October 23rd and 24th, 2019 EPA Human Studies Review Board Meeting Report


EPA-HSRB-20-1
Dr. Jennifer Orme-Zavalefa
1 IF \ Science Advisor
< )|i"tce of the Science Advisor
' ^*0 Pennsylvania Avenue, NW
Washington, DC 20460
Subject: October 23rd and 24th. 2019 EPA Human Studies Review Board Meeting Report
Dear Dr. Omie-Zavaleta.
Human Studies Review Board (HSRB) provide scientific and ethics review of two completed
studies involving human participants. The study report titled "Field Evaluation of Two Topically
Applied Insect Repellent Products Containing IR3535 Against Mosquitoes in Florida"
summarized completed research sponsored by LivFul, Iric and conducted by the University of
Florida Repellent. Testing Lab to determine the efficacy of skin-applied repellents against
mosquitoes in a field setting. The study report by the Antimicrobial Exposure Assessment Task
force II titled "A Study for Measurement of Potential Dermal and Inhalation Exposure During
the Application of Paint Containing an Antimicrobial using an. Airless Sprayer" (AEA10)
summarized completed, research to evaluate the potential dermal and inhalation, exposures of
workers using airless paint sprayer equipment to apply paints containing antimicrobial pesticides.
The HSRB's responses to the charge questions presented at the meetings on October 23rd and
24th, 2019 meetings along with detailed rationale and recommendations for their conclusions on
these studies are provided in the enclosed final meeting report.
The United States Environmental Protection Agency (EPA) requested that the
Signed,
JcniiiUi t aval lad, ScD, CIH
t 'hair { P \ Human Sutdi.es Review Board
1

-------
INTRODUCTION
On October 23-24, 2019, The United States Environmental Protection Agency (EPA or Agency)
Human Studies Review Board (I	ISRB or Board) met to address the scientific and ethical charge
questions related to a completed study titled "Field Evaluation of Two Topically Applied Insect
Repellent Products Containing 1R3535 Against Mosquitoes in Florida" and a completed study
titled "A Study for Measurement of Potential Dermal and Inhalation Exposure During the
Application of Paint Containing an Antimicrobial using an Airless Sprayer'' (AEA10), In
accordance with 40 CPR 26.1601, EPA sought HSRB review of these completed studies. Each
study is discussed more fully below,
REVIEW PROCESS
The Board conducted a public meeting on October 23-24,-2019, Advance notice of the meeting
was published in the Federal Register as "Human Studies Review Board; Notification of a
Public Meeting" (EPA, FRL-10001-G3-ORD). This Final Report of the meeting describes the
HSRB's discussion, recommendations, rationale and consensus in response to the charge
questions on ethical and scientific aspects of the completed research.
For each agenda item, the Agency staff presented their review of scientific and ethical aspects of
the research, with each presentation followed by clarifying questions from the Board. The HSRB
solicited public comments and next proceeded to address the charge questions under
consideration. The Board discussed the science and ethics charge questions and developed a
consensus response to each question in turn. For each of the charge questions, the Chair called
for the Board to vote to confirm concurrence on a summary statement reflecting the Board's
response.
For their evaluation and discussion., the Board considered materials presented at the meeting,
study reports, related materials and documents provided by the study sponsors, the Agency's
science and ethics reviews of the study, as well as oral comments from Agency staff and the
investigators during the HSRB meeting discussions. A comprehensive list of background
documents is available at https://www.epa.gov/osa/october-23-24-2019-meeting-human-studies-
review-board.
1

-------
The HSRB review of each of the reported studies is presented below, in the order in which, the
studies were reviewed at the meeting.
Field Evaluation of Two Topically Applied Insect Repellent Products Containing IR3535
Against Mosquitoes in Florida
Charge to the Board- Science:
Did the study "Field Evaluation of Two Topically Applied Insect Repellent Products Containing
IR.3535 Against Mosquitoes in Florida" generate data that are scientifically reliable and useful
for estimating the amount of time each of the products tested repels mosquitoes?
Response to the charge question.:
The HSRB concluded that the research presented in the final report "Field Evaluation of Two
Topically Applied Insect Repellent Products Containing IR3535 Against Mosquitoes in Florida"
generated data that are scientifically reliable and useful for estimating the amount of time each of
the products tested repels mosquitoes where the Complete Protection Time (CPT) for both
products can be established at 3 hours.
The HSRB also has specific comments, recommendations and additional minor points which are
described in the discussion below,
HSRB detailed response and rationale:
The 1R3535 study is a field evaluation of two topically applied insect repellent products (i.e.,
wipe and a lotion) containing IR3535 in order to estimate the amount of time each of the
products tested repels mosquitoes. Therefore, the study's aim is to determine the median
Complete Protection Time (rnCPT). The active ingredient in the product is 20% w/w of the
active product ethyl butyl acetyl aminopropionate (I.R3S35),
On July 26th, 2017, the HSRB met to review the proposed, study and EPA and the HRSB
proposed, a number of changes to the protocol and therefore study design. The proposed, changes
fell into five categories: 1) provision of more background information in the protocol document;
2) clarifications regarding the study design; 3) clarification of planned, statistical methods; 4)
1

-------
provision of further detail about study procedures; and 5) correction of minor discrepancies in
the protocol document,
Through review of the amended study protocol (2018 version), the study sponsor's report and
EPA science and ethical review, it appears that the study sponsor did an adequate job of
responding to requested changes suggested by HRSB and EPA before the start of the study in
summer of 2018, Some of the specific changes included: better description of demographic
characteristics of the population within the recruitment area, description of pre-test aspiration
and training requirements, provision of soap and how the test skin area will be cleaned, guidance
for subjects between testing to prevent contamination of test area, separation of inclusion and
exclusion criteria, testing for other vector-borne pathogens, provisions for health requirements of
subjects, and other minor consent document changes.
Performance of Study
Researchers appeared to have followed all protocol stipulations in conducting the study in terms
of pre-testing for attractiveness, capturing and testing mosquitoes for diseases (i.e., viral
pathogen testing and pre testing of species distribution);, use of aspirator training, selection of
subjects and randomization into groups, product application and testing activities throughout test
days. There are a number of aspects important to achieve during the study and those include: 1)
adequate landing pressure on controls and 2) sufficient number of subjects achieving failure
during the testing period, defined as of least seven subjects foiling (achieving confirmed landing)
before the 16-hour post application period, where the product is applied 2 hours prior to any
exposure to mosquitoes.
Test days (or Trials) 2 and 3 were excluded from the calculations due to low mosquito landing
pressure. Test day 4 was excluded from calculations of mCPT as only six subjects experienced
failure, which is below the requirement of at least seven subjects failing; however, the data from
this test day was used to estimate the CPT and considered by EPA. The mCPT was calculated
using data from test days 1 and 6 for the lotion application, and test day 5 for the wipe.
Study Deviations and Protocol Amendments
Assessment of deviations and protocol amendments made during the study are important to
consider and review as they may impact the scientific reliability of study results. It should be
1

-------
noted that some of the items noted below were deviations, but later became protocol
amendments,
1)	Researchers implemented a key protocol change following test day 1 and test day 2: To
reduce the testing requirement from landing pressure of 5 mosquito landings in 5 minutes
or less on two control subjects to 5 mosquitoes landing within 5 minutes or less on one
control subject.
a. This amendment was submitted for the purposes of continuing the study, rather
than for the calculation of mCPT.
2)	Not all landing mosquitoes could be collected via aspirator on test days 1, 2 and 3. A
protocol amendment was approved before test day 4, indicating that it was more
important to avoid bites to subjects than to capture all mosquitoes,
a. The HSRB agrees with the EPA and sponsor assessments that this did not have a
significant impact on the scientific reliability of the study.
3)	Some intervals were longer than the 30 minutes between exposure periods, due to
logistical reasons.
a. The HSRB agrees with the EPA and sponsor assessments that this did not have a
significant impact on the scientific reliability of the study.
4)	A protocol amendment was approved to clarify the language for skipping exposure
periods due to low mosquito landing pressure. Some exposure periods were skipped on
test day 1.
a. The HRSB agrees that with EPA that exposure periods should not be skipped for
low landing pressure, and only for weather conditions. As clarified by EPA, only
if a number of exposure periods have low landing pressure for all controls, the
study should he stopped (i.e., for this study that should be 5 consecutive exposure
periods). However, it is notable that at least two periods with insufficient landing
pressure did achieve CPT (test day 5). It is unclear as to whether activities before
this change affected the outcome of the study.
5)	A protocol amendment was approved to replace a spatula with a beaker for weighing of
lotion applied, given fluidity and volume needs for the lotion.
a. The HSRB agrees with the EPA and sponsor assessments that this did not have a
significant impact on the scientific reliability of the study.
1

-------
Determining a reliable and accurate mCPT for the two product applications (i.e., wipe and
lotion).
The lotion and wipe products will be marketed and must perform according to labelling under
low- or high-pressure biting conditions. Therefore, the product must be tested under adequate
landing pressure as determined by the minimum requirement for 5 mosquitoes landing within 5
minutes or less on untreated controls. The study sponsor attests that the data support a CPT of 14
hours for both'product formulations while EPA disagrees based on inadequate landing pressure
on test days 5 and 6 and supports a CPT of 3 hours. The factors related to the difference in CPT
are associated with both landing pressure as well as attractiveness of the controls, EPA identified
inadequate landing pressure on test day 5 and test day 6 during the middle of the test period
(between dawn and dusk) with inconsistency across subjects, which EPA interpreted was due to
low activity, and not just to mosquitoes' preference for either subject. The H5RB agrees with
this assessment. All subjects were tested for attractiveness and their suitability for the study was
confirmed. It is the HSRB's assessment that lack of attractiveness for a control is then
inappropriate to claim in the progression of the study. Low landing pressure can be due to a
number of other undetermined factors.
Given the current data, specifically the failure to achieve agreed-on landing pressure, the HRSB
agrees with EPA that the data do not support a CPT of 14 hours and a reasonable interpretation
of 3 hours CPT is suitable and protective of public health at this time.
Recommendations
The HSRB suggests the following changes and clarifications to study protocols:
1) For field evaluation of insect repellents, test sites should be screened for a diversity of
mosquitoes across time and adequate landing pressure prior to study initiation.
2) We urge better communication between the EPA and the Sponsor. In particular, EPA
should provide adequate guidance on the adaptation of the landing pressure rule in
OCSPP 810.3700 to longer exposure times (i.e., repellents expected to be effective for
longer durations, and which therefore require longer testing periods during efficacy
1

-------
trials), and when a study should be stopped (either for weather or the number of low
landing pressure consecutive exposure periods).
3) Since the median complete protection time (niCPT) is the time when 50% of the treated
people will remain protected from landings - this also means that 50% of the treated
people will not be protected. The mCPT may not be sufficiently -protective of public
health for CPT calculations and EPA should explore more conservative protection factors
(e.g., the 10th percentile).
4) In future studies, EPA should be concerned and work with study investigators when a
number of deviations and amendments are requested as the study is being conducted.
Neglecting to do this may lead to issues of result interpretation.
5) For both the wipe and the lotion applications, from the documents provided it is unclear
how dose is translated to a usable metric for the public, particularly given how the wipe
dose in the study was applied in a different manner than would be applied by the public.
The wipe was squeezed to release the product. This product amount was then weighed
and applied to the skin for this study. EPA and/or the study sponsor should provide the
relevant studies that support this methodology to the Board. Furthermore, the relevant
information should be referenced in the final study report.
Statistical review
The statistical design and analysis of the protocol are appropriate for EPA's intended use of the
data. Kaplan Meier survival analysis is appropriate for calculating mCPT for test day 1, 5, and 6.
Test day 4 has more than 50% of data censored so the Kaplan-Meier analysis should not be used
for computing mCPT for test day 4. The treatment of censored observations is also appropriate.
CHARGE TO THE BOARD - ETHICS
Does the available information support: a determination that the research was conducted in
substantial compliance with the applicable requirements of 40 CFR part 26, subpart Q?
Response to the charge question:
1

-------
The Board believes that this research was conducted in substantial compliance with the
applicable requirements of 40 CFR part 26, subpart Q.
HSRB detailed response and rationale:
This study was conducted to determine the complete protection time/efficacy of repellents
applied to the skin that contained 20% IR3535 (ethyl butyl acetyl aminopropionate). A lotion and
a wipe formulation were tested. Testing was conducted at the University of Florida Repellent
Testing Lab.
Independent ethics review:
The draft protocol was approved by the University of Florida Institutional Review Board (IRB)
#1 which is registered with the Office of Human Research Protections (OHRP) and accredited by
the Association for the Accreditation of Human Research Protection Programs (AAHRPP). The
draft protocol was reviewed by the EPA, and by the HSRB at the 26 July 2017 public meeting.
Changes to the protocol were proposed by both EPA and the HSRB. The researchers revised the
protocol to address the EPA and HSRB requested changes. The IRB approved the protocol,
consents and other study materials on 21 March 2017 and the study was closed on 20 December
2018.
Selection of subjects
Recruitment was conducted in the Gainesville, PL area using IRB-approved materials and was in
substantial compliance with the protocol, A total of 70 persons were recruited for the study, 68
completed the consent process and 38 subjects (20 males, 18 females) participated in at least one
test day.
Potential participants who responded to advertisements were contacted by phone using a script
and, if interested, were invited to a meeting with the study team to learn more and to consent to
participate. Study subjects were screened via questionnaire and required to be between 18 and 55
years old. Other inclusion/exclusion criteria were as defined in the protocol. Subjects were
required to speak and understand English. Pregnant and nursing women were excluded, as well
as employees of the University of Florida, the study sponsor, and their spouses.
l

-------
Next, subjects had a training visit. One deviation occurred during a training visit when a female
subject participated in the mosquito attractiveness test before completing the screening
questionnaire and pregnancy testing protocol. The subject subsequently completed the pregnancy
test and was not pregnant. This deviation was reported to the IRB but it was determined that it
was not serious and not continuing, so no action was needed. However, the study director
subsequently created an additional checklist to prevent this from occurring again.
All subjects were at least 18 years of age. Pregnancy testing of female subjects was conducted on
each day of testing and no pregnant or lactating women were enrolled in the study.
Informed consent
At the on-site recruitment meeting, the study was outlined and the group was allowed to ask
questions. It was emphasized that participation was completely voluntary. Subjects were allowed
to ask further questions in a private meeting and given the option to take the consent documents
home to review or to sign at the meeting. All participants signed consent documents at the initial
private meeting. Randomization and testing were performed satisfactorily as outlined in the
protocol, study report and EPA ethics review memo.
Minimization of risks
Safety precautions included limiting the amount and region of skin (lower legs) exposed to
mosquitoes during testing, use of protective gear (head nets and clothing) to cover areas not
involved in testing, use of lab-raised mosquitoes that had never received a blood meal during
mosquito attractiveness and aspirator training, testing in areas identified as not having mosquito-
borne pathogens, and coordination with local health departments and mosquito control districts.
A shaded, screened in area with chairs, snacks, water, and other drinks was provided to subjects
between testing episodes.
There was one adverse event (AE) when a subject reported a skin rash that appeared to be poison
ivy and that the subject believed was contracted during participation in the study. This AE was
reported to the IRB which recommended revising the consent form to note poison ivy was a risk
due to the wooded location of testing sites, but this was not done as the study was already fully
enrolled and completed.
1

-------
Confidentiality was maintained throughout the study. Subjects, including alternates, were
compensated per the protocol. The amount of compensation for each subject was consistent with
the time spent on the various study-related activities and was not excessive.
There were eight protocol amendments and these were addressed consistently with IRB and
protocol procedures. One amendment noted that epi-pens would not be available during testing
and included a change in the screening questionnaire to exclude subjects who did not know how
they reacted to mosquito bites to minimize the chances of enrolling a subject with an unknown,
severe allergy to mosquito bites.
One deviation involving a subject participating in the mosquito attractiveness test prior to
completing the screening questionnaire and pregnancy test protocol has already been discussed.
Another deviation involved the study team being unable to contact one subject despite multiple
attempts until 73 hours after a test day to inquire about adverse events. The allowed time was 72
hours, but the subject had been out-of-town with a dead phone battery. The subject reported no
adverse events.
Another deviation involved difficulty capturing "all" mosquitoes landing on test and control
subjects during the test day. This resulted in a protocol amendment stating that "attempts will be
made to capture all" mosquitoes. More emphasis was also placed on not allowing mosquitoes to
bite test and control, subjects. This deviation and subsequent amendment appropriately prioritized
the subject's safety and welfare.
Several protocol deviations were reported to the IRB related to implementation of the testing and
did not impact subject health or welfare.
EPA has noted that the effective dates of some amendments predate the dates of the IRB's
approval of the amendments. The HSRB agrees with the EPA that amendments are not effective
until they have been reviewed and approved by an IRB. The HSRB agrees with the EPA's
recommendation for future studies.
A deviation related to mosquito landing pressure on untreated controls and stopping rules was
reported and resulted in. an amendment to the protocol to clarify the stopping rules.
1

-------
A Study for Measurement of Potential Dermal and Inhalation Exposure Paring the
Application of Paint Containing an Antimicrobial using an Airless Sprayer (AEA1.0I
Charge to the Board- Science:
Did the research in study AEA10 generate scientifically reliable data, useful for assessing the
exposure of individuals who apply paint containing antimicrobial pesticides using airless
sprayers?
Response to the charge question:
The H5R.B concluded that the research presented in study AEA10 are scientifically reliable data,
useful for assessing the exposure of individuals who apply paint containing antimicrobial
pesticides using an airless sprayer.
HSRB detailed response and rationale:
HSRB reviewed information provided in advance of the meeting, as well as the EPA scientific
and. ethics presentations provided at the meeting. The HSRB was in agreement with the
assessment of the results as presented in the EPA scientific report and the superiority of the
AEATF II airless sprayer data and associated unit exposures to the existing airless sprayer data
set derived from PHED data.
Working with EPA, AEATF II adequately addressed, each of the concerns raised by EPA and
HSRB within the study protocol and .related documents including, but not limited to,
randomization of ME assignments, adequate range of variation and acknowledgement of
limitations, respirator use, justification of 30 gallons of paint maximum, and adequate drying
time for dosimeters.
The sampling procedures were performed as outlined in the protocol. Protocol and SOP
deviations were noted, appropriately handled and did not adversely affect the outcomes of the
study,
The study achieved a sample size of .18 MEs performed under two AaiH concentrations (1,200 or
12,000 PPM) and. three paint volumes (10, 15, or 30 gallons), A variety of room configurations
1

-------
were used (two home and one office simulations) with a range of surface areas (1,040 ft2 to
13,673 ft2) over 58 to 192 minutes.
The QA/QC procedures were adequately performed and accounted for in the final results.
Unit exposure results were calculated for dermal and inhalation exposure routes (Table 1, EPA
report), Test subjects wore both inner and outer whole-body dosimeters that were later sectioned
and analyzed by body part to allow for the estimation of unit exposures under various clothing
conditions including long/short pants and/or long/short sleeves.
With regard to data generalizations and limitations, the H8RB is in agreement with the EPA that:
1)	the choice of one study location by AEATF II was reasonable;
2)	the data are acceptable to use as a surrogate when assessing other chemicals with low
volatility;
3)	the extrapolation of unit exposures to larger volumes of paint is possible up to 100
gallons of paint according to EPA's analysis;
4)	the potential exposures appear to be acceptable based on the Margins of Exposure (MOE)
calculations provided;
5)	the limitations of the data include:
a.	the data are not derived from a fully stratified random sample of ME;
b,	there is a potential underestimation of the face/neck wipe samples; and
e. care must be taken to extrapolate exposure/risk to consumers, whose behaviors
may be less understood.
The HERB had no additional recommendations beyond those already presented by EPA and
summarized above.
Statistical review
The results of the lognormal simple random sampling model are most appropriate for presenting
the statistics for the unit exposures. The calculations utilizing substitution of half the LOD for
non-detected values below the LOD. or of the average of the LOD and LOQ for non-deteeted
1

-------
values between the LOD and the LOQ are most appropriate. These methodologies and results
were deemed appropriate.
CHARGE TO THE BOARD - ETHICS
Does the available information, support a determination that the research was conducted in
substantial compliance with the applicable requirements of 40 CFR part 26, subpart Q?
Response to the charge question:
The Board believes that this research was conducted in substantial compliance with the
applicable requirements of 40 CFR part 26, subpart Q.
HSRB detailed response and rationale: independent ethics review
The protocol was conditionally approved by Schulman IRB in August 2017 and discussed with
the EPA HSRB on 25 October 2017, Schulman IRB and the HSRB made a number of
recommendations that were appropriately addressed by AEATF II, the sponsor, prior to final
approval of the protocol and consent (English version) by Schulman IRB in February 2018.
Schulman provided a certified Spanish translation of the consent and all relevant documents,
Oversite of the study was transferred from Schulman to Advarra in late July 2018. Both IRBs
hold Federal Wide Assurances from OHRP and are registered with OHRP, and both are
accredited by AAHRPP.
Selection of subjects
Recruitment was accomplished via English and Spanish advertising via newspapers, radio,
printed flyers, and (following an amendment due to low enrollment) Craigslist, Both English and
Spanish speaking individuals responded and were initially screened by phone and, if they met the
inclusion criteria, were invited to consent meetings. Four potential candidates requested Spanish
communications and materials, 24 requested English communications and materials. A bilingual
researcher participated in all sessions with the individuals who requested Spanish
communications and materials.
1

-------
Potential subjects were asked to complete an online, English-only health questionnaire as part of
the medical evaluation for the respirators. For the four Spanish speaking subjects, the bilingual
researcher was present and available to translate as needed as these subjects completed this form.
Thirty-five respondents were successfully screened, 28 participated in a consent meeting and
were consented to enroll in the study. Three individuals did not pass a medical screen, and one
did not pass the respirator fit test. Two subjects withdrew from the study prior to being
monitored, one due to a family emergency and one due to scheduling difficulties. In the end, 24
subjects were enrolled in the study and 18 subjects were monitored. Five were female and 13
were male, one preferred Spanish and the rest preferred English. The six subjects who were not
monitored were alternates.
All subjects were at least 18 years old, and no pregnant or lactating women were involved in the
study.
Informed consent
Potential candidates read the informed consents, had the consents reviewed with them, by the
researcher conducting the meeting, and had their questions answered. Potential candidates were
informed that they were free to withdraw from the study at any time. Potential candidates were
also informed of the need to wear a properly fitted respirator during the study and the medical
evaluation and it testing processes were explained. Researchers asked a standard set of questions
to ensure comprehension of the consent materials.
Minimization of risks
During the study, subjects were monitored for safety purposes as outlined in the protocol, study
report and EPA ethics review memo. There was one deviation regarding handwashing that
occurred across all MEs and involved the amount of wash fluid (rather than using a pre-
measured amount for the first rinse and the remaining wash fluid for the second rinse, the same
total volume was used as described in the protocol but the first rinse did not use a pre-determined
amount) and having subjects, rather than the researchers, use the gauze provided to wipe dried
paint from the subjects' hands. This is a minor deviation that did not affect study integrity and
was more efficient and effective.
l

-------
The respirator fit test was conducted as outlined in the protocol, study report and EPA ethics
review memo, There was one minor deviation when a subject went through the fit test before
receiving approved medical recommendations. All elements of the medical clearance and fit test
were completed before the subject's monitoring event. This deviation did not affect the subject's
welfare or safety and was appropriately considered minor.
One deviation occurred when a subject removed her safety glasses when pouring paint and had
paint splashed in her eye. She was treated per protocol, recovered, and subsequently was allowed
to complete her monitoring event. EPA recommended to AEATF to consider revising the pre-
monitoring discussion to note where and when safety equipment must be worn, and the HSKB
agrees with that recommendation.
Subjects were monitored for heat stress, informed of signs of heat stress and told to inform study
staff if they experienced signs or symptoms of heat stress, reminded to take breaks as needed,
and provided with cold water and sports drinks while in the study. Hearing protection was
offered, but none of the subjects opted to use hearing protection.
One subject was found, to have dermal irritation on the forearms at the post-monitoring skin
check. This was attributed by the subject to sweat and they were not concerned. The area was
washed, and the irritation was gone later the same day.
Confidentiality was maintained throughout the study. Subjects, including alternates, were
compensated per the protocol. The amount of compensation for each subject was consistent; with
the time spend on the various study-related activities and was not excessive.
There were five protocol amendments which were handled consistent with the IRB practices and
the protocol There were seven reported protocol deviations, three of which have been previously
described. Four other deviations were reported, none of which affected the study's scientific
integrity or placed the subjects at risk.
1

-------

-------