VCT7\	EPA/635/R-22/248

IRIS Assessment Plan
www.epa.gov/iris

IRIS Assessment Plan and Protocol for Assessing Cancer Risk from
Inhalation Exposure to Cobalt and Cobalt
Compounds

(Scoping and Problem Formulation Materials)

November 2022

Integrated Risk Information System
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

DISCLAIMER

This document is a public comment draft for review purposes only. This information is
distributed solely for the purpose of public comment. It has not been formally disseminated by EPA.
It does not represent and should not be construed to represent any Agency determination or policy.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.

This document is a draft for review purposes only and does not constitute Agency policy.

ii	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

CONTENTS

AUTHORS | CONTRIBUTORS | REVIEWERS	vii

1.	INTRODUCTION	1

2.	SCOPING AND PROBLEM FORMULATION	2

2.1.	BACKGROUND	2

2.2.SCOPING SUMMARY	2

2.3.	PROBLEM FORMULATION	5

2.4.	KEY SCIENCE ISSUES	8

3.	OVERALL OBJECTIVE AND SPECIFIC AIMS	10

3.1.OVERALL OBJECTIVE	10

3.2.	SPECIFIC AIMS	10

4.	ASSESSMENT PROTOCOL	12

4.1.	ORGANIZATIONAL APPROACH FOR SUPPLEMENTAL MATERIAL	12

4.1.1.	Organization of Mechanistic Information	14

4.1.2.	Organization of ADME and PK/PBPK Model Information	15

4.2.	METHODS FOR DOSE-RESPONSE ASSESSMENT	16

4.2.1.	Selecting Endpoints for Dose-Response Assessment	16

4.2.1.1. Data Extraction and Dose Standardization	17

4.2.2.	Conducting Dose-Response Assessments	18

4.2.2.1.	Dose-Response Analysis in the Range of Observation	19

4.2.2.2.	Extrapolation: Unit Risk	20

4.2.2.3.	Extrapolation: Reference Concentrations	21

REFERENCES	R-22

APPENDIX A. CHEMICAL AND PHYSICAL PROPERTIES OF INCLUDED FORMS	A-l

A.l. KEY COMPOUNDS IDENTIFIED DURING SCOPING	A-l

A.2.	ADDITIONAL COBALT COMPOUNDS USED TO SUPPORT DERIVATION OF INHALATION

UNIT RISK ESTIMATES	A-ll

APPENDIX B. SURVEY OF EXISTING TOXICITY VALUES	B-l

B.l.	METHODS	B-l

B.2. SUMMARY OF EXISTING TOXICITY VALUES	B-5

APPENDIX C. SYSTEMATIC EVIDENCE MAP	C-l

This document is a draft for review purposes only and does not constitute Agency policy.

iii	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

C.l. SYSTEMATIC EVIDENCE MAP (SEM) SPECIFIC AIMS	C-l

C.2. POPULATIONS, EXPOSURES, COMPARATORS, AND OUTCOMES (PECO) CRITERIA AND

SUPPLEMENTAL MATERIAL TAGGING	C-2

C.3. METHODS: LITERATURE SEARCH STRATEGIES	C-8

C.3.1. Database Search Term Development	C-8

C.3.2. Database Searches	C-8

C.3.3. Searching Other Resources	C-9

C.3.4. Non-Peer-Reviewed Data	C-10

C.4. METHODS: LITERATURE SCREENING PROCESSES	C-ll

C.4.1. Title/Abstract and Full Text Screening	C-ll

C.4.2. Supplemental Material Tagging	C-12

C.4.3. Multiple Publications of the Same Data	C-12

C.4.4. Literature Flow Diagrams	C-13

C.5. METHODS: LITERATURE INVENTORY PREPARATION	C-13

C.5.1. Studies That Meet SEM PECO Criteria	C-13

C.5.2. Supplemental Material	C-18

C.6. RESULTS: LITERATURE SCREENING RESULTS	C-18

C.7. LITERATURE INVENTORY	C-22

C.7.1. Characterizing Epidemiological Studies for Dose-Response

Analysis	C-22

C.7.2. Characterizing Animal Studies for Dose-Response Analysis	C-22

ADDENDUM 1. LITERATURE SEARCH STRATEGY (DATE LIMITED TO 2019- 2021)	Addendum 1-1

ADDENDUM 2. PROCESS AND RESULTS FOR SEARCHING AND COLLECTING EVIDENCE FROM

OTHER RESOURCES	Addendum 2-1

This document is a draft for review purposes only and does not constitute Agency policy.

iv	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

TABLES

Table 2-1. EPA Program Office Interest in a Cancer Assessment of Cobalt Compounds	3

Table 2-2. Summary of Existing Cancer Hazard Conclusions for cobalt by the inhalation route	6

Table A-l. Chemical identity and physicochemical properties of cobalt	A-l

Table A-2. Chemical identity and physicochemical properties of cobalt oxide	A-2

Table A-3. Chemical identity and physicochemical properties of hexanoic acid, 2-ethyl-,

cobalt(2+) salt	A-2

Table A-4. Chemical identity and physicochemical properties of cobalt nitrate	A-3

Table A-5. Chemical identity and physicochemical properties of cobalt nitrate hexahydrate	A-4

Table A-6. Chemical identity and physicochemical properties of cobalt bromide	A-5

Table A-7. Chemical identity and physicochemical properties of cobalt carbonate	A-6

Table A-8. Chemical identity and physicochemical properties of cobalt chloride	A-7

Table A-9. Chemical identity and physicochemical properties of cobalt hydrocarbonyl	A-7

Table A-10. Chemical identity and physicochemical properties of cobalt oxide (II, III)	A-9

Table A-ll. Chemical identity and physicochemical properties of cobalt carbonyl	A-9

Table A-12. Chemical identity and physicochemical properties of cobalt sulfate	A-ll

Table B-l. Sources searched for human health reference values for cobalt and cobalt forms	B-l

Table B-2. Details on the derivation of existing inhalation cancer toxicity values for cobalt and

cobalt compounds	B-7

Table C-l. Example Populations, Exposures, Comparators, and Outcomes (PECO) Criteria	C-3

Table C-2. Categories of Potentially Relevant Supplemental Material	C-4

Table C-3. Preferred design features of animal dose-response studies of inhalation exposures to

cobalt compounds	C-14

Table C-4. Analysis of Human Studies Meeting PECO Criteria for Suitability for Dose-Response	C-24

Table C-5. Analysis of Animal Studies Meeting PECO Criteria for Suitability for Dose-Response	C-26

FIGURES

Figure 4-1. Studies identified as supplemental during literature screening. Click here to view

interactive version	13

Figure 4-2. Visual summary of overall tagging structure for mechanistic studies related to

carcinogenesis	15

Figure 4-3. Visual summary of tagging structure for ADME and PK/PBPK studies	16

Figure B-l. Available noncancer and cancer toxicity values for inhalation exposure to cobalt	B-6

Figure C-l. Overview of Integrated Risk Information System (IRIS) study evaluation process.

(a) individual evaluation domains organized by evidence type, and (b) individual
evaluation domains judgments and definitions for overall ratings (i.e., domain

and overall judgments are performed on an outcome-specific basis)	C-17

Figure C-2. Study Flow Diagram	C-20

Figure C-3. Literature tree. Click here for interactive version	C-21

This document is a draft for review purposes only and does not constitute Agency policy.

v	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

ABBREVIATIONS

ADAF	age-dependent adjustment factors

ADME	absorption, distribution, metabolism, and excretion

ATSDR	Agency for Toxic Substances and Disease Registry

BMR	benchmark response

BMDS	Benchmark Dose Software

CAA	Clean Air Act

CalEPA	California Environmental Protection Agency

CASRN	Chemical Abstract Services Registry Number

CPAD	Chemical and Pollutant Assessment Division

CPHEA	Center for Public Health and Environmental Assessment

EPA	Environmental Protection Agency

HAWC	Health Assessment Workspace Collaborative

HERO	Health and Environmental Research Online

IAP	IRIS Assessment Plan

IARC	International Agency for Research on Cancer

ICD - 8	International Classification of Diseases (8th revision)

IRIS	Integrated Risk Information System

IUR	inhalation unit risk

NTP	National Toxicology Programs

MOA	Mode of Action

OAR	Office of Air and Radiation (EPA)

ORD	Office of Research and Development (EPA)

PAC	protective action criteria

PBPK	physiologically based pharmacokinetic

PECO	populations, exposures, comparators, and outcomes

POD	point of departure

RIS	Research Information Systems

RfC	reference concentration

ROS	reactive oxygen species

SEM	systematic evidence map

SWIFT	Sciome Workbench for Interactive Computer-facilitated Text-mining

TIAB	title and abstract

TCEQ	Texas Commission on Environmental Quality

URE	unit risk estimates

URF	unit risk factor

This document is a draft for review purposes only and does not constitute Agency policy.

vi	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Chemical Manager

Pamela Noves. Ph.D.

U.S. EPA/ORD/CPHEA/CPAD

Assessment Team

Christine Cai, M.S.
Kristan Markev. B.A.
Amanda Persad, Ph.D.
Alan F. Sasso. Ph.D.
Brittany Schulz. B.S.
John Stanek, Ph.D.
Kristina Thayer. Ph.D.
Paul White, Ph.D.
lay Zhao. Ph.D.

Antony Williams, Ph.D.

U.S. EPA/ORD/CPHEA/CPAD

U.S. EPA/ORD/CCTE/CCED

Technical Contributors

lohn Bucher. Ph.D.
Diane Spencer, M.S.

NIH/NIEHS/DNTP [retired]
NIH/NIEHS/DNTP

Production and Project Management Team

Maureen Johnson
Ryan Jones
Dahnish Shams
Avanti Shirke
Jessica Soto-Hernandez
Vicki Soto
Samuel Thacker
Garland Waleko

U.S. EPA/ORD/CPHEA

Executive Direction

Wayne Cascio, M.D. (CPHEA Director)

Kay Holt, M.S. (CPHEA Deputy Director)

Samantha Jones, Ph.D. (CPHEA Associate Director)
Kristina Thayer, Ph.D. (CPAD Director)

Andrew Kraft, Ph.D. (CPAD Associate Director)
Paul White, Ph.D. (CPAD Senior Science Advisor)
Elizabeth Radke, Ph.D. (Branch Chief)

Andrew Hotchkiss, Ph.D. (Branch Chief)

Janice Lee, Ph.D. (Branch Chief)

Ravi Subramaniam, Ph.D. (Branch Chief)

Glenn Rice, Ph.D. (Branch Chief)

Viktor Morozov, Ph.D. (Branch Chief)

U.S. EPA/ORD/CPHEA

This document is a draft for review purposes only and does not constitute Agency policy.

vii	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

1. INTRODUCTION

IRIS assessments provide high quality, publicly available information on the toxicity of
environmental chemicals and pollutants to which the public might be exposed. These assessments
provide an important source of toxicity information used by the Environmental Protection Agency
(EPA), state and local health agencies, other federal agencies, tribes, and international health
organizations. Specifically, IRIS assessments provide rigorous scientific evaluations addressing the
first two steps of the 4-step risk assessment process, hazard identification and dose-response
analysis.

As part of the initial steps in assessment development, the IRIS Program undertakes scoping
and initial problem formulation activities. During scoping activities, the IRIS Program consults with
EPA programs and regional offices to identify the nature of the hazard characterization needs, the
most important exposure pathways, and the timeframe to inform Agency decisions. A broad,
preliminary literature survey (referred to as a systematic evidence map, or SEM) may also be
conducted to assist in identifying the extent of the evidence and health effects that have been
studied for the chemical of interest Based on the SEM and the scope defined by EPA, the IRIS
Program undertakes problem formulation activities to frame the scientific questions that will be the
focus of the assessment A summary of the IRIS Program's scoping and problem formulation
conclusions are contained in the IRIS Assessment Plan (IAP). Based on the IAP, an IRIS Protocol is
developed to describe the methods that will be used to address the defined scope and identified
problem formulation considerations during IRIS assessment development.

This document presents the draft IAP for the "IRIS Assessment of Cancer risk from
Inhalation Exposure to Cobalt and Compounds." The IRIS Protocol outlining the methods for
conducting the assessment is also included because the results of problem formulation indicated
the proposed analysis will be targeted, focusing on dose-response analyses of studies identified
from the SEM as being most suitable for deriving cancer inhalation toxicity values.

This document is a draft for review purposes only and does not constitute Agency policy.

1	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

2.SCOPING AND PROBLEM FORMULATION

2.1.	BACKGROUND

Section 2.1 provides a summary of background information for contextual purposes only.
This brief overview emphasizes reviews and other summary information that, unless otherwise
specifically noted, are derived from NTP (20161. TCEO (20171. OEHHA (20201. and Slack et al.
(20171: it is not intended to be a comprehensive description of the available information.

Cobalt is a metallic element that is naturally occurring as several different substances and
oxidation states, often in association with nickel, silver, lead, copper, and iron ores. Cobalt
compounds are used in a variety of industrial applications, including as catalysts, in feed
supplements, in batteries, as colorants for glass, ceramics, and paint, and as driers for inks and
paints. Cobalt is also used in alloys or composites, such as cobalt-tungsten carbide, and in cobalt-
containing prosthetics. Nanomaterials containing cobalt are used in medical tests and treatments as
well as in the textile and electronics industries. Cobalt also forms part of the structure of vitamin
B12, which plays essential roles in red blood cell formation, cell metabolism, nerve function and
DNA synthesis Osman etal. (20211: Mayo Clinic (20211.

Elemental cobalt (limited natural occurrence, generally produced during smelting) is a hard,
silvery grey metal. Cobalt reacts with other elements, such as oxygen (cobalt oxide), sulfur (cobalt
sulfate), and arsenic (cobalt arsenide). Cobalt compounds represent a large group of substances.
For example, EPA's Substance Registry Services - the central system for information about
substances that are tracked or regulated by EPA or other sources - contains over 400 cobalt
containing compounds U.S. EPA (2014b). These compounds can be organometallic or inorganic as
well as water-soluble or -insoluble. The most common oxidation states of cobalt (Co) are +2 and
+3; for most simple cobalt compounds, the valence is +2, designated as Co (II), while Co (III)
substances are generally strong oxidizers. There is only one stable isotope of cobalt, 59Co, and there
are about 26 known radioactive isotopes of cobalt, of which only two are of commercial
importance, 60Co and 57Co.

2.2.	SCOPING SUMMARY

EPA's Office of Air and Radiation (OAR) nominated a cancer assessment of water-soluble
and water-insoluble cobalt compounds to the IRIS Program. The nomination focused on inhalation
exposure and those forms most pertinent to implementing the Clean Air Act (CAA) by informing
decisions related to potential carcinogenic risks due to emissions to air of cobalt compounds during
industrial processes (Appendix A.1). This assessment activity was added to the IRIS Program
Outlook in June 2022.

This document is a draft for review purposes only and does not constitute Agency policy.

2	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Cobalt compounds represent a very large and diverse set of substances. Some uses of cobalt
compounds may result in their emissions to air. Cobalt compounds most pertinent to OAR's
implementation of the Clean Air Act (CAA) are primarily water-soluble forms [such as cobalt
aluminate, cobalt bromide, cobalt carbonate, cobalt chloride, cobalt hydrocarbonyl, cobalt naphtha,
cobalt nitrate, cobalt oxide (II, III), and hexanoic acid, 2-ethyl-, cobalt (2+) salt] and some water-
insoluble forms [such as cobalt metal and cobalt carbonyl]. These compounds were identified based
on currently available emission data U.S. EPA f2020al. Only a few cobalt compounds identified to
date have cancer toxicity data. Thus, those cobalt compounds that do have evidence for cancer due
to inhalation exposure (e.g., hydrated cobalt sulfates) are being evaluated within the scope of this
review for potential use as surrogates for other water-soluble and water-insoluble forms of cobalt.
See Appendix A for a summary of chemical and physical properties, obtained largely from the EPA
CompTox Chemicals Dashboard and PubChem, for the key compounds identified during scoping1. If
supported by the available data, EPA may develop separate cancer values for water-soluble and
water-insoluble cobalt compounds. Note that the chemicals included in Appendix A do not
represent an exhaustive list of water-soluble and water-insoluble cobalt compounds of interest to
OAR that will be addressed in the dose-response assessment. However, certain cobalt containing
substances are considered out of scope for this assessment: nanomaterials containing cobalt,
radioactive isotopes (i.e., 60Co), and vitamin B12, because their chemical and physical properties are
quite different from the forms identified during scoping as most pertinent to the CAA, and hence,
their toxicological characteristics are also expected to be different. Forms pertinent to the CAA are
those that are detected and reported in air quality monitoring.

Table 2-1. EPA Program Office Interest in a Cancer Assessment of Cobalt
Compounds.

EPA
program
or regional
office

Oral

Inhalation

Statutes/regulations

Anticipated uses/interest

OAR



V

Clean Air Act (CAA)

Cobalt compounds are listed as a hazardous air
pollutant (HAP) under section 112 (b) (42 U.S.C.§
7412) of the CAA. CAA Section 112 has a number of
regulatory requirements, including the requirement

1 The physicochemical properties in the summary tables are based on information from a variety of sources,
primarily from the EPA CompTox Chemicals Dashboard and PubChem. The data obtained from the EPA
CompTox Chemicals Dashboard are of varying quality and include both experimental and predicted data. The
data associated with the chemical substances in the CompTox Chemicals Dashboard database have been
compiled from public sources, databases and peer-reviewed literature and have varying levels of reliability
and accuracy. Predicted data in particular have significant limitations in terms of the predictions of properties
for salts, inorganic and organometallic substances. Links to many external resources are provided. Expansion,
curation, and validation of the content are ongoing. The tables presented in the Appendix were reviewed by
chemists for obvious errors and the most appropriate values available were selected.

This document is a draft for review purposes only and does not constitute Agency policy.

3	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

EPA
program
or regional
office

Oral

Inhalation

Statutes/regulations

Anticipated uses/interest









that EPA promulgate emission standards for sources
emitting HAP. Eight years after promulgation of
emission standards, EPA must perform risk and
technology reviews of emission standards that
require maximum achievable control technology
(MACT). Cobalt toxicological information developed
for this cancer assessment may be used to inform
CAA section 112 regulatory decisions.

The toxicological information may also be used for
non-regulatory purposes, such as the annual
national screening-level assessments of air toxics
(i.e., AirToxScreen).

Some cobalt containing substances are considered
out of scope for this assessment, because their
chemical and physical properties are quite different
from the cobalt forms identified as most pertinent
to the CAA during scoping, and hence, their
toxicological characteristics are also expected to be
different: nanomaterials containing cobalt,
radioactive isotopes (i.e., 60Co), and vitamin B12.

This document is a draft for review purposes only and does not constitute Agency policy.

4	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

2.3. PROBLEM FORMULATION

Multiple health agencies, including the U.S. EPA, National Toxicology Program (NTP),
International Agency for Research on Cancer (IARC), California EPA, and Texas Commission on
Environmental Quality (TCEQ), have concluded that cobalt and certain cobalt compounds are likely
to cause cancer (Table 2-2). The IRIS database does not contain a cancer classification, oral slope
factor or inhalation unit risk for cobalt. An EPA Provisional Peer Reviewed Toxicity (PPRTV)
assessment published in 2008 concluded under EPA's Guidelines for Carcinogen Risk Assessment
U.S. EPA f2005al that water-soluble cobalt compounds are "likely to be carcinogenic to humans by
the inhalation route" U.S. EPA (20081. This was based on limited evidence of carcinogenicity in
humans and sufficient evidence of carcinogenicity in animals (rats and mice) treated with a water-
soluble form of cobalt (cobalt sulfate heptahydrate2, often referenced as "cobalt sulfate" in existing
assessments) in a 2-year inhalation cancer bioassay Bucher et al. (19991; N1	8).3 Since

publication of the PPRTV assessment, the NTP has released a 2-year inhalation cancer bioassay of
cobalt metal NTP (20141. which was used by the CalEPA to develop an inhalation unit risk estimate
for cobalt metal and water-insoluble cobalt compounds (Table 2-2). CalEPA also developed an
inhalation unit risk estimate for water-soluble cobalt compounds based on the 1998 NTP cancer
bioassay study of cobalt sulfate NTP (19981. Both the 1998 and 2014 NTP cancer bioassays were
used by TCEQ to develop an inhalation unit risk estimate for cobalt and cobalt compounds. The unit
risk factor derived by TCEQ was based on the midpoint of the unit risk factors of cobalt sulfate and
metal. Additional details on derivation of the various unit risk factors are presented in Appendix B
along with a summary of non-cancer reference values for cobalt and cobalt compounds.

2	CoS04-7H20, CAS No. 10026-24-1, molecular weight 281.13. It should be noted that the experimental
conditions dehydrated the compound to cobalt sulfate hexahydrate (CoS04-6H20, molecular weight 263.09),
and that this was the chemical rodents were exposed to NTP f19981. See Section 4.2.1 for additional
information on dose standardization for NTP (19931.

3	NTP analyses Bucher et al. (19991 described statistically significant increased incidence of
alveolar/bronchiolar tumors in both sexes of rats and mice, pheochromocytomas in female rats, and
hemangiosarcomas in male mice. NTP (20141 concluded that there was clear evidence of carcinogenic activity
of cobalt metal in male rats (lung, adrenal medulla, pancreas), female rats (lung, adrenal medulla,
mononuclear cell leukemia), and male and female mice (lung); NTP (19981 concluded that there was clear
evidence of carcinogenic activity of cobalt sulfate heptahydrate in female rats (lung, adrenal medulla) and
male and female mice (lung), and some evidence of carcinogenic activity in male rats (lung).

This document is a draft for review purposes only and does not constitute Agency policy.

5	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table 2-2. Summary of Existing Cancer Hazard Conclusions for cobalt by the
inhalation route

Agency/Organization (year)

Cancer
Characterization3

Cobalt substance(s)

Inhalation unit riskb and
study on which its based

Provisional Peer-Reviewed
Toxicity Value (PPRTV)

U.S. EPA (2008)

Likely to Be
Carcinogenic to
Humans

Soluble cobalt sulfate

9.0 (mg/m3)"1
NTP (1998)
Lung tumors

California Office of
Environmental Health Hazard
Assessment

OEHHA (2019); OEHHA (2020)

Listed as under
Proposition 65 as
causing cancer

Cobalt metal and water-
insoluble compounds

8.0 (mg/m3)"1 NTP (2014)
Multisite tumor analysis

Water-soluble cobalt
compounds

0.86 (mg/m 5)1
NTP (1998)

Multisite tumor analysis

Texas Commission on
Environmental Quality

TCEQ (2017)

Likely to Be
Carcinogenic to
Humans

Cobalt and compounds

6.0 (mg/m3)"1

NTP (1998); NTP (2014)c

Lung tumors

International Agency for
Research on Cancer

Karagas et al. (2022)

Group 2A: Probably
Carcinogenic to
Humans

Cobalt metal

Cobalt sulfate and other
soluble Co(ll) salts

NA

European Chemicals Agency
(ECHA) Committee for Risk
Assessment (RAC)

ECHA (2017)

Category IB for
Carcinogenicity:
Presumed to Cause
Cancer to Humans

Cobalt and compounds

NA

National Toxicology Program

NTP (2016)

Reasonably
Anticipated to Be
Human Carcinogens

Cobalt and cobalt compounds
that release cobalt ions in
vivo

NA

American Conference of
Governmental Industrial
Hygienists

ACGIH (2001a)

Group A3: Confirmed
Animal Carcinogen
with Unknown
Relevance to Humans

Inorganic cobalt

NA

a Cancer hazard conclusions expressed using the phrasing of the specific agency or organization that conducted
the assessment and reflects terminology used at the time of the published report.

b All values normalized to cobalt content (see Section 4.2.1). It should be noted that some agencies may have
used the molecular weight of cobalt sulfate hexahydrate to convert from chemical concentrations listed in NTP
(1998) to mg/m3 elemental cobalt. This is because analysis of the chamber samples indicated that exposures
were to cobalt sulfate hexahydrate, and that the parent compound (cobalt sulfate heptahydrate) dehydrated.
However, based on a review of the assessment analytical details in the NTP report and Behl et al. (2015), it was
determined that the chemical concentrations listed in NTP (1998) were units of mg/m3 anhydrous cobalt sulfate
Bucher et a 1 . (In Press). As a result, dose-response modeling results for soluble cobalt based on data from NTP
(1998) may contain a bias due to an error in units conversion. Assuming a lower percentage of elemental cobalt
in the exposure compound would result in an overestimation of toxicity.

CTCEQ derived unit risk factors of 9.1 (mg/m3)"1 (based on NTP (1998)) and 3.0 (mg/m3)"1 (based on NTP (2014)).
The final unit risk factor was the midpoint of these two values.

This document is a draft for review purposes only and does not constitute Agency policy.

6	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

The focus of the present task is to carry out a cancer dose-response assessment and develop
values for inclusion in the IRIS database. EPA anticipates this cancer dose-response assessment will
derive an inhalation unit risk (IUR) based on previous work indicated in this document In addition,
analyses will be undertaken to evaluate support for a non-linear MOA, and, if deemed necessary, a
nonlinear approach for the dose-response will be presented. Currently, EPA does not anticipate
deriving any noncancer inhalation values. This assessment will adopt the PPRTV cancer hazard
conclusion that under EPA's Guidelines for Carcinogen Risk Assessment U.S. EPA (2005al. cobalt is
"likely to be carcinogenic to humans by the inhalation route," a conclusion consistent with other
authoritative bodies (Table 2). EPA's PPRTV concluded soluble cobalt is likely to be carcinogenic to
humans by the inhalation route. Subsequently peer reviewed assessments from other authoritative
bodies have reached this conclusion for both soluble and insoluble forms. Accordingly, this
assessment will not undertake a hazard assessment but will apply this designation to all the cobalt
forms identified within its scope. As shown in Table 2, the NTP 1998 and 2014 cancer bioassays
If	M3); NTP (20141 have consistently been considered most suitable for developing inhalation

unit risk estimates in prior assessments. A systematic evidence map (SEM, see Appendix C) was
developed to determine whether any more recent studies have been published that could plausibly
be used for dose-response. No human epidemiology or experimental animal studies were identified
that were considered at least as suitable as the NTP bioassays. Thus, for dose-response analysis, the
IRIS assessment will focus on the 1998 and 2014 NTP studies as representative of water-soluble
and water-insoluble compounds of cobalt, similar to the approach taken by CalEPA. Methods for
adjusting observed inhaled particulate exposure effect levels for interspecies dosimetric differences
will be performed according to EPA's Methods for Derivation of Inhalation Reference Concentrations
and Application of Inhalation Dosimetry	4), and implementation further refined using

EPA's MPPD Technical Support Documentation and User's Guide U.S. EPA (20221.

This document is a draft for review purposes oniy and does not constitute Agency poiicy.

7	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

2.4. KEY SCIENCE ISSUES

Based on the preliminary literature survey and review of past assessments on inhalation
exposure to cobalt, the following key scientific issues related to the mechanistic evidence for cobalt
were identified. Evaluation of these key science issues may inform facets of the dose-response
assessment, potential dependencies between different tumor types, and application of age-
dependent adjustment factors (ADAF) as appropriate in accordance with EPA's Supplemental
Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens flJ.S. EPA (2005bl.
The evaluation of these cobalt-related science issues will be informed by conclusions from prior
assessments U.S. EPA (20081: 0	); OEHHA (20201: TCEO (20171 and supplemented by

evidence obtained from new mechanistic studies identified since these assessments were
completed.

•	Tumor dependencies: Carcinogenicity studies in rats and mice conducted by NTP show
significant dose-related increases in incidences of lung tumors (alveolar/bronchiolar
adenomas and carcinomas) following 2-year inhalation exposures to insoluble cobalt metal
NTP (20141: Behl et al. (20151 and soluble cobalt sulfate heptahydrate NTP (19981: Bucher
et al. (19991. NTP (20141 and NTP (19981 also report tumor formation at sites distal to the
lung. Specifically, both cobalt compounds also caused treatment-related increases in
neoplasms of the adrenal gland in female rats. In male rats, adrenal gland tumors were also
reported with cobalt metal, but findings were equivocal in males exposed to cobalt sulfate
heptahydrate. Inhalation exposures to cobalt metal at a higher dose range also elicited a
greater spectrum of systemic tumors in rats than did cobalt sulfate heptahydrate, including
mononuclear leukemia, and tumors of the pancreas and kidneys. Assessment of
dependence or independence of the different tumor types will help to determine whether a
combined tumor analysis can be performed (i.e., combined tumor analysis is considered
invalid if it is judged that the tumors do not form independently U.S. EPA (2020b).

•	Cellular uptake and tissue disposition: The kinetics and tissue disposition of inhaled cobalt
may be affected by the specific cobalt-containing chemical compound and associated
physical-chemical properties, including solubility and particle size. Insoluble cobalt metal
and soluble cobalt sulfate heptahydrate are shown in vivo and in vitro to elicit similar
respiratory and inflammatory responses but may exhibit differing pharmacokinetics and
pharmacodynamics that can influence cobalt dosimetry and biological activity NTP (20141:
NTP (19981. Although cobalt bioavailability and its influence on carcinogenicity are not fully
understood, it is known that cellular uptake of free cobalt ion and particles occur by
differing processes U.S. EPAf20081: Lison et al. (20181: NTP (20161: NTP (20211: OEHHA
(20201. Water insoluble cobalt compounds could be absorbed into the cell via endocytosis
processes where they are solubilized in lysosomes and released in ionic form inside the cell.
As a result, some compounds that are poorly soluble in water (such as cobalt metal, cobalt
(II) oxide, and cobalt (III) oxide) have higher solubilities in serum and biological media
MAK-Commission (20071. In addition, even for sparingly soluble compounds that are
commonly termed insoluble, solubility limits may be higher than relevant biological levels
of cobalt In which case information about the rate of solubilization could inform
assessment of toxicity. Solubilized cobalt compounds release ions outside the cell after
which they are taken up into the cell through membrane-bound ion channels. Thus, many

This document is a draft for review purposes only and does not constitute Agency policy.

8	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

water-soluble and sparingly water-soluble cobalt compounds can result in the cellular
uptake or release of cobalt ions in vivo. The differences between uptake and intracellular
release rates of water-soluble and water-insoluble cobalt compounds could lead to distinct
target sites, as well as variations in systemic and intracellular concentrations. Therefore,
mechanistic information regarding cellular uptake and tissue deposition will be updated
and may inform selection and application of dosimetric adjustments or modeling
approaches Behl etal. (20151: Colognato etal. f20081: Ponti et al. f20091: Smith et al.
f 2 0141.

•	Cobalt particle toxicity: The release of cobalt ion intracellularly in lysosomes by water-
insoluble cobalt compounds is suggested to be largely responsible for mediating their
biological activity IARC f'20061: NT 1); NTP (20141: NTP f'20161: NTP f20211.
However, in addition to potential differences in particle ion uptake and distribution that
might influence tissue dosimetry, cobalt is a redox-active transition metal, and as such, it
has been suggested that cobalt particles may have a greater effect than ions in catalyzing
production of reactive oxygen species (ROS) IARC (20061: NTP (20161. Updating the
mechanistic evidence concerning whether cobalt particles may elicit direct toxicity
contributing to carcinogenesis will help inform the choice of the particle lung dose metric
used for rodent-to-human extrapolation and dose-response.

•	Proposed MOA of cobalt carcinogenicity: While not fully understood, there is evidence that
cobalt-induced neoplastic development likely involves pathways of genotoxicity, oxidative
stress (and generation/scavenging of ROS), and stabilization of hypoxia-inducible factor la
(HIF-la) IJ.S. EPA (20081: IARC (20061: NTP (20161: NTP (20211: Ton etal. (20211.
Evidence with differing water-insoluble and water-soluble cobalt compounds suggests
cobalt genotoxicity involves primarily clastogenic effects, as well as direct and indirect DNA
damage and inhibition of DNA repair U.S. EPA (20081: IARC (20061: NTP (20161. Previous
assessments have found the evidence generally inconsistent on whether inhaled cobalt
carcinogenicity involves a mutagenic MOA, and do not apply age-dependent adjustment
factors (ADAF) in unit risk estimates U.S. EPA (20081: OEHHA (20201: TCEO (20171.
Updating the current evidence in the proposed cobalt cancer MOA, including capturing any
new evidence of mechanistic responses beyond those previously described, will help inform
the dose-response analyses, pharmacokinetic evaluations, and animal-to-human
extrapolation methodologies U.S. EPA (2020c).

•	Cobalt compounds are a large and diverse group of substances. To the extent possible, the
assessment will try to describe the types of cobalt compounds for which use of this IRIS
assessment would not be appropriate. Substances that can release cobalt ions in vivo4, both
water soluble and insoluble, likely define the domain of applicability. Substances where
cobalt atoms are tightly bound and notbioavailable, such as Vitamin B12, are unlikely to
present the same carcinogenicity hazards.

4 Release of cobalt ions can involve the solubilization of Co(II) ions or, for metallic materials, reflect both
surface corrosion and release of Co (II] ions.

This document is a draft for review purposes only and does not constitute Agency policy.

9	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

3.0VERALL OBJECTIVE AND SPECIFIC AIMS

3.1. OVERALL OBJECTIVE

1	The overall objective is to carry out a dose-response assessment for water soluble and

2	water insoluble compounds of cobalt. The evaluation conducted in this assessment will use

3	relevant EPA guidelines.5

3.2. SPECIFIC AIMS

4	• Utilize the SEM (presented in Appendix C) to identify studies most suitable for the dose-

5	response modeling for water-soluble and water-insoluble compounds of cobalt.

6	° Based on the SEM and assessments conducted by others, the NTP inhalation cancer

7	bioassay studies for cobalt sulfate and cobalt metal NTP T19981: NTP T20141 were

8	considered most appropriate for dose-response analysis.

9	° As supported by the available data, EPA will consider developing separate estimates

10	for water-soluble and water-insoluble cobalt compounds.

11	• As supported by the available data, mechanistic information obtained from new studies (see

12	section 4.1.1) and prior assessments will be evaluated to inform existing conclusions on the

13	MOA and whether there are any new MOA considerations for dose-response analysis U.S.

14	EPA f2005a1: U.S. EPA f2005b1: Karagas etal. (20221

15	° MOA considerations will inform methods for deriving inhalation unit risk values for

16	water soluble and water insoluble compounds of cobalt, statistical analyses of dose-

17	response data, common dependencies between different tumor types, and application of

18	ADAF. See Section 2.4 for details.

19	• As supported by the available data, endpoints will be modeled using EPA's Benchmark Dose

20	Software6 and associated statistical dose-response methods (e.g., time-to-tumor modeling).

21	° MOA considerations will inform methods for combining multiple tumor types.

22	° Statistical considerations will inform which dose-response methods can be used for

23	each tumor type.

5The EPA guidelines have been developed over time and address the state of the science at the time they were
developed. Thus, evaluation methods may be updated as new science emerges, or when existing guidelines
are updated. EPA guideline documents can be found at: http://www.epa.gov/iris/basic-information-about-
integrated-risk-information-system#guidance/

6 Information on model fitting, model selection, and reporting of decisions and results are outlined in the
Benchmark Dose Technical Guidance Document U.S. EPA (2012bl

This document is a draft for review purposes only and does not constitute Agency policy.

10	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

1	• Points-of-departure derived from animal data will be converted to human equivalent

2	concentrations for derivation of the IUR(s).

3	° MOA and pharmacokinetic considerations will inform choice of internal dose metrics,

4	and methods for performing animal-to-human extrapolations7.

7 Methods for lung dosimetry are described in Methods for Derivation of Inhalation Reference Concentrations
and Application of Inhalation Dosimetry U.S. EPA t'19941. and in EPA's MPPD Technical Support
Documentation and User's Guide U.S. EPA (2022}.

This document is a draft for review purposes only and does not constitute Agency policy.

11	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

4.ASSESSMENT PROTOCOL

4.1. ORGANIZATIONAL APPROACH FOR SUPPLEMENTAL MATERIAL

Studies tagged as supplemental material during preparation of the SEM were grouped by
the specific category of supplemental material content (e.g., mechanistic, ADME, etc.) (Table C-2,
Figure 4-1). Additional more granular sub-tagging is undertaken in Health Assessment Workspace
Collaborative (HAWC), a web-based data content management system for human health
assessments, during draft assessment development to help address the key science issues and
inform dose-response. Full-text retrieval is reserved for studies that most directly address the key
science issues. The degree of sub-tagging depends on the extent of content for a given type of
supplemental material and needs of the assessment with respect to deriving the IUR(s). For the
cobalt assessment, more granular tagging will be conducted for supplemental content classified as
mechanistic, ADME, PK/PBPK models, and susceptibility.

Supplemental material studies identified from other assessments U.S. EPA (2008): OEHHA
("20191: OEHHA C20201: TCEO (20171: NTP (20161: ATSDR (20041 were also tagged. Tagging
judgments in HAWC are made by one assessment member and confirmed during preparation of
draft assessment by another member of the assessment team. The same study could have multiple
tags. The overall approach for supplemental material content is presented in Figure 4-1, with
details on subsequent sub-tagging presented in the following sections under the specific type of
supplemental content (i.e., mechanistic, ADME and PK/PBPK).

This document is a draft for review purposes only and does not constitute Agency policy.

12	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Supplemental

187

Mechanistifjpformalion

65

Pharmacokinetic (AOME|

0

PK/PBPK Models

\£)

Medical Therapeutic/Memetics

Q

Medical Implantation

al Implai

©

Used to ImUra Hypoxia
115

(115)

Hon-PECO Routes of Exposure

©

Non-Mammalian Models
Susceptible Populations

©

Exposure Characteristics

©

Mixture Studies

S. ©

Case Reports

©

Acute or Short-term Exposures

©

Non-cancer Keajth Outcomes

\\ vS)

No Original Data

©

Conference Abstracts

Nj>

Errata/Retracted

Figure 4-1. Studies identified as supplemental during literature screening.
Click here to view interactive version.

This document is a draft for review purposes only and does not constitute Agency policy,

13	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

4.1.1. Organization of Mechanistic Information

1	For detailed sub-tagging of mechanistic carcinogenesis evidence, studies are organized by

2	the 10 key characteristics of carcinogens (1. electrophilic or can be metabolically activated to an

3	electrophile; 2. genotoxic; 3. alters DNA repair/causes genomic instability; 4. induces epigenetic

4	alterations; 5. induces oxidative stress; 6. induces chronic inflammation; 7. immunosuppressive; 8.

5	modulates receptor-mediated effects; 9. causes cellular immortalization; 10. alters cell

6	proliferation, death, or nutrient supply) Smith etal. f20161. See Figure 4-2 for organizational

7	structure.

8	Similarly, sub-tagging will be undertaken for additional types of mechanistic evidence. This

9	sub-tagging is not based on an a priori construct Instead, it is based on the content of the available
10	studies.

This document is a draft for review purposes only and does not constitute Agency policy.

14	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Mechanism of
carcinogenesis

10 Key Characteristics
of Carcinogens

I

In vitro bacterial/lower
eukaryote

o

Genotoxic

O

Electrophilic/can be
metabohcally activated

o

	
-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

o

Biomarkers

o

In Vivo (Oral)

o

In vivo (inhalatlor)>

o

In Vitro

Figure 4-3. Visual summary of tagging structure for ADME and PK/PBPK
studies.

4.2. METHODS FOR DOSE-RESPONSE ASSESSMENT

4.2.1. Selecting Endpoints for Dose-Response Assessment

1	Based on the SEM (Appendix C) and assessments conducted by others, the NTP inhalation

2	cancer bioassay studies for cobalt sulfate and cobalt metal NTP (19981: NTP (20141 were

3	considered most appropriate for dose-response analysis. Key scientific issues related to MOA and

4	the dose response assessment are outlined in Section 2.4. In addition, statistical and biological

5	information will be used to try to identify BMR levels, and the appropriate dose metrics for animal-

6	to-human extrapolation. If supported by the available data, EPA may develop separate IURs for

7	water-soluble and water-insoluble cobalt compounds, as was done by other agencies (Table 2-2).

8	If this is done, EPA will define a water solubility limit to guide IRIS users as to which IUR to apply

o

Pharmacokinetic
(ADME)

o

Supplemental
(ADME/PBPK)

o

PK/PBPK Models

This document is a draft for review purposes only and does not constitute Agency policy.

16	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

for their specific needs. EPA may also develop additional IURs for certain subsets of cobalt
compounds or develop a single IUR to address all cobalt compounds.

Also considered is whether there are opportunities to quantitatively integrate the evidence.
Tumors of the lung and other tissues were reported in both male and female rats and mice by NTP
(20141. and NTP (19981. Examples of quantitative integration include (1) combining results for an
outcome across sex (within a study); (2) characterizing effects that occur on a continuum (e.g.,
precursors and benign tumors that progress to malignant tumors); (3) conducting a meta-analysis
or meta-regression of multiple studies; and (4) estimating the risk of getting one or more tumors
for any combination of tumors observed in a single bioassay. In addition, mechanistic evidence
that influences the dose-response analyses will be highlighted. This includes evidence related to
susceptibility or evidence informing the potential shape of the dose-response curve (i.e., linear, or
nonlinear dose response as described in the EPA Guidelines for Carcinogen Assessment U.S. EPA
(2005all. Mode(s) of action information relevant to dose-response analysis will be summarized,
including any pathway interactions relevant to understanding overall risk. For cancer dose-
response of animal data, relevant biological considerations are:

•	Is there evidence for direct mutagenicity?

•	Is there evidence of a nonlinear mechanism at low dose?

•	Does tumor latency decrease with increasing exposure?

•	If there are multiple tumor types, which cancers have longer/shorter latency periods?

•	Are incidence data or individual-level available?

•	While benign and malignant tumors of the same cell or tissue of origin are generally
evaluated together, was there an increase only in malignant tumors?

4.2.1.1. Data Extraction and Dose Standardization

Data will be extracted from the NTP inhalation cancer bioassay studies for cobalt sulfate
and cobalt metal NTP (19981:NTP (20141 into EPA's version of Health Assessment Workspace
Collaborative (HAWC, https://hawcprd.epa.gOv/l. a web-based software application designed to
manage and facilitate the process of conducting health assessments. Because the focus of the
current assessment is to develop one or more cancer IURs for inclusion in the IRIS database, tumor
data (along with any other data relevant to dose-response, such as animal survival rates and
individual-level data) will be prioritized for data extraction. Raw data for NTP studies are available
in the Chemical Effects in Biological Systems database (https://cebs.niehs.nih.gov/ceb s/). In
addition to HAWC, data will be stored in other formats necessary for dose-response modeling and
assessment data presentation (i.e., Excel, BMDS, Word). For quality control, data extraction is to be
performed by one member of the evaluation team and independently verified by at least one other

This document is a draft for review purposes only and does not constitute Agency policy.

17	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

member. Discrepancies in data extraction will be resolved by discussion or consultation with a
third member of the evaluation team.

For the dose-response assessment, exposures will be standardized to common units of
mg/m3 elemental cobalt This involves performing a molecular weight conversion from the parent
compound to cobalt The 2-year inhalation cancer bioassay of cobalt metal NTP (20141 does not
require unit conversion since concentration was measured in units of elemental cobalt. However,
the air concentrations presented in the NTP (19981 2-year inhalation cancer bioassay of cobalt
sulfate heptahydrate were in units of mg/m3 anhydrous cobalt sulfate (C0SO4), and not the
heptahydrate or hexahydrate (which it was shown to dehydrate to under the experimental
conditions). This conclusion was based on a review of the assessment analytical details in the NTP
report and Behl etal. (20151. and correspondence with study authors Bucher et al. fin Press!. To
convert from concentrations presented in NTP (19981 to concentrations of elemental cobalt, the
molecular weight ratio of Co (MW=58.933) to CoS04(MW=154.996) will be applied.

All assumptions used in performing dose conversions will be documented in the
assessment Dosimetry adjustments, including converting to continuous chronic exposure from
workday/workweek exposure used in the bioassays and application of model-derived lung
dosimetry factors, will also be documented.

4.2.2. Conducting Dose-Response Assessments

EPA uses a two-step approach for dose-response assessment that distinguishes analysis of
the dose-response data in the range of observation from any inferences about responses at lower,
potentially more environmentally relevant exposure levels U.S. EPA (2012bl: U.S. EPA f2005a. §31:

1)	The first step is an analysis of dose and response in the range of observation of the
experimental or epidemiologic studies. The preferred approach for the first step is to use
dose response modeling to incorporate as much of the data set as possible into the analysis
to derive a point of departure (POD) near the lower end of the observed dose range without
significant extrapolation.

2)	The second step is extrapolation to lower doses. The extrapolation approach considers
what is known about the agent's mode of action. When multiple estimates can be
developed, the strengths and weaknesses of each are presented. In some cases, they may be
combined in a way to best represent human cancer risk.

When sufficient and appropriate human and laboratory animal data are both available for
the same outcome, human data are generally preferred for the dose response assessment because
their use eliminates the need to perform interspecies extrapolations. Findings from human studies
were evaluated but considered less suitable for dose-response primarily due to lack of well-
characterized quantitative exposure estimates and certain study evaluation concerns (e.g., limited
duration and confounding from other exposures). Therefore, the results of the cobalt SEM (see
Appendix C) indicate that animal data represent the most appropriate evidence available for

This document is a draft for review purposes only and does not constitute Agency policy.

18	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

estimating an IUR(s) and these data will be used for dose-response analysis. When there are
multiple tumor types, the final IUR(s) will attempt to address overall cancer risk.

4.2.2.1. Dose-Response Analysis in the Range of Observation

For conducting a dose-response assessment, pharmacodynamic ("biologically based")
modeling can be used when there are sufficient data to ascertain the mode of action and
quantitatively support model parameters that represent rates and other quantities associated with
the key precursor events of the modes of action. If an applicable pharmacodynamic model is not
available to assess health effects associated with inhalation exposure to cobalt, empirical dose-
response modeling will be used to fit the data (on the apical outcomes or a key precursor events) in
the range of the observed data. For this purpose, EPA has developed a software tool (Benchmark
Dose Software, BMDS) that includes a standard set of models fhttp://www.epa.gov/ncea/bmds)
that can be applied to typical data sets, including those that are nonlinear. In situations where
there are alternative models with significant biological support, the users of the assessment can be
informed by the presentation of these alternatives along with the models' strengths and
uncertainties. The EPA has developed guidelines on modeling dose-response data, assessing model
fit, selecting suitable models, and reporting modeling results [see the EPA Benchmark Dose
Technical Guidance U.S. EPA f2012bl1.

U.S. EPA BMDS is designed to help model dose-response datasets in accordance with EPA
Benchmark Dose Technical Guidance U.S. EPA f2012b). With the nonlinear approach of cancer data
analysis based on Guidelines for Carcinogen Risk Assessment U.S. EPA f2005all. a BMCL (for
inhalation exposure data, as is the case for this assessment) is computed using a model selected
from the BMDS suite of models using statistical and graphical criteria. Linear analysis of cancer
datasets generally uses the multistage model, with degree selected following a U.S. EPA Statistical
Workgroup technical memo available on the BMDS website

fhttps://cfpub.epa.gov/ncea/bmds/recordisplay.cfm?deid=308382). Modeling of cancer data may
in some cases involve additional, specialized methods, particularly for multiple tumors or early
removal from observation. For example, when survivals are different across exposure groups
and individual-level data are available, models that include time-to-tumor information may be
useful. Also, additional judgment or alternative analyses may be used if these procedures fail to
yield results in reasonable agreement with the data. For example, modeling may be restricted to the
lower exposure levels, especially if there is competing toxicity at higher concentrations.

For each modeled response, a POD from the observed data should be estimated to mark the
beginning of extrapolation to lower exposure levels. The POD is an estimated exposure level
(expressed in human equivalent terms, e.g., PODhec for inhalation data) near the lower end of the
observed range without significant extrapolation to lower concentrations. For linear extrapolation
of cancer risk, the POD is used to calculate an inhalation unit risk (IUR), and for nonlinear
extrapolation, the POD is used in calculating an RfC. The response level at which the POD is

This document is a draft for review purposes only and does not constitute Agency policy.

19	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

calculated is guided by the severity of the endpoint. If linear extrapolation is used, selection of a
response level corresponding to the POD is not highly influential, so standard values near the low
end of the observable range are generally used (for example, 10% extra risk for cancer bioassay
data, 1% for epidemiologic data, lower for rare cancers). Nonlinear approaches consider both
statistical and biologic considerations. For dichotomous data, a response level of 10% extra risk is
generally used for minimally adverse effects, 5% or lower for more severe effects in experimental
animals. For continuous data, a response level is ideally based on an established definition of
biologic significance. In the absence of such definition, one control standard deviation from the
control mean is often used for minimally adverse effects, one-half standard deviation for more
severe effects. The point of departure is the 95% lower bound on the dose associated with the
selected response level.

EPA has developed standard approaches for determining the relevant exposure level to be
used in the dose-response modeling in the absence of appropriate pharmacokinetic modeling.

These standard approaches (limited here to inhalation cancer) also facilitate comparison across
exposure patterns and species:

•	Intermittent study exposures will be standardized to a daily average over the duration of
exposure. For chronic effects, daily exposures are averaged over the lifespan. Exposures
during a critical period, however, are not averaged over a longer duration U.S. EPA (2005a ,
S3.1.11:	91. S3.21. Note that this will typically be done after modeling because
the conversion is linear.

•	Exposure concentrations will be standardized to equivalent human terms (via a common
internal dose metric for animals and humans) to facilitate comparison of results from
different species. Inhalation exposures are scaled using dosimetry models that apply
species-specific physiologic and anatomic factors and consider whether the effect occurs at
the site of first contact or after systemic circulation U.S. EPA f2012al:	)4. §31.
The preferred approach for dosimetry extrapolation from animals to humans is through
PBPK modeling. Methods for lung dosimetry are described in Methods for Derivation of
Inhalation Reference Concentrations and Application of Inhalation Dosimetry U.S. EPA

1), and in EPA's MPPD Technical Support Documentation and User's Guide U.S. EPA
£2022).

In the absence of study specific data on, for example, inhalation rates or body weight, the
EPA has developed recommended values for use in dose response analysis U.S. EPA

f 19881.

For additional dose-response considerations specific to this assessment, see Studies that
Meet SEM PECO Criteria.

4.2.2.2. Extrapolation: Unit Risk

An IUR is calculated to facilitate estimation of human cancer risks when low-dose linear
extrapolation for cancer effects is supported, particularly for chemicals with direct mutagenic
activity or those for which the data indicate a linear component below the POD. Low-dose linear

This document is a draft for review purposes only and does not constitute Agency policy.

20	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

extrapolation is also used as a default when the data are insufficient to establish the mode of action
U.S. EPA (2005al If the currently available data on cobalt compounds (or specific tumors resulting
from cobalt exposure) are judged as sufficient to "ascertain the M0A[s] and conclude that it is not
linear at low doses and the agent [cobalt] does not demonstrate mutagenic or other activity
consistent with linearity at low doses ...Where alternative approaches with significant biological
support are available for the same tumor response and no scientific consensus favors a single
approach, [the] assessment may present results based on more than one approach (e.g., both low-
dose linear and reference concentration approaches)" U.S. EPA (2005al Both approaches may also
be used when there are multiple MOAs identified. When multiple approaches are presented, the
assessment will describe the strengths and uncertainties of each before selecting and justifying a
final estimate.

4.2.2.3. Extrapolation: Reference Concentrations

Reference value derivation is EPA's most frequently used type of nonlinear extrapolation
method. Although it is most commonly used for noncancer effects, this approach is also used for
cancer effects if there are sufficient data to ascertain the MOA and conclude that it is not linear at
low doses. For these cases, reference values for each relevant route of exposure are developed
following EPA's established practices U.S. EPA (2005a. §3.3.4): in general, the reference value is
based not on tumor incidence, but on a key precursor event in the MOA that is necessary for tumor
formation. If a reference value approach is presented as an alternative to the IUR, reference value
derivation will be performed in accordance with current EPA guidelines U.S. EPA (19981: U.S. EPA
f!9961: t n I i \ i I' 94); LU> I ! \ l I" 1); 1LSJ IW.i '0021: IJ.S. EPA (20111: U.S. EPA (201.4a1

This document is a draft for review purposes only and does not constitute Agency policy.

21	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

REFERENCES

ACGIH (American Conference of Governmental Industrial Hygienists). (2001a). Cobalt and
inorganic cobalt compounds. In Documentation of the threshold limit values and
biological exposure indices (7th ed.). Cincinnati, OH.

ACGIH (American Conference of Governmental Industrial Hygienists). (2001b). Cobalt
carbonyl. In Documentation of the threshold limit values and biological exposure
indices (7th ed.). Cincinnati, OH.

ACGIH (American Conference of Governmental Industrial Hygienists). (2001c). Cobalt
hydrocarbonyl. In Documentation of the threshold limit values and biological
exposure indices (7th ed.). Cincinnati, OH.

Alpha Chemicals (Alpha Chemicals Pty Ltd). (2020). Cobalt chloride anhydrous: Safety data
sheet. Wetherill Park, Australia, https://alphachem.com.au/wp-
content/uploads/2021/07/SDS-Cobalt-Chloride-Anhydrous.pdf.

AR.TEAM. (2022). Solubility: Cobalt(II) bromide [CoBr2]. Available online at

http://periodic-table-of-elements.org/SOLUBILITY/cobalt-II bromide (accessed
February 15, 2022).

ATSDR (Agency for Toxic Substances and Disease Registry). (2004). Toxicological profile
for cobalt [ATSDR Tox Profile]. Atlanta, GA: U.S. Department of Health and Human
Services, Public Health Service, http://www.atsdr.cdc.gov/toxprofiles/tp33.pdf.

Bannach-Brown. A: Przvbvla. P: Thomas. I: Rice. ASC: Ananiadou. S: Liao. I: Macleod. MR.

(2018). Machine learning algorithms for systematic review: reducing workload in a
preclinical review of animal studies and reducing human screening error (pp. 1-26).
bioRxiv. http://dx.doi.org/10.1101/25576Q.

Behl. M: Stout. MP: Herbert. RA: Dill. TA: Baker. GL: Havden. BK: Rovcroft. TH: Bucher. TR:

Hooth. ML (2015). Comparative toxicity and carcinogenicity of soluble and insoluble
cobalt compounds. Toxicology 333: 195-205.
http://dx.doi.Org/10.1016/j.tox.2015.04.008.

Bucher. TR: Elwell. MR: Thompson. MB: Chou. BT: Renne. R: Ragan. HA. (1990). Inhalation
toxicity studies of cobalt sulfate in F344/N rats and B6C3F1 mice. Toxicol Sci 15:
357-372.

Bucher. TR: Hailev. TR: Rovcroft. TR: Haseman. IK: Sills. RC: Grumbein. SL: Mellick. PW: Chou.
BT. (In Press) Correction to: Inhalation toxicity and carcinogenicity studies of cobalt
sulfate [Erratum]. Toxicol Sci. http://dx.doi.org/10.1093/toxsci/kfac063.

Bucher. TR: Hailev. TR: Rovcroft. TR: Haseman. IK: Sills. RC: Grumbein. SL: Mellick. PW: Chou.
BT. (1999). Inhalation toxicity and carcinogenicity studies of cobalt sulfate. Toxicol
Sci 49: 56-67. http://dx.doi.Org/10.1093/toxsci/49.l.56.

CADENAS. (2022). Cobalt(II) carbonate. Available online at

https://b2b.partcommunity.com/community/knowledge/pl/detail/2827/CobaltriI
1+carbonate (accessed February 15, 2022).

This document is a draft for review purposes only and does not constitute Agency policy.

22	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Cohen, AM; Hersh, WR; Peterson, K; Yen, FY, (2006). Reducing workload in systematic

review preparation using automated citation classification. J Am Med Inform Assoc
13: 206-219. http://dx.doi.org/10.1197/iamia.M1929.

Colognato. R: BoneHt i\«uu 1 1 arina, M: Bergamaschi, E: Sabbioni, E: Migliore, L. f20081.
Comparative genotoxicity of cobalt nanoparticles and ions on human peripheral
leukocytes in vitro. Mutagenesis 23: 377-382.
http://dx.doi.org/lQ.lQ93/mutage/genQ24.

DOE (U.S. Department of Energy). (2018). Protective Action Criteria (PAC): Chemicals with
AEGLs, ERPGs, & TEELs. Rev. 29A. Available online at https://edms.energy.gov/pac/
(accessed February 15, 2022).

ECHA (European Chemicals Agency). (2017). Committee for Risk Assessment (RAC).

Opinion proposing harmonised classification and labelling at EU level of cobalt. CLH-
0-0000001412-86-172/F. Helsinki, Finland.

https://echa.europa.eu, uments/10162/b7316bll-ae65-ldd0-2e64-

bb6ad3efbd82.

ECHA (European Chemicals Agency). (2022). Brief profile: Cobalt bis(2-ethylhexanoate).

Available online at https://echa.europa.eu/brief-profile/-/briefprofile/100.004.773
(accessed March 14, 2022).

Hong, HH: Hoenerhoff, Ml; T(	rbert. RA; Kissling, GE: Hooth, Ml; Behl, M: Witt. KL:

Smith-Roe. SL: Sills. RC: Pandiri. AR. (2015). Kras, Egfr, and Tp53 Mutations in
B6C3F1/N Mouse and F344/NTac Rat Alveolar/Bronchiolar Carcinomas Resulting
from Chronic Inhalation Exposure to Cobalt Metal. Toxicol Pathol 43: 872-882.
http://dx.doi.oi-.' It) 11
Howaul 01 Piuihj I 1 liller, K: Tandon, A: Mav, D: Shah, MR: Holmgren, S: Pelch, KE:

Walki > \ I.\hhh y AA Uacleod. M: Shah • '.aver. K. (2016). SWIFT-Review: A
text-mining workbench for systematic review. Syst Rev 5: 87.
http://dx.doi.org/lQ.1186/sl3643-Q16-Q263-z.

Howai d 01 Huihps. 1: Tandon. A: Maharana. A: Elmore. R: Mav. D: Sedvkh. A: Thayer. K:
Merrick. BA: Walker. ¥: Roonev, A: Shah. RR. (2020). SWIFT-Active Screener:
Accelerated document screening through active learning and integrated recall
estimation. Environ Int 138: 105623.
http://dx.doi.Oi " It) it) Uv i.envint.2020.105623.

IARC (International Agency for Research on Cancer). (2006). Cobalt in hard metals and

cobalt sulfate, gallium arsenide, indium phosphide and vanadium pentoxide. Lyon,
France, htti	¦iarc.fr/Book-And-Report-Series/Iarc-Monographs-On-

The-IdentiPication-Of-Carcinoge zards-To-Humans/Cobalt-In-Hard-Metals-
And-Cobah '•ulfate-Galliiim-Arsenide-Indium-PhosphhU Ami \ aaadium-Pentoxide-
2006.

Kara gas. MR: Wang. A: Dorm	rgi. CM: Svmanski. E: Ward. EM:

Arrandale, VH: Azuma, K: Brambila, E. (2022). Carcinogenicity of cobalt, antimony
compounds, and weapons-grade tungsten alloy. Lancet Oncol 23: 577-578.
http://dx.doi.oilt)lt)U} Ml t) t) J5f 2 210*) 'UJ 1.

Kerfoot. El. (1973). Chronic animal inhalation toxicity to cobalt. Cincinnati, OH: U.S.

Department of Health, Education, and Welfare, National Institute for Occupational
Safety and Health.

https://ntrl.iitis.gov/WTRL/dashboard/searchRc	?Detail/PB232247.xhtml.

This document is a draft for review purposes only and does not constitute Agency policy.

23	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Kresovich, IK; ErdaL »'lu ^ U inn, PH; Argos, M; Rauscher, GH. f20191. Metallic air
pollutants and breast cancer heterogeneity. Environ Res 177: 108639.
http://dx.doi.oi " It) lit in. i.envres.2019.108639.

Lavut. EG; Timofevev. Bl; Yuldasheva. VM. (1989). Enthalpy of formation of cobalt chloride.
J Chem Thermodyn 21: 673-676. iuxp:, ' ItimiALBcgec^CjJ^gueii\ T i\ lUi J ,
Perdrix. A. (1998). Lung cancer risk in hard-metal workers. Am J Epidemiol 148:
241-248. http://dx.dc	xfordiournals.aie.a009631.

Mur. TM: Moulii larruyer-Seinerra. MP: LaFitte. 1. (1987). A cohort mortality study

among cobalt and sodium workers in an electrochemical plant. Am J Ind Med 11: 75-
81. http://dx.doi.oi " It) it)') j{itn J t)') 1 It) it);?.

National Toxicology Program (NTP). (1998). Toxicology and carcinogenicity studies of

cobalt sulfate heptahydrate (CAS No. 10026-24-1) in F344/N rats and B6C3F1 Mice
(inhalation studies).

NCBI (National Center for Biotechnology Information). (2021). PubChem. Available online
at https://piibchem.ncbi.nlrri.nih.goy/#collection=bioassavs (accessed

NDEP (Nevada Division of Environmental Protection). (2017). Basic Comparison Levels.

Carson City, NV. https://ndep.nv.gov/uploads/documents/july-2017-ndep-bcls.pdf.

NIOSH (National Institute for Occupational Safety and Health). (2019). NIOSH pocket guide
to chemical hazards: Cobalt hydrocarbonyl (as Co).
https://www.cdc.gov/niosh/npg/npgd0148.html.

NIST (National Institute of Standards and Technology). (2021a). Cobalt,
tetracarbonylhydro. Available online at

https://webbook.nist.gov/cgi/cbook.cgi?ID=C16842038&I)nits=SI&Mas
(accessed February 16, 2022).

NIST (National Institute of Standards and Technology). (2021b). Dicobalt octacarbonyl.
Available online at

httPs://webbook.nist.gov/cgi/cbook.cgi?ID=C10210681&I)nits=SI&Mas
(accessed February 16, 2022).

NTP (National Toxicology Program). (1991). Toxicity studies of cobalt sulfate heptahydrate
(CAS no 10026-24-1) in F344/N rats and B6C3F1 mice (inhalation studies). (NIH
Publication No. 91-3124). Research Triangle Park, NC.

NTP (National Toxicology Program). (1998). Toxicology and carcinogenesis studies of

cobalt sulfate heptahydrate (CAS No. 10026-24-1) in F344/N rats and B6C3F1 mice
(inhalation studies). (NTPTR471). Research Triangle Park, NC.

This document is a draft for review purposes only and does not constitute Agency policy.

24	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

NTP (National Toxicology Program). (2014). NTP Technical Report on the Toxicology

Studies of Cobalt Metal (CAS No. 7440-48-4) in F344/N Rats and B6C3F1/N Mice
and Toxicology and Carcinogenesis Studies of Cobalt Metal in F344/NTac Rats and
B6C3F1/N Mice (Inhalation Studies). (TR-581). Research Triangle Park, NC.

httpsi//ntp.nielis.iiili.gov/iitp/litdocs/It rpts/tr581 508.pdf.

NTP (National Toxicology Program). (2016). Monograph on cobalt and cobalt compounds
that release cobalt ions in vivo. In Report on carcinogens. (CAS No. 7440-48-4).
Research Triangle Park, NC.

https://ntp.niehs.nih.gov/ntp/roc/monographs/cobalt final 508.pdf.

NTP (National Toxicology Program). (2021). Report on carcinogens, fifteenth edition:

Cobalt-related exposures. Research Triangle Park, NC: U.S. Department of Health
and Human Services, National Institutes of Health, National Toxicology Program.
https://ntp.niehs.nih.gov/ntp/roc/content/profiles/cobalt.pdf.

OEHHA (California Office of Environmental Health Hazard Assessment). (2019). Cobalt
metal powder. Available online at littps://oehha.ca.gov/chemicals/cobalt-metal-
powder (accessed February 16, 2022).

OEHHA (California Office of Environmental Health Hazard Assessment). (2020). Cobalt and
cobalt compounds cancer inhalation unit risk factors. Technical support document
for cancer potency factors: Appendix B. Sacramento, CA: California Environmental
Protection Agency, Office of Environmental Health Hazard Assessment, Air,
Community, and Environmental Research Branch, Air Toxics Hot Spots Program.
https://oehha.ca.gov/media/downloads/crnr/cobaltcpfl00220.pdf.

Osman. D: Cooke. A: Young. TR: Deei . I	nm. N1: Warren. Ml. (2021). The

requirement for cobalt in vitamin B12: A paradigm for protein metalation [Review].
Biochim Biophys Acta Mol Cell Res 1868: 118896.

Ozaki. K: Haseman. IK: Hailev. IR: Maronpot. RR: Nvska. A. (2002). Association of adrenal
pheochromocytoma and lung pathology in inhalation studies with particulate
compounds in the male F344 rat-the National Toxicology Program experience.
Toxicol Pathol 30: 263-270. Imi)://dx.doi.org/10.1080/019262302753559605.
Palmes. ED: Nelson. N: Laskin. S: Kuschner. M. (1959). Inhalation toxicity of cobalt
hydrocarbonyl. Am Ind Hyg Assoc J 20: 453-468.
http://dx.doi.org/10.1080/00028895909343751.

Pomi ^	I \limaro. B: Broggi. F: Marmot «r I t hini. F: Colognato. R: Rossi.

F, (2009). Genotoxicity and morphological transformation induced by cobalt
nanoparticles and cobalt chloride: an in vitro study in Balb/3T3 mouse fibroblasts.
Mutagenesis 24: 439-445.	¦Ldoi.org/10.1093/rniitage/gep02.

RSC (Royal Society of Chemistry). (2022). Cobalt(II) carbonate; ChemSpider ID10123.

Available online at	A^w.chemspider.com/Chemical-Structure. 10123.html

(accessed March 11, 2022).

Sauni. R: Ok	ia. A: Kerttula. R: Pukkala. E. (2017). Cancer incidence among

Finnish male cobalt production workers in 1969-2013: a cohort study. BMC Cancer
17: 340. http://dx.doi.org/10.1186/sl288S-017-3333-2.

ScholAR Chemistry. (2009). Cobalt (II, III) oxide: Material safety data sheet. Rochester, NY.
https://www.mccsd.net/cms/lib/NY02208580/Centricitv/Shared/Material%20Saf

This document is a draft for review purposes only and does not constitute Agency policy.

25	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

etv%20Data%20S!ieets%20 MSDS /MSPS%20Sheets Cob	00 QO.p

df.

Slack, IF; Kimball, BE; Shedd, KB. (2017). Cobalt: chapter F of critical mineral resources of
the United States—economic and environmental geology and prospects for future
supply. US Geological Survey Professional Papers F: F1-F40.
http://dx.doi.org/10.3133/ppll

Si.mh I [ hiu s I [ mu nal. SK: Mason. MP: Ziu u'.1 I ' 'isc, IP. (2014). The cytotoxicity
and genotoxicity of soluble and particulate cobalt in human lung fibroblast cells.
Toxicol Appl Pharmacol 278: 259-265.
http://dx.doi.Org/10.1016/i.taap.2014.05.002.

Smith. MT: Guvton, KZ: Gibbons, CF: Fritz, 1M; Portiei ^ 1: iisvii, I: tVMarini, DM: Caldwell,
h IxhIocJlKI I vUTibei IN	{V cherJBi St:e, warL B Wi B

VI: Straif. K. (2016). Key characteristics of carcinogens as a basis for organizing data
on mechanisms of carcinogenesis [Review]. Environ Health Perspect 124: 713-721.
http://dx.doi.org/10.1289/ehp.1509912.

Suh. M: Thompson. CM: Brorbv. GP: Mittai. L. iz: Proctor. DM. (2016). Inhalation cancer risk
assessment of cobalt metal [Review]. Regul Toxicol Pharmacol 79: 74-82.
http://dx.doi.Org/10.1016/i.yrtph.2016.05.009.

TCEO (Texas Commission on Environmental Quality). (2017). Development support
document: Cobalt and cobalt compounds.

https://www.tcea.texas.gOv/assets/public/implementation/tox/dsd/Final/cobalt.p
df.

Ton. TT: Ko\ i I 1 ^ dada. TN: Chhabria. RM: Shocklev. KR: Flagh i \ rrish. KE:

Herbert. RA: Behl. iiierhoff. Ml: Sills, RC: Pandiri, AR. f20211. Cobalt-induced
oxidative stress contributes to alveolar/bronchiolar carcinogenesis in B6C3F1/N
mice. Arch Toxicol 95: 3171-3190. lmi)://dx.doi.org/10.1007/s00204-021-U * 1 Ju
5.

Tuch.M u ^ U usen. MV: Villadsen ^	!•]. (1996). Incidence of lung cancer among

cobalt-exposed women. Scand J Work Environ Health 22: 444-450.

U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and
documentation of biological values for use in risk assessment [EPA Report].
(EPA600687008). Cincinnati, OH.
http://cfpub.epa.gov/ncea/cfm/recordisplay.

U.S. EPA (U.S. Environmental Protection Agency). (1991). Guidelines for developmental
toxicity risk assessment. Fed Reg 56: 63798-63826.

U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of

inhalation reference concentrations and application of inhalation dosimetry [EPA
Report]. (EPA600890066F). Research Triangle Park, NC.

https://cfpub.epa.gov/ncea/risk/recordisplay.cfin	H-'B&CFII^ ' 11 J029&

> ' ^ M ^ ' 'U)63'l7.

U.S. EPA (U.S. Environmental Protection Agency). (1996). Guidelines for reproductive
toxicity risk assessment (pp. 1-143). (EPA/630/R-96/009). Washington, DC: U.S.
Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.gov/sites/production/files/2014-
11 Uocumem "nidelines repro toxicitv.pdf.

This document is a draft for review purposes only and does not constitute Agency policy.

26	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

U.S. EPA (U.S. Environmental Protection Agency). (1998). Guidelines for neurotoxicity risk

assessment [EPA Report] (pp. 1-89). (ISSN 0097-6326
EISSN 2167-2520

EPA/630/R-95/001F). Washington, DC: U.S. Environmental Protection Agency, Risk
Assessment Forum, http://www.epa.gov/risk/guide1ines-neurotoxi city-risk-
assessment.

U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose
and reference concentration processes. (EPA630P02002F). Washington, DC.
https://www.epa.gov/sites/production/files/2014-12/documents/rfd-final.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2005a). Guidelines for carcinogen risk
assessment [EPA Report]. (EPA630P03001F). Washington, DC.
https://www.epa.gov/sites/production/files/2Q13-
09/docume ncer guidelines final 3-25-05.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2005b). Supplemental guidance for
assessing susceptibility from early-life exposure to carcinogens [EPA Report].
(EPA/630/R-03/003F). Washington, DC: U.S. Environmental Protection Agency,

Risk Assessment Forum, https://www.epa.gov/risk/supplemental-guidance-
assessing-susceptibilitv-early-life-exposure-carcinogens.

U.S. EPA (U.S. Environmental Protection Agency). (2008). Provisional peer reviewed

toxicity values for cobalt (CASRN 7440-48-4) [EPA Report]. (EPA/690/R-08/008F).
Cincinnati, OH. https://cfpub.epa.gov/ncea/pprtv/recordisplay.cf 1=338894.
U.S. EPA (U.S. Environmental Protection Agency). (2011). Recommended use of body
weight 3/4 as the default method in derivation of the oral reference dose.
(EPA100R110001). Washington, DC.

https://www.epa.gov/sites/production/files/2013-09/documents/recommended-

use-of-bw34.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2012a). Advances in inhalation gas
dosimetry for derivation of a reference concentration (RfC) and use in risk
assessment (pp. 1-140). (EPA/600/R-12/044). Washington, DC.
https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=244650&CFID=50l
&.€ FTP K E N = 17139189.

U.S. EPA (U.S. Environmental Protection Agency). (2012b). Benchmark dose technical
guidance [EPA Report]. (EPA100R12001). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum, https://www.epa.gov/risk/benchmark-
dose-technical-guidance.

U.S. EPA (U.S. Environmental Protection Agency). (2014a). Guidance for applying

quantitative data to develop data-derived extrapolation factors for interspecies and
intraspecies extrapolation [EPA Report]. (EPA/100/R-14/002F). Washington, DC:
Risk Assessment Forum, Office of the Science Advisor.

https://www.epa.gov/sites/production/files/2015-01/docun'	7-final.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2014b). Substance registry services
[Database]. Washington, D.C. Retrieved from

https ipiib.epa.gov/sor internet/registry/substreg/searchandretrieve/substa
ncesearch/search.do

U.S. EPA (U.S. Environmental Protection Agency). (2015). Peer review handbook [EPA
Report] (4th ed.). (EPA/100/B-15/001). Washington, DC: U.S. Environmental

This document is a draft for review purposes only and does not constitute Agency policy.

27	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Protection Agency, Science Policy Council, https://www.epa.gov/osa/peer-review-
handbook-4th-edition-2015.

U.S. EPA (U.S. Environmental Protection Agency). (2018). An umbrella Quality Assurance
Project Plan (QAPP) for PBPK models [EPA Report]. (0RD QAPP ID No: B-0030740-
QP-1-1). Research Triangle Park, NC.

U.S. EPA (U.S. Environmental Protection Agency). (2019). ChemView [Database]. Retrieved
from https://chemview.epa.gov/chemview

U.S. EPA (U.S. Environmental Protection Agency). (2020a). 2017 National Emissions

Inventory (NEI) data (April 2020 version) (Version April 2020). Washington, DC: US
Environmental Protection Agency. Retrieved from https://www.epa.gov/air-
emissions-inventories/2017-national-emissions-inventorv-nei-data

U.S. EPA (U.S. Environmental Protection Agency). (2020b). Benchmark Dose Software

(BMDS). Version 3.2: User guide [EPA Report]. (EPA/600/R-20/216). Washington,
DC: U.S. Environmental Protection Agency, Office of Research and Development.
https://nepis.epa.gov/Exe/ZyPURL.cgi?Dock<	3T2.txt.

U.S. EPA (U.S. Environmental Protection Agency). (2020c). ORD staff handbook for

developing IRIS assessments (public comment draft) [EPA Report]. (EPA/600/R-
20/137). Washington, DC: U.S. Environmental Protection Agency, Office of Research
and Development, Center for Public Health and Environmental Assessment.
https://cfpub.epa.gov/ncea/iris drafts/recordisplay.cfm?deid=350086.

U.S. EPA (U.S. Environmental Protection Agency). (2020d). Toxicology testing in the 21st
century (Tox21). Available online at https://ntp.niehs.nih.gov/go/tox21 (accessed
May 4, 2021).

U.S. EPA (U.S. Environmental Protection Agency). (2021). CompTox chemicals dashboard.
Washington, DC. Retrieved from https://comptox.epa.gov/dashboard

U.S. EPA (U.S. Environmental Protection Agency). (2022). Multiple-path Particle Dosimetry
(MPPD) model: EPA technical support documentation and user's guide (EPA MPPD
2022 v.2.0).

Whit s!; ¦-v 1 > ien. KM: Niehoff. NM: Carroll. R: Saudi > (2019). Metallic air pollutants
and breast cancer risk in a nationwide cohort study. Epidemiology 30: 20-28.

http://dx.doi.Oi ItMtH S H00000000000917.

This document is a draft for review purposes only and does not constitute Agency policy.

28	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

APPENDIX A. CHEMICAL AND PHYSICAL
PROPERTIES OF INCLUDED FORMS

A.l. KEY COMPOUNDS IDENTIFIED DURING SCOPING

Table A-l. Chemical identity and physicochemical properties of cobalt

Characteristic or property

Value3

Reference

Chemical structure

Co

U.S. EPA (2021)

CASRN

7440-48-4

U.S. EPA (2021)

Synonyms

cobalt element

U.S. EPA (2021)

Color/form

hard, lustrous, silver-gray metal

U.S. EPA (2021)

Molecular formula

Co

U.S. EPA (2021)

Molecular weight (g/mol)

58.933

U.S. EPA (2021)

Density (g/cm3)

8.9 at 20°C

ATSDR (2004)

Boiling point (°C)

3,000

U.S. EPA (2021)

Melting point (°C)

1,500

U.S. EPA (2021)

Heat of formation (kJ/mol)

427.7 (gas)

NCBI (2021)

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant
(atm-m3/mol)

ND

NA

Solubility in water (g/L)

2.9 x 10"3

OEHHA (2020)

Vapor pressure (mmHg)

1 at 1,910 °C

ATSDR (2004)

NA = not applicable; ND = no data.

a When available, average experimental values are reported from U.S. EPA (2021) Chemicals Dashboard (Cobalt
DTXSID1031040): https://comptox.epa.gov/dashboard/chemical/details/DTXSID1031040.

This document is a draft for review purposes only and does not constitute Agency policy.

A-l	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table A-2. Chemical identity and physicochemical properties of cobalt oxide

Characteristic or property

Value

Reference

Chemical structure

Co2+ O2"

U.S. EPA (2021)

CASRN

1307-96-6

U.S. EPA (2021)

Synonyms

cobalt(ll) oxide, cobaltous oxide, FCO
178, (oxido)cobalt, Zaffre, C.I. 77322,
C.I. Pigment Black 13, cobalt black,
cobalt monoxide, cobaltoxid

U.S. EPA (2021)

Color/form

olive-green or gray solid

U.S. EPA (2021)

Molecular formula

CoO

U.S. EPA (2021)

Molecular weight (g/mol)

74.932

U.S. EPA (2021)

Density (g/cm3)

6.45

ATSDR (2004)

Boiling point (°C)

ND

NA

Melting point (°C)

1,935

NCBI (2021)

Heat of formation (kJ/mol)

-237.9

NCBI (2021)

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

4.88 x lO"3 at 20°C

NCBI (2021)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.

Table A-3. Chemical identity and physicochemical properties of hexanoic acid,
2-ethyl-, cobalt(2+) salt

Characteristic or property

Value3

Reference

Chemical structure

H 0

h3c^ ch3

U.S. EPA (2021)

CASRN

136-52-7

U.S. EPA (2021)

Synonyms

cobalt(2+) bis(2-ethylhexanoate); 2-ethylhexanoic
acid cobalt(2+) salt; bis(2-ethylhexanoate) de
cobalt; cobalt 2-ethylhexanoate; cobalt bis(2-

U.S. EPA (2021)

This document is a draft for review purposes only and does not constitute Agency policy.

A-2	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Characteristic or property

Value3

Reference



ethylhexanoate); cobalt(ll) 2-ethylhexanoate;
cobalt octoate; cobaltous 2-ethylhexanoate;
cobaltous octoate; hexanoate, 2-ethyl-, cobalt;
Octlife Co 12; Octlife Co 8; Versneller NL49



Color/form

blue liquid

NCBI (2021)

Molecular formula

C16H30C0O4

U.S. EPA (2021)

Molecular weight (g/mol)

345.345

U.S. EPA (2021)

Density (g/cm3)

1.01

NIP (2016)

Boiling point (°C)

decomposes at 90

NCBI (2021)

Melting point (°C)

53-84 at 100.5- 101.325 kPa

ECHA (2022)

Heat of formation (kJ/mol)

ND

NA

Log Kow

2.96 at 20°C

ECHA (2022)

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

40.3 at 20°C

ECHA (2022)

Vapor pressure (Pa)

5

ECHA (2022)

NA = not applicable; ND = no data.

Table A-4. Chemical identity and physicochemical properties of cobalt nitrate

Characteristic or property

Value

Reference

Chemical structure

0" 0

/ -

G=N+ O	N

\ \

0 0

Co +

U.S. EPA (2021)

CASRN

10141-05-6

U.S. EPA (2021)

Synonyms

cobalt(ll) nitrate; cobalt dinitrate; cobalt
bis(nitrate); cobaltous nitrate; nitric acid,
cobalt(2+) salt

U.S. EPA (2021)

Color/form

red solid

ATSDR (2004)

Molecular formula

Co(NOb)2

U.S. EPA (2021)

Molecular weight (g/mol)

182.941

U.S. EPA (2021)

This document is a draft for review purposes only and does not constitute Agency policy.

A-3	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Characteristic or property

Value

Reference

Density (g/cm3)

2.49

ATSDR (2004)

Boiling point (°C)

NA

NA

Melting point (°C)

decomposes at 100-105

ATSDR (2004)

Heat of formation (kJ/mol)

-420.5

NCBI (2021)

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

670

OEHHA (2020)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.







Table A-5. Chemical identity and physicochemical properties of cobalt nitrate
hexahydrate



Characteristic or property

Value

Reference



Chemical structure

O O"

// /

O	M+ 0	N* Co '

\ . \\
0 0

h2o h2o h2o
h2o h2o h2o

U.S. EPA (2021)



CASRN

10026-22-9

U.S. EPA (2021)



Synonyms

cobalt(2+) nitrate--water

U.S. EPA (2021)



Color/form

red solid

NCBI (2021)



Molecular formula

Co(N03)2x6 H20

U.S. EPA (2021)



Molecular weight (g/mol)

291.031

U.S. EPA (2021)



Density (g/cm3)

1.88

NCBI (2021)



Boiling point (°C)

decomposes at 74

NCBI (2021)



Melting point (°C)

55

NCBI (2021)



Heat of formation (kJ/mol)

ND

NA



Log Kow

ND

NA



Koc(L/kg)

ND

NA



This document is a draft for review purposes only and does not constitute Agency policy.

A-4	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Characteristic or property

Value

Reference

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

1,338 at 0°C

NCBI (2021)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.

Table A-6. Chemical identity and physicochemical properties of cobalt
bromide

Characteristic or property

Value

Reference

Chemical structure

9 +

Rr Co
br Br

U.S. EPA (2021)

CASRN

7789-43-7

U.S. EPA (2021)

Synonyms

cobalt(ll) bromide, cobalt
dibromide, cobaltous bromide

U.S. EPA (2021)

Color/form

green solid

U.S. EPA (2021)

Molecular formula

CoBr2

U.S. EPA (2021)

Molecular weight (g/mol)

218.741

U.S. EPA (2021)

Density (g/cm3)

4.909

NCBI (2021)

Boiling point (°C)

927

AR.TEAM (2022)

Melting point (°C)

678

NCBI (2021)

Heat of formation (kJ/mol)

-220.9

NCBI (2021)

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

1,132 at 20°C

NCBI (2021)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.

This document is a draft for review purposes only and does not constitute Agency policy.

AS	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table A-7. Chemical identity and physicochemical properties of cobalt
carbonate

Characteristic or property

Value

Reference

Chemical structure

0

_ 2 +

Co

U.S. EPA (2021)

CASRN

513-79-1

U.S. EPA (2021)

Synonyms

carbonic acid, cobalt(2+) salt (1:1),
cobalt(ll) carbonate

U.S. EPA (2021)

Color/form

reddish paramagnetic solid

U.S. EPA (2021)

Molecular formula

C0CO3

U.S. EPA (2021)

Molecular weight (g/mol)

118.941

U.S. EPA (2021)

Density (g/cm3)

4.13

CADENAS (2022)

Boiling point (°C)

ND

NA

Melting point (°C)

decomposes at 427

CADENAS (2022)

Heat of formation (kJ/mol)

-722.6

CADENAS (2022)

Log Kow

-1.192

RSC (2022)

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

11.4 x lO"3

OEHHA (2020)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.

This document is a draft for review purposes only and does not constitute Agency policy.

A-6	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table A-8. Chemical identity and physicochemical properties of cobalt
chloride

Characteristic or property

Value3

Reference

Chemical structure

cr co2+ cr

U.S. EPA (2021)

CASRN

7646-79-9

U.S. EPA (2021)

Synonyms

cobalt(ll) chloride, cobalt
dichloride, cobaltous chloride

U.S. EPA (2021)

Color/form

blue solid

ATSDR (2004)

Molecular formula

C0CI2

U.S. EPA (2021)

Molecular weight (g/mol)

129.83

U.S. EPA (2021)

Density (g/cm3)

3.4

NCBI (2021)

Boiling point (°C)

1,050

U.S. EPA (2021)

Melting point (°C)

411

U.S. EPA (2021)

Heat of formation (kJ/mol)

-311.07

Lavut et al. (1989)

Log Kow

0.8494

Alpha Chemicals (2020)

Koc(L/kg)

23.74

Alpha Chemicals (2020)

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

450

OEHHA (2020)

Vapor pressure (mmHg)

75 at 818°C

NCBI (2021)

NA = not applicable; ND = no data.

a When available, average experimental values are reported from U.S. EPA (2021) Chemicals Dashboard (Cobalt
chloride DTXSID9040180): https://comptox.epa.gov/dashboard/chemical/details/DTXSID9040180.

Table A-9. Chemical identity and physicochemical properties of cobalt
hydrocarbonyl

Characteristic or property

Value

Reference

Chemical structure

0+

cr

_ J

0—C—Co—C—0+

iiv

o+

U.S. EPA (2021)

CASRN

16842-03-8

U.S. EPA (2021)

This document is a draft for review purposes only and does not constitute Agency policy.

A-7	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Characteristic or property

Value

Reference

Synonyms

carbon monooxide-cobalt

U.S. EPA (2021)

Form

flammable gas with offensive odor

ACGIH (2001c)

Molecular formula

C4HC0O4

U.S. EPA (2021)

Molecular weight (g/mol)

171.981

U.S. EPA (2021)

Relative gas density

5.93

NIOSH (2019)

Boiling point (°C)

10

DOE (2018)

Melting point (°C)

-26

ACGIH (2001c)

Heat of formation (kJ/mol)

-569.2

NIST (2021a)

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

0.5

ACGIH (2001c)

Vapor pressure (atm)

>1

NIOSH (2019)

NA = not applicable; ND = no data.

This document is a draft for review purposes only and does not constitute Agency policy.

A-8	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table A-10. Chemical identity and physicochemical properties of cobalt oxide
(II, HI)

Characteristic or property

Value

Reference

Chemical structure

C0O.C02O3

NCBI (2021)

CASRN

1308-06-1

U.S. EPA (2021)

Synonyms

cobaltic-cobaltous oxide, cobalto-
cobaltic oxide, cobalto-cobaltic
tetroxide, cobaltosic oxide, cobalt
tetraoxide, tricobalt tetraoxide

U.S. EPA (2021)

Color/form

black antiferromagnetic solid

U.S. EPA (2021)

Molecular formula

C03O4

U.S. EPA (2021)

Molecular weight (g/mol)

240.797

NCBI (2021)

Density (g/cm3)

6.07

ATSDR (2004)

Boiling point (°C)

decomposes at 900

ScholAR Chemistrv (2009)

Melting point (°C)

895

ScholAR Chemistrv (2009)

Heat of formation (kJ/mol)

ND

NA

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

1.6 x 10"3

OEHHA (2020)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.

Table A-ll. Chemical identity and physicochemical properties of cobalt
carbonyl

Characteristic or property

Value

Reference

Chemical structure

o+

0+^ c- ^0+

0 = C^ l2^C=0

Co

0^ C" ^0+
0+

U.S. EPA (2021)

CASRN

10210-68-1

U.S. EPA (2021)

Synonyms

dicobalt octacarbonyl

U.S. EPA (2021)

This document is a draft for review purposes only and does not constitute Agency policy.

A-9	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Characteristic or property

Value

Reference

Color/form

orange solid, white when pure

ATSDR (2004)

Molecular formula

Co2(CO)8

U.S. EPA (2021)

Molecular weight (g/mol)

341.946

U.S. EPA (2021)

Density (g/cm3)

1.73 at 18°C

ATSDR (2004)

Boiling point (°C)

decomposes at 52

ACGIH (2001b)

Melting point (°C)

51

ACGIH (2001b)

Heat of formation (kJ/mol)

-1,249.3

NISI (2021b)

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water

insoluble

ACGIH (2001b)

Vapor pressure (torr)

1.5

ACGIH (2001b)

NA = not applicable; ND = no data.

This document is a draft for review purposes only and does not constitute Agency policy.

A-10	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

A.2. ADDITIONAL COBALT COMPOUNDS USED TO SUPPORT DERIVATION
OF INHALATION UNIT RISK ESTIMATES

Table A-12. Chemical identity and physicochemical properties of cobalt
sulfate

Characteristic or property

Value3

Reference

Chemical structure

Co2+ 0_

0

0— =0

M

0

U.S. EPA (2021)

CASRN

10124-43-3

U.S. EPA (2021)

Synonyms

cobalt(ll) sulfate, cobalt
monosulfate, cobalt sulphate,
cobaltous sulfate, sulfuric acid,
cobalt (2+) salt

U.S. EPA (2021)

Color/form

red or pink solid

NCBI (2021)

Molecular formula

C0SO4

U.S. EPA (2021)

Molecular weight (g/mol)

154.99

U.S. EPA (2021)

Density (g/cm3)

3.71

NCBI (2021)

Boiling point (°C)

735 - decomposition temperature3

NCBI (2021)

Melting point (°C)

97a

NCBI (2021); U.S. EPA (2021)

Heat of formation (kJ/mol)

-888.3

NCBI (2021)

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

383

NCBI (2021)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.

a Several online databases, including PubChem and the Hazardous Substances Databank, contain conflicting data
including that 735 °C is the melting point and decomposition temperature for cobalt (II) sulfate (while also
reporting 97 °C as a melting point).

This document is a draft for review purposes only and does not constitute Agency policy.

A-ll	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)
33 Table A-13. Chemical identity and physicochemical properties of cobalt sulfate heptahydrate

Characteristic or property

Value3

Reference

Chemical structure

h2o

h2o

o h2o
u 0 II

o=s—cr

h2o 0 c°2+ h2o
h2o

U.S. EPA (2021)

CASRN

10026-24-1

U.S. EPA (2021)

Synonyms

cobalt(ll) sulfate heptahydrate;
cobalt monosulfate heptahydrate;
cobaltous sulfate heptahydrate;
sulfuric acid, cobalt(2+) salt,
heptahydrate

U.S. EPA (2021)

Color/form

pink or red crystalline solid

NCBI (2021)

Molecular formula

CoS04 x 7 H2O

NCBI (2021)

Molecular weight (g/mol)

281.09

U.S. EPA (2021)

Density (g/cm3)

1.95

NCBI (2021)

Boiling point (°C)

Becomes anhydrous at 420 (°C),
turning into cobalt sulfate15

NCBI (2021)

Melting point (°C)

NDb

NA

Heat of formation (kJ/mol)

ND

NA

Log Kow

ND

NA

Koc(L/kg)

ND

NA

Henry's law constant (atm-m3/mol)

ND

NA

Solubility in water (g/L)

604 at 3°C

NCBI (2021)

Vapor pressure (mmHg)

ND

NA

NA = not applicable; ND = no data.

a When available, average experimental values are reported from U.S. EPA (2021) Chemicals Dashboard (Cobalt sulfate
heptahydrate DTXSID7020340): https://comptox.epa.gov/dashboard/chemical/details/DTXSID7020340.
bSeveral online databases, including PubChem and the Hazardous Substances Databank, contain conflicting data
including that 735 °C is the melting point and decomposition temperature for cobalt (II) sulfate (while also reporting 97
°C as a melting point).

This document is a draft for review purposes only and does not constitute Agency policy.

A-12	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

APPENDIX B. SURVEY OF EXISTING TOXICITY
VALUES

B.l. METHODS

1	Table B-l lists websites which were searched for relevant human health reference values

2	for various compounds of cobalt, along with indications of the results of the search. In addition to

3	these sources, the ToxVal database on the Chemicals Dashboard

4	fhttps://comptox.epa.gov/dashboard/chemical lists/TOXVAL V5] was also searched for both

5	reference values and potential points of departure (PODs) for development of values.

Table B-l. Sources searched for human health reference values for cobalt and
cobalt forms

Source

Search Results

Query and/or link

ACGIH

See table of non-cancer values
in HAWC

ACGIH. 2001. 2001 TLVs and BEIs: Based on documentation of
the threshold limit values for chemical substances and physical
agents and biological exposure indices. Cincinnati, OH:
American Conference of Governmental Industrial Hygienists.

AIHA

See table of non-cancer values
in HAWC

AIHA. 2019. 2019 ERPG/WEEL Handbook. Fairfax, VA: American
Industrial Hygiene Association. [Latest list of values.]

AIHA. 2002 (and updates). 2002 Emergency Response Planning
Guidelines. Fairfax, VA: American Industrial Hygiene Association.
[Details used in deriving values.]

ATSDR

See table of non-cancer values
in HAWC

http://www. atsdr.cdc.gov/toxDrofiles/index.asD

httos://www.atsdr.cdc.gov/mrls/mrllist.asD

EPA

CompTox
Chemicals
Dashboard

See table of non-cancer values
in HAWC and Table B-2

httDs://comotox.eoa.gov/dashboard

CT DEEP

See table of non-cancer values

https://eregulations. ct.gov/eRegsPortal/Browse/getDocument?g

in HAWC

uid={00D6A654-0300-CC47-9B95-397D2AD21304}

DFG

No values found

httos://series. Dublisso.de/sites/default/files/documents/series/

mak/lmbv/Vol2021/lss2/Doc002/mbwl 2021 eng.odf

EPA/NRC
AEGL

No values found

https://www.epa.gov/aegl/access-acute-exposure-guideline-

levels-aegls-values#chemicals

Health
Canada

No values found

https://publications.gc.ca/collections/collection 2021/sc-

hc/H 129-108-2021-eng.pdf

This document is a draft for review purposes only and does not constitute Agency policy.

B-l	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Source

Search Results

Query and/or link





https://www.canada.ca/en/services/health/publications/healthy

-living,html

http://publications.gc.ca/site/archivee-

archived.html?url=http://publications.gc.ca/collections/Collectio

n/H46~2~36~134E,Ddf

HSA

See table of non-cancer values

https://www.hsa.ie/eng/publications and forms/publications/c

in HAWC

hemical and hazardous substances/chemical agents and carci



nogens code of practice 2021.html

IDEM

See table of non-cancer values
in HAWC

https://www.in.gov/idem/toxic/2343.htm

ID DEQ

24-h acceptable ambient
concentrations for cobalt
(0.0025 mg/m3), cobalt
carbonyl, and cobalt
hydrocarbonyl (0.005 mg/m3)

https://adminrules.idaho.gov/rules/current/58/580101.pdf



IFA

See table of non-cancer values
in HAWC

https://limitvalue.ifa.dguv.de/WebForm gw2,aspx

IRIS

No values found

http://www.epa.gov/iris/

JSOH

No values found

https://www.sanei.or. ip/?mode=view&cid=328

MassDEP

No values found

https://www.mass.gov/service-details/massdep-ambient-air-

toxics-guidelines

MDH

No values found

https://www.health.state.mn.us/communities/environment/risk

/guidance/air/table, html

MlEGLE

See table of non-cancer values

https://www.michigan.gov/documents/dea/dea-rrd-chem-

in HAWC

CleanupCriteri pdf

NATICH

Compendium of state values
based on prior occupational
exposure limits, last updated in
1993

https://nepis.epa.gov/Exe/ZyPDF.cgi/2000NS7S. PDF?Dockey=200

0NS7S.PDF

NC DEQ

No values found

https://files.nc.gov/ncdea/Air%20Quality/rules/rules/D1104.pdf



NDEP

See table of non-cancer values

https://ndep.nv.gov/resources/risk-assessment-and-toxicology-

in HAWC and Table B-2

basic-comparison-levels

NIOSH

See table of non-cancer values
in HAWC

http://www.cdc.gov/niosh/npg/npgdcas.html

https://www.cdc.gov/niosh/pubs/criteria date desc nopubnum

bers.html

https://www.cdc.gov/niosh/idlh/intridl4.html

NYSDEC

No values found

https://www.dec.ny.gov/docs/remediation hudson pdf/techsup

pdoc.pdf

OAQPS

No unique results

https://www.epa.gov/fera/dose-response-assessment-assessing-

health-risks-associated-exposure-hazardous-air-pollutants

This document is a draft for review purposes only and does not constitute Agency policy.

B-2	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Source

Search Results

Query and/or link

OEHHA

See table of non-cancer values
in HAWC and Table B-2

http://www.oehha.ca.gov/tcdb/index.asp

https://oehha.ca.gov/air

Ontario MOL

See table of non-cancer values

https://www.labour.gov.on.ca/english/hs/pubs/oel table.php

in HAWC



ORDEQ

See table of non-cancer values
in HAWC

https://www.oregon.gov/deq/FilterDocs/airtox-abc.pdf

OS HA

See table of non-cancer values
in HAWC

https://www.osha.gov/chemicaldata/

PAC

Database

See table of non-cancer values
in HAWC

https://edms.energy.gov/pac/Search

PPRTV

See table of non-cancer values

https://www.epa.gov/pprtv/provisional-peer-reviewed-toxicity-

in HAWC and Table B-2



Publications
Quebec

See table of non-cancer values
in HAWC

http://legisquebec.gouv.qc.ca/en/showdoc/cr/S-
2.1,%20r,%2013?csi scan 9222d36c6a354dc6=BO9xyrMZ+270U



P3iOMGuODOkZjgFAAAAXrM3HA==&bcsi scan filename=S-

2.l,%20r.%2013&bcsi scan 3222d36c6a3S4dc6=KXzmpPueuN0L

lAjnJOBlZerr85YMAAAAyhrPTg==&bcsi scan filename=S-

20r.%2013

Rl DEM

See table of non-cancer values

http://www.dem.ri.gov/programs/benviron/air/pdf/airtoxgl.pdf

in HAWC



RIVM

No values found

https://www.rivm.nl/bibliotheek/rapporten/711701092.pdf

https://www.rivm.nl/bibliotheek/rapporten/609021044.pdf



See table of non-cancer values

https://www.rivm.nl/bibliotheek/rapporten/711701025.pdf

in HAWC



Safe Work
Australia

See table of non-cancer values
in HAWC

https://www.safeworkaustralia.gov.au/exposure-

standards#exposure-standards-in-australia

SWCAA

24-h acceptable source impact
levels for cobalt metal
(0.00017 mg/m3), cobalt
carbonyl, and cobalt
hydrocarbonyl (0.00033
mg/m3)

http://www.swcleanair.org

TCEQ

See table of non-cancer values
in HAWC and Table B-2

https://www.tcea.texas.gov/toxicology/dsd/final

https://www.tcea.texas.gov/remediation/trrp/trrppcls.html



USAPHC

Critical, marginal, and
negligible military exposure
guidelines based on other
agencies' values

https://phc.amedd.army.miI/topics/envirohealth/hrasm/Pages/T

G230,aspx

VT DEC

See table of non-cancer values
in HAWC

https://dec.vermont.gov/sites/dec/files/aac/laws-

regs/documents/AQCD%20Regulations%20ADOPTED Decl3201



8,pdf#page=127

This document is a draft for review purposes only and does not constitute Agency policy.

B-3	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Source

Search Results

Query and/or link

WA State
Dept. of
Ecology

24-h acceptable source impact
level of 0.0001 mg/m3

Iittpsi//apps,Ieg,wa,gow/WAC/defauIt,aspx?cite=173-460-lS0



Worksafe

See table of non-cancer values

https://worksafe.govt.nz/topic-and-industrv/work-related-

in HAWC

health/monitoring/exposure-standards-and-biologica (-exposure-



indices/

ACGIH = American Conference of Governmental Industrial Hygienists; AEGL = Acute Exposure Guideline Levels;
AIHA = American Industrial Hygiene Association; ATSDR = Agency for Toxic Substances and Disease Registry; BEI =
biological exposure index; CT DEEP = Connecticut Department of Energy & Environmental Protection; DFG =
Deutsche Forschungsgemeinschaft, German Research Foundation; EPA = Environmental Protection Agency; ERPG
= Emergency Response Planning Guideline; HSA = Health and Safety Authority; IDEM = Indiana Department of
Environmental Management; ID DEQ= Idaho Department of Environmental Quality; I FA = Institutfur
Arbeitsschutz, The Institute for Occupational Safety and Health; IRIS = Integrated Risk Information System; JSOH =
Japan Society for Occupational Health; MassDEP = Massachusetts Department of Environmental Protection; MDH
= Minnesota Department of Health; Ml EGLE = Michigan Environment, Great Lakes & Energy; MOL = Ministry of
Labour; NATICH = National Air Toxics Information Clearinghouse; NC DEQ = North Carolina Department of
Environmental Quality; NDEP = Nevada Division of Environmental Protection; NIOSH = National Institute for
Occupational Safety and Health; NRC = National Research Council; NYSDEC = New York State Department of
Environmental Conservation; OAQPS = Office of Air Quality Planning and Standards; OEHHA = California Office of
Environmental Health Hazard Assessment; OR DEQ = Oregon Department of Environmental Quality; OSHA =
Occupational Safety and Health Administration; PAC = Protective Action Criteria; PPRTV = Provisional Peer-
Reviewed Toxicity Value; Rl DEM = Rhode Island Department of Environmental Management; RIVM =
Rijksinstituut voor Volksgezondheid en Milieu, The Netherlands Institute for Public Health and the Environment;
SWCAA = Southwest Clean Air Association; TCEQ = Texas Commission on Environmental Quality; TERA -
Toxicology Excellence for Risk Assessment; TLV = threshold limit value; USAPHC = United States Army Public
Health Center; VT DEC = Vermont Department of Environmental Conservation; WEEL = Workplace Environmental
Exposure Level.

This document is a draft for review purposes only and does not constitute Agency policy.

B-4	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

B.2. SUMMARY OF EXISTING TOXICITY VALUES

1	A summary of inhalation reference values and cancer risk ranges is presented in Figure B-l.

2	Details on the derivation of the inhalation cancer toxicity values are presented in Table B-2. Details

3	on the available non-cancer values displayed in Figure B-l can be found in HAWC. see "Non-cancer

4	reference values for inhalation exposure to cobalt and compounds" under "Attachments."

This document is a draft for review purposes only and does not constitute Agency policy.

B-5	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Cobalt Inhalation Reference Values and Cancer Risk Estimates

February 2022

10,000.0
1,000.0
100.0
10.0
1.0
0.1
0.01
0.001
0.0001
0.00001
0.000001
0.0000001
0.00000001

£
O)

E,

6

E

o
o

ACUTE

Short Term

0)
>
n
a

~ PAC-3 CoC03
PAC-3 Co(N03)2 x 6 H20

"N. /v- PAC-2 C0CO3

PAC-3 Co(N03)2

-	PAC-3 CoBr2 PAC.3 Cob|2
NIOSH IDLH* r PAC-3 CoO

Jk- PAC-3 CO3O4
¦!— paC-3 Co

-	PAC-2 CoCI, -W- PAC-3 CBCo20B
PAC-2 CofNOsfcAL PAC-2 CofNOjkx 6H20

HAC-2 CoBr2 	p*p 9 p ^

PAC-2 CoO -4R- "AC-2 C03O4
PAC-2 C8Co208/^'~ ERPG-3 C4CoO,

~ PAC-2 Co -'&¦	C4Co04

PAC-1 Co(N03)2 x 6 H20
PAC-1 CsCOzOs <"> PAC-1 C4CoO„

PAC-1 Co(N03)2* PAC-1 CoBr2

PAC-1 C0CO3 Z PAC-1 Co 0> I				

PAC-1 C03O; -T- PAC-1 CoCI2 O —4	

PAC-1 CoO	^	|

ra	ACGIH-TLV (TWA)* inorganic Co

Subchronic

Chronic

OSHA-PEL Cal/OSHA-PEL (TWA)*
(TWA)* C8Co208 & C4C0O4

NIOSH-REL (TWA)*
C8Co208 & C4C0O4

ACGIH-TLV (TWA)*
C8Co208 & C4C0O4

| Avg. of Other State Values^.

TX-ReV (1 hr)

-0

NIOSH-REL (TWA)* inorganic Co |

Cal/OSHA-PEL (TWA)* inorganic Co

I

OEHHA Cancer Range soluble Cov
RIVM TCA



TX-ReV (24 hr)

| Avg. of Other State Values ] t

ATSDR-MRL (> 1yr)

PPRTV p-RfC (Subchronic)

TX-ReV (Chronic)

Tf"

PPRTV p-RfC (Chronic)

TCEQ Cancer Range

| Avg. of Other State Values |— ~

PPRTV Cancer Range
OEHHA Cancer Range insoluble Co

r|

ERPG-3

ERPG-2

PAC-3

PAC-2

PAC-1

NIOSH IDLH*

NIOSH-REL (TWA)*

ACGIH-TLV (TWA)*

OSHA-PEL (TWA)*

Cal/OSHA-PEL (TWA)*

TX-ReV (1 hr)

TX-ReV (24 hr)

-	TX-ReV (Chronic)

ATSDR-MRL (> 1yr)

-	PPRTV p-RfC (Subchronic)

-	PPRTV p-RfC (Chronic)

-	RIVM TCA

Avg. of Other State Values
PPRTV Cancer Range
TCEQ Cancer Range
OEHHA Cancer Range insoluble Co
OEHHA Cancer Range soluble Co

O	<"

c	£

®	Q

2 a

®	VI

F	®

IN	K

<0

a

D

O

o
O

n
3
Q-

~tc

i—

a?
c

<13

o

10	100	1,000

Duration (hours)

10,000 100,000 1,000,000

Indicates an occupational value; expert judgement necessary prior to applying these values to the general public.

Figure B-l. Available noncancer and cancer toxicity values for inhalation exposure to cobalt.

This document is a draft for review purposes only and does not constitute Agency policy,
B-6	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table B-2. Details on the derivation of existing inhalation cancer toxicity values for cobalt and cobalt compounds

Toxicity

Cobalt Form(s)

Toxicity

Health Effect

Point of

Qualifier

Source

Notes on

Review

Value Name

Value

Departure

Derivation

Status

NDEP BCL

Cobalt

3.12 x 10"7

Cancer

9 (mg/m3)1

PPRTV IUR

U.S. EPA

Calculated3

Final





mg/m3







(2008)



NDEP

















(2017)

PPRTV IUR

Soluble cobalt

9 (mg/m3)1

Alveolar/bronchiolar

0.3 mg/m3

NOAEL

Bucheret

Duration

Provisional



sulfate



adenomas and





al. (19991

adjusted,

U.S. EPA



hexahydrate,



carcinomas in female

0.012 mg Co/m3

NOAELadj

and NIP

MW adjustment15

(2008)



applied to



rats exposed to cobalt





(1998)







additional



sulfate hexahydrate

0.0095 mg Co/m3

NOAELhec



HEC adjusted0





compounds























0.011 mg Co/m3

BMDL







OEHHAIUR

Cobalt metal and

7.7

Alveolar/bronchiolar

1.25 mg/m3

NOAEL

NIP(2014)

Duration

Final



water-insoluble

(mg/m3)1

adenomas and







adjusted:

OEHHA



compounds



carcinomas in male

0.23 mg/m3

NOAELadj



(6.2-h/24-h) x

(2020)







mice exposed to







(5-d/7-d)









cobalt metal

0.26 mg/kg-d

ADD





















ADD adjustedd











0.01122 mg/kg-d

BMDLos





















CSFa = 0.05 -f











4.46 (mg/kg-d)1

CS Fa



BMDLos











27 (mg/kg-d)1

CSFh



CSFh calculated6

















IUR calculated'



This document is a draft for review purposes only and does not constitute Agency policy.

B-7	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Toxicity
Value Name

Cobalt Form(s)

Toxicity
Value

Health Effect

Point of
Departure

Qualifier

Source

Notes on
Derivation

Review
Status



Water-soluble
cobalt compounds

0.86

(mg/m3)1

Lung and adrenal
tumors in female rats
exposed to
aerosolized cobalt
sulfate

0.01504 mg/kg-d
3.32 (mg/kg-d)1
13.41 (mg/kg-d)1
3.0 (mg Co/kg-d)1

BMDLos
CS Fa
CSFh

MW-adjusted
CSF

NIP (1998)

CSFa = 0.05 -f
BMDLos

CSFh calculated8
MW adjustedh
III R calculated1



TCEQIUR

Cobalt compounds

6 (mg/m3)1

Alveolar/bronchiolar
adenomas and
carcinomas in female
rats exposed to cobalt
sulfate hexahydrate

0.3 mg/m3
0.012 mg Co/m3
0.0095 mg Co/m3
0.011 mg Co/m3

NOAEL
NOAELadj
NOAELhec
BMDLio

NIP (1998)
and U.S.
EPA (2008)

Duration
adjusted),
MW adjustment

HEC adjustedk

Calculated1

Final

TCEQ
(2017)







Alveolar/bronchiolar
adenomas and
carcinomas in female
rats exposed to
aerosolized cobalt
metal

1.25 mg/m3
0.223 mg/m3
0.132 mg/m3
0.108 mg/m3

LOAEL
LOAELadj
LOAELhec
BMDL

NIP(2014)
and Suli et
al. (2016)





This document is a draft for review purposes only and does not constitute Agency policy.

B-8	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Toxicity
Value Name

Cobalt Form(s)

Toxicity
Value

Health Effect

Point of
Departure

Qualifier

Source

Notes on
Derivation

Review
Status

ADD = average daily dose; ADJ = adjusted; AT = averaging time; BCL = basic comparison level; BMDL = benchmark dose level; BWa = animal body weight; BWh
= human body weight; Co = cobalt; C0SO4X 6 H2O = cobalt sulfate hexahydrate; CSFa = animal cancer slope factor; CSFh = human cancer slope factor; ED =
exposure duration; EF = exposure frequency; EPA = Environmental Protection Agency; ET = exposure time; HEC = human equivalent concentration; IR =
inhalation rate; IIIR = inhalation unit risk; LOAEL = lowest-observed-adverse-effect level; MW = molecular weight; NDEP = Nevada Division of Environmental
Protection; NOAEL = no-observed-adverse-effect level; NTP = National Toxicology Program; OEHHA = California Office of Environmental Health Hazard
Assessment; PPRTV = Provisional Peer-Reviewed Toxicity Value; RDDR = regional deposited dose ratio; TCEQ = Texas Commission on Environmental Quality;
TR = target risk; URF = unit risk factor.

a BCL = TR x AT 4 (ET x EF x ED x URF) = (106 x 70 y x 365 d/y x 24 h/d) 4 [24 h/d x 350 d/y x 26 y x 9 (nig/m3)"1] = 3.12 x 10"7 mg/m3.
b NOAELadj = NOAEL x (6 h 4 24 h) x (5 d 4 7 d) x [Co atomic mass 4 (C0SO4x 6 H20) MW] = 0.3 mg/m3 x (6-h 4 24-h) x (5-d 4 7-d) x (58.933 g/mol 4 263.08
g/mol) = 0.012 mg Co/m3.

c NOAELhec = NOAEUdj x RDDR = 0.012 mg Co/m3 x 0.79 = 0.0095 mg Co/m3.

d ADD = 0.0345 m3/d x(BWt 0.025 kg)2/3 x NOAELadj 4- BW = 0.0345 m3/d x (0.0485 kg 4 0.025 kg)2/3 x 0.23 mg/m3 4 0.0485 kg = 0.26 mg/kg-d.
e CSFh = CSFa x (BWh 4 BWa)1/4 = 4.46 (mg/kg-d)1 x (70 kg 4 0.0485 kg)1/4 = 27 (mg/kg-d)-1.
f IUR = CSFh x IR 4 BW = 27 (mg/kg-d)^x 20 m3/d 4 70 kg = 7.7 (mg/m3) ^
sCSFh = CSFa x (BWh 4 BWa)1/4 = 3.32 (mg/kg-d)1 x (70 kg 4 0.2633 kg)1/4= 13.41 (mg/kg-d)-1.

h MW-adjusted CSF = CSFh x [Co atomic mass 4 (C0SO4X 6 H2O) MW] = 13.41 (mg/kg-d)-1 x (58.9 g/mol 4 263.1 g/mol) = 3.0 (mg Co/kg-d)1.

' IIIR = CSFx IR 4 BW = 3.0 (mg Co/kg-d^x 20 m3/d 4 70 kg = 0.86 (mg Co/m3)1.

J NOAELadj = NOAEL x (6 h 4 24 h) x (5 d 4 7 d) x [Co atomic mass 4 (C0SO4x 6 HzO) MW] = 0.3 mg/m3 x (6-h 4 24-h) x (5-d 4 7-d) x (58.933 g/mol 4 263.08
g/mol) = 0.012 mg Co/m3.

k NOAELhec = NOAEUdj x RDDR = 0.012 mg Co/m3 x 0.79 = 0.0095 mg Co/m3.

LOAELhec = LOAELadj x RDDR = 0.223 mg Co/m3 x 0.592 = 0.132 mg Co/m3.

1 NTP 1998 IUR = 0.1 4 BMDLio= 0.140.011 mg/m3 =9.1 (mg/m3)"1.

NTP 2014 IUR = 0.32 4 BMDL = 0.32 4 0.108 mg/m3 = 3 (mg/m3)1.

The two derived lURs were averaged to arrive at the final value: [9.1 (mg/m3) 1+ 3 (mg/m3)"1] 4-2 = 6 mg/m3.

This document is a draft for review purposes only and does not constitute Agency policy.

B-9	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

APPENDIX C. SYSTEMATIC EVIDENCE MAP

C.l. SYSTEMATIC EVIDENCE MAP (SEM) SPECIFIC AIMS

•	Develop a systematic evidence map (SEM) to identify epidemiological (i.e., human) and
toxicological (i.e., experimental animal) literature that report reporting effects of inhalation
exposure to cobalt or cobalt compounds and cancer.

o The SEM includes searches for studies published since the October 2020 inhalation
'unit risk estimates' (URE) or 'inhalation unit risk' (IUR) developed by California
EPA OEHHA (2020). The SEM also includes a survey of prior assessments U.S. EPA
("20081: OEHHA C20191: OEHHA C20201: TCEO C20171: NTP C20161: ATSDR C20041
to ensure consideration of studies cited to develop cancer hazard conclusions or
develop inhalation unit risk estimates.8

•	Evaluate studies that meet SEM PECO criteria to identify studies most suitable for deriving
an inhalation unit risk (IUR) for water-soluble and water-insoluble compounds of cobalt
Prioritized studies from this evaluation are those that appear at least as suitable for IUR
derivation as the NTP rodent cancer bioassays NTP (2014.1998) used in prior assessments
U.S. EPA (~2Q08"):OEHHA f20191:QEHHA f20201:TCEQ f20171.

•	Conduct study evaluation (evaluating risk of bias and sensitivity) and data extraction for
prioritized epidemiological and toxicological studies.

•	Identify supplemental material in the literature published since October 2020 or cited in the
prior assessments listed above that may potentially inform dose-response analysis, clarify
what is known currently about the cancer mode of action, inform conclusions on potential
susceptibility, or help elucidate key science issues. Supplemental material content includes
mechanistic in vitro, in vivo, ex vivo, or in silico studies; toxicokinetic and absorption,
distribution, metabolism, and excretion (ADME) studies; pharmacokinetic (PK) or
physiologically based pharmacokinetic (PBPK) model studies; studies using non-inhalation
route of exposure; non-mammalian model systems; exposure assessment studies with no
health outcomes reported; mixture studies; human case studies and case reports; animal
cancer studies using less than subchronic duration exposures; studies or reports with no
original data; and conference/symposium abstracts or poster presentations, and studies
assessing noncancer health outcomes. Studies considered PECO-relevant that also contain
supplemental information are tagged as such.

8 The full 2022 IARC Monograph on" Carcinogenicity of cobalt, antimony compounds, and weapons-grade
tungsten alloy" was not publicly released at the time of preparing this SEM but will be surveyed for any
missing citations when it becomes available.

This document is a draft for review purposes only and does not constitute Agency policy.

C-l	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

C.2. POPULATIONS, EXPOSURES, COMPARATORS, AND OUTCOMES
(PECO) CRITERIA AND SUPPLEMENTAL MATERIAL TAGGING

1	PECO criteria are used to focus the research question(s), search terms, and

2	inclusion/exclusion criteria used in a SEM or systematic review. The SEM PECO criteria are

3	presented in Table C-l. In addition, studies containing supplemental material are inventoried

4	during the literature screening process using the categories presented in Table C-2.

This document is a draft for review purposes only and does not constitute Agency policy.

C-2	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table C-l. Example Populations, Exposures, Comparators, and Outcomes (PECO) Criteria





Populations

Human: Any population and lifestage (occupational or general population, including in pregnant women, infants, children, adolescents and
adults).

Animal: Nonhuman mammalian animal species (whole organism) of any lifestage (including fetal, early postnatal, adolescents and adults).

Studies of transgenic animals are tracked as mechanistic studies under "potentially relevant supplemental material".

Note: Studies meeting PECO criteria may also contain information on susceptible populations. When this occurs, these studies are also tagged

as having information pertinent to susceptible populations. This typically happens during preparation of the literature inventory or full text

extraction.

Exposures

Relevant forms for Clean Air Act: cobalt aluminate (1345-16-0), cobalt bromide (7789-43-7), cobalt carbonate (513-79-1), cobalt
carbonyl (10210-68-1), cobalt chloride (7646-79-9), cobalt (7440-48-4), cobalt hydrocarbonyl (16842-03-8), cobalt naphtha (61789-51-
3), cobalt nitrate (10141-05-6), cobalt oxide (1307-96-6), cobalt oxide (II, III) (1308-06-1), andhexanoic acid, 2-ethyl-, cobalt(2+) salt
(136-52-7). Many of these compounds do not have cancer toxicity information, thus other water-soluble and water-insoluble cobalt
compounds that do have inhalation cancer evidence are included within the scope of this review, e.g., cobalt sulfate, cobalt hydroxide, and
cobalt sulfide. Radioactive isotopes (i.e., 60Co) and vitamin B12 are considered out of scope.

Human: Anv quantitative exposure to cobalt via the inhalation route, aside from acute or verv short fdavsl duration. Studies of
developmental exposure are also included. Studies will also be included if biomarkers of exposure are evaluated (e.g., measured compound
or metabolite levels in tissues or bodily fluids) and the exposure route can be inferred as primarily inhalation.

Animal: Anv quantitative exposure to cobalt via the inhalation route for anv subchronic and chronic exposure duration. Studies of
developmental exposure are also included. Studies involving exposures to mixtures will be included only if they include exposure to a
relevant form of cobalt alone. Non-inhalation routes, including oral, dermal or intravenous, are tracked as "potentially relevant
supplemental information."

Comparators

Human: Referent populations exposed to lower fwithin the studvl levels of cobalt. The results of the comparisons must be presented with
sufficient detail of quantitative modeling (e.g., regression coefficients presented with statistical measure of variation). Case reports
describing findings in 1-3 people are tagged as "potentially relevant supplemental information."

Animal: A concurrent control sroup exposed to vehicle-onlv treatment and/or untreated control.

Outcomes

Any cancer-related effect on any system.

This document is a draft for review purposes only and does not constitute Agency policy.

C-3	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table C-2. Categories of Potentially Relevant Supplemental Material

Category (Tag)

Description

Typical Assessment Use

Pharmacokinetics Data Potentially Informative to Assessment Analyses

Classical pharmacokinetic (PK) or
physiologically based
pharmacokinetic (PBPK) model
studies

Classical Pharmacokinetic or Dosimetry Model Studies: Classical PK or dosimetry
modeling usually divides the body into just one or two compartments, which are not
specified by physiology, where movement of a chemical into, between, and out of
the compartments is quantified empirically by fitting model parameters to ADME
(absorption, distribution, metabolism, and excretion) data. This category is for
papers that provide detailed descriptions of PK models but are not PBPK models.
The data are typically the concentration time-course in blood or plasma after oral
and or intravenous exposure, but other exposure routes can be described.
Physiologically Based Pharmacokinetic or Mechanistic Dosimetry Model Studies:
PBPK models represent the body as various compartments (e.g., liver, lung, slowly
perfused tissue, richly perfused tissue) to quantify the movement of chemicals or
particles into and out of the body (compartments) by defined routes of exposure,
metabolism, and elimination, and thereby estimate concentrations in blood or target
tissues.

A defining characteristic is that key parameters are determined from a substance's
physicochemical parameters (e.g., particle size and distribution, octanol-water
partition coefficient) and physiological parameters (e.g., ventilation rate, tissue
volumes).

PBPK and PK model studies are included
in the assessment and evaluated for
possible use in conducting quantitative
extrapolations. PBPK/PK models are
categorized as supplemental material
with the expectation that each one will be
evaluated for applicability to address
assessment extrapolation needs and
technical conduct. Specialized expertise
is required for their evaluation.

Standard operating procedures for
PBPK/PK model evaluation and the
identification, organization, and
evaluation of ADME studies are outlined
in An umbrella Quality Assurance Project
Plan (QAPP) for PBPK models U.S. EPA
(2018).

Pharmacokinetic (ADME)

Pharmacokinetic (ADME) studies are primarily controlled experiments, where defined
exposures usually occur by intravenous, oral, inhalation, or dermal routes, and the
concentration of particles, a chemical, or its metabolites in blood or serum, other
body tissues, or excreta are then measured.

These data are used to estimate the amount absorbed (A), distributed (D),
metabolized (M), and/or excreted (E).

ADME data can also be collected from human subjects who have had environmental
or workplace exposures that are not quantified or fully defined.

ADME data, especially metabolism and tissue partition coefficient information, can
be generated using in vitro model systems. Although in vitro data may not be as
definitive as in vivo data, these studies should also be tracked as ADME. For large
evidence bases it may be appropriate to separately track the in vitro ADME studies.

ADME studies are inventoried and
prioritized for possible inclusion in an
ADME synthesis section on the chemical's
PK properties and for conducting
quantitative adjustments or
extrapolations (e.g., animal-to-human).
Specialized expertise in PK is necessary
for inventory and prioritization.

Standard operating procedures for
PBPK/PK model evaluation and the
identification, organization, and
evaluation of ADME studies are outlined
in An umbrella Quality Assurance Project

This document is a draft for review purposes only and does not constitute Agency policy.

C-4	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Category (Tag)

Description

Typical Assessment Use



*Studies describing environmental fate and transport or metabolism in bacteria or
model systems that are not applicable to humans or animals should not be tagged.

Plan (QAPP) for PBPK models U.S. EPA
(2018).

Supplemental Evidence Potentially Informative to Assessment Analyses

Mechanistic endpoints

Studies that do not meet PECO criteria but report measurements that inform the
biological or chemical events associated with phenotypic effects related to a health
outcome. Experimental design may include in vitro, in vivo (by various routes of
exposure; includes all transgenic models), ex vivo, and in silico studies in mammalian
and nonmammalian model systems. Studies using New Approach Methodologies
(NAMs, e.g., in vitro high throughput testing strategies, read across applications) are
also categorized here. Studies where the chemical is used as a laboratory reagent
(e.g., as a chemical probe used to measure antibody response) generally should not
be tagged.

Mechanistic evidence can also help identify factors contributing to susceptibility;
these studies should also be tagged "susceptible populations."

[Notes: During screening, especially at the title and abstract (TIAB) level, it may not
be readily apparent for studies that meet P, E, and C criteria if the endpoint(s) in a
study are best classified as phenotypic or mechanistic with respect to the 0 criteria. In
these cases, the study should be screened as "unclear" during TIAB screening, and a
determination made based on full-text review (in consultation with a content expert
as needed). Full-text retrieval is performed for studies of transgenic model systems
that meet E and C criteria to determine if they include phenotypic information in
wildtype animals that meet P and 0 criteria that is not reported in the abstract.]

Prioritized studies of mechanistic
endpoints are described in the
mechanistic synthesis sections; subsets of
the most informative studies may become
part of the units of analysis. Mechanistic
evidence can provide support for the
relevance of animal effects to humans
and biological plausibility for evidence
integration judgments (including MOA
analyses, e.g., using the MOA framework
in the US EPA Cancer Guidelines 2005a)).

Non-PECO animal model

Studies that report outcomes in animal models that meet the outcome criteria but do
not meet the population criteria in the PECO.

Depending on the endpoints measured in these studies, they can also provide
mechanistic information (in these cases studies should also be tagged "mechanistic
endpoints").

This categorization generally does not apply to studies that use species with limited
human health relevance (e.g., ecotoxicity-focused studies are typically excluded).

Studies of non-PECO animals, exposures,
or durations can be summarized to inform
evaluations of consistency (e.g., across
species or routes or durations),
coherence, or adversity; subsets of the
most informative studies may be included
in the unit of analysis. These studies may
also be used to inform evidence
integration judgments of biological
plausibility and/or MOA analyses and thus

Non-PECO route of exposure

Epidemiological or animal studies that use a non-PECO route of exposure, e.g.,
injection studies or dermal studies if the dermal route is not part of the exposure
criteria.

This document is a draft for review purposes only and does not constitute Agency policy.

C-5	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Category (Tag)

Description

Typical Assessment Use



This categorization generally does not apply to epidemiological studies where the
exposure route is unclear; such studies are considered to meet PECO criteria if the
relevant route(s) of exposure are plausible, with exposure being more thoroughly
evaluated at later steps.

may be summarized as part of the
mechanistic evidence synthesis.

Acute or short-term duration
exposures

Given the focus on cancer, acute exposure durations (defined as animal studies of
<1 d) or short-term (defined as animal studies of <90 d/13 weeks) are considered
supplemental.

Susceptible populations

Studies that help to identify potentially susceptible subgroups, including studies on
the influence of intrinsic factors such as sex, lifestage, or genotype to toxicity, as well
as some other factors (e.g., health status). These are often co-tagged with other
supplemental material categories, such as mechanistic or ADME. Studies meeting
PECO criteria that also address susceptibility should be co-tagged as supplemental.
*Susceptibility based on most extrinsic factors, such as increased risk for exposure due
to residential proximity to exposure sources, is not considered an indicator of
susceptible populations for the purposes of IRIS assessments.

Provides information on factors that
might predispose sensitive populations or
lifestages to a higher risk of adverse
health effects following exposure to the
chemical. This information is summarized
during evidence integration for each
health effect and is considered during
dose-response, where it can directly
impact modeling decisions.

Background Information Potentially Useful to Problem Formulation and Protocol Development

(These studies fall outside the scope of IRIS assessment analyses)

Human exposure and
biomonitoring (no health
outcome)

Information regarding exposure monitoring methods and reporting that are
unrelated to health outcomes, but which provide information on the following:
methods for measuring human exposure, biomonitoring (e.g., detection of chemical
in blood, urine, hair), defining exposure sources, or modeled estimates of exposure
(e.g., in occupational settings). Studies that compare exposure levels to a reference
value, risk threshold or assessment points of departure are also included in this
category. Studies related to environmental fate and transport are typically tagged as
background materials unless otherwise described in the assessment-specific
protocol.

* Assessment teams may want to subtag studies that describe or predict exposure
levels versus those that present exposure assessment methods.

This information may be useful for
developing exposure criteria for study
evaluation or refining problem
formulation decisions.

Notably, providing an assessment of
typical human exposures (e.g., sources,
levels) falls outside the scope of an IRIS
assessment.

Mixture study

Mixture studies use methods that do not allow investigation of the health effects of
exposure to the chemical of interest by itself (e.g., animal studies that lack exposure

Mixture studies are tracked to help
inform cumulative risk analyses, which
may provide useful context for risk

This document is a draft for review purposes only and does not constitute Agency policy.

C-6	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Category (Tag)

Description

Typical Assessment Use



to chemical of interest alone or epidemiology studies that do not evaluate
associations of the chemical of interest with relevant health outcome(s)).

* Methods used to assess investigation of the exposure by itself may not be clear
from the abstract, in particular for epidemiology studies. When unclear, the study is
advanced to full-text review to determine eligibility.

assessment but fall outside the scope of
an IRIS assessment.

Case reports or case series

Human studies that present an investigation of a single exposed individual or group
of < 3 subjects that describe health outcomes after exposure but lack a comparison
group (i.e., do not meet the "C" in the PECO) and typically do not include reliable
exposure estimates.

Tracking case studies can facilitate
awareness of potential human health
issues missed by other types of studies
during problem formulation.

Noncancer health outcomes

Studies assessing noncancer health outcomes.

Out of scope for the assessment but
tracked to facilitate any assessment work
conducted by others in understanding
potential non-cancer health publication
trends.

Reference Materials

Records with no original data

Records that do not contain original data, such as other agency assessments,
informative scientific literature reviews, editorials, or commentaries.

Studies that are tracked for potential use
in identifying missing studies, background
information, or current scientific opinions
(e.g., hypothesized MOAs).

Posters or conference abstracts

Records that do not contain sufficient documentation to support study evaluation
and data extraction.

This document is a draft for review purposes only and does not constitute Agency policy.

C-7	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

C.3. METHODS: LITERATURE SEARCH STRATEGIES

C.3.1.Database Search Term Development

The literature search focused on the chemical name (and synonyms, trade names, and
metabolites/degradants of interest) and was date limited to studies published after 2019
(Addendum 1). The literature search was completed on December 16, 2021. This date was selected
to cover new studies published since the 2020 CalEPA cobalt assessment OEHHA (20201. which is
the most recent US Federal or State assessment conducted. No language restrictions were applied.
Chemical synonyms were identified by using the "Find Chemical Synonyms" feature in SWIFT
(Sciome Workbench for Interactive computer-Facilitated Text-mining) Review Howard et al.
(20161. In brief, this feature automatically creates a PubMed-formatted chemical search using
(1) the common name for the chemical as presented in the Tox21 chemical inventory list U.S. EPA
(2020dl: (2) the Chemical Abstract Services Registry Number (CASRN); (3) synonyms from the
ChemlDPlus database, which currently contains chemical names and synonyms for over 400,000
chemicals; and (4) removal of ambiguous or short alphanumeric terms that could lead to false
positives. This search is manually reviewed to ensure that any synonyms listed in EPA's Dashboard
U.S. EPA (20211 as "valid" or "good" are included. The PubMed search created from SWIFT Review,
along with additional synonyms identified from EPA's Dashboard, is shared with EPA information
specialists to develop search strategies tailored for each of the databases below, as each database
has its own search architecture. Full details of the search strategy for each database are presented
in the Addendum 1.

C.3.2.Database Searches

The databases listed below are searched by an EPA information specialist Retrieved
references are imported into the EPA's Health and Environmental Research Online (HERO)
database and undergo a round of deduplication in HERO9.

•	Web of Science (Thomson Reuters)

•	PubMed (National Library of Medicine)

The literature search is updated throughout SEM development. In addition to the databases
listed below, a variety of other resources are subsequently searched using customized processes
(see "Other Resources"). One process described in "Other Resources" is to review prior

9 Deduplication in HERO involves first determining whether a matching unique ID exists (e.g., PMID, WoSID,
or DOI). If one matches one that already exists in HERO, HERO will tag the existing reference instead of
adding the reference again. Second, HERO checks if the same journal, volume, issue and page number are
already in HERO. Third, HERO matches on the title, year, and first author. Title comparisons ignore
punctuation and case

This document is a draft for review purposes only and does not constitute Agency policy.

C-8	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

assessments of cobalt carcinogenicity to identify studies meeting the current SEM PECO criteria
that would have been missed by the date limited database search described above.

The unique studies are imported into SWIFT Review software Howard et al. f20161 to
identify those references most likely to be applicable to a human health risk assessment In brief,
SWIFT Review has pre-set literature search strategies ("filters") developed by information
specialists that can be applied to identify studies that are more likely to be useful for identifying
human health content from those that likely do not (e.g., environmental fate). The filters function
like a typical search strategy where studies are tagged as belonging to a certain filter if the terms in
the filter literature search strategy appear in title, abstract, keyword or Medical Subject Headings
[MeSH) fields content The details of the search strategies that underlie the filters are available
online. For this SEM, filters for human, animal (human health models) and in vitro evidence were
used. Studies not retrieved using the search strategies are not considered further. Studies that
include one or more of the search terms in the title, abstract, keyword, or MeSH fields are exported
as a Research information Systems (RIS) file for uploading into the screening software as described
below in "Screening Process." Application of the SWIFT Review evidence stream filters to the initial
search results (12/16/2021) reduced the number of studies for title and abstract screening from
29,833 to 4,589.

C.3.3.Searching Other Resources

The literature search strategies described above are designed to be broad, but like any
search strategy, studies may be missed (e.g., cases where the specific chemical is not mentioned in
title, abstract, or keyword content; ability to capture "gray" literature that is not indexed in the
databases listed above). Thus, in addition to the database searches, the sources below are used to
identify studies that may have been missed based on the database search. References that appear to
meet the PECO criteria are uploaded into the screening software, annotated with respect to source
of the record, and screened according to PECO as described below. Searching of these sources is
summarized to include the source type or name, the search string (when applicable), the URL
(when available and applicable), number of results, and number of unique references not otherwise
identified from database searching (Addendum 2).

• For studies screened as 'included' based on full text review, manual review of the citation
list of each study was then conducted at the title and abstract level.

• Review of the reference list from final or publicly available draft or finalized assessments
(e.g., EPA IRIS [Integrated Risk Information System], EPA PPRTV [Provisional Peer
Reviewed Toxicity], ATSDR [Agency for Toxic Substances and Disease Registry]
Toxicological Profile, NTP [National Toxicology Program], California EPA, TCEQ [Texas
Commission on Environmental Quality], IARC [International Agency for Research on
Cancer]). Assessments are identified from the database search, the resources listed in
Appendix B, or from the EPA CompTox Chemicals Dashboard ToxVal database U.S. EPA
(20211. Citation review of these materials is focused on the most pertinent section, i.e.,

This document is a draft for review purposes only and does not constitute Agency policy.

C-9	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

presentation of the human health literature, focusing on primary data studies pertinent to
cancer.

•	European Chemicals Agency (ECHA) registration dossiers to identify data submitted by
registrants fattoi //echa.europa.eu/information-on-chemicals /information-from-existing-

substances-regulation.

•	EPA ChemView database U.S. EPA (20191 to identify unpublished studies, information
submitted to EPA under Toxic Substances Control Act (TSCA) Section 4 (chemical testing
results), Section 8(d) (health and safety studies), Section 8(e) (substantial risk of injury to
health or the environment notices), and FYI (For Your Information, voluntary documents).
Other databases accessible via ChemView include EPA's High Production Volume (HPV)
Challenge database (https://iaspub.epa.gov/oppthpv/public_search.html_page) and the
Toxic Release Inventory database.

•	National Toxicology Program (NTP) Chemical Effects in Biological Systems (CEBS) database
of study results and research projects.

•	The Organisation for Economic Cooperation and Development (OECD) eChemPortal to
retrieve results for OECD Screening Information DataSet (SIDS) and High Production
Volume (HPV) Chemicals (https://www.echemportal.org/echemportal/).

•	References identified by technical consultants, during peer-review, and during public
comment periods (when applicable).

C.3.4.Non-Peer-Reviewed Data

IRIS assessments rely mainly on publicly accessible, peer-reviewed studies. However, it is
possible that unpublished data directly relevant to the PECO may be identified during assessment
development In these instances, the EPA will try to get permission to make the data publicly
available (e.g., in HERO); data that cannot be made publicly available are not used in IRIS
assessments. In addition, on rare occasions where unpublished data would be used to support key
assessment decisions (e.g., deriving a toxicity value), EPA may obtain external peer review if the
owners of the data are willing to have the study details and results made publicly accessible, or if an
unpublished report is publicly accessible (or submitted to EPA in a non-confidential manner) U.S.
EPA (20151. This independent, contractor driven, peer review would include an evaluation of the
study similar to that for peer review of a journal publication. The contractor would identify and
select at least three scientists knowledgeable in scientific disciplines relevant to the topic as
potential peer reviewers. Persons invited to serve as peer reviewers would be screened for conflict
of interest In most instances, the peer review would be conducted by letter review. The study and
its related information, if used in the IRIS assessment, would become publicly available. In the
assessment, EPA would acknowledge that the document underwent external peer review managed
by the EPA, and the names of the peer reviewers would be identified. In certain cases, IRIS will
assess the utility of a data analysis of accessible raw data (with descriptive methods) that has

This document is a draft for review purposes only and does not constitute Agency policy.

C-10	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

undergone rigorous quality assurance/quality control review (e.g., ToxCast/Tox21 data, results of
NTP studies not yet published) but that have not yet undergone external peer review.

Unpublished data from personal author communication can supplement a peer-reviewed
study as long as the information is made publicly available. If such ancillary information is acquired,
it will be documented in the Health Assessment Workspace Collaborative (HAWC,
https://hawcprd.epa.gOv/l or HERO project page (depending on the nature of the information
received). HAWC is a web-based software application designed to manage and facilitate the
process of conducting health assessments.

C.4. METHODS: LITERATURE SCREENING PROCESSES

CA.l.Title/Abstract and Full Text Screening

The studies identified from the database searches and application of SWIFT Review filters
are imported into SWIFT-Active Screener (https://www.sciome.eom/swift~activescreener/l for
title and abstract (TIAB) screening. SWIFT-Active Screener is a web-based collaborative software
application that utilizes active machine learning approaches to reduce the screening effort Howard
et al. [20201. TIAB screening is conducted by two independent reviewers and any screening
conflicts are resolved by discussion between the primary screeners with consultation by a third
reviewer, if needed. For citations with no abstract, articles are initially screened based on the
following: title relevance (title should indicate clear relevance), and page length (articles two pages
in length or less are assumed to be conference reports, editorials, or letters). Eligibility status of
non-English studies is assessed using the same approach with online translation tools or
engagement with a native speaker.

The machine learning screening process is designed to prioritize references that appear to
meet PECO-criteria or supplemental material content for manual review (i.e., both types of
references are screened as "include" for machine learning purposes). Screening continues until
SWIFT-Active Screener indicates that it was likely at least 95% of the relevant studies are
identified, a percent identification often used to evaluate the performance of machine learning
applications and considered comparable to human error rates Bannach-Brown et al.
f2Q181:Howard et al. f20161:Cohen etal. (20061. Any studies with "partially screened" status at the
time of reaching the 95% threshold are then fully screened. Studies identified as meeting PECO
criteria "unclear" or supplemental material during TIAB screening in SWIFT-Active Screener are
then imported into DistillerSR software fhttos://www.evidencepartners.com/products/distillersr-
svstematic-review-software /1. In DistillerSR, these studies underwent another round of TIAB
screening to separate PECO-relevant studies from studies containing only supplemental material.
The utility of studies classified as "unclear" was determined. Studies that met PECO or a specific
type of supplemental content were tagged accordingly and added to the evidence stream.

In DistillerSR, both TIAB and full-text screening is conducted by two independent reviewers
and any screening conflicts resolved by discussion between the primary screeners with

This document is a draft for review purposes only and does not constitute Agency policy.

C-ll	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

consultation by a third reviewer, if needed. Conflicts between screeners in applying the
supplemental tags, which primarily occur at the TIAB level, are resolved by similarly, erring on the
side of over-tagging based on TIAB content. Full-text references are sought through the EPA's HERO
database for studies screened as meeting PECO criteria or "unclear" based on the TIAB screening.
References that are not able to be procured within 45 days of attempt are determined to be
unavailable.

The screening decisions are then imported into HAWC's Literature Review Module, where
the screening and tagging results are visualized in interactive literature tag trees where additional
tagging can be conducted, e.g., more details on the nature of mechanistic or ADME studies.
C.4.2.SuppIementaI Material Tagging

Supplemental material records (Table C-2) can be identified at either the TIAB or full-text
levels. Conflicts between screeners in applying the supplemental material tags are resolved by
discussion and consultation with a third reviewer (as needed), erring on the side of over-including
at the TIAB level when the article content is relatively unclear.

It is important to emphasize that articles tagged as supplemental material are not
necessarily excluded from consideration in an assessment. The tagging structure is designed to
ensure that supplemental material studies are categorized for easy retrieval while conducting the
assessment Studies that meet the PECO criteria are those most likely to be used to derive toxicity
values and thus will undergo subsequent individual study evaluation and data extraction. In
contrast, the impact on the assessment conclusions of individual studies tagged as supplemental
material is often difficult to assess during the screening phase of the assessment These studies
could emerge as being critically important to the assessment and need to be evaluated and
summarized at the individual study level (e.g., cancer MOA or ADME studies). Supplemental
materials might be helpful to provide context (e.g., summarize current levels of exposure, provide
hazard evidence from routes or durations of exposure not pertinent to the PECO) or they might not
be cited by the assessment (e.g., individual studies that contribute to a well-established scientific
conclusion). The tagging inventory is intended to inform a systematic identification of key science
issues and refine the assessment evaluation plan (i.e., approach for analysis of mechanistic and
ADME/PK/PBPK content, or consideration of susceptible populations). When tagged during title
and abstract screening, it may not be clear whether the chemical of interest is reported in the study
(i.e., abstracts might not describe all chemicals investigated). In such cases, studies are still tagged
with the expectation that additional screening would clarify if the studies are considered pertinent
to address the specific aims of the assessment
C.4.3.Multiple Publications of the Same Data

When there are multiple publications using the same or overlapping data, all publications
will be included, with one selected for use as the primary study; the others will be considered as
secondary publications with annotation in HAWC indicating their relationship to the primary
record during data extraction. For epidemiology studies, the primary publication is most often the

This document is a draft for review purposes only and does not constitute Agency policy.

C-12	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

one with the longest follow-up, the largest number of cases, or a factor relevant to study evaluation.
For animal studies, the primary publication will typically be the one with the longest duration of
exposure, or with the outcome(s) most informative to the PECO. For both epidemiology and animal
studies, the assessments will include relevant data from all publications of the study, although if the
same data are reported in more than one study, the data will only be extracted once. For
corrections, retractions, and other companion documents to the included publications, a similar
approach to annotation is taken and the most recently published data are incorporated in the
assessments.

C.4.4.Literature Flow Diagrams

The results of the screening process are posted on the project page for the assessment in
the HERO database fhttps://heronetepa.gov/heronet/index.cfm/proiect/page/project id/14781.
Results are also summarized in a literature study flow diagram and interactive HAWC literature
trees (where additional tagging can be documented and visualized, e.g., more details on the nature
of mechanistic or ADME studies).

C.5. METHODS: LITERATURE INVENTORY PREPARATION

During title/abstract or full-text level screening in DistillerSR, studies that meet SEM PECO
criteria or a category of supplemental information are categorized based on evidence type (human,
animal, mechanistic, PBPK, etc.). Next, study design details for studies that meet SEM PECO criteria
are summarized and a more granular tagging of supplemental material is conducted as described
below. The results of this tagging are referred to as a literature inventory.

C.S.l.Studies That Meet SEM PECO Criteria

Human and animal studies that met SEM PECO criteria after full-text review are briefly
summarized in tabular format Summaries are done by one team member and quality checked by at
least one other team member. For non-English studies online translation tools (e.g., Google
translator) or engagement with a native speaker can be used to summarize studies at the level of
the SEM literature inventory. Fee-based translation services for non-English studies are typically
reserved for studies considered potentially informative for dose response, a consideration that
typically occurs subsequent to the SEM during preparation of the draft assessment

Assessing Suitability for Dose-Response Based on Study Design Considerations

The studies that meet SEM PECO criteria are evaluated with respect to the considerations
below to identify studies that may be suitable for developing an IUR.

•	Studies with chronic exposure durations or including exposure during reproduction or
development, are prioritized over studies with shorter-term exposure durations.

•	Animal studies using a species that is considered a relevant human surrogate.

This document is a draft for review purposes only and does not constitute Agency policy.

C-13	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

•	Studies with a broad exposure range and multiple exposure levels are preferred to the
extent that they can provide information about the shape of the exposure-response
relationship [see the EPA Benchmark Dose Technical Guidance, §2.1.1 U.S. EPA (2012b)]
and facilitate extrapolation to more relevant (generally lower) exposures.

•	For human studies, studies for which quantitative exposure measurements were available
and exposure-response results are presented in sufficient detail (e.g., standardized
mortality rate or relative risks, numbers of cases/controls, etc) are prioritized. Studies
based exclusively on duration of exposure analyses (i.e., longer versus shorter exposure
duration) are typically not considered suitable for dose response unless additional
information on exposure can be incorporated.

•	For epidemiological studies, studies that used biomarker measurements in tissues or bodily
fluids as the metric for exposure were only considered suitable for dose-response analysis if
data or PBPK models are available to extrapolate between the reported biomarker
measurement and the level of exposure.

•	For both animal and human studies, whether the nature of the outcomes/endpoints
assessed were interpretable with respect to potential adversity, was considered. Typically,
apical or clinical measures ("phenotypic") are preferred over other endpoints for dose
response. However, "mechanistic" endpoints can be useful in dose-response analyses when
they can be reasonably established as predictive of, or strongly associated with, phenotypic
outcomes interpreted as adverse.

•	High or medium confidence studies are highly preferred over low confidence studies (see
"Study Evaluation" below).

In addition to the broad criteria presented above, attributes of animal studies that met the
SEM PECO criteria are compared to the NTP inhalation cancer bioassays for soluble and insoluble
cobalt compounds NTP f2014.1998) used by prior assessments to develop cancer inhalation unit
risk values OEHHA f2Q2Q. 2019: TCEO f2017: U.S. EPA [20081 Only studies considered to be
comparable to (or an improvement over) the NTP studies will be considered for dose-response. Key
study attributes of the NTP studies are presented in (Table C-3).

Table C-3. Preferred design features of animal dose-response studies of
inhalation exposures to cobalt compounds.

Attribute

Preferred design feature

Rationale

Exposure duration

At least 2 years

Tumors in NTP (2014,1998) were late-onset. Prefer chronic
exposures to observe tumors.

Exposure design

Cyclical daily or workweek
exposure

Prefer studies to inform chronic continuous exposure. NTP (2014,
1998) exposed animals for 6 hours/day, 5 days/week.

Measurement of
exposure

Measures of particle size
(i.e., MMAD). Analytical
validation of chamber air
concentration

Particle size information is necessary for inhalation dosimetry, dose-
response modeling, and human extrapolation. Analytical validations
should be comparable to NTP protocols.

Number of
exposure groups

At least 3 (excluding
controls)

NTP {2014,1998) utilized 3 groups.

This document is a draft for review purposes only and does not constitute Agency policy.

C-14	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Attribute

Preferred design feature

Rationale

Animal sex

Both male and female

NTP (2014.1998) utilized both sexes.

Animal

species/strain

A species that is a relevant
or reliable human surrogate

NTP (2014.1998) utilized F344 rats and B6C3F, mice.

Number of
animals/groups

At least 50

NTP (2014,1998) utilized 50 animals per group.

Dose range

At least two concentration
groups below 5 mg Co/m3

5 mg/m3 is the highest concentration group in the NTP (2014) study
of insoluble cobalt metal. Tumor incidences were high at this
concentration. Data at lower levels (which are more environmentally
relevant and near the modeled benchmark response rate) are
preferred.

Measurement of
health outcome

Tumor incidence per group,
with adenomas/carcinomas
listed separately.

NTP (2014,1998) reports tumor incidence per group, with adenoma
and carcinomas presented separately.

Individual-level
data

Individual-level animal
tumor and survival data.

NTP (2,014,1998) provides individual-level data and poly-3 survival
statistic. Individual-level data are needed for time-to-tumor
modeling. NTP also reported changes in survival rate as a function of
concentration and time.

Study evaluation

Tumor data considered
medium or high confidence

Both NTP reports NTP f1998): NTP C2014) were considered hiah
confidence

(https://hawc.epa.gov/summary/visual/assessment/100500295/NTP-
Cancer-Bioassays/)

Study Evaluation

Epidemiological or animal studies that are prioritized from the analysis of suitability for
dose response will undergo study evaluation. When available, study evaluations from prior
assessments (e.g., RoC Monograph) were used to identify major limitations that would preclude the
study from being considered suitable for dose-response in this assessment Studies considered
suitable for dose-response - such as the NTP rodent cancer bioassays N]	8); NTP (20141 -

undergo full study evaluation using IRIS methodology - a domain-based approach to evaluate
studies. The detailed approaches are described in the Office of Research and Development (ORD)
Staff Standard Operating Procedures for Developing Integrated Risk Information System (IRIS)
Assessments (Version 1.0, October 2020, referred to as the "IRIS Handbook") U.S. EPA f2020cl.

The key concerns for the review of studies are potential bias (factors that affect the
magnitude or direction of an effect in either direction) and insensitivity (factors that limit the
ability of a study to detect a true effect; low sensitivity is a bias towards the null when an effect
exists). Each outcome or grouping of related outcomes within a study is judged independently by
two or more ORD staff reviewers using the HAWC Study Evaluation module. Reviewers reach a
consensus judgment (with conflict resolution by an additional reviewer, as needed) for each
evaluation domain and overall confidence determination. Judgments could differ from one outcome
to another within the same study, and with the overall study confidence determination. During
review, for each evaluation, domain reviewers reach a consensus judgment of good, adequate,
deficient, not reported, or critically deficient. It is important to emphasize that evaluations are
performed in the context of the study's utility for identifying individual hazards. Limitations

This document is a draft for review purposes only and does not constitute Agency policy.

C-15	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

specific to the usability of the study for dose-response analysis are useful to note, but they do not
contribute to the study confidence classifications. Once the evaluation domains have been rated, the
identified strengths and limitations are considered collectively to reach a study confidence
classification of high, medium, or low confidence, or uninformative for a specific health outcome.

This classification is based on the reviewer judgments across the evaluation domains and considers
the likely impact that inadequate reporting or the noted deficiencies in bias and sensitivity have on
the outcome-specific results. The specific limitations identified during study evaluation are carried
forward to help inform the synthesis within each body of evidence for a given health effect. Health
outcomes evaluated as uninformative are considered unusable for hazard and dose-response given
that the findings of interest are considered to be uninterpretable based on the identified flaws.
These studies have no impact on evidence synthesis or integration conclusions but may be used to
highlight research gaps.

This document is a draft for review purposes only and does not constitute Agency policy.

C-16	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

(a) Individual evaluation domains

Epidemiology

Animal

In vitro

•	Exposure measurement

•	Outcome ascertainment

•	Participant selection

•	Confounding

•	Analysis

•	Selective reporting

•	Sensitivity

•	Allocation

•	Observational bias/blinding

•	Confounding
¦ Attrition

•	Chemical administration and
characterization

•	Endpoint measurement

•	Results presentation

•	Selective reporting

•	Sensitivity

¦	Observational bias/blinding

¦	Variable control

•	Selective reporting

•	Chemical administration and
characterization

•	Endpoi nt measurement

•	Results presentation

•	Sensitivity

(b) Domain level judgments and overall study rating

Domain judgments

Judgment

Interpretation

0 Good
Adequate

Deficient

® Critically
Deficient

Appropriate study conduct relating to the domain and minor
deficiencies not expected to influence results

A study that may have some limitations relating to the domain, but
they are not liKely to be severe or to have a notable impact on results

Identified biases or deficiencies interpreted as liKely to have had a
notable impact on the results or prevent reliable interpretation of
study findings

A serious (law identified that makes the observed effect(s)
uninterpretabie. Studies with a critical deficiency are considered
"uninformative" overall.

Overall study rating for an outcome

Rating

Interpretation

High

Medium

Low

Uninformative

Mo notable deficiencies or concerns identified: potential for bias
unlikely or minimal; sensitive methodology

Possible deficiencies or concerns noted but they are unlikely to have a
significant impact on results.

Deficiencies or concerns were noted, and trie potential for substantive
bias or inadequate sensitivity could have a significant impact on trie
study results or their interpretation

Serious flaw(s) makes study results uninterpretabie but may be used
to highlight possible research gaps

Figure C-l. Overview of Integrated Risk Information System (IRIS) study
evaluation process, (a) individual evaluation domains organized by evidence type.,
and [b] individual evaluation domains judgments and definitions for overall ratings
(i.e., domain and overall judgments are performed on an outcome-specific basis),

1	Data Extraction of Study Methods and Results

2	Data will be extracted from prioritized studies into EPA's version of Health Assessment

3	Workspace Collaborative (HAWC, https://hawcprd.epa.gOv/l. a web-based software application

4	designed to manage and facilitate the process of conducting health assessments. Because the focus

5	of the current assessment is to develop a cancer IUR for inclusion in the IRIS database, tumor data

This document is a draft for review purposes only and does not constitute Agency policy,

C-l7	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

(along with any other data relevant to dose-response, such as animal survival rates and individual-
level data) are prioritized for data extraction. Data are also be stored in other formats (i.e., Excel,
BMDS, Word). See Section 4.2.1 "Selecting Endpoints for Dose-Response Assessment"

For quality control, data extraction is be performed by one member of the evaluation team
and independently verified by at least one other member. Discrepancies in data extraction are
resolved by discussion or consultation with a third member of the evaluation team.
C.5.2.SuppIementaI Material

The results of the supplemental material tagging (Table C-2) conducted in DistillerSR are
imported into the Literature Inventory module in HAWC, where more granular sub-tagging within a
type of supplemental material content category is conducted during assessment development
(including after preparation of the SEM). A single study can have multiple tags. Tagging judgements
in HAWC are made by one assessment member and confirmed during preparation of draft
assessment by another member of the assessment team.

C.6. RESULTS: LITERATURE SCREENING RESULTS

The database searches yielded 29,833 references in HERO after duplicate removal
(Figure C-l). Application of the SWIFT Review literature search filters (available online from
Sciome Company) for "human", "animal (human health models)", and "in vitro" evidence reduced
the number of studies for consideration to 4,588 after duplicate removal. The studies were
screened in SwiftActive Screener using predictive relevance, resulting in 2095 studies being
manually screened to identify 742 studies that were considered potentially PECO relevant or
supplemental ("included" for the purposes of machine learning) and 1353 references that were
manually excluded. After manually reviewing these 2095 references, screening was stopped
because SWIFT ActiveScreener indicated at least 95% of the relevant studies are identified, a
percent identification often used to evaluate the performance of machine learning applications and
considered comparable to human error rates Bannach-Brown et al. f20181:Howard et al.
f2Q16):Cohen etal. f2006). More specifically, in this project screening stopped when a predicted
96% of relevant studies were identified.

Separately, over 1600 unique records were identified from the other sources searched and
compared to the 4588 that were initially uploaded into SWIFTActive Screener, yielding 502 unique
records. These 502 studies, as well as the 742 studies previously identified as potentially PECO
relevant or supplemental, were imported into DistillerSR for a total of 1244 studies screened at
TIAB level. During TIAB screening in DistillerSR, 62 were included for full-text review, 826 were
tagged as supplemental material, and 399 were excluded as not relevant to PECO.

During full-text review, 19 studies were considered PECO relevant (11 animal studies and 8
human studies), 22 studies were excluded, and 22 studies were tagged as supplemental material.
The PECO relevant human and animal studies were then assessed for suitability for dose response

This document is a draft for review purposes oniy and does not constitute Agency poiicy.

C-18	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

(Table C-4, Table C-5). Literature search results are summarized graphically in Figure C-l and in an
interactive version in Figure C-2.

This document is a draft for review purposes only and does not constitute Agency policy.

C-19	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Cobalt Compounds Literature Searches (December 2021)

PubMedr 12/16/2021
(n = 7,442)

WoS, 12/16/2021
(n = 28,676)

Following duplicate removal, SWIFT Review used to analyze 29,833 records from database searches

Identification of potentially relevant records based on application of SWIFT-Review evidence stream tags for
human, animal (human health models), and in vitro evidence, n = 4,589.

Following additional duplicate removal in SWIFT Active, n = 4,588.

TIAB Screen in SWIFT Active (n = 4,588)

742 records met SEM PECO or considered
supplemental material

Excluded (n = 3846]

1353 records manually screened and excluded
2493 records predicted as not relevant in SWIFT
Active (and not manually screened)

Records identified from other sources {n = 1834)

Reference list from
existing assessments
(ASTDR, CalEPA, NTP, PPRTV)
	(n = 1834)	

Reference list from
included studies (n = 93}

ECHA
(n = 303)

NTP CEBS
(11=10)

QECD Echem
(n = 4)

EPA Chemyiew
(n = 14)

Comptox
(n = 0)

Identification of potentially relevant records based on citation from assessment cancer sections or potential PECO
relevance based on Title scan, n = 502.

T

Title & Abstract Screen in DistillerSR^

(n = 1244)

Included after Full-Text Screening (n = 19)

Animal {n = 11); Human (n = S)

Full-Text Screen in DistHlerSR (n = 62)

Excluded, did not meet all PECO criteria (n = 377)

Tagged as supplemental material (n = 304]

Sum of TIAB excluded or supplemental [n = 1,181)

Excluded (n = 22)

Not PECO relevant (n = 17)
Unable to obtain full text {n = 5}

Tagged as supplemental material (n = 22)

Sum of full-text excluded or supplemental (n = 44)

Tagged as Supplemental Material
TIAB + Full text + Inventory (n = 826)

•	Mechanistic (including in vivo/in vitro/ex vivo/in silico studies,}
(n = 187)

•	Non-mammalian model systems (n = 7)

•	Non-PECO route of exposure (n = 115}

•	Noncancer epidemiology or in vivo sub chronic/chronic animal
study (n = 235)

•	Cancer studies with less than subchronic exposure (n = 3)

•	Medical therapeutic studies (n = 18)

•	Medical implantation studies (n = 15)

•	Hypoxia-induction studies {n = 115)

•	ADME (n = 65)

•	PBPK (n = 0)

•	Exposure characteristics (no health outcome assessment) (n =
49)

•	Mixture studies (n = 44)

•	Records or other assessments with no original data (n = 66)

•	Case reports (n = 34)

•	Conference abstract (n = 4)

•	Errata/retracted (n= 1)

Figure C-2. Study Flow Diagram

This document is a draft for review purposes only and does not constitute Agency policy.

C-20	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Studies can be tagged to multiple supplemental tags, therefore, total number of supplemental subtags is
greater than the total number of supplemental references.

©

Animal

^19

Inclusion



826

Supplemental

Cobalt (inhalation) (2022}

Human

4245

Q

Manually Excluded

Exclusion

2493

Excluded by Machine Learning

Figure C-3. Literature tree. Click here for interactive version.

This document is a draft for review purposes only and does not constitute Agency policy.

C-21	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

C.7.LITERATURE INVENTORY

C. 7.1. Characterizing Epidemiological Studies for Dose-Response Analysis

Six epidemiological studies were identified that met the SEM PECO criteria, which were
developed to identify studies of cancer in relation to quantitative estimates of exposure Mur et al.
(1987]; Moulin et al. (1998]; Tuchsen et al. (1996]; Sauni et al. [2017]; White etal. [2019]; Kresovich
et al. (20191 (Table C-5). Four of the epidemiological studies involved workers, and included
evaluations of: malignant tumor (ICD-8 140-209] mortality in an electrochemical plant workers
Mur et al. (19871: lung cancer in a case-control study nested in a cohort study of workers in the
French hard-metal industry Moulin et;	8]; lung cancer in women exposed to cobalt-

aluminum spinel in a retrospective cohort study Tuchsen	3]; and multiple cancer types

(including lung] in Finnish cobalt production workers Sauni etal. (20171. The remaining two
studies assessed breast cancer in relation to environmental exposure to air pollutants including
cobalt (participants of the U.S.-wide Sister Study White etal. (20191. and participants of the Cancer
Care in Chicago study Kresovich et al. (201911.

Among the epidemiological studies, 3 had been included in the NTP RoC Monograph Cobalt
and Cobalt Compounds that Release Cobalt Ions In Vivo NTP (20161: the summary of study strengths
and limitations presented in the RoC Monograph Mur et a	[]; Moulin et al. (19981: Tuchsen et

al. (19961: Kresovich et al. (20191 were used to evaluate this set of studies for suitability for dose-
response analysis. For the 3 studies published after the RoC Monograph Sauni etal. (20171: White
et al. (20191: Kresovich et al. (20191. a targeted evaluation based on the considerations outlined in
the IRIS Handbook U.S. EPA (202Pel.and summarized in section 8.5 was performed. This targeted
evaluation revealed concerns in all 3 studies that precluded their use for dose-response, namely the
lack of individual-level exposure information, and the potential for confounding by co-exposures to
other carcinogens. These limitations are summarized in Table C-4. As the 3 earlier studies
evaluated in the RoC Monograph also had limitations, none of the human studies were deemed to
be more suitable for dose-response compared to the NTP animal cancer bioassay studies.
C. 7.2. Characterizing Animal Studies for Dose-Response Analysis

Eleven animal studies were identified that met SEM PECO criteria, including three NTP
Toxicity Reports NTP (19911: N	98]; NTP (20141 and six associated publications Bucher etal.

(19901: Bucher etal. (19991: Ozaki etal. (20021: Behl etal. (20151: i et al. (20151: Ton etal.
(20211. The two remaining publications had been considered in prior assessments Kerfoot (19731:
Palmes el	) and no new cancer bioassays were identified. The NTP rodent cancer

bioassays NTP (19981:NTP (20141 were both considered high confidence (Figure C-3], and the six
associated publications Bucher et al. (19901:Bucher et al. (19991:Ozaki etal. (20021:6efal etal.
(2015]:Hong et al. (20151:Ton et al. (20211 weNational Toxicology Program (Nr	08351: NTP

(20141 based on comparisons outlined in Table C-3. All other studies were determined to be
inadequate for dose-response for multiple reasons, with short exposure durations being the most

This document is a draft for review purposes only and does not constitute Agency policy.

C-22	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

1	common rationale (see Table C-5). The three subchronic studies fNTP (1991). Kerfoc	IX and

2	Palmes	) contained no tumor dose-response data. In addition, Kerfoot (19731. and

3	Palmes	had insufficient study designs and data reporting.

This document is a draft for review purposes only and does not constitute Agency policy.

C-23	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Table C-4. Analysis of Human Studies Meeting PECO Criteria for Suitability for Dose-Response.

Study

Study design and
population

Exposure

Endpoints

Study evaluation observation

Suitability for dose-
response

Muretal. (19871

Cohort of

electrochemical plant
workers producing
cobalt and sodium
(1950-1980)

Occupational
categories (whole
cohort, general
services,

maintenance, sodium
production, cobalt
production)

Malignant
tumor

mortality, lung

cancer

mortality

"Exposure duration: 60% worked greater than 10
years; 75% hired before 1975. Confounding: Likely
inadequate control for smoking; however, likely co-
exposure to nickel and arsenic with no control for
coexposures. Strengths: Cobalt production workers
exposed primarily to cobalt compounds.
Limitations: Small number of exposed cases; high
loss to follow-up (20%); potential for selection bias
due to left truncation" (page 49 from RoC
Monograph, NIP (2016).

Study quality concerns identified in the
confounding and sensitivity domains (page 47 from
RoC Monograph, NIP (2016).

Not suitable for dose-
response

Main limitations
related to
confounding,
sensitivity and
selection bias

Moulin et al.

(1998)

Nested case control
study of French hard-
metal industry workers
(10 facilities, 1968-
1991). 5777 males,
1682 females

Job-exposure matrix,
320 job periods and
semi-quantitative
estimation of
exposure to cobalt
and to tungsten
carbide

All cancer
mortality, lung
cancer

"No information on actual exposure level or
average exposure duration for the cohort.
Confounding: Potential concern for exposure to
other lung carcinogens, which were not controlled
in the cobalt alone analyses. Strengths: Exposure-
response analyses with multiple exposure metrics;
JEM validated for atmospheric concentrations of
cobalt; incident cohort reducing the potential for
left truncation; internal analysis reducing the
impact of the reported HWE; and lagged analysis.
Limitations: Potential confounding by coexposures
classified only as "ever/never" in the JEM" (page 51
from RoC Monograph, NIP (2016)

Study quality concerns identified in the
confounding domain (page 47 from RoC
Monograph, NIP (2016)

Not suitable for dose-
response

Main limitations
related to
confounding from
exposure to other
carcinogens.

Tuchsen et al.
(1996)

Retrospective cohort of
two Danish porcelain
factories, 874 women

Dust and airborne
concentrations (only
for certain years)

All-cause
mortality,
organ- specific

"Employment in factories/departments with or
without cobalt. Confounding: No control for
smoking; however, smoking data on subset of

Not suitable for dose-
response

This document is a draft for review purposes only and does not constitute Agency policy.

C-24	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Study

Study design and
population

Exposure

Endpoints

Study evaluation observation

Suitability for dose-
response



occupationally exposed
to cobalt (and 520
women not exposed)



cancer
incidence
(including lung
and breast
cancer)

workers suggests that smoking was not associated
with exposure. Strengths: Population exposed
primarily to cobalt compounds alone; only female
population with data on cobalt. Limitations: Small
number of exposed cases. Differential selection out
of the cohort could have occurred as the authors
mentioned that records of ill persons may have
been removed potentially resulting in an
underestimate of the true incidence of cancer."
(Page 47 from RoC Monograph, NIP (2016) 'This
study had low sensitivity to detect an effect
because of (1) small numbers of exposed cases in
this relatively small cohort and (2) potentially
combining workers with high and low exposures
together, which could dilute any effect and bias the
results towards the null. In addition, no lagged
analyses were reported. A concern about
differential selection also exists in this study. The
authors suggested that removal of records of ill
persons was known to take place in Danish
manufacturing. The possibility of differential
selection out of the cohort could have resulted in
an underestimation of the true incidence of lung
cancer in this study."

Study quality concerns identified in the sensitivity
domain (page 47 from RoC Monograph, NIP

(2016)

Main limitations
related to low
sensitivity

Sauni et al.
(20171

Cohort study of male
cobalt production
workers (Finland, 1969-
2013). 995 men with
26083 person-years.

Occupational
categories

Cancer
incidence
(including lung,
tongue, other
cancer types)

Male worker cohort stratified by age and exposure
level. Strengths: routine stationary measurements
and personal sampling with worker history verified.
Smoking data available. Limitation: potential
confounding by other carcinogens, namely nickel.
No information on alcohol consumption.

Study quality concerns: confounding and sensitivity

Not suitable for dose-
response

Main limitations
related to potential
confounding and
limited

generalizability.

This document is a draft for review purposes only and does not constitute Agency policy.

C-25	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Study

Study design and
population

Exposure

Endpoints

Study evaluation observation

Suitability for dose-
response

White et al.
(2019)

The Sister Study (US-
wide prospective
cohort) of 50,884
women.

U.S. EPA National Air
Toxics Assessment

Breast cancer

General population study with estimated exposure
to ambient toxic pollutants. Strengths: Large study
population. Limitations: Exposure to cobalt
estimated based on national air pollutant data. No
measurement of actual cobalt exposure levels.
Potential confounding by other air pollutants.
Study quality concerns: specificity of exposure and
confounding

Not suitable for dose-
response

Main limitations
related to potential
confounding due to
exposure to other
carcinogens.

Kresovich et al.
(2019)

Breast Cancer Care in
Chicago (population-
based cohort study)
study of 696 women.

U.S. EPA National Air
Toxics Assessment

Breast cancer

General population study with estimated exposure
to ambient toxic pollutants. Strengths: Health
outcome (breast cancer) medically verified.
Limitations: Exposure to cobalt estimated based
on national air pollutant data. No measurement of
actual cobalt exposure levels. Potential
confounding by other air pollutants. Study quality
concerns: specificity of exposure and confounding

Not suitable for dose-
response

Main limitations
related to potential
confounding due to
exposure to other
carcinogens.

1

Table C-5. Analysis of Animal Studies Meeting PECO Criteria for Suitability for Dose-Response.

Study

Species,
strain, sex

Dur.

Design

Air

measurements

Sample
size/group

Cone
(mg/m3)

Outcome
measure

Suitability for dose-response

NTP (1998)*

F344 rats,
B6C3Fi mice
M, F

2 yr

6h/day,
5d/week

Particle size and
mg/m3 validation

50

0

0.114

0.38

1.14

Tissue pathology
(quantitative)

Suitable for dose-response.

Chronic study. Tumors observed. Individual

animal data available. Large sample size.

NTP (2014)*

F344 rats,
B6C3Fi mice
M, F

2 yr

6h/day,
5d/week

Particle size and
mg/m3 validation

50

0

1.25

2.5

5.0

Tissue pathology
(quantitative)

Suitable for dose-response.

Chronic study. Tumors observed. Individual

animal data available. Large sample size.

NTP (1991)

F344 rats,
B6C3Fi mice
M, F

90 d

6h/day,
5d/week

Particle size and
mg/m3 validation

10

0

0.114

0.38

1.14

3.8

Tissue pathology
(quantitative).
No tumors
observed.

Not suitable for dose-response. Subchronic
study. No tumors observed. Small sample
size limits power to observe rare effects.

This document is a draft for review purposes only and does not constitute Agency policy.

C-26	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)













11.4





Kerfoot {1973}

Mini, swine
(sex not
specified)

90 d

6h/day,
5d/week

No particle or air

validation

presented

5

0

0.1

1

Tissue pathology
(qualitative). No
tumors
observed.

Not suitable for dose-response.

Insufficient data (animal specification, air
and outcome quantitation). Subchronic
study. No tumors observed. Small sample
size and few exposure groups. No individual
animal data available.

Palmes et al.

{1959}

Albino rats
(M), guinea
pigs, dogs

90 d

6h/day,
5d/week

Gaseous cobalt
hydrocarbonyl. Air
mg/m3validation

41 control,
75 exposed
(rats)

0
9

Tissue pathology

(qualitative),

hematology,

pharmacokinetic

s

Not suitable for dose-response. Insufficient
data (outcome quantitation). Subchronic
study. No tumors observed. Single high
exposure group (above 5 mg/m3). No
individual animal data available.

* Related studies include Bucher et al. {1990), Bucher et al. {1999), Ozaki et al. {2002), Be

il et al. {2015), Hong et al. {2015), and Ton et al. {2021).

1

This document is a draft for review purposes only and does not constitute Agency policy.

C-27	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

ADDENDUM 1. LITERATURE SEARCH STRATEGY
(DATE LIMITED TO 2019- 2021)

Search

Search Strategy

Results
and Date

wos

(TS=("cobalt" OR "7440-48-4" OR "10124-43-3" OR "Cobaltsulfat" OR "7646-79-9" OR
"Cobaltous chloride" OR "Dichlorocobalt" OR "1317-42-6" OR "71-48-7" OR "6147-53-1" OR
"917-69-1" OR "513-79-1" OR "10210-68-1" OR "21041-93-0" OR "21158-51-0" OR "61789-51-
3" OR "10141-05-6" OR "10026-22-9" OR "1308-04-9" OR "1307-96-6" OR "1308-06-1" OR
"10026-24-1" OR "Cobaltic acetate" OR "Dicobalt octacarbonyl" OR "Cobalt(ll) hydroxide" OR
"Cobaltous hydroxide" OR "Cobalt(ll) acetate" OR "Cobalt(ll) acetate tetrahydrate" OR
"Cobalt(lll) acetate" OR "Cobalt(ll) carbonate" OR "Cobalt(ll) chloride" OR "Cobalt(ll)
hydroxide" OR "Cobalt(ll) mesoporphyrin" OR "Cobalt(ll) naphthenate" OR "Cobalt(ll) nitrate"
OR "Cobalt(ll) nitrate hexahydrate" OR "Cobalt(ll) oxide" OR "Cobalt(lll) oxide" OR "Cobalt(ll)
sulfate" OR "Cobalt(ll) sulfate heptahydrate" OR "Naftolite" OR "Cobaltdinitrat" OR "Cobaltous
nitrate" OR "Cobaltous oxide" OR "C.I. Pigment Black 13" OR "Cobaltoxid" OR "Cobaltic oxide"
OR "Dicobalt oxide" OR "Cobaltosic oxide" OR "Cobaltic-cobaltous oxide" OR "Cobalto-cobaltic
oxide" OR "Tetraoxyde de tricobalt" OR "Tricobalttetraoxid" OR "tricobalt tetraoxide" OR
"Tricobalt tetraoxide" OR "Tricobalt tetroxide" OR "Cobaltous sulfate heptahydrate" OR
"cobalt element" OR "cobalto") AND (PY=2019-2021))

28,676
12/16/2021

PubMed

"cobalt"[tw] OR "7440-48-4"[rn] OR "10124-43-3"[tw] OR "Cobaltsulfat"[tw] OR "7646-79-
9"[tw] OR "Cobaltous chloride"[tw] OR "Dichlorocobalt"[tw] OR "1317-42-6"[tw] OR "71-48-
7"[tw] OR "6147-53-l"[tw] OR "917-69-l"[tw] OR "513-79-l"[tw] OR "10210-68-l"[tw] OR
"21041-93-0"[tw] OR "21158-51-0"[tw] OR "61789-51-3"[tw] OR "10141-05-6"[tw] OR "10026-
22-9"[tw] OR " 1308-04-9"[tw] OR "1307-96-6"[tw] OR "1308-06-l"[tw] OR "10026-24-l"[tw]
OR "Cobaltic acetate"[tw] OR "Dicobalt octacarbonyl"[tw] OR "Cobalt(ll) hydroxide"[tw] OR
"Cobaltous hydroxide"[tw] OR "Cobalt(ll) acetate"[tw] OR "Cobalt(ll) acetate tetrahydrate"[tw]
OR "Cobalt(lll) acetate"[tw] OR "Cobalt(ll) carbonate"[tw] OR "Cobalt(ll) chloride"[tw] OR
"Cobalt(ll) hydroxide"[tw] OR "Cobalt(ll) mesoporphyrin"[tw] OR "Cobalt(ll) naphthenate"[tw]
OR "Cobalt(ll) nitrate"[tw] OR "Cobalt(ll) nitrate hexahydrate"[tw] OR "Cobalt(ll) oxide"[tw] OR
"Cobalt(lll) oxide"[tw] OR "Cobalt(ll) sulfate"[tw] OR "Cobalt(ll) sulfate heptahydrate"[tw] OR
"Naftolite"[tw] OR "Cobaltdinitrat"[tw] OR "Cobaltous nitrate"[tw] OR "Cobaltous oxide"[tw]
OR "C.I. Pigment Black 13"[tw] OR "Cobaltoxid"[tw] OR "Cobaltic oxide"[tw] OR
"Dicobalt oxide"[tw] OR "Cobaltosic oxide"[tw] OR "Cobaltic-cobaltous oxide"[tw] OR
"Cobalto-cobaltic oxide"[tw] OR "Tetraoxyde de tricobalt"[tw] OR "Tricobalttetraoxid"[tw] OR
"tricobalt tetraoxide"[tw] OR "Tricobalt tetraoxide"[tw] OR "Tricobalt tetroxide"[tw] OR
"Cobaltous sulfate heptahydrate"[tw] OR "cobalt element"[tw] OR "cobalto"[tw])
AND (2019/01/01:3000[dp])

7,442
12/16/2021



Unique items were discovered using the search strategy above.

29,833



Number of records after application of SWIFT Review tags for human, animal (human health
models), and in vitro evidence

4,589

TOTAL

Number of records after an addition round of de-duplication SWIFT Active

4,588

1

This document is a draft for review purposes only and does not constitute Agency policy.

Addendum 1-1	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

ADDENDUM 2. PROCESS AND RESULTS FOR
SEARCHING AND COLLECTING EVIDENCE FROM
OTHER RESOURCES

Process

Review of reference lists from existing assessments (final or publicly available draft) and
journal studies considered relevant to PECO based on full-text screening

Citations from cancer sections of prior assessments were compiled and reviewed manually
by scanning the titles for those that appear to meet the PECO criteria. Any unique records
identified from these sources are formatted in an RIS file format, imported into DistillerSR,
annotated with respect to source, and screened as outlined previously in "Literature Screening
Processes".

Reference lists from journal articles are also reviewed manually by scanning the titles for
those that appear to meet the PECO criteria. This is only done for journal articles that meet PECO
criteria based on full-text review and not for journal articles tagged as supplemental material.

European Chemicals Agency

A search of the ECHA-registered substances database is conducted using the CASRN. The
registration dossier associated with the CASRN number is retrieved. The general information page
and all subpages included under the Toxicological Information tab are downloaded in PDF format,
including all nested reports that have unique URLs.

At this stage, each study summary is reviewed for inclusion on the basis of the PECO
criteria. When a study summary considers relevant reported data from a study or lab report, a
citation for the full study is generated in HERO, and it is verified that the study is not already
identified from the database search (or searches of "other sources consulted") prior to moving
forward to screening.

EPA ChemView

A search of the EPA ChemView database U.S. EPA f20191 using the chemical CASRN is
conducted. The prepopulated CASRN match and the "Information Submitted to EPA" output option
filter is selected before generating results. If results are available, the square-shaped icon under the
"Data Submitted to EPA" column is selected, and the following records are considered:

• High Production Volume Challenge Database (HPVIS)

This document is a draft for review purposes only and does not constitute Agency policy.

Addendum 2-1	DRAFT-DO NOT CITE OR QUOTE


-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

•	Human Health studies (Substantial Risk Reports)

•	Monitoring (Includes environmental, occupational and general entries)

•	TSCA Section 4 (Chemical testing results)

•	TSCA Section 8(d) (Health and safety studies)

•	TSCA Section 8(e) (Substantial Risk)

•	FYI (Voluntary documents)

All records for ecotoxicology and physical & chemical property entries are excluded. When
results are available, extractors navigate into each record until a substantial risk report link is
identified and saved as a PDF file. If the report cannot be saved, due to file corruption or broken
links, the record is excluded during full-text review as "unable to obtain record." Most substantial
risk reports contain multiple document IDs; thus, citations are derived by concatenating the unique
report numbers (OTS, 8EHD Num, DCN, TSCATS RefID, CIS) associated with each document along
with the typical author organization, year, and title. Once a citation is generated, the study is moved
forward to DistillerSR, where it is screened according to PECO criteria.

NTP Chemical Effects in Biological Systems

This CEBS database is searched using the chemical CASRN
fhttPsi //manticore.niehs.nih.gov/cebssearchl All non-NTP data are excluded using the "NTP Data
Only" filter. Data tables for reports undergoing peer review are also searched for studies that have
not been finalized (httpsi / /ntp.niehs.nih.gov/data/tables /index.html) on the basis of a manual
review of chemical names.

OECD Echem Portal

The OECD Echem Portal fhttps://hpvchemicals.oecd.org/Ul/Search.aspxl is searched using
the chemical CASRN to retrieve results for OECD Screening Information DataSet (SIDS) and High
Production Volume (HPV) Chemicals (https://www.echemportal.org/echemportal/). Only database
entries from those resources are included, and entries from all other databases are excluded in the
search. Final assessment reports and other relevant SIDS reports embedded in the links are
captured and saved as PDF files.

This document is a draft for review purposes only and does not constitute Agency policy.

Addendum 2-2	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

1 Results of Searching Other Resources

Source

Source address

Search terms

Search date

Total unique
number of

results
retrieved

Records not
otherwise
identified that were
screened in
DistillerSR

Review of reference lists of studies
considered relevant to PECO based
on full-text screening.

NA

NA

7/15/2022

93

34

Review of reference lists from
existing assessments (final or
publicly available draft)

NA

NA

3/24/22

1,834

465

EPA CompTox Chemicals Dashboard
version to retrieve a summary of any
ToxCast orTox21 high throughput
screening information

Iittps://comptox, epa.gov/dashboard

7440-48-4; 1345-16-0
7789-43-7; 513-79-1;
10210-68-1; 16842-03-8;
7646-79-9; 61789-51-3;
1307-96-6; 1308-06-1;
136-52-7; 10141-05-6;
10026-22-9; 10124-43-3

3/16/2022

0

0

ECHA

https://echa.europa.eu/da/informatio
nicals/registered-substances

7440-48-4; 1345-16-0;
7789-43-7; 513-79-1;
10210-68-1; 16842-03-8;
7646-79-9; 61789-51-3;
1307-96-6; 1308-06-1;
136-52-7; 10141-05-6;
10026-22-9; 10124-43-3

3/17/2022

303

0

EPA ChemView

https://chemview.epa.gov/chemview

7440-48-4; 1345-16-0;
7789-43-7; 513-79-1;
10210-68-1; 16842-03-8;
7646-79-9; 61789-51-3;
1307-96-6; 1308-06-1;
136-52-7; 10141-05-6;
10026-22-9; 10124-43-3

3/15/2022

14

0

This document is a draft for review purposes only and does not constitute Agency policy.
Addendum 2-3	DRAFT-DO NOT CITE OR QUOTE


-------
IRIS Assessment Plan and Protocol for Cobalt and Cobalt Compounds (Cancer, Inhalation)

Source

Source address

Search terms

Search date

Total unique
number of

results
retrieved

Records not
otherwise
identified that were
screened in
DistillerSR

NTPCEBS

httpsi//manticore,niehs, nih.gov/cebss
earch/

7440-48-4; 1345-16-0;
7789-43-7; 513-79-1;
10210-68-1; 16842-03-8;
7646-79-9; 61789-51-3;
1307-96-6; 1308-06-1;
136-52-7; 10141-05-6;
10026-22-9; 10124-43-3

3/16/2022

10

0

OECD Echem Portal

https://hpvchemicals.oecd.org/ll I/Sear
ch.aspx

7440-48-4; 1345-16-0;
7789-43-7; 513-79-1;
10210-68-1; 16842-03-8;
7646-79-9; 61789-51-3;
1307-96-6; 1308-06-1;
136-52-7; 10141-05-6;
10026-22-9; 10124-43-3

3/17/2022

4

4

PECO = Populations, Exposures, Comparators, and Outcomes; NA = not applicable; POD = point of departure; ECHA = European Chemicals Agency; NTP
CEBS = National Toxicology Program Chemical Effects in Biological Systems; OECD = Organisation for Economic Co-operation and Development.

This document is a draft for review purposes only and does not constitute Agency policy.
Addendum 2-4	DRAFT-DO NOT CITE OR QUOTE


-------