Cetiedil Citrate Injection, containing 5 mg/mL cetiedil citrate, sodium chloride, and Water for Injection in Type I glass ampules, having an unadjusted pH of 3.3 to 3.7 was insufficiently stable if stored at controlled room temperature (20º to 25º). Shelf life predictions indicated a 1.66 to 1.8 year expiry period. A solution kinetics study showed maximum stability at pH between 3.5 and 3.8. No advantage was observed in the stabilities when the concentration was increased 10-fold, or one of three buffers (citrate-phosphate, acetate, or succinate) were used. The only alternative was storage at refrigeration or Controlled Cold Storage, 2º to 8º.

Arrhenius plots were used to estimate the rate constants at the higher extremes of Controlled Cold (8º) and Controlled Room Temperatures (25º). Eight degrees showed the best promise for satisfactory expiry dating. The experimental and derived temperatures and rate constants were applied to 2 process models that defined all of the steps in the manufacture, testing, release, packaging, transportation and storage, tracking the hypothetical stability in terms of % cetiedil remaining after each step in the process. The models were identical except that one included refrigeration storage and the other did not. The refrigerated solution met stability criteria longer than the non-refrigerated product. However, several processing steps including autoclaving and transportation demanded the application of higher temperatures for a short duration and affected the stability of the product. The data from this study indicate that a 3 year expiry date should be applied.

•    pH Stability Profile
•    Arrhenius Plots
•    Stability vs. Processes and Duration

Chrzanowsk F (2020) Cetiedil Citrate Injection, Resolve Instability: Case Study. J Clin Pharm 4(1): 1016.

Cetiedil Citrate, 2-(azepan-1-yl)ethyl-2-cyclohexyl-2- thiophen-3-ylacetate; 2-hydroxypropane-1,2,3-tricarboxylic acid salt (Figure 1).

Figure 1: Cetiedil Citrate.

Cetiedil Citrate Injection was an in licensed product under development for the treatment of sickle cell crises. Cetiedil at a dose of 0.4 mg/kg was found to be significantly superior to placebo in reducing the number of painful sites present on all four treatment days and in shortening the total time in crisis [1].

The formulation contained cetiedil citrate 5 mg/mL expressed as the free base, sodium chloride, and Water for Injection in 10 mL Type I glass ampules. The solution was isotonic, contained no other ingredients and was sterilized by terminal sterilization in an autoclave at 120º/14 psi for 20 minutes. The pH was in the range 3.3 to 3.7 (its native pH). Clinical supply batches were on stability testing in 40º , 30º , and 25º chambers with protocols for appropriate testing at monthly intervals for 6 months, quarterly until 18 months, and semiannually up to 48 months.

An early assessment of the stability of the clinical supplies after 3 months storage (Table 1)

Table 1: Cetiedil Citrate Injection, 5 mg/mL, 10 mL ampules, Clinical Supplies Stability Data.

indicated that solution stability was becoming a problem. The accelerated 3 month 40º samples had 96.2% remaining. The 30º samples had 98.5% and the 25º or Controlled Room Temperature samples had 99.0%. The shelf life specification was not less than 95.0% remaining. The stability problem was confirmed at 6 months, with 93.0% at 40º , 97.3% at 30º , and 98.0% at 25º . The data indicated a shorter than 2 year expiry dating or shelf life which would be problematic both in the clinical development program and eventual commercialization.

The intended release specifications are 98 to 102%, and a shelf life of not less than 95.0 % remaining. A minimum shelf life of 2 years would be acceptable; however, 3 years or more would be preferred.

Preformulation testing of cetiedil citrate drug substance had been initiated and was ongoing at the time of discovery of the stability problem. Cetiedil has a pKa = 9.56 [2]. Cetiedil citrate has usable solubility within the range pH 1 to 5. The solution kinetics were under investigation in the pH 1 to 5 range at a concentration of 0.2 mg/mL in 0.1 N hydrochloric acid, pH 2, 3, 4 and 5 citratephosphate buffers at constant ionic strength, µ = 0.5 M, water with and without ionic strength adjustment at temperatures of 80º, 60º, 40º and 30º. The effect of cetiedil concentration at 10-fold the 0.2 mg/mL concentration, and buffers were added after discovery of the stability problem with the formulation on stability testing.

Arrhenius plots were created from data collected from two pH values to obtain estimates of the rate constants at 25º and 8º . These temperatures represent the upper temperature limits of the USP/NF defined Controlled Room Temperature (20º to 25º ) and Controlled Cold Temperature (2º to 8º ) [3]. The USP/ NF further states a refrigerator is a cold place in which the temperature is maintained thermostatically between 2º and 8º .

The stability and degradation of a hypothetical batch of cetiedil citrate ampules carried through the manufacturing, packaging, warehouse storage and delivery to a pharmacy was evaluated assuming the absence of refrigeration storage and the inclusion of refrigeration storage to demonstrate the advantages of using refrigeration and validate the selection of using refrigeration for storage and transport of Cetiedil Citrate Injection by the manufacturer.

Kinetics

Cetiedil citrate solutions containing 0.2 mg/mL were prepared at pHs of 1, 2, 3, 4 and 5 using 0.1 N hydrochloric acid (pH 1), McIlveen’s Citrate-Phosphate pH 2, 3, 4 and 5 Buffers and adjusted to constant ionic strength, µ = 0.5 M [4] and water with and without ionic strength adjustment for use in the pH Stability Profile determination. The pH range was limited to pH 1 to 5 because the solubility of cetiedil citrate was less 0.2 mg/mL at pH 6 and higher. Additional 0.2 mg/mL solutions were prepared using acetate [5], succinate [6] and the citrate phosphate buffers adjusted to pH 3.7, and cetiedil citrate in water at a 10-fold concentration of 2.0 mg/mL.

The cetiedil concentrations remaining (%) vs. time (days) for each solution were plotted on semi-logarithmic graph paper. The rate constant (k) was determined by fitting the concentration vs. time data to the following non-linear model equations for first order kinetics using a kinetics variation of the NONLIN program developed by Niebergall et al [7].

where:

= concentration (%) remaining at time, t (days)

= an initial concentration

k = rate constant (days-1)

t = time in days

The time in which a solution stored at the same temperature reaches 95.0%, () can be calculated from its rate constant (k).

=ln (0.95)/k

= - 0.05129/k

Arrhenius plots

The rate constants from 80º and 60º storage of the initial 0.2 mg/mL 0.1 N hydrochloric acid, citrate-phosphate buffer solutions, and water with and without ionic strength adjustment were used to create a pH-Stability Profile. The rate constants for 80º , 60º , 40º and 30º for pH 3.7 (water) and 5.0 (citrate-phosphate) were used to create graphical plots of the Arrhenius Equation as described below to obtain estimates for the degradation rates at 25º and 8º by extrapolation.

Arrhenius Plots

where

k = rate constant at a given temperature and pH,

A = pre-exponential factor

Ea = Activation energy

R = Gas Constant

T = Absolute temperature, K (degrees Kelvin)

and

ln(k) = ln (A) – (Ea/R) (1/T)

The linear version, ln(k) = ln(A) –(Ea/R)(1/T), was evaluated using linear regression because there was insufficient data at each pH to fit the Arrhenius equation to a non-linear model. Reasonable estimates for the values of A and Ea were not available.

Process models

A list of the processes, lengths of time and upper temperature limits starting at mix, fill and finish stage up to dispensing to a patient was created assuming no storage at temperatures less than ambient room temperature (25º ). A second similar list was created assuming that refrigerated storage (2º to 8º ) was available during the life of the product and its initial transportation from the manufacturing facility. The upper limits of the temperature ranges and duration of each steep were used to calculate the concentrations of cetiedil remaining after each step in the proposed life of the product.

Kinetics

Semi-logarithmic plots of cetiedil % remaining vs. time for all of the solutions indicated that cetiedil degradation followed a pseudo first-order model with respect to cetiedil concentration. Plots representing the kinetics at pH 3.7 and 5.0 are presented in Figure 2.

Figure 2: Cetiedil Solution Kinetics pH 3.7 and 5.0 at 80º, 60º, and 40º.

Non-linear regression analyses of the data provided rate constant values supporting pseudo first order degradation dependent upon the concentration of cetiedil in all of the test solutions. The pH-Stability Profile created using the rate constants at 80º and 60º is presented as Figure 3: Cetiedil pH Stability Profile, 80º and 60º .

Figure 3: Cetiedil pH Stability Profile, 80º and 60º.

Maximum rates of degradation were observed at the extremes, pH 1 and pH 5, with minimal degradation in range pH 3.3 to 3.8, which happens to be the pH range of the 5 mg/ mL injection formulation. The optimal pH range is the existing pH range of the formulated product.

The degradation rate was not affected by a 10 fold increase in cetiedil concentration, 0.2 to 2 mg/mL. Stability could not be increased or improved by changes in pH, ionic strength, adding buffers such as citrate-phosphate, acetate, or succinate.

Arrhenius Plots

The Arrhenius plots for pH 3.7 and 5.0 are presented in Figure 4

Figure 4: Arrhenius Plots, rate constant k vs. 1/TºK for pH 3.7 and 5.0.

The rate constants used to create Figure 4 and the extrapolated rate constants are presented in Table II also lists the t95 for pH 3.7 and 5.0 at 8º as 11.4 and 2.9 years respectively. The t95 at 25º and pH 3.7 is 1.8 years.

Process Models

The t95 value for pH 3.7 and storage at 8º suggests incredible stability, 11.4 years. However, other processes and activities, and the temperatures and durations must be considered to guarantee satisfactory prediction of stability gained by the use of 8º storage. The maximum duration for a process, analyses, inspections, transportation, packaging and storage were applied with the maximum temperature and its respective rate constant to demonstrate the stability or lack of sufficient stability to meet a preferred 3 year shelf life, i.e. at least 3 years expiry dating after manufacture. Shorter times and lower temperatures might have been sufficient for some of the procedures. However, if a shorter time period could not be guaranteed, the model would not be as dependable and predictive. Then the model would be inadequate to predict an expiry date. Procedures such as Warehouse and Pharmacy storage were automatically assigned 180 days duration because that is what happens at these locations, waiting to be ordered and shipped, or waiting to be needed and dispensed.

The hypothetical solutions are assumed to have a concentration equal to 100% after formulation and 1% degradation due to sterilization under autoclave conditions. Laboratory and storage experiences were assigned a 25º temperature. Shipping was assigned a 40º temperature because it is equivalent to 104º F, a temperature that might be encountered during the summer in the northeast and central US. The calculations were made using the non-linear equation A = A0 e-kt. The value of A0 for each process was the value of A, the % remaining at the end of the last procedure. The tables list the procedures, durations and cumulative durations, the assigned temperatures and rate constants, the ratios of A/A0 , percentage remaining and cumulative time in years.

In the absence of storage and transportation under refrigerated conditions the data in Table 2

Table 2: Cetiedil degradation rates (days-1) at pH 3.7 and 5.0 vs. temperatures and t95 at 8º.

demonstrate that the hypothetical batch is at 95.1% after 1.66 years, although the calculation from its rate constant at 25º was 1.8 years, a difference of 55 days, showing that expiry dating calculated with only one temperature such as ambient room, i.e. ~ 25º may overestimate the expiry date.

A review of the results of this study lead to a decision to provide refrigerated storage and transportation of the product by the manufacturing organization, and require refrigerated storage for the product, with allowable excursions during transport. The specification that the product could only be stored in the Cold could be enforced at warehousing and in the pharmacy inventories. However, requiring refrigeration for transport from the packaging or wholesaler organizations would be difficult to enforce. Table 4 describes the processes, duration, assigned temperature and rate constant, the ratio and percentage of cetiedil remaining when cold storage/refrigeration is used. After 1.66 years, the time for 95.1% remaining in the previous model, the hypothetical product would contain 97.7% intact cetiedil. Because a Cold Place is required for all storage, time spent in the warehouses and pharmacy is interchangeable. Table 4 contains 8 entries listed as Pharmacy / WWH indicating the equivalence. The Ship to Pharmacy Truck captures the effect of the last transportation of the product and could appear at a different time in the history of the product. However, its existence is captured at 425 days or 1.16 years, although it could occur at any time up to 1865 days or 5.11 years and 96.2% of the cetiedil would have remained in this hypothetical example.

Table 3: Calculated Stability for Cetiedil Citrate Injection vs. manufacturing, testing, release, distribution, and storage processes with no refrigeration.

Table 4: Calculated Stability for Cetiedil Citrate Injection vs. manufacturing, testing, release, packaging, storage, and distribution processes that include Cold Storage.

Table 3 and Table 4 show that for a parenteral product such as cetiedil citrate, it takes about 58 days or about 2 months to manufacture and have available for distribution to wholesalers and pharmacies. This should be included in any consideration of establishing an expiry date.

The data in this study demonstrated that the native pH range of the existing formulation provided maximum stability for the product. Changes in pH, ionic strength, and buffer did not improve the stability of Cetiedil Injection. The only viable alternative for improving stability is storage in a Cold Storage environment. The data in Table 3 and Table 4 show the advantages of lower temperature storage in a Cold Place. A corporate decision was made that storage under refrigeration would be provided by corporate at the manufacturing and warehouse facilities and corporate trucks for delivery to the packaging facility

The data in Table IV demonstrate that only about 3% degradation occurs after 3 years. Had the initial batch contained only 98.0%, which is within the release rate criteria, the cetiedil percent remaining would approach 95.0% at 3 years. The shelf life specification for this product is 95.0%. The data in this study support a 3 year maximum expiry dating.

The work described in this study was performed while the author was employed at McNeil Pharmaceutical, Welsh and McKean Roads, Spring House PA. The author thanks Barry Fegely, Lucy Ulissi and Beth Ann Ahlswede for their assistance and to Irwin Gibbs for suggesting consideration of including processing conditions and their effects on stability.

1. Benjamin LJ, Berkowitz LR, Orringer E, Mankad VN, Prasad AS, Lewkow LM, et al. A collaborative, double blind randomized study of cetiedil citrate in sickle cell crisis. 1986; 67: 1442-1447.

2. Chrzanowski FA, Ulissi LA. pKa of Cetiedil. Acta Pharm Technol. 1987; 33: 35-37.

3. General Notices. The United States Pharmacopeia The National Formulary Twenty-sixth edition. USP 31st Revision, NF 26th edition. Rockville: US Pharmacopeia. 2008; 1: 10.

4. Elving PJ, Markowitz M, Rosenthal I. Preparation of Buffer Systems of Constant Ionic Strength. 1956; 28: 1179 - 1180.

5. Walpole GS. CCXXXIII. Hydrogen Potentials of mixtures of acetic acid and sodium acetate. 1914; 105: 2501-2521.

6. Gomori G, Colowick SP, Kaplan NO. Preparation of buffers for use in enzyme studies. [ed.] NO Kaplan SP Colowick. Methods in Ezymology. New York: Academic Press. 1955; 1: 138-146.

7. Niebergall PJ, Schnaare RL, Sugita ET. Nonlinear regression applied to non-newtonian flow. JPharmSci. 1971; 60: 1393-1402.