Annals of Occupational Hygiene Advance Access originally published online on October 26, 2004
Annals of Occupational Hygiene 2004 48(8):673-681; doi:10.1093/annhyg/meh066
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© British Occupational Hygiene Society Published by Oxford University Press;
Evaluation of Exposure and Health Care Worker Response to Nebulized Administration of tgAAVCF to Patients with Cystic Fibrosis
1 University of Washington, Department of Environmental and Occupational Health Sciences, Field Research & Consultation Group, 4225 Roosevelt Way NE, Suite 100, Seattle, WA 98105-6099, USA; 2 Targeted Genetics Corporation, 1100 Olive Way, Suite 100, Seattle, WA 98101, USA; 3 Stanford University, 701 Welch Road, Suite 3328, Department of Pediatrics, Palo Alto, CA 94305, USA; 4 The Johns Hopkins University, Pediatric Respiratory Sciences, 600 N. Wolfe St., Park 316, Baltimore, MD 21287, USA
* Author to whom correspondence should be addressed. Tel: +1-206-543-9711; fax: +1-206-616-6240; e-mail: croteau{at}u.washington.edu
Received 2 February 2004; in final form 9 July 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
A study was conducted to assess health care worker exposure to tgAAVCF during the aerosolized administration of this experimental gene transfer agent in clinical trials for the treatment of cystic fibrosis (CF). tgAAVCF is a recombinant adeno-associated virus (AAV) genetically engineered to contain the human CF transmembrane conductance regulator cDNA. Study subjects included eight health care workers involved in the administration of tgAAVCF in a phase II study and 12 control health care workers who were involved with the treatment of CF patients, but not administration of the study drug. The exposure assessment entailed the determination of personal and area airborne tgAAVCF concentrations. In addition, serologic status of the health care workers was evaluated throughout the study for the presence of antibodies to AAV. A symptom survey was also completed by both the active and control health care workers. Air samples were analyzed by an infectivity assay (active vector) and a DNA polymerase chain reaction amplification procedure (vector DNA). Air monitoring was conducted during 13 tgAAVCF and seven placebo administrations. Active vector and vector particles were detected in four of 51 and 48 of 51 air samples collected during the administration of tgAAVCF, respectively. Based on the airborne vector particle concentration, the workers' exposure was estimated to be 0.0006% of the administered dose. At this level of exposure, the prevalence of symptoms was very low, the spectrum was similar in both study groups and did not result in any reported negative health effects.
Keywords: aerosol sampling • airborne virus sampling • cystic fibrosis • health care worker
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Health care workers can potentially be exposed to drugs administered as an aerosol to a patient's pulmonary system. This drug administration method entails the use of a nebulizer and an air pump to aerosolize an aqueous formulation of the drug. The aerosolized drug can become airborne directly by escaping from the nebulizer or indirectly as a result of exhalation from the patient. This type of occupational exposure has been examined previously with respect to the administration of ribavirin (Harrison et al., 1988
; Shults et al., 1996; Krilov, 2002) and pentamidine (Balmes et al., 1995). tgAAVCF is a gene transfer agent under development for the treatment of cystic fibrosis (CF) that health care workers may potentially be exposed to when the drug is administered to patients as an aerosol. The efficacy and safety of this drug is currently being assessed in clinical trials, one of which included this occupational exposure assessment.
CF is one of the most common autosomal recessive diseases, affecting between 1 in 2000 and 1 in 4500 children of Caucasian origin. The gene involved in CF, the CF transmembrane conductance regulator (CFTR), encodes for a protein that functions, at least in part, as a cAMP-regulated chloride channel on the epithelium surface. This mutation of the CFTR gene results in a deficiency of cAMP-regulated epithelial cell chloride secretion (Koch and Hoiby, 1993; Rosenfeld et al., 1993).
Gene transfer holds the promise of addressing the primary defect in CF by reconstituting proper chloride transport in vivo. Adeno-associated virus (AAV) vectors have utility as in vivo gene transfer agents as they are capable of high-frequency, stable gene transfer and expression in a variety of cells including CF bronchial and nasal epithelial cells (Carter, 1992; Egan et al., 1992; Flotte et al., 1992, 1993). In addition, AAV is not known to cause any disease, is not a transforming or oncogenic virus and human cell line integration of AAV does not substantially alter cell growth properties or morphological characteristics. Finally, highly purified and concentrated AAV vectors can be produced using standard biopharmaceutical manufacturing methods. tgAAVCF is a modified AAV, genetically engineered to contain the CFTR cDNA.
During earlier safety and efficacy evaluations of tgAAVCF, aerosolized material was delivered to patients enclosed in mist tents. Refinement in aerosol delivery systems that limit fugitive emissions now allows tgAAVCF to be administered without containment, thus improving patient comfort. Once fully developed, tgAAVCF will most likely be administered by a health care worker in a hospital or CF clinic, raising the possibility of low-level worker exposure to aerosolized vector. The magnitude and effect of this exposure has not been described previously. This study was conducted to provide a preliminary assessment of health care worker exposure to tgAAVCF through the collection of personal and area samples during nebulization. The study also determined whether there was any change in the serologic status of health care workers as a result of exposure.
|
METHODOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This 22-month study was conducted in conjunction with a multicenter, double-blind, placebo-controlled, phase II study of aerosolized tgAAVCF for the treatment of CF patients with mild lung disease. In the phase II study, which was 20 months in length, 37 patients with CF received up to three doses of 1 x 1013 DNase resistant particles (DRP) tgAAVCF (n = 20) or placebo (n = 17) delivered at 30 day intervals via oral aerosol (Moss et al., 2004
). All study participants were blinded to which study drug (tgAAVCF or placebo) was being administered in the phase II study.
Study sites and participants
The exposure assessment study was conducted at six of the eight CF centers that participated in the phase II study. Health care worker participants were recruited from all six sites. Air sampling was conducted at three of the sites. Twenty health care worker participants were enrolled in the study. The active group consisted of eight health care workers who administered study drug to CF patients. The control group consisted of 12 health care workers who provided care to CF patients in the study center, but were not administering study drug and had not worked with any AAV products in the preceding 6 months. During drug administration, the health care workers used the following respiratory protective equipment: surgical masks at Sites 1, 3 and 5; N-95 filtering face-pieces at Sites 2 and 4 and N-100 filtering face-pieces at Site 6. All health care workers gave written informed consent prior to study participation.
Study drug delivery system
The study drug was aerosolized and delivered using a Pari LC Plus nebulizer and Proneb compressor (PARI, Richmond, VA). The compressor had a rated output of 4.0 l/min, resulting in a study drug delivery rate of 0.32 ml/min and an aerosol with a mass median aerodynamic diameter of 4.0 µm (Pari literature). The nebulizer was equipped with an AutoNeb controller (Vortran Medical Technologies, Sacramento, CA), which limits fugitive emissions that might occur during delivery. A valve, which provides for unidirectional airflow, ensures that aerosolized drug is only delivered during inspiration and exhaled breath is passed through a filter housed in the nebulizer before being released into the treatment room.
Air sampling
Air sampling to assess health care worker exposure to AAV vector was conducted during 20 administrations of study drug, which included 13 administrations of tgAAVCF and seven administrations of placebo. Each air-sampling train consisted of a midget impinger, containing 15 ml of formulation buffer, connected to a personal air-sampling pump by way of plastic tubing. A high efficiency particulate air (HEPA) filter cartridge was placed between the impinger and the sampling pump to ensure that any vector passing through the impinger would not contaminate the sampling pump. The airflow rate through each sampling train was set to 1.5 l/min using either a rotameter or piston airflow measuring device. At the conclusion of air sampling, the airflow rate through the sampling train was measured with the same equipment used to set the airflow sampling rate.
Air monitoring entailed the collection of a personal sample from the breathing zone of the health care worker during drug administration. The personal air sample was collected by a midget impinger placed in the left breast pocket of a lab coat worn by the health care worker who moved about the room during study drug administration. In addition, area samples (3) were collected from a location adjacent to the patient (<60 cm) and at locations 1.5 and 3 m away from the patient. At each sampling event, the midget impingers used for area sampling were placed at the same height, which ranged from 1.2 to 1.7 m above the floor between the different sites.
Upon completion of air sampling, the impinger contents were transferred to prelabeled conical centrifuge tubes, which were immediately placed in a cooler containing dry ice and were ultimately stored in a –60°C freezer. Samples were shipped to the analytical lab in coolers containing dry ice using a standard chain of custody procedures. Upon receipt, samples were transferred to a –60°C freezer where they were stored until analysis. Data collected during sampling included sampling location, sampling start and stop times, pre- and post-sampling airflow rate, volume of formulation buffer in impinger at start and end of sampling, and start and stop times of study drug administration. At the completion of each drug administration, the impingers and plastic tubing were disposed of as biohazardous waste. The air-sampling pumps and HEPA filters were wiped with a 10% sodium hypochlorite solution.
Treatment room characterization
Characteristics of the room, including ambient temperature, relative humidity, room number, room dimensions and air handling (positive, negative or neutral pressure), were recorded. At Site 3, the room ventilation rate was determined after each study drug administration, by measuring the airflow rate from a diffuser. At Sites 1 and 5, the airflow rate was only determined prior to the first air-sampling event. Building or room ventilation rate is expressed as air exchanges per hour, which is the volume of air delivered to the room in an hour divided by the room volume. The treatment room door was closed during the period when the study drug was administered. At study Sites 3 and 5, the same room was used for each drug administration, whereas, three different rooms were used at Site 1.
Air sample analysis
Vector concentration in air was measured in two ways. Active vector, or vector that could potentially transduce human cells, was measured using an infectivity assay (Atkinson et al., 1998; Moss et al., 2004). Briefly, dilutions of impinger fluid samples were placed onto cells carrying the AAV2 rep and cap genes that were pre-infected with adenovirus type 5. Infectious vector present in the sample would replicate and could then be detected by hybridization. The titer of the sample was reported by averaging all the dilutions that were within the standard curve range of the assay.
Vector particles were measured using a more sensitive DNA polymerase chain reaction (PCR) amplification procedure (Moss et al., 2004). Impinger samples were analyzed using vector-specific primers. The presence of vector genomes was detected and quantitated in real time using a fluorescent vector-specific probe.
The concentration of active vector and vector particles in air were calculated for each sampling site using equation (1).
 | (1) |
Antibodies to AAV2
Antibodies to AAV2 were measured in the serum of both active and control health care worker participants. Serum was collected from the active study group at study entry and every 30 (±3) days thereafter until the conclusion of the study. Serum was collected in the control study group at study entry and at the conclusion of the study.
Using an infectivity bioassay, serum samples were analyzed for the presence of antibodies that can neutralize AAV2 (Atkinson et al., 1998; Moss et al., 2004). An ELISA assay was also used to detect antibodies present in serum that can bind to AAV2 (anti-AAV2 IgG antibodies). The change in titer between the first and subsequent samples was calculated and determined to be biologically significant if a 4-fold or greater rise was observed.
Symptom surveys
Symptoms were assessed through the use of a survey administered to both the active and control health care worker participants. The active study group was surveyed at study entry and every 30 (±3) days thereafter until the conclusion of the study. The control study group was assessed at study entry and at the conclusion of the study. The survey solicited a positive or negative response regarding the presence of 13 symptoms. The list of symptoms included those reported by patients in early stage trials and those observed during animal trials. In contrast to the development of AAV2 antibodies, the presence of symptoms may not have been caused by exposure to tgAAVCF.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Baseline demographics
The demographics of the active (n = 8) and control (n = 12) study participants were similar (Table 1). Most were white females about 40 years of age, and approximately half had children (<18 years old) at home. There were no significant differences in sex, race or age between the two groups (P < 0.05). Three of the eight health care workers in the active study group had administered tgAAVCF in the phase II study prior to the initiation of this study. None of the control group participants had been exposed to AAV vectors prior to or during their enrollment in this study. All participants completed the study. The active study group administered the tgAAVCF 8.4 ± 4.3 (SD) times and placebo 7.4 ± 5.4 (SD) times throughout the duration of the study.
|
Air sampling
Air samples were collected during 13 tgAAVCF and seven placebo administrations at three of the six study sites, resulting in 80 possible samples. A single sample was lost during air sampling and assay results were inconclusive for one sample tested for active vector and two samples tested for vector particles. The final air monitoring data set was comprised of 78 and 77 data points for the active vector and vector particle analysis, respectively. The data was found to be approximately log normally distributed, and was log10 transformed prior to analysis. The mean sample collection time (±SD) was 36.1 (±7.9) min per study drug administration.
With one exception, neither active vector nor vector particles were detected in air samples collected during the administration of placebo. The exception occurred on one occasion at Site 1, where vector particles were detected at a concentration of 2484 copies/l of air sampled at the 3 m monitoring location and active vector was found at the patient, at 1.5 and 3 m sampling locations at concentrations of 495, 488 and 658 i.u./l of air sampled, respectively. The detection of tgAAVCF during the administration of placebo was attributed to specimen contamination, possibly from a tgAAVCF administration to another patient in a different room by the same health care worker at the same site, approximately 3 h earlier. During this prior administration of tgAAVCF, active vector was detected at a concentration of 723 i.u./l air sampled at the 3 m location, and vector particles were found at the patient, at 1.5 and 3 m monitoring locations at 1673, 1436 and 2303 copies/l of air sampled, respectively. Unless otherwise noted, all discussion from this point on will be in reference to air monitoring conducted during administration of tgAAVCF only.
Active vector was detected in only four of the 51 air samples collected during the administration of tgAAVCF (Table 2). Of the four detectable samples, three were from Site 3 (two from the same tgAAVCF administration) and one was from Site 1. The airborne active vector concentrations for the four detectable samples were 405, 489, 723 and 971 i.u./l of air sampled. The number of vector particles in these samples were 3029, 1855, 2303, and 2245 copies/l air sampled, respectively, corresponding to vector particle to active vector ratios of 7.5, 3.8, 3.2 and 2.3, suggesting that the vector was not inactivated by the nebulizer. In addition, this low concentration of active vector in air suggests that the possibility of transducing a person is low, which is borne out by the serum AAV2 antibody data.
|
Vector particles, in contrast to active vector, were detected in 48 of 51 samples (Table 2). Only detectable samples were included in subsequent data analysis. By monitoring location, the geometric mean airborne vector particle concentration ranged from 601 copies/l air sampled at the 1.5 m monitoring location at Site 5 to 6942 copies/l air sampled at the 3 m monitoring location at Site 1. The overall geometric mean airborne vector particle concentration for all data was 1984 copies/l of air sampled. Overall, there was considerable variability in the vector particle air monitoring results.
Boxplots were constructed to identify possible relationships between airborne vector particle concentration and the independent variables of study site and sampling location (Fig. 1). Mean airborne vector particle concentrations at Site 1 are noted to be approximately three times that of Sites 3 and 5. The geometric mean vector particle concentration of all samples for Site 1 was 3839 copies/l air sampled, whereas it was 1314 and 933 copies/l air at Sites 3 and 5, respectively. An ANOVA model of vector particle (log10 transformed) as a dependent variable and study site and sampling location as independent variables confirmed this observation, as study site was found to be a significant (P < 0.05) predictor of exposure.
|
Boxplots (Fig. 1) of the geometric mean vector particle concentration, clustered by monitoring location for each study site, do not indicate a strong relationship between proximity of the area monitoring locations to the source (patient) and vector particle concentration. However, at Site 1, the airborne vector particle concentration is observed to be greater at the 1.5 and 3 m area monitoring locations than that of the patient area monitoring location. At Sites 3 and 5, no apparent trend was noted. Sampling location was not found to be a significant predictor (P > 0.05) of airborne vector particle concentration in the ANOVA model described above. This observation indicates that no measurable airborne vector particle concentration gradient was present and further suggests that airborne vector particles were well mixed throughout the room volume.
Treatment room environment
Room temperature during study drug administration at the three study sites was noted to be similar (Table 3), ranging from 21.3°C at Site 5 to 22.6°C at Site 3. The relative humidity at Sites 3 and 5 was noted be similar at 32.4 and 34.8%, respectively. A higher relative humidity of 47.2% was observed at Site 1. Given the substantial variability in humidity and the small differences in temperature across sites, it seems unlikely that these environmental factors had a substantial effect on airborne vector sampling results.
|
Considerable differences were observed in room ventilation between the different study sites. With respect to the net flow of air through the treatment room into the adjacent hallway, Sites 1, 3 and 5 were observed to be neutral, positive and negative, respectively. There is insufficient data to comment on how this difference may have affected airborne exposure levels.
The room air exchange rate for Sites 3 and 5 at 8.8 and 11.5 air exchanges per hour respectively, are noted to be similar. At Site 1, no ventilation air was entering Room A through the primary diffuser during the initial study drug administration. Ventilation for this room was provided solely through the bathroom ventilation diffuser, resulting in a very low estimated ventilation rate of 3.9 air exchanges per hour. At some unknown point in time, the primary ventilation to this room was restored. The ventilation system for the adjacent Room B, which was used twice as a treatment room during this study, was operating during the initial study drug administration and the ventilation rate was found to be 19.9 air exchanges per hour.
AAV2 antibodies
None of the active or control study group participants were observed to have a 4-fold or greater increase in either neutralizing or IgG antibodies for AAV2 between their baseline and subsequent study visits. Positive titers (
1 : 4) to AAV2 neutralizing antibody were noted for three of the control participants at baseline and final study visit. Likewise, positive titers (1 : 64) to anti-AAV2 IgG antibodies were noted for two of the control participants at baseline; however, only one of the control participants exhibited a positive titer to anti-AAV2 IgG antibody at study conclusion. None of the active study group participants had a positive titer for either neutralizing or IgG antibodies for AAV2 at baseline or any of the subsequent serum analyses.
These results indicate that the active study group's exposure to tgAAVCF was not great enough to elicit a detectable antibody response. In previous studies, seroconversion for neutralizing antibodies, defined as a 4-fold or greater titer increase, was consistently observed in CF patients exposed to 1 x 1013 DRP of tgAAVCF, whereas CF patients exposed to lower doses did not exhibit an increase in neutralizing antibodies (Aitken et al., 2001; Moss et al., 2004). Potential worker exposure to tgAAVCF during nebulization was estimated to be 
6.1 x 107 DRP during each administration (Table 4). However, the dose required to induce an antibody response in healthy individuals is unknown and may be different from that required in CF subjects because of a number of factors, including differing deposition of inhaled dose, differing availability of antigen to immune cells and differing responsiveness of the immune system to the antigen.
|
Symptom survey
The symptom survey was administered 122 times, 98 times to the eight active study participants and 24 times to the 12 control study participants. Symptoms were reported on 37 of 122 surveys (30%). The most commonly reported symptoms, from all 122 surveys, were runny nose (21% of surveys), headache (12%), cough (10%), increased sputum (6%) and sore throat (4%). Fever, nausea and generalized joint or muscle pain were reported one time each. Bloody sputum, pleuritic chest pain, vomiting, shortness of breath and bleeding from nose or throat were never reported. At the baseline and final study visits (40 surveys), the prevalence of symptoms was very low, and the spectrum was similar in both study groups.
To determine if these symptoms were related to administration of tgAAVCF during the study, the 30 day period prior to each symptom survey was reviewed for all study participants. Active study participants were administered tgAAVCF once (n = 25) or twice (n = 6) in the 30 days prior to the symptom survey. Symptoms were reported 11 out of 31 times (34%) when the health care worker had administered tgAAVCF within the preceding 30 days, and 20 of 85 times (24%) when the health care worker had not administered tgAAVCF within the preceding 30 days. The proportion of symptom surveys that were positive was not found to be statistically different between the two groups (P = 0.47).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
With the advance of technology in science and medicine, clinical trials to assess new biopharmaceutical agents, such as recombinant DNA vectors, are occurring more commonly. The administration of biopharmaceuticals as an aerosol pose a greater exposure risk to health care workers than the administration of such drugs orally or intravenously. Consequently, occupational exposure assessment should be considered during safety and efficacy testing of biopharmaceuticals that are delivered as an aerosol. Risk–benefit evaluation is standard for study participants. What is not known, however, is the level at which the health care workers who participate in administering the study drug are exposed, and whether this exposure poses a health risk. If an aerosolized study drug is successfully developed, it is essential that health care workers' exposure to the drug, the health risk associated with the exposure and methods to control exposure are understood.
The study results indicate that relatively low airborne concentrations of tgAAVCF were detected in the treatment rooms where the study drug was administered. Vector particles were detected in 94% of the air samples collected. However, active vector was only detected in 8% of the air samples collected. The health care workers' exposure to tgAAVCF resulting from the administration of the study agent to the patient was estimated to be about 105 times less than the patient's administered dose of 1 x 1013 DRP.
Although the health care workers were in an environment documented to have vector in the air, none of the active participants developed a 4-fold or greater increase in either binding or neutralizing antibodies to AAV2. This supports the conclusion that the exposure level to tgAAVCF was less than that required to elicit a detectable antibody response in the health care workers who administered the vector. Similarly, results from the symptom survey did not exhibit any statistically significant clinical symptomology that would indicate that either infection or an inflammatory response had occurred as a result of exposure.
Increase in neutralizing antibody titer against AAV is a measure of exposure to AAV particles. Seroepidemiology studies indicate that AAV2 circulates in nature with its helper virus, adenovirus, and 20–60% of individuals have antibodies to AAV, with 10–30% of the population exhibiting neutralizing antibodies (Blacklow et al., 1968a,b, 1971; Chirmule et al., 1999; Moss et al., 2004). Thus, the observed neutralizing antibody titer against AAV2 at baseline in this group of participants may be due to prior AAV2 exposure. Lack of change in the AAV2 neutralizing antibody titer in the study participants is indicative of a low probability of exposure to AAV2 during the course of the study. It was beyond the scope of this study to evaluate whether or not the health care workers had adenoviral infections during the study, or if prior or current adenoviral infection increased or decreased the likelihood of seroconversion to AAV.
Applicability of the study results to health care workers administering tgAAVCF on a large scale, is limited by the small population of workers studied and the low number of tgAAVCF administrations. The active study population comprised eight health care workers who administered the study agent an average of 8.4 times during the study. However, if the study agent is commercialized, some health care workers could administer the agent more frequently than was observed in this study, resulting in a substantially greater drug administration rate for individual health care workers. Multiple study drug administrations per day and several times a week by health care workers is a possibility.
The accuracy of the airborne tgAAVCF exposure levels observed in this study may have been affected by the inlet sampling and particle removal efficiency of the midget impinger. Inlet sampling efficiency is the ability of the sampling device to entrain particles without bias with respect to particle size, shape and density. Particle removal efficiency is the ability of the sampling device to retain or trap particles after they become entrained. Overall, the inlet sampling efficiency of the midget impinger and the all-glass impinger is optimal for particles with a diameter of about 1.0 µm (Grinshpun et al., 1994, 1997; Spanne et al., 1997). However, inlet sampling and particle removal efficiency decline substantially with both increasing and decreasing particle diameter (Lyons, 1992; Grinshpun et al., 1994, 1997; Lin et al., 1997; Spanne et al., 1999; Streicher et al., 2000).
The actual collection efficiency achieved by the midget impingers in this study was dependent on the particle size distribution of airborne tgAAVCF in the treatment room, which was not determined. Given that no uncontrolled release of tgAAVCF from the nebulizer into the room was observed during any of the study agent administrations, exhaled breath from the patient is the likely source of airborne tgAAVCF and was probably largely comprised of particles with a submicron particle diameter. Consequently, low inlet sampling and particle removal efficiencies were probably obtained and the monitoring results presented may underestimate the actual airborne tgAAVCF concentrations.
Although the collection efficiency of the midget impinger is compromised, the use of aqueous media ensures a high biological recovery and its small size allows the device to be used for the collection of personal samples. For these reasons, the midget impinger was selected as an air-sampling device in this study. Currently, there are limited alternatives for the sampling of airborne viruses, especially if a live cell culture assay analysis is being used and there is a need to maintain virus viability. The Biosampler (SKC, Inc., Eighty Four, PA), a proprietary impinger sampling device, has been designed to provide better inlet sampling and particle removal efficiency than the midget impinger. However, the size of the Biosampler and associated air-sampling pump prohibit it from being used as a personal sampling device. A filter-type sampling device, such as the inhalable particulate sampler developed by the Institute of Occupational Medicine (IOM, Edinburgh, Scotland), could potentially be used in conjunction with PCR-based assays, but collection of the virus on a filter would reduce viability, prohibiting the use of an IOM sampler for live culture assays.
Health care worker's exposure to airborne tgAAVCF resulting from study agent administration is, to some degree, dependent on the treatment room environment. More specifically, removal of airborne tgAAVCF from the treatment room is dependent on the mechanical ventilation rate. In general, the rate of aerosolized tgAAVCF removal from a room increases concomitantly with the room's ventilation rate, resulting in a lower exposure level for the health care worker. In addition, particle size distribution as well as vector viability, are affected by the evaporation rate, which in turn is largely dependent on temperature and relative humidity of the treatment room.
The high airborne vector particle concentrations observed in Treatment Room A (Table 5) at Site 1 on 15 August 2001 may have been a result of this room's general ventilation system not operating that day. The Engineering Department at this site was notified and the system was subsequently repaired, although the exact date the system was repaired is not known. Airborne tgAAVCF levels in Treatment Room A were observed to be lower on 4 and 27 September 2002, possibly as a result of the ventilation system being repaired. These monitoring results suggest that the treatment room general ventilation system is an important factor for reducing the concentration of airborne vector particles during the administration of tgAAVCF.
|
Another observation of note during the study was the presence of airborne tgAAVCF in a treatment room in which the placebo was being administered. The same health care worker who administered placebo, also administered the study agent in a different treatment room 3 h previously. Consequently, it appears that the health care worker may have come in contact with tgAAVCF during the earlier administration and cross-contaminated the air samples obtained during the placebo administration, through direct contact. This observation suggests that tgAAVCF can remain viable for some period after administration, potentially resulting in exposure of individuals not associated with the tgAAVCF administration outside of the treatment area. The level of exposure to such individuals is likely to be considerably less than that of the health care workers and would therefore probably present a very low health risk.
Although the health care workers' exposure to airborne tgAAVCF was found to be low and the health risk associated with this exposure minimal, controls, like those used in this study, should be utilized to ensure that exposure levels to tgAAVCF during aerosolized administration remain low. The use of a nebulizer that limits both emissions directly from the device, as well as from the patient's exhaled breath, is highly recommended. In addition, the treatment room should have a ventilation system that provides at least 10, and preferably 20, air changes per hour. The ventilation system should also create a negative pressure environment with the exhaust being treated through an appropriate filter, ensuring that any tgAAVCF released in the treatment room is captured. Based on the experience gained in this study, health care workers should assess whether the ventilation system is operating. This can be accomplished simply by holding up a piece of paper to the room air diffusers to check if air is being delivered. To limit the dissemination of the study agent through contact, health care workers should use clothing and gloves that are disposed of after the study agent has been administered. Surfaces that may have been contaminated with the study agent should also be wiped down with a 10% sodium hypochlorite solution. The exposure levels and safety testing conducted to date do not appear to warrant the use of respiratory protection.
The health care workers' occupational exposure to tgAAVCF was estimated to be five orders of magnitude less than a dosage that has been found not to cause any deleterious health effects in CF patients. This low occupational exposure to tgAAVCF did not result in any negative health effects. Therefore, adherence to the control practices briefly described above should provide adequate protection to health care workers. However, as demonstrated in this study, health care workers and management staff need to be vigilant of breaches in exposure control practices, which could increase inadvertent exposure.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The authors would like to acknowledge the health care workers at the six research centers that participated in this study. We would also like to recognize the efforts of the Bioanalytics and Quality Assurance Departments at Targeted Genetics Corporation, especially Linda Wilson and Tara Allen.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Aitken ML, Moss RB, Waltz DA et al. (2001) A phase I study of aerosolized administration of tgAAVCF to CF subjects with mild lung disease. Hum Gene Ther; 12: 1907–16.[CrossRef][Web of Science][Medline]
Atkinson EM, Debelack DJ, Hart LA et al. (1998) A high-throughput hybridization method for titer determination of viruses and gene therapy vectors. Nucleic Acids Res; 26: 2821–3.
Balmes JR, Estacio PL, Quinlan P et al. (1995) Respiratory effects of occupational exposure to aerosolized pentamidine. J Occup Env Med; 37: 145–50.[CrossRef][Web of Science][Medline]
Blacklow NR, Hoggan MD, Kapikian AZ et al. (1968a) Epidemiology of adenovirus-associated virus infection in a nursery population. Am J Epidemiol; 88: 368–78.
Blacklow NR, Hoggan MD, Rowe WP. (1968b) Serologic evidence for human infection with adenovirus-associated viruses. J Natl Cancer Inst; 40: 319–27.[Web of Science][Medline]
Blacklow NR, Hoggan MD, Sereno MS et al. (1971) A seroepidemiologic study of adenovirus-associated virus infection in infants and children. Am J Epidemiol; 94: 359–66.
Carter BJ. (1992) Adeno-associated virus vectors. Curr Opin Biotechnol; 3(5): 533–9.[CrossRef][Medline]
Chirmule N, Propert K, Magosin S et al. (1999) Immune responses to adenovirus and adeno-associated virus in humans. Gene Ther; 6: 1574–83.[CrossRef][Web of Science][Medline]
Egan M, Flotte T, Afione S et al. (1992) Defective regulation of outwardly rectifying chloride channels by protein kinase A corrected by insertion of CFTR. Nature; 358: 581–4.[CrossRef][Medline]
Flotte TR, Solow R, Owens RA et al. (1992) Gene expression from adeno-associated virus vectors in airway epithelial cells. Am J Respir Cell Mol Biol; 7: 349–56.
Flotte TR, Afione SA, Conrad C et al. (1993) Stable in vivo expression of the cystic fibrosis transmembrane regulator with an adeno-associated virus vector. Proc Natl Acad Sci USA; 90: 10610–7.
Grinshpun SA, Chang CW, Nevalainen A et al. (1994) Inlet characteristics of bioaerosol samplers. J Aerosol Sci; 25: 1503–22.[CrossRef]
Grinshpun SA, Willeke K, Ulevicius V et al. (1997) Effect of impaction, bounce and reaerosolization on the collection efficiency of impingers. Aerosol Sci Tech; 26: 326–42.
Harrison R, Bellows J, Rempel D et al. (1988) Assessing exposures of health-care workers to aerosols of ribavirin: California. MMWR; 37: 560–3.[Medline]
Koch C, Hoiby N. (1993) The pathogenesis of cystic fibrosis. Lancet; 341: 1065–9.[CrossRef][Web of Science][Medline]
Krilov LR. (2002) Safety issues related to the administration of ribavirin. Pediatr Infect Dis J; 21: 479–81.[CrossRef][Web of Science][Medline]
Lin X, Willeke K, Ulevicus V et al. (1997) Effect of sampling time on the collection efficiency of all glass impingers. Am Ind Hyg Assoc J; 58: 480–8.
Lyons CP. (1992) Sampling efficiency of all glass midget impinger. J Aerosol Sci; 23: S559–S602.[CrossRef]
Moss RB, Rodman D, Spencer LT et al. (2004) Repeated adeno-associated virus serotype 2 aerosol mediated cystic fibrosis transmembrane regulator gene transfer to the lungs of patients with cystic fibrosis. Chest; 125: 509–21.
Rosenfeld MA, Ronald G, Crystal RG. (1993) Gene therapy for pulmonary diseases. Pathol Biol (Paris); 41: 677–80.[Medline]
Shults RE, Baron S, Decker J et al. (1996) Health care worker exposure to aerosolized ribavirin: Biological and air monitoring. J Occup Environ Med; 38: 257–63.[CrossRef][Web of Science][Medline]
Spanne M, Gryzbowski P, Bohgard M. (1999) Collection efficiency for submicron particles of a commonly used impinger. Am Ind Hyg Assoc J; 60: 540–4.
Streicher RP, Reh CM, Key-Schwartz RJ et al. (2000) Determination of airborne isocyanate exposure: Considerations in method selection. Am Ind Hyg Assoc J; 61: 544–56.
US EPA. (1997) Exposure factors handbook, EPA/600/P-95/002Fa. Cincinnati, OH.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||