RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Thermal Stability of Simvastatin under Different Atmospheres 1 ˜ ˜ OLIVEIRA,2 CARLOS CORDEIRO,1 RICARDO G. SIMOES, HERM´INIO P. DIOGO,2 ANA DIAS,2 MARIA CONCEIC ¸ AO 1,2 1 CARLOS E. S. BERNARDES, MANUEL E. MINAS DA PIEDADE 1
Centro de Qu´ımica e Bioqu´ımica e Departamento de Qu´ımica e Bioqu´ımica, Faculdade de Ciˆencias, Universidade de Lisboa, Lisbon 1749-016, Portugal 2 Centro de Qu´ımica Estrutural, Instituto Superior T´ecnico, Universidade de Lisboa, Lisbon 1049-001, Portugal Received 2 August 2013; revised 25 October 2013; accepted 25 October 2013 Published online 22 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.23787
ABSTRACT: Simvastatin (SV) is a widely used drug for the treatment of hypercholesterolemia in humans. Nevertheless, serious efforts are still being made to develop new SV formulations with, for example, improved tabletability or bioavailability properties. These efforts frequently involve heating the compound well above ambient temperature or even fusion. In this work, the thermal stability of solid SV under different atmospheres was investigated by using isothermal tests in glass ampules, differential scanning calorimetry, and Calvetdrop microcalorimetry experiments. These tests were combined with analytical data from diffuse reflectance infrared Fourier-transform spectroscopy and liquid chromatography coupled with tandem mass spectrometry or Fourier transform ion cyclotron resonance mass spectrometry (LC–FT–ICR–MS). No decomposition was observed when the sample was kept at a temperature ≤373 K under N2 or reduced pressure (13.3 Pa) atmospheres. Thermal degradation was, however, observed for temperatures ≥353 K in the presence of pure or atmospheric oxygen. The nature of the two main oxidative degradation products was determined through MS/MS experiments and accurate mass measurements of the precursor ions using FT–ICR–MS. The obtained results indicated that the decomposition process involves the C 2013 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm oxidation of the hexahydronaphthalene fragment of SV. Sci 103:241–248, 2014 Keywords: calorimetry; mass spectrometry; thermal analysis; degradation products; diffuse reflectance; solid-state stability
INTRODUCTION Statins are the main class of active pharmaceutical ingredients currently in use to control hypercholesterolemia in humans. They reduce plasma levels of low-density lipoprotein cholesterol (LDL-c) particles by inhibiting hydroxymethylglutaryl-CoA reductase (EC 1.1.1.34).1–3 Several studies have shown that high plasma concentration of LDL-c lead to atherosclerosis, a predisposing factor for the development of cardiovascular diseases. Therefore, statin therapy has been playing a major role in the pharmaceutical armamentarium for the prevention and treatment of such diseases.3–5 An important member of the statin family is simvastatin (SV) (Fig. 1; C25 H38 O5 , CAS number [79902–63–9], 1S,3R,7S,8S,8aR) - 8 - {2 - [(2R,4R) - 4-hydroxy-6-oxotetrahydro2H-pyran - 2 - yl]ethyl} - 3,7 - dimethyl - 1,2,3,7,8,8a-hexahydronaphthalen-1-yl 2,2-dimethylbutanoate), which was approved for marketing by the United States Food and Drug Administration in 1991 and is still widely prescribed to treat hypercholesterolemia.1–3,6 It is normally administered in the lactone form (Fig. 1), which is traditionally prepared by direct alkylation of lovastatin, a similar statin obtained as a secondary metabolite from the filamentous fungus Aspergillus terreus.7–9 Thermal stability studies of crystalline10,11 and amorphous11–14 SV from ambient temperature (∼298 K) to the melting point (∼412 K)10,11,15–19 are scarce, and, to the Correspondence to: Manuel E. Minas da Piedade (Telephone: +351-217500866; Fax: +351-21-7500088; E-mail: [email protected]) This article contains supplementary material available from the authors upon request or via the Internet at http://onlinelibrary.wiley.com/. Journal of Pharmaceutical Sciences, Vol. 103, 241–248 (2014)
C 2013 Wiley Periodicals, Inc. and the American Pharmacists Association
Figure 1. Molecular structure of simvastatin (SV) lactone form.
best of our knowledge, a detailed characterization of degradation products has not yet been reported.20 This information is, however, very important for various aspects of SV use, such as the reliable definition of storage conditions or the development of new formulations that imply heating. There have been, for example, serious efforts to develop efficient methodologies for solubility enhancement of SV because its bioavailability is essentially limited by a poor solubility in aqueous media.2,21 Some promising strategies involve the production of SV glasses or solid dispersions in hydrophilic carriers (e.g., polyvinylpyrrolidone).2,21 Such methods require fusion of SV or at least heating the sample well above ambient temperature. Organic pharmaceuticals often degrade through oxidation reactions22 and the picture emerging from reports on the thermal behavior of solid SV indicate that the compound may undergo an exothermic oxidative degradation in air, which is very slow at ambient temperature and has an onset at approximately 401 K when the sample is heated in a differential
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
241
242
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
scanning calorimeter at a rate of 2 K/min.20 Under inert atmosphere (e.g., N2 ), no thermal events are observed between 298 K and the fusion temperature (412 K), except for a glass transition at approximately 306 K if an amorphous phase is present.10,11,13,14 Thermogravimetry experiments carried out at 10 K/min indicate that after fusion SV undergoes pyrolysis with onset at approximately 443 or 476 K in dynamic air or dynamic N2 atmospheres, respectively.10,23 In this work, a comprehensive experimental study of the thermal stability of SV under different atmospheres with identification of the major degradation products is reported. In addition to isothermal tests carried out in glass ampules, dynamic and isothermal experiments were also carried out by differential scanning calorimetry (DSC) and Calvet-drop microcalorimetry, respectively. These two types of technique are widely used in accelerated stability tests of solid-state APIs, and provide important information such as the temperature onsets and exothermic or endothermic nature of the degradation processes.24
MATERIALS AND METHODS Materials The SV sample (Jubilant Organosys, Mysore, Karnataka, India) was previously characterized in terms of chemical purity (98.88±0.12%), phase purity (form I, orthorhombic, P21 21 21 , Z /Z = 1/4), and morphology by elemental analysis, diffuse reflectance infrared Fourier-transform (DRIFT) spectroscopy, 1 H and 13 C NMR, X-ray powder diffraction, scanning electron microscopy, DSC, thermogravimetry, and LC–ESI/MS.18 Acetonitrile (LC-MS grade; Fisher Scientific, Loughborough, Leicestershire, U.K.), acetic acid (p.a.; Sigma–Aldrich, Sintra, Portugal), and deionized water (Millipore Simplicity Simpak 2, R = 18.2 M cm, U.S.A.) were used in the LC–MS analysis. Nitrogen (Alphagaz N2 -1; , Ar L´ıquido, Lisbon, Portugal), oxygen (Alphagaz O2 -1; Ar L´ıquido, Lisbon, Portugal), and dry air obtained from a compressed air apparatus equipped with a desiccant dryer element were used. Water saturated air (air+H2 O), nitrogen (N2 +H2 O), or oxygen (O2 +H2 O) atmospheres were produced by passing the gases through a water bubbler at a flow of 26±1 cm3 /s. The relative humidity of atmospheric air (57%–60%) was monitored with a TFA 30502 sensor. R
DRIFT Spectroscopy DRIFT measurements were performed in the range 400– 4000 cm−1 , with a resolution of 2 cm−1 , on a Nicolet 6700 spectrometer (Thermo Fisher Scientific, Waltham, MA, U.S.A.). The samples were dispersed in KBr. Liquid Chromatography–Tandem Mass Spectrometry Sample solutions (about 0.4 mg/cm3 ) prepared in acetonitrile were analyzed with a ProStar 410 autosampler, two 210-LC chromatography pumps, a ProStar 335 diode array detector, and a 500-MS ion trap mass spectrometer equipped with an electrospray ionization (ESI) source (Varian, Palo Alto, CA, U.S.A.). Data acquisition and processing were performed using the Varian MS Control 6.9.3 software. The samples were injected onto the column via a Rheodyne injector with a 100 :L loop in the pick-up injection mode. Separations were carried out with a Phenomenex Luna C18 (2) column (150 × 2 mm2 , 3 :m), at a flow rate of 200 :L/min, using a 5 min linear gradient from ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
50% to 70% (v/v) acetonitrile in 2×10−3 mol/dm3 ammonium acetate, pH 4.0, followed by a 10 min linear gradient to 100% acetonitrile, and an 8 min isocratic elution with acetonitrile. The UV absorbance was monitored at 238 nm. The mass spectrometer was operated in the positive and negative ESI mode, with the following optimized parameters: ion spray voltage, ±4.5 kV; capillary voltage, 20 V, and RF loading, 80%. Nitrogen was used as nebulizing and drying gas, at pressures of 35 and 10 psi, respectively; the drying gas temperature was set at 623 K. The positive tandem mass spectra (MS/MS) were obtained with an isolation window of 2.0 Da, excitation energy values between 0.9 and 1.2 V, and an excitation time of 10 ms. Fourier Transform Ion Cyclotron Resonance Mass Spectrometry High-resolution mass measurements were performed on an Apex Qe FTICR 7 Tesla mass spectrometer (Bruker Daltonics, Bremen, Germany) equipped with an Apollo II dual ESI/MALDI source (Bruker Daltonics, Bremen, Germany). Spectra were acquired in positive ESI mode with external calibration. For LC–MS analysis, an Agilent 1200 capLC system (Agilent Technologies, Santa Clara, CA, U.S.A.) was used, at a flow rate of 10 cm3 /min. The gradient program and the solvents were identical to those used in the LC–MS/MS analysis, the column was a Zorbax C18 150 × 0.5 mm2 . Spectra were processed with the data Analysis 4.0 software package from Bruker Daltonics. Thermal Stability Studies in Glass Ampules Glass ampules (ca. 90 mm length and 7 mm internal diameter) were loaded with 51–380 mg of SV and closed under the following atmospheres: reduced pressure of 13.3 Pa, atmospheric air (57%–60% relative humidity), air+H2 O, N2 , N2 +H2 O, O2 , and O2 +H2 O. The ampules were sealed by means of a blow-torch, except in the case of the oxygen atmospheres where a rubber stopper reinforced with an Apiezon Q seal was employed. All weightings were performed with a precision of ±0.01 mg in a Mettler AT201 balance (Mettler-Toledo, Columbus, OH, U.S.A.). The reduced pressure atmosphere was attained by pumping the ampule containing the SV sample with a rotary pump for 30 min before sealing. The ampule charged with atmospheric air was sealed under normal laboratory conditions (295±1 K, 57%–60% relative humidity). The dry oxygen and all water saturated atmospheres were achieved by purging the ampule for 30 min with the appropriate gas, before sealing. The N2 atmosphere was obtained by connecting the ampule to a vacuum/N2 line and performing three cycles consisting of pumping to 13.3 Pa and then filling with N2 at approximately 1 bar. The thermal stability of SV under the various atmospheres mentioned above was first tested by immersing the ampules containing the sample in a silicon oil bath kept at 373±1 K for 14 h. The effect of increasing the duration of the experiment at 373±1 K to 26 h was investigated in air and N2 atmospheres. Experiments with fixed 18 h duration were performed at different temperatures (343±1 K, 353±1 K, 363±1 K, and 373±1 K) for atmospheric air. Finally, a test was made where a sample under reduced pressure was kept at 423 K (above the fusion temperature) for 5 min, and left to cool to ambient temperature after removal from the bath. DOI 10.1002/jps.23787
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Differential Scanning Calorimetry
RESULTS AND DISCUSSION
The DSC experiments were carried out on a DSC 7 from PerkinElmer (Waltham, MA, U.S.A.). The DSC studies included (1) the comparison of the materials present inside the ampules at the end of the thermal stability experiments with the original SV sample and (2) the thermal decomposition of SV under atmospheric air. In the first case, the samples with 0.3–4.9 mg mass were sealed in air, inside aluminum crucibles. Each crucible was transferred to the apparatus and heated at a rate of 10 K/min in the range 300–420 K. Nitrogen (Air Liquide N45), at a flow rate of 0.5 cm3 /s was used as the purging gas. In the second case, the mass of the SV samples was 1–3 mg but the crucibles were not sealed and no purging gas was used. The heating rate was 2 K/min and the range 300–420 K. The temperature and heat flow scales of the instrument were calibrated at the same heating rates with indium (PerkinElmer, Waltham, MA, U.S.A.; mass fraction: 0.99999; Tfus = 429.75 K, fus ho = 28.45 J/g). All weightings were performed with a precision of ±0.1 :g on a Mettler XP2U ultra-microbalance.
Thermal Stability Studies in Glass Ampules
Calvet-Drop Microcalorimetry The thermal stability of SV in air (57%–60% relative humidity) and N2 atmospheres was also investigated with an electrically calibrated twin-cell Calvet microcalorimeter.25,26 In a typical experiment, the sample with a mass of 0.23–1.59 mg was placed into a small glass capillary and weighed with a precision of ±0.1 :g on a Mettler XP2U ultra-microbalance. The capillary was equilibrated for approximately 600 s, at 298.3±0.4 K, inside a furnace placed above the entrance of the sample cell, and dropped into the cell under air or N2 atmospheres. The measuring curve subsequently observed was recorded for periods that varied from 3 h to 1 week in different experiments. The temperature of the cell was set to 352 K, 363 K, 367 K, 369 K, or 375 K in the experiments with air and to 375 K for the N2 atmosphere. This temperature was controlled to better than ±0.1 K with a Eurotherm 2404 PID unit and measured with a precision of ±0.1 K by a Tecnisis 100 platinum resistance thermometer.
Table 1.
The results of the thermal stability studies on SV performed in sealed glass ampules and of the analysis of the final products by LC–MS/MS are summarized in Table 1, where m is the mass of sample, T is the temperature of the thermostatic oil bath, and t the duration of the experiment. In the case of runs 8 and 13, where a heterogeneous product consisting of a white powder and yellow glass was obtained, fractions of both phases were independently analyzed. The identification of species D1 and D2 mentioned in Table 1 is described below in the section on characterization of oxidative degradation products by LC–MS/MS. Observation of the materials inside the ampules at the end of the experiments indicated that: (1) when SV was kept at 373 K under a dry or water saturated nitrogen atmosphere (runs 3–5), no change of the sample (e.g., color, morphology) was noted, regardless of the duration of the experiment. This is illustrated in Figure 2a for run 5, with 26 h duration. The previous conclusion is corroborated by the results of LC–MS/MS analysis (Table 1), which showed no evidence of degradation products except in the case of run 4 where traces of D1 were detected. (2) No decomposition was also observed in runs 1 and 2, carried out at 373 and 423 K, respectively, under reduced pressure (13.3 Pa). However, in the case of run 2, where SV was heated above its fusion temperature, a glass phase was formed (Fig. 2b) on cooling to ambient temperature (293±2 K). (3) In the presence of oxygen (runs 6–14), different extents of decomposition were noted. When T = 343 K no color changes or degradation products were observed. For T = 353 K, the sample seemed unaltered but traces of D1 were found by LC– MS/MS. Above 353 K, decomposition was clearly evidenced by the formation of a yellow glass (Fig. 2c) and further confirmed by the detection of degradation products D1 and D2 in the LC–MS/MS analysis. The products of the thermal stability experiments summarized in Table 1 were also analyzed by DSC. The corresponding results are listed in Table 2, where m is the mass of sample, Tg represents the glass transition temperature, Ton and Tmax are the onset and maximum temperatures of the fusion peak,
Results of the Thermal Stability Studies Carried Out in Glass Ampulesa
Run number 1 2 3 4 5 6 7 8 9 10 11 12 13 14
243
m (mg)
T (K)
Atmosphere
t (h)
Product Appearance
LC–MS/MS Analysis
86.77 107.60; 125.57 82.61 81.53 369.12 70.79 50.58 63.60 209.05 124.39 115.57 132.97 78.03 384.26
373 423 373 373 373 373 373 373 343 353 363 373 373 373
Reduced pressure (13.3 Pa) Reduced pressure (13.3 Pa) N2 N2 +H2 O N2 O2 O2 +H2 O Air Air Air Air Air Air+H2 O Air
14 0.08 14 14 26 14 14 14 18 18 18 18 14 26
White powder Colorless glass White powder White powder White powder Yellow glass Yellow glass White powder + yellow glass White powder White powder Yellow glass Yellow glass White powder + yellow glass Yellow glass
No degradation products No degradation productsb No degradation products Traces of D1 No degradation products Formation of D1 and D2 Formation of D1 and D2 Formation of D1 and D2 No degradation products Traces of D1 Formation of D1 and D2 Formation of D1 and D2 Formation of D1 and D2 Formation of D1 and D2
a Air refers to atmospheric air with 57%–60% relative humidity and air+H2 O, N2 +H2 O, and O2 +H2 O to water saturated atmospheres. D1 and D2 refer to the products of thermal decomposition discussed in the section on characterization of the oxidative degradation products of simvastatin by LC–MS/MS (see below). b The colorless glasses obtained in the two runs were mixed and analyzed by LC–MS/MS.
DOI 10.1002/jps.23787
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
244
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 2. Images of the samples in the ampules at the end of (a) run 5, (b) run 2, and (c) run 6 (see Table 1).
Table 2. Results of the DSC Analysis on the Simvastatin Starting Material and the Products of Thermal Stability Studies in Glass Ampules (see also Table 1) Fusion Run Number Starting material 1 2a 3 4 5 6 7 8 (white fraction) 8 (yellow fraction) 9 10 11 12 13 (white fraction) 13 (yellow fraction) 14 a
m (mg) Tg (K) 0.839 0.422 4.864 4.362 0.767 0.276 0.388 0.982 0.530 0.783 1.391 1.031 1.298 1.396 0.502 0.555 0.742 0.589
Ton (K) 411.6 408.6
Tmax (K) fus ho (J/g) 413.3 410.8
67.90 64.76
302 303
302 304 309
303 295 318 320
406.6 408.7 408.8 345.0 367.1 407.8 361.9 406.2 404.6 360.5 346.3 406.7 380.3 354.3
409.6 410.4 410.4 370.1 372.8 410.4 386.8 409.9 408.9 374.4 350.8 411.0 394.8 374.1
52.95 65.46 61.23 39.20 8.36 60.38 25.40 42.19 42.63 23.12 13.30 45.73 27.49 11.93
Results corresponding to each duplicate of run 2 in Table 1.
respectively, and fus ho is the standard specific enthalpy of fusion. A comparison between the DSC measured curves of the starting material (scan rate of 10 K/min) and of the different crystalline and glassy products obtained after 14 h at 373 K under different atmospheres is given as Supporting Information. The formation of an amorphous phase and decomposition products upon thermal treatment of SV in the presence of oxygen is reflected in the DSC results by (1) a notable decrease of the onset temperature (31–67 K) and enthalpy of fusion (29–60 J/g) relative to the starting material, when a yellow glass is obtained (Table 2); (2) the observation of a glass transition in the range 295–320 K for those samples. The fact that a 3–14 J/g decrease in fus ho is noted, even when no degradation products were found (runs 1, 3, and 5), also suggests that the thermal treatment may consistently lead to some degree of amorphization of the sample. The amorphous nature of the yellow glass products is further evidenced by a notable broadening of the OH stretching band ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
in the DRIFT spectra (Fig. 3). This band is located at 3552 cm−1 in the spectrum of the starting material. Finally, the onset of thermal decomposition when the samples are subjected to increasing temperatures under air can be noted at T > 353 K in the DRIFT and DSC results shown in Figure 4. Nevertheless, these two analyzes do not seem to be sensitive to the traces of D1 detected by LC–MS/MS at 353 K (Table 1). Thermal Stability Studies by DSC and Calvet-Drop Microcalorimetry The thermal stability of SV under N2 and air atmospheres was also studied in non-isothermal and isothermal conditions by using DSC and Calvet-drop microcalorimetry, respectively. DSC experiments carried out in air (non-sealed aluminum crucibles) at a rate of 2 K/min showed an exothermic event with onset at Ton = 397±3 K (see Supporting Information). Here, the indicated uncertainty corresponds to the standard error of the mean of four independent determinations. This thermal event had been previously observed by DSC and assigned to oxidative decomposition of SV, albeit without identification of the decomposition products.20 Similar conclusions were drawn from the results of a series of isothermal Calvet-drop microcalorimetry experiments with 18 h duration. The corresponding measured curves are given as Supporting Information. No thermal event was observed when N2 atmosphere was present inside the calorimetric cell and the temperature was set to 375 K. This was also true for an experiment with approximately 1 week duration. In contrast, for temperatures higher than 367 K and under air, an exothermic peak was observed, which can be assigned to the oxidative degradation process described in the previous section. The fact that the peak becomes broader and the induction time for the onset of decomposition increases as the temperature decreases is compatible with the expected decrease of the reaction rate as the temperature decreases. The decomposition of SV above 367 K was evidenced by the glassy and yellowish aspect of the material inside the capillary removed from the calorimetric cell at the end of each run. This observation corroborates the conclusions of the studies made in glass ampules (Table 1). Although no peak could be noted in the measuring curve corresponding to 363 K, partial decomposition of the sample was, nevertheless, suggested by the yellowish color of the material obtained at the end of the experiment. DOI 10.1002/jps.23787
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
245
Figure 3. DRIFT spectra of the original simvastatin sample and of the materials obtained after thermal treatment (14 h, 373 K) under different atmospheres. The result for run 2 refers to the mixture of the two colorless glass products obtained in the two independent experiments. All spectra were normalized relative to the most intense peak.
It can therefore be concluded that the observations of both the DSC and Calvet-drop microcalorimetry measurements are consistent with the results of the studies carried out in glass ampules (Table 1). Characterization of the Oxidative Degradation Products of SV by LC–MS/MS Liquid chromatography–ESI/MS chromatograms of a control sample of simvastatin showed only a major peak corresponding to SV, at retention time of 12.5 min, along with minor signals related with the main impurities previously identified in the sample by LC–MS/MS (see Supporting Information for details).18 The same LC–MS method was applied to the analysis of the products of the thermal stability studies summarized in Table 1. As indicated in Table 1, no significant degradation was observed when the starting material was subjected to thermal treatment under nitrogen atmosphere. In the presence of pure or atmospheric oxygen, two main oxidative degradation DOI 10.1002/jps.23787
products with m/z 433 (peaks D1 and D1 ) and m/z 435 (peaks D2 and D2 ) were detected by the LC–MS at retention times between 4.5 and 6.8 min. The presence of two peaks with the same m/z value and close retention times is indicative of isomeric degradation products. This is illustrated in Figure 5 for the product of run 6. The protonated molecules corresponding to these signals showed a mass increase of 14 and 16 Da relative to SV. The UV spectrum of D2 exhibited, as for SV, an absorbance maximum at 241 nm characteristic of cyclodienes, whereas that of D1 presented an absorbance maximum at 290 nm (see Supporting Information). This shift toward a higher wavelength suggests an increase of conjugation in the cyclodiene chromophore structure of D1 relative to the parent molecule. The identification of the oxidative degradation products D1 and D2 was achieved by (1) structural characterization based on a comparison of the MS/MS fragmentation patterns of a protonated SV molecule (previously reported27 and further confirmed in ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
246
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 4. DCS measured curves (a) and DRIFT spectra (b) of the original simvastatin sample and of the materials obtained after 18 h under air at different temperatures. The heat flow in the DSC curves was scaled by mass (J /g) and all DRIFT spectra were normalized relative to the most intense peak.
this work) and of the oxidative degradation products; (2) accurate mass measurements of the precursor ions using FT–ICR– MS, which led to the following results (M = measured monoisotopic mass of precursor ion; C = calculated monoisotopic mass of precursor ion; = deviation in ppm; F = deduced elemental formula of protonated molecule): (1) SV, M = 419.2795, C = 419.2797, = 0.48, F = C35 H39 O5 ; (2) product D1, M = 433.2589, C = 433.2590, = 0.23, F = C35 H37 O6 ; (3) product D2, M = 435.2742, C = 435.2747, = 1.15, F = C35 H39 O6 . The tandem mass spectra of the protonated molecules of D1 (m/z 433), D2 (m/z 435), and SV (m/z 419) are given as Supporting Information (Figure S7). The MS2 spectra of D1 and D2 follow a similar dissociation pattern (data not shown). The general fragmentation patterns of D1 and D2 ions are the same as those for SV. The main fragment ions at m/z 303, 285, and 225 for SV increased by 14 and 16 Da to become the product ions at m/z 317, 299, and 239 for D1, and m/z 319, 301, and 241 for D2, respectively. These results clearly indicated that the degradation products were formed through oxidation of the hexahydronaphthalene portion of SV. This assumption is also supported by the presence of two small peaks at m/z 173 (175) and 143 (145) in the tandem mass spectrum of D1 (D2), which correspond to the species indicated in Figures S8 and S9 of the Supporting Information. It is accepted that SV is susceptible to oxidation at the conjugated diene presumably through a mechanism involving radical initiation, propagation, and termination steps. The main pathway is based on the formation of a resonance-stabilized free radical (FR1 or FR2) whose resonance forms may account for the oxidation at different positions. Oxidation at a tertiary position will yield a stable alcohol (D2a and D2d), whereas oxidation at an allylic position will yield an alcohol (D2b and D2c) prone to further oxidation to a dienone (D1b and D1c), with a higher conjugated structure. A proposed mechanism is given as Supporting Information ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
(Figure S6). The formation of the species D2b and D1b is reported on earlier studies of autoxidation of SV in solution.28 Schemes of the fragmentation mechanism for the oxidative degradation products D1 and D2 are given as Supporting Information.
CONCLUSIONS The experiments carried out in this work indicated that, provided SV is handled under N2 or reduced pressure atmospheres (13.3 Pa) at a temperature ≤373 K, no thermal decomposition is to be expected, at least within 1 week. Significant thermal degradation does, however, occur at temperatures ≥353 K in the presence of pure or atmospheric oxygen. The two main oxidative degradation products obtained in this case result from the oxidation of the hexahydronaphthalene fragment of SV. It is therefore hoped that the results here reported will provide useful guidelines for the development of SV solid formulations that require thermal treatment.
ACKNOWLEDGMENTS ˜ para a Ciˆencia e a This work was supported by Fundac¸ao Tecnologia (FCT), Portugal (Projects PTDC/QUI-QUI/098216/ 2008, PEst-OE/QUI/UI0612/2013, PEst-OE/QUI/UI0100/2013, REDE/1501/REM/2005, and REDE/1502/REM/2005). PhD (SFRH/BD/48410/2008) and Post-Doctoral (SFRH/BPD/43346/ 2008) grants from FCT are also gratefully acknowledged by R. G. Sim˜oes and C. E. S. Bernardes, respectively. We also thank Prof. Matilde Marques (CQE-IST) for helpful discussions about the oxidation mechanism of SV. DOI 10.1002/jps.23787
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
247
Figure 5. LC–DAD–ESI/MS chromatograms of the material obtained in run 6 (see Table 1): (a) LC-DAD profile at 8 = 238 nm, (b) total ion chromatogram and mass spectrum of the protonated molecule of simvastatin (SV), (c) extracted ion chromatogram and mass spectrum of m/z 433 ion, (d) extracted ion chromatogram and mass spectrum of m/z 435 ion.
REFERENCES 1. Li JJ. 2009. Triumph of the heart. The story of statins. New York: Oxford University Press. 2. Tiwari R, Pathak K. 2011. Statins therapy: A review on conventional and novel formulation approaches. J Pharm Pharmacol 63:983– 998. 3. Taylor F, Ward K, Moore THM, Burke M, Davey Smith G, Casas JP, Ebrahim S, Eds. 2012. Statins for the primary prevention of cardiovascular disease. New York: John Wiley. 4. Vance DE, Van den Bosch H. 2000. Cholesterol in the year 2000. Biochem Biophys Acta 1529:1–8.
DOI 10.1002/jps.23787
5. Roger VL, Go AS, Lloyd-Jones DM, Benjamin EJ, Berry JD, Borden WB, Bravata DM, Dai S, Ford ES, Fox CS, Fullerton HJ, Gillespie C, Hailpern SM, Heit JA, Howard VJ, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Makuc DM, Marcus GM, Marelli A, Matchar DB, Moy CS, Mozaffarian D, Mussolino ME, Nichol G, Paynter NP, Soliman EZ, Sorlie PD, Sotoodehnia N, Turan TN, Virani SS, Wong ND, Woo D, Turner MB, Comm AHAS, Subcomm SS. 2012. Heart disease and stroke statistics—2012 update. A report from the American Heart Association. Circulation 125:E2–E220. 6. Baxendale IR, Hayward JJ, Ley SV, Tranmer GK. 2007. Pharmaceutical strategy and innovation: An academics perspective. ChemMedChem 2:768–788.
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
248
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
7. Hoffman WF, Alberts AW, Anderson PS, Chen JS, Smith RL, Willard AK. 1986. 3-Hydroxy-3-methylglutaryl-coenzyme a reductase inhibitors. 4. Side-chain ester derivatives of mevinolin. J Med Chem 29:849–852. 8. Askin D, Verhoeven TR, Liu TMH, Shinkai I. 1991. Synthesis of Synvinolin—Extremely high conversion alkylation of an ester enolate. J Org Chem 56:4929–4932. 9. Barreiro EJ, Kummerle AE, Fraga CAM. 2011. The methylation effect in medicinal chemistry. Chem Rev 111:5215–5246. ˜ MM, Silva MCD, Soledade LEB, Souza AG. 10. Souza MAF, Conceic¸ao 2007. Thermal and kinetic study of statins. J Therm Anal Calorim 87:859–863. 11. Graeser KA, Strachan CJ, Patterson JE, Gordon KC, Rades T. 2008. Physicochemical properties and stability of two differently prepared amorphous forms of simvastatin. Cryst Growth Des 8:128–135. 12. Ambike AA, Mahadik KR, Paradkar A. 2005. Spray-dried amorphous solid dispersions of simvastatin, a low Tg drug: In vitro and in vivo evaluations. Pharm Res Dordr 22:990–998. 13. Graeser KA, Patterson JE, Zeitler JA, Gordon KC, Rades T. 2009. Correlating thermodynamic and kinetic parameters with amorphous stability. Eur J Pharm Sci 37:492–498. 14. Graeser KA, Patterson JE, Rades T. 2009. Applying thermodynamic and kinetic parameters to predict the physical stability of two differently prepared amorphous forms of simvastatin. Curr Drug Deliv 6:374–382. 15. Ambike AA, Mahadik KR, Paradkar A. 2005. Physico-chemical characterization and stability study of glassy simvastatin. Drug Dev Ind Pharm 31:895–899. 16. Jun SW, Kim MS, Kim JS, Park HJ, Lee S, Woo JS, Hwang SJ. 2007. Preparation and characterization of simvastatin/hydroxypropyl-betacyclodextrin inclusion complex using supercritical antisolvent (SAS) process. Eur J Pharm Biopharm 66:413–421. 17. Oliveira MA, Yoshida MI, Gomes ECL, Mussel WN, Vianna-Soares ´ ˜ CD, Pianetti GA. 2010. Analise t´ermica aplicada a` caracterizac¸ao
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
da sinvastatina em formulac¸o˜ es farmacˆeuticas. Quim Nova 8:653– 1657. 18. Sim˜oes RG, Bernardes CES, Diogo HP, Agapito F, Minas da Piedade ME. 2013. Energetics and structure of simvastatin. Mol Pharm 10:2713–2722. 19. Nti-Gyabaah J, Chan V, Chiew YC. 2009. Solubility and limiting activity coefficient of simvastatin in different organic solvents. Fluid Phase Equilibr 280:35–41. 20. Ellison DK, Moore WD, Petts CR. 1993. Simvastatin. In Analytical profiles of drug substances and excipients; Brittain HG, Ed. San Diego: Academic Press. 21. Murtaza G. 2012. Solubility enhancement of simvastatin: A review. Acta Pol Pharm 69:581–590. 22. Hovorka SW, Sch¨oneich C. 2001. Oxidative degradation of pharmaceuticals: Theory, mechanisms and inhibition. J Pharm Sci 90:253–269. 23. Proc´opio JVV, Souza VG, Costa RA, Correia LP, Souza FS, Macˆedo RO. 2011. Application of thermal analysis and pyrolysis coupled to GC/MS in the qualification of simvastatin pharmaceutical raw material. J Therm Anal Calorim 106:665–670. 24. Craig DQM, Reading M. 2006. Thermal analysis of pharmaceuticals. Boca Raton: CRC Press. 25. Kiyobayashi T, Minas da Piedade ME. 2001. The standard molar enthalpy of sublimation of 05 -bis-pentamethylcyclopentadienyl iron measured with an electrically calibrated vacuum-drop sublimation microcalorimetric apparatus. J Chem Thermodyn 33:11–21. 26. Bernardes CES, Santos LMNBF, Minas da Piedade ME. 2006. A new calorimetric system to measure heat capacities of solids by the drop method. Meas Sci Technol 17:1405–1408. 27. Wang H, Wu YH, Zhao ZX. 2001. Fragmentation study of simvastatin and lovastatin using electrospray ionization tandem mass spectrometry. J Mass Spectrom 36:58–70. 28. Smith GB, Dimichele L, Colwell LF, Dezeny GC, Douglas AW, Reamer RA, Verhoeven TR. 1993. Autooxidation of simvastatin. Tetrahedron 49:4447–4462.
DOI 10.1002/jps.23787
Thermal Stability of Simvastatin under Different Atmospheres 1 ˜ ˜ OLIVEIRA,2 CARLOS CORDEIRO,1 RICARDO G. SIMOES, HERM´INIO P. DIOGO,2 ANA DIAS,2 MARIA CONCEIC ¸ AO 1,2 1 CARLOS E. S. BERNARDES, MANUEL E. MINAS DA PIEDADE 1
Centro de Qu´ımica e Bioqu´ımica e Departamento de Qu´ımica e Bioqu´ımica, Faculdade de Ciˆencias, Universidade de Lisboa, Lisbon 1749-016, Portugal 2 Centro de Qu´ımica Estrutural, Instituto Superior T´ecnico, Universidade de Lisboa, Lisbon 1049-001, Portugal Received 2 August 2013; revised 25 October 2013; accepted 25 October 2013 Published online 22 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.23787
ABSTRACT: Simvastatin (SV) is a widely used drug for the treatment of hypercholesterolemia in humans. Nevertheless, serious efforts are still being made to develop new SV formulations with, for example, improved tabletability or bioavailability properties. These efforts frequently involve heating the compound well above ambient temperature or even fusion. In this work, the thermal stability of solid SV under different atmospheres was investigated by using isothermal tests in glass ampules, differential scanning calorimetry, and Calvetdrop microcalorimetry experiments. These tests were combined with analytical data from diffuse reflectance infrared Fourier-transform spectroscopy and liquid chromatography coupled with tandem mass spectrometry or Fourier transform ion cyclotron resonance mass spectrometry (LC–FT–ICR–MS). No decomposition was observed when the sample was kept at a temperature ≤373 K under N2 or reduced pressure (13.3 Pa) atmospheres. Thermal degradation was, however, observed for temperatures ≥353 K in the presence of pure or atmospheric oxygen. The nature of the two main oxidative degradation products was determined through MS/MS experiments and accurate mass measurements of the precursor ions using FT–ICR–MS. The obtained results indicated that the decomposition process involves the C 2013 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm oxidation of the hexahydronaphthalene fragment of SV. Sci 103:241–248, 2014 Keywords: calorimetry; mass spectrometry; thermal analysis; degradation products; diffuse reflectance; solid-state stability
INTRODUCTION Statins are the main class of active pharmaceutical ingredients currently in use to control hypercholesterolemia in humans. They reduce plasma levels of low-density lipoprotein cholesterol (LDL-c) particles by inhibiting hydroxymethylglutaryl-CoA reductase (EC 1.1.1.34).1–3 Several studies have shown that high plasma concentration of LDL-c lead to atherosclerosis, a predisposing factor for the development of cardiovascular diseases. Therefore, statin therapy has been playing a major role in the pharmaceutical armamentarium for the prevention and treatment of such diseases.3–5 An important member of the statin family is simvastatin (SV) (Fig. 1; C25 H38 O5 , CAS number [79902–63–9], 1S,3R,7S,8S,8aR) - 8 - {2 - [(2R,4R) - 4-hydroxy-6-oxotetrahydro2H-pyran - 2 - yl]ethyl} - 3,7 - dimethyl - 1,2,3,7,8,8a-hexahydronaphthalen-1-yl 2,2-dimethylbutanoate), which was approved for marketing by the United States Food and Drug Administration in 1991 and is still widely prescribed to treat hypercholesterolemia.1–3,6 It is normally administered in the lactone form (Fig. 1), which is traditionally prepared by direct alkylation of lovastatin, a similar statin obtained as a secondary metabolite from the filamentous fungus Aspergillus terreus.7–9 Thermal stability studies of crystalline10,11 and amorphous11–14 SV from ambient temperature (∼298 K) to the melting point (∼412 K)10,11,15–19 are scarce, and, to the Correspondence to: Manuel E. Minas da Piedade (Telephone: +351-217500866; Fax: +351-21-7500088; E-mail: [email protected]) This article contains supplementary material available from the authors upon request or via the Internet at http://onlinelibrary.wiley.com/. Journal of Pharmaceutical Sciences, Vol. 103, 241–248 (2014)
C 2013 Wiley Periodicals, Inc. and the American Pharmacists Association
Figure 1. Molecular structure of simvastatin (SV) lactone form.
best of our knowledge, a detailed characterization of degradation products has not yet been reported.20 This information is, however, very important for various aspects of SV use, such as the reliable definition of storage conditions or the development of new formulations that imply heating. There have been, for example, serious efforts to develop efficient methodologies for solubility enhancement of SV because its bioavailability is essentially limited by a poor solubility in aqueous media.2,21 Some promising strategies involve the production of SV glasses or solid dispersions in hydrophilic carriers (e.g., polyvinylpyrrolidone).2,21 Such methods require fusion of SV or at least heating the sample well above ambient temperature. Organic pharmaceuticals often degrade through oxidation reactions22 and the picture emerging from reports on the thermal behavior of solid SV indicate that the compound may undergo an exothermic oxidative degradation in air, which is very slow at ambient temperature and has an onset at approximately 401 K when the sample is heated in a differential
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
241
242
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
scanning calorimeter at a rate of 2 K/min.20 Under inert atmosphere (e.g., N2 ), no thermal events are observed between 298 K and the fusion temperature (412 K), except for a glass transition at approximately 306 K if an amorphous phase is present.10,11,13,14 Thermogravimetry experiments carried out at 10 K/min indicate that after fusion SV undergoes pyrolysis with onset at approximately 443 or 476 K in dynamic air or dynamic N2 atmospheres, respectively.10,23 In this work, a comprehensive experimental study of the thermal stability of SV under different atmospheres with identification of the major degradation products is reported. In addition to isothermal tests carried out in glass ampules, dynamic and isothermal experiments were also carried out by differential scanning calorimetry (DSC) and Calvet-drop microcalorimetry, respectively. These two types of technique are widely used in accelerated stability tests of solid-state APIs, and provide important information such as the temperature onsets and exothermic or endothermic nature of the degradation processes.24
MATERIALS AND METHODS Materials The SV sample (Jubilant Organosys, Mysore, Karnataka, India) was previously characterized in terms of chemical purity (98.88±0.12%), phase purity (form I, orthorhombic, P21 21 21 , Z /Z = 1/4), and morphology by elemental analysis, diffuse reflectance infrared Fourier-transform (DRIFT) spectroscopy, 1 H and 13 C NMR, X-ray powder diffraction, scanning electron microscopy, DSC, thermogravimetry, and LC–ESI/MS.18 Acetonitrile (LC-MS grade; Fisher Scientific, Loughborough, Leicestershire, U.K.), acetic acid (p.a.; Sigma–Aldrich, Sintra, Portugal), and deionized water (Millipore Simplicity Simpak 2, R = 18.2 M cm, U.S.A.) were used in the LC–MS analysis. Nitrogen (Alphagaz N2 -1; , Ar L´ıquido, Lisbon, Portugal), oxygen (Alphagaz O2 -1; Ar L´ıquido, Lisbon, Portugal), and dry air obtained from a compressed air apparatus equipped with a desiccant dryer element were used. Water saturated air (air+H2 O), nitrogen (N2 +H2 O), or oxygen (O2 +H2 O) atmospheres were produced by passing the gases through a water bubbler at a flow of 26±1 cm3 /s. The relative humidity of atmospheric air (57%–60%) was monitored with a TFA 30502 sensor. R
DRIFT Spectroscopy DRIFT measurements were performed in the range 400– 4000 cm−1 , with a resolution of 2 cm−1 , on a Nicolet 6700 spectrometer (Thermo Fisher Scientific, Waltham, MA, U.S.A.). The samples were dispersed in KBr. Liquid Chromatography–Tandem Mass Spectrometry Sample solutions (about 0.4 mg/cm3 ) prepared in acetonitrile were analyzed with a ProStar 410 autosampler, two 210-LC chromatography pumps, a ProStar 335 diode array detector, and a 500-MS ion trap mass spectrometer equipped with an electrospray ionization (ESI) source (Varian, Palo Alto, CA, U.S.A.). Data acquisition and processing were performed using the Varian MS Control 6.9.3 software. The samples were injected onto the column via a Rheodyne injector with a 100 :L loop in the pick-up injection mode. Separations were carried out with a Phenomenex Luna C18 (2) column (150 × 2 mm2 , 3 :m), at a flow rate of 200 :L/min, using a 5 min linear gradient from ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
50% to 70% (v/v) acetonitrile in 2×10−3 mol/dm3 ammonium acetate, pH 4.0, followed by a 10 min linear gradient to 100% acetonitrile, and an 8 min isocratic elution with acetonitrile. The UV absorbance was monitored at 238 nm. The mass spectrometer was operated in the positive and negative ESI mode, with the following optimized parameters: ion spray voltage, ±4.5 kV; capillary voltage, 20 V, and RF loading, 80%. Nitrogen was used as nebulizing and drying gas, at pressures of 35 and 10 psi, respectively; the drying gas temperature was set at 623 K. The positive tandem mass spectra (MS/MS) were obtained with an isolation window of 2.0 Da, excitation energy values between 0.9 and 1.2 V, and an excitation time of 10 ms. Fourier Transform Ion Cyclotron Resonance Mass Spectrometry High-resolution mass measurements were performed on an Apex Qe FTICR 7 Tesla mass spectrometer (Bruker Daltonics, Bremen, Germany) equipped with an Apollo II dual ESI/MALDI source (Bruker Daltonics, Bremen, Germany). Spectra were acquired in positive ESI mode with external calibration. For LC–MS analysis, an Agilent 1200 capLC system (Agilent Technologies, Santa Clara, CA, U.S.A.) was used, at a flow rate of 10 cm3 /min. The gradient program and the solvents were identical to those used in the LC–MS/MS analysis, the column was a Zorbax C18 150 × 0.5 mm2 . Spectra were processed with the data Analysis 4.0 software package from Bruker Daltonics. Thermal Stability Studies in Glass Ampules Glass ampules (ca. 90 mm length and 7 mm internal diameter) were loaded with 51–380 mg of SV and closed under the following atmospheres: reduced pressure of 13.3 Pa, atmospheric air (57%–60% relative humidity), air+H2 O, N2 , N2 +H2 O, O2 , and O2 +H2 O. The ampules were sealed by means of a blow-torch, except in the case of the oxygen atmospheres where a rubber stopper reinforced with an Apiezon Q seal was employed. All weightings were performed with a precision of ±0.01 mg in a Mettler AT201 balance (Mettler-Toledo, Columbus, OH, U.S.A.). The reduced pressure atmosphere was attained by pumping the ampule containing the SV sample with a rotary pump for 30 min before sealing. The ampule charged with atmospheric air was sealed under normal laboratory conditions (295±1 K, 57%–60% relative humidity). The dry oxygen and all water saturated atmospheres were achieved by purging the ampule for 30 min with the appropriate gas, before sealing. The N2 atmosphere was obtained by connecting the ampule to a vacuum/N2 line and performing three cycles consisting of pumping to 13.3 Pa and then filling with N2 at approximately 1 bar. The thermal stability of SV under the various atmospheres mentioned above was first tested by immersing the ampules containing the sample in a silicon oil bath kept at 373±1 K for 14 h. The effect of increasing the duration of the experiment at 373±1 K to 26 h was investigated in air and N2 atmospheres. Experiments with fixed 18 h duration were performed at different temperatures (343±1 K, 353±1 K, 363±1 K, and 373±1 K) for atmospheric air. Finally, a test was made where a sample under reduced pressure was kept at 423 K (above the fusion temperature) for 5 min, and left to cool to ambient temperature after removal from the bath. DOI 10.1002/jps.23787
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Differential Scanning Calorimetry
RESULTS AND DISCUSSION
The DSC experiments were carried out on a DSC 7 from PerkinElmer (Waltham, MA, U.S.A.). The DSC studies included (1) the comparison of the materials present inside the ampules at the end of the thermal stability experiments with the original SV sample and (2) the thermal decomposition of SV under atmospheric air. In the first case, the samples with 0.3–4.9 mg mass were sealed in air, inside aluminum crucibles. Each crucible was transferred to the apparatus and heated at a rate of 10 K/min in the range 300–420 K. Nitrogen (Air Liquide N45), at a flow rate of 0.5 cm3 /s was used as the purging gas. In the second case, the mass of the SV samples was 1–3 mg but the crucibles were not sealed and no purging gas was used. The heating rate was 2 K/min and the range 300–420 K. The temperature and heat flow scales of the instrument were calibrated at the same heating rates with indium (PerkinElmer, Waltham, MA, U.S.A.; mass fraction: 0.99999; Tfus = 429.75 K, fus ho = 28.45 J/g). All weightings were performed with a precision of ±0.1 :g on a Mettler XP2U ultra-microbalance.
Thermal Stability Studies in Glass Ampules
Calvet-Drop Microcalorimetry The thermal stability of SV in air (57%–60% relative humidity) and N2 atmospheres was also investigated with an electrically calibrated twin-cell Calvet microcalorimeter.25,26 In a typical experiment, the sample with a mass of 0.23–1.59 mg was placed into a small glass capillary and weighed with a precision of ±0.1 :g on a Mettler XP2U ultra-microbalance. The capillary was equilibrated for approximately 600 s, at 298.3±0.4 K, inside a furnace placed above the entrance of the sample cell, and dropped into the cell under air or N2 atmospheres. The measuring curve subsequently observed was recorded for periods that varied from 3 h to 1 week in different experiments. The temperature of the cell was set to 352 K, 363 K, 367 K, 369 K, or 375 K in the experiments with air and to 375 K for the N2 atmosphere. This temperature was controlled to better than ±0.1 K with a Eurotherm 2404 PID unit and measured with a precision of ±0.1 K by a Tecnisis 100 platinum resistance thermometer.
Table 1.
The results of the thermal stability studies on SV performed in sealed glass ampules and of the analysis of the final products by LC–MS/MS are summarized in Table 1, where m is the mass of sample, T is the temperature of the thermostatic oil bath, and t the duration of the experiment. In the case of runs 8 and 13, where a heterogeneous product consisting of a white powder and yellow glass was obtained, fractions of both phases were independently analyzed. The identification of species D1 and D2 mentioned in Table 1 is described below in the section on characterization of oxidative degradation products by LC–MS/MS. Observation of the materials inside the ampules at the end of the experiments indicated that: (1) when SV was kept at 373 K under a dry or water saturated nitrogen atmosphere (runs 3–5), no change of the sample (e.g., color, morphology) was noted, regardless of the duration of the experiment. This is illustrated in Figure 2a for run 5, with 26 h duration. The previous conclusion is corroborated by the results of LC–MS/MS analysis (Table 1), which showed no evidence of degradation products except in the case of run 4 where traces of D1 were detected. (2) No decomposition was also observed in runs 1 and 2, carried out at 373 and 423 K, respectively, under reduced pressure (13.3 Pa). However, in the case of run 2, where SV was heated above its fusion temperature, a glass phase was formed (Fig. 2b) on cooling to ambient temperature (293±2 K). (3) In the presence of oxygen (runs 6–14), different extents of decomposition were noted. When T = 343 K no color changes or degradation products were observed. For T = 353 K, the sample seemed unaltered but traces of D1 were found by LC– MS/MS. Above 353 K, decomposition was clearly evidenced by the formation of a yellow glass (Fig. 2c) and further confirmed by the detection of degradation products D1 and D2 in the LC–MS/MS analysis. The products of the thermal stability experiments summarized in Table 1 were also analyzed by DSC. The corresponding results are listed in Table 2, where m is the mass of sample, Tg represents the glass transition temperature, Ton and Tmax are the onset and maximum temperatures of the fusion peak,
Results of the Thermal Stability Studies Carried Out in Glass Ampulesa
Run number 1 2 3 4 5 6 7 8 9 10 11 12 13 14
243
m (mg)
T (K)
Atmosphere
t (h)
Product Appearance
LC–MS/MS Analysis
86.77 107.60; 125.57 82.61 81.53 369.12 70.79 50.58 63.60 209.05 124.39 115.57 132.97 78.03 384.26
373 423 373 373 373 373 373 373 343 353 363 373 373 373
Reduced pressure (13.3 Pa) Reduced pressure (13.3 Pa) N2 N2 +H2 O N2 O2 O2 +H2 O Air Air Air Air Air Air+H2 O Air
14 0.08 14 14 26 14 14 14 18 18 18 18 14 26
White powder Colorless glass White powder White powder White powder Yellow glass Yellow glass White powder + yellow glass White powder White powder Yellow glass Yellow glass White powder + yellow glass Yellow glass
No degradation products No degradation productsb No degradation products Traces of D1 No degradation products Formation of D1 and D2 Formation of D1 and D2 Formation of D1 and D2 No degradation products Traces of D1 Formation of D1 and D2 Formation of D1 and D2 Formation of D1 and D2 Formation of D1 and D2
a Air refers to atmospheric air with 57%–60% relative humidity and air+H2 O, N2 +H2 O, and O2 +H2 O to water saturated atmospheres. D1 and D2 refer to the products of thermal decomposition discussed in the section on characterization of the oxidative degradation products of simvastatin by LC–MS/MS (see below). b The colorless glasses obtained in the two runs were mixed and analyzed by LC–MS/MS.
DOI 10.1002/jps.23787
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
244
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 2. Images of the samples in the ampules at the end of (a) run 5, (b) run 2, and (c) run 6 (see Table 1).
Table 2. Results of the DSC Analysis on the Simvastatin Starting Material and the Products of Thermal Stability Studies in Glass Ampules (see also Table 1) Fusion Run Number Starting material 1 2a 3 4 5 6 7 8 (white fraction) 8 (yellow fraction) 9 10 11 12 13 (white fraction) 13 (yellow fraction) 14 a
m (mg) Tg (K) 0.839 0.422 4.864 4.362 0.767 0.276 0.388 0.982 0.530 0.783 1.391 1.031 1.298 1.396 0.502 0.555 0.742 0.589
Ton (K) 411.6 408.6
Tmax (K) fus ho (J/g) 413.3 410.8
67.90 64.76
302 303
302 304 309
303 295 318 320
406.6 408.7 408.8 345.0 367.1 407.8 361.9 406.2 404.6 360.5 346.3 406.7 380.3 354.3
409.6 410.4 410.4 370.1 372.8 410.4 386.8 409.9 408.9 374.4 350.8 411.0 394.8 374.1
52.95 65.46 61.23 39.20 8.36 60.38 25.40 42.19 42.63 23.12 13.30 45.73 27.49 11.93
Results corresponding to each duplicate of run 2 in Table 1.
respectively, and fus ho is the standard specific enthalpy of fusion. A comparison between the DSC measured curves of the starting material (scan rate of 10 K/min) and of the different crystalline and glassy products obtained after 14 h at 373 K under different atmospheres is given as Supporting Information. The formation of an amorphous phase and decomposition products upon thermal treatment of SV in the presence of oxygen is reflected in the DSC results by (1) a notable decrease of the onset temperature (31–67 K) and enthalpy of fusion (29–60 J/g) relative to the starting material, when a yellow glass is obtained (Table 2); (2) the observation of a glass transition in the range 295–320 K for those samples. The fact that a 3–14 J/g decrease in fus ho is noted, even when no degradation products were found (runs 1, 3, and 5), also suggests that the thermal treatment may consistently lead to some degree of amorphization of the sample. The amorphous nature of the yellow glass products is further evidenced by a notable broadening of the OH stretching band ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
in the DRIFT spectra (Fig. 3). This band is located at 3552 cm−1 in the spectrum of the starting material. Finally, the onset of thermal decomposition when the samples are subjected to increasing temperatures under air can be noted at T > 353 K in the DRIFT and DSC results shown in Figure 4. Nevertheless, these two analyzes do not seem to be sensitive to the traces of D1 detected by LC–MS/MS at 353 K (Table 1). Thermal Stability Studies by DSC and Calvet-Drop Microcalorimetry The thermal stability of SV under N2 and air atmospheres was also studied in non-isothermal and isothermal conditions by using DSC and Calvet-drop microcalorimetry, respectively. DSC experiments carried out in air (non-sealed aluminum crucibles) at a rate of 2 K/min showed an exothermic event with onset at Ton = 397±3 K (see Supporting Information). Here, the indicated uncertainty corresponds to the standard error of the mean of four independent determinations. This thermal event had been previously observed by DSC and assigned to oxidative decomposition of SV, albeit without identification of the decomposition products.20 Similar conclusions were drawn from the results of a series of isothermal Calvet-drop microcalorimetry experiments with 18 h duration. The corresponding measured curves are given as Supporting Information. No thermal event was observed when N2 atmosphere was present inside the calorimetric cell and the temperature was set to 375 K. This was also true for an experiment with approximately 1 week duration. In contrast, for temperatures higher than 367 K and under air, an exothermic peak was observed, which can be assigned to the oxidative degradation process described in the previous section. The fact that the peak becomes broader and the induction time for the onset of decomposition increases as the temperature decreases is compatible with the expected decrease of the reaction rate as the temperature decreases. The decomposition of SV above 367 K was evidenced by the glassy and yellowish aspect of the material inside the capillary removed from the calorimetric cell at the end of each run. This observation corroborates the conclusions of the studies made in glass ampules (Table 1). Although no peak could be noted in the measuring curve corresponding to 363 K, partial decomposition of the sample was, nevertheless, suggested by the yellowish color of the material obtained at the end of the experiment. DOI 10.1002/jps.23787
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
245
Figure 3. DRIFT spectra of the original simvastatin sample and of the materials obtained after thermal treatment (14 h, 373 K) under different atmospheres. The result for run 2 refers to the mixture of the two colorless glass products obtained in the two independent experiments. All spectra were normalized relative to the most intense peak.
It can therefore be concluded that the observations of both the DSC and Calvet-drop microcalorimetry measurements are consistent with the results of the studies carried out in glass ampules (Table 1). Characterization of the Oxidative Degradation Products of SV by LC–MS/MS Liquid chromatography–ESI/MS chromatograms of a control sample of simvastatin showed only a major peak corresponding to SV, at retention time of 12.5 min, along with minor signals related with the main impurities previously identified in the sample by LC–MS/MS (see Supporting Information for details).18 The same LC–MS method was applied to the analysis of the products of the thermal stability studies summarized in Table 1. As indicated in Table 1, no significant degradation was observed when the starting material was subjected to thermal treatment under nitrogen atmosphere. In the presence of pure or atmospheric oxygen, two main oxidative degradation DOI 10.1002/jps.23787
products with m/z 433 (peaks D1 and D1 ) and m/z 435 (peaks D2 and D2 ) were detected by the LC–MS at retention times between 4.5 and 6.8 min. The presence of two peaks with the same m/z value and close retention times is indicative of isomeric degradation products. This is illustrated in Figure 5 for the product of run 6. The protonated molecules corresponding to these signals showed a mass increase of 14 and 16 Da relative to SV. The UV spectrum of D2 exhibited, as for SV, an absorbance maximum at 241 nm characteristic of cyclodienes, whereas that of D1 presented an absorbance maximum at 290 nm (see Supporting Information). This shift toward a higher wavelength suggests an increase of conjugation in the cyclodiene chromophore structure of D1 relative to the parent molecule. The identification of the oxidative degradation products D1 and D2 was achieved by (1) structural characterization based on a comparison of the MS/MS fragmentation patterns of a protonated SV molecule (previously reported27 and further confirmed in ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
246
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 4. DCS measured curves (a) and DRIFT spectra (b) of the original simvastatin sample and of the materials obtained after 18 h under air at different temperatures. The heat flow in the DSC curves was scaled by mass (J /g) and all DRIFT spectra were normalized relative to the most intense peak.
this work) and of the oxidative degradation products; (2) accurate mass measurements of the precursor ions using FT–ICR– MS, which led to the following results (M = measured monoisotopic mass of precursor ion; C = calculated monoisotopic mass of precursor ion; = deviation in ppm; F = deduced elemental formula of protonated molecule): (1) SV, M = 419.2795, C = 419.2797, = 0.48, F = C35 H39 O5 ; (2) product D1, M = 433.2589, C = 433.2590, = 0.23, F = C35 H37 O6 ; (3) product D2, M = 435.2742, C = 435.2747, = 1.15, F = C35 H39 O6 . The tandem mass spectra of the protonated molecules of D1 (m/z 433), D2 (m/z 435), and SV (m/z 419) are given as Supporting Information (Figure S7). The MS2 spectra of D1 and D2 follow a similar dissociation pattern (data not shown). The general fragmentation patterns of D1 and D2 ions are the same as those for SV. The main fragment ions at m/z 303, 285, and 225 for SV increased by 14 and 16 Da to become the product ions at m/z 317, 299, and 239 for D1, and m/z 319, 301, and 241 for D2, respectively. These results clearly indicated that the degradation products were formed through oxidation of the hexahydronaphthalene portion of SV. This assumption is also supported by the presence of two small peaks at m/z 173 (175) and 143 (145) in the tandem mass spectrum of D1 (D2), which correspond to the species indicated in Figures S8 and S9 of the Supporting Information. It is accepted that SV is susceptible to oxidation at the conjugated diene presumably through a mechanism involving radical initiation, propagation, and termination steps. The main pathway is based on the formation of a resonance-stabilized free radical (FR1 or FR2) whose resonance forms may account for the oxidation at different positions. Oxidation at a tertiary position will yield a stable alcohol (D2a and D2d), whereas oxidation at an allylic position will yield an alcohol (D2b and D2c) prone to further oxidation to a dienone (D1b and D1c), with a higher conjugated structure. A proposed mechanism is given as Supporting Information ˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
(Figure S6). The formation of the species D2b and D1b is reported on earlier studies of autoxidation of SV in solution.28 Schemes of the fragmentation mechanism for the oxidative degradation products D1 and D2 are given as Supporting Information.
CONCLUSIONS The experiments carried out in this work indicated that, provided SV is handled under N2 or reduced pressure atmospheres (13.3 Pa) at a temperature ≤373 K, no thermal decomposition is to be expected, at least within 1 week. Significant thermal degradation does, however, occur at temperatures ≥353 K in the presence of pure or atmospheric oxygen. The two main oxidative degradation products obtained in this case result from the oxidation of the hexahydronaphthalene fragment of SV. It is therefore hoped that the results here reported will provide useful guidelines for the development of SV solid formulations that require thermal treatment.
ACKNOWLEDGMENTS ˜ para a Ciˆencia e a This work was supported by Fundac¸ao Tecnologia (FCT), Portugal (Projects PTDC/QUI-QUI/098216/ 2008, PEst-OE/QUI/UI0612/2013, PEst-OE/QUI/UI0100/2013, REDE/1501/REM/2005, and REDE/1502/REM/2005). PhD (SFRH/BD/48410/2008) and Post-Doctoral (SFRH/BPD/43346/ 2008) grants from FCT are also gratefully acknowledged by R. G. Sim˜oes and C. E. S. Bernardes, respectively. We also thank Prof. Matilde Marques (CQE-IST) for helpful discussions about the oxidation mechanism of SV. DOI 10.1002/jps.23787
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
247
Figure 5. LC–DAD–ESI/MS chromatograms of the material obtained in run 6 (see Table 1): (a) LC-DAD profile at 8 = 238 nm, (b) total ion chromatogram and mass spectrum of the protonated molecule of simvastatin (SV), (c) extracted ion chromatogram and mass spectrum of m/z 433 ion, (d) extracted ion chromatogram and mass spectrum of m/z 435 ion.
REFERENCES 1. Li JJ. 2009. Triumph of the heart. The story of statins. New York: Oxford University Press. 2. Tiwari R, Pathak K. 2011. Statins therapy: A review on conventional and novel formulation approaches. J Pharm Pharmacol 63:983– 998. 3. Taylor F, Ward K, Moore THM, Burke M, Davey Smith G, Casas JP, Ebrahim S, Eds. 2012. Statins for the primary prevention of cardiovascular disease. New York: John Wiley. 4. Vance DE, Van den Bosch H. 2000. Cholesterol in the year 2000. Biochem Biophys Acta 1529:1–8.
DOI 10.1002/jps.23787
5. Roger VL, Go AS, Lloyd-Jones DM, Benjamin EJ, Berry JD, Borden WB, Bravata DM, Dai S, Ford ES, Fox CS, Fullerton HJ, Gillespie C, Hailpern SM, Heit JA, Howard VJ, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Makuc DM, Marcus GM, Marelli A, Matchar DB, Moy CS, Mozaffarian D, Mussolino ME, Nichol G, Paynter NP, Soliman EZ, Sorlie PD, Sotoodehnia N, Turan TN, Virani SS, Wong ND, Woo D, Turner MB, Comm AHAS, Subcomm SS. 2012. Heart disease and stroke statistics—2012 update. A report from the American Heart Association. Circulation 125:E2–E220. 6. Baxendale IR, Hayward JJ, Ley SV, Tranmer GK. 2007. Pharmaceutical strategy and innovation: An academics perspective. ChemMedChem 2:768–788.
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
248
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
7. Hoffman WF, Alberts AW, Anderson PS, Chen JS, Smith RL, Willard AK. 1986. 3-Hydroxy-3-methylglutaryl-coenzyme a reductase inhibitors. 4. Side-chain ester derivatives of mevinolin. J Med Chem 29:849–852. 8. Askin D, Verhoeven TR, Liu TMH, Shinkai I. 1991. Synthesis of Synvinolin—Extremely high conversion alkylation of an ester enolate. J Org Chem 56:4929–4932. 9. Barreiro EJ, Kummerle AE, Fraga CAM. 2011. The methylation effect in medicinal chemistry. Chem Rev 111:5215–5246. ˜ MM, Silva MCD, Soledade LEB, Souza AG. 10. Souza MAF, Conceic¸ao 2007. Thermal and kinetic study of statins. J Therm Anal Calorim 87:859–863. 11. Graeser KA, Strachan CJ, Patterson JE, Gordon KC, Rades T. 2008. Physicochemical properties and stability of two differently prepared amorphous forms of simvastatin. Cryst Growth Des 8:128–135. 12. Ambike AA, Mahadik KR, Paradkar A. 2005. Spray-dried amorphous solid dispersions of simvastatin, a low Tg drug: In vitro and in vivo evaluations. Pharm Res Dordr 22:990–998. 13. Graeser KA, Patterson JE, Zeitler JA, Gordon KC, Rades T. 2009. Correlating thermodynamic and kinetic parameters with amorphous stability. Eur J Pharm Sci 37:492–498. 14. Graeser KA, Patterson JE, Rades T. 2009. Applying thermodynamic and kinetic parameters to predict the physical stability of two differently prepared amorphous forms of simvastatin. Curr Drug Deliv 6:374–382. 15. Ambike AA, Mahadik KR, Paradkar A. 2005. Physico-chemical characterization and stability study of glassy simvastatin. Drug Dev Ind Pharm 31:895–899. 16. Jun SW, Kim MS, Kim JS, Park HJ, Lee S, Woo JS, Hwang SJ. 2007. Preparation and characterization of simvastatin/hydroxypropyl-betacyclodextrin inclusion complex using supercritical antisolvent (SAS) process. Eur J Pharm Biopharm 66:413–421. 17. Oliveira MA, Yoshida MI, Gomes ECL, Mussel WN, Vianna-Soares ´ ˜ CD, Pianetti GA. 2010. Analise t´ermica aplicada a` caracterizac¸ao
˜ et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:241–248, 2014 Simoes
da sinvastatina em formulac¸o˜ es farmacˆeuticas. Quim Nova 8:653– 1657. 18. Sim˜oes RG, Bernardes CES, Diogo HP, Agapito F, Minas da Piedade ME. 2013. Energetics and structure of simvastatin. Mol Pharm 10:2713–2722. 19. Nti-Gyabaah J, Chan V, Chiew YC. 2009. Solubility and limiting activity coefficient of simvastatin in different organic solvents. Fluid Phase Equilibr 280:35–41. 20. Ellison DK, Moore WD, Petts CR. 1993. Simvastatin. In Analytical profiles of drug substances and excipients; Brittain HG, Ed. San Diego: Academic Press. 21. Murtaza G. 2012. Solubility enhancement of simvastatin: A review. Acta Pol Pharm 69:581–590. 22. Hovorka SW, Sch¨oneich C. 2001. Oxidative degradation of pharmaceuticals: Theory, mechanisms and inhibition. J Pharm Sci 90:253–269. 23. Proc´opio JVV, Souza VG, Costa RA, Correia LP, Souza FS, Macˆedo RO. 2011. Application of thermal analysis and pyrolysis coupled to GC/MS in the qualification of simvastatin pharmaceutical raw material. J Therm Anal Calorim 106:665–670. 24. Craig DQM, Reading M. 2006. Thermal analysis of pharmaceuticals. Boca Raton: CRC Press. 25. Kiyobayashi T, Minas da Piedade ME. 2001. The standard molar enthalpy of sublimation of 05 -bis-pentamethylcyclopentadienyl iron measured with an electrically calibrated vacuum-drop sublimation microcalorimetric apparatus. J Chem Thermodyn 33:11–21. 26. Bernardes CES, Santos LMNBF, Minas da Piedade ME. 2006. A new calorimetric system to measure heat capacities of solids by the drop method. Meas Sci Technol 17:1405–1408. 27. Wang H, Wu YH, Zhao ZX. 2001. Fragmentation study of simvastatin and lovastatin using electrospray ionization tandem mass spectrometry. J Mass Spectrom 36:58–70. 28. Smith GB, Dimichele L, Colwell LF, Dezeny GC, Douglas AW, Reamer RA, Verhoeven TR. 1993. Autooxidation of simvastatin. Tetrahedron 49:4447–4462.
DOI 10.1002/jps.23787
