Downloaded from journal.pda.

org on July 4, 2019

Vapor Phase Hydrogen Peroxide Uptake by Silicone Tubing and

Primary Packaging Components during Protein Drug Product
Aseptic Filling: Impact by Pre-treatment and Sterilization
Process
Yuh-Fun Maa, Devon Roshan Eisner, Aaron Hubbard, et al.

PDA Journal of Pharmaceutical Science and Technology 2019,

Access the most recent version at doi:10.5731/pdajpst.2019.009928
 Downloaded from journal.pda.org on July 4, 2019

Vapor Phase Hydrogen Peroxide Uptake by Silicone Tubing and Primary Packaging

Components during Protein Drug Product Aseptic Filling: Impact by Pre-treatment

and Sterilization Process

Devon Roshan Eisner, Aaron Hubbard, Kirk Eppler, Vassia Tegoulia, and Yuh-Fun Maa*

Pharmaceutical Processing and Technology Development, Genentech, Inc., 1 DNA Way, South

San Francisco, CA 94080

*Corresponding author: Yuh-Fun Maa. Genentech, a member of the Roche Group, 1 DNA Way

South San Francisco, CA 94080. Telephone: +1-650-225-3499. Fax: +1-650-742-1504. e-mail:

maay@gene.com

1
 Downloaded from journal.pda.org on July 4, 2019

ABSTRACT

In the vapor phase hydrogen peroxide (VPHP)-sanitized environment, VPHP uptake by product-

contacting components could eventually lead to undesired oxidation of biological drug products.

Silicone tubing and primary packaging materials are prominent examples of such product-

contacting surfaces that are typically processed/sterilized prior to use. This study investigated the

VPHP-uptake tendency of these components and how their respective processing/sterilization

methods affect uptake behaviors. Silicone tubing that was sterilized via autoclave or gamma

irradiation exhibited different VPHP uptake patterns; decreased uptake rates post autoclaving vs.

increased uptake rates post gamma irradiation. The reduced uptake tendency of autoclaved

tubing is maintained 14 days after sterilization while the uptake tendency of irradiated tubing

was mostly reversed to normal levels 1 month after irradiation. Empty glass vials adsorbed

hydrogen peroxide via the diffusion of VPHP into the vial with high vial-to-vial variability. Vial

pre-treatment (i.e., depyrogenation) and surface hydrophilicity/hydrophobicity impacted uptake

tendency. Stoppers and empty syringes also adsorbed hydrogen peroxide but at a relatively low

level. The uptake behavior of these components appeared to correlate with water levels at the

surface (i.e., hydrophilicity). This study would provide process development scientists and

engineers an in-depth understanding of VPHP uptake by critical product-contacting surfaces to

mitigate the impact on drug product quality.

KEYWORDS: Vapor phase hydrogen peroxide, Monoclonal antibody, VPHP uptake, Primary

packaging components, Silicone tubing, Sterilization, Gamma irradiation, Autoclave, VHP,

Vaporized hydrogen peroxide.

2
 Downloaded from journal.pda.org on July 4, 2019

LAY ABSTRACT

This study investigated vapor phase hydrogen peroxide (VPHP) absorption by biopharmaceutical

drug products via VPHP uptake by critical product-contacting components during the aseptic

manufacturing process with a focus on various pre-treatments and processing of these

components. Sterilization of silicone tubing by gamma irradiation or autoclaving resulted in

different VPHP uptake profiles with different lasting effects. Primary packaging components,

such as vials, syringes, and stoppers, also contributed to different levels of VPHP uptake with

surface hydrophilicity/hydrophobicity playing a critical role. These outcomes suggested that

VPHP uptake is a complex phenomenon and should be carefully considered to minimize its

impact on product quality. The approach and outcome of this study can benefit scientists and

engineers who develop biological product manufacturing processes by providing an in-depth

understanding of drug product process risks.

3
 Downloaded from journal.pda.org on July 4, 2019

1. Introduction

Processing and packaging facilities intended for aseptic filling of biopharmaceutical drug

product into sterilized containers require consistent methods of cleaning and sanitization to

ensure the removal of microorganisms. Vapor phase hydrogen peroxide (VPHP) is a common

and highly effective sanitizing agent against spores, bacteria, and viruses (1-4). A general

concern for manufacturing biopharmaceutical products in a VPHP-sanitized environment is the

migration of residual VPHP into the drug product (DP) solution in the form of hydrogen

peroxide (H2O2), because H2O2 is an effective oxidizing agent that is known to oxidize proteins

(5-9).

The mechanism whereby residual VPHP enters the DP solution during production is critical

technical knowledge that can guide proper manufacturing process controls to ensure that VPHP

uptake levels in final DP containers (e.g., vials, prefilled syringes, etc.) are within acceptable

limits. Hubbard and Eppler (10) provided an overview of potential H2O2 uptake sources in the

isolator and the restricted access barrier system (RABS) filling lines. They identified four likely

sources that may eventually lead to H2O2 absorption by the DP solution: empty glass surfaces

(vials or syringes); the exposed product at the tip of the filling needle; open (unstoppered), filled

vials/syringes; and the tubing of the filling line (10).

When considering two of the identified sources, i.e., filled vials/syringes and the exposed

product at the tip of the filling needle, it is well known that H2O2 is absorbed by the DP solution

directly from the atmosphere. However, the mechanisms driving the transfer of H2O2 from the

product-contacting components (i.e., empty vials, syringes, and tubing) to the DP solution are not

yet understood. The product-contacting tubing flowpath is considered the most prominent source

4
 Downloaded from journal.pda.org on July 4, 2019

due to the high surface-area-to-volume ratio. The primary packaging components (i.e., vials,

syringes, and stoppers) are the final contact surfaces with the DP solution, and these components

may have a more direct impact on product quality.

Most DP filling lines use platinum-cured silicone tubing in the product flowpath, especially in

peristaltic filling systems due to its advantageous mechanical properties (11-12). Silicone tubing

(polydimethylsiloxane) is highly permeable to low molecular weight species and non-polar

substances (12, 13). VPHP is a low molecular weight compound and can easily permeate the

tubing wall into the DP solution via diffusion and partitioning. A recent study by Hubbard and

coworkers (14) demonstrated that VPHP uptake by the silicone tubing in a peristaltic filling line

inside a VPHP-decontaminated isolator plays a critical role in H2O2 migration into the DP.

However, there are various aseptic operations in which flowpath components are sterilized by

various methods and assembled in the filling barrier systems before or after H2O2 sanitization.

Pharmaceutical tubing is typically sterilized by autoclaving (steam sterilization), gamma or

electron-beam (e-beam) ionizing irradiation, or ethylene oxide (EtO) gas treatment (15, 16). The

choice of sterilization method should consider factors beyond sterilization efficacy, including the

impact on tubing properties, the effect on product quality, and operation logistics (17-21). If the

flowpath (tubing) assembly is set up prior to production, the tubing and associated parts are

typically sterilized by autoclaving or steam sterilization at the manufacturing site. For pre-

assembled tubing systems, the assembly is typically pre-sterilized by irradiation (usually gamma

irradiation) at a contract manufacturing location and subsequently shipped to the DP

manufacturing site. Because autoclaving and gamma irradiation are the most commonly used

tubing sterilization methods, we investigated how these two methods would affect the tubing’s

VPHP uptake tendency.

5
 Downloaded from journal.pda.org on July 4, 2019

It is less known whether primary packaging components adsorb H2O2 from VPHP as they are

made of various materials – typically glass for vials and syringes and rubber for stoppers – and

their surfaces may be coated to minimize the interactions with the DP (22-25). Like the tubing,

these primary packaging components are processed (i.e., washed and sterilized) prior to being

filled with DP solution. In this study, we evaluated the H2O2 uptake tendency of these

components and determined their potential impact on DP quality.

2. Materials and Methods

The components used in this study are summarized in Table I. They include tubing, glass vials,

syringes, and stoppers that were either pre-treated (or sterilized) in house or at a contract

laboratory prior to the VPHP uptake experiments.

2.1 Components pre-treatment or sterilization

Tubing was sterilized by either autoclaving or gamma irradiation as listed in Table I and as

described in section 2.2.1. Glass vials were pre-treated by washing and depyrogenating or

conditioning in a humidity chamber. Stoppers were pre-treated by autoclaving.

2.2 VPHP uptake experiments

A Pharmaceutical Safety Isolator (PSI-M, Skan AG, Switzerland) was used to test the filling

components in a controlled VPHP environment. In addition to normal isolator functionality, this

isolator is able to maintain a constant H2O2 concentration in the atmosphere by continuously

vaporizing H2O2 while all other functions of the isolator are in normal production mode. By

changing the dosing rate and the concentration of the H2O2 stock bottle, any VPHP level can be

targeted and maintained. This level was monitored by a Picarro H2O2 sensor (Model G1114 or

G2114, Santa Clara, CA).

6
 Downloaded from journal.pda.org on July 4, 2019

All uptake experiments were performed inside the isolator. The concentration of absorbed H2O2

was determined using a qualified Amplex® UltraRed Hydrogen Peroxide assay.

2.2.1 VPHP uptake by silicone tubing

To test H2O2 absorption by silicone tubing, a Flexicon PD12I (Flexicon Corp., Bethlehem, PA)

pump head and a Flexicon MC12 pump controller were placed outside the isolator. Sample

tubing of a pre-determined length was assembled into the pump head and run into the isolator

through tri-clamp fitting ports. The two ends of the sample tubing ran outside the isolator

through separate fitting ports; one end was connected to a water bottle (upstream) and the other

to a filling needle (downstream). The section of sample tubing inside the isolator could be

exposed to various concentrations of VPHP. The sample tubing was sterilized either by

autoclaving or by exposure to gamma irradiation. Control tubing, which was not sterilized by

either method, was processed similarly to, and tested side by side with, its respective sterilized

counterpart.

For sterilization by autoclave, the silicone tubing (1.6 mm internal diameter × 1.6 mm wall

thickness) was autoclaved at 121°C for 60 min following a pre-defined program ending with a

drying step. The autoclaved tubing was then cut to a pre-determined length and exposed to 100

ppb VPHP in the isolator. VPHP exposure was performed immediately or days (e.g., 2 and 14

days) after autoclaving. Each piece of tubing was then primed with ultrapure water and incubated

in the VPHP environment for up to 60 min prior to taking a water sample for Amplex® testing.

For sterilization by gamma irradiation, the silicone tubing (3.2 mm & 1.6 mm internal diameter ×

1.6 mm wall thickness) was sterilized with 40 kGy of gamma irradiation in the original sealed

plastic bag. Samples of the 3.2 mm irradiated tubing were exposed to 100 ppb VPHP in the

7
 Downloaded from journal.pda.org on July 4, 2019

isolator 3 and 8 days after irradiation. After 29 days, the bag of 1.6 mm irradiated tubing was

opened and exposed to 100ppb VPHP in the isolator. During each VPHP exposure, the samples

of tubing were primed with ultrapure water and incubated in the isolator for various durations up

to 90 minutes prior to taking a water sample for Amplex® testing.

2.2.2 VPHP uptake by glass vials

To test H2O2 adsorption by glass vials, open vials were introduced into the isolator via the rapid

transfer port (RTP) after the VPHP in the isolator reached a constant level of 500 parts per

billion (ppb). The vials were positioned upright on the floor in the center of the isolator. After

VPHP exposure of various time intervals, the vials were removed from the isolator via the RTP.

Each VPHP-exposed vial was immediately pipetted with ultrapure water (or water hereafter) and

stoppered with a clean stopper that had not been exposed to VPHP. The amount of water added

was vial-size dependent: 0.5 mL into 2 cc vials, 1.0 mL into 6 cc vials, 1.5 mL into 15 cc vials,

2.0 mL in 20 cc vials, and 3.0 mL in 50 cc vials. Each vial was rolled on its side and turned

upside down so that water would contact all interior surfaces for complete H2O2 absorption in the

water. All vials were left upside down in a dark environment for 1 h before being tested using the

Amplex® UltraRed assay.

2.2.3 VPHP uptake by glass syringes

To test H2O2 adsorption by glass syringes, ready to use syringes with staked in needle and rigid

needle shield were used. These open syringes were introduced into the isolator via the RTP after

VPHP in the isolator reached a constant level of 500 ppb. Syringes stood with the tip down in a

stand on the floor in the center of the isolator. After VPHP exposure of various time intervals,

the syringes were removed from the isolator via the same RTP. Each syringe was immediately

8
 Downloaded from journal.pda.org on July 4, 2019

pipetted with 0.35 mL of ultrapure water and stoppered using a hand stoppering tool with a clean

plunger stopper that had not been exposed to VPHP. Each syringe was turned end-to-end for a

few rotations and shaken vigorously to ensure all interior surfaces had been in contact with the

water. Syringes were left on their side in a dark environment for 1 h before testing. The contents

of the syringe were expelled into a 2 cc vial and analyzed using the Amplex® UltraRed assay.

2.2.4 VPHP uptake by butyl rubber stoppers

To test H2O2 adsorption (or absorption) by butyl rubber stoppers, the stoppers were introduced

into the isolator via the RTP after VPHP in the isolator reached a constant level of 500 ppb. All

stoppers were placed on the floor of the isolator with the product-contacting side facing upward.

After VPHP exposure of various time intervals, the stoppers were removed from the isolator via

the RTP. Each VPHP-exposed stopper was immediately seated onto a clean, unexposed 2 cc or 6

cc vial filled with 0.5 mL or 1.0 mL of ultrapure water, respectively. Each vial was then turned

upside down to allow the water to be in full contact with the stopper. Stopper-containing vials

were placed upside down in a dark environment for 1 h before the water sample was tested for

H2O2 concentration using the Amplex® UltraRed assay.

2.3 Determination of H2O2 concentration in water

H2O2 concentrations in water were measured using the fluorometric Amplex® UltraRed assay.

This method has been previously described (26).

9
 Downloaded from journal.pda.org on July 4, 2019

3. Results and Discussion

3.1 VPHP uptake by silicone tubing: effect of sterilization on uptake behaviors

We previously reported a VPHP uptake case study involving isolator operation (14). To

minimize human manipulation, the entire filling flowpath, including silicone tubing, was pre-

assembled in the isolator prior to VPHP sanitization. The flowpath was sterilized via clean-in-

place (CIP) and steam-in-place (SIP) processes during isolator aeration after the decontamination

phase. In this case, the tubing assembly absorbed a substantial amount of VPHP during the

decontamination phase; the absorbed H2O2, however, was carried away by WFI and steam and

was depleted after CIP/SIP. The VPHP uptake behavior of the tubing assembly was eventually

dictated by the residual VPHP concentration in the isolator atmosphere during the filling

operation.

In another barrier filling option, featuring a RABS, the liquid flowpath components would be

sterilized in advance and then assembled on the filler (post VPHP decontamination and aeration)

immediately before the start of the filling operation. In this case, the tubing components would

be autoclaved (or steam-sterilized) at the filling site shortly (within a few days) before filling. If

a pre-assembled single-use filling line is used, the tubing assembly is typically gamma irradiated

by the vendor and shipped to the manufacturing site. Thus, gamma irradiated tubing may be used

for manufacturing weeks (or months) after sterilization.

3.1.1 Tubing sterilized by autoclaving

The autoclaved tubing absorbed substantially less H2O2 than the control, at approximately 20%

of the control, when the VPHP uptake experiment was performed 2 days after autoclaving

(Figure 1a). When the VPHP uptake experiment was performed 14 days post autoclaving, the

10
 Downloaded from journal.pda.org on July 4, 2019

uptake tendency of the autoclaved tubing recovered somewhat (to 39%; Figure 1b) relative to the

control. Since autoclaved tubing is typically used within a few days after sterilization, it is

expected that the autoclaving process would present a favorable condition in absorbing less

H2O2.

This decreased VPHP uptake tendency of the autoclaved tubing might be attributed to reduced

water content in the tubing as the result of the drying process at the end of the autoclaving cycle.

Silicone tubing is permeable to gas molecules due to free volume or “openings” created in

association with the high flexibility (movement) of the silicone-oxygen chain in silicone (27).

Free volume permits gas diffusion. Because silicone is hydrophobic, tubing permeability can be

decreased if water is removed from the free volume of the tubing material.

Additional studies were performed to validate this hypothesis. In the first experiment, the

autoclaved tubing was filled with water overnight to allow for water re-absorption prior to the

100 ppb VPHP uptake experiment. The autoclaved tubing showed an increase in VPHP uptake

but was not fully recovered, at roughly 42%, compared to the control (Table II).

Two studies were performed to assess the impact of dehydration of non-autoclaved tubing on

VPHP uptake behaviors. Tubing dehydration took place by either incubating at 130 °C for 12 h

or by vacuum drying. The dehydrated tubing showed reduced VPHP uptake tendency, at 50%–

60% of the control, but not to the level of the autoclaved tubing (Table II). This suggested that

the water content in the tubing does impact VPHP uptake behaviors, but other unknown factors

associated with autoclaving also play a role in the reduced tendency of VPHP uptake.

11
 Downloaded from journal.pda.org on July 4, 2019

3.1.2 Tubing sterilized by gamma irradiation

Three days after irradiation, the 3.2 mm tubing was tested and absorbed substantially higher

amounts of H2O2 than the control (Figure 2a). The H2O2 level for the irradiated tubing was 3

times higher than the control after a 30 min incubation. The difference after the 90 min

incubation was even more significant, but the amount of H2O2 absorbed in the irradiated tubing is

uncertain as the detected concentration of H2O2 exceeded the qualified range of the assay.

In a separate experiment 8 days after irradiation, another sample of the irradiated tubing was

removed and evaluated for VPHP uptake in the same way except for having three incubation

groups at 15, 30, and 60 min. Interestingly, the irradiated tubing still absorbed more H2O2 than

the non-irradiated counterpart but to a lesser extent when compared with the tubing tested 3 days

after irradiation (Figure 2b), indicating that the impact of irradiation on the VPHP uptake rate is

reversible. This observation was confirmed in a third experiment with smaller tubing (1.6 mm

internal diameter × 1.6 mm wall thickness), which was left in the original sealed bag until being

tested for VPHP uptake 29 days after irradiation. In this final experiment, the amount of H2O2

absorbed by the irradiated and the non-irradiated tubings were almost identical for each

incubation time (Figure 2c).

Overall, these data suggested that gamma irradiation alters the H2O2 absorption behaviors of

silicone tubing, which are, however, reversible approximately one month after irradiation. The

DP manufacturer may not realize this phenomenon, as, from the perspective of manufacturing

scheduling and shipping, it is unlikely that the vendor-irradiated tubing will be used for

manufacturing within 30 days of irradiation.

12
 Downloaded from journal.pda.org on July 4, 2019

Changes in the mechanical properties of irradiated silicone tubing have been reported. Gamma

irradiation is known to induce changes in the molecular architecture of silicone rubber,

increasing its molecular weight (cross-linking) and decreasing elasticity (16). Higher doses of

gamma irradiation and longer treatment cycles have been shown to cause higher crosslink

densities (28). However, the relationship between the enhanced permeability and the increased

crosslink structure of gamma-irradiated silicone tubing is difficult to comprehend, particularly in

association with the reversibility feature due to the permanent nature of the crosslinked structure.

Mazor and Zilberman (29) reported a decrease in water vapor transmission rate with increased

irradiation dose for a wound dressing consisting of a barrier layer made of poly(DL-lactic-co-

glycolic acid). Although in this case, gamma irradiation still increased polymer crosslinking,

they attributed the increased water vapor transmission rate to the formation of micro-fracture in

the polymer matrix (i.e., creating pores allowing for better water vapor permeation). If this

phenomenon is applicable to irradiated silicone, the same hypothesis can explain the increased

H2O2 permeability but not the reversibility behavior.

3.2 VPHP uptake by empty glass vials

3.2.1 Adsorption of H2O2 on glass surfaces

During filling operations, empty vials can sit for hours on the line (while exposed to a VPHP

atmosphere) due to process interruptions before being filled. Can an inert surface like glass

adsorb H2O2 from the atmospheric VPHP during the filling operation, and how fast and how

much can it adsorb? To address these questions, a set of 50 cc vials were placed in the isolator

and exposed to three levels of VPHP concentration in the following sequence: 50, 100, 500, and

back down to 50 ppb (the blue profile in Figure 3). Three vials were removed at multiple time

points within each level of VPHP exposure for incubation with ultrapure water and Amplex®

13
 Downloaded from journal.pda.org on July 4, 2019

testing. The result (red line in Figure 3) showed that glass vials could indeed adsorb H2O2 in a

reversible fashion: ~50 ng/vial while exposed to 50 ppb VPHP, ~80 ng/vial to 100 ppb, ~200

ng/vial to 500 ppb, and back to ~50 ng/vial to 50 ppb. The level of adsorption and desorption

closely followed the VPHP concentration (shown as the red profile in Figure 3), reflecting the

possible correlation of surface adsorption by the glass and the VPHP level in the isolator

atmosphere.

A follow-up study was performed. Empty vials (20 cc and 50 cc) were placed in the isolator and

exposed to a VPHP concentration at 500 ppb for up to 48 h. Six vials were removed at each time-

point for Amplex® testing. The correlation of H2O2 adsorbed to the glass surface and VPHP

exposure time is summarized in Figures 4a (50 cc vials) and 4b (20 cc vials), which display a

slight upward trend with a steep uptake initially. After 24 h exposure, the amount of H2O2

adsorbed was in the approximate range of 225 ng per 50 cc vial and 40 ng per 20 cc vial. For

oxidation-sensitive proteins with a low fill volume in a large vial, this level of H2O2

concentration is likely to impact DP quality. However, the overall data were difficult to interpret

due to significant vial-to-vial variability. For example, when reviewing the 24 h time point for

the 50 cc vial, the average uptake of the six 50 cc vials is 207 ng/vial with the minimum uptake,

the maximum uptake, and relative standard deviation (RSD) at 93 ng, 279 ng, and 39%,

respectively. This level of variability was also observed in other vial sizes (2, 6, and 15 cc).

Excluding the variability of the Amplex® assay (~10% RSD), the data suggested other factors

may contribute to this high level of variability.

14
 Downloaded from journal.pda.org on July 4, 2019

3.2.2 Root cause of variability in H2O2 adsorption by vials

VPHP uptake variability was initially considered to be related to the air-flow pattern in the

isolator. Ansys Fluent® computer modeling software was used to simulate air flow inside and

around the vial. It assumed downward airflow in the isolator at a velocity of 0.7 m/s (Figure 5).

Based on the model, air velocity inside the vial opening was minimal, suggesting that air barely

flowed into the vial, and VPHP would enter the vial mostly through diffusion. Thus, variable

H2O2 adsorption on the glass surface should not be attributed to isolator air flow.

To minimize procedure-related variabilities, the study (50 cc vials exposed to 500 ppb VPHP)

was repeated for a single time point using improved procedures, consistent timing for vial

placement in the isolator, vial removal from the isolator, vial incubation, and analytical testing.

All vials in the isolator were placed in a pre-defined zone away from the walls, where air flow

might be uneven. There was still a high level of variability, ~40% RSD (for six vials), in the

amount of H2O2 adsorbed by the vials even after the elimination of potential procedural

variability. The overall analysis suggested that VPHP uptake variability may be due to

heterogeneity of the glass surface for H2O2 adsorption.

3.2.3 Non-uniform adsorption of VPHP by vial glass surface

An experiment was performed that subjected 19 vials (50 cc) to repeated VPHP exposure (five

times). Before each VPHP exposure in the isolator (500 ppb for 4 h), all 19 vials were pre-

conditioned using the same depyrogenation conditions. All procedures for vial placement,

removal, incubation, and analytical testing were executed consistently to avoid procedural

variations (see section 3.2.2). The results demonstrate that each vial showed day-to-day

variations consistently; the vial ranking for the amount of adsorbed H2O2 being very similar in

15
 Downloaded from journal.pda.org on July 4, 2019

each of the 5 days of testing (Figure 6). The same trend was observed for all other vial sizes (data

not shown). This shows that VPHP uptake variability is due to vial variability.

A hypothesis to explain this vial-to-vial variability may be that the glass surface consists of

patches of hydrophilic and hydrophobic zones due to non-homogeneous chemical composition.

Hydrophilic zones bind water, which increases the vial surface water content and facilitates H2O2

adsorption. Hydrophobic zones discourage water binding, so H2O2 adsorption tendency in these

zones is reduced. In Figure 6, the amount of adsorbed H2O2 trends downward with the number of

VPHP exposures (and conditioning/depyrogenation cycles) for all 19 vials. The trend is likely

due to the high temperature of depyrogenation during conditioning, which decreased the glass

surface’s water content.

To test the hypothesis that vial surface water content dictates the amount of H2O2 adsorbed, five

sets of 50 cc vials were placed in the isolator (500 ppb) for 4 hours. The five vial sets are listed

below, the control or standard processed vial and then in order of decreasing surface water

content:

Group 1. Standard glass vials were washed and depyrogenated (350 °C for 3 h). This is

standard vial processing prior to filling.

Group 2. Standard glass vials were washed and placed in a humidity chamber (75% relative

humidity for 80 h). High humidity exposure is expected to increase the total

surface water content of the vial.

Group 3. Standard glass vials were washed and over depyrogenated (350 °C for 80 h).

Extended depyrogenation is expected to decrease the total surface water content of

the vial.

16
 Downloaded from journal.pda.org on July 4, 2019

Group 4. Vials with hydrophobic coating (SCHOTT TopLyo®) were used without

processing. The coating increases the hydrophobic zones on the surface and

decreases the total surface water content of the vial.

Group 5. Vials with hydrophobic coating (SCHOTT TopLyo®) were washed and

depyrogenated (350 °C for 3 h). Depyrogenation of the coated vial may further

decrease the total surface water content of the vial.

The VPHP uptake results are summarized in Figure 7. Vials in Group 1 served as the control

with standard wash and depyrogenation conditioning. As shown in Figure 7, vials with higher

surface water content (Groups 1 and 2) adsorbed much higher amounts of H2O2 than the dried

(Group 3) and the more hydrophobic vials (Groups 4 and 5). Compared with the control group

(~150 ng/vial), incubating the standard vials in a high humidity environment (Group 2)

facilitated VPHP uptake to ~350 ng/vial with a high variability (40% RSD), while all dried and

hydrophobic vials discouraged H2O2 adsorption to below 50 ng/vial and showed low vial-to-vial

variability. Drying or depyrogenating vials at a high temperature for a long duration (i.e.,

extended depyrogenation for Group 3) could severely dehydrate glass surfaces. Applying a

hydrophobic coating to vials (Groups 4 and 5) increases hydrophobic zones on the glass surface

preventing water from binding to the glass surfaces. All the data from this experiment support

the hypothesis that an increase in glass surface moisture promotes H2O2 adsorption and exhibit

high levels of vial-to-vial variability.

3.2.4 Contribution of empty vial to the overall H2O2 absorption of filled vial

A paper calculation based on existing experimental uptake results was performed to evaluate the

contribution of uptake in the empty vial to the total H2O2 uptake in a filled vial. The scenario

17
 Downloaded from journal.pda.org on July 4, 2019

included exposure of empty and filled vials to 100 ppb VPHP isolator air. Under this condition,

empty 50 cc vials would adsorb approximately 80 ng per vial (Figure 3). Considering a worst-

case low fill volume at 5 mL (10% of the vial size) of a monoclonal antibody formulation, which

has a H2O2 limit of 100 ng/mL, the liquid formulation would need to absorb 500 ng of H2O2

before unacceptable H2O2-induced oxidation occurred. In this worst case, H2O2 from empty vials

would contribute 16% of the maximum allowable H2O2 absorption. As 50 cc vials are typically

filled with more than 10 mL, the H2O2 contribution from empty vials becomes relatively

insignificant.

3.3 VPHP uptake by glass syringes

Empty syringes of two sizes, 1 mL long and 2.25 mL, were placed in the isolator and exposed to

500 ppb VPHP. The 1 mL long syringes were exposed for up to 24 h with three syringes sampled

at each time-point for Amplex® testing. The results showed very low levels of VPHP uptake for

all syringes, in the range of 1–2 ng H2O2 per syringe (Figure 8a), which is below assay precision.

There was no increasing uptake trend with exposure time, and syringe-to-syringe variability was

very low (<10% RSD which is within the range of the Amplex® assay). Two sets of 2.25 mL

syringes were tested separately on different days (nine syringes on each day) for VPHP exposure

for 4 h in the isolator. Consistent results were observed: low uptake (2–3 ng H2O2 per vial) and

low syringe-to-syringe variability (Figure 8b).

VPHP uptake behaviors of glass syringes are consistent with those of hydrophobic vials (section

3.2.3). The inner surfaces of these glass syringes are indeed hydrophobic as they have been

coated with a thin layer of silicone oil for the purpose of reducing injection force. Given this

assessment, the contribution of the empty glass syringe to overall H2O2 absorption by the DP

18
 Downloaded from journal.pda.org on July 4, 2019

solution is considered negligible. Syringes requiring no silicone oil coating, such as Crystal

Zenith® syringes made of cyclic olefin polymer, should be assessed separately for H2O2

absorption/adsorption, which is outside the scope of this study.

3.4 VPHP uptake by butyl rubber stoppers

Three types of vial stoppers (20 mm liquid and lyophilization stoppers and 13 mm liquid

stoppers) were placed in the isolator and exposed to 500 ppb VPHP for up to 48 h. All stoppers

had low levels of uptake, <7 ng H2O2 per 20 mm stopper (liquid and lyophilization) and <4 ng

per 13 mm stopper with low stopper-to-stopper variability (<10% RSD) (Figures 9a and 9b).

Again, the amount of H2O2 detected was below assay precision. All of these stoppers were

coated with hydrophobic Flurotec® coating, so they showed consistent uptake similar to

hydrophobic glass vials and syringes. The quantity of H2O2 that each stopper absorbed is

negligible. Although many rubber stoppers for pharmaceutical applications are coated to

minimize stopper-drug interactions, un-coated stoppers should be evaluated separately to

understand their H2O2 absorption/adsorption behaviors, which is beyond the scope of this study.

In addition, syringe stoppers were not tested in this study, but the same Flurotec ® coating is

expected to yield similar results with negligible VPHP uptake.

4. Conclusions

Among the product-contacting components investigated in this study, primary packaging

components adsorb H2O2 regardless of their material of construction. Although primary

packaging components may not be considered major contributors to overall VPHP uptake, their

H2O2 adsorption may still present product quality risks and should be assessed based on product

specific configurations. Silicone tubing plays a more critical role and can contribute significantly

19
 Downloaded from journal.pda.org on July 4, 2019

to overall H2O2 absorption by the DP solution. Silicone tubing displayed unique H2O2 absorption

patterns in response to sterilization methods and demonstrated the complicated nature of VPHP

uptake behaviors. A common phenomenon shared by all components is that any measure capable

of lowering component water content or making component surfaces more hydrophobic would

reduce their VPHP uptake tendency and minimize adsorption variability. The knowledge gained

from this study could help development scientists and engineers prioritize their study strategy

and better select product-contacting materials for protein products that are sensitive to H2O2-

induced oxidation.

Acknowledgments

The authors would like to thank Dr. Philippe Lam for his work on air-flow computer modeling.

Conflict of Interest Declaration

The authors declare that they have no competing interests.

References

1. Müller, H.J.; Sommermeyer, K.; Cech, F. Permeation of gaseous hydrogen peroxide and

peracetic acid into IV bags during their surface sterilization. Hospital Pharmacist. 2003,

10, pp. 81-86.

2. Meszaros, J.E., Antloga, K., Justi, C., Plesnicher, C., McDonnell, G. Area fumigation

with hydrogen peroxide vapor. Applied Biosafety. 2005, 70, pp. 91-100.

3. Unger-Bimczok, B.; Kottke, B.; Hertel, V.; Rauschnabel, J. The influence of humidity,

hydrogen peroxide concentration, and condensation on the inactivation of Geobacillus

20
 Downloaded from journal.pda.org on July 4, 2019

Stearothermophilus spores with hydrogen peroxide vapor. J. Pharm. Innov. 2008, 36, pp.

123-133.

4. Krishna, A.K.; Lodhi, S.A.; Harris, M.R. Isolation technology for research and

development applications: from concept to production. Pharm. Dev. Technol. 2000, 5, pp.

507-520.

5. Thirumangalathu, R.; Sampathkumar, K.; Bondarenko, P. Oxidation of methionine

residues in recombinant human interleulin-1 receptor antagonist: Implication of

conformational stability on protein oxidation kinetics. Biochemistry. 2007, 46, pp. 6213-

6224.

6. Liu, J.L.; Lu, K.V.; Eris, T.; In Vitro Methionine Oxidation of Recombinant Human

Leptin. Pharm. Res. 1998, 15, pp. 632-640.

7. Simat, T.J.; Steinhart, H. Oxidation of Free Tryptophan and Tryptophan Residues In

Peptides and Proteins. J Agric Food Chem. 1998, 46, pp. 490-498.

8. Requena, J.R.; Dimitrova, M.N.; Legname, G.; Teijeira, S.; Prusiner, S.B.; Levine, R.L.

Oxidation of Methionine Residues in the Prion Protein by Hydrogen Peroxide. Arch

Biochem Biophys. 2004, 432, pp. 188-195.

9. Mulinacci, F.; Poirier, E.; Capelle, M.A.; Gurny, R.; Arvinte, T. Influence of Methionine

Oxidation On the Aggregation of Recombinant Human Growth Hormone. Eur. J Pharm.

Biopharm. 2013, 85, pp. 42-52.

21
 Downloaded from journal.pda.org on July 4, 2019

10. Hubbard, A.; Eppler, K. Measuring Hydrogen Peroxide Residuals and Product Exposure

on Isolator and RABS Filling Lines. BioPharma Asia. 2015, 4, pp. 26-33.

11. Saller, V., Matilainen, J., Rothkopf, C., Serafin, D., Bechtold-Peters, K., Mahler, H.-C.,

Friess, W. Preservative Loss from Silicone Tubing during Filling Processes. Eur.

J Pharm. Biopharm. 2017, 12, pp. 109-118

12. Colas, A., Malczewski, R., Ulman, K. Silicone Tubing for Pharmaceutical Processing.

PharmaChem. 2004, 30-36.

13. Wijmans, J.G., Baker, R.W. The Solution-Diffusion Model: A review. J Membr. Sci.

1995, 107, 1-21

14. Hubbard, A., Roedl, T., Hui, A., Knueppel, S., Eppler, K., Lehnert, S., Maa, Y.-F. Vapor

Phase Hydrogen Peroxide Sanitization of an Isolator for Aseptic Filling of Monoclonal

Antibody Drug Product – Hydrogen Peroxide Uptake and Impact on Protein Quality,

PDA J Pharm. Sci. Technol. 2018, 72, 348-366.

15. LaVerne, L. Sterilization - Tough on Germs, Tough on Plastics Too. Plastics Des. Forum

1994, 19, 16-22.

16. Gautriand, E., Stafford, K.T., Adamchuk, J., Simon, M.W. Effect of Sterilization on the

Mechanical Properties of Silicone Rubbers. BioProcess International Supplement, 2010,

April, Chapter 3.

17. Rogers, W.J., Sterilisation of Polymer Healthcare Products. 2005. in: Rapra Technology

(edited by Strewbury, U.K), pp. 24-143.

22
 Downloaded from journal.pda.org on July 4, 2019

18. Palsule, A.S., Clarson, S.J., Widenhouse, C.W. Gamma Irradiation of Silicones. Journal

of Inorganic and Organometallic Polymers and Materials, 2008, 18, 207-221.

19. Kiminami, H., Krueger, A.B., Abe, Y., Yoshino, K., Carpenter, J.F. Impact of

Sterilization Method on Protein Aggregation and Particle Formation in Polymer-Based

Syringes. J Pharm Sci, 2017, 106, 1001-1007.

20. Chen, L., Sloey, C., Zhang, Z., Bondarenko, P.V., Kim, H., Ren, D., Kanapuram, S.

Chemical Modifications of Therapeutic Proteins Induced by Residual Ethylene Oxide. J

Pharm Sci, 2015, 104, 731-739.

21. da Cunha Mendes, G.C., da Silva Brandao, T.R., Silva, C.L.M. Ethylene Oxide Potential

Toxicity. Expert Rev. Med. Devices, 2008, 5, 323-328.

22. Konda, N., Prashanth, P., Arvind, G., Shah, S. Pharmaceutical Development and

Compatibility Studies on Cytarabine Injection. Asian J. Pharm. Clin. Res., 2013, 6, 142 –

145.

23. American Coatings Association website, Pharmaceutical Vial Coatings Make a Big

Difference, https://www.paint.org/article/pharmaceutical-vial-coatings-make-big-

difference (accessed February 5, 2018)

24. Schott website, SCHOTT TopLyo ® Vials,

http://www.schott.com/d/pharmaceutical_packaging/7f629b7e-e978-4417-a15e-

8621f969d225/1.1/schott-datasheet-schott-toplyo-english-14062017.pdf (accessed

February 5, 2018).

23
 Downloaded from journal.pda.org on July 4, 2019

25. Westpharma website, FluroTec Barrier Film® Stoppers,

https://www.westpharma.com/products/vial-containment-solutions/stoppers/flurotec-

barrier-film-stoppers (accessed February 5, 2018).

26. Hui, A.; Lam, X.M.; Kuehl, C.; Grauschopf, U.; Wang, Y.J. Kinetic Modeling of

Methionine Oxidation in Monoclonal Antibody from Hydrogen Peroxide Spiking

Studies. PDA J Pharm Sci Technol. 2015, 69, pp. 511-525.

27. Zhang, H.; Cloud, A. The permeability characteristics of silicone rubber. SAMPE Fall

Technical Conference on Global Advances in Materials and Process Engineering,

November 6-9, 2006, Dallas, Texas, USA.

28. Traeger, R.K.; Castonguay, T.T. Effect of γ-Radiation on the Dynamic Mechanical

Properties of Silicone Rubbers. J. Appl. Polym. Sci. 1966, 10, pp. 535–550.

29. Mazor, E.; Zilberman, M. Effect of Gamma-Irradiation Sterilization on the Physical and

Mechanical Properties of a Hybrid Wound Dressing. Polym Adv Technol, 2017, 28, pp.

41-52.

24
TABLE I
Components used in VPHP uptake experiments
Components /
Model Dimensions Manufacturer Coating Pre-treatment/Sterilization
Material
Tubings
1.6 mm & 3.2 mm Autoclaved: 121 °C for 60 min and
Silicone Sani-Tech ID (both with 1.6 Saint Gobain NA dried under vacuum
mm wall thickness) Gamma-irradiated: 25–40 KGy
Vials
Clear tubing Washed and depyrogenated, humidity
NA 20, 50 mL Schott NA
borosilicate glass chamber, over-depyrogenated
TopLyo
Clear tubing From manufacturer, coated, washed
NA 50 mL Schott Hydrophobic
borosilicate glass and depyrogenated
Coating
Syringes
HYPAKSCF1
Borosilicate glass 1 mL BD Silicone Ready to use from manufacturer
MLLS27G1/2
HYPAKSCF22
Borosilicate glass 2.25 mL BD Silicone Ready to use from manufacturer
5MLLS27G1/2
Stoppers (for vials or syringes)
Butyl rubber
19459 13 mm liquid West / Daikyo Flurotec/RB2-40 Autoclaved
D777-1 (vial)
Butyl rubber
17070 20 mm liquid West / Daikyo Flurotec/RB2-40 Autoclaved
D777-1 (vial)
Butyl rubber
17069 20 mm lyo West / Daikyo Flurotec/RB2-40 Autoclaved
D777-1 (vial)
Butyl rubber
RSH 6.75F RS 1mL Daikyo RB2-40 Autoclaved
D777-7 (syringe)
Butyl rubber
RSH 9.0F RS 2.25mL Daikyo RB2-40 Autoclaved
D777-7 (syringe)

25
ID: internal diameter; NA: not available

26
TABLE II

Impact of hydration and dehydration on VPHP uptake of autoclaved and non-autoclaved tubing

VPHP uptake of processed 1.6 mm ID tubing relative

to non-autoclaved control (%)
Pre-process
1 day post 2 days post 6 days post 14 days post
process process process process
Autoclaved tubing
28 ± 1 20 ± 1 37 ± 2 39 ± 3
(121 °C for 60 min)
Water incubated up to 7 h
42 ± 4 NT NT NT
in autoclaved tubing
Drying at 130 °C for 12 h 54 ± 5 NT NT NT
Vacuum 59 ± 7 58 ± 6 69 ± 5 67 ± 4
NT: Not tested

27
 Downloaded from journal.pda.org on July 4, 2019

Figure Captions

Figure 1: (a) VPHP uptake into silicone tubing (1.6 mm ID) 2 days after autoclaving. Tubing

was primed with water and exposed to 100 ppb VPHP in the isolator for 30, 60, and 90

min, prior to sampling and Amplex® testing; (b) VPHP uptake into silicone tubing (1.6

mm ID) 14 days after autoclaving. Tubing was primed with water and exposed to 100

ppb VPHP in the isolator for 30, 60, and 90 min prior to sampling and Amplex®

testing.

Figure 2: a) VPHP uptake into silicone tubing (3.2 mm ID) 3 days after gamma irradiation.

Tubing was primed with water and exposed to 100 ppb VPHP for 30 and 90 min prior

to sampling and Amplex® testing; (b) VPHP uptake into silicone tubing (3.2 mm ID) 8

days after gamma irradiation. Tubing was primed with water and exposed to 100 ppb

VPHP for 15, 30, and 60 min prior to sampling and Amplex® testing; (c) VPHP uptake

into silicone tubing (1.6mm ID x 1.6mm wall) 29 days after gamma irradiation.

Tubing was primed with water and exposed to 100 ppb VPHP for 15, 30, and 60 min

prior to sampling and Amplex® testing.

Figure 3: VPHP uptake by 50 cc glass vials (red) in the isolator in response to three levels of

VPHP concentration in the sequence of (1) 50 ppb, (2) 100 ppb, (3) 500 ppb, and back

to (4) 50 ppb (blue). Three vials were analyzed by Amplex® testing at multiple time-

points within each level of VPHP exposure.

28
 Downloaded from journal.pda.org on July 4, 2019

Figure 4: VPHP uptake by (a) empty 50 cc glass vials (b) and 20 cc glass vials in the isolator

after exposure to 500 ppb VPHP for different durations (up to 48 h). Six vials were

analyzed at each time-point by Amplex® testing.

Figure 5: Computer modeling simulating isolator air flow inside and around the vial with uni-

directional air blowing down toward the vial opening at a velocity of 0.7 m/s.

Figure 6: Day-to-day variability of VPHP uptake by 19 empty 50 cc glass vials (each line

representing one vial). All vials were pre-conditioned using the same depyrogenation

conditions and the same procedures for vial placement, removal, incubation, and

analytical testing to avoid procedural variations.

Figure 7: VPHP uptake tendency of 50 cc vials after 4 h exposure to 500 ppb in the isolator. The

five vial groups are listed as the control or standard processed vial, and then in order of

decreasing glass surface water content. Each vial group was an average of 10 vials.

Figure 8: (a) VPHP uptake tendency by empty 1 mL long glass syringes after exposure to 500

ppb VPHP concentration in the isolator for different durations (up to 24 h) with three

samples taken at each time-point for Amplex® testing; (b) VPHP uptake by empty 2.25

mL glass syringes tested separately on two different days (9 syringes on each day) for

VPHP exposure of 4 h in the isolator.

29
 Downloaded from journal.pda.org on July 4, 2019

Figure 9: VPHP uptake by three types of vial stoppers – (a) 20 mm liquid and lyophilization

stoppers and (b) 13 mm liquid stoppers after exposure to 500 ppb VPHP concentration

in the isolator for different durations (up to 48 h).

30
 Downloaded from journal.pda.org on July 4, 2019

Figure 1

(a)

(b)

31
 Downloaded from journal.pda.org on July 4, 2019

Figure 2

(a)

(b)

32
 Downloaded from journal.pda.org on July 4, 2019

(c)

Figure 3

33
 Downloaded from journal.pda.org on July 4, 2019

Figure 4

(a)

(b)

34
 Downloaded from journal.pda.org on July 4, 2019

Figure 5

Figure 6

35
 Downloaded from journal.pda.org on July 4, 2019

Figure 7

36
 Downloaded from journal.pda.org on July 4, 2019

Figure 8

(a)

(b)

37
 Downloaded from journal.pda.org on July 4, 2019

Figure 9

(a)

(b)

38
 Downloaded from journal.pda.org on July 4, 2019

An Authorized User of the electronic PDA Journal of Pharmaceutical Science and

Technology (the PDA Journal) is a PDA Member in good standing. Authorized Users are
permitted to do the following:

·Search and view the content of the PDA Journal

·Download a single article for the individual use of an Authorized User
·Assemble and distribute links that point to the PDA Journal
·Print individual articles from the PDA Journal for the individual use of an Authorized User
·Make a reasonable number of photocopies of a printed article for the individual use of an
Authorized User or for the use by or distribution to other Authorized Users

Authorized Users are not permitted to do the following:

·Except as mentioned above, allow anyone other than an Authorized User to use or access the
PDA Journal
· Display or otherwise make any information from the PDA Journal available to anyone other than an
Authorized User
·Post articles from the PDA Journal on Web sites, either available on the Internet or an Intranet, or in
any form of online publications
·Transmit electronically, via e-mail or any other file transfer protocols, any portion of the PDA
Journal
·Create a searchable archive of any portion of the PDA Journal
·Use robots or intelligent agents to access, search and/or systematically download any portion of the
PDA Journal
·Sell, re-sell, rent, lease, license, sublicense, assign or otherwise transfer the use of the PDA
Journal or its content
·Use or copy the PDA Journal for document delivery, fee-for-service use, or bulk reproduction or
distribution of materials in any form, or any substantially similar commercial purpose
·Alter, modify, repackage or adapt any portion of the PDA Journal
·Make any edits or derivative works with respect to any portion of the PDA Journal including any text
or graphics
·Delete or remove in any form or format, including on a printed article or photocopy, any copyright
information or notice contained in the PDA Journal