EPA-HSRB-11-02

Paul Anastas, PhD

EPA Science Advisor

Office of the Science Advisor

1200 Pennsylvania Avenue, NW

Washington, DC 20460 

Subject: April 13-14, 2011 EPA Human Studies Review Board Meeting Report

Dear Dr. Anastas,

	The United States Environmental Protection Agency (EPA or Agency)
requested that the Human Studies Review Board (HSRB) provide scientific
and ethics reviews of several completed studies and scenarios involving
intentional exposure of human subjects to pesticides or other EPA
regulated compounds.

	The first of these scenarios, a set of three interrelated studies
measuring dermal and inhalation exposure of professional agricultural
workers who spray pesticides using open-cab airblast equipment, was
conducted by the Agricultural Handler Exposure Task Force, LLC (AHETF).
These studies (AHE62, AHE63 and AHE64) were conducted after publication
of the EPA’s expanded final rule for protection of subjects in human
research (40 CFR 26) on February 6, 2006 (71 Federal Register 24, 6137).
These studies were conducted under a single scenario monograph (MRID
48326701), which incorporates the data from these three studies into a
single dataset. An additional set of data, collected as part of a
third-party sponsored protocol (AHE07) conducted prior to publication of
the EPA’s expanded final rule for protection of subjects in human
research, will also be included in this data set. This data set will be
posted to the Agricultural Handlers Exposure Database (AHED®), and used
generically to estimate daily dermal and inhalation exposures of workers
who treat agricultural crops with conventional pesticides using open-cab
airblast equipment.

	In addition, the Board reviewed a completed study of dermal and
inhalation exposure of professional janitorial workers who clean indoor
surfaces with an antimicrobial pesticide product using  ready-to-use
wipes or a trigger spray bottle and wipe, conducted by the Antimicrobial
Exposure Assessment Task Force II (AEATF II). This study (AEA-02) was
also conducted after publication of the EPA’s expanded final rule for
protection of subjects in human research. The data will be posted to the
Biocide Handlers Exposure Database (BHED®), and used generically to
estimate daily dermal and inhalation exposures of those who wipe indoor
surfaces with antimicrobial pesticides. 

	The Board also reviewed a published study involving intentional human
exposure to the zinc oxide contained in sunscreens. The Agency proposes
to rely on this study, conducted after to publication of the EPA’s
expanded final rule for protection of subjects in human research, for
regulatory actions.

	Finally, the Board considered additional sponsor-provided information
regarding the Kisicki et al. (1999) study on chlorpyrifos, which was
originally discussed by the HSRB in June 2009 (EPA HSRB 2009). As the
review and discussion of this additional information was conducted
independently rather than at the request of the Agency, the Board’s
conclusions are attached to this meeting report as Appendix 1.

	 The enclosed report provides the Board’s response to EPA charge
questions presented at the April 13-14, 2011 meeting.

Assessment of Completed AHETF Research Studies AHE62, AHE63 and AHE64:
Determination of Dermal and Inhalation Exposure to Workers During
Airblast Applications of Liquid Sprays Using Open Cab Equipment (MRID
48289611, 48289612, 48289613, 48289614, 48289615, 48289616 and
48326701).

Science

	The Board concluded that the research reported in the completed
monograph, associated field study reports, and associated supplemental
documents was conducted in a manner that was reasonably faithful to the
design and objectives of the protocol and governing documents of the
AHETF. 

	The Board also concluded that the Agency has adequately, but not
completely, considered the limitations on these data that should be
considered when using the data in estimating the dermal and inhalation
exposure of those who apply conventional pesticides with open-cab
airblast equipment. Additional limitations and concerns have been
identified by the Board, and the conclusion as to the generalizability
of these data requires further consideration and analysis.

Ethics

The Board concluded that the study was conducted in substantial
compliance with subparts K and L 40 CFR 26.

Assessment of Completed AEATF II Research Study AEA02: Measurement of
Potential Dermal and Inhalation Exposure during Application of a Liquid
Antimicrobial Pesticide Product Using Trigger Spray and Wipe or
Ready-to-Use Wipes for Cleaning Indoor Surfaces (MRID 48375601).

Science

	The Board concluded that the research reported in the completed
monograph, associated field study reports, and associated supplemental
documents was conducted in a manner that was reasonably faithful to the
design and objectives of the protocol and governing documents of the
AEATF II. 

	The Board also concluded that the Agency has adequately, but not
completely, considered the limitations on these data that should be
considered when using the data in estimating dermal and inhalation
exposures of handlers who are applying antimicrobial pesticides using a
trigger spray or ready-to-use wipes.

Ethics

The Board concluded that the study was conducted in substantial
compliance with subparts K and L 40 CFR 26.

Assessment of Published Research Study MRID 48387301: Gulson et al.
(2010) Small Amounts of Zinc from Zinc Oxide Particles in Sunscreens
Applied Outdoors Are Absorbed through Human Skin.

Science

	The Board concurred with the Agency’s assessment that the Gulson et
al. (2010) study provides some potentially useful data on the dermal
absorption of zinc from zinc oxide nanoparticles in sunscreen applied to
human skin, despite the multiple limitations identified by the HSRB.
However, the Board advised the Agency to proceed with caution when using
these data in risk assessment, as these data cannot be used as a
stand-alone set to assess dermal absorption of zinc oxide.

Ethics

	The Board concluded that there was insufficient information available
at the time of the April meeting to determine whether the Gulson et al.
(2010) study was conducted in substantial compliance with procedures at
least as protective as those in subparts A - L of EPA’s regulation at
40 CFR Part 26. The Board recommended that EPA seek additional
information from Macquarie University or the study investigator
(including a copy of the research protocol and unsigned informed consent
form) and that information be provided to the HSRB for re-consideration.

Sincerely,

 

Sean Philpott, PhD, MSBioethics

Chair

EPA Human Studies Review Board



NOTICE

This report has been written as part of the activities of the EPA Human
Studies Review Board, a Federal advisory committee providing advice,
information and recommendations on issues related to scientific and
ethical aspects of human subjects research.  This report has not been
reviewed for approval by the Agency and, hence, the contents of this
report do not necessarily represent the view and policies of the
Environmental Protection Agency, nor of other agencies in the Executive
Branch of the Federal government, nor does the mention of trade names or
commercial products constitute a recommendation for use.  You may obtain
further information about the EPA Human Studies Review Board from its
website at   HYPERLINK "http://www.epa.gov/osa/hsrb" 
http://www.epa.gov/osa/hsrb .  You may also contact the HSRB Designated
Federal Officer, via e-mail at   HYPERLINK "mailto:ord-osa-hsrb@epa.gov"
 ord-osa-hsrb@epa.gov 

	In preparing this document, the Board carefully considered all
information provided and presented by the Agency presenters, as well as
information presented by public commenters.  This document addresses the
information provided and presented within the structure of the charge by
the Agency.



US ENVIRONMENTAL PROTECTION AGENCY

HUMAN STUDIES REVIEW BOARD

Chair

Sean Philpott, PhD, MSBioethics, Director for Research Ethics, The
Bioethics Program of Union Graduate College and the Mount Sinai School
of Medicine, Schenectady, NY

Vice Chair

Janice Chambers, PhD, DABT, Fellow ATS, William L. Giles Distinguished
Professor, Director, Center for Environmental Health Sciences, College
of Veterinary Medicine, Mississippi State University, Mississippi State,
MS

Members

George Fernandez, PhD, Professor of Applied Statistics, Director of the
University of Nevada-Reno Center for Research Design and Analysis,
University of Nevada-Reno, Reno, NV

Vanessa Northington Gamble, MD, PhD, University Professor of Medical
Humanities, Professor of Health Policy and American Studies, The George
Washington University, Washington, DC

Sidney Green, Jr., PhD, Fellow of the ATS, Professor, Department of
Pharmacology, Howard University College of Medicine, Washington, DC

Dallas E. Johnson, PhD, Professor Emeritus, Department of Statistics,
Kansas State University, Manhattan, KS

Michael D. Lebowitz, PhD, FCCP, Professor (NTE) of
Medicine-Research/Scholar, University of Arizona, Tucson, AZ 

Jose E. Manautou, PhD, Associate Professor of Toxicology, Department of
Pharmaceutical Science, University of Connecticut School of Pharmacy,
Storrs, CT

Jerry A. Menikoff, MD, JD, Director, Office for Human Subjects Research,
Office of the

Secretary, Department of Health and Human Services, Rockville, MD

*Rebecca Parkin, PhD, MPH, Professorial Lecturer (EOH) Human Risk
Assessment, School of Public Health and Human Services, The George
Washington University, Washington, DC

William Popendorf, PhD, MPH, Professor Emeritus, Department of Biology,
Utah State University, Logan, UT 

 

Virginia Ashby Sharpe, PhD, Medical Ethicist, National Center for Ethics
in Health Care, Veterans Health Administration, Washington, DC

Linda J. Young, PhD, Professor, Department of Statistics, Institute of
Food and Agricultural Sciences, University of Florida, Gainesville, FL 

Human Studies Review Board Staff

Jim Downing, Executive Director, Human Studies Review Board Staff,
Office of the Science Advisor, United States Environmental Protection
Agency, Washington, DC

* Not in attendance at the April 13-14, 2011 Public Meeting



INTRODUCTION 

	On April 13-14, 2011, the United States Environmental Protection
Agency’s (EPA or Agency) Human Studies Review Board (HSRB) met to
address scientific and ethical issues concerning two completed study
monographs. One study measured levels of dermal and inhalation exposure
received by janitorial workers who apply antimicrobial pesticides using
trigger spray or ready-to-use wipes. The second was comprised of three
completed, interrelated studies measuring levels of dermal and
inhalation exposure received by pesticide applicators who spray
pesticides with open-cab airblast equipment. These studies, along with a
fourth study conducted prior to publication of the EPA’s expanded
final rule for protection of subjects in human research, were compiled
into a single scenario monograph (MRID 48326701), which incorporates the
data from these studies into a single data set. In accordance with 40
CFR 26.1602, EPA sought HSRB review of these two completed study
monographs. The completed studies are discussed more fully below.

	In addition, the Agency sought HSRB review of one published study of
dermal absorption of zinc (Zn) from zinc oxide (ZnO) nanoparticles in
sunscreen applied to human skin. This study, conducted after publication
of the EPA’s expanded final rule for protection of subjects in human
research, was identified by Agency scientists from the peer reviewed
literature. This study, which the Agency proposes to rely upon for
regulatory actions, is discussed in detail below.

	Finally, the HSRB considered additional sponsor-provided information
regarding a pre-Rule intentional exposure study of chlorpyrifos,
originally discussed by the Board in June 2009. The Board’s review and
discussion of this additional information was conducted independently
rather than at the request of the EPA. The Board’s discussion and
conclusions are attached to this meeting report as Appendix 1.

REVIEW PROCESS

On April 13-14, 2011, the Board conducted a public face-to-face meeting
in Arlington, Virginia. Advance notice of the meeting was published in
the Federal Register as “Human Studies Review Board; Notice of Public
Meeting” (76 Federal Register 59, 17122).

	Following welcoming remarks from Agency officials, the Board heard
presentations from EPA on the following topics: three completed studies
(compiled into a single monograph) measuring dermal and inhalation
exposure received by pesticide applicators who spray pesticides with
open-cab airblast equipment, one completed study measuring levels of
exposure received by janitorial workers when applying antimicrobial
pesticides using a trigger spray or ready-to-use wipes, and one
published study measuring dermal absorption of zinc (Zn) from zinc oxide
(ZnO) nanoparticles in sunscreen applied to human skin.

	The Board also asked clarifying questions of several study sponsors
and/or research investigators, including:

Ms. Megan Boatwright, Analytical Coordinator, Golden Pacific
Laboratories

Dr. Victor Cañez, Technical Chair, Agricultural Handler Exposure Task
Force, LLC (AHETF).

Dr. Richard Collier, Administrative Committee Chair, AHETF.

Dr. Has Shah, Manager, Antimicrobial Exposure Assessment Task Force II
(AEATF II).

Mr. Robert Testman, Vice-President, Golden Pacific Laboratories.

Public oral comments were provided by: 

	Dr. Victor Cañez, Technical Chair, AHETF.

	Dr. Richard Collier, Administrative Committee Chair, AHETF.

Dr. Has Shah, Manager, AEATF II.

	One written public comment was submitted by Ms. Carol Dansereau,
Executive Director of the Farm Worker Pesticide Project, Seattle, WA. 

For their deliberations, the Board considered the materials presented at
the meeting, oral comments, and Agency background documents (e.g.,
published literature, sponsor and investigator research reports, study
protocols, data evaluation records, and Agency science and ethics
reviews of proposed protocols and completed studies). A comprehensive
list of background documents is available online at   HYPERLINK
"http://www.regulations.gov"  http://www.regulations.gov . 

CHARGE TO THE BOARD AND BOARD RESPONSE

Assessment of Completed AHETF Research Studies AHE62, AHE63 and AHE64:
Determination of Dermal and Inhalation Exposure to Workers During
Airblast Applications of Liquid Sprays Using Open Cab Equipment.

Overview of the Study

	Three separate field studies were conducted, each monitoring dermal and
inhalation exposure of workers to commercially available pesticides
while spraying tree or trellis crops in three different U.S. states
where open-cab airblast equipment is commonly used in production
agriculture. A total of 13 professional agricultural handlers were
monitored as they applied pesticides using open-cab airblast equipment:
three adult men applying pesticides to grape vines in California
(AHE62), five adult men applying pesticides to grape vines in New York
(AHE63), and five adult men applying pesticides to pecan trees in
Oklahoma (AHE64). The scenario design, protocols for the three studies,
SOPs and governing documents were reviewed favorably by the HSRB at its
October 21-22, 2008 (EPA HSRB 2008b).

	Monitored on actual days of work, study participants handled from 7 to
90 lbs of active ingredient (carbaryl or malathion), spraying 3 to 24
acres in 1.4 to 10.6 hours. Dermal exposure was measured using hand
washes, face/neck wipes, and whole body dosimeters (100% cotton union
suits) for the remainder of the body (torso, arms, and legs). Inhalation
exposure was measured using personal air sampling pumps and OVS mounted
on the shirt collar. Results represent dermal exposure with and without
chemical-resistant hats while wearing a long-sleeved shirt, pants,
shoes/socks and chemical-resistant gloves, and inhalation exposure
without respiratory protection.

	The Agency proposes to use data from these three studies, along with a
previous open-cab airblast study involving 15 male participants applying
pesticides to apple, orange, peach and pear trees in Florida, Georgia
and Idaho (AHE07; Smith 2005), to estimate generically daily dermal and
inhalation exposures of workers who treat agricultural crops with
conventional pesticides using open-cab airblast equipment.

	

Science

Charge(s) to the Board

	1. Was the research reported in the AHETF completed monograph report
and associated field study reports faithful to the design and objectives
of the protocol, SOPs and governing documents?

	2.  Has EPA adequately characterized, from a scientific perspective,
the limitations on these data that should be considered when using the
data in estimating exposure of those who apply pesticides with open-cab
airblast equipment?

Board Response to the Charge(s)

HSRB Recommendation 

	The Board concluded that the research reported in the completed
monograph, associated field study reports, and associated supplemental
documents was conducted in a manner that was reasonably faithful to the
design and objectives of the protocol and governing documents of the
AHETF. 

	The Board also concluded that the Agency has adequately, but not
completely, considered the limitations on these data that should be
considered when using the data in estimating the dermal and inhalation
exposure of those who apply conventional pesticides with open-cab
airblast equipment. Additional limitations and concerns have been
identified by the Board, and the conclusion as to the generalizability
of these data requires further consideration and analysis.

HSRB Detailed Recommendations and Rationale

	The HSRB concluded that the research reported in the completed
monograph report and the associated field study reports of Open Cab
Airblast Applicator Exposures (Bruce 2010a;  Bruce 2010b; Bruce 2010c;
Klonne and Holden 2010, Smith 2005) was faithful to the design and
objectives of the protocol, SOPs, and governing documents. Four
limitations were suggested as potential additions to those outlined in
the Agency’s Reviews of the monograph and associated field study
reports (Crowley 2011; Crowley and Sarkar 2011).  The magnitude of one
of these limitations cannot be determined but could significantly impact
the accuracy of the measured exposures in handlers who are not wearing a
chemical resistant (CR) hat.

	The AHETF seems to have responded well to most of the concerns
expressed by the HSRB during their prior reviews of the protocol (EPA
HSRB 2006; EPA HSRB 2008b).  For instance, the use of an informal
survey/poll of local experts that identified the absence of the
“Kinkelder” airblast sprayer, the use of a refined and better
defined recruitment process, and the inclusion of descriptive records of
applicator behavior. In the end, the analytical chemistry methods seemed
appropriate and well chosen, the quality assurance (QA) considerations
appeared to be in place, and the desired study conditions were largely
obtained.

	Although no formal statistical test for differences among crops or
clusters was provided, the visual evidence in the Science Review (Figure
1; Crowley 2011, 12) is strong support for their similarity.  The Board
supports the Agency’s plan to combine the datasets and to use a
mixed-model approach for statistical analysis to account for potential
data clustering for both the existing data and new monitoring.  The
Board also agrees with a similar approach and conclusion regarding the
negligible effect of canopies used by six of the 28 monitoring units
(MUs).

	The use of head patches inside and outside CR hats was innovative.  Its
use obviated the need for an MEA for the portion of the head covered by
the hat, probably increased the accuracy of measuring exposure with a
hat, and yielded potentially useful estimates of the protection from
exposure that the hat provided.  At the same time, the use of this
technique presents two limitations in estimating exposures without such
a hat.

	First, it could be argued that when estimating exposure without such a
CR hat that the amount reaching the inner patch should not be counted. 
This argument would apply if the pathway by which the active ingredient
that was deposited onto the inner patch was via a micro-scale form of
drift under the hatband (“penetration”) and not directly through the
hat’s material (“permeation”).  The effect on the overall dermal
exposure of this sort-of double counting the patch data is
insignificantly small (on the order of 0.1%) due to the very small
deposition measured on the inner patch.

	Second, the use of this innovation may have reduced the accuracy of
measuring exposure without a hat because the presence of a wide-brimmed
hat probably also reduced deposition onto the exposed face and neck from
what it would have been had the hat not been there at all.  The
magnitude of this reduction is undetermined and seems undeterminable
with the information available.  The brim is likely to have acted
somewhat like a canopy, only much closer.  Its effect is likely to have
been greater on the portions of the face and neck closest to the
underside of the hatband and to decrease on the portions further away,
such as the tip of the nose. The effect of the brim’s protection is
probably clear but not overwhelming.

	The Agency might reconsider applying a 1/2 limit of quantitation (LOQ)
factor when no active ingredient is found in the backup portion of the
OVS air samples.  A proper sample should have no airborne chemical in
that portion.  In fact, finding active ingredient there implies some
degree of overloading of the front portion of the sampling media. 
Consistently assigning a non-detect measurement in the backup portion of
a value of 1/2 limit of detection (LOD) results in a biased result. 

	Four additional limitations were suggested.  The first is to include
such qualitative data as the absence of the “Kinkelder” airblast
sprayer from the array of equipment included within the study.  This
sprayer is an older model noted by the local experts to be prone to
spray drift.  The implications of higher drift on handler exposures
depend on a combination of environmental conditions and work practices. 
However, the broader point is that such qualitative information could
become a valuable resource to future users of the exposure database. 

	The second limitation is the potential underestimation of spray
actually deposited onto the wiped portion of the subjects’ head due to
shielding by the brim of the CR hats that were always worn.  

	The third limitation is the possible exceptional status of one
monitoring unit involved in exposure monitoring for open-cab spraying of
Oklahoma pecans (Study AHE64; Monitoring Unit A5).  This MU used nozzle
pressure that was twice as high as that recommended for the type of
nozzle (and three to four times as high as reported by any other MU
using this type of nozzle). The Board recognizes that increasing nozzle
pressure will cause a decrease in droplet size and an increase in spray
drift.  This MU also changed from spraying in both directions to
spraying in only one direction for the last third of his application
time. This was the only MU for which such a comment was made.  Two
reasons were suggested by the Board for why an applicator might make
such a change in spraying direction: either to navigate sloping terrain
or/and to avoid windblown drift. No information is available regarding
the terrain, but the initial one way pattern resulted in this MU
spraying only when driving upwind.  Both of these reasons could cause
this MU’s exposures to be low.  Whether these differences would
otherwise disqualify this MU is unknown. 

	The fourth limitation is the need to rely on model accuracy for
estimates to be interpretable. The primary objective is to estimate the
geometric mean, the arithmetic mean, and the 95th percentile of
normalized exposures (exposure per pound of active ingredient handled
[AaiH]) within 3-fold with 95% confidence. The secondary objective is to
assess proportionality (ln(exposure) =  ln(AaiH) ) against the
alternative of independence in ln(exposure) and ln(AaiH) with 80% power.
 It is critical to carefully evaluate the validity of the model to
ensure that the mean of the normalized exposures on the log scale (and
the geometric mean on the data scale) are interpretable. Because
clusters are part of the study’s design, a random cluster effect
should be included in the model. How the presence of a cluster effect
impacts interpretation should be considered carefully. 

	In the Agency’s Review (Crowley and Sarkar 2011), an alternative
approach was taken in which the normalized exposures were modeled using
a lognormal distribution. The validity of this approach depends on two
assumptions. First, the expectation of ln(exposure/AaiH) must be linear.
It is important to assess the validity of the linear relationship
between ln(exposure) and ln(AaiH). Second, for the approach to be valid,
there cannot be a cluster effect. Otherwise, the mean of the
distribution would change with cluster so that a single lognormal
distribution would not describe the normalized exposures.  Perhaps this
is why some of the Q-Q plots for the lognormal distribution (Appendix B;
Crowley and Sarkar 2011, 24-5) indicate some lack of fit, especially in
the tails. (It is not possible to be sure that these correspond to the
inhalation data for which a significant cluster effect was present
because the graphs are not labeled.) 

	The Agency would benefit from carefully considering all assumptions
underlying a proposed statistical approach and then checking the
validity of each. Care should be taken to ensure that the results can be
meaningfully interpreted. The term “arithmetic mean” should be
changed to “mean,” especially when a model is used to estimate the
population mean. It would be helpful for the axes of all graphs to be
labeled. Lastly, in the write-up of results, implying that the failure
to reject a null hypothesis indicates that the null hypothesis is true,
should be avoided.

	The Agency should note that more detailed suggestions and analyses may
be provided by some Board members independent of this consensus Board
report.	

Ethics

Charge to the Board

	Does the available information support a determination that the studies
were conducted in substantial compliance with subparts K and L of 40 CFR
Part 26? 

	

Board Response to the Charge

HSRB Recommendation

The Board concurred with the Agency’s assessment (Sherman 2011a) that
the study was conducted in substantial compliance with subparts K and L
40 CFR 26.

HSRB Detailed Recommendation and Rationale

The documents provided include reports of each of three field studies
conducted on behalf of the AHETF (Bruce 2010a; Bruce 2010b; Bruce
2010c). The submitted study documents state that the study was conducted
in substantial compliance with the ethical and regulatory standards of
40 CFR 26, Subparts K and L. The requirements of FIFRA §12(a)(2)(P) and
Title 3, § 6710 of the California Code of Regulations also apply. 

The three protocols were each reviewed and approved pursuant to the
standards of the Common Rule (45 CFR Part 46, Subpart A) by an
independent human subjects review committee, Independent Institutional
Review Board, Inc. (IIRB, Inc.) of Plantation, FL. Minutes of IIRB, Inc.
meetings and a copy of IIRB, Inc. policies and procedures were provided
(Bruce 2010d; Bruce 2010e; Bruce 2010f; IIRB 2010a; IIRB 2010b; Sherman
2011a). IIRB, Inc. is registered with the US Office for Human Research
Protections (OHRP; Reg. # IORG0002954) and is fully accredited by the
Association for the Accreditation of Human Research Protection Programs,
Inc. (AAHRPP).

1.	The Board concurred with the conclusions and factual observations
relating to the study, as detailed in the EPA’s Ethics Review (Sherman
2011a) and summarized briefly below. 

a. 	Prior HSRB and Agency Review. 

	Because each of these three studies was initiated after April 7, 2006,
prior submission of the protocol and supporting materials to EPA was
required by 40 CFR §26.1125.  The requirements of 40 CFR §26.1125 for
prior submission of the protocol to EPA and of §26.1601 for HSRB review
of the protocol were satisfied. The scenario designs and study protocols
were initially approved by IIRB, Inc. in July 2008. The HSRB discussed
these three protocols at its October 2008 meeting, concurring with the
Agency’s assessment that these three proposed open cab airblast field
study protocols, if revised as suggested by the Agency and the HSRB,
would meet the applicable requirements of 40 CFR Part 26, subparts K and
L (EPA HSRB 2008b).

b.	Responsiveness to HSRB and Agency Recommendations.

	The initial ethics reviews by the Agency (Carley and Evans 2008a;
Carley and Evans 2008b; Carley and Evans 2008c) and by the HSRB (EPA
HSRB 2008b) provided nine recommendations with regard to these three
protocols. All of those recommendations, and the responses made with
regard to them, are detailed in the submitted documents (Sherman 2011,
25). The HSRB agreed with the Agency that the comments by the Agency and
HSRB were addressed satisfactorily. 

2.  The Board concluded that this study, as conducted, met all
applicable ethical requirements for research involving human
participants, in accordance with the following criteria:

a.	Acceptable risk-benefit ratio. 

	The risks to study participants were minimized appropriately and were
justified by the potential societal benefits, particularly data on the
dermal and inhalation exposure of professional pesticide applicators to
the liquid pesticides they apply to orchard and trellis crops using an
airblast sprayer drawn by a vehicle with no cab. These data could be
used to develop mechanisms to protect future persons who apply these
liquid pesticides.

	Minors and pregnant or lactating women were excluded from
participation, with pregnancy confirmed by over-the-counter pregnancy
testing on the day of study or by opt-out. The potential of stigma
resulting from study exclusion was also appropriately minimized.

	Clear stopping rules and medical management procedures were in place,
and no adverse events or other incidents of concern related to product
exposure were reported.

	The study was designed to minimize the risks of exposure to the test
compounds, subject to being able to accomplish the purposes of the
study.

b.	Voluntary and informed consent of all participants.

	The study protocol included several mechanisms designed to minimize
coercive recruitment and enrollment.

	Monetary compensation was not so high as to unduly influence
participation.

	In addition, it is worth noting that twelve of the thirteen subjects
were owner-operators, and thus less likely to be as vulnerable to
coercion or undue influence as an employee might be.

3.	Eight minor protocol deviations were reported. In addition, there was
one unreported but minor protocol deviation. These are documented in
detail in the EPA Ethics Review (Sherman 2011a, 8-10). The Board agrees
with the Agency’s assessment that these deviations did not compromise
the informed consent process or put the study participants at increased
risk.

ovember 2009―after this research had been conducted―the SOP for
these types of studies was revised to lower the acceptable heat index
threshold from 120º F to 105º F. However, as noted in Table 4 of the
EPA Ethics Review (Sherman 2011a, 15), the heat index values for the
subjects enrolled in this study never exceeded 105º F, so this was not
an issue for this research.

Assessment of Completed AEATF II Research Study AEA02: Measurement of
Potential Dermal and Inhalation Exposure during Application of a Liquid
Antimicrobial Pesticide Product Using Trigger Spray and Wipe or
Ready-to-Use Wipes for Cleaning Indoor Surfaces.

Overview of the Study

	The objective of this study was to measure individual exposures to a
surrogate antimicrobial pesticide (didecyl dimethyl ammonium chloride;
DDAC) while cleaning surfaces with a trigger spray and wipe or with
ready-to-use wipes. A total of 36 volunteers participated in the study,
18 cleaning surfaces using a trigger spray and wipe and 18 clearing
surfaces using ready-to-use wipes in one of three building types (an
office building, a Rite Aid pharmacy building, or a retired teacher’s
memorial building in Fresno, CA). Participants cleaned surfaces for one
of six pre-determined wiping times (30-60, 60-90, 90-120, 120-150,
150-180 and 180-210 minutes total cleaning time, respectively). The
study protocol, SOPs and governing documents were reviewed favorably by
the HSRB at its April 9-10, 2008 meeting (EPA HSRB 2008a).

 

	Dermal and inhalation exposure monitoring was conducted while study
participants cleaned surfaces. Dermal exposure was measured by inner and
outer body dosimeters. Airborne concentrations of the surrogate compound
were monitored in the participant’s breathing zone using an OVS tube
sample collector connected to a personal sampling pump. Environmental
conditions were also recorded at the time of monitoring, and observers
made notes, photographs and videos of participant activity throughout
the wiping period.

	These exposure data will be used to populate a database representing
inhalation and dermal exposure during a number of antimicrobial handler
scenarios. A scenario is defined as a pesticide handling task based on
activity (e.g., application) and equipment type (e.g., mop & bucket,
ready-to-use wipes, pressure treatment of wood facilities, painting).
These data will be used by the Agency to estimate dermal and inhalation
exposures of antimicrobial handlers who are applying pesticides using a
trigger spray and wipe or ready-to-use wipe under a variety of
scenarios.

	

Science

Charge(s) to the Board

	1. Was the research reported in the AEATF II completed wipe study
report faithful to the design and objectives of the protocol and
governing documents of the AEATF II?

	2.  Has EPA adequately characterized, from a scientific perspective,
the limitations on these data that should be considered when using the
data in estimating exposure of those who clean indoor surfaces with
antimicrobial pesticides using a trigger-bottle and wipes or
ready-to-use wipes?

Board Response to the Charge(s)

HSRB Recommendation 

	The Board concluded that the research reported in the completed
monograph, associated field study reports, and associated supplemental
documents was conducted in a manner that was reasonably faithful to the
design and objectives of the protocol and governing documents of the
AEATF II. 

	The Board also concluded that the Agency has adequately, but not
completely, considered the limitations on these data that should be
considered when using the data in estimating dermal and inhalation
exposures of antimicrobial handlers who are applying pesticides using a
trigger spray or ready-to-use wipes.

HSRB Detailed Recommendations and Rationale

	

	The Board concurred with the EPA’s assessment (Leighton 2011) that
the AEATF II made the appropriate changes to the protocol proposed by
the EPA and HSRB and has executed the study in a way that is reasonably
faithful to the design and objectives of the protocol and governing
documents of the AEATF II. Although several protocol deviations
occurred, they were unlikely to have adversely affected the reliability
of the data. The Agency’s analysis of the data, especially their use
of adjustments for non-detects in the fortified samples (of which there
were more than a few) was an improvement. 

	There was one potentially important shortcoming of the study. The
calculated rates at which the active ingredient was applied to the
treated surfaces did not achieve the stated target application rate of 7
mg of active ingredient per m2 of treated surface. Only one of the
participants using ready-to-use wipes and only two of the participants
using a trigger spray achieved the 7 mg/m2 target.  Neither the origin
nor the significance of not meeting this target is known. While the
protocol failed to achieve this technical target, this target was not
recognized heretofore to potentially be in conflict with achieving the
important goal of letting the subjects “wipe surfaces as they would
normally work” (Selim 2011, 43). By all appearances, the study did
achieve the latter goal.

	The Board believes that the EPA has insufficiently characterized from a
scientific perspective the limitations on these data. There were
limitations that were not satisfactorily addressed. As an example, there
were too many non-detects in the whole body dosimeters, which created
the positive regression lines for the dermal exposures by anchoring at
the low end. The assumptions of their modeling should be tested. Then,
the appropriate regression models should be run on the results of the
hand wipes (exposure by AaiH). If this is done, the effects of the
non-detects can be evaluated in a step-wise fashion. 

	The results of the non-parametric bootstrap had narrower confidence
limits and should be used instead of the parametric bootstrap results.
Interpretations of the results should be re-evaluated, as they are
currently too strong given the lack of any statistical significance. The
lack of any relationship of inhalation exposure to AaiH should be
accepted. The results of evaluating the K-factor, especially for
inhalation, show the study to be underpowered to evaluate this
association. Conclusions that estimates at the high end of AaiH are
overestimates are definitely incorrect in the threshold model, which
shows values below the estimated line at high AaiH and higher values in
the middle of the range of AaiH that are elevating the curve.

	Two potential new limitations to the study were also identified.  One
is an outgrowth of a search to explain the result that the dermal unit
exposure value for the ready-to-use wipe applications is twice as large
as that for the trigger spray bottle. The difference in results may be
due to the different ways in which the amount of active ingredient
handled was calculated in the two studies, and this explanation has
implications to the validity of extrapolations based on AaiH. The two
methods treated about the same amount of surface area in the same amount
of time and resulted in about the same exposure, but the amount of
biocide handled by the trigger spray bottle applicators was about twice
that handled by ready-to-use wipe applicators. The study was not
designed to tell how much active ingredient (ai) actually reached the
target surface nor if the two methods were equally effective in treating
the surfaces.  Wiping has no intrinsic means to control the rate at
which the active ingredient is applied.  The rate depends completely on
the applicator.  The methods used to calculate the amount of ai handled
may not directly relate to the amount. Neither method accounts for the
amount of ai that stayed on the wipe or rag at the time it was
discarded, and this unknown may have affected the calculated amount of
ai handled. The possibility that the almost two-fold difference in the
amounts handled (and in unit exposure values) is attributable to the
difference in the amount left on the wipes and rags suggests a
limitation to the data that should be considered when using the data.
The need to extrapolate via unit exposures may be elevated due to the
fact that the test conditions were well less than the target application
rates of 7 mg ai/m2.

	The second limitation stems from the predominance of hand exposures. 
Noting that the hands account for 92% of the dermal exposure for the
trigger spray bottle uses (Selim 2011, 107) and 98% for the ready-to-use
wipe users (Selim 2011, 98) brings attention to a potentially important
limitation of the recovery correction factor based on the DDAC dermal
exposure and recovery study conducted by Boatwright (2007).  Although
this study investigated the recovery at two dermal application rates, it
appears that it did not investigate the effect of retention time on
recovery from the skin.  Without data one can only conjecture that the
behavior of DDAC might mimic the behavior of the few insecticides in
which similar tests have included time.  In their January 2007 report,
the EPA’s Scientific Advisory Panel pointed out the importance of a
chemical’s residence time on the skin in the recovery efficiency (EPA
SAP 2007).  (This report was also cited by the HSRB in their June 2007
policy on MEAs.)  Parts of that discussion implied that, in general,
recoveries from skin are (or appear to be) higher soon after deposition
and to decrease as either the chemical resides on the skin for longer
durations or continuous or repeated applications saturate the ability of
the skin to retain as much of the ai deposited later as it could
earlier.  Both of these scenarios result in what appear to be decreases
in the recovery efficiency of those chemicals from skin as exposure time
increases.

The Agency should note that more detailed suggestions and analyses may
be provided by some Board members independent of this consensus Board
report.

Ethics

Charge to the Board

	Does the available information support a determination that the study
was conducted in substantial compliance with subparts K and L of 40 CFR
Part 26? 

	

Board Response to the Charge

HSRB Recommendation

The Board concurred with the Agency’s assessment (Sherman 2011b) that
the study was conducted in substantial compliance with subparts K and L
40 CFR 26.

HSRB Detailed Recommendation and Rationale

The documents provided include a report of a completed wipe study
conducted on behalf of the AEATF II (Selim 2011). The submitted study
documents state that the study was conducted in substantial compliance
with the ethical and regulatory standards of 40 CFR 26, Subparts K and
L. The requirements of FIFRA §12(a)(2)(P) and Title 3, § 6710 of the
California Code of Regulations also apply.

	The protocol was reviewed and approved pursuant to the standards of the
Common Rule (45 CFR Part 46, Subpart A) by an independent human subjects
review committee, IIRB, Inc. of Plantation, FL. Minutes of IIRB, Inc.
meetings and a copy of IIRB, Inc. policies and procedures were provided
(IIRB 2010a; IIRB 2010b; Selim 2011; Sherman 2011b). As stated
previously, IIRB, Inc. is registered with OHRP and is fully accredited
by AAHRPP. 

	 

1.	The Board concurred with the conclusions and factual observations
relating to the study, as detailed in the EPA’s Ethics Review (Sherman
2011b) and summarized briefly below. 

a. 	Prior HSRB and Agency Review. 

	Because this study was initiated after 7 April 2006, prior submission
of the protocol and supporting materials to EPA was required by 40 CFR
§26.1125.  The requirements of 40 CFR §26.1125 for prior submission of
the protocol to EPA and of §26.1601 for HSRB review of the protocol
were satisfied.  The HSRB reviewed the protocol at its April 2008
meeting, concluding that “if the proposed mop and wipe scenario
design, protocol, and supporting documentation is revised as suggested
in EPA’s review, the research would meet the applicable requirements
of 40 CFR part 26, subparts K and L” (EPA HSRB 2008a, 2). 

b.	Responsiveness to HSRB and Agency Recommendations.

	The HSRB noted two instances where the sponsor did not address comments
raised by the HSRB in its April 2008 report. In the first instance, the
HSRB recommended that the consent form be revised to explain that the
“underlying purpose of the study will be to collect information that
will be provided to the EPA, and that the EPA would use that information
to determine the appropriate standards for allowable exposures to
products such as the test compound” (EPA HSRB 2008a, 25). In the
second instance, the Board recommended that the recruitment flyer should
explain that the research would measure inhalation as well as dermal
exposure (EPA HSRB 2008a, 26).  The sponsors could not explain the lack
of responsiveness to these two recommendations. The Board concluded that
the lack of responsiveness to these two HSRB recommendations did not
compromise the ethical conduct of the study, but recommended that the
Agency follow up with the sponsor to determine why these recommendations
were not addressed.

2.  The Board concluded that this study, as conducted, met all
applicable ethical requirements for research involving human
participants, in accordance with the following criteria:

a.	Acceptable risk-benefit ratio. 

	The risks to study participants were minimized appropriately and were
justified by the potential societal benefits, particularly data on the
dermal and inhalation exposure of professional janitorial workers to the
liquid antimicrobials they apply. These data could be used to develop
mechanisms to protect future persons who apply these liquid
antimicrobials.

	Minors and pregnant or lactating women were excluded from
participation. All study participants were at least 18 years old.  All
females self-administered an over-the-counter pregnancy test on the day
of monitoring, and all such tests were negative. No nursing mothers were
enrolled in the study.

	The study was designed to minimize the risks of exposure to the test
compounds, subject to being able to accomplish the purposes of the
study. The study involved the use of a commonly used, commercially
available cleaning product, in a matter consistent with its labeling. 

	Clear stopping rules and medical management procedures were in place.
There was one report of a participant leaving early because he was
feeling ill. Although the observation log does not indicate whether the
worker was feeling ill as a result of his participation in the research
or whether his illness was unrelated to the study, documentation
provided to the Agency and the HSRB subsequent to the April 2011 meeting
suggests that this participant was unwell prior to study initiation.
However, there was no indication that there had been any follow up of
the participant. Although there is little reason to think that this
participant would develop a serious and unexpected medical problem
related to his participation after leaving the study site, the Board
recommended that the Agency work with sponsors to develop a SOP for
following up with participants who leave a study early because of
illness.

b.	Voluntary and informed consent of all participants.

	The study protocol included several mechanisms designed to minimize
coercive recruitment and enrollment.

	Monetary compensation was not so high as to unduly influence
participation.

3.	Two minor protocol deviations were reported. In addition, there were
two unreported but minor protocol deviations. The reported deviations
were that: 1) the protocol called for study participants to take
10-minute rest breaks, but most participants declined or took breaks
that were less than 10 minutes in length; and 2) a photograph of a
participant’s face was taken at one monitoring site, although no
images were included in the report.

	The Agency review noted one unreported deviation (Sherman 2011b). This
deviation involved the enrollment of three participants (one who was
monitored and two who were alternates) who self-reported that their
health was only “fair,” despite the requirement that all
participants be in “good health”.  In response to a similar protocol
deviation that occurred during the related AEATF II mop study discussed
at the October 2010 HSRB meeting, the Board recommended the AEATF II
develop a SOP to clarify its health status reporting and inclusion
criteria (EPA HSRB 2010). The Board noted that this study was conducted
prior to the October 2010 HSRB meeting.

	The HSRB also discovered an additional unreported deviation, namely the
use of an “IRB-unapproved form to collect personal information about
volunteers and to sign a ‘Worker Health Statement’” (Selim 2011,
3). These forms were stored separately from the field phase data.  The
sponsors could not explain this deviation from the protocol. 

	The HSRB concluded that these protocol deviations did not compromise
the informed consent process or put the study participants at increased
risk.

Assessment of Published Research Study by Gulson et al. (2010): Small
Amounts of Zinc from Zinc Oxide Particles in Sunscreens Applied Outdoors
Are Absorbed through Human Skin.

Overview of the Study

	In the Gulson et al. (2010) study, 20 human volunteers were exposed to
sunscreens containing stable isotopes of zinc oxide (ZnO) for the
purpose of measuring dermal absorption of zinc (Zn) from particles of
varying size. Such metal oxide nanoparticles are being used more and
more frequently in sunscreens because they reflect and absorb
ultraviolet (UV) light as transparent coatings, rather than as opaque
coatings as with traditional metal oxide sunscreens. 

	The study was conducted in Australia in March 2009. Ten male and ten
female subjects, ranging from 19 to 66 years of age, were enrolled.
Study participants wore UV protective upper body clothing with a patch
removed to expose a small section of skin on the back. Sunscreen
containing stable isotope 68Zn oxide, formulated as “bulk” particles
(110 ± 46 nanometers [nm] in size) or as nanoparticles (19 ± 8 nm in
size), was applied twice daily for a period of five days to the exposed
section of skin on the back. Eleven participants (5 males and 6 females)
applied 68Zn oxide formulated as nanoparticles and nine participants (5
males and 4 females) applied 68Zn oxide formulated as bulk particles.
Participants were encouraged to wear other non-zinc types of sunscreen
on the areas of skin not covered by the UV-resistant clothing. The test
material was removed from each subject’s back at the end of each day
with an alcohol-lanolin wipe.

	Urine and blood samples were taken eight days before sunscreen
application, immediately before the first application, at the end of
each of the five days of application, and six days after the last
application. Dermal absorption of Zn was calculated from the changes in
the ratios of 68Zn to 64Zn in urine and blood samples during the course
of the experiment; Zn was measured using Multi-Collector Inductively
Coupled Mass Spectrometry (MC-ICP-MS).

	

Science

Charge(s) to the Board

	1. Is the Gulson et al. (2010) study scientifically sound, providing
reliable data?

	2. If so, is the Gulson et al. (2010) study relevant for qualitative
use in support of an assessment of the absorption of zinc oxide through
the skin?

Board Response to the Charge(s)

HSRB Recommendation 

	The Board concurred with the Agency’s assessment (Ryman and Dole
2011) that the Gulson et al. (2010) study provides some potentially
useful data on the dermal absorption of zinc from ZnO nanoparticles in
sunscreen applied to human skin, despite the multiple limitations
identified by the HSRB. However, the Board advised the Agency to proceed
with caution when using these data in risk assessment, as these data
cannot be used as a stand-alone set to assess dermal absorption of zinc
oxide.

HSRB Detailed Recommendations and Rationale

	The Gulson et al. (2010) study provides some potentially useful data on
the dermal absorption of zinc from zinc oxide nanoparticles in sunscreen
applied to human skin. The analytical methods used in the study appeared
to be sound. The use of the 68Zn as a tracer for Zn from the sunscreen
was an innovative approach for assessing absorption and the presence of
a chemical in body fluids (e.g., blood or urine) when the same chemical
is also present endogenously.  The Board also agreed with the comments
of the Agency (Ryman and Dole 2011) that the distinction between bulk
and nanoparticles was not as absolute as indicated in the study since
the lower range of size for bulk particles is within the range of the
size for nanoparticles.  This overlap in size between the groups of
particles should not affect the outcome of the studies and the
interpretation of the results significantly. However, the Board raised a
number of caveats, including lack of a proper control group, small
sample size, and sample contamination, which limits the utility of this
study in risk assessment of zinc oxide-based sunscreen.

	While the analytical approaches used in the Gulson et al. (2010) study
to measure Zn in blood and urine were solid, the study design itself was
faulty.  For example, there were no untreated controls to verify the
consistency of Zn levels in untreated individuals.  The presence of
increases in 68Zn in the individual who applied the sunscreen (hereafter
referred to as “the applicator”) creates questions regarding the
changes that might occur with time independent of sunscreen application.
 The design was not a carefully controlled experimental trial, and
therefore was subject to a variety of behavioral differences among the
subjects as they pursued their beach recreation activities that are
likely to have influenced the results.

	

	Despite the fact that the tracer analytical technique that measured Zn
in blood and urine did an excellent job in distinguishing endogenous Zn
from Zn absorbed from the sunscreen, this procedure did not distinguish
between dermal penetration of soluble Zn from penetration of intact
nanoparticles, or from absorption via dermal versus oral routes.
Therefore, this study has many uncertainties associated with the
interpretation of the data, and does not provide information regarding
the penetration of nanoparticles through intact, healthy skin.  The
study thus is of limited value in judging the level of intact
nanoparticle absorption through intact skin.

	

	The sunscreen formulation contained isopropyl myristate, which is a
known chemical enhancer of skin penetration, while the wash off solution
used contained alcohol and lanolin. The impact of these two variables on
Zn absorption was not considered, nor did the study report explain how
much Zn was removed by the daily alcohol and lanolin wash off procedure.
 Another source of concern is the potential impact of
ethylenediaminetetraacetic acid (EDTA), a commonly used preservative and
stabilizer, on the zinc oxide nanoparticle formulation.  The authors did
not address the possibility that EDTA can chelate ionic Zn released from
the nanoparticle formulation.  This is an important consideration since
chelation of free zinc in the skin by EDTA can influence the degree of
dermal absorption.  EDTA is routinely used in this type of sunscreen
formulation, but its impact on dermal Zn absorption was not addressed.  

	

	The final study design was based on a pilot study in which the zinc
oxide active ingredient was 51%, while the 68Zn isotope in the final
study was 99% enriched 68Zn.  The authors provide insufficient
information on the pilot study (e.g., age, gender, or skin type of
subjects).  The Board concurs with the Agency’s assessment of lack of
clarity on how the pilot study was used in the design of the final study
and subject selection.

	

	Other variables not accounted for include the frequency and time of
garment changes and the type of activities the subjects engaged in after
the first 30-minute exposure interval.  There was no mention either of
the total length of the subjects’ sun exposure and the timing between
the first and second daily exposure.  Although UV exposure was estimated
to be the same for all subjects, differences in length and intensity of
physical activity likely contributed to the degree of sweating and blood
flow to the skin.  No information was provided on the time of the second
application of the sunscreen in relationship to the first application. 
Description of dosing intervals was missing.

	

	The authors of the Gulson et al. (2010) study provided no justification
or criteria for selecting subjects 4, 8, 9 and 13 used for complete
analysis of all samples collected (Figure 8; 2010, 146).  The use of
retention rates to address the issue of urine sample contamination was a
very good approach; however, there was no mention of how these potential
contaminations occurred and the potential for hand contamination to have
caused hand-to-mouth dosing of the subjects.  The investigators also did
not consider the effect of the ZnO that was retained in the
applicator’s hand(s) on the quantities of the dosing rates reported in
Table 1 of the manuscript (Gulson et al. 2010, 143).

	

  In addition, it appears that the authors did not attempt to minimize,
control for, or account for the possibility of the contaminated
hand-to-mouth route of exposure among their subjects.  While the authors
provided a rather convincing post-hoc series of arguments to support a
hypothesis of contaminated urine samples (Gulson et al. 2010, 146-7),
they provided no description of a mechanism for how the zinc would have
gotten from a subject’s back into the urine sample.  Since the
collection containers were pre-cleaned and the beach towels provided to
the subject were renewed daily in order to eliminate any potential
source of sample contamination, the reader is left to infer that six of
the ten female subjects somehow got a portion of the sunscreen from
their backs onto their hands and subsequently from their hands into the
urine collection containers.  If this transport of sunscreen occurred
with these subjects, then it is also plausible that 68Zn from the
sunscreen was on the hands of all the subjects but that not all subjects
contaminated their urine samples.  If that were the mechanism, then it
also seems likely that sometime during the day these subjects had one or
more occasions to transfer sunscreen into their mouths via food,
beverages, mouth wiping and other hand-to mouth activities.  Thus, it is
plausible, if not likely, that the hands of many or all subjects were
contaminated with 68Zn, and that either some (if not most) of the 68Zn
measured in blood and urine was the result of the transfer from the
subject’s back into their hands and then to their mouths, leading to
gastrointestinal absorption of 68Zn.  This alternative plausible
mechanism could completely negate the validity of the authors’
conclusions regarding dermal absorption.

	

	Gulson and his colleagues may have also made significant errors in the
quantities of the doses reported in Table 1 (Gulson et al. 2010, 143) by
not considering the amount of ZnO that was retained on the
applicator’s hands after dosing each subject.  The report does not say
whether or not the applicator was wearing gloves or using the same
one-finger application technique as described for their pilot study. 
Gloves would have been important in the pilot study to avoid particulate
contamination of the samples being generated for examination by scanning
electron microscopy.  It is reasonable to expect that a significant
amount of sunscreen would have been retained on the applicator’s glove
or hand after each application.  It is also reasonable to expect that a
larger fraction of the first application was retained on the
applicator’s hand than the fraction retained from subsequent
applications.  Thus, the effect that the retention of sunscreen on the
applicator’s gloves or hands would have had on individual subjects
cannot be determine⹤ऍ

	While the source of the Δ68Zn% in the blood from the applicator
reported in Figure 4B (Gulson et al. 2010, 144)  is unknown, the
similarity of the Δ68Zn% in her blood to that reported for the blood of
the test subjects indicates that the applicator’s absorbed dose was
similar to that of the test subjects.  Both of the following possible
sources of the applicator’s dose have serious implications to the
results.  If the source were via the hand-to-mouth route, contamination
due to field practices would seem to be even more widespread than just
intra-subject.  The source of her 68Zn could be from dermal absorption
through her hands, but if the absorption rate through hands is less than
through the skin on the back (as has been reported in studies assessing
anatomical variations in dermal absorption of OP insecticides), the
similarity of the blood and urinary excretion rates for the applicator
and the subjects would imply that the amount retained on the
applicator’s hands could have been several times larger than the
dermal doses reported to be applied to the subject’s backs in Table 1
(Gulson et al. 2010, 143). Thus, such dose values may be inaccurate.

	

	While the authors did not even mention isotope fractionation in their
article, the effect of isotope fractionation probably caused the authors
to slightly over-estimate their calculated Zn absorption factor. 
Isotope fractionation refers to the separation of one isotope in
comparison to either a lighter or heavier isotope of the same element;
in this case, the human body’s preferential absorption or retention of
68Zn versus 64Zn. This phenomenon was reported to occur in both of the
two “other studies of biological samples” cited by Gulson et al.
(2010, 144).  Isotopic fractionation may have been the unrecognized
cause of the finding by Gulson et al. that the mean 68Zn/64Zn ratio in
their pre-exposure blood samples (0.4158) was 7% higher than the
18.8%/48.6% = 0.3868 natural background referred to in the manuscript
(2010, 141). However, the effect of isotope fractionation may only
reduce the68Zn/64Zn ratio by a few percent.

	

	The researchers also failed to establish the temporal stability of an
individual’s 68Zn/64Zn ratio upon which their conclusion of an
increase of that ratio in blood or urine is based (Gulson et al. 2010). 
The only example of time-series data that was located in the material
presented to the Board was the results by Ohno et al. (2005), which
measured the 68Zn/64Zn ratio in blood samples collected once per month
from just one person in which the average month-to-month variation was
reported to be ±6.6% (a coefficient of variation of ±6.2%).  This
natural variation is considerably more than the 0.42% and 0.23% mean
post-exposure increases reported by Gulson et al. (2010) over 13 and 19
days from day 0 to the end of the exposures and the six-day follow-up,
respectively.  The magnitude of this natural variation and the lack of
an untreated control group suggest that the accuracy of the reported
absorption factor could have been erroneous.

	

	Regarding gender-related differences in dosing, the authors provide
insufficient information for the reader to tell whether there was any
interaction between the subject’s individual body weights (not
reported) and “a significant difference between doses for males and
females with a mean of 4.6 mg/cm2 for males versus 3.7 mg/cm2 for
females” (Gulson et al. 2010, 142).  Without any implication of
probability, it is possible that the fact that significant differences
between genders were found only when comparing their change in 68Zn/64Zn
ratios without body weight (reported in Figure 5A) but not when
comparing their total Zn content in whole blood using the subject’s
weight (reported in Figure 5B) was a spurious association.  However, the
author’s

Δ68Zn% did take into account differences in gender and body weight by
adjusting for fat-free body mass via BMI.

	

	Regarding study sample selection, there are two major statistical
concerns with the sample’s demographic constitution.  The report
indicates that the data came from 20 subjects split between the
nanoparticle group, with 6 females and 5 males and the bulk group with 4
females and 5 males (Table 1; Gulson et al. 2010, 143).  This table,
listing the subjects’ demographics, showed greater similarities among
the subjects due to family connections.  Therefore, the question arises
whether the sample of individuals used in the study represents a random
sample of the reference population or whether is it a convenience
sample.  Because the experimental error or within treatment variation
could be underestimated in a non-random, convenient, and small sample,
the statistical validity of the data used in this study is questionable.
 The second issue is related to the known outlier (subject 7, enrolled
in the nanoparticle group).  This subject’s observed data point was
relatively larger than all other data points (e.g., Figures 4 and 5;
Gulson et al. 2010, 144).  Also, an adverse or allergic reaction was
mentioned in the report as the likely reason for the large response
observed in the case of subject 7.  Therefore, this single observation
could be the main cause of the observed skewedness in the experimental
data. The researchers performed a log transformation to correct this
high degree skewedness, which may be caused by this single outlier.  The
statistical significance reported in the paper (gender x treatment) may
have disappeared if they had excluded this observation from the analysis
or performed any robust weighted linear model analysis to adjust for
this influential data. 

	

	Based on these concerns, the statistical validity of the data used and
the conclusion derived may be questionable. Thus, the results reported
could be treated as exploratory in nature and further studies are needed
to confirm the conclusion.	

Ethics

Charge to the Board

	Is there adequate information to determine that the Gulson et al.
(2010) study was conducted in substantial compliance with procedures at
least a protective as those in subparts A - L of EPA’s regulation at
40 CFR Part 26? 

	

Board Response to the Charge

HSRB Recommendation

	The Board concluded that there was insufficient information available
at the time of the April meeting to determine whether the Gulson et al.
(2010) study was conducted in substantial compliance with procedures at
least as protective as those in subparts A - L of EPA’s regulation at
40 CFR Part 26. The Board recommended that EPA seek additional
information from Macquarie University or the study investigator,
including a copy of the research protocol and an unsigned informed
consent form, and that this information be provided to the HSRB for
re-consideration.

HSRB Detailed Recommendation and Rationale

	This is the first time EPA has asked the HSRB to review a study from
the published literature that was conducted after EPA’s amended Rule
for the Protection of Human Subjects of Research. The study was
conducted in Sydney, Australia and received approval from the ethics
committee of Macquarie University, as required by the Australian Code
for the Responsible Conduct of Research (2007).

	As noted in EPA’s ethics review (Parsons 2011), 40 CFR Part 26,
subparts A through L do not apply; the research was neither conducted or
supported by EPA nor was it conducted by a person with the intention to
submit the results to EPA. FIFRA §12(a)(2)(P) does not apply to this
research since the compound tested is not a pesticide. Thus, the
applicable acceptance standards are as follows:

40 CFR §26.1703. Prohibition of reliance on research involving
intentional exposure of human subjects who are pregnant women (and
therefore their fetuses), nursing women, or children. Except as provided
in §26.1706, in actions within the scope of §26.1701 EPA shall not
rely on data from any research involving intentional exposure of any
human subject who is a pregnant woman (and therefore her fetus), a
nursing woman, or a child.

40 CFR §26.1705 Prohibition of reliance on unethical human research
with non-pregnant, non-nursing adults conducted after April 7, 2006.
Except as provided in §26.1706, in actions within the scope of
§26.1701, EPA shall not rely on data from any research initiated after
April 7, 2006, unless EPA has adequate information to determine that the
research was conducted in substantial compliance with subparts A through
L of this part, or if conducted in a foreign country, under procedures
at least as protective as those in subparts A through L of this part.
This prohibition is in addition to the prohibition in §26.1703.

	The HSRB determined that most of the information that the Board has
about the ethical conduct of the Gulson et al. (2010) study is without
supporting documentation and is thus inferential. Although the Board
thought that there is no reason to believe that the conduct of the
reported research was fundamentally unethical, in the absence of
supporting documentation like the informed consent form and research
protocol, the Board concluded that it did not have adequate information
to determine that participant selection was voluntary, that information
about risks and benefits was sufficiently communicated, that study
participants understood the nature of the research or their right to
withdraw from the study, or that the female participants were neither
pregnant nor nursing. Thus, the Board concluded that it lacked adequate
information to determine that the research was conducted under
procedures at least as protective as those in subparts A through L of 
40 CFR Part 26.

	In the absence of the informed consent form, the Board had insufficient
information to determine how study risks and benefits may or may not
have been characterized, whether participants were informed of their
right to withdraw from the study, whether participation involved
reasonable or unreasonable compensation or pressure to participate with
family members. The Board offers the following reasoning for its
recommendation:

1.	Acceptable risk-benefit ratio.

The Board agreed with the EPA review (Parsons 2011) that the main risks
to participants were in blood sampling and possible adverse reaction to
the sunscreen formulations.  According to Gulson et al., the sunscreen
formulations contained an enriched level of a stable isotope of zinc,
which does not pose a higher risk than naturally occurring zinc and
“many” commercially available sunscreens contain “nanoparticulate
TiO2 and/or ZnO” (2010, 140). Thus, although the risk-benefit balance
was not discussed in the study report, the risks appear to be low.

No minors were enrolled in the study. However, apart from an email from
the study investigator to the Agency (Parsons 2011), the Board had no
evidence that pregnant or lactating women were excluded from
participation. 

The study reports that one participant (Subject 7) had an adverse
reaction to the sunscreen. The nature and severity of the adverse
reaction was not described.

According to the published study, the risks were mitigated as follows:
1) risks of sun exposure were mitigated with UV-protective clothing and
advice to wear sunscreen; 2) Subject 7 ceased participation in the
application portion of the study after an adverse reaction; and 3) a
trained phlebotomist drew participants’ blood.

The Board concluded that although the risks of the study were low, the
“adequacy of information” standard should not be strictly correlated
with risk. That is, the threshold for adequacy of information should not
be understood in proportion to the amount of risk posed by the study (if
the risk is low, then the threshold for adequacy can be low). The Board
determined that such a standard would exclude other requirements of
ethical research including voluntary informed consent, respect for
participants, and fair participant selection.

2.	Voluntary and informed consent of all participants. 

The Board concluded that it did not have adequate information to
determine that participant selection in the Gulson et al. (2010) study
was conducted in substantial compliance with procedures at least as
protective as those in subparts A - L of 40 CFR Part 26.

Board members raised a concerns about the fairness and voluntariness of
participant selection, including: 1)	the investigator holding
recruitment discussions with potential participants in his home; 2)
selecting participants who were “personal contacts” of the
investigator; 3) inclusion of participants from the same family,
suggesting that the study was conducted in a context where subjects may
have experienced undue pressure to participate; and 4) a lack of
information about the type and level of compensation that participants
were offered. In addition, although the published paper states that
“venous blood and urine samples were collected 8 days before
exposure” (Gulson et al. 2010, 140) an email from the study
investigator to the Agency (Parsons 2011) stated that “the subjects
attended a meeting at my house 1 week before the trial started and I
described the trial and any risks.” This apparent inconsistency in the
timing of the informed consent process and the collection of
study-related samples raises some concerns.

	Although the Board recognizes that this research was reviewed and
approved by the Macquarie University Human Research Ethics Committee,
which is registered with the Australian National Health and Medical
Research Council, it concluded that ethics committee approval alone does
not provide adequate information to determine that the Gulson et al.
(2010) study was conducted in substantial compliance with procedures at
least as protective as those in subparts A - L of EPA’s regulation at
40 CFR Part 26. The Board thus recommended that EPA seek additional
information from Macquarie University or the study investigator,
including a copy of the research protocol and unsigned informed consent
form, and that this information be provided to the HSRB for
re-consideration.REFERENCES

Australian National Health and Medical Research Council, Australian
Research Council, Australian Vice-Chancellor’s Committee. 2007.
National Statement on Ethical Conduct in Human Research. Canberra, AUS:
Australian National Health and Medical Research Council. Available
online at   HYPERLINK
"http://www.nhmrc.gov.au/publications/synopses/e72syn.htm" 
http://www.nhmrc.gov.au/publications/synopses/e72syn.htm . Accessed
April 14, 2011.

Boatwright, M. 2007. Determination of Removal Efficiency of Didecyl
Dimethyl Ammonium

Chloride (DDAC) from Hand Surfaces Using Isopropyl Alcohol/Water Wash.
Study Number 050211. Unpublished study prepared by Golden Pacific
Laboratories. MRID 47214801.

Bruce, E.D., for the Agricultural Handler Exposure Task Force (AHETF).
2010a. Determination of Dermal and Inhalation Exposure to Workers During
Airblast Applications of Liquid Sprays Using Open Cab Equipment in
California Trellis Crops. Study Number AHE62. Dated November 3, 2010.
Unpublished document prepared by the AHETF. MRID 48289611.

Bruce, E.D., for the AHETF. 2010b. Determination of Dermal and
Inhalation Exposure to Workers During Airblast Applications of Liquid
Sprays Using Open Cab Equipment in New York Trellis Crops. Study Number
AHE63. Dated November 3, 2010. Unpublished document prepared by the
AHETF. MRID 48289612.

Bruce, E.D., for the AHETF. 2010c. Determination of Dermal and
Inhalation Exposure to Workers During Airblast Applications of Liquid
Sprays Using Open Cab Equipment in Oklahoma Tree Nuts. Study Number
AHE64. Dated November 3, 2010. Unpublished document prepared by the
AHETF. MRID 48289613.

Bruce, E.D., for the AHETF. 2010d. IIRB Correspondence Report for
Cluster Report AHE62. Dated November 4, 2010. Unpublished document
prepared by the AHETF. MRID 48289614.

Bruce, E.D., for the AHETF. 2010e. IIRB Correspondence Report for
Cluster Report AHE63. Dated November 4, 2010. Unpublished document
prepared by the AHETF. MRID 48289615.

Bruce, E.D., for the AHETF. 2010f. IIRB Correspondence Report for
Cluster Report AHE64. Dated November 4, 2010. Unpublished document
prepared by the AHETF. MRID 48289616.

Carley, J., and J. Evans. 2008a. Science and Ethics Review of AHETF
Scenario Design and Protocol AHE62 for Exposure Monitoring in California
Trellis Crops. Dated September 23, 2008. Unpublished document prepared
by the Health Effects Division, United States Environmental Protection
Agency.

Carley, J., and J. Evans. 2008b. Science and Ethics Review of AHETF
Scenario Design and Protocol AHE63 for Exposure Monitoring in New York
Grapes. Dated September 23, 2008. Unpublished document prepared by the
Health Effects Division, United States Environmental Protection Agency.

Carley, J., and J. Evans. 2008c. Science and Ethics Review of AHETF
Scenario Design and Protocol AHE64 for Exposure Monitoring in Oklahoma
Pecans. Dated September 23, 2008. Unpublished document prepared by the
Health Effects Division, United States Environmental Protection Agency.

Crowley, M. 2011. Review of Agricultural Handler Exposure Task Force
(AHETF) Open Cab Airblast Applicator Exposure Monitoring Studies: AHE62,
AHE63, AHE64. Dated March 1, 2011. Unpublished document prepared by the
Health Effects Division, United States Environmental Protection Agency. 

Crowley, M., and B. Sarkar. 2011. Review of Agricultural Handler
Exposure Task Force (AHETF) Monograph: Open Cab Airblast Application of
Liquid Sprays. Dated January 11, 2011. Unpublished document prepared by
the Health Effects Division, United States Environmental Protection
Agency.

EPA Human Studies Review Board. 2006. June 27-30, 2006 Human Studies
Review Board Meeting Report.

EPA Human Studies Review Board. 2008a. April 9-10, 2008 Human Studies
Review Board Meeting Report. 

EPA Human Studies Review Board. 2008b. October 21-22, 2008 Human Studies
Review Board Meeting Report.

EPA Human Studies Review Board. 2009. June 24-25, 2009 Human Studies
Review Board Meeting Report.

EPA Human Studies Review Board. 2010. October 27-28, 2010 Human Studies
Review Board Meeting Report.

EPA Scientific Advisory Panel. 2007. January 7-12, 2007 Scientific
Advisory Panel Meeting Report.

Gulson, B., M. McCall, M. Korsch, et al. 2010. Small Amounts of Zinc
from Zinc Oxide Particle in Sunscreens Applied Outdoors are Absorbed
Through Human Skin. Toxicological Sciences 118(1): 140-9. MRID 48387301.

Independent Institutional Review Board, Inc. (IIRB, Inc.). 2010a. Human
Research Protection Plan. Unpublished materials prepared by IIRB, Inc.

IIRB, Inc. 2010b. Current Membership Roster. Unpublished materials
prepared by IIRB, Inc.

Kisicki, J, C. Seip, and M. Combs. A Rising Dose Toxicological Study to
Determine the No-Observable-Effect-Levels (NOEL) For Erythrocyte
Acetylcholinesterase (AChE) Inhibition and Cholinergic Signs and
Symptoms of Chlorpyrifos at Three Dose Levels. Dated April 15-19, 1999.
Unpublished study prepared by MDS Harris Laboratories under Project No.
21438 and Dow AgroSciences Study No. DR K-0044793-284. MRID 44811002.

Klonne, D.R., and L.R. Holden for the AHETF. 2010. Agricultural Handler
Exposure Scenario Monograph: Open Cab Airblast Application of Liquid
Sprays. Dated December 15, 2010. Unpublished document prepared by the
AHETF. MRID 48326701. 

Leighton, T. 2011. Science Review of the AEATF II Wipe Human Exposure
Monitoring Study

(Trigger Spray & Wipe and Ready-to-Use Wipes Scenarios). Dated March 17,
2011. Unpublished document prepared by the Office of Pesticide Programs,
United States Environmental Protection Agency.

Macquarie University Ethics Review Committee. 2011. Research Ethics
Correspondence and Approval Letters. Unpublished document submitted to
the United States Environmental Protection Agency.

Ohno, T., A. Shinohara, M. Chiba, and T. Hirata. 2005. Precise Zn
Isotopic Ration Measurements of Human Red Blood Cell and Hair Samples by
Multiple Collector-ICP-Mass Spectrometry. Analytical Sciences 21(4):
425-8. 

Parsons, L. 2011. Ethics Review of Intentional Exposure Human Toxicity
Study. Dated March 16, 2011. Unpublished document prepared by the Office
of Pesticide Programs, United States Environmental Protection Agency.

Ryman, J. 2011. Data Evaluation Record: 68Zinc Oxide
Nanoparticles/888502. Dated March 15, 2011. Unpublished document
prepared by the Health Effects Division, United States Environmental
Protection Agency. 

{

|

U

V

¹

º



 

 

(

)



(

)

e

j



€

$

 document prepared by Golden Pacific Laboratories, LLC. MRID 48375601.

Sherman, K. 2011a. Ethics Review of Completed AHETF Open Cab Airblast
Scenario Worker Exposure Monitoring Study. Dated March 8, 2011.
Unpublished document prepared by the Office of Pesticide Programs,
United States Environmental Protection Agency.

Sherman, K. 2011b. Ethics Review of Completed AEATF II Wipe Scenario
Worker Exposure Monitoring Study. Dated March 17, 2011. Unpublished
document prepared by the Office of Pesticide Programs, United States
Environmental Protection Agency.

Smith, L.D. 2005. Determination of Dermal and Inhalation Exposure to
Workers During Application of a Liquid Pesticide Product by Open Cab
Airblast Application to Orchard Crops. Study Number AHE07. Amended
August 23, 2005. Unpublished document prepared by the AHETF. MRID
46448201.

Final Draft Document Dated May 20, 2011

Page   PAGE  2  of   NUMPAGES  30 

Page   PAGE  5  of   NUMPAGES  30 

It was the study investigators, not the Agency, that speculated how this
could occur. The Agency merely included it in the review. 

DRAFT

