Statistical Evaluation of the Chemical Stability of Ziconotide Solutions During Simulated Intrathecal Administration

Article Outline

• Methods
• Results
• Comment
• Acknowledgments
• Reference
• Copyright

To the Editor:

Ziconotide is indicated for the management of severe chronic pain in patients for whom intrathecal (IT) therapy is warranted and who are intolerant of, or have pain that is refractory to, other treatments, such as systemic analgesics, adjunctive therapies, or IT morphine.¹ The 25-mcg/mL formulation is used for pump priming and for the initial fill of pumps that have never been exposed to ziconotide; the 100-mcg/mL formulation can be diluted with 0.9% sodium chloride, or it can be administered undiluted after a satisfactory therapeutic dose has been identified.¹ After the initial fill of a ziconotide-naïve pump with the 25 mcg/mL solution, the current prescribing information recommends that the pump be emptied and refilled within 14 days. Subsequent refills are recommended every 84 days (or less) for pumps containing undiluted ziconotide formulations and every 40 days (or less) for pumps containing diluted ziconotide solutions. These refill intervals were determined from ziconotide clinical trials and from 15 chemical stability studies during simulated IT administration. This report combines the results of the 15 stability studies with six additional stability studies, which were performed after the prescribing information was released, to provide a more complete characterization of ziconotide chemical stability during simulated clinical use.

Back to Article Outline

Methods 

Solutions of 100 mcg/mL formulated ziconotide, 25 mcg/mL formulated ziconotide, and 25 mcg/mL diluted ziconotide were each evaluated in seven studies. Each study was composed of two experiments; the first experiment was performed by filling pumps that had not been previously exposed to ziconotide (i.e., naïve pumps) with one of the three ziconotide solutions, and the second experiment was performed by refilling the pumps (i.e., exposed pumps) with a fresh preparation of the same ziconotide solution. Each experiment used 20mL of ziconotide solution per pump.

Ziconotide (Elan Pharmaceuticals, Inc., South San Francisco, CA) was obtained from commercial inventory in 5-mL vials containing 100 mcg/mL ziconotide or 20-mL vials containing 25 mcg/mL ziconotide; both commercial formulations included 50 mcg/mL of methionine to reduce ziconotide oxidation. The 25 mcg/mL diluted ziconotide solution was prepared by gently mixing one part 100 mcg/mL formulated ziconotide solution with three parts preservative-free, sterile 0.9% sodium chloride injection (USP). Studies performed with the 100 mcg/mL and 25 mcg/mL formulated ziconotide solutions used the commercially available preparations.

A total of 126 new, previously unused pumps (72 SynchroMed® II and 54 SynchroMed EL pumps [Medtronic, Inc., Minneapolis, MN]) were obtained for these studies. The pumps were warmed at 37°C in a low-temperature incubator and then filled according to the recommendations in the prescribing information. Upon completion of the naïve-pump experiment (four weeks later), the pumps were emptied and refilled with 20mL of freshly prepared ziconotide solution for use in the exposed-pump experiment. Pumps were first sampled within minutes of filling (Week 0). Subsequent samples were obtained on Weeks 1, 2, 3, and 4 for naïve pumps and on Weeks 2, 4, 8, 10, and 12 for exposed pumps. Two experiments using the 100-mcg/mL ziconotide formulation in exposed pumps also included samples obtained on Week 6, and the Week 1 sampling was omitted in three experiments with naïve pumps. The sample collection vials were transferred to a high-performance liquid chromatograph and analyzed within 24hours of collection.

The reverse-phase high-performance liquid chromatograph used was an Agilent 1100 model (Agilent Technologies, Inc., Santa Clara, CA) equipped with a Vydac C18, 5-μ, 250X 4.6-mm column (Grace Vydac, Hesperia, CA) maintained at 40°C ± 2°C as the stationary phase. The mobile phase consisted of gradient elutions starting with 100% mobile phase A (consisting of 96% water, 4% isopropyl alcohol, and 0.1% trifluoroacetic acid) and transitioning to 25% mobile phase B (consisting of 4% water, 96% isopropyl alcohol, and 0.1% trifluoroacetic acid) over the course of 35minutes. The flow rate used throughout the procedure was 1mL/min. The ultraviolet detector was set at 214nm. Sample solutions and the ziconotide reference standard (100 mcg/mL) were injected in 150 mcL volumes.

Stability was defined as the time period during which pump samples retained at least 90% of the ziconotide concentration measured at Week 0. A linear regression analysis was conducted by using statistical software package SAS version 8.2. With the use of a fitted regression line, the predicted average percentages of ziconotide remaining, along with the lower 95% confidence interval, were calculated, and the corresponding time points were obtained. The estimated in-pump ziconotide stability was calculated as the length of time during which the lower bound of the 95% confidence interval for the predicted average percentage of ziconotide concentration remained at or above 90% of the initial ziconotide concentration; values that exceeded the duration of the study were extrapolated assuming linear degradation.

Back to Article Outline

Results 

At each time point (with the exception of Week 6), at least 35 pumps were sampled for each solution and pump type. At Week 6, 12 exposed pumps were sampled for the 100 mcg/mL formulated ziconotide solution, and 0 exposed pumps were sampled for the 25 mcg/mL formulated and diluted ziconotide solutions. Missing data were the result of pump failures and programming errors. In naïve pumps filled with the 100-mcg/mL formulated ziconotide solution, the mean ziconotide concentration at the end of the study (Week 4) exceeded 95% of the initial concentration (range, 92.1%–98.7%; SD, 1.2%) (Fig. 1a). At study termination, samples from naïve pumps filled with the 25-mcg/mL formulated ziconotide solution had a mean ziconotide concentration of 88.3% of the initial concentration (range, 78.0%–94.2%; SD, 4.3%), whereas samples from naïve pumps filled with the 25-mcg/mL diluted ziconotide solution had a mean ziconotide concentration of 81.1% of the initial concentration (range, 67.6%–94.9%; SD, 5.1%). In naïve pumps, the lower bound of the 95% confidence interval for ziconotide concentration in the 100 and 25-mcg/mL commercial formulations and the 25-mcg/mL diluted solution exceeded 90% of the initial ziconotide concentration for 57, 22, and 13 days, respectively.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1 

  (a) The stability of three ziconotide solutions over time when stored in ziconotide-naïve pumps. The concentration of ziconotide in new, unused Medtronic SynchroMed^® pumps stored at 37°C was expressed as a percentage of the concentration measured at study initiation, and the mean percentage measured each week is displayed for each of three ziconotide solutions: the commercially available 100-mcg/mL formulated solution (diamonds), the commercially available 25-mcg/mL formulated solution (squares), and a 25-mcg/mL ziconotide solution prepared by diluting the commercially available 100-mcg/mL formulation with preservative-free, sterile 0.9% sodium chloride (USP) (triangles). Error bars represent 1 standard deviation below the mean. (b) The stability of three ziconotide solutions over time when stored in pumps that had been previously exposed to ziconotide. Upon completion of the experiments displayed in (a), the Medtronic SynchroMed^® pumps were refilled with a fresh ziconotide solution prepared to match the solution used in the first series of experiments, and the pumps were then stored at 37°C. The concentration of ziconotide is expressed as a percentage of the concentration measured at study initiation, and the mean percentage measured on each sampling week is displayed for each of three ziconotide solutions: the commercially available 100-mcg/mL formulated solution (diamonds), the commercially available 25-mcg/mL formulated solution (squares), and a 25-mcg/mL ziconotide solution prepared by diluting the commercially available 100-mcg/mL formulation with preservative-free, sterile 0.9% sodium chloride (USP) (triangles). Error bars represent 1 standard deviation below the mean.

In exposed pumps filled with the 100-mcg/mL formulated ziconotide solution, the mean ziconotide concentration at the end of the study (Week 12) measured 95.0% of the initial concentration (range, 93.3%–96.8%; SD, 0.9%) (Fig. 1b). At study termination, samples from exposed pumps containing the 25 mcg/mL formulated ziconotide solution had a mean ziconotide concentration of 92.4% of the initial concentration (range, 85.0%–99.6%; SD, 2.4%), whereas samples from exposed pumps filled with the 25-mcg/mL diluted ziconotide solution had a mean ziconotide concentration of 85.1% of the initial concentration (range, 78.9%–92.8%; SD, 3.5%). In exposed pumps, the 100 and 25-mcg/mL commercial formulations and the 25-mcg/mL diluted solution maintained 90% stability for 158, 105, and 53 days, respectively.

Back to Article Outline

Comment 

The chemical stability of an IT drug is important for the maintenance of IT therapy because drugs with lower chemical stabilities may require more frequent pump refills. The original recommended refill intervals in the ziconotide prescribing information are conservative estimates. Our findings incorporate data from as many as 42 different pumps for each ziconotide solution and indicate that the stability of ziconotide exceeds the refill intervals recommended in the current ziconotide prescribing information. Interpretation of this finding is limited by the in vitro nature of the studies, and clinical factors unrelated to the chemical stability of a ziconotide preparation also must be considered when determining the frequency of pump refills.

Back to Article Outline

Acknowledgments 

The authors wish to acknowledge Jennifer Aclan, Matthew Ragusa, and Rob Howard for their contributions to these studies. This work was sponsored by Elan Pharmaceuticals, Inc. All authors were employees of Elan Pharmaceuticals, Inc., during the conduct of the studies.

Back to Article Outline

Reference 

1. PRIALT®. [package insert] San Diego, CA: Elan Pharmaceuticals, Inc.; 2007;
   • View In Article