R

utilize lactose for a wide range of solid dosage formulations in pharmaceutical industries worldwide. Literature has shown that there is a lack of information about the involvement of a tertiary amine

NU

drug, such as doxepin in Maillard type interaction with lactose.

Thus, in this research, the possibility of the Maillard reaction in doxepin, with only one tertiary amine

MA

moiety will be evaluated using simple and sophisticated analytical techniques. Finally, non-isothermal DSC methods will be utilized to assess the kinetic information in different drug-excipient mixtures, to

2. Materials and methods

CE P

2.1. Materials

TE

D

be able to predict the remaining drugs in various conditions.

Doxepin (3-(6H-benzo[c][1]benzoxepin-11-ylidene)-N,N-dimethylpropan-1-amine) was obtained from Dipharma Francis Pharmaceutical Co. (Baranzate, Italy). Monohydrate and anhydrous lactose

AC

were provided from DMV Chemical Co. (Veghal, Netherlands). Acetonitrile and formic acid were purchased from Merck (Darmstadt, Germany). All other chemicals were of HPLC grade and were obtained from Labscan Analytical Science (Dublin, Ireland). Generic preparations of doxepin named Brand 1–3 were acquired in local pharmacies in Iran.

2.2. Methods 2.2.1. DSC (Differential Scanning Calorimetry) A DSC-60, Shimadzu differential scanning calorimeter (Kyoto, Japan), with TA-60 software (version 1.51) was used for thermal analysis of drug and excipients alone, or in binary mixtures. Binary mixtures (10 g) were prepared from equal masses of doxepinand each excipient which were weighed

ACCEPTED MANUSCRIPT individually into amber glass flasks and uniform blending was ensured by tumbling method. Then, 5 mg of each sample was weighed and compressed in the DSC aluminum pan, and pressed using a cap. Consequently, it was scanned in the temperature range of 25–300 °C, with different heating rates (2.5,

SC R

2.2.2. FTIR (Fourier-transform infrared) spectroscopy

IP

T

5, 7.5, 10 and 15 °C/min).

Doxepin and excipients were blended in 1:1 mass ratios. They were mixed with 20% ( v/w) water and

NU

were stored in closed vials at 70 °C for 24 hours [9].

FTIR spectra were obtained immediately after mixing and also after incubation at elevated

MA

temperatures at predetermined time intervals, using potassium bromide disc preparation method (Bomem, MB-100 series, Quebec, Canada). The spectrum was an average of ten sequential scans on

D

the same sample, and the resolution was kept constant at 4 cm-1. FTIR data was processed by

CE P

2.2.3. Mass spectrometry

TE

GRAMS/32 version 3.04 (Galactic Industries Corporation, Salem, NH).

Mass analysis was performed on the Waters 2695 (Milford, Massachusetts, USA) Quadrupole Mass

AC

system, at electron-spray ionization mode, positive ionization, capillary voltage 3.5 V, cone voltage 30 V, extractor voltage 3 V, RF lens voltage 0.50V, source temperature (80 ˚C) desolvation temperature (200 ˚C), desolvation gas flow( 310 L / h) and cone gas flow (60L / h).

2.2.4. TLC TLC method was as described in our previous study [10]. A mixture of ethyl acetate and methanol (1:3 v/v) containing 0.25% v/v glacial acetic acid was used as the mobile phase. Lactose with a concentration of (1mg/mL) in diluent solution (methanol: water (2:3 v/v)), was spotted as the reference standard. Twenty units of each brand tablets (brand2 and 3) or capsules (brand 1) were weighed and the mean weight was calculated. Assuming that the entire excipient content of the average weight is lactose, an equivalent of 25 mg lactose in powdered tablets and capsules content

ACCEPTED MANUSCRIPT was transferred to a 25-mL volumetric flask and was diluted to 1 mg/mL. Standard and test solutions (2µL) were spotted on a thin-layer chromatographic plate individually (2020, Silica gel-60 F254, 0.25 mm thickness prepared from Merck (Darmstadt, Germany).

IP

T

The spots were dried and placed in a separation chamber, which was previously saturated with the solvent. The plate was removed from the chamber before the solvent front reached the top of the

SC R

stationary phase. It was dried with a stream of hot air, sprayed uniformly with staining solution containing 0.5g thymol in 95mL alcohol and 5mL sulfuric acid. Later, the plate was heated at 130 ºC

NU

for 5 min. As shown in Fig. 2, the presence of lactose was approved when the main spot resulted from these brands were similar to the standard solution in appearance and Rf (Retention Factor) values.

MA

Lactose was present in one domestic brand, and also in a foreign innovator brand.

2.3. Kinetic study:

TE

D

Fig. 2. TLC results of (S) lactose, (A) Brand-1, (B) Brand-2, and (C) Brand-3.

CE P

Comparative thermal stability of doxepin and lactose (monohydrate and anhydrous) mixtures was determined using the kinetic parameters in non−isothermal condition. Calculations were made on the resultant DSC curves, using differential models, such as the Friedman method [11, 12] and also

[13].

AC

integral modes, such as Kissinger–Akahira–Sunrose (KAS) and Flynn–Wall–Ozawa (FWO) methods

3. Results and discussion: 3.1. Analytical methods: 3.1.1. DSC DSC (Differential Scanning Calorimetry) can successfully evaluate the drug-excipient compatibility, and consequently provide important information, such as drug purity and stability. It can also provide information about the physicochemical characteristics, such as existence of different polymorphic forms and stability [14-18].

ACCEPTED MANUSCRIPT Selected DSC curves of drug, excipient and drug-excipient mixtures are shown in Fig.3 and 4. Thermal behavior of pure drug and pure excipients, as well as the binary mixture, were analysed in the DSC curves.

IP

T

According to Fig.3A, doxepin presented its melting point at 193.8 °C, and revealed a heat of fusion equal to 38.1 J/g (β=10).

SC R

The endothermic peak of pure anhydrous lactose appeared at 239.1°C, while lactose monohydrate showed two endothermic peaks at 152.1 and 218.3 °C, respectively. The first transition in lactose

NU

monohydrate is related to water loss. The other one is related to lactose melting (β=10). The endothermic peak of doxepin melting disappeared in the doxepin–anhydrous lactose and

MA

doxepin–monohydrate lactose binary mixture, which can only indicate the drug-excipient incompatibility (Fig.3B and 4A). A new endothermic peak also appeared at 167.9 °C in doxepin–

D

anhydrous lactose mixture, which may be related to an interaction between the binary mixtures

TE

components.

According to Fig. 3C and 4B, when heating rates increased, DSC curves were shifted to higher

CE P

temperatures.

It has been previously concluded that the heating rate variation has considerable effects on

AC

the temperature range, and the shape of curves for the process of decomposition, and an interaction between drug and excipients. Generally by increasing the heating rate, a right hand shift happens in the Heat flow versus Temperature curve while the left hand shift occurs in the Heat flow versus Time curve. This means that by increasing the heating rate the peak area of the melting endotherm increases and the melting happens in a shorter time but at higher temperatures [19, 20].

Fig. 3.Selected DSC curves of (A) doxepin at different heating rates (β=2.5, β=5, β=7.5, β=10, β=15). (B) doxepin, lactose anhydrous and doxepin-lactose anhydrous 1:1 W/W binary mixture (β=10). (C) doxepin-lactose anhydrous physical mixture with 1:1 mass ratio at different heating rates (β=2.5, β=5, β=7.5, β=10, β=15).

ACCEPTED MANUSCRIPT

SC R

IP

T

Fig. 4. Selected DSC scans of (A) doxepin, lactose monohydrate and doxepin-lactose monohydrate 1:1 W/W binary mixtures (β=10). (B) doxepin-lactose monohydrates 1:1 W/W binary mixtures at different heating rates (β=2.5, β=5, β=7.5, β=10, β=15).

3.1.2. FTIR

Fourier Transform Infrared (FT-IR) spectroscopy is another simple, fast and precise technique used

NU

for the assessment of drug- excipient incompatibility. In such cases, the use of FT-IR is based on the integral analysis of the spectra and determination of any changes in the position or the intensity of the

MA

peaks, as well as appearance or disappearance of the absorption band(s). Appearance of a new absorption band is an incontrovertible subject, which can refer to drug-excipient interaction and

D

provide valuable data in drug-excipient interaction mechanisms [21, 22].

TE

IR spectra of doxepin, lactose, doxepin-lactose (monohydrates and anhydrous) blends immediately

CE P

after mixing, and 24 hours after incubation in an oven (t=70 °C). This is shown in Fig. 5 and 6.

AC

Fig. 5. FTIR spectra of (a) doxepin, (b) lactose anhydrous (c) doxepin-lactose anhydrous1:1 W/W binary mixture immediately after mixing, and (d) binary mixture with 20% added water after 24 hours incubation at70 °C Fig. 6. FTIR spectra of (a) doxepin, (b) lactose monohydrate (c) doxepin–lactose monohydrate immediately after mixing, (d) binary mixture with 20% added water after 24 hours incubation at 70

Doxepin IR’s main signals appeared at ~ 2953 and 3019 cm-1(weak signal Ar ─ H bonding), 2953cm-1(─CH3), 1108, 1158 and 1253 cm-1(C─ N stretch), 1601cm-1 (aromatic systems) and principal peaks at wave numbers 750, 1006, 1198, 1290, 1219 cm−1[23]. Lactose anhydrous IR main signals: ~3459, 3292and 2811 cm-1 (OH) 2879cm-1 (CH2, CH3). Lactose monohydrate IR main peaks including: 3530, 3382 cm-1(OH), 2931 and 2897cm-1 (CH2, CH3).

Lactose anhydrous– doxepin mixture and lactose monohydrate– doxepin

ACCEPTED MANUSCRIPT mixture’s main signals were corresponding to the sum of each component’s peaks. In lactose anhydrous and monohydrate mixture with doxepin, new peaks are generated in 1649 and

T

1648 cm-1, respectively, which can be related to the generation of a C=N covalent band.

IP

The probable interaction between doxepin and lactose is supposed to be a Maillard type

SC R

reaction, which would lead to imine band generation in the FTIR spectrum. The C=N stretching band appears at 1640-1690 cm-1 in the infrared spectra of imine-containing compounds, including the Schiff’s base, which would form during the Maillard reaction [24,

NU

25]. Visual changes in the heated binary mixture are presented in Fig.7.

TE

3.1.3. Mass spectrometry

D

MA

Fig. 7. Visual changes in A) doxepin and lactose monohydrate binary mixture immediately after mixing B) doxepin and lactose anhydrous binary mixture immediately after mixing C) doxepin and lactose monohydrate binary mixture with 20% added water and stored at 70ºC after 30 days D) doxepin and lactose anhydrous binary mixture with 20% added water and stored at 70 ºC after 30 days.

CE P

Physical mixture of doxepin and lactose anhydrous (1:1), with 20% added water was prepared and stored at 90 °C. After 30 days, this sample dissolved in a mass compatible diluent (acetonitrile: formic acid 0.1% (40:60)), and was injected into the mass system. Mass spectra are presented in Fig.8.

AC

Fig. 8. Positive ion mode electrospray mass spectrum of (A) doxepin, (B) doxepin and lactose anhydrous (1:1) with 20% added water and stored at 90 °C for 30 days.

The full-scan positive ion electrospray product’s ion mass spectra showed that the molecular ion of doxepin was the protonated molecule, [M+H] + of m/z 280 (Fig.8A).

Fig. 9. Proposition for the complex of Maillard reaction condensation products.

ACCEPTED MANUSCRIPT The same molecular and daughter ions (m/z=235) have been previously reported for doxepin LC– MS/MS analysis [26]. The suggested structures for Maillard type interaction products have been presented in Fig. 9. The m/z value at 593.6 is related to [M+H] + of C29H39NO12, which is labeled as

IP

T

the condensation product A in Fig. 9.

Liquid chromatography/mass spectrometry (LC/MS) technique presents exclusive ability for

SC R

pharmaceutical analysis. LC/MS methods are enforceable to a wide range of pharmaceutical ingredients. They prominently indicate analytical figures of advantages, such as sensitivity,

NU

selectivity, speed of analysis and cost effectiveness [27]. In a research, Harmon et al. have reported the lactose–hydrochlorothiazide condensation product (m/z = 622 and 620) by MS [28]. Also,

MA

lactose–metoclopramide and lactose–fluoxetine condensation products have been detected by Qiu et al. and Wirth et al, respectively [29, 30]. We have previously studied the compatibility of acyclovir,

D

baclofen and gabapentin with lactose in physical mixtures, and commercial tablets using mass spectrometry [10, 31, 32]. More recently, Szalka et al. have confirmed the structures of intermediates

3.2. Kinetic study

CE P

TE

of various reducing carbohydrates –bisoprolol fumarate, in Maillard reactions by the LC/MS [33].

Determination of kinetic parameters by non-isothermal methods presents benefits over conventional

AC

isothermal studies. Specifically, kinetic analysis in non–isothermal conditions has been proposed as a rapid method for evaluating thermal behavior of pharmaceutical products. In kinetic studies, multiple scan method at different heating rates is an alternative to conventional non-isothermal single scan method [14]. Multiple scan method using iso-conversional calculation procedures have recently been applied by many researchers in solid-state reaction monitoring. Friedman (FR), Kissinger–Akahira– Sunose (KAS) and Flynn–Wall–Ozawa (FWO) methods have recently been used to study the kinetic parameters in solid state chemical interactions [34, 35]. The isoconversional Friedman method is based on the equation 1: Eq. 1

ACCEPTED MANUSCRIPT in which, T is the temperature, α is the extent of conversion, β is heating rate (˚C/min), E is activation is the reaction model, A is the pre-exponential factor, and R is the gas constant.

IP

T

energy

SC R

In which α is the conversion fraction, ∆H(t) is the area under curve at temperature (t), and ∆H(Max) is the total peak area [36].

NU

If α is kept constant and various heating rates (β) were applied, the plot of

MA

linear.

vs. (1/T) is

The values of activation energy (E) were obtained from the slopes of the straight lines in Fig. 10 for

D

doxepin -lactose anhydrous and doxepin -lactose monohydrate, respectively (Table 1and 2).

TE

Fig. 10. Friedman’s plot for (A) doxepin and lactose anhydrous (B) doxepin and lactose monohydrates at different heating rates and various conversion degrees

CE P

Flynn–Wall–Ozawa’s isoconversional method is based on measurement of the temperature needs to reach to determined values of the conversion α, for experiments performed at different heating rates

AC

(β). The equation 2 corresponding to this method is: Eq.2

The plot of lnβ vs. (1/T) is shown in Fig.11 for anhydrous and monohydrate lactose. According to the diagram, the correlation is linear. Activation energy (E) was obtained from the slopes of the straight lines. They are listed in Table 1 and 2.

Fig. 11. The Flynn–Wall–Ozawa (FWO) diagrams for (A) doxepin and lactose anhydrous (B) doxepin and lactose monohydrate at different heating rates and various conversion degrees.

ACCEPTED MANUSCRIPT The Kissinger–Akahira–Sunose (KAS) method is based on measuring the temperature related to fixed values of conversion for experiments, at different heating rates (β). Equation 3 is corresponding

IP

Eq. 3

T

to this method:

were plotted vs.

Fig.12

SC R

This method is one of the best iso-conversional methods.The values of

refers to these plots for doxepin -lactose anhydrous, and doxepin-lactose monohydrate, respectively. Activation energy (E) was calculated from the slope of the straight line and was listed in table 1 and

NU

2.

MA

Fig. 12. The Kissinger–Akahira–Sunose diagrams for (A) doxepin and lactose anhydrous (B) doxepine and lactose monohydrates at different heating rates and various conversion degrees.

TE

D

Table 1. Activation energy values for doxepin and lactose anhydrous obtained by the Friedman, Flynn–Wall–Ozawa and Kissinger–Akahira–Sunose methods.

CE P

Table 2. Activation energy values for doxepin and lactose monohydrates obtained by the Friedman, Flynn–Wall–Ozawa and Kissinger–Akahira–Sunose methods.

AC

As shown in Table 1 and 2, the values obtained by the three methods are in good agreement. According to the results, the activation energy of doxepin and lactose monohydrate mixture is less than that of doxepin and lactose anhydrous mixture. This can be explained by the fact that water has a significant role in the interaction of reactants [1]. The kinetic reaction provides valuable data about the reaction characteristics, and is used to compare the different reaction conditions. The most common method to evaluate the reaction kinetics is based on HPLC (high performance liquid chromatography) technique. This method is sensitive and reliable. However, it is time consuming and expensive. For example, the analyst should wait until an acceptable degradation takes place, based on the reaction rate, even at elevated temperatures. The method is set up. Several injections and large sample sizes are the other main drawbacks of conventional HPLC based kinetic evaluations. Overall, for an expert analyst, the time required to

ACCEPTED MANUSCRIPT evaluate the kinetic study for only one binary mixture is at least 30 days. Surprisingly, DSC kinetic evaluation can be performed in less than one week, and the sample size required for DSC method is too small. This is valuable, where the available amount of API is too low.

IP

T

Despite the aforementioned advantages, DSC data should be interpreted more carefully. According to the results, a simple conclusion can be made, in case of the reaction rate. However, there are some

SC R

other important parameters that a formulator analyst in pharmaceutical industries should take into consideration for final decisions. For example, in case of the different dosage forms, the behavior of

NU

Active pharmaceutical ingredient with anhydrous lactose can be treated differently. We have shown in our previous study that tablets prepared with anhydrous lactose will be harder than those prepared

MA

with monohydrate lactose, due to strong compaction of anhydrous lactose and better particle interlocking between baclofen and this excipient [31]. Compaction and interlocking of anhydrous

D

lactose favors the reaction in the tablet dosage forms, as compared to lactose monohydrate. This

4. Conclusion

CE P

TE

explains why in the binary mixtures, the case is completely reversed [31].

Although the Maillard reaction of amines with reducing agents is a well-known reaction but it has

AC

been shown in this study that in some formulations of doxepin the manufacturers have utilized lactose. The main reason for this may be related to the fact that in the pharmaceutical industries, lactose is an appropriate choice as a filler because it has excellent compressibility properties and is also cost benefit. Compatibility studies were performed using different analytical methods. DSC, FTIR and mass spectrometry provided useful compatibility information. DSC and FTIR results revealed the presence of incompatibility between doxepin and tested excipients. Activation energy of the proposed reaction was estimated using DSC data at different heating rates. According to the results, DSC provides fast and reliable kinetic data, in order to predict the remaining drugs in various conditions.

ACCEPTED MANUSCRIPT Mass results supported the drug- excipient condensation product. There is some safety information in the available literature [37, 38]. This says that Maillard’s reaction products may be genotoxic, carcinogenic, or cytotoxic. Therefore, it is recommended to avoid a combination of doxepin with

IP

T

lactose in pharmaceutical formulations. More detailed research is still needed to verify doxepinlactose Maillard reaction products safety in vivo.

SC R

5. Acknowledgments

This paper was extracted from a PhD thesis (No: 91) submitted to faculty of Pharmacy, Tabriz

NU

University of Medical Sciences and financially supported by the same University. 6. Conflict of interest

MA

The authors have no conflicts of interest. 7. References

AC

CE P

TE

D

[1] S.S. Bharate, S.B. Bharate, A.N. Bajaj, Incompatibilities of Pharmaceutical Excipients with Active Pharmaceutical Ingredients: A Comprehensive Review, Journal of Excipients and Food Chemicals, 1 (2010) 3-26. [2] F. Monajjemzadeh, F. Ebrahimi, P. Zakeri-Milani, H. Valizadeh, Effects of Formulation Variables and Storage Conditions on Light Protected Vitamin B12 Mixed Parenteral Formulations, Advanced pharmaceutical bulletin, 4 (2014) 329. [3] P.C. Buxton, R. Jähnke, S. Keady, Degradation of glucose in the presence of electrolytes during heat sterilisation, European journal of pharmaceutics and biopharmaceutics, 40 (1994) 172-175. [4] H. Jing, D. Kitts, Comparison of the antioxidative and cytotoxic properties of glucose-lysine and fructose-lysine Maillard reaction products, Food Research International, 33 (2000) 509-516. [5] G. Hajak, A. Rodenbeck, U. Voderholzer, D. Riemann, S. Cohrs, F. Hohagen, M. Berger, E. Ruther, Doxepin in the treatment of primary insomnia: a placebo-controlled, double-blind, polysomnographic study, Journal of Clinical Psychiatry, 62 (2001) 453-463. [6] F. Use, H. Use, T. Side-effects, W. Contra-indications, Martindale: the complete drug reference, (2007). [7] D. Markov, K. Doghramji, Doxepin for insomnia, Current Psychiatry, 9 (2010) 67. [8] L.V. Allen Jr, Meperidine Hydrochloride and Promethazine Hydrochloride Capsules, US Pharm, 35 (2010) 36-37. [9] A. Serajuddin, A.B. Thakur, R.N. Ghoshal, M.G. Fakes, S.A. Ranadive, K.R. Morris, S.A. Varia, Selection of solid dosage form composition through drug–excipient compatibility testing, Journal of pharmaceutical sciences, 88 (1999) 696-704. [10] F. Monajjemzadeh, D. Hassanzadeh, H. Valizadeh, M.R. Siahi-Shadbad, J.S. Mojarrad, T.A. Robertson, M.S. Roberts, Compatibility studies of acyclovir and lactose in physical mixtures and commercial tablets, European Journal of Pharmaceutics and Biopharmaceutics, 73 (2009) 404-413. [11] W.L. Chang, Decomposition behavior of polyurethanes via mathematical simulation, Journal of applied polymer science, 53 (1994) 1759-1769. [12] H.L. Friedman, Kinetics of thermal degradation of char‐forming plastics from thermogravimetry. Application to a phenolic plastic, Journal of Polymer Science Part C: Polymer Symposia, Wiley Online Library, 1964, pp. 183-195. [13] T. Ozawa, Kinetics of non-isothermal crystallization, Polymer, 12 (1971) 150-158.

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

[14] A. Fulias, T. Vlase, G. Vlase, Z. Szabadai, G. Rusu, G. Bandur, D. Tita, N. Doca, Thermoanalytical study of cefadroxil and its mixtures with different excipients, reactions, 4 (2010) 11. [15] N.R. Pani, L.K. Nath, S. Acharya, Compatibility studies of nateglinide with excipients in immediate release tablets, Acta Pharmaceutica, 61 (2011) 237-247. [16] Y. Huang, Y. Cheng, K. Alexander, D. Dollimore, The thermal analysis study of the drug captopril, Thermochimica acta, 367 (2001) 43-58. [17] L. Burnham, D. Dollimore, K.S. Alexander, Kinetic study of the drug acetazolamide using thermogravimetry, Thermochimica acta, 392 (2002) 127-133. [18] F. Monajjemzadeh, F. Ghaderi, Thermal Analysis Methods in Pharmaceutical Quality Control, J Mol Pharm Org Process Res, 3 (2015) e121. [19] B. Tiţa, A. Fuliaş, G. Bandur, G. Rusu, D. Tiţa, Thermal stability of ibuprofen. Kinetic study under non-isothermal conditions, Rev Roum Chim, 55 (2010) 553-558. [20] P.J. Haines, Principles of thermal analysis and calorimetry, Royal society of chemistry2002. [21] A. Fuliaş, I. Ledeţi, G. Vlase, C. Popoiu, A. Hegheş, M. Bilanin, T. Vlase, D. Gheorgheosu, M. Craina, S. Ardelean, Thermal behaviour of procaine and benzocaine Part II: compatibility study with some pharmaceutical excipients used in solid dosage forms, Chemistry Central Journal, 7 (2013) 140. [22] K. Liltorp, T.G. Larsen, B. Willumsen, R. Holm, Solid state compatibility studies with tablet excipients using non thermal methods, Journal of pharmaceutical and biomedical analysis, 55 (2011) 424-428. [23] A.C. Moffat, M.D. Osselton, B. Widdop, Clarke's analysis of drugs and poisons. [24] H. Nazir, C. Arici, K.C. Emregül, O. Atakol, A crystallographic and spectroscopic study on the imine-amine tautomerism of 2-hydroxyaldimine compounds, Zeitschrift für Kristallographie, 221 (2006) 699-704. [25] H. Namli, O. Turhan, Background defining during the imine formation reaction in FT-IR liquid cell, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 64 (2006) 93-100. [26] C. Coulter, M. Taruc, J. Tuyay, C. Moore, Antidepressant drugs in oral fluid using liquid chromatography-tandem mass spectrometry, Journal of analytical toxicology, 34 (2010) 64-72. [27] M.S. Lee, E.H. Kerns, LC/MS applications in drug development, Mass Spectrometry Reviews, 18 (1999) 187-279. [28] P.A. Harmon, W. Yin, W.E. Bowen, R. Tyrrell, R.A. Reed, Liquid chromatography–mass spectrometry and proton nuclear magnetic resonance characterization of trace level condensation products formed between lactose and the amine‐containing diuretic hydrochlorothiazide, Journal of pharmaceutical sciences, 89 (2000) 920-929. [29] Z. Qiu, J.G. Stowell, K.R. Morris, S.R. Byrn, R. Pinal, Kinetic study of the Maillard reaction between metoclopramide hydrochloride and lactose, International journal of pharmaceutics, 303 (2005) 20-30. [30] D.D. Wirth, S.W. Baertschi, R.A. Johnson, S.R. Maple, M.S. Miller, D.K. Hallenbeck, S.M. Gregg, Maillard reaction of lactose and fluoxetine hydrochloride, a secondary amine, Journal of pharmaceutical sciences, 87 (1998) 31-39. [31] F. Monajjemzadeh, D. Hassanzadeh, H. Valizadeh, M.R. Siahi-Shadbad, J.S. Mojarrad, T. Robertson, M.S. Roberts, Assessment of feasibility of Maillard reaction between baclofen and lactose by liquid chromatography and tandem mass spectrometry, application to pre formulation studies, AAPS PharmSciTech, 10 (2009) 649-659. [32] F. Monajjemzadeh, D. Hassanzadeh, H. Valizadeh, M.R. Siahi-Shadbad, J.S. Mojarrad, T.A. Robertson, M.S. Roberts, Detection of gabapentin-lactose Maillard reaction product (Schiff's Base): Application to solid dosage form preformulation. Part 1, pharmind: Die Pharmazeutische Industrie, 73 (2011) 174-177. [33] M. Szalka, J. Lubczak, D. Naróg, M. Laskowski, K. Kaczmarski, The Maillard reaction of bisoprolol fumarate with various reducing carbohydrates, European Journal of Pharmaceutical Sciences, 59 (2014) 1-11.

ACCEPTED MANUSCRIPT

NU

SC R

IP

T

[34] S. Vyazovkin, D. Dollimore, Linear and nonlinear procedures in isoconversional computations of the activation energy of nonisothermal reactions in solids, Journal of chemical information and computer sciences, 36 (1996) 42-45. [35] G. He, B. Riedl, A. Aït‐Kadi, Model‐free kinetics: Curing behavior of phenol formaldehyde resins by differential scanning calorimetry, Journal of applied polymer science, 87 (2003) 433-440. [36] L. Núñez-Regueira, C. Gracia-Fernández, S. Gómez-Barreiro, Use of rheology, dielectric analysis and differential scanning calorimetry for gel time determination of a thermoset, Polymer, 46 (2005) 5979-5985. [37] I.B.Z. Díaz, V.I. Chalova, C.A. O'Bryan, P.G. Crandall, S.C. Ricke, Effect of soluble maillard reaction products on cadA expression in Salmonella typhimurium, Journal of Environmental Science and Health Part B, 45 (2010) 162-166. [38] C. Delgado-Andrade, F.J. Morales, I. Seiquer, M. Pilar Navarro, Maillard reaction products profile and intake from Spanish typical dishes, Food Research International, 43 (2010) 1304-1311.

AC

CE P

TE

D

MA

Figures:

Fig. 1. Structure of doxepin

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Fig. 2. TLC results of (S) lactose, (A) Brand-1, (B) Brand-2 (Innovator), and (C) Brand-3

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 3.Selected DSC thermograms of (A) doxepin at different heating rates (β=2.5, β=5, β=7.5, β=10, β=15). (B) doxepin, lactose anhydrous and doxepin-lactose anhydrous 1:1 W/W binary mixture (β=10). (C) doxepin-lactose anhydrous physical mixture with 1:1 mass ratio at different heating rates (β=2.5, β=5, β=7.5, β=10, β=15).

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 4. Selected DSC scans of (A) doxepin, lactose monohydrate and doxepin-lactose monohydrate 1:1 W/W binary mixtures (β=10). (B) doxepin-lactose monohydrates 1:1 W/W binary mixtures at different heating rates (β=2.5, β=5, β=7.5, β=10, β=15).

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Fig. 5. FTIR spectra of (a) doxepin, (b) lactose anhydrous (c) doxepin-lactose anhydrous1:1 W/W binary mixture immediately after mixing, and (d) binary mixture with 20% added water after 24 hours incubation at70 °C

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 6. FTIR spectra of (a) doxepin, (b) lactose monohydrate (c) doxepin–lactose monohydrate immediately after mixing, (d) binary mixture with 20% added water after 24 hours incubation at 70 °C

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Fig. 7. Visual changes in A) doxepin and lactose monohydrate binary mixture immediately after mixing B) doxepin and lactose anhydrous binary mixture immediately after mixing C) doxepin and lactose monohydrate binary mixture with 20% added water and stored at 70ºC after 30 days D) doxepin and lactose anhydrous binary mixture with 20% added water and stored at 70 ºC after 30 days.

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 8. Positive ion mode electrospray mass spectrum of (A) doxepin, (B) doxepin and lactose anhydrous (1:1) with 20% added water and stored at 90 °C for 30 days.

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Fig. 9. Proposition for the complex of Maillard reaction condensation products.

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 10. Friedman’s plot for (A) doxepin and lactose anhydrous (B) doxepin and lactose monohydrates at different heating rates and various conversion degrees.

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 11. The Flynn–Wall–Ozawa (FWO) diagrams for (A) doxepin and lactose anhydrous (B) doxepin and lactose monohydrate at different heating rates and various conversion degrees.

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Fig. 12. The Kissinger–Akahira–Sunose diagrams for (A) doxepin and lactose anhydrous (B) doxepine and lactose monohydrates at different heating rates and various conversion degrees

ACCEPTED MANUSCRIPT

Tables:

0.4

0.5

0.6

0.7

0.8

0.9

value

385.01

348.85

324.04

298.59

286.79

269.55

246.69

217.65

202.52

286.63

±10.36

±3.50

±5.26

±7.53

±7.16

±6.53

±13.89

±7.19

±8.47

±2.98

396.03

359.10

335.35

304.64

295.10

271.92

250.58

227.19

201.20

293.46

±12.54

±3.96

±6.69

±11.11

±11.21

±9.69

±10.22

±8.91

±9.92

±2.60

394.49

350.44

325.63

293.88

288.27

263.49

238.96

218.01

205.94

286.57

±7.94

±3.22

±5.01

±7.30

±7.59

±8.77

±5.97

±7.00

±7.09

±1.68

MA

NU

0.3

D

KAS

0.2

TE

FWO

Main

0.1

CE P

FR

E, (kJ mol–1), for conversion degree, α

AC

Method

SC R

IP

T

Table 1. Values for the activation energy of doxepin and lactose anhydrous obtained by the Friedman, Flynn–Wall–Ozawa and Kissinger–Akahira–Sunose methods.

ACCEPTED MANUSCRIPT Table 2. Values of the activation energy of doxepin and lactose monohydrates obtained by the Friedman, Flynn–Wall–Ozawa and Kissinger–Akahira–Sunose methods.

value

190.77

T

175.50

181.35

±7.03

±11.44

±2.60

205.76

192.24

169.83

181.87

±4.93

±5.96

±4.95

±10.58

±3.00

208.48

203.42

192.24

176.76

183.54

±5.89

±4.95

±9.17

±1.72

0.4

0.5

0.6

0.7

141.40

152.50

168.77

190.69

205.84

203.53

203.12

±5.36

±9.68

±7.17

±6.58

±7.09

±2.23

±9.70

143.26

154.10

165.74

192.66

204.75

208.48

±4.73

±2.53

±11.45

±4.88

±8.36

145.01

152.49

170.63

192.66

209.65

±10.36

±4.81

±5.16

±4.88

±8.10

SC R

NU

MA

±4.92

0.8

IP

0.3

D

KAS

0.9

0.2

TE

FWO

Main

0.1

CE P

FR

E, (kJ mol–1), for conversion degree, α

AC

Method

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Graphical abstract

ACCEPTED MANUSCRIPT Highlights Dealing with a tertiary amine (Doxepin) in Maillard reaction



Introducing DSC kinetic evaluation for the observed incompatibility



Combining 3 different analytical methods to prove the incompatibility

AC

CE P

TE

D

MA

NU

SC R

IP

T

