Jerry Mizell Steven Genovese Jerry Neal Metrics Contract Services
Recently, a Phase I immediate-release solid oral dosage formulation of a pro-drug in gelatin capsules failed to completely dissolve during a three-month accelerated stability study. Scientists at Metrics Contract Services were assigned the task of identifying why the product failed and resolving the issue.
Initial testing
The scientists performed dissolution testing on the product, supplying laboratory personnel with a validated and replicable Phase I dissolution method with high-performance liquid chromatography (HPLC) finish as detailed in Table 1. The initial dissolution testing of the drug product demonstrated full dissolution release at 45 minutes.
Prototype stability study
The scientists then placed samples of the drug product in stability chambers at ambient conditions (25°C/60 percent relative humidity [RH]) and accelerated conditions (40°C/75 percent RH) for a three month prototype stability study, per US FDA and International Conference on Harmonisation guidelines for stability studies.
The scientists completed a dissolution analysis of the stored capsules at the one-month time point and observed that the dissolution release for samples stored in the accelerated conditions was below the expected range. Moreover, samples stored in the accelerated conditions continued to exhibit a significantly slower dissolution release compared to the initial results at all three analysis points during the three-month prototype stability study, with the capsules not fully dissolving, as shown in Figure 1.
For the samples stored in ambient conditions, on the other hand, the dissolution release showed no significant change compared to the initial data, with the capsules fully dissolving (Figure 1).
Based on these results and observations, the scientists suspected that the low dissolution results were a product of crosslinking. Crosslinking describes the process of two or more proteins chemically joining1, 2. It occurs when molecules with two or more reactive sites attach themselves to specific functional groups located on the proteins, linking them together with a covalent bond.
Investigation
The scientists designed an experiment to determine if interactions between the API and excipients were causing the suspected crosslinking and reductions in dissolution release rate. The experiment consisted of filling one sample of gelatin capsules with 60 milligrams of API and another sample with 60 milligrams of a placebo blend. Once filled, these capsules were stored at accelerated conditions (40°C/75 percent RH) for one month. Empty capsule shells were also stored as a control. The scientists then analyzed the stressed capsules per the dissolution procedure previously described.
After 45 minutes, both the empty and placebo-filled capsule shells were fully dissolved. The capsules filled with API were not dissolved at 30 minutes and exhibited signs of crosslinking, as shown in Photo 1. The experiment showed that an API-gelatin interaction was the root cause for the crosslinking.
Literature Review
The scientists initiated a literature review to further understand their observations. They inspected the chemical structure of the API used in the formulation and found a carbamate functional group to be present. The carbamate is comprised of a N-[(Acyloxy)]methyl group, which has the general formula RR’N-CHR’’-OCOR’’’ where R’’ = H or methyl and COR’’’ is the acyl group. This carbamate group is subject to degradation across the ester bridge followed by the chemical breakdown of the N-(hydroxymethyl) intermediate.
The literature review also revealed that formaldehyde is liberated when R’’ = H3. Inspection of this API chemical structure shows that R’’ is a hydrogen atom. The scientists suspected that the API was generating the following hydrolytic degradation products: degraedant 1, degradant 2, carbon dioxide, and formaldehyde. The proposed degradation pathway, as shown in Figure 2, would explain the crosslinking of the capsule shells due to the formation of an aldehyde, which has been shown to cause crosslinking in gelatin capsules4.
To demonstrate that the proposed degradation pathway is accurate, the scientists needed to observe the suspected degradation products in stressed capsules.
Figure 3 shows the amount of degradants 1 and 2 present in capsules stored at accelerated conditions for three months. The data was compiled from assay and impurities testing, with the structures of degradants 1 and 2 being confirmed by comparison to marker preparations for each degradant. Figure 3 also shows the correlation between the degradant levels observed and the dissolution rate for the stressed capsules. As these impurity peaks increased, the percent dissolved in dissolution analysis decreased. From the impurity data, the scientists further suspected the presence of formaldehyde.
To confirm the presence of formaldehyde after stressing, the scientists added approximately 10 milligrams of API to a GC headspace vial and crimped immediately. The vial was placed in an oven at 40°C for one month. The stressed sample and an unstressed control were prepared via the following derivatization procedure inside the headspace vials at a sample concentration of 0.5 mg/mL.
Derivatization procedure
Scientists added 1.0 milliliter of acetonitrile to the vial then prepared a 2,4-dinitrophenylhydrazine stock solution by adding 124 milligrams to a 100-milliliter volumetric flask, adding 50 milliliters of acetonitrile to dissolve, and diluting to volume with acetonitrile. They then added 1.0 milliliter of the 2,4-dinitrophenylhydrazine stock solution and 0.2N hydrochloric acid to the GC headspace vial, mixed the vial, and allowed the mixture to stand for approximately five minutes.
The scientists then added 1.0 milliliter of acetonitrile and 16.0 milliliters of high purity water. The vial was mixed well and allowed to equilibrate to room temperature to derivatize. They prepared a formaldehyde standard at a concentration of 4 μg/mL and analyzed both the standard and the samples using the same HPLC finish described earlier.
The derivatization results showed a significant increase in formaldehyde concentration compared to the control sample, which confirms the generation of formaldehyde during degradation.
Figure 4 shows an overlay chromatogram for standard and sample preparations unstressed and stressed.
Conclusion
As the stress studies showed, degradation of the API liberated formaldehyde, which caused crosslinking between the API and the gelatin capsules in the stressed samples in the drug product stability study.
With this degradation pathway confirmed, the scientists recommended that the formulation be changed from gelatin capsules to hydroxypropyl methyl cellulose (HPMC) capsules to avoid the crosslinking problem.
References
1. https://www.thermofisher.com/us/en/home/lifescience/ protein-biology/protein-biology-learning-center/ protein-biology-resource-library/pierce-protein-methods/ chemistry-crosslinking.html.
2. Thermo Fisher Scientific. “Thermo Scientific Crosslinking Technical Handbook.” Life Science Research. 2012. http://tools.thermofisher.com/content/sfs/bro chures/1602163-Crosslinking-Reagents-Handbook.pdf.
3. Bernard Testa and Joachim M. Mayer. “The hydrolysis of carboxylic acid ester prodrugs.” Hydrolysis in Drug and Prodrug Metabolism: Chemistry, Biochemistry, and Enzymology. Chapter 8. Verlag Helvetica Chimica Acta. 2003. pages 501-503.
4. G. A. Digenis, T. B. Gold, and V. P. Shah. “Crosslinking of gelatin capsules and its relevance to their in vitro-in vivo performance.” Journal of Pharmaceutical Sciences. 1994. Vol. 83, No. 7. pages 915-921.
Jerry Mizell is director of analytical services, Steven Genovese is an analytical chemist II, and Jerry Neal is an analytical chemist III at Metrics Contract Services, a full-service global contract development and manufacturing organization specializing in oral dosage forms.