Research ArticleResearch

Methods To Determine the Silicone Oil Layer Thickness in Sprayed-On Siliconized Syringes

Viviane Loosli, Oliver Germershaus, Henrik Steinberg, Sascha Dreher, Ulla Grauschopf and Stefanie Funke
PDA Journal of Pharmaceutical Science and Technology May 2018, 72 (3) 278-297; DOI: https://doi.org/10.5731/pdajpst.2017.007997

Abstract

The silicone lubricant layer in prefilled syringes has been investigated with regards to siliconization process performance, prefilled syringe functionality, and drug product attributes, such as subvisible particle levels, in several studies in the past. However, adequate methods to characterize the silicone oil layer thickness and distribution are limited, and systematic evaluation is missing. In this study, white light interferometry was evaluated to close this gap in method understanding. White light interferometry demonstrated a good accuracy of 93–99% for MgF2 coated, curved standards covering a thickness range of 115–473 nm. Thickness measurements for sprayed-on siliconized prefilled syringes with different representative silicone oil distribution patterns (homogeneous, pronounced siliconization at flange or needle side, respectively) showed high instrument (0.5%) and analyst precision (4.1%). Different white light interferometry instrument parameters (autofocus, protective shield, syringe barrel dimensions input, type of non-siliconized syringe used as base reference) had no significant impact on the measured average layer thickness. The obtained values from white light interferometry applying a fully developed method (12 radial lines, 50 mm measurement distance, 50 measurements points) were in agreement with orthogonal results from combined white and laser interferometry and 3D-laser scanning microscopy. The investigated syringe batches (lot A and B) exhibited comparable longitudinal silicone oil layer thicknesses ranging from 170–190 nm to 90–100 nm from flange to tip and homogeneously distributed silicone layers over the syringe barrel circumference (110– 135 nm). Empty break-loose (4–4.5 N) and gliding forces (2–2.5 N) were comparably low for both analyzed syringe lots. A silicone oil layer thickness of 100–200 nm was thus sufficient for adequate functionality in this particular study. Filling the syringe with a surrogate solution including short-term exposure and emptying did not significantly influence the silicone oil layer at the investigated silicone level. It thus appears reasonable to use this approach to characterize silicone oil layers in filled syringes over time. The developed method characterizes non-destructively the layer thickness and distribution of silicone oil in empty syringes and provides fast access to reliable results. The gained information can be further used to support optimization of siliconization processes and increase the understanding of syringe functionality.

LAY ABSTRACT: Silicone oil layers as lubricant are required to ensure functionality of prefilled syringes. Methods evaluating these layers are limited, and systematic evaluation is missing. The aim of this study was to develop and assess white light interferometry as an analytical method to characterize sprayed-on silicone oil layers in 1 mL prefilled syringes. White light interferometry showed a good accuracy (93–99%) as well as instrument and analyst precision (0.5% and 4.1%, respectively). Different applied instrument parameters had no significant impact on the measured layer thickness. The obtained values from white light interferometry applying a fully developed method concurred with orthogonal results from 3D-laser scanning microscopy and combined white light and laser interferometry. The average layer thicknesses in two investigated syringe lots gradually decreased from 170–190 nm at the flange to 100–90 nm at the needle side. The silicone layers were homogeneously distributed over the syringe barrel circumference (110–135 nm) for both lots. Empty break-loose (4–4.5 N) and gliding forces (2–2.5 N) were comparably low for both analyzed syringe lots. Syringe filling with a surrogate solution, including short-term exposure and emptying, did not significantly affect the silicone oil layer. The developed, non-destructive method provided reliable results to characterize the silicone oil layer thickness and distribution in empty siliconized syringes. This information can be further used to support optimization of siliconization processes and increase understanding of syringe functionality.

1. Introduction

Over the last decades, prefilled syringes (PFSs) have been increasingly used as drug delivery formats for therapeutic proteins, which need to be administered parenterally—for example, subcutaneously (e.g., insulin, somatropin, adalimumab) and intravenously (e.g., granulocyte-colony stimulating factor, epoetin alpha) (1⇓⇓⇓⇓⇓⇓–8). PFSs show several advantages compared to vials and ampoules, for which additional handling steps are required by the end user prior to injection. Patients and health care professionals benefit from high dosage accuracy, easy and fast handling (particularly in case of emergencies), and a lower risk of product contamination (1⇓⇓⇓–5, 9⇓–11). However, the combination of storage container and device functions within PFSs results in a more complex primary packaging system compared to vials and ampoules. To ensure optimal functionality during injection (2, 9, 12⇓–14), the inner surface of the glass syringe barrel is coated with silicone oil. PFS performance is typically assessed by monitoring of break-loose and gliding forces required to move the plunger through the syringe barrel (2, 4, 13).

Silicone oil is applied to the inner syringe barrel surface either by spray-on or bake-on siliconization processes. Spray-on siliconization is typically used for PFSs with staked-in needle where a temperature-sensitive adhesive is used to fix the needle into the glass barrel tip. In such cases silicone oil—in the form of dimethicone (13) or silicone oil used as lubricant (15)—is used as spray medium (12, 13). Silicone oil used for PFS lubrication typically has a viscosity of 1000 centistokes (25 °C) and a refractive index of 1.405 (16). Sprayed-on silicone oil levels described in the literature were between 0.2 and 1.0 mg/barrel, and reported silicone oil layer thicknesses ranged from 100 to 700 nm (9, 13, 14, 17). Bake-on siliconization can be performed by spraying a diluted silicone oil emulsion onto the inner syringe barrel followed by heat treatment. Bake-on siliconization is limited to temperature-insensitive systems such as luer tip syringes and cartridges. Compared to spray-on siliconized syringes, substantially lower target silicone oil levels of below 0.1 mg/barrel with resulting silicone oil layer thicknesses below 80 nm are typically applied in the case of baked-on siliconization (9, 18).

Independent of the siliconization method, the silicone oil layer may interact with the therapeutic protein formulation (21). Silicone oil droplet migration has been associated with an increase in turbidity and subvisible particle levels (1, 2, 21, 22), thus negatively affecting important drug product quality characteristics. Additionally, inhomogeneous silicone oil layers, especially with an insufficient siliconization towards the needle side, were shown to induce stalling of the plunger and incomplete drug product injection (23), potentially resulting in failure of the application device if the syringe plunger does not move steadily through the syringe barrel (2). Incomplete injections thus represent an unacceptable failure of PFSs (18). Hence, the silicone oil layer thickness and its homogeneity not only affect drug product quality characteristics but are also important functionality-related characteristics of PFSs.

The currently available methods to investigate the silicone oil layer thickness and distribution in syringe barrel are limited, and systematic evaluation is missing. The following paragraph provides a brief overview of the available techniques. A simple but destructive method to visually assess the silicone oil distribution is the powder method. Glass, talcum, or aluminum oxide powder sticks to the silicone oil–coated glass surface areas provided that the silicone oil layer is sufficiently thick (17, 24, 25). By this means, the distribution of silicone oil can be roughly assessed. Time of flight secondary ion mass spectrometry yields layer thickness information as low as 10 nm, but only small areas (150 × 150 μm) can be characterized with high technical effort (10, 26). 3D-laser scanning microscopy (3D-LSM) is able to non-destructively visualize thin silicone layers. Thickness measurements down to 10 nm are feasible, but analysis is time-consuming and destructive (9). The two latter methods can be used to characterize layer thickness distribution along the syringe barrel if multiple areas are analyzed individually. Schlieren optics is used to visualize sprayed-on silicone oil layers: Inhomogeneities are qualitatively detected by the formation of silicone oil droplets upon contact with aqueous media (9, 27). Testing of break-loose and gliding forces indirectly characterizes the lubrication quality by verifying syringe functionality (17, 19, 28).

All the abovementioned methods are challenging to implement for routine analysis of silicone oil layer thickness and layer thickness distribution due to high time effort, destructive sample preparation, highly complex instrumentation, or limited sensitivity, or a combination thereof. The aim of this study was therefore to develop and assess a method based on white light interferometry (WLI) for the analysis of sprayed-on silicone oil layers in PFSs. White light (λ = 400–800 nm, halogen light source) is perpendicularly sent through the syringe barrel where it is reflected from both the glass–coating and coating–air interfaces. The superposition of the reflected light leads to a typical interference spectrum that depends on the layer thickness and the wavelength. The spectra are detected by a photodiode detector and mathematically converted into the corresponding layer thickness (3, 19, 29). The measurement spot has a diameter of approximately 45 μm, and the detected thickness range is 80–4000 nm. Using an automated sample holder, multiple spots can be measured. To summarize, WLI is a non-destructive and quantitative method (19) providing information about the radial and longitudinal silicone oil layer distribution (18). In addition, the measurement range can be extended to cover layer thicknesses down to 20 nm by applying a laser light source (λ = 660 nm) (30). The main influencing parameters in laser interferometry (LAI) measurements are the refractive indices of the glass and the silicone oil layer, where the measurement sensitivity is increased with an increased difference in refractive indices (29, 31). Combined white light and laser interferometry (WLI+LAI) eventually covers layer thicknesses from 20 nm up to 4000 nm (31). For further details on the measurement principle of WLI, the reader is referred to the paper by Lankers (19).

The developed WLI method was challenged in terms of its analytical properties such as accuracy, instrument, and analyst precision. The adequate setting of associated method parameters such as baseline, autofocus (AF), protective shield position, and entered syringe barrel dimensions was also investigated. WLI was additionally compared to 3D-LSM and a combination of WLI+LAI. The developed method was further used for comprehensive syringe batch characterization and for evaluating the influence of syringe filling on the silicone oil layer thickness.

2. Materials and Methods

2.1. Materials

Measurements were performed with two batches (lot A and lot B) of spray-on siliconized 1 mL long PFSs (barrel length 54 mm, inner diameter 6.35 mm, outer diameter 8.15 mm) made of Fiolax clear® glass with a target silicone oil level of 0.5 mg/barrel. Non-siliconized syringes with dimensions comparable to the siliconized syringes and different glass types (Fiolax clear® glass, Borofloat® 33) were used for baseline measurements.

The surrogate solution contained 20 mM L-histidine/L-histidine HCl, 240 mM sucrose, 10 mM L-methionine and 0.05% (w/v) polysorbate 20, pH 5.5. L-histidine, L-histidine-HCl, and methionine were purchased from Ajinomoto Inc. (Tokyo, Japan), sucrose from Pfanstiehl GmbH (Zug, Switzerland), and polysorbate 20 from Croda International (Snaith, United Kingdom). The buffer was filtered through a 0.2 μm polyvinylidene fluoride membrane. Water for chromatography from Merck Millipore (Darmstadt, Germany) was used to rinse the syringes after emptying.

2.2. White Light Interferometry (WLI) Method Development

Three different syringes (syringe I, syringe II, syringe III) from lot A with typical silicone oil distribution patterns were analyzed during method development. Syringe I showed a homogeneous average layer thickness (ALT) over the entire measured line along the barrel length. The silicone oil layer in syringe II was thicker at the flange and decreased towards the needle, while in syringe III the ALT sharply increased in the middle of the barrel and therefore yielded thicker silicone oil layers at the needle side.

The silicone oil layer thickness of the three syringes I, II, and III was determined by WLI (Layer Explorer, rap.ID Particle Systems GmbH, Berlin, Germany). The number of radial lines, angle of rotation, measuring length, and the measurement points per line were the main method inputs. Non-siliconized syringes with dimensions and glass type (Fiolax clear®) comparable to the siliconized syringes were used for base (“setup base” software feature) and base control measurements [performed with one line, 0°, 50 mm, 50 measurement points, all values below the limit of detection (LOD) of 80 nm].

2.2.1. Instrument Accuracy over Time:

Four curved, magnesium fluoride–coated glass standards with certified layer thicknesses ranging from 115 ± 1 to 473 ± 9 nm were analyzed weekly in order to assess the instrument accuracy over time (“standard base” and “standard one point” software features were used).

2.2.2. Analyst and Instrument Precision:

One measurement line on the barrel circumference (one line, 0°, 50 mm) with 50 measurement points along the syringe barrel length from the flange towards needle side were measured five consecutive times for each syringe to evaluate the instrument precision. The analyst precision was assessed by analyzing the same line (one line, 0°, 50 mm, 50 measurement points) five times while the syringes were newly loaded onto the sample holder before each new line measurement.

2.2.3. Impact of Autofocus (AF), Protective Shield (PS), and Syringe Barrel Dimensions:

The silicone oil layer thickness (one line, 0°, 50 mm, 50 measurement points) was measured with and without AF by an experienced user, with open and with closed protective shield (PS) and with four different instrument methods (A, B, C, and D), covering the upper and lower range of the inner and outer diameter specified for the syringe barrel. Those results are for information only and are only depicted in the supporting information (Figure S1, supporting information). They will not be further discussed because the measured ALTs were mainly comparable. Therefore, further measurements were consistently performed with AF, with open PS, and with nominal syringe diameters.

Figure S1
Figure S1

ALT per sector of the syringes I, II, and III measured (a) with AF and (b) with PS closed. ALT per sectors with different syringe barrel dimensions inputs (method A–D) for syringe I (c), II (d), and III (e). LOD values were replaced by 70 nm (n = 1).

2.2.4. Baseline Impact:

Different types of non-siliconized syringes (made of Fiolax clear® glass, Borofloat® 33 glass, and de-siliconized syringes) were set as the base followed by measuring the silicone oil layer thickness (one line, 0°, 50 mm, 50 measurement points). Those results are only depicted in the supporting information (Figure S2, supporting information) and will not be further discussed because the measured ALTs were independent from the type of non-siliconized syringe used for base measurements.

Figure S2
Figure S2

ALT of the syringes I (a), II (b), and III (c) measured after different baselines. LOD values were replaced by 70 nm (n = 1, *P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.001).

Syringes were de-siliconized by extraction with heptane from VWR International GmbH (Darmstadt, Germany): The syringes were closed using a rubber stopper at the needle side and filled 10 times with heptane. After 30 s extraction, the syringes were emptied by pouring the heptane over the flange side or alternatively by removing the rubber stopper from the needle.

2.2.5. Number of Radial Lines and Measurement Points:

The number of radial lines ranged from three to 16 (50 mm, 50 measurement points) to determine the number of radial lines necessary to sufficiently describe the radial silicone oil distribution in the syringe barrel (for different angles of rotation please refer to graphs in the Results section). The minimal required number of measurement points per line was determined by analyzing eight radial lines with between five and 100 measurement points each.

2.2.6. Head-to-Head Comparison: 3D-Laser Scanning Microscopy (3D-LSM):

Two chrome-coated step height standards of 17.3 ± 1.6 mm and 89.4 ± 1.2 mm from John P. Kummer GmbH (Augsburg, Germany) were used to verify the accuracy of the laser microscope. Five randomly selected siliconized syringes from the abovementioned lot A were analyzed by WLI (12 lines; 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, 330°; 50 mm; 50 measurement points) and subsequently cut lengthwise into halves by a water-cooled diamond wire from a Well Vertical Diamond Wire Saw 3500, Well diamond wire saws (Le Locle, Switzerland).

The ALT along the syringe barrel was determined using a VK-X210 laser scanning microscope from Keyence Deutschland GmbH (Neu Isenburg, Germany) and applying a previously described 3D-LSM method (9) at five points per syringe, one point per 10 mm barrel length.

2.2.7. Head-to-Head Comparison: Combination of White Light and Laser Interferometry (WLI+LAI):

Four curved, half-coated (MgF2) glass standards with certified layer thicknesses ranging from 44 ± 3 nm to 540 ± 13 nm were analyzed by WLI+LAI [Layer Explorer Ultra Thin (UT) 2.0, rap.ID Particle Systems GmbH, Berlin, Germany] to assess the instrument accuracy. Standards were measured with two lines, 0°, 10 mm, and 100 measurement points after setting up the base for the non-siliconized standard halves. The LAI raw signal of the corresponding non-siliconized standard halves was determined (one line, 0°, 2 mm, 100 measurement points) and set as fixed UT base (range between 2.37 and 2.40 V for individual standards) prior to measuring the coating thickness. The “set-up base” and “standard one point” software features were used for WLI.

For sample analysis, the LAI raw signal of the corresponding 1 mL non-siliconized syringes was determined in triplicate with one line, 0°, 2 mm, 100 measurement points and was subsequently used as the fixed UT base (2.52 ± 0.07 V). The refractive indices (RIs) were set to RI 1.485 for Fiolax® glass (based on recommendations from rap.ID Particle Systems GmbH) and to RI 1.404 for silicone oil (16).

A control measurement of the corresponding non-siliconized syringes was performed before and after sample analysis (one line, 0°, 20 mm, 20 measurements points). The fixed UT base was considered valid if the measured ALT for the non-siliconized reference samples was ≤30 nm [LOD and not defined (“n.d.”) values were assumed to be 20 nm as worst case]. The ALT determined for the non-siliconized syringes was 22 ± 3 nm and thus confirmed a theoretical LOD of approximately 20 nm for the WLI+LAI method in this particular study. Sample measurements (one line, 0°, 45 mm, 45 measurements points) were only analyzed if the number of LOD and n.d. values was ≤20%; otherwise measurements were rejected. Samples were randomly selected from lot A (n = 5).

2.3. Syringe Batch Characterization

2.3.1. Silicone Layer Thickness:

The ALT of 20 randomly selected syringes—10 per batch (lot A and lot B with the abovementioned characteristics) —was analyzed by WLI (12 lines, 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, 330°, 50 mm, 50 measurement points).

2.3.2. Mechanical Performance:

The same syringes used for silicone layer thickness determination (section 2.3.1) were assembled with syringe plungers using a semi-automatic stoppering machine from Bausch + Ströbel®, Maschinenfabrik Ilshofen GmbH + Co. KG (Ilshofen, Germany). A plunger position of 10 mm below the finger flange (in order to cover a maximum barrel length) was used for five syringes, and a position of 24 mm below the finger flange (typical plunger position in filled 1 mL syringes) was used for the five remaining syringes of each batch.

Break-loose and gliding forces in the empty syringes were analyzed using tensile/compression force equipment from Zwick GmbH & Co. KG (Ulm, Germany). The compression rate was set to 100 mm/min, and the measurement endpoint was 50 N, that is, when the plunger reached the needle side.

2.4 Impact of Syringe Filling and Emptying

Four randomly selected syringes from lot A were analyzed by WLI (12 lines, 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, 330°, 50 mm, 50 measurement points) and subsequently filled with 1.40 mL surrogate buffer. Syringe plungers were manually assembled close to the finger flange (plunger edge was on the same level as finger flange), and air was carefully expelled through the needle. The plunger was not inserted more than absolutely necessary to maintain the longest possible measurement distance for WLI. The filled syringes were measured by WLI and immediately afterwards emptied through the flange side. Filling time did not exceed 1 h. The emptied syringes were immediately rinsed three times with 1.40 mL water to remove buffer residues and finally air-dried. The emptied syringes were then analyzed by WLI.

The filled syringes did not yield plausible results (data not shown), which can be attributed to the currently available equipment set-up with limited applicability of the mathematical algorithm to calculate layer thicknesses in a filled syringe. Due to the mechanistic influence of stopper placement and removal, the first 10 measurement points (sector 1) were not considered for data analysis.

2.5 Statistical Evaluation

Comparisons between results were conducted using a two-sample Student's t-test using GraphPad Prism 6. Significant differences were labelled as follows: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

3. Results

3.1. White Light Interferometry (WLI) Method Development

3.1.1. Instrument Accuracy over Time:

Four curved, MgF2-coated standards were analyzed weekly over a period of 4 weeks. The layer thicknesses were consistently measured over time—standard deviation (SD) between 0 and 1 nm—and comparable to the certified values with accuracies ranging from 93 to 99%, as shown in Table I. The correlation coefficient R2 of the calibration curve amounted to 1.000.

View this table:
TABLE I

Measured MgF2-Coated Standards Compared to the Certified Layer Thicknesses (n = 20)

3.1.2. Analyst and Instrument Precision:

In order to assess the analyst precision, one line (0°, 50 measurement points) was measured five consecutive times while the syringes were newly loaded onto the sample holder before each measurement. The layer thickness determined at each measurement point (Figure S3a, supporting information) contributes to the mean layer thickness along the syringe barrel. The SD between these mean values of the five measured lines was 2–5 nm [relative standard deviation (RSD) from 1.7 to 4.1%].

Figure S3
Figure S3

Determination of (a) analyst precision and (b) instrument precision using ALT of syringe I, II, and III per measurement point with all LOD values replaced by 70 nm (n = 5).

Sectors have been introduced to improve data interpretation: Each sector represented the average of the measurement points of 1 cm barrel length. If 50 points were analyzed per barrel, a sector thus represented 10 measurement points. Sector 1 described the ALT at the flange, while sector 5 depicted the ALT at the needle side. Based on the 3D-LSM results (see discussion in 4.1.4), all values below the LOD were replaced by 70 nm (Figure 1).

Figure 1
Figure 1

Determination of analyst precision. ALT of syringe I, II, and III per sector (n = 5) with all LOD values replaced by 70 nm.

To assess instrument precision, the same line (0°, 50 measurement points) was measured five consecutive times (layer thickness for individual measurement point Figure S3b, supporting information). The ALTs of the five measured lines showed a SD of 0 to 1 nm, corresponding to RSDs between 0.1 and 0.5% (Figure S3b, supporting information).

Significant differences in sectors 4 and 5 for both syringe I and II between the measurements for analyst and instrument precision were observed (Figure 1 and Figure 2a): ALTs of syringes I and II fell below or exceeded the LOD in sectors 4 and 5. The WLI microscope revealed an inhomogeneous micro-structure of silicone layers with round micro-droplets (Figure 2b).

Figure 2
Figure 2

Determination of instrument precision. (a) ALT of syringe I, II, and III per sector (n = 5) with all LOD values replaced by 70 nm. (b) Representative images per sector of the 0° line for syringes I, II, and III; the scale bar applies to all images.

3.1.3. Number of Radial Lines and Measurement Points:

Syringes I, II, and III were analyzed with three to 16 radial lines and five to 100 measurement points per line to identify the minimum number of lines and measurement points required for representative measurements. ALTs were calculated per sector to assess the longitudinal silicone oil distribution along the syringe barrel. ALTs per measurement line characterized the radial silicone oil distribution over the syringe barrel circumference. Multiple radial lines were required to fully describe the silicone oil distribution on the glass barrel due to the inhomogeneous micro-structure observed in the microscopic images (Figure 2b).

The mean polygon areas yielded in the different experiments (three to 16 radial lines measured) were compared (Table II, Figure 3a–c). Significant differences between eight and 12 lines were found, indicating that eight lines were not able to sufficiently describe the radial silicone oil distribution. However, a greater number of radial lines, for example, 16, did not yield significantly different results regarding the radial silicone oil distribution. No significant differences between the ALTs per sector could be seen, no matter whether five or 100 points were measured along the syringe barrel length (Figure S4, supporting information).

Figure S4
Figure S4

Determination of number of measurement point. Longitudinal silicone oil distribution shown as ALT per sector of syringe I (a), II (b), and III (c) and derived from measurements with eight lines, and five to 100 points per line (n = 1). All LOD values were replaced by 70 nm.

View this table:
TABLE II

Determination of the Number of Radial Lines. Comparison of the Mean Theoretical Relative Polygon Areas Derived from Polygons for Syringe I, II, and III Shown in Figure 3a–c

Figure 3
Figure 3

Radial silicone oil distribution shown as ALT per line of syringe I (a), II (b), and III (c) measured with three to 16 lines (n = 1) and all LOD values replaced by 70 nm. Legend indicates number of measured radial lines.

3.1.4. Head-to-Head Comparison to 3D-Laser Scanning Microscopy (3D-LSM) and to a Combination of White Light and Laser Interferometry (WLI+LAI):

Chrome-coated step height standards were measured by 3D-LSM, while MgF2 standards were measured by WLI+LAI. The measured layer thicknesses were comparable to the certified values with accuracies ranging from 90% to 98% (Table III). Five syringes were analyzed by WLI with 12 lines and 50 measurement points per line. The ALTs were higher at the flange (approximately 160 nm) than at the needle side (approximately 90 nm) (Figure 4a). The radial silicone oil distribution over the syringe diameter was homogeneous with ALTs per line ranging from 100 to 130 nm (Figure 4b). Both the longitudinal silicone layer thickness along the syringe barrel and the radial ALT determined by WLI, 3D-LSM, and a combined method of WLI+LAI were not significantly different.

View this table:
TABLE III

Measured Chrome-Coated Step Height Standards Compared to the Certified Layer Thicknesses (n = 3) for 3D-LSM Measurements and Measured MgF2 Standards Compared to the Certified Layer Thicknesses (n = 2) for WLI+LAI Measurements

Figure 4
Figure 4

Comparison between ALT (a) per sector and (b) per line yielded by WLI (n = 5), 3D-LSM (n = 5), and WLI+LAI (n = 5). LOD values in WLI results were replaced by 70 nm.

3.2. Syringe Batch Characterization

3.2.1. Silicone Layer Thickness:

ALTs in syringes from lot A gradually decreased from 170 nm at the flange to 90 nm at the needle side (Figure 5a). The silicone layers were homogeneously distributed over the syringe barrel circumference with ALTs per line from 110 to 120 nm (Figure 5b). These trends also applied to the measured syringes of lot B: ALTs per sector were similar and covered a range from 100 to 190 nm. The SD was higher than the one measured in the first batch, especially when the radial ALTs per line (ranging from 115 to 135 nm) were compared (Figure 5b). The high SD resulted mainly from one data point (syringe 05, measurement point at 20 mm) with ALTs close to 600 nm measured at all 12 radial line,s while all other measurement results ranged from 100 to 200 nm. This may result from a material change such as a silicone droplet entrapped in the coated layer or a micro-scratch in the glass barrel.

Figure 5
Figure 5

Silicone oil layer thickness of two syringe batches by WLI. ALT (a) per sector and (b) per line of lot A and lot B syringes. LOD values were replaced by 70 nm (n = 10).

RSDs between the different syringes from lot A ranged from 19% to 41% (average 31%) for the ALTs per sector (longitudinal silicone layer distribution) and from 36% to 41% (average 38%) for the ALTs per line (radial silicone layer distribution), respectively. RSDs for syringes of lot B ranged from 37% to 68% (ALT per sector, average 41%) and from 47% to 70% (ALT per line, average 55%).

3.2.2. Mechanical Performance:

Break-loose forces for both lot A and B amounted to 4–4.5 N, while the average gliding forces were between 2 and 2.5 N independent of the plunger stopper position (Figure 6). Both syringe batches showed comparable force profiles during plunger movement. The force increase at the end of the plunger movement resulted from the plunger stopper hitting the needle side of the syringe barrel. A higher variability was observed when the plunger stopper was positioned 24 mm below the finger flange compared to a stopper position of 10 mm below the flange.

Figure 6
Figure 6

Average force profiles of two syringe batches. Syringe plunger positions for lot A and lot B were 10 mm and 24 mm below finger flange, respectively (n = 5).

3.3. Impact of Syringe Filling and Emptying

The ALTs per sector of empty syringes gradually decreased from 120 nm in sector 2 near the flange to 90 nm in sector 5 at the needle side (Figure S5a, supporting information). The radial distribution over the syringe barrel circumference was homogeneous, exhibiting silicone layers from 100 to 135 nm (Figure S5b, supporting information). After filling with surrogate solution, including short-term exposure and subsequent emptying, the silicone layer thicknesses ranged from 90 nm to 100 nm in sectors 2 to 5 and were thus comparable to the values prior to filling. The silicone oil was still homogeneously distributed over the syringe barrel circumference (ALTs per line from 105 nm to 125 nm). The silicone layer showed a similar appearance with micro-droplet structures in both empty syringes and after filling with surrogate buffer followed by emptying (Figure S5c, supporting information).

Figure S5
Figure S5

Impact of syringe filling with surrogate solution and emptying. ALT (a) per sector and (b) per line of syringes empty and after filling+emptying (n = 4). LOD values were replaced by 70 nm. Sector 1 results were not considered for analysis (position of syringe plunger). (c) Exemplary microscopic images for one syringe before filling and after filling+emptying with surrogate solution.

4. Discussion

4.1. White Light Interferometry (WLI) Method Development

4.1.1. Instrument Accuracy over Time:

A correlation coefficient of 1.000 (Table I) indicated a highly linear calibration (32). The accuracy of 93–99% was lower than the typically expected accuracy of 98–102% for analytical methods (22) but was still within the limits provided by the manufacturer (90–100%). Due to the low deviation of the measurement results over time (SD within 0 nm to 1 nm), a monthly rather than a weekly accuracy control of the curved MgF2-coated standards by WLI was considered to be sufficient.

4.1.2. Analyst and Instrument Precision:

The RSD for instrument precision with multiple measurements of one measurement line ranged from 0.1% to 0.5%. When the syringes were newly loaded onto the sample holder to determine analyst precision, the RSD increased to 1.7–4.1%. The stringent criterion of ≤2% RSD that is conventionally applied (32) was not met. However, both the analyst and instrument precision showed similar or even lower RSDs than the ones reported for WLI by Lankers [RSD for coated standards 3% to 5.5%, and for silicone layers 5% to 7% (19)]. The precision additionally met the specifications described by the manufacturer with RSDs below 5% (3). Overall, one line can thus be measured with sufficient analyst and instrument precision by WLI.

The ALT of one measurement line could not be used to describe the characteristic layer thickness profile along the syringe barrel length (Figure 1, Figure 2a). One average value would not represent the complete silicone layer distribution properly. Data points outside the measurement range of 80 nm to 4000 nm were neither shown in the raw data nor included in the average values. In this study, the ALTs of one measurement line were used as an example to calculate the analyst and instrument precision as detailed before. For all other experiments, the silicone layer thickness of the individual measurement points along one measurement line was summed up to ALTs per sector. This procedure facilitated the mapping of the silicone layer distribution along the syringe barrel and enabled a more straightforward data interpretation compared to evaluating all individual measurement points. For some sectors, especially for sector 4 and 5 close to the needle side of the syringe, thin silicone layers and thus a high number of LOD values were obtained, which would limit the reliability of the ALT per sector. Hence, all LOD values were replaced by 70 nm (see section 4.1.4).

The observed differences in sectors 4 and 5 between instrument (Figure 2a) and analyst precision (Figure 1) for sample II and III may be due to inhomogeneous silicone micro-droplets (Figure 2b). Although the sample position was kept as constant as possible, minimal variation of the sample position may yield different signals depending on the exact location of the measurement point. In a theoretical scenario, in one measurement the peak of a silicone droplet might be measured exactly, whereas the next measurement might map the descending drop area. Consequently, ALTs close to 80 nm may be once measured as effective signals and once fall just below the LOD.

4.1.3. Number of Radial Lines and Measurement Points:

The silicone oil distribution in the syringe barrel strongly depends on the siliconization method, particularly on parameters such as nozzle position, nozzle kinetics, or spray amount used to spray the silicone oil or silicone emulsion in the syringe barrels (6, 12). Fixed nozzle positions likely result in a pronounced siliconization at the syringe flange (3, 9, 17). A more homogeneous coating was described for diving nozzle kinetics (26, 33). Independent of the spray parameters, micro-droplets and plaque-like structures surrounded by the thin film layer have been observed for sprayed-on silicone layers using silicone oil compared to a more uniform coating yielded by bake-on siliconization (10, 19, 23, 25).

In this study, the silicone layers of the characterized syringe lots showed similar inhomogeneous micro-droplets and plaque-like structures (Figure 2b), which made layer thickness mapping challenging. Twelve radial lines were found to be optimal to describe the radial silicone oil distribution (Figure 3a–c). One measurement point per millimeter barrel length was used in the described measurements, as this corresponds to the actual recommendations in the literature (19). However, fewer measurement points did not yield significantly different ALTs per sector (Figure S4, supporting information). Measurements time thus potentially could be reduced by minimizing the number of measurement points along the syringe barrel length.

Compared to the literature, the chosen measurement set-up with 12 radial lines and 50 measurement points allowed an improved mapping of the longitudinal and radial silicone layer distribution. Lankers used three radial lines to roughly characterize the silicone layers and recommended to set one measurement point per millimetre of barrel length (19). Others measured silicone layer thicknesses with eight radial lines and 50 points per line (10, 18).

4.1.4. Head-to-Head Comparison to 3D-Laser Scanning Microscopy (3D-LSM) and to a Combination of White Light and Laser Interferometry (WLI+LAI):

Accurate measurements with 3D-LSM, WLI, and WLI+LAI (96–97%, 93–99%, and 90–98%, respectively (Tables I and III), were ensured by measuring the corresponding certified standards. All three methods were thus considered applicable as orthogonal methods to measure and compare the silicone layer thickness in spray-on siliconized syringes.

The LOD of 3D-LSM was determined to be 10 nm (9). Hence, thin silicone layers can be measured with high sensitivity, which is an important advantage compared to WLI, with a LOD of 80 nm. In the WLI values below, 80 nm are thus left out from final analyses by the WLI instrument software. This led to non-representative ALTs, particularly in sector 4 and 5 close to the needle side—especially for syringe II due to a low number of actual measurement values above 80 nm. In 3D-LSM the thin silicone layers in sector 4 and 5 exhibited a layer thickness of 79 ± 23 nm and 68 ± 33 nm, respectively (Figure 4a). Therefore, LOD results yielded by WLI were replaced by a theoretical layer thickness of 70 nm based on the 3D-LSM results.

The primary disadvantage of 3D-LSM is its destructive measurement. Fewer measurement points along the barrel length result in less representative results: 3D-LSM results from rather inhomogeneous sprayed-on silicone oil layers thus need to be carefully evaluated. More homogeneous baked-on silicone oil layer thicknesses, however, can be more reliably characterized by 3D-LSM because only a few measurement points are needed to represent the silicone oil distribution (9). Furthermore, measuring time is longer than for WLI and WLI+LAI techniques (Table IV); and the handling—for example, determining the measuring points—is less reproducible, as no syringe sample holder predetermines the measuring line. A major advantage of 3D-LSM is the ability to visualize thin silicone oil layers with high resolution. Thin silicone layers may be either visualized after container breakage (34) or non-destructively (9) by focussing through the glass barrel onto the silicone oil layer (9, 25). Tracking the radial silicone oil distribution with multiple lines in 3D-LSM is challenging due to the destructive nature of the technique and requires high effort in sample preparation. Overall, less radial lines can be measured compared to WLI.

View this table:
TABLE IV

Comparison of WLI, 3D-LSM, and Combined WLI+LAI for the Characterization of the Silicone Oil Layer Thickness in Empty Glass Syringes

Alternatively, LAI could fill the gap for silicone layer thicknesses below 80 nm in WLI. A combination of WLI+LAI within one instrument thus presents a viable alternative to additional 3D-LSM measurements. However, LAI measurements need to be performed with caution and careful set-up of baseline values, and measurements steps are required. A combined WLI+LAI method approximately triples the measurement time to roughly 15 s per measurement spot compared to 4 s for WLI. For routine measurements, though, a combined WLI+LAI method seems more feasible compared to additional 3D-LSM measurements if visualization of the measured layer is not necessary. A final method comparison can be found in Table IV.

4.2. Syringe Batch Characterization

4.2.1. Silicone Oil Layer Thickness:

The silicone oil layer thickness of two syringe batches (lots A and B) was compared with respect to its radial and longitudinal silicone oil distribution within the syringe barrel. The longitudinal ALTs ranged from 90 nm to 170 nm for the first and from 100 nm to 190 nm for the second batch, with silicone oil layers being thicker at the flange and decreasing towards the needle side (Figure 5a). For diving nozzle positions as applied for these syringes, an opposite silicone oil layer distribution with thicker layers at the needle side was reported (3, 17). A pronounced siliconization at the flange was more commonly observed for syringes siliconized with static nozzle positions (3, 17). However, tip-up storage of empty siliconized syringes has been shown to induce a redistribution of silicone oil towards the flange (3, 9, 18, 23, 25), where an accumulation and even dripping onto the insert sheet of the tub may occur (18, 19, 23). Wen et al. measured syringes with such an extent of migration that the measurement points near the needle did not show any silicone oil (23). Redistribution has been observed after 3 days (19), which may explain the altered silicone oil distribution of both syringe batches siliconized in 2015.

The target silicone oil level of the syringes ranged from 0.3 to 0.7 mg per barrel. These values corresponded to a theoretical layer thickness between 310 and 720 nm using a cylindrical shape and neglecting the surface at the needle side as an approximation for the syringe barrel. The measured ALTs were below the minimal theoretical silicone oil layer thickness, which may indicate a redistribution of silicone oil towards the flange and silicone oil loss on the insert sheet. Additionally, an accumulation of silicone oil in the syringe “shoulder” is likely to happen during diving nozzle spray-on siliconization (25), thus giving rise to thinner silicone oil layers as compared to the theoretical values.

The radial silicone oil layer thickness was rather homogeneous for both batches between 100 and 200 nm (Figure 5b), which is reasonable for tip-up stored syringes (assuming homogeneous siliconization process) as opposed to horizontally stored syringes, where silicone oil is expected to accumulate on one side of the circumference (18).

The averaged variability between 10 syringes from lot A was 31% when the ALTs within the different sectors were compared. The averaged variability of the silicone layer thickness measured at the different radial lines was 38% between the 10 syringes. The intra-batch variability for layer thicknesses thus approximately corresponds to the inherent variability given by the target silicone oil level range (40%). The intra-batch variability in the second batch was higher with an averaged RSD of 41% (ALT per sector) and 55% (ALT per line) between the 10 analyzed syringes. This was attributed to one syringe with a different layer thickness at one measurement point (see section 3.2.1). As all other measurement points between the 10 syringes from lot A and 10 syringes from lot B were similar for both batches, which were thus considered to be comparable.

Overall, compared to the literature thin homogeneous silicone oil layers in the range of 100 to 200 nm were observed for a representative number of syringes. For single syringes, highly varying silicone oil layer thicknesses ranging from 900 nm at the flange to 0 nm at the needle side were reported (26), while Wen et al. described syringes with layer thicknesses ranging from approximately 100 to 700 nm (23).

4.2.2. Mechanical Performance:

Average break-loose (4–4.5 N) and gliding forces (2–2.5 N) were comparable for both syringe batches and were independent of plunger stopper position (Figure 6). Both syringe batches showed a homogeneous force profile along the syringe barrel. Functionality of the PFSs was thus considered guaranteed as the measured forces were below the 10 N set as maximum forces for empty syringes by the International Organization for Standardization (ISO) 7886-1 norm (35). Others reported patient-friendly gliding forces between 15 and 20 N (36) and maximum forces have been fixed at 30 N for filled syringes (37). Wen et al. observed an increase of the gliding force or even stalling of the syringe plunger towards the end of the syringe barrel due to a lack of silicone oil at the needle side (23). In this study, slightly decreasing silicone oil layers towards the needle side still facilitated smooth gliding of the syringe plunger. A silicone oil layer thickness of 100 to 200 nm can thus be considered adequate for functionality at this particular measurement time point.

Higher SDs were observed when the plunger stopper was placed at 24 mm below the finger flange compared to 10 mm below the flange. This could be attributed to the slightly thicker silicone oil layers measured at the flange, which were most likely displaced towards the needle side during stopper movement starting at 10 mm below the flange (25). This displacement of silicone oil from the flange potentially further improved smooth and consistent stopper movement along the syringe barrel. At a plunger stopper position of 24 nm below the flange, the silicone layer at the flange would not be affected. Thinner silicone oil layers at the middle and top of the syringe still enabled adequate piston gliding, but it was less smooth. Depending on the siliconization profile, the applied plunger stopper position is thus crucial to obtain force measurements representative for the final product. With slightly thicker silicone oil layers at the flange, which likely appear in syringes stored tip-up, the displacement of silicone oil from flange to top may ease gliding and result in lower or less variable forces. These results may, however, not be representative for products with low plunger stopper positions, for example, those used with low filling volumes.

4.3. Impact of Syringe Filling

WLI is limited to measuring the silicone oil layer thickness in empty syringes. Sample preparation is more challenging for filled syringes because filling and emptying may change the silicone oil layer. Schlieren visualization can be applied to qualitatively characterize the silicone oil distribution in filled syringes. The obtained results map surface coverage with silicone oil droplets, which have formed upon the contact with an aqueous filling medium, rather than yielding quantitative results for layer thickness and its distribution (9, 33). Therefore, WLI was evaluated to assess filled PFSs. The ALT of empty syringes and after filling/emptying with surrogate solution followed the trend of a slightly decreasing silicone oil layer from the flange (∼160 nm) towards the needle side (∼90 nm) (Figure S5a, supporting information). The distribution over the syringe barrel circumference was homogeneous with 105–125 nm (Figure S5b, supporting information). Gerhardt et al. proposed a disruption of the coated silicone oil layer and release of silicone oil droplets in filled syringes subjected to agitation due to the movement of the air bubble along the syringe wall (38). Silicone oil droplets may reattach to hydrophobic silicone surfaces, which could contribute to redistribution of silicone oil. The movement of emptying through the flange could be compared to such an air bubble, being a mechanical force that moves across the silicone oil coating. In this study, the silicone oil layer remained intact with comparable silicone layer thicknesses and visual appearance after filling and emptying (Figure S5c, supporting information). It thus appears reasonable to use this approach to characterize silicone oil layers in filled syringes over time (10). Based on these results, the impact of storage for a longer period can be evaluated by excluding the impact of an inherent mechanical disruption of the silicone oil layer.

5. Conclusion

WLI demonstrated a good accuracy of 93–99% for MgF2-coated, curved standards and a high instrument (0.5%) and analyst precision (4.1%), both of the latter assessed for empty, siliconized PFSs. WLI measurements of PFSs were optimized regarding number of radial lines scanned (3–12 lines, 12 lines found as optimum) and number of measurement points per line (5–100 points, 50 found as optimum). Raw data was aggregated to represent the ALTs within five sectors, each along the barrel length. This procedure allowed improved mapping of the silicone layer distribution along the barrel compared to using a single average thickness value for an entire line and enabled straightforward data interpretation compared to evaluating all individual measurement points.

A head-to-head comparison between WLI, 3D-LSM, and combined WLI+LAI yielded similar ALTs. All three methods thus appear suitable as orthogonal methods to compare the silicone layer thickness in spray-on siliconized syringes. Combined WFI+LAI further reduced the LOD to approximately 20 nm in this study and is a valid alternative to time-consuming and destructive 3D-LSM measurements.

PFSs from two different batches were characterized using the WLI method and exhibited comparable longitudinal silicone oil layers ranging from 170–190 nm to 90–100 nm from flange to tip and homogeneously distributed silicone layers over the barrel circumference (110–135 nm), which is in contrast to data exemplarily shown in the literature. The intra-batch variabilities were between 30% and 55% and approximately corresponded to the inherent variability given by the target silicone level (0.5 mg/barrel ± 40%). Nonetheless, the inherent variability in both batches clearly indicated that analysing a representative number of syringes is crucial to properly characterize the silicone oil layer thickness and distribution.

Characterization of mechanical performance showed overall low break-loose (4–4.5 N) and average gliding forces (2–2.5 N) for both syringe batches. Plunger stopper positions for functionality measurements have to be carefully chosen, as they were shown to affect variability in the force profile. Overall, a silicone oil layer thickness of 100 to 200 nm was sufficient for adequate functionality in this particular study.

The ALTs in syringes after filling, short-term exposure to a surrogate solution, and emptying were not significantly different to the ALTs of empty syringes, indicating that the filling process, the short-term exposure to surrogate solution, and the emptying process over the flange did not alter the silicone layer thickness. This approach could thus be used to characterize silicone oil layers in filled syringes over time.

The WLI method provided reliable, detailed results to non-destructively characterize the silicone oil layer thickness and distribution in sprayed-on siliconized syringes within 40 min per sample. The obtained information can be used to support optimization of siliconization processes and helps to better understand the functionality of siliconized PFSs.

Conflict of Interest Declaration

The authors declare that they have no competing interests.

Abbreviations

3D-LSM, 3D-laser scanning microscopy; AF, autofocus; ALT, average layer thickness; LAI, laser interferometry; LOD, limit of detection; LSM, laser scanning microscopy; n.d., not defined; PFS, prefilled syringe; PS, protective shield; RIs, refractive indices, RSD, relative standard deviation; SD, standard deviation; UT, ultra-thin; WLI, white light interferometry; WLI+LAI, (combined) white light and laser interferometry.

Acknowledgements

The authors gratefully acknowledge support from rap.ID Particle Systems GmbH (Berlin, Germany) for providing access to the Layer Explorer Ultra Thin 2.0 instrument. We particularly thank Lisa Krapf for the detailed insight into the different interferometry methods and her valuable help during the measurements. We also thank Mathieu Rigollet, Dominique Lavergnat, Carolin Schrank, and Marcel Fröhlich from F. Hoffmann-La Roche Ltd. for providing material for analysis. We are also grateful to Lars Abel (F. Hoffmann-La Roche Ltd.) for his valuable technical expertise.

Reference

  1. 1.↵
    1. Sacha G.,
    2. Rogers J. A.,
    3. Miller R.
    Pre-Filled Syringes: A Review of theHhistory, Manufacturing and Challenges. Pharm. Dev. Technol. 2015, 20 (1), 1–11.
    OpenUrl
  2. 2.↵
    1. Sacha G.,
    2. Saffell-Clemmer W.,
    3. Abram K.,
    4. Akers M. J.
    Practical Fundamentals of Glass, Rubber, and Plastic Sterile Packaging Systems. Pharm. Dev. Technol. 2010, 15 (1), 6–34.
    OpenUrlPubMed
  3. 3.↵
    F. Hoffmann-la-Roche Ltd. Qualifizierungsplan Layer Explorer Qsp0902, 2009.
  4. 4.↵
    1. Rathore N.,
    2. Pranay P.,
    3. Eu B.,
    4. Ji W.,
    5. Walls E.
    Variability in Syringe Components and Its Impact on Functionality of Delivery Systems. PDA J. Pharm.Sci. Technol. 2011, 65 (5), 468–480.
    OpenUrl
  5. 5.↵
    1. Jezek J.,
    2. Darton N. J.,
    3. Derham B. K.,
    4. Royle N.,
    5. Simpson I.
    Biopharmaceutical Formulations for Pre-Filled Delivery Devices. Expert Opin. Drug Del. 2013, 10 (6), 811–828.
    OpenUrl
  6. 6.↵
    1. Vugmeyster Y.,
    2. Xu X.,
    3. Theil F.-P.,
    4. Khawli L. A.,
    5. Leach M. W.
    Pharmacokinetics and Toxicology of Therapeutic Proteins: Advances and Challenges. World J. Biol. Chem. 2012, 3 (4), 73–92.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Seidl A.,
    2. Hainzl O.,
    3. Richter M.,
    4. Fischer R.,
    5. Böhm S.,
    6. Deutel B.,
    7. Hartinger M.,
    8. Windisch J.,
    9. Casadevall N.,
    10. London G. M.,
    11. Macdougall I.
    Tungsten-Induced Denaturation and Aggregation of Epoetin Alfa During Primary Packaging as a Cause of Immunogenicity. Pharm. Res. 2012, 29 (6), 1454–1467.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Gascon P.
    Presently Available Biosimilars in Hematology-Oncology: G-CSF. Targeted Oncology 2012, 7 (1), 29–34.
    OpenUrl
  9. 9.↵
    1. Funke S.,
    2. Matilainen J.,
    3. Nalenz H.,
    4. Bechtold-Peters K.,
    5. Mahler H. C.,
    6. Friess W.
    Analysis of Thin Baked-On Silicone Layers by FTIR and 3D-Laser Scanning Microscopy. Eur. J. Pharm. Biopharm. 2015, 96, 304–313.
    OpenUrl
  10. 10.↵
    1. Depaz R. A.,
    2. Chevolleau T.,
    3. Jouffray S.,
    4. Narwal R.,
    5. Dimitrova M. N.
    Cross-Linked Silicone Coating: A Novel Prefilled Syringe Technology That Reduces Subvisible Particles and Maintains Compatibility with Biologics. J. Pharm. Sci. 2014, 103 (5), 1384–1393.
    OpenUrlPubMed
  11. 11.↵
    1. Krayukhina E.,
    2. Tsumoto K.,
    3. Uchiyama S.,
    4. Fukui K.
    Effects of Syringe Material and Silicone Oil Lubrication on the Stability of Pharmaceutical Proteins. J. Pharm. Sci. 2014, 104 (2), 527–535.
    OpenUrl
  12. 12.↵
    1. Colas A.,
    2. Siang J.,
    3. Ulman K.
    Siliconization of Parenteral Packaging Components. In Silicones in Pharmaceutical Applications; Dow Corning Corporation: Midland, MI, 2006.
  13. 13.↵
    Parenteral Drug Association. Lubrication of Packaging Components Task Force. In Siliconization of Parenteral Drug Packaging Components. Parenteral Drug Association: Bethesda, MD, 1988; Vol 12.
  14. 14.↵
    1. Badkar A.,
    2. Wolf A.,
    3. Bohack L.,
    4. Kolhe P.
    Development of Biotechnology Products in Pre-Filled Syringes: Technical Considerations and Approaches. AAPS PharmSciTech 2011, 12 (2), 564–572.
    OpenUrlPubMed
  15. 15.↵
    European Pharmacopoeia Commission. European Pharmacopoeia; Council of Europe: Strasbourg, France, 2004; Vol 5.
  16. 16.↵
    Dow Corning Corporation. Dow Corning 360 Medical Fluid product information (safety data sheet), 2015.
  17. 17.↵
    1. Chan E.,
    2. Hubbard A.,
    3. Sane S.,
    4. Maa Y.
    F.Syringe Siliconization Process Investigation and Optimization. PDA J. Pharm. Sci. Technol. 2012, 66 (2), 136–150.
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    1. Bee J. S.,
    2. Frey V. V.,
    3. Javed U.,
    4. Chung J.,
    5. Corcoran M. L.,
    6. Roussel P. S.,
    7. Krause S. O.,
    8. Cash P. W.,
    9. Bishop S. M.,
    10. Dimitrova M. N.
    Characterization of the Initial Level and Migration of Silicone Oil Lubricant in Empty Prefilled Syringes for Biologics Using Infrared Spectroscopy. PDA J. Pharm. Sci. Technol. 2014, 68 (5), 494–503.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Lankers M.
    Analyse von Silikonschichtdicken bei der Herstellung von Fertigspritzen mit Hilfe von Reflektometriemessungen. Pharmazeutische Industrie 2010, 72 (12).
  20. 20.
    1. Khandke L.,
    2. Malone R.,
    3. Yang X.,
    4. Han H.,
    5. Look J. L.,
    6. Jin Z.,
    7. Seid R. C. Jr..,
    8. Chen Y.
    Novel Formulations Which Stabilize and Inhibit Precipitation of Immunogenic Compositions. U.S. Patent 2011/0172393 A1, 2011.
  21. 21.↵
    1. Yoshino K.,
    2. Arisa Yamashita N.,
    3. Abe Y.,
    4. Iwasaki K.,
    5. Kanazawa Y.,
    6. Funatsu K.,
    7. Yoshimoto T.,
    8. Suzuki S.
    Functional Evaluation and Characterization of a Newly Developed Silicone Oil-Free Prefillable Syringe System. J. Pharm. Sci. 2014, 103 (5), 1520–1528.
    OpenUrlPubMed
  22. 22.↵
    1. Funke S.,
    2. Matilainen J,
    3. Nalenz H,
    4. Bechtold-Peters K,
    5. Mahler H. C,
    6. Friess W.
    Silicone Migration from Baked-On Silicone Layers. Particle Characterization in Placebo and Protein Solutions. J. Pharm. Sci. 2016, 105 (12), 3520–3531.
    OpenUrl
  23. 23.↵
    1. Wen Z.-Q.,
    2. Vance A.,
    3. Vega F.,
    4. Cao X.,
    5. Eu B.,
    6. Schulthesis R.
    Distribution of Silicone Oil in Prefilled Glass Syringes Probed with Optical and Spectroscopic Methods. PDA J. Pharm. Sci. Technol. 2009, 63 (2), 149–158.
    OpenUrlAbstract/FREE Full Text
  24. 24.↵
    European Committee for standardization. EN ISO 11040-4:2015 Prefilled syringes—Part 4: Glass barrels for injectables and sterilized subassembled syringes ready for filling. 2015.
  25. 25.↵
    1. Funke S.,
    2. Matilainen J,
    3. Nalenz H,
    4. Bechtold-Peters K,
    5. Mahler H. C,
    6. Friess W.
    Optimization of the Bake-On Siliconization of Cartridges. Part I: Optimization of the Spray-On Parameters. Eur. J. Pharm. Biopharm. 2016, 104, 200–215.
    OpenUrl
  26. 26.↵
    1. Felsovalyi F.,
    2. Janvier S.,
    3. Jouffray S.,
    4. Soukiassian H.,
    5. Mangiagalli P.
    Silicone-Oil-Based Subvisible Particles: Their Detection, Interactions, and Regulation in Prefilled Container Closure Systems for Biopharmaceuticals. J. Pharm. Sci. 2012, 101 (12), 4569–4583.
    OpenUrlPubMed
  27. 27.↵
    1. Reuter B.
    Silicone Oil and Its Applications for Parenteral Products. PDA Europe Workshop, Cologne, Germany, 2010.
  28. 28.↵
    1. Ingle R.,
    2. Agarwal A.
    Pre-Filled Syringe—A Ready-To-Use Drug Delivery System: A Review. Expert Opin. Drug Deliv. 2014, 11 (9), 1391–1399.
    OpenUrl
  29. 29.↵
    Rap.ID Particle Systems GmbH. Technical Paper—How the Layer Explorer Works; Berlin, Germany, 2012.
  30. 30.↵
    Rap.ID Particle Systems GmbH. Layer Explorer Operator's Manual. 2012.
  31. 31.↵
    Rap.ID Particle Systems GmbH. Silicone Oil Layer Control. Available online: http://www.rap-id.com/premium_download/1006_DS_LE_e.pdf, accessed on April 13, 2018.
  32. 32.↵
    1. Green P.,
    2. Silverman B.
    Nonparametric Regression and Generalized Linear Models: A Roughness Penalty Approach; CRC Press: Boca Raton, FL, 1993.
  33. 33.↵
    1. Reuter B.,
    2. Petersen C.
    Syringe Siliconization—Trends, Methods, Analysis Procedures. TechnoPharm 2012, 2 (4), 238–244.
    OpenUrl
  34. 34.↵
    1. Funke S.,
    2. Matilainen J.,
    3. Nalenz H.,
    4. Bechtold-Peters K.,
    5. Mahler HC.,
    6. Vetter F.,
    7. Müller C.,
    8. Bracher F.,
    9. Friess W.
    Optimization of the Bake-On Siliconization of Cartridges. Part II: Investigations into Burn-In Time and Temperature. Eur. J. Pharm. Biopharm. 2016, 105, 209–222.
    OpenUrl
  35. 35.↵
    European Committee for Standardization. EN ISO 7886-3 Sterile hypodermic syringes for single use—Part 3: Auto-disable syringes for fixed-dose immunization. 2015.
  36. 36.↵
    1. Joseph L.,
    2. Martin S.,
    3. Kohle P.,
    4. Muthurania K.
    Development of Syringeability Guide for Subcutaneous Protein Formulations. 2010; Available online: http://www.readbag.com/aapsj-abstracts-am-2010-w5265, accessed on June 15, 2016.
  37. 37.↵
    1. Burckbuchler V.,
    2. Mekhloufi G.,
    3. Giteau A. P.,
    4. Grossiord J. L.,
    5. Huille S.,
    6. Agnely F.
    Rheological and syringeability properties of highly concentrated human polyclonal immunoglobulin solutions. Eur. J. Pharm. Biopharm. 2010, 76 (3), 351–356.
    OpenUrlPubMed
  38. 38.↵
    1. Gerhardt A.,
    2. Mcgraw N. R.,
    3. Schwartz D. K.,
    4. Bee J. S.,
    5. Carpenter J. F.,
    6. Randolph T. W.
    Protein aggregation and particle formation in prefilled glass syringes. J. Pharm. Sci. 2014, 103 (6), 1601–1612.
    OpenUrl
Back to top

In This Issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Methods To Determine the Silicone Oil Layer Thickness in Sprayed-On Siliconized Syringes
Viviane Loosli, Oliver Germershaus, Henrik Steinberg, Sascha Dreher, Ulla Grauschopf, Stefanie Funke
PDA Journal of Pharmaceutical Science and Technology May 2018, 72 (3) 278-297; DOI: 10.5731/pdajpst.2017.007997
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords