Abstract
For pharmaceutical parenteral packaging the glass compositions have always been either Type I borosilicate or Type III soda-lime glass. As both the compositions and certain chemical and physical properties are mandated by international standards, there has not been room for any changes. However, by applying only minor adjustments, a borosilicate glass was developed that showed improved chemical stability. The chemical composition is still in the range of currently used borosilicate glasses, which makes it a Type I glass according to all current pharmacopeia. A study was performed on glass vials comparing the new glass with the standard FIOLAX® and two other publicly available glasses. In an extraction study with water at 121 °C the new glass showed the highest chemical stability with the lowest amount of extractables. In an accelerated ageing study, which was done with water, phosphate, and carbonate buffer at 40 °C for 12 months, the new glass also proved to have the lowest amount of leachables. In this article the new glass and the results from the studies are presented, showing the reader how much of an effect can be attained with only minor adjustments if the scientific fundamentals are clear.
LAY ABSTRACT: The pharmaceutical market has been quite constant and risk-oriented due to the high impact on the safety of the patient. As any change necessitates a complicated change process, this has, in consequence, lead the industry to resist changing the parenteral primary packaging material for decades. The main glasses have either been Type I borosilicate or Type III soda-lime glass. On the other hand, a combination of improved inspection systems and the development of more sensitive biologically based drugs has elevated the standards for parental packaging materials. For example, the measurement of extractables and leachables from the packaging material steadily came into focus. In this article, a new glass is presented that still belongs to the group of Type I borosilicate glasses according to all pharmacopeia. However, with some minor adjustments in the chemical composition it was possible to increase the chemical stability measurably. To prove this several studies were performed, of which the extraction study with water at 121 °C and the accelerated ageing study with water, phosphate, and carbonate buffer at 40 °C for 12 months are presented here.
- Glass
- Primary packaging
- Borosilicate glass
- Chemical stability
- Extractables and leachables
- Accelerated ageing study
- Phosphate buffer
- Carbonate buffer
Introduction
The Current Situation and Motivation
In pharmaceutical primary packaging there are two glass compositions that are used: borosilicate glass and soda-lime glass. The compositions of these glasses are precisely defined in various standards: for example, the American Society for Testing and Materials (ASTM) E438-92 (1) or International Organization for Standardization (ISO) 12775 (2). In these standards, both the chemical composition and certain relevant physical properties are described for Type I (borosilicate) and Type III (soda-lime) glasses. The Type I glasses are again substructured according to their coefficient of mean linear thermal expansion (CTE or α). Glasses with a CTE of around 3.3·10−6 K−1 (20 °C; 300 °C) belong to the 3.3 group, glasses with a CTE of around 4.9–5.5·10−6 K−1 (20 °C; 300 °C) belong to the 5.0 group, and glasses with a CTE of around 6.3–7.5·10−6 K−1 (20 °C; 300 °C) belong to the 7.0 group.
The chemical compositions shown in Table I represent the range of borosilicate 3.3, 5.0, and 7.0 groups and soda-lime-silica clear glasses currently used in the pharmaceutical industry. Additionally, the CTE of glasses as they are mainly found on the market is given. These data result from long years of experience based on the analysis of samples from the market.
Overview of Currently Used Borosilicate and Soda-Lime Glasses with Their Corresponding Classification in Established Standards and Pharmacopeia, Their Chemical Composition, and Their Physical Properties
According to the current pharmacopeia, glass containers for pharmaceutical use are intended to come into direct contact with pharmaceutical products. As more and more newly developed drugs are quite sensitive toward changes in their chemical environment, regulatory agencies have started setting a stronger focus on drug-container interactions. These interactions are significantly dependent on the glass composition, the converting process, and the drug product. The latter two factors are not the topic of this article, which will focus rather on the composition of the glass packaging itself.
The suitability of a glass as a primary packaging material can be evaluated by performing accelerated ageing studies. Hereby the final drug product is stored for a certain amount of time (e.g., 6, 12, or 18 months) at elevated temperatures (e.g., 40 °C, depending on the drug product's original storage conditions) whereas in parallel the intactness of the drug product is monitored. Along the way the chemical stability of the glass can also be evaluated, for example by measuring the glass elements being dissolved in the drug solution, that is, the amount of extractables and leachables.
As the reactions of the drug solution with the glass material vary depending on the chemical nature of the drug product—mainly the buffer solution—it is crucial to perform the studies with the original formulation.
By using glass as a primary packaging material, such studies are significantly facilitated, owing to the fact that glass has only a limited number of components and the composition is known and is publicly available. However, to what extent the various elements are released from the glass is dependent on different factors. The glass matrix is a quite complicated interplay between the oxides that are used. This interplay defines the strength of the silicate network and the chemical and physical properties. Minor compositional changes can have a significant impact on the chemical and physical properties. But as the pharmaceutical industry is highly regulated, and changes in composition are tied to a complicated change management process, the composition of the most widely used Type I glasses stayed rather constant over the years. Generally, there are two ways to develop a different glass packaging material. One way is by developing a completely new glass composition. This requires a higher number of studies than for already existing glass compositions. The other way is to make minor adjustments to an established glass composition, while obtaining a substantial impact on chemical resistance.
Description of the Newly Developed Glass
In this article, a new glass composition (borosilicate Type I according to current USP and European Pharmacopoeia) (3, 4) is presented. Due to an optimized interplay between the elements, it offers greater chemical stability and a lower total extractable/leachable level under the described conditions. It was discovered, for instance, that the potassium concentration in a glass composition contributes to a higher chemical stability within a very narrow range: the range of 0.1 to 0.9 weight percent potassium oxide resulted in the highest hydrolytic resistance (higher chemical stability in terms of water attack). The reason for this is as follows: Alkali ions (such as lithium, sodium, potassium, etc.) occupy the interstitial sites within the silicate network, thus “closing” the structure. Sodium ions occupy the smaller gaps whereas potassium ions—due to their larger ionic radius—occupy the larger gaps in the glass structures. In this quite sensitive balance, the art is to find the optimum ratio of sodium and potassium. If the potassium content is too high, the silicate network is dilated, which leads to a facilitated release of the smaller ions from the network. The principle of the optimum ratio is of course valid for all elements but will not be focused on further in this article. To further enhance the chemical durability of the glass, barium is not used as it has been shown to precipitate with certain drug product solutions. A more detailed characterization can be found in U.S. Patent 20140323287 A1 (dated October 30, 2014) (5).
Out of the leading global glass tubing producers for pharmaceutical applications, three were chosen for comparison with the new glass development. In Table II, the composition of the glasses used herein is shown. Glass A is the newly developed glass, whereas B represents FIOLAX®, while C and D are comparative neutral borosilicate glass compositions currently used for pharmaceutical applications.
Glass Composition of the Newly Developed Glass (Glass A), FIOLAX® Glass (Glass B), and Two Comparative Neutral Borosilicate Glasses (Glass C and Glass D); All Glasses Were Characterized by Using X-Ray Fluorescence Spectroscopy
Description of the Studies Performed
From the number of studies that were performed on this glass, a few representative ones were selected to be discussed here, especially with a focus on the extractables from an extraction study and the leachables from accelerated ageing studies.
The extraction study with 1 h autoclaving at 121 °C was chosen because autoclaving is a standard terminal sterilization method used in parenteral packaging processing. Furthermore, autoclaving is a quite aggressive procedure on the glass, which quickly reveals the chemical stability of the inner glass surface (it should be kept in mind that this is also altered during the converting process). The concentration of extractables is important as they can enhance possible interactions with the drug. If the drug is sensitive toward any changes in its chemical environment, the amount of extractables should be kept as low as possible.
For the accelerated ageing studies, water at pH 5.5 was chosen in addition to commonly used buffers like phosphate at pH 7.0 and carbonate at pH 8.0. With this selection, a whole range of pH values, from acidic to alkaline, was covered to test the chemical stability under a variety of possible realistic conditions (Table III).
Overview of Studies and the Corresponding Conditions
Materials and Methods
Water: Ultrapure water, pH 5.5, received from a Sartorius atrium® pro UV/DI water purification system (filter 0.2 μm, 18.2 MΩ·cm, 25 °C).
Carbonate: 8.4% sodium bicarbonate (Sigma Aldrich, Germany) solution, pH 8.0.
Phosphate: 10 mM phosphate (Sigma Aldrich, Germany) solution, pH 7.0 with 150 mM NaCl solution and 0.005% (v/v) Tween 20 (Sigma Aldrich, Germany) solution.
Stoppers: WEST 4110/40 Westar RS B2-40 Fluro-Tec-Plus; autoclaved at 121 °C for 30 min.
Autoclave: A Systec Type DX 150 autoclave was used to follow the procedure described in the European Pharmacopeia 8.2, chapter 3.2.1 inner surface test, 1 h at 121 °C.
Inductively coupled plasma mass spectrometry (ICP-MS): Agilent Type 7500ce; a repeat determination with the solution of 5 pooled vials was done for each measurement point.
Inductively coupled plasma optical emission spectrometry (ICP-OES): Agilent Type 725; a repeat determination with the solution of 5 pooled vials was done for each measurement point.
Measured elements: For both the extraction and the accelerated ageing studies the following elements were measured: ICP-MS: Mg, Al, K, Ca; ICP-OES: Ba, Na, Si.
Production of vials: Glasses A–D were used with compositions according to Table II. Glass tubing was converted into 2R vials (outer diameter 16 mm, wall thickness 1.0 mm) on the same converting machines using comparable machine settings according to the physical requirements of the individual glasses.
Preparation of samples: 470 2R vials of each glass type were washed three times with hot tap water, three times with ultrapure water (see Materials and Methods: Water), and finally dried with filtered nitrogen. They were filled with 2 mL sample solution (water, phosphate, carbonate) and stored for 12 months at 40 °C without humidity control. Samples were taken and measured after 1 day (18 h), 3 months, 6 months, and 12 months. An additional set of 13 2R vials filled with ultrapure water was autoclaved at 121 °C for 1 h and subsequently measured for extractables in the solution.
Results and Discussion
Extraction Study
In Figure 1, the results of the extraction study with water are shown. It can clearly be seen that with 4 ppm (mg/L), Glass A offers the lowest amount of total extractables in the solution, followed by Glass B with 7 ppm. Glasses C and D exhibit the highest amount with 10 ppm and 14 ppm. Glass D releases a three times higher amount of extractables into the solution than does Glass A. When packaging, for example water for injection, a terminal sterilization at 121 °C is the usual method of choice. A more detailed breakdown is given in Figure 2 where the individual glass elements are shown per glass type.
Extraction results of Glasses A, B, C, and D with water (pH 5.5) and autoclaving for 1 h at 121 °C; extractables measured: Mg, Al, K, Ca, Ba, Na, and Si by ICP-MS and ICP-OES respectively and given in milligrams per liter (mg/L, ppm); measurement uncertainty: 10%.
Extraction results on an individual element basis of Glasses A, B, C, and D with water (pH 5.5) and autoclaving for 1 h at 121 °C; extractables measured: Mg, Al, K, Ca, Ba, Na, and Si by ICP-MS and ICP-OES respectively and given in mg/L (ppm); y-axis truncated at 4 mg/L for a clearer representation of the lower concentrated elements; measurement uncertainty: 10%.
As visible silicon forms the biggest part of the extracted elements, it represents the major component of the glasses.
Accelerated Ageing Study
In Figure 3, the total extractable elements at 40 °C in ultrapure water, pH 5.5, is shown. Throughout the 12 months of storage Glass A offered the lowest amount of leachables with a maximum of around 4 ppm (mg/L) after 12 months. Glass D exhibited the highest amount with around 9 ppm. As the reaction of water with glass is an ion exchange reaction where the alkali and alkaline earth ions from the glass are exchanged for hydrogen ions from the water, it is not surprising that, for example, sodium makes a very high portion of the leached elements (data not shown).
Results from the accelerated ageing study of Glasses A, B, C, and D with water (pH 5.5) and storage at 40 °C for up to 12 months; elements measured: Mg, Al, K, Ca, Ba, Na, and Si by ICP-MS and ICP-OES respectively and given in mg/L (ppm); measurement uncertainty: 10%.
Figure 4 shows the total extracted glass elements in phosphate solution. Glasses A and B consistently present lower amounts of total released glass elements with around 10–11 ppm (mg/L). After 12 months of storage, Glasses C and D show higher amounts (with more than 14 ppm [mg/L]) of elements measured in the solution. Phosphate solution reacts with the inner surface of the glass, chelating especially aluminum, which is why aluminum is measured only to a very low extent in the solution (data not shown).
Results from the accelerated ageing study of Glasses A, B, C, and D with phosphate buffer (pH 7.0) and storage at 40 °C for up to 12 months; elements measured: Mg, Al, K, Ca, Ba, Na, and Si by ICP-MS and ICP-OES respectively and given in mg/L (ppm); measurement uncertainty: 10%.
In Figure 5, it is apparent that Glass A has the lowest concentration of leached glass elements, with a maximum of around 14 ppm, whereas Glass D has the highest level with approximately 26 ppm. It can be seen that the sum of leached elements in carbonate buffer is higher than the one in phosphate buffer or water. In water the maximum amount was around 9 ppm, in phosphate buffer it was around 14 ppm, whereas in carbonate buffer a concentration of around 26 ppm was measured. This indicates the different chemical reactions with aqueous solutions. At higher pH values the alkaline solution attacks the glass surface and dissolves the silicate network. The result is a higher amount of all glass components in the solution. With an acidic aqueous solution, an ion exchange takes place, resulting in a higher concentration of alkali and alkaline earth metals in the solution.
Results from the accelerated ageing study of Glasses A, B, C, and D with carbonate buffer (pH 8.0) and storage at 40 °C for up to 12 months; elements measured: Mg, Al, K, Ca, Ba, Na, and Si by ICP-MS and ICP-OES respectively and given in mg/L (ppm); measurement uncertainty: 10%.
While still being within the compositional range of Type I borosilicate glasses, it was possible to develop a glass composition with an increased chemical stability toward water attack at an acidic pH level, attack from phosphoric acid at a neutral pH of 7.0, and carbonic acid with an alkaline pH value of 8.0.
The advantage of not exceeding the compositional ranges for Type I glass is that fewer compatibility studies are required to gain regulatory approval. However, the final risk assessment lies in the hands of the pharmaceutical company and differs from company and drug to company and drug.
Conflict of Interest Declaration
The authors declare that they have no competing interests.
- © PDA, Inc. 2016
Jump to section
Related Articles
Cited By...
- Assay for characterizing adsorption-properties of surfaces (APS) used for sample preparation prior to quantitative omics
- Materials in Manufacturing and Packaging Systems as Sources of Elemental Impurities in Packaged Drug Products: An Updated Literature Review
- Investigating the Effects of the Chemical Composition on Glass Corrosion: A Case Study for Type I Vials
- Assessment of Extractable Elements from Elastomers
- Historical Review of Glasses Used for Parenteral Packaging