Welcome

This third edition of the Science and Technology newsletter will continue our discussion on beyond-use dates and how to assign them based on United States Pharmacopeia criteria for both nonsterile and sterile preparations. This issue will discuss the topic of interpreting published stability studies.

Beyond-use Dates, Part 3
Interpreting Published Stability Studies

Introduction
One of the most important activities of a compounding pharmacist is the assignment of a beyond-use date (BUD) for a compounded preparation. It is essential that the date be reasonably determined from either laboratory testing or using official default standards. Published studies are widely used in establishing BUDs.

Stability is the extent to which a product retains, within specified limits and throughout its period of storage and use (i.e., shelf life), the same properties and characteristics that it possessed at the time of its manufacturer. There are five types of stability which should be of concern to pharmacists.

1. Chemical stability is where each active ingredient retains its chemical integrity and labeled potency within the specified limits.
2. Physical stability is where the original physical properties, including appearance, palatability, uniformity, dissolution, and suspendability are retained.
3. Microbiologic stability includes sterility or resistance as well as the situation where resistance to microbial growth is retained according to the specified requirements. Antimicrobial agents retain effectiveness within specified limits.
4. Therapeutic stability is where the therapeutic effect remains unchanged.
5. Toxicologic stability is where there is no significant increase in toxicity.

Chemical stability is important for selecting storage conditions (temperature, light, humidity), selecting the proper container for packaging and dispensing (glass vs plastic, clear vs amber or opaque, cap liners), and anticipating interactions when mixing drugs and dosage forms.

Stability, expiration dates, and BUDs are based on reaction kinetics, the study of the rate of chemical change and the way this rate is influenced by concentration of reactants, products, and other chemical species, and by factors such as solvent, pressure, and temperature.

Study Designs
Most stability studies result from questions involving commonly used dosages in children. It is routine to study only one, or maybe two, concentrations of a drug in the selected vehicles. Temperatures commonly used in these studies include room temperature and/or refrigerated temperature. The preparations are generally placed in light-resistant containers and sampled at predetermined intervals. The preparations may be prepared from either commercial dosage forms or from United States Pharmacopeia (USP)-grade substances.

Stability-indicating assays, generally high-performance liquid chromatography (HPLC), are routinely developed, validated, and used for these studies. A standard cutoff of 90% of the initial concentration remaining is generally used in assigning a BUD. It is customary to provide the actual analyzed concentration at the initial time point, and the remaining time intervals are expressed as a percent of the original drug remaining. In the actual studies, the standard deviations must be provided to assist in evaluating the cutoff time for the recommended BUD.

In general, the procedures involve the following steps:

1. Selection and development of the drugs to be tested and the formulation for the study
2. Development and validation of a stability-indicating assay for the active drug in the formulation
3. Development of the study protocol, including containers, storage temperatures, etc.
4. Preparation of the study samples
5. Pulling the initial samples and placing the containers in the appropriate storage conditions
6. Pulling samples as indicated
7. Analyzing the active drug
8. Data analysis and presentation

The Stability Study
For a study to be conducted, the sample must be properly prepared. The formulation should be one that is widely used or can be widely compounded by pharmacists. The ingredients should be commonplace and the formulation relatively simple. Sufficient sample should be prepared to allow for multiple containers for different storage conditions and sampling times.

Storage conditions should include at least room temperature and refrigerated temperature. Frozen temperature is a good addition, and other temperatures as required. Sampling times should include an initial sample right after preparation (time-zero), and then at representative time intervals based upon the desired potential storage times. This can include 1, 3, 7, 14, 30, 60, 90, 120, 180 days if that is reasonable.

Sufficient material should be prepared for at least triplicate containers at each storage condition. From each of these triplicate containers, at least one or two samples can be obtained at the specific sampling interval. Each obtained sample is analyzed individually. It is apparent that if only triplicate containers are involved, and only one sample is obtained, then 3 data points are all that are obtained at each sample interval. If one is spoiled, then that is a problem. However, if two or three samples are obtained from each container at each time point and there are triplicate containers, then 6 or 9 data points would be possible, and, if one or more is spoiled, there is still sufficient numbers for interpretation. Sample-handling must be considered to ensure the sample is not decomposed or altered between the time of sampling and analysis.

Once the study experimentation is complete, the next step is data analysis and interpretation. Generally, descriptive statistics are obtained and placed into a table format as shown in the examples below.

It should be noted that there can be confusion in the way stability studies are referred to in the literature. For example, if a study was only conducted for 7 days and no longer and at the end of that study 99.5% of the active drug remained, it may be presented as the drug is "stable for 7 days." Actually, the drug is stable for "at least 7 days." Therefore, it is important to review the original article when establishing the BUD for the compounded preparation.

It should also be noted that there is a difference in the purpose of potency tests and stability tests. Potency testing, quantitative testing, is designed to determine how much of an active drug is present in a sample. Stability tests are used to determine an expiration date or a BUD for a preparation. Methods of determining potency may or may not be stability indicating. Stability can be determined only by a stability-indicating method, which can determine both potency and stability. However, a potency test may or may not be able to determine stability.

Stability-indicating Methods
The purpose of a stability-indicating assay method is to accurately quantitate the intact drug or drugs in the presence of decomposition products and other components/excipients. It is best that all components in the formulation be present to confirm there is not any peak overlap between the excipients, degradants, and the active drug. This is essential to provide accurate and reliable results.

Generally, the development of a stability-indicating assay method involves subjecting the samples containing the drug (and preferably the complete formulation) to the following conditions:

1. Extremes of pH (generally using strong acids and bases)
2. Extremes of temperature, especially high temperature (boiling)
3. Addition of an oxidation agent (hydrogen peroxide)
4. Exposure to light

The complete drug formulation should be exposed to these conditions, in addition to just the drug in aqueous solution. The purpose of this activity is to ensure that the peak representing the intact/nondegraded drug in a chromatogram is only due to the drug itself and not due to a degradation product or an excipient(s). Some HPLCs have "peak purity" software that can be of great value in this procedure.

High-performance Liquid Chromatography
HPLC is the most common method used in pharmaceutical analysis today. It is based on differential solubility between the components in the sample for the solvent system (mobile phase) and the stationary phase. The peak height or peak area is proportional to the concentration of the entities being detected. A test for linearity is also required to confirm that the increase in drug concentration is linear with the peak height/area. It is a matter of experimentation to obtain the proper conditions (mobile phase, stationary phase, flow rate, temperature, detection method, and/or wavelength) for the analysis. The process, once completed, then requires standards for the development of a standard curve which is used for quantitation purposes.

Data Presentation
Generally, the data that is presented in studies will be in tabular form and sometimes, in graphic form. The time-zero sample is customarily set to 100% and each subsequent data value compared to the time-zero value. Generally, the subsequent data is either around 100% or below. Occasionally, a data value greater than 100% is obtained after time-zero. This can be related to improper container mixing when sampling, improper sample mixing prior to analysis, evaporation of solvent, analytical variability, and other factors.

In looking at the data, one looks at the standard deviations to make sure the data is reasonably "tight." Then, looking at the means, one observes whether there is a tendency for drug degradation or if the drug appears to be relatively stable. If the values stay close to 100%, then the formulation appears stable. If the values decrease over time, then one has to determine what is happening at about the 90% value. The drug preparation must retain at least 90% of the time-zero value to be considered stable at a respective time point; and, this includes the standard deviation value. For example, if a data value is 92.2% +/- 1.3, then the value ranges from 90.9% to 93.5%, which is acceptable. However, if the data value is 92.2% +/- 3.1, then the value ranges from 89.1% to 95.3%; since the low value is less than 90%, this cannot be used as the time for which the drug is still stable. Interpretation between sampling times is possible but must be based upon knowledge of whether the degradation occurs as zero order, first order, or other. It is generally best to look at the data and then assign a BUD conservatively as appropriate.

In addition to drug concentration, pH determinations and physical observations should be made at each sampling time. Physical observations can include general appearance, color, ease of resuspending (if a suspension), clarity, uniformity, gas formation, odor, etc.

Examples
Let's look at two examples of stability studies where BUDs have been assigned. These were reported in Volume 15, Number 1 of Secundum Artem (available at www.PaddockLabs.com). These examples will demonstrate how a reasonable "BUD" can be obtained from the data presented.

Example 1. Valacyclovir Hydrochloride
The valacyclovir hydrochloride 50-mg/mL oral liquid was prepared using the caplets and a porcelain mortar; the caplets were first crushed to a fine powder. The suspension vehicle (Ora-Sweet or Ora-Sweet SF with Ora-Plus) was added with mixing between additions. The preparation was transferred to an amber glass bottle where the final vehicle was added by rinsing the mortar and adding to the final container (5 rinses); the bottles were stored at refrigerated temperature and sampled weekly for four weeks. The data is shown in Table 1.

Table 1. Stability of Valacyclovir 50-mg/mL Oral Liquids at Refrigerated Temperatures.

One can reasonably state that the data shows both preparations are stable in the refrigerator for 14 days. The standard deviations are too great to extend the storage period to 21 days, as the actual range may include 89.2% to 94.0% for the Ora-Sweet/Ora-Plus combination and 86.3% and 93.9% for the Ora-Sweet SF/Ora-Plus combination. Since the values drop below 90.0%, that time cannot be used. It may be reasonable to interpolate between 14 days and 21 days if the order of the reaction is known. The pH values remained unchanged as did the physical observations.

Example 2. Terbinafine Hydrochloride
Terbinafine hydrochloride 25 mg/mL was prepared using the terbinafine tablets. The tablets were crushed to a fine powder in a mortar and a small quantity of the vehicle (Ora-Plus and Ora-Sweet; 1:1) was used to make a smooth paste. Additional volumes of the vehicle were added, and the preparation transferred to a graduate where it was brought to final volume. The suspension was packaged in amber polyethylene prescription bottles and stored at both room and refrigerated temperatures. Samples were withdrawn for up to 91 days. The results are shown in Table 2.

Table 2. Stability of Terbinafine Hydrochloride 25 mg/mL in Amber Polyethylene Prescription Bottles is Stable at Both Room and Refrigerated Temperatures for Up to 42 Days.

The data shows that terbinafine hydrochloride 25 mg/mL is stable for up to 42 days in polyethylene prescription bottles at both room and refrigerated temperatures. The pH of the suspension decreased only very slightly over 91 days, from an initial pH 5.6 to 5.5. It may be reasonable to interpolate between 42 days and 56 days if the order of the reaction is known.

Summary
In evaluating stability studies for appropriateness, it is important to check for specific topics to be addressed in the studies. A cardinal rule is that one of similar education and ability should be able to reproduce the results from the information provided in the published study. Some specific issues to address in stability studies of extemporaneous compounded preparations are listed in Table 3. Generally, the proposed BUD assigned should be determined at less than about a 7% loss. Also, the appropriate BUD is no longer than the length of the study. Since excipients may affect stability, the source of the drug must be specified; is it the bulk drug substance, capsules, tablets, injections?

Table 3. General Guidelines for Appropriateness of a Stability Study to be Used in Assigning Beyond-use Dates.

1. Is the complete formula composition provided, including whether or not commercial products or bulk substances were used?
2. Are samples evaluated with >1 intermediate time point?
3. Are the sample storage conditions described (temperature, light, container)?
4. Is the analytical method used stability indicating?
5. Is the study published in a peer-reviewed journal or equivalent?
6. Are the lot numbers for components of the study preparation provided?
7. Are the equipment and supplies used specified with manufacturer and model numbers?
8. Is there sufficient detail about the study for reproducibility?
9. Are all components used United States Pharmacopeia, National Formulary, FCC, or ACS grade?
10. Is there an initial time-zero evaluation?
11. Are the time intervals between assays reasonable and useful?
12. Is the sample schedule for evaluation reasonable?
13. Was time-zero testing performed on at least 3 replicates?
14. Was there any overlap or interference with the intact drug peak?
15. Was accelerated decomposition to loss of 25% to 75% of the drug performed?
16. Are sample chromatograms or elution times provided?
17. Was a reference standard calibration curve range 80% to 120% utilized?
18. Was the reference standard calibration curve linearity >0.995?
19. Is the limit of quantification of the analyte stated?
20. Is the method repeatability for >6 replicates and >2%?
21. Is the intraday and interday coefficients of variation <3%?
22. Was there an evaluation for unacceptable particulate matter?
23. Was there any visible particulate matter (for solutions)?
24. Was there an evaluation for microparticulate formation (for injections and ophthalmics)?
25. Were there any changes in pH, color, or turbidity?
26. Was an evaluation of uniformity, settling, caking, resuspension performed (for suspensions)?
27. Was there an evaluation of uniformity, creaming, layering, etc. (for emulsions)?
28. Did the analytical replicates exceed 2% variation?
29. Were any anomalous results adequately explained?
30. Do the conclusions overreach the data?

For further information:
www.PaddockLabs.com
Stability of Extemporaneously Prepared Pediatric Formulations: Part I (Vol. 5, No. 4); Part II (Vol. 6, No. 1); Part III (Vol. 6, No. 2); Part IV (Vol. 14, No. 1); Part V (Vol. 14, No. 3); Part VI (Vol. 15, No. 1); Part VII (Vol. 16, No. 1); Part VIII (Vol. 16, No. 3).

…that www.CompoundingToday.com is the world's largest database for compounding information? It includes Trissels 2 Clinical Pharmaceutics Database, containing the most complete collection of compatibility and stability information for both sterile and nonsterile compounding available anywhere.

David W. Newton, FAPhA, earned BSP, MSP, and PhD degrees from the University of Florida in 1967, 1972, and 1976. In 1969, as a U.S. Army artillery observer and medic in Vietnam combat, he was awarded Bronze Star, Air, and Army Commendation Medals. In the early 1970s, he implemented intravenous admixture services in a U.S. Army and a municipal hospital. He has authored 110 journal articles and made 40 national presentations mostly on pharmaceutical compounding and stability topics. From 2000-2010, he chaired the U.S. Pharmacopeial (USP) committee that developed the original (2004) and first revision (2008) of United States Pharmacopeia Chapter <797> Pharmaceutical Compounding-Sterile Preparations. He serves on the 2010-2015 USP Compounding Expert Committee, and has served on the NABPLEX®/NAPLEX® Review Committee since 1983. He was a 1983 member of the inaugural Am J Hosp Pharm, AJHP, Editorial Board serving through 1997, and he authored the first AJHP paper for CE credit in 1978 (pages 1213-1222). Since 1996, he has been a professor at the Bernard J. Dunn School of Pharmacy of Shenandoah University in Winchester, Virginia, following faculty appointments at three other schools from 1976-1996.

Future Topics (2011)