AAPS PharmSciTech, Vol. 11, No. 2, June 2010 ( # 2010)
`DOI: 10.1208/s12249-010-9434-8
`
`Research Article
`
`Influence of Formulation and Processing Factors on Stability of Levothyroxine
`Sodium Pentahydrate
`
`Jarrod W. Collier,1,2 Rakhi B. Shah,1 Abhay Gupta,1 Vilayat Sayeed,3
`Muhammad J. Habib,2 and Mansoor A. Khan1,4
`
`Received 25 August 2009; accepted 9 April 2010; published online 8 May 2010
`Abstract. Stability of formulations over shelf-life is critical for having a quality product. Choice of
`excipients, manufacturing process, storage conditions, and packaging can either mitigate or enhance the
`degradation of the active pharmaceutical
`ingredient (API), affecting potency and/or stability. The
`purpose was to investigate the influence of processing and formulation factors on stability of
`levothyroxine (API). The API was stored at long-term (25°C/60%RH), accelerated (40°C/75%RH),
`and low-humidity (25°C/0%RH and 40°C/0%RH) conditions for 28 days. Effect of moisture loss was
`evaluated by drying it (room temperature, N2) and placed at 25°C/0%RH and 40°C/0%RH. The API
`was incubated with various excipients (based on package insert of marketed tablets) in either 1:1, 1:10, or
`1:100 ratios with 5% moisture at 60°C. Commonly used ratios for excipients were used. The equilibrium
`sorption data was collected on the API and excipients. The API was stable in solid state for the study
`duration under all conditions for both forms (potency between 90% and 110%). Excipients effect on
`stability varied and crospovidone, povidone, and sodium laurel sulfate (SLS) caused significant API
`degradation where deiodination and deamination occurred. Moisture sorption values were different
`across excipients. Crospovidone and povidone were hygroscopic whereas SLS showed deliquescence at
`high RH. The transient formulation procedures where temperature might go up or humidity might go
`down would not have major impact on the API stability. Excipients influence stability and if possible,
`those three should either be avoided or used in minimum quantity which could provide more stable tablet
`formulations with minimum potency loss throughout its shelf-life.
`
`KEY WORDS: excipients; formulation; levothyroxine sodium pentahydrate; moisture sorption; stability.
`
`INTRODUCTION
`
`Levothyroxine sodium pentahydrate is the sodium salt of
`the levo-isomer of thyroxine, an active physiological sub-
`stance found in the thyroid gland. Synthetic version is
`primarily used in the treatment of hypothyroidism and as a
`thyroid stimulating hormone suppressant, in the treatment of
`various types of euthroid goiters (1). Since its first inception
`in the market, there have been various recalls for these drug
`products from various manufacturers. The primary reason for
`these recalls is due to sub-potency before the expiration date
`of the drug product because of stability failures (2).
`
`1 Division of Product Quality Research, Office of Testing and
`Research, Office of Pharmaceutical Sciences, Center for Drug
`Evaluation and Research, Food and Drug Administration, 10903
`New Hampshire Ave, Life Sciences Bldg 64, Silver Spring, Maryland
`20993-0002, USA.
`2 Department of Pharmaceutical Sciences, School of Pharmacy,
`Howard University, Washington, District of Columbia, USA.
`3 Division of Chemistry III, Office of Generic Drugs, Office of
`Pharmaceutical Sciences, Center for Drug Evaluation and Research,
`Food and Drug Administration, Silver Spring, Maryland, USA.
`4 To whom correspondence should be addressed. (e-mail: Mansoor.
`khan@fda.hhs.gov)
`
`Although expiration dating is based on the scientific data
`at normal and stressed conditions, not all the batches and
`strengths undergo stability testing and for this reason under-
`standing the factors that affects the product stability is critical.
`Levothyroxine has been a subject of advisory committee
`meetings at FDA due to its potency and stability issues (3).
`Levothyroxine has a complex stability profile and is sensitive
`to various environmental
`factors such as light, air, and
`humidity, among others (4–8). Effects of some excipients
`have been studied by Patel et al. (8) and Gupta et al. (9).
`However, in both the reports, the impurities remain largely
`unidentified. Thus stability of levothyroxine has not been
`systematically characterized in the presence of processing and
`formulation factors by a stability-indicating method which can
`identify majority of its degradation products.
`Hydrate formation and the dehydration of the product
`may occur during the manufacturing or storage of pharma-
`ceuticals and this phenomenon of hydration and dehydration is
`known to be is very sensitive and variable (10). Since biological
`effects of hydrates sometimes differ from that of anhydrates, it
`is important that these behavioral differences be known for the
`development of stable pharmaceutical formulations. As the
`phase transition that occurs on hydration or dehydration is
`accompanied by a change in the physicochemical properties, it
`is important to understand the mechanisms of these transitions
`
`1530-9932/10/0200-0818/0 # 2010 American Association of Pharmaceutical Scientists
`
`818
`
`Mylan Ex 1009, Page 1
`
`

`

`Factors on Stability of Levothyroxine Sodium Pentahydrate
`
`819
`
`under various conditions (11–13). Currently, there are no
`studies that investigate the stability of levothyroxine solid state
`under hydrous and anhydrous conditions.
`The compatibility of drug and its excipients is extremely
`critical in the formulation of a quality drug product. The
`formulation of a stable and effective dosage form requires
`careful selection of excipients used to facilitate administra-
`tion, promote consistent release and bioavailability of the
`drug, and to promote the active moiety from the environ-
`ment. Although often regarded as ‘inert’, excipients can in
`fact readily interact with drugs (14). Lack of compatibility
`between levothyroxine and its excipients can lead to stability
`problems and inadequate bioavailability/bioequivalence
`issues that are critical for patients in need of treatment for
`hypothyroidism. It is important to know that the product
`development phase of a drug and its excipients is just as
`significant as the stability properties of the finished product.
`Environmental conditions of 40°C/75%RH are typically used
`for 3–6 months as accelerated conditions. This is a lengthy
`process when trying to evaluate the stability of formulations
`in a more efficient and timely manner. Another method that
`is commonly employed for evaluating the drug–excipient
`compatibility is isothermal stress testing of binary drug–
`excipient mixtures. The method involves storing the drug–
`excipient blends with or without moisture at high temperature
`and determining the drug content (15–17).
`The current research was aimed to fill the knowledge
`gaps which still exist in understanding the stability of this
`active pharmaceutical ingredient (API) by conducting stabil-
`ity studies by a stability-indicating method with degradation
`impurities under hydrous and anhydrous conditions. Also, the
`effect of commonly used formulation excipients in the
`presence of moisture was studied to understand the stability
`profile of the drug substance.
`
`MATERIALS AND METHODS
`
`Materials
`
`L-Thyroxine sodium (L-T4) was obtained from KV Pharma-
`ceutical (St Louis, MO, USA). 3,3′,5-tri-iodo-L-thyronine; 3,5-
`diiodo-L-thyronine; 3,5-diiodo-L-tyrosine; 3-iodo-L-tyrosine;
`L-thyronine; L-tyrosine; 3,3′,5-tri-iodo-L-thyroacetic acid;
`3,3′,5,5′-tetra-iodo-L-thyroacetic acid, corn starch (CS), acacia,
`sucrose, magnesium stearate (MS), mannitol (M), and sodium
`laurel sulfate (SLS) were purchased from Sigma-Aldrich [St.
`Louis, MO, USA]. Microcrystalline cellulose (MCC) and
`croscarmellose sodium (CCS) were obtained from FMC
`Biopolymer [Philadelphia, PA, USA]. Lactose monohydrate
`(LM) [Kerry BioScience, Chicago, IL, USA], povidone (P)
`[BASF, Florham Park, NJ, USA], talc [Spectrum Chemicals,
`Gardena, CA, USA], crospovidone (CP) [ISP Technologies
`Inc, Wayne, NJ, USA], sodium starch glycolate (SSG)
`[Explotab, Patterson, NY, USA], colloidal silicon dioxide
`(CSD) [Aerosil, Evonik Degussa, Orange, CA, USA], and
`confectioner’s sugar (CFS) [Domino’s sugar, Baltimore, MD,
`USA] were used as received. Methanol, acetonitrile, 0.01 M
`NaOH, 0.1% trifluoroacetic acid (TFA),
`theophylline
`reagents, and Fisherbrand low adhesion specialty tips (21-
`381-83) were purchased from Fisher Scientific (Suwanee,
`
`GA, USA). For all studies, distilled and deionized water
`was used.
`
`Stability of Levothyroxine Sodium
`
`Stability of Pentahydrate Form
`
`The pure drug was weighed and placed in open amber
`glass containers to equilibrate with temperature and humidity
`conditions. The stability conditions were as follows: 25°C/0%
`RH and 40°C/0%RH (Fisher Scientific, St. Louis, MO, USA)
`and 25°C/60%RH and 40°C/75%RH (HotPak, Baltimore,
`MD, USA). For conditions of 25°C/0%RH and 40°C/0%RH,
`the samples were placed in desiccators using indicating
`Drierite anhydrous calcium sulfate dessicants and positioned
`under the appropriate temperature condition. Humidity
`values were measured using a hygrometer (Fisher Sci,
`Suwanee, GA, USA).
`To determine the chemical changes of levothyroxine
`sodium hydrate, the samples were analyzed at predetermined
`days (0, 3, 6, 10, 14, and 28). Samples were weighed (10 mg)
`in three 100-mL amber glass volumetric flasks followed by the
`addition of 20 mL of sample diluent (10 mM NaOH–MeOH;
`1:1 v/v), sonicated for 10 min, and filled to volume with
`sample diluent. One milliliter of each sample was then
`transferred to individual 10-mL amber volumetric flasks.
`One milliliter of theophylline internal standard (0.1 mg/mL)
`was then added and solution filled to volume with sample
`diluent. The contents of each solution were transferred to an
`automatic injector for HPLC analysis. The determination of
`the API was carried out by a previously determined validated
`method (18).
`
`Stability of Dehydrated Form
`
`The API was weighed (50 mg) on a calibrated balance
`and placed in eight 1.5-mL amber glass septum vials for each
`stability condition (25°C/0%RH and 40°C/0%RH). Samples
`were then placed in vacuum/gas compartments containing
`Fisher Scientific thermo-hygro readers (11-661-13, Suwanee,
`GA, USA) and color-indicating desiccants. A vacuum was
`used to remove the air from each compartment followed by a
`purge of nitrogen in each compartment. Compartments were
`then closed and placed in their temperature chambers,
`respectively.
`The samples were analyzed at predetermined days (0, 3,
`6, 10, 14, and 28). For each temperature condition, a vial is
`removed from the vacuum/gas chamber and immediately
`purged with a low flow of nitrogen to assure the prevention of
`moisture entering the vials. Vial is then sealed with septum
`vial screw caps containing Tuf-Bond disks for secured closure
`and placed in a desiccator. The desiccator is taken to a
`compact glove box (Plaslabs, Lansing, MI, USA) where
`nitrogen is slowly being purged in the closed atmosphere. It
`is here that the weighing is performed for HPLC, differential
`scanning calorimetry (DSC), and thermogravimetric analysis
`(TGA) studies. Samples are then placed back in desiccator
`and removed from compact glove box for their respective
`analyses.
`
`Mylan Ex 1009, Page 2
`
`

`

`820
`
`Drug–Excipient Mixtures Studies
`
`The API was weighed on a calibrated balance and was
`then transferred to 15-mL glass culture test tubes. Excipients
`were weighed in either 1:1, 1:10, or 1:100 w/w ratios to the
`drug, ideally depending upon the normal strength in which
`they are used in the formulations. Deionized water was added
`to each test tube to obtain a 5% moisture content and
`vortexed (20 s) for a more homogenous mixture of compo-
`nents. Samples were then closed using screw top closures and
`transferred to a stability chamber (Fisher Scientific, St. Louis,
`MO, USA) set at an accelerated test condition of 60°C±1°C
`(drug–excipient ratios are included in Table I). It is known
`that levothyroxine degrades rapidly at higher temperature
`(5). In present study, the objective was to screen various
`formulation excipients for their effects on stability of levo-
`thyroxine. If the tight conditions were not maintained,
`it
`would be challenging to understand whether the degradation
`was caused by the excipient or due to slightly elevated
`temperature. Therefore, a very tight range was used to
`perform the study.
`The samples were analyzed at predetermined times for a
`total of 28 (0, 1, 2, 5, 9, 13, and 28) days to observe the
`chemical change. Those results were compared with samples
`collected at time zero. For sample analysis, the contents were
`washed into individual 100-mL amber volumetric flasks,
`sonicated for 10 min, and filled to volume with sample diluent
`(10 mM NaOH–MeOH; 1:1 v/v). One milliliter of each
`sample was then transferred to individual 10-mL amber
`volumetric flasks. One milliliter of
`theophylline internal
`standard (0.1 mg/mL) was then added and solution filled to
`volume with sample diluent. The contents of each solution
`were transferred to an automatic injector for HPLC analysis.
`The determination of the API was determined by a pre-
`viously determined validated method.
`
`HPLC System
`
`A validated HPLC method was used to analyze the
`samples (18). The HP 1100 HPLC equipment from Agilent
`
`Collier et al.
`
`(Wilmington, DE, USA) consisted of a quaternary pump, an
`automatic injector, a diode array detector, and a column
`oven. Data was collected using Agilent ChemStation soft-
`ware. Separation was achieved with a reverse phase Inertsil
`ODS 2 column (250×4.6 mm, 5 µm, 150 A) with an Inertsil
`ODS Security Guard cartridge (4.0×3.0 mm, 10 µm). It
`provided baseline separation with gradient conditions with
`0.1% TFA (A) and acetonitrile (B) from 92% to 8% A in
`25 min, at 8% A from 25 to 30 min, from 8% to 92% A from
`30 to 35 min run time of 40 min in a single chromatographic
`run. The flow rate was 0.8 mL/min, column temperature was
`25°C, and the injection volume was 50 µL. The UV detection
`wavelength was set at 215, 223, 228, 232, and 240 nm.
`However, all calculations were performed at 223 nm.
`
`Differential Scanning Calorimetry
`
`Approximately 1–2 mg of sample was weighed and
`placed in an aluminum pan (no seal) with an empty open
`aluminum pan as a reference. The sample was scanned at a
`heating rate of 10°C/min from 25–300°C (TA Instruments
`Model 2920). The cell was purged with nitrogen during the
`run.
`
`Thermogravimetric Analysis
`
`Approximately 5–10 mg of sample was heated in an open
`platinum pan from 25–300°C (TA Instruments Model 2950) at
`10°C/min. The cell was purged with nitrogen during the run.
`
`Equilibrium Moisture Content
`
`Equilibrium moisture content of L-T4 and individual
`excipients listed in Table I at different relative humidity (RH)
`levels was determined gravimetrically at 25°C on a Sym-
`metrical Gravimetric Analyzer (Model SGA-100, VTI Cor-
`poration, Hialeah, FL, USA). Sorption profiles were collected
`by increasing the instrument RH from 5% to 95%, while
`desorption profiles were collected by decreasing the RH from
`95% to 5% for the same sample. At each RH level, the
`
`Excipient
`
`Abbreviations
`
`Drug–excipient ratio
`
`Amount of water (mL)
`
`Table I. Drug–Excipient Ratios
`
`Colloidal silicon dioxide
`Crospovidone
`Magnesium stearate
`Mannitol
`Microcrystalline cellulose
`Povidone
`Sodium lauryl sulfate
`Sucrose
`Acacia
`Confectioner’s sugar
`Lactose monohydrate
`Talc
`Croscarmellose sodium
`Corn starch
`Sodium starch glycolate
`
`CSD
`CP
`MS
`M
`MCC
`P
`SLS
`S
`A
`CFS
`LM
`T
`CCS
`CS
`SSG
`
`1:1
`1:10
`1:10
`1:10
`1:10
`1:1
`1:1
`1:10
`1:1
`1:10
`1:100
`1:10
`1:1
`1:10
`1:1
`
`1
`5.5
`5.5
`5.5
`5.5
`1
`1
`5.5
`1
`5.5
`50.5
`5.5
`1
`5.5
`1
`
`Mylan Ex 1009, Page 3
`
`

`

`Factors on Stability of Levothyroxine Sodium Pentahydrate
`
`821
`
`sample was allowed to reach moisture equilibrium before the
`RH was changed to the next level. The equilibrium criteria
`was based on the hygroscopicity of the sample but was always
`tighter than <0.05% weight change in 20 min. The sample size
`also varied between 3 and 30 mg for the same reason. All
`samples were dried in the instrument at 40/60°C for 1 h prior
`to water sorption experiment, except L-T4 and MS which
`were dried at 25°C since these two materials undergo
`physical/chemical changes at higher temperature.
`
`RESULTS AND DISCUSSION
`
`Influence of Processing Factors on Stability of Levothyroxine
`Sodium Pentahydrate
`
`Levothyroxine sodium is available in pentahydrate form
`and is used in the tablet dosage form. Influence of processing
`factors on moisture level has not yet been determined
`systematically to determine degradation products by a
`stability indicating method. Therefore, it was investigated at
`various conditions in the current study. The drug substance
`was placed in stability chamber under various conditions as
`described in the “MATERIALS AND METHODS” section.
`In addition to long-term and accelerated stability conditions,
`samples were also exposed at 0% RH at 25°C and 40°C.
`Levothyroxine sodium pentahydrate loses water molecules
`at those temperatures when the RH is below 30% (Fig. 1).
`This is consistent with the report
`in the literature (8).
`Whether loss of moisture is related to destabilization of the
`crystal lattice and thus affecting its potency is not yet known
`and therefore, in the present work, there was an effort to
`determine the effect of moisture loss on the stability of the
`drug.
`Potency of levothyroxine was determined at various time
`points under various stability conditions and the results are
`listed in Table II. It was seen that the pentahydrate form was
`stable in all the conditions for the duration of the study. The
`moisture content was analyzed by TGA and the moisture loss
`(%) data for the samples are presented in Table III. There
`
`Fig. 1. Moisture sorption isotherm of levothyroxine sodium pentahydrate
`
`was no moisture loss when the samples were left at higher
`humidity values; however, there was about 3% moisture loss
`when the humidity was low. DSC thermogram for the drug
`substance at 0 time point is shown in Fig. 2. The DSC
`thermogram showed that levothyroxine sodium pentahydrate
`showed two exotherms (at 120°C and 159°C) and an
`endotherm at 209°C. The two exotherms might correspond
`to the moisture loss from the hydrate form. The TGA
`profile showed the weight
`loss of 10% initially which
`corresponds to the total amount of moisture including
`hydrated water amount
`in the levothyroxine sodium
`pentahydrate. The moisture content was in close agreement
`with the value obtained from Karl-Fisher analysis which
`was about 9.8%.
`The study was also conducted by placing the pentahy-
`drate form into the anaerobic chamber where all the air was
`removed and the chamber was flushed with nitrogen. The
`samples were then removed from the desiccator, weighed in
`the chamber respectively for both TGA and DSC, and
`transferred back to a desiccator. Samples were then taken
`out of chamber for moisture content determination. This was
`used as a starting material as the dehydrated form and the
`samples were kept in low-humidity conditions at two temper-
`atures (25°C and 40°C). The data for potency and moisture
`loss are shown in Tables II and III, respectively. It was found
`that there was a significant moisture loss for these samples at
`the end of study duration. For dehydrated form, only one
`sample was evaluated for moisture content at various time
`points. It was mainly done to see whether the moisture
`content changed at various time points. The data is highly
`variable but is mainly due to sample preparation techniques.
`As soon as dehydrated form comes in contact with the
`moisture, it absorbs moisture. Care was taken not to expose
`material to environment before analyzing them by TGA;
`however, it was beyond control once the sample is on the
`TGA pan and on the instrument when the thermostat will
`close the system. The data although variable, gave a
`qualitative estimate that the moisture levels were very low
`throughout the study period. However, the potency was not
`affected, thus confirming that removal of moisture from the
`molecule does not give rise to an unstable preparation.
`Moreover, the HPLC chromatograms did not indicated the
`presence of any impurity. Many of
`the manufacturing
`procedures such as drying, milling, and compression oper-
`ations might give rise to low-humidity conditions or increase
`in the temperature. If the duration of exposure is short
`enough it might not cause any degradation of the drug
`substance, which was shown by the current results.
`
`Influence of Formulation Factors on Stability of Levothyroxine
`Sodium Pentahydrate
`
`Using accelerated conditions gives the opportunity to
`conduct stability studies in a shorter period of time. Com-
`monly, conditions such as 40°C/75%RH are used as accel-
`erated and last between 3 and 6 months. Increasing the
`temperature to 60°C places L-thyroxine in its threshold
`temperature which causes rapid degradation. Screening of
`excipients was performed at a higher temperature of 60°C as
`
`Mylan Ex 1009, Page 4
`
`

`

`822
`
`Collier et al.
`
`Table II. Stability of Levothyroxine Sodium Pentahydrate and Dehydrated Form Under Various Stability Conditions
`
`Conditions
`
`0
`
`3
`
`Pentahydrate form
`25°C/60%RH
`25°C/0%RH
`40°C/75%RH
`40°C/0%RH
`Dehydrated form
`25°C/0%RH
`40°C/0%RH
`
`102.88±0.77
`105.04±0.75
`102.88±0.77
`102.88±0.77
`
`92.32±2.03
`92.32±2.03
`
`99.75±5.39
`100.90±0.54
`101.77±1.54
`102.80±3.02
`
`99.00±0.99
`99.85±4.55
`
`The results are expressed as average±SD for n=3
`
`Days
`
`6
`
`97.42±5.81
`103.52±2.68
`101.39±1.01
`98.97±0.64
`
`98.11±4.58
`91.30±2.47
`
`10
`
`14
`
`28
`
`101.59±0.39
`100.91±0.36
`99.23±2.17
`97.67±3.41
`
`104.59±6.81
`96.21±3.95
`
`101.23±1.23
`102.80±0.83
`102.76±1.27
`101.43±0.38
`
`96.82±1.42
`95.36±2.20
`
`101.48±1.00
`95.13±0.84
`100.68±2.03
`99.44±1.09
`
`96.20±0.80
`95.43±2.40
`
`opposed to accelerated stability studies which are usually
`performed at 40°C for 6 months. The higher temperature
`would accelerate the degradation of levothyroxine as it was
`observed in another study (5) that levothyroxine degrades
`rapidly at higher temperatures. Thus, a short term of 28 days
`was used in the study in order to determine the excipients
`which caused instability of levothyroxine. The combination of
`drug and excipient in this temperature environment allows for
`expedited drug degradation studies. Moreover, the mecha-
`nism of degradation does not change at higher temperatures
`(5). Another purpose of conducting this study was to
`evaluate manufacturing conditions such as wet granulation
`followed by drying which is done at higher temperatures,
`although it did not last for more than a few hours. Thus,
`conducting a short-term stability study at higher temperature
`would give an evidence of instability during manufacturing
`conditions.
`isothermal stress testing approaches to
`Conventional
`drug–excipient compatibility evaluation typically involve
`challenging binary drug–excipient mixtures (in realistic ratios)
`with moisture as the majority of drug degradation reactions
`involve moisture (19). Increasing the moisture content may
`be accomplished by increasing the relative humidity of the
`environment, or by adding a fixed amount of water. With
`
`exposure of the samples to high humidities, the excipient–
`drug interaction depends upon the free moisture present and
`relative hygroscopicities (16). Drug degradation thus may
`vary depending on the hygroscopicity of
`the excipients
`(16,20). Therefore, it is suggested that a constant amount of
`water be added to facilitate interactions between the exci-
`pient and drug, and to surround undissolved particles with an
`aqueous layer saturated with drug, excipient, and any impurities
`present, as well as the microenvironmental pH. Literature
`values range from 5% to 20% added water (16,19–22). In the
`current study, 5% moisture was used for all the drug–excipient
`mixtures.
`Although levothyroxine was found to be stable in solid
`state at accelerated temperature conditions, it degraded up to
`40% in the presence of moisture when exposed to higher
`temperature. It is sensitive to moisture which causes its
`degradation rapidly as compared to its dry form. Different
`excipients influenced the stability of levothyroxine sodium
`pentahydrate in slurries to varying extents (Fig. 3a and b).
`The smallest amount of degradation, i.e., API remaining at
`the end of 28 days >50% was observed with CSD (61.3%),
`MS (90.1%), A (76.3%), LM (96.4%), CCS (71.6%), CS
`(65.0%), and SSG (65.8%). The following excipients dis-
`played close to 50% degradation: M (50.0%), MCC (51.5%),
`
`Table III. Moisture Loss (%) Data by TGA for Levothyroxine Sodium Pentahydrate and Dehydrated Form Under Various Stability
`Conditions
`
`Conditions
`
`0
`
`3
`
`Pentahydrate form
`25°C/60%RH
`25°C/0%RH
`40°C/75%RH
`40°C/0%RH
`Dehydrated forma
`25°C/0%RH
`40°C/0%RH
`
`9.77±0.07
`9.77±0.07
`9.77±0.07
`9.77±0.07
`
`7.21
`7.21
`
`9.69±0.09
`8.31±2.05
`10.12±0.18
`8.86±0.24
`
`3.06
`2.25
`
`The results are expressed as average±SD for n=3
`a For dehydrated form n=1
`b Data not available due to instrument error
`
`Days
`
`6
`
`9.77±0.16
`9.30±0.24
`10.26±0.17
`8.01±0.48
`
`N/Ab
`4.53
`
`10
`
`14
`
`28
`
`8.81±1.61
`6.06±0.71
`9.43±1.13
`8.21±0.33
`
`N/Ab
`4.65
`
`9.59±0.50
`7.64±1.12
`10.16±0.37
`7.35±0.16
`
`3.91
`4.77
`
`9.48±0.67
`8.52±0.19
`9.96±0.50
`7.24±0.33
`
`2.62
`2.99
`
`Mylan Ex 1009, Page 5
`
`

`

`Factors on Stability of Levothyroxine Sodium Pentahydrate
`
`823
`
`Fig. 2. DSC thermogram a and TGA profile b of levothyroxine
`sodium pentahydrate
`
`Fig. 4. Stability of levothyroxine sodium pentahydrate in presence of
`crospovidone
`
`and CFS (53.8%). Unfavorable degradation, i.e., API remain-
`ing at the end of 28 days <20% was shown with CP (0%), P
`(3.2%), and SLS (14.1%). In another study also, CP was
`found to affect stability of levothyroxine sodium pentahydrate
`at 50°C after 1 month (8). Figures 4, 5, and 6 shows the
`degradation products of levothyroxine API in presence of CP,
`P, and SLS, respectively. The identification was based on the
`retention times of these impurities on chromatograms and
`comparing with the chromatogram of the API with eight
`different impurities (18).
`Half-lives were calculated for levothyroxine for all the
`drug–excipient mixtures (Table IV). The datapoint from 0 to
`5 h were used which followed a first-order kinetics. Since
`levothyroxine undergoes biphasic degradation, only the first
`phase was used for the half-life calculations. This was due to
`faster degradation rate than the second phase. Degradation
`rate constant (k) was calculated and the following equation
`was used to calculate the half-life.
`k ¼ 0:693T1=2
`This suggests that these three excipients should either be
`avoided, replaced, or their use can be minimized in solid
`dosage forms of levothyroxine sodium. Alternatively, care
`should be taken to avoid exposure to moisture.
`The equilibrium moisture studies were carried out to
`understand hygroscopicity of the excipients. The equilibrium
`moisture varied from one excipient to the other at all RH
`levels (Fig. 7a and b). The highest moisture uptake was
`observed with P, CP, SSG, A, CCS, and CS, all of which
`showed >10% moisture sorption above 50% RH. The lowest
`
`ð1Þ
`
`Fig. 3. Levothyroxine pentahydrate stability in presence of 5%
`moisture and excipient. (a) colloidal silicon dioxide, crospovidone,
`magnesium stearate, mannitol, microcrystalline cellulose, povidone,
`sodium laurel sulfate, no excipient. (b) sucrose, acacia, confectioner's
`sugar, lactose monohydrate, talc, croscarmellose sodium, corn starch,
`sodium starch glycolate
`
`Fig. 5. Stability of levothyroxine sodium pentahydrate in presence of
`povidone
`
`Mylan Ex 1009, Page 6
`
`

`

`824
`
`Collier et al.
`
`Fig. 6. Stability of levothyroxine sodium pentahydrate in presence of
`sodium laurel sulfate
`
`moisture uptake was observed with LMH, M, S, T, SLS, and
`MS, all of which showed <1% moisture sorption up to 75%
`RH. Povidone and crospovidone are very hygroscopic and
`therefore they have a tendency to absorb any moisture
`present, either from environment or from API. Thus it might
`act as a catalyst
`if
`the moisture is present and might
`accelerate the degradation reaction as was seen in stability
`samples. SLS, on the other hand, is not very hygroscopic at
`low RH. But at very high RH, it showed deliquescence. This
`might have caused acceleration in its potential to degrade
`levothyroxine. Thus, it is a combination of factors, such as
`processing as well as formulation, which might be very
`important
`to consider while formulating levothyroxine
`sodium pentahydrate tablets as some of them influences
`negatively on its stability.
`The type of excipients used in the manufacturing of
`levothyroxine sodium pentahydrate tablets will influence its
`stability. Certain excipients which caused rapid degradation of
`levothyroxine in this study should be avoided to have a more
`stable preparation. Thus excipients should be judiciously
`selected which will not cause any degradation of levothyrox-
`
`Table IV. Degradation Half-Lives for Drug–Excipient Mixture
`
`Drug with excipient
`
`Degradation half-life (days)
`
`CSD
`CP
`MS
`M
`MCC
`P
`SLS
`Control
`S
`A
`CFS
`LM
`T
`CCS
`SSG
`
`32
`3
`33
`23
`18
`3
`3
`26
`21
`25
`40
`93
`26
`27
`34
`
`Fig. 7. Moisture sorption isotherms of excipients used in levothyrox-
`ine sodium tablets. (a) povidone, crospovidone, sodium starch
`glycolate, acacia, croscarmellose sodium, corn starch, colloidal silicon
`dioxide, microcrystalline cellulose and (b) magnesium stearate,
`sodium laurel sulfate, talc, sucrose, mannitol, lactose monohydrate
`
`ine. The primary degradation pathway in solid-state levothyr-
`oxine was identified as deiodination and deamination (5).
`These findings are consistent with the previous findings of
`Patel et al., Andre et al., and Kazemifard et al. (6,8,23).
`Thus influence of processing and formulation factors
`should be carefully evaluated before formulating the tablets.
`Lot-to-lot variability, potency, and stability issues could be
`limited by choosing the factors which do not cause or catalyze
`the degradation of levothyroxine sodium. The main purpose
`of the manuscript was to evaluate the effect of formulation
`and manufacturing variables on the stability of levothyroxine,
`specifically, to understand the causes of potency loss in some
`marketed tablet formulations as compared to initial potency,
`and to screen the excipients for their protective effects. It
`would be of interest to see the effect of tabletting, but it is
`beyond the scope of the present study.
`
`CSD colloidal silicon dioxide, CP crospovidone, MS magnesium
`stearate, M mannitol, MCC microcrystalline cellulose, P povidone,
`SLS sodium laurel sulfate, S sucrose, A acacia, CFS confectioner's
`sugar, LM lactose monohydrate, T talc, CCS croscarmellose sodium,
`SSG sodium starch glycolate
`
`CONCLUSIONS
`
`Levothyroxine sodium pentahydrate drug substance was
`stable at long-term as well as accelerated stability conditions
`for the study period of 28 days. It loses some of its moisture
`
`Mylan Ex 1009, Page 7
`
`

`

`Factors on Stability of Levothyroxine Sodium Pentahydrate
`
`825
`
`rapidly when placed in a dry environment; however, it is
`stable in the dehydrated form as well as seen from the studies
`conducted by placing a dried form under the stability
`conditions of 0% RH at 25°C and 40°C. The processing
`conditions while formulating it in tablet forms might encoun-
`ter such transient conditions and, therefore, it was important
`to evaluate effects of
`these factors on the stability of
`levothyroxine. Formulation factors such as excipients were
`also found to influence its stability. Crospovidone, povi-
`done, and sodium laurel sulfate were found to be unsuit-
`able excipients when formulated in the presence of
`moisture as they cause degradation of levothyroxine. The
`degradation pathways for levothyroxine in presence of
`these excipients were deiodination and deamination. There-
`fore, a careful selection of excipients might prevent potency
`loss over the shelf-life of the tablets which could be a
`significant issue for some of the formulations of levothyr-
`oxine sodium pentahydrate.
`
`Disclaimer The opinions expressed in this work are only of the
`authors, and do not necessarily reflect the policy and statements of
`the FDA.
`
`REFERENCES
`
`1. Goodman L, Gilman A. The pharmacological basis of therapeu-
`tics. 10th ed. New York: McGraw-Hill; 2001.
`2. Rhodes C. Regulatory aspects of the formulation and evaluation
`of levothyroxine tablets. Clin Res Regul Aff. 1998;15:173–86.
`3. FDA. Regulatory history and current issues. Health & Human
`Services; 2006.
`4. Post A, Warren R. Sodium levothyroxine. In: Florey K, editor.
`Analytical profiles of drug substances. New York: Academic
`Press; 1976. p. 226–81.
`5. Won CM. Kinetics of degradation of levothyroxine in aqueous
`solution and in solid state. Pharm Res. 1992;9:131–7.
`6. Kazemifard AG, Moore DE, Aghazadeh A. Identification and
`quantitation of sodium-thyroxine and its degradation products by
`LC using electrochemical and MS detection. J Pharm Biomed
`Anal. 2001;25:697–711.
`7. Wortsman J, Papadimitriou DC, Borges M, Defesche CL.
`Thermal inactivat