DRAFT                                                        EPA/600/P-97/001A
DO NOT CITE OR QUOTE                                            June 17, 1997
                                                            External Review Draft
                  Carcinogenic Effects of Benzene:
                                An Update
                                   NOTICE

THIS DOCUMENT IS A PRELIMINARY DRAFT. It has not been formally released by the
U.S. Environmental Protection Agency and should not at this stage be construed to represent
Agency policy. It is being circulated for comment on its technical accuracy and policy
implications.
                    National Center for Environmental Assessment
                         Office of Research and Development
                        U.S. Environmental Protection Agency
                                Washington, DC

-------
                                         DISCLAIMER

 1            This document is an external review draft for review purposes only and does not
2     constitute U.S. Environmental Protection Agency policy. Mention of trade names or commercial
3     products does not constitute endorsement or recommendation for use.
      6/17/97                                   ii        DRAFT-DO NOT CITE OR QUOTE

-------
                              CONTENTS


1. INTRODUCTION	1
           1.1. HISTORY OF THE 1985 INTERIM DOCUMENT	1
           1.2. PROPOSED 1996 GUIDELINES FOR CARCINOGEN RISK
           ASSESSMENT 	2

2. HAZARD ASSESSMENT AND CHARACTERIZATION  	5
           2.1. HUMAN DATA	5
           2.2. LABORATORY ANIMAL DATA	10
           2.3. MODE-OF-ACTION INFORMATION	10
      2.3.1. Mutagenicity and Genotoxicity  	11
      2.3.2. Metabolism  	14
      2.3.3. Pathogenesis  	16
  2.4.  HAZARD CHARACTERIZATION SUMMARY	17

3. DOSE-RESPONSE ASSESSMENT AND CHARACTERIZATION	19
      3.1. DESCRIPTION OF DIFFERENT RISK ASSESSMENTS	20
      3.2. SHAPE OF THE DOSE-RESPONSE FUNCTION AT LOW DOSES	23
      3.3. DOSE-RESPONSE CHARACTERIZATION 	26

4. CHILDREN'S RISK CONSIDERATIONS  	31

5. FUTURE RESEARCH NEEDS	32

6. REFERENCES  	34
6/17/97                              iii      DRAFT-DO NOT CITE OR QUOTE

-------
                                  LIST OF TABLES
1.      Relative risk as a function of cumulative exposure  	6
2.      Standardized mortality ratios for deaths from leukemia
       among Pliofilm workers based on the estimated cumulative
       exposure of the selected investigators	8
3.      Estimated relative risks of leukemia derived by the proportional hazards
       dose-response model according to the estimated cumulative
       exposure (ppm-years) of the selected investigators	8
4.      Risk estimates calculated on the basis of Pliofilm
       workers by various investigators  	22
5.      Evidence that benzene-induced leukemia is nonlinear
       at low doses	24
                                  LIST OF FIGURES
1.      Key metabolic activation pathways in benzene toxicity	15
6/17/97                                   iv        DRAFT-DO NOT CITE OR QUOTE

-------
                 AUTHORS, CONTRIBUTORS, AND REVIEWERS
      This document was prepared by the National Center for Environmental Assessment-
Washington Office (NCEA-W) of EPA's Office of Research and Development.
AUTHORS

David L. Bayliss, NCEA-W
Chao Chen, NCEA-W
Babasaheb Sonawane, NCEA-W
Lawrence Valcovic, NCEA-W
CONTRIBUTORS

NCEA

Annie Jarabek, NCEA-W
Robert McGaughy, NCEA-W
James Walker, NCEA-W

Outside Contributor

Martyn T. Smith (contractor), University of California, Berkeley, CA


REVIEWERS

NCEA

Michael Callahan
David Cleverly
James Cogliano
William Farland
Sue Perlin
Charlie Ris
John Schaum
Chon Shoaf

Other EPA Offices

Linda Birnbaum, NHEERL, RTF
Pam Brodowicz, QMS, RTF


6/17/97                                v        DRAFT-DO NOT CITE OR QUOTE

-------
                                          1. INTRODUCTION

  1            In 1992, the U.S. Environmental Protection Agency's (EPA's) Office of Mobile Sources
  2     (OMS) requested the National Center for Environmental Assessment (NCEA) to provide an
  3     updated characterization of the cancer risk of benzene to humans.  The previous characterization
  4     of the carcinogenic risk of exposure to benzene was done in 1985 by the Office of Health and
  5     Environmental Assessment (the predecessor organization to NCEA). Additional scientific data
  6     relevant to the carcinogenicity of benzene have been published in the literature since that time.
  7     This has brought into question the relevancy  of the earlier quantitative cancer risk estimates. The
  8     1985 cancer unit risk estimates were based on assumptions about the effects of low-level benzene
  9     exposure on humans derived from occupational health studies.
 10            The regulatory authority (Clean Air Act Amendments, 1990) for controlling fuel emissions
  11     in vehicles resides in OMS. Before OMS exercises its regulatory authority, OMS has asked
 12     NCEA for scientific supporting documents based on health implications of continued  exposure to
 13     benzene.
 14            The scope of this report is limited to issues related to the carcinogenicity of benzene.
 15     Specifically, this report evaluates and discusses studies published since 1985 to ascertain if there
 16     has been sufficient new scientific information that would significantly alter the 1985 interim
 17     benzene unit cancer risk estimate.

 18     1.1.  HISTORY OF THE 1985 INTERIM DOCUMENT
 19            In 1985, the Office of Research and Development prepared estimates of the inhalation unit
20     risk for benzene (U.S.  EPA, 1985) at the request of OAQPS.  The previous cancer risk
 21     assessment on benzene by the Agency was completed in January 1979 (U.S. EPA, 1979).  In
22     subsequent years, this assessment became out of date as new scientific information became
23     available. In response to the need to update the 1979 assessment, the 1985 Interim Quantitative
24     Cancer Unit Risk Estimate Due to Inhalation of Benzene was developed. It reviewed and
25     incorporated information from the three most recent epidemiologic studies at the time (Rinsky et
26     al.,  1981; Ott et al., 1978; Wong et al., 1983). In addition, animal inhalation studies in male rats
27     and mice (Goldstein et al., 1982) and in male and female rats (Maltoni et al., 1983) constituted the
28     expanded scientific information base.
29            Data from the occupational cohorts of Rinsky et al. (1981) and Ott et al. (1978) were
30     pooled and analyzed by Crump and Allen (1984) to provide exposure (cumulative dose) estimates
 31     for use in the development of a benzene cancer risk assessment for the Occupational Safely and
32     Health Administration (OSHA, 1987) independently of EPA.  These exposure estimates were
33     available for use by the Agency. Crump and Allen (1984) made their exposure estimates using
        6/17/97                                     1         DRAFT-DO NOT CITE  OR QUOTE

-------
  1     three separate approaches (cumulative, weighted cumulative, and window) and used two risk
  2     models (absolute and relative).
  3            The cumulative dose approach assumes that the risk depends on the air concentration
  4     times duration of exposure. The weighted cumulative dose approach assumes that the
  5     contribution of an exposure to risk varies depending on when exposure occurred to the individual.
  6     The window approach assumes that benzene exposure for longer than 15 years induces no
  7     additional risk but that exposure between 5 and 10 years induces a risk proportional to the air
  8     concentration and exposure duration. All  exposure estimate approaches assume a latency period
  9     that begins at the beginning of exposure and during which there is assumed to be no increased
 10     risk. An absolute risk model assumes that the risk from exposure is independent of the
  11     background  risk of disease, whereas a relative risk model assumes that the risk from exposure is
 12     proportional to the background  incidence  of the disease (see section 3.1).
 13            The Agency concluded that the cumulative and the weighted cumulative exposure
 14     estimates were both valid and preferable to the window approach.  EPA also concluded that the
 15     absolute and relative risk models had equal validity. It was decided to calculate the geometric
 16     mean of the  four resulting estimates derived from the different exposure estimates and risk models
 17     and then  multiply this by a correction factor based on the epidemiologic data of Wong et al.
 18     (1983).  This correction factor (1.23) was  the ratio of risk estimates (under the relative risk model
 19     and cumulative exposure estimate) when all three studies (Rinsky et al., 1981; Ott et al., 1978;
20     Wong et al., 1983) are used to the risk estimate generated when  only the Rinsky et al. (1981) and
 21     Ott et al.  (1978) studies are used under the same relative risk model and cumulative exposure
22     estimate.  (Note:  The Wong et  al. [1983]  study cannot be used under the absolute risk model
23     because no person-year information was provided in the report.) The resulting quantitative
24     cancer unit risk of 2.6 x 10"2 per ppm air concentration was about 10 times greater than the
25     human risk estimate based on the three animal inhalation studies  and 1.5 times higher than the
26     pooled estimates from the three gavage studies. This estimate compares well with the original
27     estimate from the 1979 benzene risk document (U.S. EPA, 1979) of 2.41  x  10"2, which was based
28     on the geometric mean of three  unit risk estimates derived from the occupational cohort studies of
29     Infante et al. (1977), Aksoy (1976, 1977), Aksoy et al. (1994), and Ott et al. (1977).

30     1.2. PROPOSED 1996 GUIDELINES FOR CARCINOGEN RISK ASSESSMENT
 31            The Agency recently published its  Proposed Guidelines for Carcinogen Risk Assessment
32     (U.S. EPA, 1996). When final,  these guidelines will supersede the existing Guidelines for
33     Carcinogen Risk Assessment published in 1986 (U.S. EPA, 1986).  The 1996 proposed guidelines
34     include a number of changes that accommodate a more detailed understanding of the carcinogenic
35     process (i.e., chemical and gene interactions) and provide a framework for the use of mechanistic
        6/17/97                                    2         DRAFT-DO NOT  CITE OR QUOTE

-------
  1     data. It should be noted, however, that the results of an assessment under the new guidelines will
  2     not differ greatly from those under the 1986 guidelines, unless new kinds of information are
  3     forthcoming from research on mechanisms and toxicokinetics (Wiltse and Dellarco, 1996).
  4            The revisions are intended to provide for greater flexibility in evaluating the rapidly
  5     increasing new scientific data available on cancer research in decisions to implement the Agency's
  6     regulatory authority. Technical characterizations are important components of the new guidelines
  7     and serve to explain the key lines of evidence and conclusions, discuss the strengths and
  8     weaknesses of the evidence, present alternative conclusions, and point out significant issues and
  9     uncertainties deserving serious consideration. A risk characterization summary would integrate
 10     technical  characterizations of exposure, hazard, and dose response to form the overall synthesis
  11     and conclusions about human health risk. This document is limited to discussions of the hazard
 12     and dose-response characterizations.
 13            The hazard assessment component emphasizes use of information about an agent's mode
 14     of action to reduce the uncertainty in describing the likelihood of harm and to provide insight on
 15     appropriate extrapolation procedures.  Mode of action is defined as the agent's influence on
 16     molecular, cellular,  and physiological functions. Because it is the sum of the biology of the
 17     organism and the chemical properties of an agent that leads to an adverse effect, the evaluation of
 18     the entire range of data (i.e., physical, chemical, biological, and toxicological information) allows
 19     one to arrive at a reasoned judgement of an agent's mode of action. Although cancer is a
20     complex and diverse process, a risk assessment must operationally dissect the presumed critical
 21     events, at least those that can be measured experimentally, to derive a reasonable approximation
22     of risk (Wiltse and Dellarco, 1996). Understanding the mode of action helps interpret the
23     relevancy of the laboratory animal data and guides the dose-response extrapolation procedure,
24     i.e., it helps to answer the question of the shape of the dose-response function at low doses. The
25     conditions (i.e., route, duration, pattern, and magnitude of exposure) under which the
26     carcinogenic effects of the agent may be expressed should also be considered in the hazard
27     characterization.
28            The weight-of-evidence narrative for  the hazard characterization includes classification
29     descriptors. Three standard categories of descriptors ("known/likely," "cannot be determined,"
30     and "not likely") were proposed to replace the six letter categories used in the 1986 guidelines
 31     (i.e., A-E).  Because of the wide variety of data sets encountered on agents, these descriptors are
32     not meant to stand alone; rather, the narrative context is intended to provide a transparent
33     explanation of the biological evidence and how the conclusions were derived.
34            The dose-response assessment under  the new guidelines is a two-step process.  In the first
35     step, the response data are modeled in the range of empirical observation. Modeling in the
36     observed  range is done with biologically based or appropriate curve-fitting models. The second
        6/17/97                                     3        DRAFT-DO NOT CITE OR QUOTE

-------
  1      step, extrapolation below the range of observation, is accomplished by modeling if there are
 2      sufficient data or by a default procedure.
 3             This evaluation and review of benzene health risk assessment issues is being conducted
 4      under the standing guidance of the 1986 guidelines but with a recognition of these areas of
 5      emphasis in the 1996 proposed guidelines. Thus, this review of benzene health risk assessment
 6      issues contains a discussion of how recent evidence on mode of action can be incorporated into
 7      hazard  characterization and dose-response approaches. Earlier dose-response assumptions or
 8      alternative approaches will be  discussed in this context.
 9             The major issue addressed in this document involves the magnitude of the risk of cancer to
10      humans exposed to low levels of benzene. Occupational  studies provide the bulk of evidence of
 11      benzene's carcinogenicity, and workers are exposed at much higher levels than is the general
12      public.  The 1996 proposed guidelines recommend a detailed discussion of the basis for
13      developing the quantitative unit risk estimate drawing on mode-of-action, metabolism, and
14      pharmacokinetics information replete with uncertainty discussions as appropriate. The 1985
15      interim estimate calculation for benzene was based on science policy using a procedure
16      incorporating the geometric mean of maximum likelihood estimates because little information was
17      available regarding the mode of action of carcinogenicity at low exposure levels.
        6/17/97                                     4        DRAFT-DO NOT CITE OR QUOTE

-------
                       2.  HAZARD ASSESSMENT AND CHARACTERIZATION

  1            The "known/likely" category of the proposed 1996 cancer guidelines includes agents for
  2     which adequate epidemiologic evidence (known) or a combination of epidemiologic and
  3     experimental evidence demonstrates an association between human exposure and cancer.
  4            It has been clearly established and accepted that exposure to benzene and its metabolites
  5     causes acute nonlymphocytic leukemia and a variety of other blood-related disorders in humans
  6     (ATSDR, 1996; IARC, 1982; U.S. EPA, 1979).  The existing Group A classification of benzene
  7     based on the 1986 guidelines would be replaced with a narrative incorporating the "known/likely"
  8     descriptor under the 1996 proposed guidelines.  The narrative should discuss the uncertainties
  9     about the following: the shape of the dose-response curve at low doses, mode of action, and
 10     exposure in human studies; these topics are addressed in this section.

  11     2.1.  HUMAN DATA
 12            Epidemiologic studies provide clear evidence of a causal association between exposure to
 13     benzene and leukemia, especially acute nonlymphocytic (myelogenous) leukemia and, to a lesser
 14     extent, chronic nonlymphocytic leukemia as well as chronic lymphocyte leukemia and multiple
 15     myeloma (Aksoy, 1976, 1977; Aksoy et al.,1974; Infante et al., 1977; Rinsky et al., 1981, 1987;
 16     Vigliani and Saita, 1964; IARC, 1982; ATSDR, 1996).  Lymphocyte leukemia, a form of
 17     leukemia commonly found in children, may have a genetic  component as well as an
 18     environmental exposure component (Linet, 1985). A role for benzene  and other environmental
 19     chemicals cannot be ruled out. A higher risk of multiple myeloma also may be associated with
20     exposure to benzene (DeCoufle et al.,  1983; Rinsky  et al., 1987).
 21            The study of Pliofilm rubber workers at three facilities in Ohio (Rinsky et al., 1981)
22     provides the best published set of data to date for evaluating human cancer risks from exposure to
23     benzene. Since the 1985 assessment, this cohort has been expanded (Rinsky et al., 1987) to
24     include workers who were employed at least 1 day between January 1, 1940, and December 31,
25     1965. (In the previous study, employment after December 31, 1950, was not considered.)
26            Three questions have been raised by NCEA concerning the impact of these more recent
27     data to the present assessment of benzene and its use in a quantitative risk assessment. First, does
28     the 1987 Rinsky et al. update lead to any substantial  changes in the estimated relative risk ratios
29     that were derived in the 1981 Rinsky et al. report? Second, one of the  major problems with
30     exposure estimates used by Rinsky et al. (1981, 1987) and others in deriving relative risk
 31     estimates for use in developing quantitative unit risk estimates is that no ambient air
32     measurements of exposure to benzene in the workplace of the Pliofilm workers were taken in the
33     years before 1946.  The first known measurements were taken in 1946, and then there were only
        6/17/97                                    5        DRAFT-DO NOT CITE OR QUOTE

-------
  1     four samples measured. The absence of definitive ambient air measurements during this time has
  2     led to a flourish of quantitative risk estimates by numerous investigators over the past several
  3     years that have differed based partially on differences in the assumptions made about what those
  4     earlier exposures to benzene were. Do the various approaches used to estimate exposure in the
  5     those early years lead to estimates that differ by a substantial amount?  Third, because the Rinsky
  6     et al. (1987) Pliofilm cohort is currently the best set of data available for estimating exposure and
  7     the risk of leukemia, would it be advisable to calculate the quantitative unit risk estimates utilizing
  8     that cohort only, and what effect would discarding the Ott et al. (1978) and Wong et al. (1983)
  9     epidemiologic studies have on the calculation of a unit risk estimate?
 10            To answer the first question, the first study (Rinsky et al., 1981) of Pliofilm workers in
  11     the rubber industry covered three facilities in Ohio and consisted of 1,165 male workers who had
 12     been employed sometime between 1940 and 1965 and followed through 1981. The second  study
 13     (Rinsky et al., 1987) included an additional  6.5 years of follow-up from the earlier study.  It also
 14     included  individual estimates of personal exposure, which were not included in previous versions.
 15     Duration of employment in combination with personal exposure estimates during that employment
 16     was used to generate risk estimates based on grouped data.  The updated version made it possible
 17     to evaluate dose-response relationships and  estimate risks at low exposure levels in terms of ppm-
 18     years of exposure.  One myeloblastic leukemia was subsequently added after the additional
 19     follow-up. However, because of the compensating increase in expected deaths due to the
20     additional person-years of follow-up, only a small change occurred in the overall relative risk.
 21     Altogether, 9 leukemias were observed versus 2.66 expected in this cohort by December 31, 1981
22     (Rinsky et al., 1987).
23            The relative risks were found to increase with cumulative exposure as shown  in table 1.
24            To answer the second question, after 1946, some measurements were available that made
25     it possible to calculate rough estimates of personal cumulative exposure for each member of the
26     cohort. These estimates tended to be similar among different investigators. However, Rinsky et
                        Table 1.  Relative risk as a function of cumulative exposure
Cumulative exposure
(ppm-years)
0-40
40-200
200-400
More than 400
Relative risk
1.1
3.2
11.9
66.4
        6/17/97
DRAFT-DO NOT CITE OR QUOTE

-------
  1     al. (1981, 1987), Crump and Allen (1984), and Paustenbach et al. (1992, 1993) employed various
  2     assumptions to estimate personal exposure levels before 1950, when exposures were most
  3     intense. The estimates of exposure made by Rinsky et al. (1981, 1987) were generally the lowest,
  4     thus giving rise to the highest risk estimates, but there is no consistent pattern among the
  5     estimates for particular years.
  6            Paustenbach et al. (1992, 1993) used a variety of assumptions to derive the highest
  7     estimates of personal exposure of any of the investigators. They cited seven factors that
  8     influenced their estimates as follows: (1) inaccuracy of devices used for monitoring airborne
  9     concentrations of benzene, (2) length of the work week, (3) rubber shortages during World War
 10     II, (4) installation of local exhaust systems to reduce airborne concentrations of benzene, (5)
  11     additional exposure to benzene by skin contact, (6) ineffectiveness of respiratory devices, and (7)
 12     medical evidence of overexposure of workers to benzene.  These factors tended to provide an
 13     incentive for the authors to conclude that these pliofilm workers were exposed to the highest
 14     levels during the early years of exposure.
 15            Rinsky et al. (1981, 1987), on the other hand, after analyzing data from various sources
 16     (Industrial Commission of Ohio in 1946 and 1955,  Ohio Department of Health in 1956, the
 17     University of North Carolina in 1974, NIOSH in 1976, and company surveys from 1946 to 1950
 18     and 1963 to 1976), assumed that the levels of benzene as measured by the 8 h time-weighted
 19     average (TWA) exposure of the workers were close to recommended standards for specific years
20     as follows:  100 ppm (1941), 50 ppm 8 h TWA (1947), 35 ppm 8 h  TWA (1948), 25 ppm 8 h
 21     TWA (1957 and 1963), and 10 ppm TWA (1969).  They produced the lowest set of estimates.
22            Crump and Allen (1984) developed a third  set of exposure estimates based on the concept
23     that benzene levels declined as progressively more  restrictive standards were implemented in the
24     workplace.  These estimates lie somewhere between those of Rinsky et al. (1981, 1987) and
25     Paustenbach et al.  (1993). These same estimates of Crump and Allen (1984) were used in
26     deriving the quantitative unit risk estimates in EPA's Interim Quantitative Cancer Unit Risk
27     Estimates Due to Inhalation  of Benzene (U.S. EPA, 1985). Even with the differences in the
28     exposure levels produced by utilizing these three sets of estimates of exposure for the employees,
29     the cumulative standardized mortality ratios (SMRs) differed from the Crump and Allen estimates
30     by no more than a factor of two (table 2).
 31            When using the proportional hazards dose-response model, such as was used by Paxton et
32     al. (1992) and Paxton (1996), the estimated relative risks differed by no more than a factor of four
33     within each cumulative dose-response category from the Crump and Allen (1984) estimates  (table
34     3).  Hence,  the use of Rinsky et al. (1981, 1987) or Paustenbach et al. (1993) exposure estimates
35     would affect the quantitative risk estimate little.

        6/17/97                                    7         DRAFT-DO NOT CITE OR QUOTE

-------
               Table 2.  Standardized mortality ratios for deaths from leukemia among Pliofilm
               workers based on the estimated cumulative exposure of the selected investigators
Investigators
Rinskyetal., 1981,1987
Crump and Allen, 1984
Paustenbachetal., 1993
0-5 ppm-yrs
2.0
0.9
1.3
5-50 ppm-yrs
2.3
3.2
1.8
50-500 ppm-yrs
6.9
4.9
2.8
>500 ppm-yrs
20
10.3
11.9
          Source: Paxton, 1996.

               Table 3.  Estimated relative risks of leukemia derived by the proportional hazards
               dose-response model according to the estimated cumulative exposure (ppm-years) of
               the selected investigators
Investigators
Rinskyetal., 1981, 1987
Crump and Allen, 1984
Paustenbach et al., 1993
4.5 ppm-yrs
1.02
1.00
1.01
45 ppm-yrs
1.19
1.04
1.07
90 ppm-yrs
1.41
1.07
1.14
450 ppm-yrs
5.5
1.43
1.96
            Source: Paxton etal., 1992.

  1             More recently, a new analysis has been provided (Schnatter et al., 1996) of the Pliofilm
 2      cohort that continues to use the three main sets of exposure estimates described above and the
 3      median of the three to develop a new set of indices of exposure per person.  This technique,
 4      however, differs from the standard method of measuring total exposure to benzene (i.e.,
 5      cumulative exposure = length of exposure * concentration) in that an "average" total
 6      concentration per person is determined from the job with the greatest exposure (maximally) and
 7      of longest duration from the exposure estimates above.  This method enables the researcher to
 8      isolate subgroups with less exposure to specified concentrations of benzene and then calculate the
 9      risk of leukemia in those subgroups. In theory, these subgroups were less likely to be exposed to
10      concentrations greater than a specified concentration.
 11             The results of the Schnatter et al. (1996) analysis indicate that for the lowest exposure
12      estimates (Rinsky et al., 1981, 1987), the "critical" concentration is between 20 and 25 ppm for
13      the risk of acute myelogenous leukemia (AJVIL) "to be expressed," and for the median, the risk is
14      between 50 and 60 ppm, although there appears to be instability in both these risk estimates.
15      Interestingly, for total leukemia, the "critical" concentrations for the median are lower and appear
16      to fall in the range of 35 to 40 ppm and the risk estimates appear somewhat less erratic.
        6/17/97
DRAFT-DO NOT CITE OR QUOTE

-------
  1            These figures are not inconsistent with estimates from Wong (1995), who utilized the
  2     parameter cumulative exposure to estimate the risk of AML in pliofilm workers.  However, the
  3     Schnatter et al. (1996) analysis suffers from the same problems that the Wong (1995) and Rinsky
  4     et al. (1987) studies suffer in utilizing pliofilm workers to estimate risks at low levels of exposure
  5     to benzene:  the data lack sensitivity.  To assume that a critical concentration exists at the levels
  6     indicated and from which a "threshold" could be inferred is unwarranted based solely on this data
  7     set. In fact, the lower estimates  of the critical concentration based on the sum total of leukemia
  8     deaths versus just those deaths from AML seem to suggest that there may be a lower critical
  9     region for AML if a larger data set were available.
 10            To answer the third question, the net result of discarding the Ott et al. (1978) and the
  11     Wong et al. (1983) studies would be to change the unit risk estimate little.  The Ott et al. (1978)
 12     cohort and its later update (Bond et al., 1986a) rely on a smaller data set.  Both the Ott study and
 13     its update by Bond have insufficient power to detect a risk of leukemia at low doses.
 14     Furthermore, Bond et al. (1986a) also state that their data for risk assessment purposes should not
 15     be used because of several factors (i.e., small number of events, competing exposures to other
 16     potentially hazardous materials,  and the uncertain contribution of unquantified brief exposures).
 17            While the Wong et al. (1983) cohort has ample power to detect a risk of leukemia, and the
 18     update (Wong, 1987) includes estimates of personal exposure to benzene, the estimates
 19     apparently are not reliable.  Wong (1987) states that the estimated historical industrial hygiene
20     data were not precise enough for absolute quantitative risk assessment. The Rinsky et al. (1981,
 21     1987) cohort has ample power, latency, and better  estimates of later exposure to airborne
22     benzene. However, during certain time frames (i.e., levels of ambient air benzene before 1950),
23     the actual airborne measurements of benzene in the workplace were either meager or nonexistent.
24            In the 1985 interim benzene document (U.S. EPA, 1985), a decision was made to
25     calculate a single overall unit risk estimate by obtaining the geometric mean of four maximum
26     likelihood unit risk estimates generated from the Ott et al. (1978) and Rinsky et al.  (1981, 1987)
27     studies, both absolute and relative risk models, and then "correcting" this mean by multiplying it
28     by the ratio of the largest unit risk estimate from the four separate unit risk numbers above to the
29     unit risk estimate calculated from the Wong (1987) cohort.  The result was a probability  of 2.6 x
30     10"2, which is close to that calculated by Crump (1992) assuming  similar conditions, that is, a
 31     linear model and Crump and Allen (1984) exposure estimates but excluding Ott at al. (1978),
32     Bond et al. (1986a), and Wong (1987).  These numbers range from 1.1 x icr2 to 2.5 x icr2 and
33     can be found in section 3 (table 4).  By inspection,  the inclusion of data from Ott at al. (1978),
34     Bond et al. (1986a), or Wong (1987) changes these unit risk estimates little.
        6/17/97                                     9         DRAFT-DO NOT CITE OR QUOTE

-------
  1            It is apparent that the calculation of a new unit risk estimate based on a reordering of the
  2     assumptions about what the earlier distribution of ambient air measurements of benzene might
  3     have been, or from the elimination of data sets that add little to the knowledge of risk at low
  4     doses and have questionable validity, will likely result in little change from the 1985 estimate
  5     based on the epidemiologic data alone.

  6     2.2.  LABORATORY ANIMAL DATA
  7            Studies on the carcinogen!city of benzene in rodents include inhalation exposures in
  8     Sprague-Dawley rats, C57BL/6 mice, AKR mice, CD-I mice, and CBA mice and gavage
  9     treatment of Sprague-Dawley rats, Wistar rats, F344 rats, RF/J mice, Swiss mice, and B6C3FJ
 10     mice (Cronkite et al., 1989; Goldstein et al., 1982; Huff et al., 1989; Maltoni et al., 1983, 1988;
  11     NTP, 1986; Snyder et al., 1980, 1982, 1984). Inhalation concentrations ranged from 0 to 1,000
 12     ppm and gavage doses ranged from 0 to 200 mg/kg. Upon exposure via inhalation, benzene was
 13     found to be carcinogenic in rats and mice in multiple target organs including oral and nasal
 14     cavities, liver, forestomach, preputial gland, lung, ovary, and mammary gland. It is noted that in
 15     humans the cancer induced by benzene exposure is  predominantly acute nonlymphocytic
 16     leukemia, while in rodents, lymphocyte leukemia was induced in two series of experiments in
 17     C57BL/6 mice (Snyder et al., 1980) and CBA/Ca (Cronkite et al., 1989).  While the reason for
 18     the difference in lineage of hematopoetic cancers induced in the mouse and humans is not fully
 19     understood, it may be related  to differences in hematopoesis.  Lymphocytes are a larger portion of
20     the nucleated cells in mouse bone marrow than in human bone marrow (Parmley, 1988) and could
 21     simply represent a larger target cell population for benzene metabolites. The target organs for
22     benzene carcinogen!city in rodents are rich in enzymes that may confer tissue sensitivity to
23     benzene, as is the human bone marrow (Low et al., 1989,  1995).  The bone marrow, Zymbal
24     gland, and Harderian gland all contain peroxidases, which can activate phenols to toxic quinones
25     and free radicals. Sulfatases,  which remove conjugated sulfate and thus reform free phenols, are
26     also present at high levels in these target organs. The selective distribution of these two types of
27     enzymes in the body may explain the accumulation of free phenol, hydroquinone, and catechol in
28     the bone marrow and the target organ toxicity of benzene in humans and animals. Therefore, the
29     animal bioassay results have some relevance to human leukemia.

30     2.3. MODE-OF-ACTION INFORMATION
 31            Much of the toxicology research summarized herein has focused on elucidating the nature
32     of the mechanisms through which benzene exerts its leukemogenic effects. The central issue to
33     integrating the mechanistic data from the laboratory animal experiments with the occupational
34     epidemiologic data to estimate risk of the anticipated ambient low-level human scenario is to
        6/17/97                                    10        DRAFT-DO NOT CITE OR QUOTE

-------
  1     establish whether the mechanisms that are operative in laboratory animals are similar to
  2     mechanisms operative in humans and how to account for the dose dependency of those
  3     mechanisms. That is, understanding the mode of action permits rational extrapolation across
  4     species and from high to low doses.  Characterization of dosimetry, i.e., description of the uptake,
  5     internal disposition, and translation of an exposure concentration to the effective dose at the
  6     target site is necessary. This requires an understanding and description of both physiologically
  7     and metabolically driven pharmacokinetic processes. Pharmacodynamic characterization is also
  8     necessary, i.e., description of the key mechanisms through which the dose at the target site elicits
  9     the ultimate adverse response. Processes such as altered gene regulation, cytotoxicity, and cell
 10     proliferation are processes thought to be important for benzene leukemogenesis. A quantitative
  11     understanding of the mechanisms involved along the exposure-dose-response continuum can aid
 12     in integrating the available data for risk assessment purposes.
 13            Benzene has been established as a human leukemogen, but the mode of action by which it
 14     produces leukemia remains unclear.  This section is devoted to discussing recent data, including
 15     evidence on the role of dosimetry and toxicant-target interactions, that may help to elucidate a
 16     mode of action for benzene-induced leukemia.

 17     2.3.1.  Mutagenicity and Genotoxicity
 18            Benzene generally has yielded negative results in gene mutations assays in bacteria or in
 19     vitro mammalian cell systems (Ashby et al., 1985; Oberly et al., 1984,  1990). However, Ward et
20     al. (1992) reported dose-related increases in mutations at the hprt locus in lymphocytes of CD-I
 21     mice exposed to benzene (40, 100, and 1,000 ppb) by inhalation for 6 weeks (22 h/day, 7
22     days/week). Also, Mullin et al. (1995) detected increased mutant frequencies in the lad
23     transgene from lung and spleen but not liver from C57BL/6 mice exposed to 300 ppm benzene for
24     6 h/day, 5 days/week for  12 weeks. The literature on the genotoxic effects of benzene is
25     extensive with more than 220 publications with original data.  Reviews of the earlier literature
26     (Dean, 1978, 1985) present clear evidence that benzene exposure results in chromosome
27     aberrations in a variety of in vitro and in vivo assays and in persons occupationally exposed to
28     benzene over long periods of time.
29            Aneuploidy, the loss and gain of whole chromosomes, is common in myeloid malignancy.
30     Patients with benzene-induced leukemia, rodents, and human cells treated in vitro display
 31     increased aneuploidy.  Numerical changes in the C-group chromosomes 6-12 and X have been
32     detected in the blood and bone marrow of patients with benzene-induced myelogenous leukemia,
33     myelodysplastic syndrome, and pancytopenia (Vigliani and Forni, 1976). A recent report by
34     Zhang et al. (1996a) showed that the induction of aneuploidy of chromosome 9 as measured by
35     fluorescence in  situ hybridization (FISH) in interphase lymphocytes from benzene-exposed
        6/17/97                                    11        DRAFT-DO NOT CITE OR QUOTE

-------
  1     workers is significantly elevated only at high levels of exposure (>31 ppm in air).  However, an
  2     unpublished study has shown that the induction of aneuploidy of other chromosomes (e.g.,
  3     chromosome 7) occurs at lower doses and that the effect of benzene on hyperdiploidy of
  4     chromosomes 7, 8, and 9 shows a significant linear trend (Zhang et al.,  1996b).  The human
  5     evidence for aneuploidy induction also is supported by in vitro experiments.  Hydro quinone and
  6     1,2,4-benzenetriol induce aneuploidy of chromosomes 7 and 9 in human cells (Zhang et al., 1994;
  7     Eastmond et al., 1994). Eastmond and co-workers also have reported that micronuclei containing
  8     centromeres are formed in spleen cells following oral benzene exposure in mice (Chen, et al.,
  9     1994). Centromere-containing micronuclei are thought to be formed when a whole chromosome
 10     is lost during mitosis.  Thus, considerable evidence supports the assertion that benzene and its
  11     metabolites are able to produce aneuploidy in a variety of systems.
 12            In addition to causing loss and gain of whole chromosomes, benzene exposure causes
 13     clastogenicity.  Recent studies using new methods have shown that benzene metabolites induce
 14     strand breaks in human cells (Plappert et al., 1994; Anderson et al., 1995). Further, experiments
 15     in rodents have provided consistent evidence from a number of studies that benzene exposure
 16     causes increased frequency of micronucleated cells (summarized in ATSDR, 1996).  Micronuclei
 17     also are seen in human cells exposed in vitro to various metabolites and combinations of
 18     metabolites (Zhang et  al., 1993; Eastmond, 1993; Yager et al., 1990; Hogstedt et al., 1991;
 19     Robertson et al., 1991). Synergistic increases in micronuclei were induced by catechol and
20     hydroquinone, but not catechol and phenol or phenol and hydroquinone (Robertson et al., 1991).
 21     However, in mice treated intraperitoneally with binary or ternary mixtures of these three
22     metabolites, synergistic effects resulted only from mixtures of phenol and hydroquinone
23     (Marrazzini et al., 1994); adding catechol to the mixture was no more effective than hydroquinone
24     alone in inducing micronuclei. Chen and Eastmond (1995) corroborated the phenol and
25     hydroquinone synergy. Using an antikinetichore-specific antibody and FISH, they demonstrated
26     that both chromosome breakage and loss were induced and that the relative  frequency of these
27     events were indistinguishable whether mice were treated with benzene (440  mg/kg) or the binary
28     mixture of hydroquinone and phenol (60/160 mg/kg).  There is also human evidence of
29     clastogenicity from reports of unstable chromosome aberrations in exposed humans (Aksoy,
30     1989; Forni, 1971; Sarto et al., 1984; Sasiadek, 1992; Tompa et al.,  1994; Van den Berghe et al.,
 31     1979). For example, one report found that lymphocytes of exposed workers had increased
32     frequency of chromosome aberrations, most of which were acentric fragments, presumably the
33     products of double strand breakage (Sarto et al., 1984).
34            Growing evidence is beginning to implicate benzene in producing the chromosomal
35     rearrangements  associated with AML and myelodysplastic syndromes, such as interstitial deletion,
36     inversion, or translocation. Earlier studies of patients with benzene-induced hematopoietic
        6/17/97                                    12        DRAFT-DO NOT CITE OR QUOTE

-------
  1     disorders demonstrated increased chromosome aberrations in lymphocytes and bone marrow cells
  2     (Dean, 1985).  The rearrangements observed included stable and unstable aberrations.  Recent
  3     evidence that may relate to the ability of benzene to induce rearrangement comes from  studies in
  4     which the glycophorin A (GPA) gene mutation assay was used to examine the type of mutations
  5     produced by benzene in human bone marrow (Rothman et al., 1995).  The GPA assay measures
  6     somatic cell mutation frequency in peripheral erythrocytes.  Because mature erythrocytes lack a
  7     nucleus, mutations expressed in these cells must have occurred in precursor erythroid cells or
  8     stem cells in the bone marrow.  The assay detects a spectrum of mutational mechanisms, but
  9     results show that the most significant increase in benzene-exposed persons were changes that
 10     arose through gene conversion, such that heterozygous individuals became homozygous for one
  11     of their two alleles. Precisely how this type of genetic change occurs is not measured by the
 12     assay, but a likely mechanism is mitotic recombination after damage of one allele.  These data do
 13     not directly address the question of whether benzene can cause translocations and other structural
 14     aberrations; however, changes at the GPA locus of this type do indicate that interchromosomal
 15     exchange occurred. Therefore, while the evidence is not currently strong,  there is reason to
 16     suspect that benzene may be able to induce rearrangement in concert with its clastogenic  and
 17     anueploidogenic properties.
 18            DNA adducts of phenol, hydroquinone, or benzoquinone have been reported in a number
 19     of in vitro systems (Reddy et al., 1990; Levay et al., 1993; Bodell et al.,  1993). Reddy et al.
20     (1990) did not detect DNA adducts in rat bone marrow, Zymbal gland, liver, or spleen after four
 21     daily  gavage treatments of phenol or a 1:1 mixture of phenol and hydroquinone. Subsequently,
22     the  same group (Reddy et al., 1994) did not detect DNA adducts in liver, bone marrow, or
23     mammary glands of mice sacrificed  after receiving four daily intraperitoneal (i.p.) injections of 500
24     mg/kg benzene.  Using the same Pi-enhanced P32-postlabeling procedure, Pathak et al. (1995)
25     performed a series of experiments using concentrations ranging from 25  to 880 mg/kg and
26     treatments ranging from a single i.p. injection to daily injections for up to 14 days as well as in
27     vitro experiments with hydroquinone or 1,2,4-benzenetriol. One major and two minor DNA
28     adducts were detected in the bone marrow of mice receiving i.p.  injections of 440 mg/kg  of
29     benzene twice  a day for 3 days. No adducts were seen with any treatment regimen involving only
30     a single injection per day, even at 880 mg/kg for 3 days. Co-chromatography indicated that the
 31     adducts were identical to those seen after in vitro treatment of bone marrow with hydroquinone.
32     Using the same treatment regimen, the same adducts were detected in white blood cells of mice
33     (Levay et al., 1996). More recently, data obtained using accelerated mass  spectrometry, which
34     has tremendous sensitivity, show that the formation of protein and DNA adducts in mouse bone
35     marrow is linear over 8 orders of magnitude to doses as low as 700 pg/kg  (Turtletaub et al.,
36     1996). Indeed, the dose range that can be studied by this technique is remarkable, and  benzene
        6/17/97                                    13        DRAFT-DO NOT CITE OR QUOTE

-------
  1     was shown to produce a linear dose-response curve for adduct formation between a dose
  2     equivalent to one cigarette to doses equivalent to high occupational levels of exposure. Using the
  3     "comet assay," which detects strand breaks and alkaline labile sites in DNA, Plappert et al. (1994)
  4     observed damage in bone marrow with 100 ppm benzene in mice exposed 6 h/day for 5 days.

  5     2.3.2.  Metabolism
  6            The first critical event in benzene carcinogen!city is conversion of benzene to active
  7     metabolites (figure 1). Benzene is first metabolized in the liver, mainly via cytochrome P4502E1.
  8     The major oxidation product is phenol, which is either conjugated, primarily to phenyl sulfate in
  9     humans, or further hydroxylated by P4502E1 to hydro quinone  (Snyder and Kalf,  1994).
 10     Hydroquinone is then secondarily converted to other highly toxic products (discussed below).
  11     Other major products of primary benzene metabolism include catechol and trans, trans-muconic
 12     acid (Witz et al., 1990a). The latter is presumed to be formed from the ring opening of benzene
 13     epoxide via benzene oxepin, or perhaps benzene dihydrodiol. The intermediate product trans,
 14     fram'-muconaldehyde has genotoxic properties and could play a role in benzene toxicity (Witz et
 15     al., 1990b).  However, the selective toxicity of benzene to blood and bone marrow is unlikely to
 16     be explained by trans, /ram'-muconaldehyde alone. Further, it is unlikely that significant
 17     quantities of trans, /ram'-muconaldehyde escape hepatic glutathione and are transported to the
 18     bone marrow. Studies have shown that little or no trans, /rara'-muconaldehyde is likely to leave
 19     the liver, bringing its role in benzene toxicity into question (Brodfuehrer et al., 1990).
20            Secondary metabolism of the phenolic products of benzene generally is regarded as a
 21     critical aspect of benzene toxicity. All of the phenolic metabolites of benzene can be oxidized by
22     myeloperoxidase and other peroxidase enzymes to their active quinones and semiquinone radicals
23     (Smith et al., 1989; Subrahmanyam et al., 1991).  These species are highly toxic by directly
24     binding to cellular macromolecules and/or generating oxygen radicals through redox cycling.
25     There is now strong evidence that the quinone products of secondary metabolism and related free
26     radicals are the ultimate toxic metabolites of benzene.  Specifically, 1,4-benzoquinone and its
27     semiquinone radical, derived from hydroquinone, are likely to be the most critical toxic
28     intermediates. The conversion of phenol to diphenoquinone and radical intermediates also could
29     play an important role, as could oxidation products of 1,2,4-benzenetriol. Although formed in
30     small amounts,  1,2,4-benzenetriol has potent effects (Zhang et al.,  1993, 1994).  The role, if any,
 31     of 1,2-benzoquinone derived from catechol remains unclear at this time.
        6/17/97                                     14        DRAFT-DO NOT CITE OR QUOTE

-------
       1,2,4-Benzenetriol
       NQ01
 I	1
 I MPO |
,1	1
                                     Benzene
                                 CYP2E1
                                  Benzene Oxide
                                                                 Toxicity
                                     Phenol
                                 CYP2E1
                                  Hydroquinone
                                                                   \
                                                            t,t - Muconaldehyde
                      t,t - Muconic acid
                         Catechol
                                NQO1
I	1
I MPO |
I	1
                                                           NQ01
                               I	1
                               |  MPO |
                              ,L_J
   2-hydroxy-1,4-benzoquinone      1,4-Benzoquinone
                     1,2-Benzoquinone
                                  TOXICITY
            Figure 1. Key metabolic activation pathways in benzene toxicity.
6/17/97
15
                                DRAFT-DO NOT CITE OR QUOTE

-------
  1            Recent human studies show that the dose-response curve for benzene-induced leukemia
  2     has been shown to be supralinear because the formation of toxic metabolites plateaus above 25
  3     ppm benzene in air (Rothman et al., 1996; Bechtold et al., 1996). The risk ratios for
  4     hematopoietic malignancies tend to remain somewhat constant albeit significantly elevated at
  5     exposure levels ranging from less than 10 ppm-years to over 400 or more ppm-years based on
  6     Chinese cohort data (table 2 in Hayes et al., 1996). On the other hand, the Rinsky et al. (1981,
  7     1987) data show a definite dose-response relationship in pliofilm workers (table 1) from 40 to
  8     over 400 ppm-years of exposure. The Chinese data are not inconsistent with the knowledge that
  9     massive exposure to benzene suppresses the hematopoietic system.  However, with respect to
 10     levels of benzene below 10 ppm, the dose is predicted to be linear with the risk of leukemia (Bois
  11     et al., 1996).  This is  based on three cases, however, and there are insufficient human data at such
 12     low levels to validate the supposition.  Most human studies deal with subjects exposed to much
 13     greater levels of benzene.
 14            There has been a considerable amount of progress in understanding and quantifying the
 15     factors that contribute to the distribution and metabolism of benzene and its metabolites in
 16     experimental animal  species (Schlosser et al., 1993, 1995; Medinsky et al., 1994; Low et al.,
 17     1995). The quantity  of benzene metabolites produced is the result of subtle interplay of oxidation
 18     and conjugation pathways and distribution of enzyme systems in the liver and other organs as well
 19     as relative rates of perfusion in different organs and different species.  These differences have been
20     explored using a physiologically based pharmacokinetic model (Schlosser et al., 1995; Medinsky
 21     et al., 1996), but their application in predicting metabolism and dosimetry in humans remains a
22     subject of considerable debate.

23     2.3.3. Pathogenesis
24            Lymphohematopoietic neoplasia can be defined as uncontrolled proliferation or expansion
25     of lymphohematopoietic cells that no longer have the capacity to differentiate normally to form
26     mature blood cells. Clones derived from the myeloid lineage are designated as chronic or acute
27     leukemias. Within these general classes, leukemias represent a heterogeneous group of diseases.
28     Heterogeneity is apparent even within the group classified as acute myelogenous leukemia
29     (AML). Myelodysplastic syndromes (MDS)  consist of a group of blood disorders with defects in
30     hematopoietic maturation.  They are considered as  preleukemic because a significant portion of
 31     these progress to frank leukemia (Wright, 1995). Consistent with present models for the origin
32     and progression of neoplasia, development of leukemia is thought to be a multistep process that
33     involves several independent genetic and epigenetic events.  Cell survival, differentiation, and
34     proliferation are regulated processes under coordinated control by multiple factors in normal
35     hematopoiesis.  Irons and Stillman (1996) have summarized much of the extensive literature that
        6/17/97                                     16        DRAFT-DO NOT CITE OR QUOTE

-------
  1     exists relating to secondary leukemia involving either therapy or occupational exposures. It is
  2     generally recognized that chromosomal aberrations or deletions can alter the regulation and
  3     function of protooncogenes and other growth-promoting genes. Clonal chromosome aberrations
  4     involving more than 30 different abnormalities have been identified in the majority of patients
  5     diagnosed with AML (Caligiuri et al., 1997). In secondary leukemias associated with alkylating
  6     agent antineoplastic therapy, loss of genetic material from chromosomes 5 and 7 are found in the
  7     great majority, while leukemias following topoisomerase II inhibitory drugs more frequently
  8     involve aberrations involving chromosome band 1 Iq23 (Pedersen-Bjergaard et al., 1995).  Several
  9     interleukin genes (IL-3, IL-4, IL-5), granulocyte/macrophage-colony-stimulating factor (GM-
 10     CSF), and other regulatory genes  are tightly linked on chromosome 5. Irons and Stillman (1996)
  11     described a model for benzene-induced leukemia based on the disrupted functions of these  genes.
 12     Young and Saha (1996) discuss several different translocations, all involving 1 Iq23. The gene at
 13     this location has been sequenced and has been designated MLL (mixed-lineage leukemia) and
 14     while the normal function of this gene has yet to be determined, it shares homology with  the
 15     Drosophila trx gene that regulates transcription of genes for normal development. Although
 16     many leukemias have one chromosomal rearrangement in all cells, cytogenetically unrelated
 17     clones are more frequently found in secondary leukemias than in de novo leukemias (Heim, 1996).
 18     Despite these complexities, a growing knowledge of the function and role of cytokines, their
 19     receptors, protooncogenes, and suppressor genes can provide a useful framework for analysis of
20     the respective roles of altered cell growth and differentiation in leukemogenesis.

 21     2.4.  HAZARD CHARACTERIZATION SUMMARY
22            This document reconfirms that benzene is a known human carcinogen by all routes  of
23     exposure (U.S. EPA, 1979, 1985). This finding is supported by evidence from three different
24     areas: human epidemiologic studies,  animal data, and improvement in understanding of
25     mechanisms of action. Human epidemiologic studies of highly exposed occupational cohorts have
26     demonstrated unequivocally that exposure to benzene can cause acute nonlymphocytic leukemia
27     and other blood disorders, that is,  preleukemia and aplastic anemia (Aksoy, 1976, 1977; Aksoy et
28     al., 1974; Infante et al., 1977; Rinsky et al., 1981, 1987; Vigliani and Saita, 1964; IARC, 1982;
29     ATSDR, 1996).  It is also likely that exposure is associated with a higher risk of chronic
30     lymphocyte leukemia and multiple myeloma (DeCoufle et al., 1983; Rinsky et al., 1987). In
 31     experimental animal species, benzene exposure (both inhalation and oral routes) has been found to
32     cause cancer in multiple target organ sites such as oral and nasal cavities, liver, forestomach,
33     preputial gland, lung, ovary, and mammary gland (section 2.2). It is likely that these responses
34     are due to interactions of the metabolites of benzene (section 2.3.1).  Recent evidence suggests

        6/17/97                                    17        DRAFT-DO NOT CITE OR QUOTE

-------
  1     that there are likely multiple mechanistic pathways leading to cancer and in particular
  2     leukemogenesis from exposure to benzene (section 3.2).
  3            Additionally, changes in blood and bone marrow consistent with hematotoxicity are
  4     recognized in humans and experimental animals.  Clinical outcomes observed are leukopenia,
  5     thrombocytopenia, anemia, and aplastic anemia (ATSDR, 1996).  Benzene induces peripheral
  6     blood abnormalities and disrupts hematopoiesis at separate compartments of blood cell formation
  7     (i.e., white, platelet, and red) (ATSDR,  1996). Granulocytic and erythropoietic progenitor cells
  8     are significantly depressed.  Chromosomal breakage and loss are increased in mice from exposure
  9     to benzene or its metabolites, a mixture of phenol and hydroquinone (section 2.3.2).
 10            The metabolic studies summarized herein suggest that in both laboratory animals and
  11     humans, benzene metabolism exhibits dose-dependent behavior, with the proportion  of the
 12     metabolites formed changing considerably depending on the dose of benzene administered.
 13     Benzene metabolism also has been reported to be modulated by coexposure  or prior exposure to
 14     other organic chemicals (Medinsky et al., 1994).
 15            Benzene affects bone marrow cells in  several different ways.  These effects are produced
 16     by synergistic interaction of multiple metabolites.  Genotoxic effects are a critical component of
 17     the leukemogenic properties of benzene. As more information becomes available about the
 18     epigenetic effects of benzene and the role these effects play in the leukemogenic process in
 19     general, it is likely that these will be shown to have an important role. Evidence supports the
20     hypothesis that more than one toxic effect contributes to the leukemogenic process, especially
 21     because benzene metabolic products may be able to cause general disruption of protein functions
22     in bone marrow cells. Protein damage is likely to result in pleiotropic effects, including general
23     toxicity, alteration of growth factor responses, and DNA damage. Therefore, the overall picture
24     of benzene-induced leukemogenesis is an increased rate of genetic damage to hematopoietic cells
25     that occurs in the context of disrupted bone marrow biology. This situation could encourage not
26     only the production of cells with key genetic changes, but also the selection and expansion of such
27     cells due to the abnormal marrow. However, data are not sufficient at this time to state precisely
28     which of the various documented effects, genotoxic or otherwise, are the critical ones for
29     benzene-induced leukemogenicity.
        6/17/97                                     18         DRAFT-DO NOT CITE OR QUOTE

-------
                  3. DOSE-RESPONSE ASSESSMENT AND CHARACTERIZATION

  1            In the earlier EPA benzene risk assessment document, (U.S. EPA, 1985), the lifetime
  2     leukemia risk due to 1 ppm of benzene in air was estimated to be 2.6 x 10"2.  This risk number is
  3     the geometric mean of risk estimates that were calculated on the basis of data from one study on
  4     pliofilm workers (Rinsky et al., 1981) and two studies of chemical workers (Wong et al., 1983;
  5     Ott et al., 1978).  On the basis of Rinsky et al.'s (1981) data alone, the risk due to 1 ppm of
  6     benzene in air was estimated to be 4.1 x 10"2 when the  relative risk model was used, and 1.8 x 10"
  7     2 when the additive risk model was used.
  8            Subsequently, several risk assessments on the basis of Rinsky et al.'s (1981) cohort have
  9     become available (Thorslund 1988, 1993; Brett et al., 1989;  Crump 1992; Paxton et al., 1994;
 10     Cox, 1996).  More than  100 individual risk  estimates using varying assumptions and/or models
  11     have been presented, with outcomes ranging more than 6 orders of magnitude at 1 ppb exposure.
 12            Two dose-response models, relative and absolute risk models, were used to calculate
 13     benzene risk estimates using epidemiologic data in the  1985  EPA document.  In fitting the dose-
 14     response models, person-years of observation are divided into subgroups according to the
 15     benzene dose (ppm-year). Let O; be the number of leukemia deaths observed in group I, E; the
 16     expected number of leukemia deaths in the ith group based on the mortality rates in a comparison
 17     population, d; the average benzene dose in the ith group, and  Y; the number of person-years in the
 18     ith group. The relative risk model is of the form

 19                                        E(O;) = aE;(l+bd;)

20     and absolute risk model  of the form

 21                                      E(0;) = E; + (a+bd;)Y;

22     where E(O;) is the expected number of leukemia deaths in the ith dose group under the respective
23     model. The parameters  a and b are estimated from cohort data under the assumption that the
24     number of observed leukemia deaths, O;, is  a Poisson random variable with the expected value
25     given by one of the two  models above.  The parameter b represents the potential of benzene to
26     induce leukemia per unit dose (ppm-year).  Once an estimate of the parameter b is obtained, it
27     was translated into a unit risk (i.e., lifetime risk per unit of ambient air exposure in ppm or |ig/m3)
28     by a straightforward mathematical manipulation that depends on whether the model is absolute or
29     relative risk.


        6/17/97                                   19        DRAFT-DO NOT CITE OR QUOTE

-------
  1            The unit risk estimate of 2.6E-2 per ppm was based on the report by Crump and Allen
  2     (1984).  Because of the lack of information on exact exposure conditions for individual members
  3     of the cohort, cumulative dose (ppm-years) was used by Crump and Allen to construct dose-
  4     response models. Clearly, the use of cumulative dose is less desirable than the use of actual
  5     concentration (ppm). Its impact on risk estimates, however, is difficult to assess without knowing
  6     the exact exposure concentrations for individuals in the cohort.

  7     3.1.  DESCRIPTION OF DIFFERENT RISK ASSESSMENTS
  8            Differences between these risk estimates largely derive from differences in the
  9     determination of the exposure estimates used in the dose-response modeling.  Rinsky et al. (1981,
 10     1987), Crump and Allen (1984), and Paustenbach et al. (1992, 1993) chiefly center on the levels
  11     that existed in the plants where the Pliofilm workers were employed before 1946.  Paustenbach et
 12     al. (1992, 1993) assumed that the samples taken after 1946 underestimated actual levels chiefly
 13     because inadequate measuring devices were used, asserting that these devices consistently
 14     underestimated exposure by as much as 50%. It was further assumed that the working week was
 15     on the average 51 hours for the Pliofilm workers, not the 40 hours usually assumed. Other
 16     assumptions are also given to justify his high exposure estimates (Paustenbach et al., 1992, 1993).

 17            Much controversy exists concerning the levels of benzene that permeated the workplace
 18     during the early employment years of the Pliofilm workers. It has not been established what those
 19     levels were in the period from the late 1930s until 1946.  Actual measurements do not exist before
20     1946 when most of the Pliofilm workers were employed including most of the leukemia victims.
 21     After 1946 and into the 1960s, few measurements of actual benzene exposure were taken and in
22     many instances they were taken in areas where it was known that high levels of benzene would be
23     found. Rinsky et al. (1981) maintains that the average exposure to the workers were "within the
24     limits considered permissible at the time of exposure." Rinsky agrees that peak exposure to high
25     levels of benzene probably  did occur but, unfortunately, there is no information regarding when
26     these peak exposures occurred and how large they were for individual members of the cohort.
27     Several leukemia victims are listed as being exposed to as much as 40 ppm 8-hr TWA during their
28     early years with the company. It is believed that actual levels were probably within the range of
29     35 to 100 ppm during the course of their employment during those early years.  These levels
30     tended to drop in time as efforts to improve air quality in the plants were implemented.
 31            Using a simplified version of a model developed by Moolgavkar and Knudson (1981),
32     Thorslund (1988) presented several risk estimates, all of which are at least an order of magnitude
33     smaller than the EPA risk numbers.

        6/17/97                                   20        DRAFT-DO NOT CITE OR QUOTE

-------
  1            Both Brett et al. (1989) and Paxton et al. (1994) assumed that rate ratio (RR) is related to
  2     exposure (ppm-year) by RR(d)=exp(b*d), where d is exposure in ppm-year, and b is a parameter
  3     to be estimated (the two assessments differ in the way the parameter b was estimated). Only risk
  4     estimates due to occupational exposure (i.e.,  8/h day, 5 days/week, 50 weeks/year) were
  5     presented. To calculate the lifetime risk due  to continuous exposure of 1 ppm (i.e., d=76 ppm-
  6     years), the parameter b is multiplied by a factor of (24/8)x(7/5)x(52/50). The resultant risks at 1
  7     ppb and 1 ppm are  given in table 4.
  8            Crump (1992) presented 96 dose-response analyses by considering different factors such
  9     as (1) different disease end points, (2) additive or multiplicative models, (3) linear/nonlinear
 10     exposure-response  relationships, (4) two exposure measurements (Crump and Allen [1984] vs.
  11     exposure estimates by Paustenbach eventually published in Paustenbach et al. [1993]), and (5)
 12     cumulative or weighted exposure measurements. The risk estimates range from 8.6 x 10"11 to 2.6
 13     x 10"5 at 1 ppb of benzene air concentration and 8.6 x 10"5 to 2.5 x 10"2 at 1 ppm of benzene air
 14     concentration.  The largest deviation from the EPA risk number (U.S. EPA, 1985) was obtained
 15     when a nonlinear model and Paustenbach et al. (1993) exposure estimates were used. When a
 16     linear model was used, risk estimates ranged  from 7.1 x icr3 to 2.5 x 10"2 at 1 ppm, regardless of
 17     which exposure measurements were used. When a linear model and Crump and Allen (1984)
 18     exposure measurements were used, the risk at 1 ppm ranged from  1.1 x icr2 to 2.5 x 10"2. These
 19     are close to the 1985 EPA risk estimates. As previously stated, the use of the updated Rinsky et
20     al. (1987) cohort would not significantly alter risk estimates if the same exposure-response model
 21     and exposure estimates are used. The  single  factor that affects the risk estimate most is the
22     assumption of nonlinearity. If low-dose linearity is assumed, consideration of other factors (e.g.,
23     new exposure estimates) will result in no more than a fivefold difference from the existing EPA
24     risk number. A subset of calculations  of Crump (1992) appear in Crump (1994).
25            Thorslund (1993), departing from his 1988 risk assessment, later used a more
26     conventional approach (i.e., additive risk model) to calculate risks. The newer risk estimate at 1
27     ppm (Thorslund, 1993) with the linear-quadratic model was increased about eight times from his
28     previous report (Thorslund, 1988) of 1.0 x 10'3 to 7.8 x 10'3, utilizing the same (Crump and
29     Allen, 1984) exposure data. Also, the newer estimate,  7.8 x 10"3 per ppm, is one-half that of the
30     1985 EPA estimate of 1.8 x 10'2 (based only  on the Rinsky  et al. [1981] cohort and Crump and
 31     Allen [1984] exposure estimates) when the same (additive) model was used.  Only AML was used
32     in Thorslund's (1993) calculations, although other leukemia cell types may be associated with
33     benzene exposure.  When the linear-quadratic model was used with the Paustenbach et al. (1993)
34     exposure estimates, the risk at 1 ppm was estimated to be 4.7 x 10"3 (note that a 95%
        6/17/97                                    21        DRAFT-DO NOT CITE OR QUOTE

-------
Table 4. Risk estimates calculated on the basis of Pliofilm workers by various investigators
Source
U.S. EPA, 1985
Thorslund, 1988
Brett etal., 1989
Paxtonetal., 1994
Crump, 1992
Thorslund, 1993
Risk at 1 ppm
l.Sxlcr2
4.1xlO'2
1.4xlO'4
l.OxlO'3
3.2xlCT3
3.5xlO'3
5.2xlO-3to2.5xlO-2
4.3xl04to8.1xl04
2.2xlCr3
4.6xlO'3
l.SxlO'2
I.lxl0-2to2.5xl0-2
5.4xlO-3to2.5xlO-2
7.1xlO-3tol.5xlO-2
8.6xlO-5to6.5xlO-3
1.2xlCr2
7.8xlCr3
5.5xlO'3
4.7X10-31
Risk at 1 ppb
l.Sxlcr5
4.1xlO'5
1.4xl040
l.OxlO'6
3.2xlCT6
3.5xlO'6
3.9xlO-6tol.lxlO-5
2.9xlO-5to3.4xlO-5
1.9xlCr6
3.5xlO'6
8.9xlO'6
l.lxlO-5to2.5xlCr5
4.5xlO-6to2.6xlO-5
7.2xlO-6tol.6xlO-5
8.6xl041 to 5.6xlCT6
1.2xlO'5
7.8xlO'6
S.SxlO'6
4.7xlO'6
Exposure


Crump and Allen, 1984
Crump and Allen, 1984
Crump and Allen, 1984
Crump and Allen, 1984
Rinsky et al., 1984
Rinsky et al., 1984
Crump and Allen, 1984
Paustenbach et al., 1993
Rinsky et al., 1984
Crump and Allen, 1984
Crump and Allen, 1984
Paustenbach et al., 1993
Paustenbach et al., 1993
Crump and Allen, 1984
Crump and Allen, 1984
Paustenbach et al., 1993
Paustenbach et al., 1993
Model
Additive risk
Relative risk
Quadratic
Linear quadratic
One stage/one
hit
Two stage/two
hit
Conditional
logistic
Conditional
logistic



Linear
Nonlinear
Linear
Nonlinear
Linear, additive
Linear quadratic,
additive
Linear, additive
Linear quadratic,
additive
a95% upper bound is provided because of instability of the maximum likelihood estimate (linear coefficient
was estimated to be 0).
6/17/97
22
DRAFT-DO NOT CITE OR QUOTE

-------
  1     upper bound must be used for the later risk number because of the instability of the point estimate;
  2     the linear coefficient was estimated to be 0).  These risk estimates are about two to four times
  3     smaller than the corresponding EPA additive risk estimate of 1.8 x 10"2, depending on whether the
  4     Crump and Allen (1984) or the Paustenbach et al. (1993) exposure estimates were used.
  5            Recently, Cox (1996) reassessed benzene risks using internal doses and Monte-Carlo
  6     uncertainty analysis. He reexamined the physiologically based pharmacokinetic models of
  7     benzene metabolism in animals and humans, and a Monte-Carlo uncertainly analysis based on
  8     maximum-entropy probabilities.  Bayesian conditioning was used to develop an entire probability
  9     distribution for the true but unknown dose-response function.  He concluded that the excess risk
 10     due to benzene exposure may be nonexistent (or even negative) at sufficiently low doses.
  11            A need exists to further support these conclusions based on additional research on
 12     biological  mechanisms of benzene-induced hematopoiesis and leukemia rather than on statistical
 13     modeling uncertainties alone.

 14     3.2.  SHAPE OF THE DOSE-RESPONSE FUNCTION AT LOW DOSES
 15            Too many questions remain about the mode of action for benzene-induced leukemia for
 16     the shape of the dose-response function to be known with certainty. While much progress has
 17     been made in the past few years and a reasonable hypothesis can be generated for the mechanism
 18     of benzene-induced leukemia, it remains simply a hypothesis.  Arguments for and against the
 19     dose-response curve being nonlinear at low doses are presented in summary form in table 5.
20            Analysis of the Rinsky et al. (1987) data shows that at doses less than 40 ppm-years, the
 21     SMR for leukemia was 1.1 and is not significantly elevated. This has prompted some
22     investigators to suggest that benzene has a threshold for leukemia induction of about 40 ppm-
23     years. However, this analysis of leukemia dose-response is based on only nine cases of leukemia,
24     limiting its value for dose-response analysis.  In addition, only six of these cases were AML.
25     Further, Rinsky et al. (1987) showed a clearly increased SMR  for multiple myeloma at doses
26     below 40 ppm-years, and in a larger Chinese study, involving more than 30 cases, leukemogenic
27     effects of benzene were observed at exposures well below 200 ppm-years (Yin et al., 1989).
28     These observations suggest, as expected, that it is difficult to determine the shape of the dose-
29     response function based on occupational studies alone.
30            As indicated previously, benzene is not a classic carcinogen, that is, its metabolites are not
 31     genotoxic  in simple mutation assays.  It most likely produces leukemia by chromosomal damage
32     rather than simple point mutations.  An argument can be made for nonlinearity on the basis that
33     the induction of chromosome damage by benzene and its metabolites is nonlinear and
        6/17/97                                    23        DRAFT-DO NOT CITE OR QUOTE

-------
              Table 5.  Evidence that benzene-induced leukemia is nonlinear at low doses
Category
Genotoxicity
Theoretical
Pro
Micronucleus induction by benzene and its
metabolites in the mouse bone marrow and in human
cells in vitro is nonlinear.
The induction of aneuploidy of chromosome 9 is
nonlinear and is significant only at high levels of
exposure (>31 ppm in air) (Zhang et al., 1996a).
DNA adduct formation is observed by
P32-postlabelling only at high doses.
Oxidative DNA damage may contribute to benzene
genotoxicity (Kolachana et al., 1993) but has a high
rate of repair.
Hematotoxicity is required for leukemia induction,
and this will have a threshold.
If aneuploidy is critical, then leukemia induction is
likely to have a threshold. (Numerous molecules of
benzene metabolites will be required to disrupt
microtubules.)
The cells in the bone marrow have numerous
defense mechanisms.
Con
Micronucleus assay is relatively insensitive and may not show
effects at low doses.
Induction in aneuploidy of other chromosomes (e.g., 7) occurs at
lower doses, and effect of benzene on hyperdiploidy of
chromosomes 7, 8, and 9 shows a significant linear trend.
Data obtained using accelerator mass spectrometry shows that the
formation of DNA adducts in mouse bone marrow is linear to very
low doses.
Errors during repair may cause point mutations.
Hematotoxicity may increase risk of malignancy but has not been
shown to be a prerequisite.
There is a high background of exposure to benzene and its
metabolites. Additional environmental exposure will simply add to
this and be linear. There are also numerous mechanisms of
aneuploidy induction, and aneuploidy is not the only mechanism of
suppressor gene loss and oncogeny activation.
There is a high background exposure to benzene and its
metabolites, so additional exposure could escape defenses.
  1     in some instances shows a threshold.  However, it should be pointed out that the micronucleus
 2     assay of chromosomal damage is relatively insensitive and may not show effects at low doses,
 3     even though some chromosomal damage is occurring.  A recent report by Zhang et al. (1996a)
 4     showed that the induction of aneuploidy of chromosome 9 as measured by FISH in interphase
 5     lymphocytes from benzene-exposed workers is significantly elevated only at high levels of
 6     exposure (>31 ppm in air).  However, as yet unpublished studies have shown that the induction of
 7     aneuploidy of other chromosomes (e.g., chromosome 7) occurs at lower doses and that the effect
 8     of benzene on hyperdiploidy of chromosomes 7, 8, and 9 shows a significant linear trend (Zhang
 9     etal., 1996b).
10            As discussed earlier, bone marrow DNA adducts as detected by P32 postlabeling after in
 11     vivo exposure to benzene correspond with adducts formed by in vitro treatment with
12     hydroquinone or 1,2,4-benzentriol (Pathak et al., 1995). Also, recent data obtained using
13     accelerated mass spectrometry show that the formation of protein and DNA adducts in mouse
14     bone marrow is linear over 8 orders of magnitude to doses as low as 700 pg/kg (Turtletaub et al.,
       6/17/97
24
DRAFT-DO NOT CITE OR QUOTE

-------
  1     1996). Indeed, the dose range that can be studied by this technique is remarkable, and benzene
  2     was shown to produce a linear dose-response curve for adduct formation between a dose
  3     equivalent to one cigarette to doses equivalent to high occupational levels of exposure.
  4            It also has been demonstrated that oxidative DNA damage may contribute to benzene
  5     genotoxicity and thus benzene-induced leukemia (Kolachana et al., 1993; Lagorio et al., 1994).
  6     Because oxidative damage has a high rate of repair and studies in benzene-exposed mice and
  7     human cells in vitro showed that the oxidative DNA damage was rapidly repaired, it could be
  8     argued that this high level of repair will produce a threshold or nonlinearity at low doses.
  9     However, it is errors during this repair process that cause point mutations from oxidative DNA
 10     damage.  Further, because there is already a considerable background level of oxidative damage
  11     (Ames and Shigenaga, 1992), additional damage caused by benzene exposure may induce a linear
 12     increase in point mutations.
 13            It also could be proposed that hematotoxicity is required for leukemia induction. Because
 14     hematotoxicity is likely to have a threshold, it is therefore possible that benzene-induced leukemia
 15     will have a threshold and be nonlinear at low doses. In theory, hematotoxicity may increase the
 16     risk of benzene-induced leukemia, because it could cause quiescent stem cells to enter the cycling
 17     feeder cell stage, thereby expressing any genetic damage. However,  there is no evidence that
 18     hematotoxicity is a prerequisite for leukemia induction. Cases of leukemia following benzene
 19     exposure without previous hematotoxicity have been reported, but the thoroughness of
20     monitoring for hematological effects is always a  question. Benzene recently also has been shown
 21     to have hematological effects below 10 ppm (Ward et al., 1996), and thus the relevance of a
22     threshold for hematotoxicity has decreased in most investigators' estimation.
23            Irons, Subrahmanyam, Eastmond and their co-workers have argued that the induction of
24     aneuploidy is a component of leukemia induction by benzene (Irons and Neptune, 1980;
25     Subrahmanyam et al.,  1991; Eastmond, 1993). If this is true, then it  could be argued that
26     leukemia induction has a threshold because numerous molecules of benzene metabolites would be
27     required to disrupt microtubules and cause aneuploidy. However, it  should be pointed out that
28     there is a high level of background exposure to benzene and its metabolites. Benzene and its
29     metabolites are present in our diet and in cigarette smoke. Additional environmental exposure
30     will simply add to this background. Indeed, McDonald and co-workers have shown that proteins
 31     in both the blood and bone marrow of humans and animals contain high levels of benzene
32     metabolite adducts and that the exposure of animals to benzene causes a linear increase in 1,4-
33     benzoquinone adducts on top of this background (McDonald et al., 1993, 1994).  This additional
34     benzene exposure from the environment is likely to have a linear additional effect on the
35     background. Further, there are numerous mechanisms of aneuploidy induction that do not
36     necessarily involve binding to microtubules, and  aneuploidy is not the only genetic mechanism of
        6/17/97                                   25        DRAFT-DO NOT CITE OR QUOTE

-------
  1     suppressor gene loss and oncogeny activation.  Care must therefore be exercised in claiming that
  2     benzene is nonlinear on the basis of aneuploidy involvement.
  3            Another theoretical argument that can be made is the fact that the cells in the bone
  4     marrow have numerous defense mechanisms to cope with toxic benzene metabolites. However,
  5     as discussed above, there is a high background exposure to benzene and its metabolites and so
  6     additional exposures could actually escape defenses. Indeed, we have calculated that there are
  7     approximately 10,000 benzene molecules per bone marrow cell following normal environmental
  8     background exposures to benzene. The addition of further molecules from environmental or
  9     occupational exposures will simply add to this and may easily overwhelm or escape defense
 10     mechanisms.
  11            Even if there are threshold levels at which each individual experiences increased leukemia
 12     risk, population variability will almost certainly dictate that there is no one threshold dose that
 13     applies across the population of people exposed to benzene.  The data on susceptibility factors for
 14     benzene toxicity and leukemogenicity are growing and will likely shed some light on population
 15     variability in sensitivity to benzene's adverse effects.

 16     3.3.  DOSE-RESPONSE CHARACTERIZATION
 17            The major finding from this update is that it reaffirms the benzene interim unit risk
 18     estimates derived in EPA's 1985 interim risk assessment (U.S. EPA, 1985). The 1985 interim
 19     risk assessment established the probability of humans developing cancer from exposure to 1 ppm
20     of benzene.  Two main concerns had to be addressed before this conclusion could be reached.
 21     The first involves using the updated epidemiologic data from Rinsky et al.'s (1987) cohort of
22     Pliofilm workers as well as selecting the appropriate estimates of exposure for the derivation of
23     the unit risk estimate.  The second major issue involves the continued application of the low-dose
24     linearity concept to the model from which the unit risk estimates are generated.  At present, there
25     is insufficient evidence to reject this concept.
26            The update of Rinsky et al.'s (1987) cohort could have only a limited impact on the
27     existing EPA (1985) interim risk estimates if the same exposure-response (linear) model and
28     Crump and Allen (1984) exposure measurements were used. When the Paustenbach et al. (1993)
29     higher estimated exposure measurements were substituted, the corresponding risk estimates
30     (SMRs) were reduced by  only a factor of, at most, two. The linear-quadratic exposure-response
 31     model used by Thorslund (1993) deals with the concept of low-dose linearity,  and the resultant
32     risk estimates also are not markedly different from the  1985 interim risk EPA estimate.  None of
33     the approaches toward estimating exposure have greater scientific support than any  other because
34     ambient air benzene exposure data did not exist before 1946 in the Pliofilm workers.  There is no
35     clear basis for choosing a single best estimate.  Rather, these sets of risk estimates reflect  both the
        6/17/97                                     26        DRAFT-DO NOT CITE OR QUOTE

-------
  1     inherent uncertainties in the applied model as well as the limitations of the exposure
  2     characterization and response information in the epidemiologic data.
  3            Without extensive analyses of the raw data, only theoretical discussions of the possible
  4     impact on risk estimates under various exposure assumptions and presumed etiologic mechanisms
  5     can be provided. There are two approaches to address this issue; one is to assume a biological
  6     mechanism of benzene-induced leukemia (e.g. to assume that benzene-induced leukemia involves
  7     a sequence of genetic and epigenic changes, and that these steps are effected by benzene
  8     exposure). Another approach is to assume a linear model, as was used in this assessment. Using
  9     the linear model, the use of cumulative exposure would have less impact on the resultant risk
 10     estimate if the concentration was roughly constant during the work history of the cohort.
  11     However, if exposure concentrations in early work history were higher than the later years, the
 12     unit risk could be over-estimated.
 13            While the risk estimates would be significantly different if a nonlinear exposure response
 14     model was found to be more plausible, characterizing the shape (i.e., the nonlinearity) of the
 15     exposure-response curve would still require a better understanding of the biological mechanisms
 16     of benzene-induced leukemia. Some recent evidence suggests the possibility that the low dose
 17     curve could be supra-linear since the formation of toxic metabolites plateaus above 25 ppm
 18     benzene in air (Rothman et al., 1996; Bechtold et al., 1996).  This pattern is similar to that seen in
 19     laboratory animals (Sabourin et al.,  1989) where the effect per unit dose of benzene is less at high
20     doses than at low doses. Thus, it is possible that the unit risk is underestimated if linearity is
 21     assumed at low doses.  The arguments made in favor of benzene-induced leukemia being
22     nonlinear at low doses can be matched by arguments opposing this viewpoint. Currently, there is
23     insufficient evidence to reject a linear dose-response curve for benzene  in the low-dose region,
24     and there is insufficient evidence to demonstrate that benzene is, in fact, nonlinear in its effects.
25     Even if the dose-response relationship were nonlinear, the shape remains to be determined.
26     Because this knowledge is not available at the present time, the Agency's approach of using a
27     model with low-dose linearity is still recommended. Of the various approaches employing a linear
28     assumption, the risk at  1 ppm ranges from 4.7 x 10"3 to 2.5 x 10"2 (table 4).
29            Based on the Rinsky et al. (1987)  study, the risk of leukemia is significantly elevated
30     (SMR=1,186, 95% C.I.=133-4,285) at a dose of 200 to 400 ppm-years (a person exposed to a
 31     level of 5 to 10 ppm for 40 years). This assumes that exposure occurred for only 8 h each day.
32     However, Rinsky et al.'s (1987) data suggest that a rise in the SMR begins at levels under 40
33     ppm-years, although the trend does not attain statistical  significance until the dose of 200 to 400
34     ppm-years is reached.  Therefore, we are  not as confident that the risk begins to rise below 40
35     ppm-years (1 ppm for 40 years).  However, this may be a matter of the lack of power to detect a
36     risk as significant below this level. Wong (1995), in a separate analysis of the risk of only AML in
        6/17/97                                    27       DRAFT-DO NOT CITE OR QUOTE

-------
  1     the Rinsky et al. (1981) cohort, calculated an SMR of 0.91 (1 observed, 1.09 expected) in the
  2     exposure category under 200 ppm-years. However, because AML is a subtype within the
  3     leukemia category, the sensitivity for detecting a significant risk at that level of exposure is much
  4     lower.
  5            On the other hand, recent data from the Chinese cohort (Yin et al., 1989) implies that the
  6     risk of AML is well below 200 ppm-years, although the data analysis is still incomplete.  Out of
  7     30 identified leukemia cases reported in that study, 11 reported cumulative exposure of under 200
  8     ppm-years, and of these 11,7 were subject to average levels of under 5 ppm-years, during the
  9     time that they were exposed.  In fact, Hayes et al.  (1996) added 12 leukemias to this total in an
 10     update of the Yin et al. (1989) study. Interestingly, dosimetry data were calculated  on selected
  11     causes of death in this very same cohort. Hayes et al. (1996) reported that excess risks of death
 12     from hematopoietic malignancies were found at the level of 10 ppm-years cumulative exposure (9
 13     observed vs. 2.5 expected). Unfortunately, length of employment was not provided in the
 14     Chinese cohort.
 15            In addition, Bond et al. (1986a) reported five cases of myelogenous leukemia, four with
 16     cumulative doses of benzene exposure between 1.5 ppm-years and 54 ppm-years. Their average
 17     yearly exposure ranged from 1.0 ppm to 18 ppm.  The Wong study (1987) reported that six of
 18     seven cases of leukemia had cumulative benzene exposures of between 0.6 ppm-years and 113.4
 19     ppm-years. Their average yearly exposure ranged from 0.5 ppm to 7.6 ppm. The seventh case
20     had no measured cumulative dose.  It is possible that peak exposures could have occurred at any
 21     time during employment, however, that information is unavailable.
22            Although the authors of these studies developed dose-response data for some members of
23     their respective cohorts, it was clear that these same workers were subject to other concomitant
24     exposures that were also present in the workplace.  These also might affect the risk  of cancer.
25     Methodological problems also existed in these studies. These considerations precluded the use of
26     these studies in the unit risk calculations.
27            Based on observations from Rinsky et al. (1981, 1987) and recent studies, the Agency is
28     fairly confident that exposure to benzene increases the risk of leukemia at the level of 40 ppm-
29     years of cumulative exposure.  However, below 40 ppm-years, the shape of the dose-response
30     curve cannot be determined based on the current epidemiologic data.  Because of this lack of data
 31     at low levels, the Agency could not confidently say that the risk begins to increase at exposures
32     below 40 ppm-years, although some data seem to suggest that.  Based on the above discussion,
33     the point of departure (POD) is determined at a level of 40 ppm-years in the occupational
34     environment, which assumes 8 h of exposure per working day. Assuming that a 24  h exposure
35     would incur  a threefold risk increase, then from an 8 h exposure, the POD is likely to be one-third
36     of 40 ppm-years, or about 13 ppm-years.
        6/17/97                                   28        DRAFT-DO NOT CITE OR QUOTE

-------
  1            To put this POD of 190 ppb into perspective, environmental surveys completed around
  2     the United States have provided a variety of information on monitored levels of benzene, using
  3     both ambient measurements and personal exposure measurements. (ATSDR [1996] provides a
  4     convenient summary of much of the data.)
  5            Ambient measurements have been made both outdoors and indoors.  Shah and Singh
  6     (1988) report that the Volatile Organic Compound National Ambient Database (1975-1985)
  7     contains the following daily median benzene air concentrations:  workplace air (2.1 ppb), indoor
  8     air (1.8 ppb), urban ambient (1.8 ppb), suburban ambient (1.8 ppb), rural ambient (0.47 ppb), and
  9     remote (0.16 ppb).  The EPA (1987) reports data from 44 sites in 39 cities of the United States,
 10     taken during the 6 to 9  a.m. "morning rush hour" periods during June-September of 1984,  1985,
  11     and 1986. The median concentrations at these sites ranged from 4.8 to 35 ppb, with the authors
 12     noting that mobile sources (motor vehicles) were the major source of ambient benzene in these
 13     samples.  In industrialized areas, Pellizzari (1982) reports outdoor levels of 0.13 to 5 ppb in
 14     Iberville Parish, Louisiana, and Cohen et al. (1989) report median outdoor levels in the Kanawha
 15     Valley region of West Virginia as 0.78 ppb.  Cohen et al. (1989) also report that mean indoor
 16     levels in the study were 2.1 ppb (median = 0.64 ppb, maximum = 14.9 ppb).
 17            The EPA's Total Exposure Assessment Methodology (TEAM) studies showed
 18     consistently that personal exposures to benzene were higher than ambient indoor levels, and that
 19     indoor levels, in turn, were higher than outdoor levels. Wallace (1989) reported that the overall
20     mean personal benzene exposure (smokers and nonsmokers) from the  TEAM data was 4.7 ppb,
 21     compared to an overall mean outdoor ambient level  of 1.9 ppb. Median levels of benzene indoors
22     were broken out by those homes without smokers (mean = 2.2 ppb) and those where one or more
23     smokers were present (mean = 3.3 ppb).  The TEAM authors frequently suggest smoking as a
24     source for indoor benzene concentrations (Wallace,  1987, 1989).  Brunnemann et al. (1989)
25     report that indoor air samples at a smoke-filled bar ranged from 8.1 to 11.3 ppb of benzene.
26     Wester et al. (1986) noted that benzene in the breath of smokers was higher than that of
27     nonsmokers, but that both were higher than the concentrations in outdoor ambient air.
28            Other measurements of benzene concentrations include transient levels of benzene
29     approaching 1 ppm outside a vehicle while refueling (Bond et al., 1986b). Within a parking
30     garage, (Flachsbart, 1992) found a maximum level of 21  ppb. The estimated maximum level in a
 31     basement during the Love Canal situation was about 160 ppb, with levels of about 60 ppb
32     estimated around an uncontrolled hazardous waste site (Bennett, 1987; Pellizzari, 1982). The
33     maximum single personal monitoring  sample, representing one night's exposure, during the 1981
34     New Jersey TEAM study, was 159.6 ppb.
35            Using 190 ppb as the POD, the margin  of exposure (MOE) can be calculated for several
36     of these levels.  Since the 190 ppb is a 70-year lifetime value, this should be  compared with
        6/17/97                                   29        DRAFT-DO NOT CITE OR QUOTE

-------
 1      average levels in whichever exposure scenario is used.  For example, if one assumes that 4.7 ppb
2      is the long-term average exposure for the general population, the MOE would be 190/4.7, or
3      about 40. If one assumes that the ambient indoor levels cited above of 2.2 ppb represent actual
4      exposures to nonsmokers, their MOE would be 190/2.2, or 86.  If one were to construct a
5      hypothetical scenario where a person spent their entire life in a smoke-filled bar, the MOE would
6      drop to a range of 17 to 23.  The MOE for other exposure scenarios can be similarly calculated.
       6/17/97                                    30        DRAFT-DO NOT CITE OR QUOTE

-------
                               4.  CHILDREN'S RISK CONSIDERATIONS

  1            The effects from exposure to benzene can be quite different among subpopulations.
  2     Children may have a higher unit body weight exposure because of their heightened activity
  3     patterns which can increase their exposures, as well as different ventilation tidal volumes and
  4     frequencies, factors that influence uptake.  This could entail a greater risk of leukemia and other
  5     toxic effects to children if they are exposed to benzene at similar levels as adults.  Infants and
  6     children may be more vulnerable to leukemogenesis because their hematopoietic cell populations
  7     are differentiating and undergoing maturation. Many confounding factors may affect the
  8     susceptibility of children to leukemia, for example, nutritional status, lifestyle, ethnicity, and place
  9     of residence. Furthermore, in children, the predominant type of leukemia is lymphatic, while in
 10     adults it is a combination of myeloid and lymphatic. Leukemia formerly classified as a single
  11     disease now has been recognized as several different distinct malignancies that are characterized
 12     by varying age, race, sex, and ethnic group patterns; different secular trends; and  different
 13     etiologic factors (Linet, 1985).
 14            There exist very limited data on children from environmental exposure to  benzene.
 15     Weaver et al. (1996) conducted a pilot study that evaluated the feasibility of using trans, trans-
 16     muconic acid as a biomarker of environmental benzene exposure in urban children. The authors
 17     concluded that muconic acid could be used as a biomarker in children for environmental exposure,
 18     but there have been no studies found that used this biomarker to determine actual benzene
 19     exposure to children.
20            In summary, children may represent a subpopulation at increased risk  due to factors that
 21     could increase their susceptibility to effects on benzene exposure (e.g., activity patterns), on key
22     pharmacokinetic processes (e.g., ventilation rates, metabolism rates and capacities), or on key
23     pharmacodynamic processes (e.g., toxicant-target interactions in the immature hematopoietic
24     system). However, the data to make quantitative adjustments for these factors do not exist at this
25     time.
        6/17/97                                     31        DRAFT-DO NOT CITE OR QUOTE

-------
                                    5.  FUTURE RESEARCH NEEDS

  1            Data insufficiencies in several areas have been noted. Additional research into these areas
  2     will promote a better understanding of how benzene causes cancer, particularly the mechanism of
  3     benzene-induced leukemia. Several classes of data are needed on humans, that is, more extensive
  4     epidemiologic data with good exposure estimation, to permit verification and validation of the
  5     prediction models. Additionally, data on the preleukemic hematology of benzene-exposed
  6     persons, such as the abnormal monoclonality and blood cell counts seen in such persons, would be
  7     a significant contribution.  Specific measures of early genetic damage in humans with known
  8     exposure to benzene will help define the biological events leading up to the disease by providing
  9     internal markers of its progression. This could be potentially useful in risk prediction and assist in
 10     the identification of the steps leading to leukemia induced by exposure to benzene.  Such
  11     information may be forthcoming in the near future from a large cohort of benzene-exposed
 12     workers under study in China.  Investigators from the National Cancer Institute in the United
 13     States, the Chinese Academy of Preventive Medicine, and the University of California at Berkeley
 14     are currently developing such biomarker information as well as gathering clinical data on
 15     hematologic abnormalities.
 16            A need exists to further validate toxicokinetic models and to assess metabolic
 17     susceptibility factors in human subjects. The collection of such information is problematic at best
 18     because it requires exposure of human volunteers to a known carcinogen.  However, data now
 19     being collected in the Chinese cohort on the urinary metabolites of benzene and in vitro studies of
20     cell-specific metabolism and toxicity in defined human bone marrow cell populations may be of
 21     use.
22            Continued basic research in hematopoiesis and leukemia biology is critical. Issues of
23     importance are questions about the cell population that contains targets for leukemic
24     transformation, such as cell number and rate of division, quiescence patterns, maturation,
25     regulation, and apoptotic behavior.  Future understanding of the phenotypic consequences of
26     common genetic aberration in myelodysplastic syndromes (MDS) and AML also is needed to
27     assist in identifying the stages of leukemic transformation.
28            Current uncertainties limit the ability of modeling to explicitly consider all relevant
29     mechanisms, such as the formation of several types of genetic aberrations;  disruption of
30     proliferation, differentiation,  or apoptotic behaviors through genetic change or epigenetic
 31     chemical interference; and the extremely complex and subtle regulation of hematopoietic
32     processes under normal feedback systems.  The future research would be able to quantitatively
33     describe benzene pharmacokinetics in humans, relate dose measures to the above
34     pharmacodynamic mechanisms, and account for observed epidemiologic features of benzene-
        6/17/97                                    32        DRAFT-DO NOT CITE OR QUOTE

-------
  1     induced leukemia, such as patterns of latency and susceptibility. For any mechanistic model of
  2     leukemogenesis to be validated, it must be applied to existing data that relate known human
  3     exposures to the probability of contracting MDS/AML.  While there are epidemiologic data for
  4     benzene, estimation of exposure is a complex task, with considerable uncertainty. Therefore, a
  5     suggested approach is to first develop a biologically based risk model for AML (t-AML).  It
  6     should be recognized that in modeling benzene-induced leukemia in the general population there
  7     is considerable interindividual variability that may influence risk. Some of the genetic factors
  8     important in metabolic variability are becoming known, but other aspects of susceptibility are less
  9     well characterized.  For example, the factors controlling whether patients who suffer benzene-
 10     induced myelosuppression progress to AML or recover after exposure is reduced or removed are
  11     unknown.  To what extent susceptibility factors will dictate leukemia risk and to what extent
 12     leukemia is a manifestation of stochastic processes are not known.
 13            Particular emphasis should be placed on research on those subpopulations who are
 14     believed to be at increased risks (e.g., infants and children, the elderly).  Research is needed to
 15     show how growth, development, and aging affects the risk to humans. In addition, environmental
 16     and epidemiological studies are needed to better determine the environmental exposure levels that
 17     sensitive subpopulations such as pregnant women, infants and children, and the elderly are likely
 18     to encounter on potential benzene risks. Studies  are needed to better understand how the
 19     absorption, distribution, metabolism, and elimination of benzene varies with age, gender, race, or
20     ethnicity and how this information can be modeled to  predict risk to sensitive subpopulations.
 21
        6/17/97                                     33        DRAFT-DO NOT CITE OR QUOTE

-------
                                          6. REFERENCES

  1     Agency for Toxic Substances and Disease Registry. (1996) Toxicological profile for benzene.
  2     Update. Public Health Service, U.S. Department of Health and Human Services, Atlanta, GA.
  3     Draft for public comment.

  4     Aksoy, M. (1976) Types of leukemia in chronic benzene poisoning. A study in thirty-four
  5     patients. ActaHaemat 55:65-72.

  6     Aksoy, M. (1977) Testimony of Mazaffer Aksoy, M.D., to Occupational Safety and Health
  7     Administration, U.S. Department of Labor, July 13, 1977, 13 pp.

  8     Aksoy, M. (1989) Hematotoxicity and carcinogen!city of benzene. Environ Health Perspect 82:
  9     193-197.

 10     Aksoy, M; Erden, S; Dincol, G. (1974) Leukemia in show-workers exposed chronically to
  11     benzene.  Blood 44(6):837-841.

 12     Ames, BN; Shigenaga, MK.  (1992) Oxidants are a major contributor to aging.  Ann NY Acad Sci
 13     663:85-96.

 14     Anderson, D; Yu, TW; Schmezer, P. (1995) An investigation of the DNA-damaging ability of
 15     benzene and its metabolites in human lymphocytes, using the comet assay. Environ Mol Mutagen
 16     26:305-314.

 17     Ashby, JA; de Serres, FJ; Draper, M; et al. (1985) Overview and conclusions of the IPCS
 18     collaborative study on in vitro assay systems.  In: Ashby, J; de Serres, FJ; Draper, M; Ishidate,
 19     M, Jr; Margolin, BH; Matter, BE; Shelby, MD, eds. Evaluation of short-term tests for
20     carcinogenesis: report of the  International Programme on Chemical Safety's collaborative study
 21     on in vitro assays. Vol. 5.  Amsterdam: Elsevier Science Publishers, pp. 117-174.

22     Bechtold, WE; Rothman, N;  Yin, SN; et al. (1996) Urinary excretion of phenol, catechol,
23     hydroquinone, and muconic acid by workers occupationally exposed to benzene. Occup Environ
24     Med (submitted).

25     Bennett, GF. (1987) Air quality aspects of hazardous waste landfills. Haz Waste Haz Mat 4:119-
26     138.

27     Bodell, WJ; Levay, G; Pongracz, K. (1993) Investigation of benzene DNA adducts and their
28     detection in human bone marrow. Environ Health Perspect 99:241-244.

29     Bois, F; Jackson, E; Pekari, K. (1996) Population toxicokinetics of benzene. Environ Health
30     Perspect 104 (Suppl  6): 1405-1411.

 31     Bond, GG;  McLaren, EA; Baldwin, CL; et al.  (1986a) An update of mortality among chemical
32     workers exposed to benzene. Br J Ind Med 43:685-691.

        6/17/97                                   34       DRAFT-DO NOT  CITE OR QUOTE

-------
  1     Bond, AE; Thompson, VL; Ortman, GC; et al. (1986b) Self service station vehicle refueling
  2     exposure study. In: Proceedings of the 1986 EPA/APCA symposium on measurements of toxic
  3     air pollutants. Pittsburgh, PA: Air Pollution Control Association, 458-466.

  4     Brett, S; Rodricks, J; Chinchilli, V.  (1989) Review and update of leukemia risk potentially
  5     associated with occupational exposure to benzene.  Environ Health Perspect 82:267-281.

  6     Brodfuehrer, JI; Chapman, DE; Wilke, TJ; et al. (1990) Comparative studies of the in vitro
  7     metabolism and covalent binding of 14C-benzene by liver slices and microsomal fraction of
  8     mouse, rat, and human. Drug Metab Dispos 18:20-27.

  9     Brunnemann, KD; Kagan, MR; Cox, JE; et al. (1989) Determination of benzene, toluene and 1,3-
 10     butadiene in cigarette smoke by GC-MSD. Exp Pathol 37:108-113.

  11     Caligiuri, MA; Strout, MP; Gilliland, DG. (1997) Molecular biology of acute myeloid leukemia.
 12     Semin Oncol 24:32-44.

 13     Chen, HW; Eastmond, DA. (1995) Topoisomerase inhibition by phenolic metabolites: a potential
 14     mechanism for benzene's clastogenic effects. Carcinogenesis 16:2301-2307.

 15     Chen, H; Rupa, DS; Tomar, R; et al. (1994) Chromosomal loss and breakage in mouse bone
 16     marrow and spleen cells exposed to benzene in vivo. Cancer Res 54:3533-3539.

 17     Clean Air Act Amendments of 1990.  Subchapter II (42 U.S.C. 7521-7590, as amended).

 18     Cohen, MA; Ryan, PB; Yanagiasawa, Y; et al. (1989) Indoor/outdoor measurements of volatile
 19     organic compounds in Kanawha Valley of West Virginia.  J Air Pollut Control Assoc 39:1086-
20     1093.

 21     Cox, LA, Jr. (1996) Reassessing benzene risks using internal doses and Monte-Carlo uncertainty
22     analysis. Environ Health Perspect 104(6): 1413-1429.

23     Cronkite, EP; Drew, RT; Inoue, T; et al. (1989) Hematotoxicity and carcinogenicity of inhaled
24     benzene. Environ Health Perspect 82:97-108.

25     Crump, KS. (1992) Exposure-response analyses of pliofilm cohort. Work supported by Western
26     States Petroleum Association. Draft.

27     Crump, KS. (1994) Risk of benzene-induced leukemia: a sensitivity analysis of the pliofilm cohort
28     with additional follow-up and new exposure estimates. J Toxicol Environ Health 42:219-242.

29     Crump, KS; Allen, BC. (1984) Quantitative estimates of risk of leukemia from occupational
30     exposure to benzene.  Prepared for the Occupational Safety and Health Administration by Science
 31     Research Systems, Inc., Ruston, Louisiana.
        6/17/97                                   35        DRAFT-DO NOT CITE OR QUOTE

-------
  1     Dean, BJ. (1978) Genetic toxicology of benzene, toluene, xylenes, and phenols.  Mutat Res 47:
  2     75-97.

  3     Dean, BJ. (1985) Recent findings on the genetic toxicology of benzene, toluene, xylenes, and
  4     phenols. Mutat Res 154:153-181.

  5     DeCoufle, P; Blatter, WA; Blair, A. (1983) Mortality among chemical workers exposed to
  6     benzene and other agents.  Environ Res 30:16-25.

  7     Eastmond, DA. (1993) Induction of micronuclei and aneuploidy by the quinone-forming agents
  8     benzene and o-phenylphenol. Toxicol Lett 67:105-118.

  9     Eastmond, DA; Rupa, DS; Hasegawa, LS. (1994) Detection of hyperdiploidy and chromosome
 10     breakage in interphase human lymphocytes following exposure to the benzene metabolite
  11     hydroquinone using multicolor fluorescence in situ hybridization with DNA probes. Mutat Res
 12     322:9-20.

 13     Flachsbart, PG. (1992) Human exposure to motor vehicle air pollutants.  In:  Motor vehicle air
 14     pollution: public health impacts and control measures. Mage, D; Zali, O, eds. ECOTOX, Service
 15     of Dept. of Ecotoxicology, Dept. of Public Health, World Health Organization, Geneva,
 16     Switzerland, pp. 85-113.

 17     Forni, A. (1971) Chromosome studies in workers exposed to benzene or toluene or both. Arch
 18     Environ Health 22:373-378.

 19     Goldstein, BD; Snyder, CA; Laskin, S; et al. (1982) Myelogenous leukemia in rodents inhaling
20     benzene. Toxicol Lett 13:169-173.

 21     Hayes, RB; Yin, SN; Dosemeci, M; et al. (1996) Mortality among benzene-exposed workers in
22     China. Environ Health Perspect 104(6): 1349-1352.

23     Heim, S. (1996) Clonal chromosome abnormalities in neoplastic cells: evidence of genetic
24     instability? Cancer Surv 28:247-260.

25     Hogstedt, B; Holmen, A; Karlsson, A; et al. (1991) Gasoline pump mechanics had increased
26     frequencies and sizes of micronuclei in lymphocytes stimulated by pokeweed mitogen. Mutat Res
27     263:51-55.

28     Huff, JE; Haseman, JK; DeMarini, DM; et al.  (1989) Multiple-site carcinogenicity of benzene in
29     Fischer 344 rats and B6C3Fj mice. Environ Health Perspect 82:125-163.

30     Infante, PF; Rinsky, RA; Wagoner, JK; et al. (1977) Leukemia in benzene workers. Lancet, 2:76-
 31     78.
        6/17/97                                   36        DRAFT-DO NOT CITE OR QUOTE

-------
  1     International Agency for Research on Cancer. (1982) IARC monographs on the evaluation of
  2     carcinogenic risks of chemicals to humans: some industrial chemicals and dyestuffs. Vol. 29,
  3     Lyon, France: IARC, pp. 93-148.

  4     Irons, RD; Neptune, DA. (1980) Effects of the principal hydroxy-metabolites of benzene on
  5     microtubule polymerization. Arch Toxicol 45:297-305.

  6     Irons, RD; Stillman, WS. (1996) The process of leukemogenesis. Environ Health Perspect
  7     104:1239-1246.

  8     Karp, JE; Smith, MA. (1997)  The molecular pathogenesis of treatment-induced (secondary)
  9     leukemias: foundations for treatment and prevention. Semin Oncol 24:103-113.

 10     Kolachana, P; Subrahmanyam, VV; Meyer, KB; et al. (1993) Benzene and its phenolic
  11     metabolites produce oxidative DNA damage in HL60 cells in vitro and in the bone marrow in
 12     vivo. Cancer Res 53:1023-1026.

 13     Lagorio S; Tagesson, C; Forastiere, F; et al. (1994) Exposure to benzene and urinary
 14     concentrations of 8-hydroxydeoxyguanosine, a biological marker of oxidative damage to DNA.
 15     Occup Environ Med 51:739-743.

 16     Levay, G; Ross, D; Bodell,  WJ. (1993) Peroxidase activation of hydroquinone results in the
 17     formation of DNA adducts in HL-60 cells, mouse bone marrow macrophages and human bone
 18     marrow.  Carcinogenesis 14:2329-2334.

 19     Levay, G; Pathak, DN; Bodell, WJ. (1996) Detection of DNA adducts in the white blood cells of
20     B6C3Fj mice treated with benzene. Carcinogenesis 17:151-153.

 21     Linet, MS. (1985) The leukemias: epidemiologic aspects. New York: Oxford University Press,
22     Inc., 293 p.

23     Low, LK; Meeks, J; Norris, KJ; et al. (1989) Pharmacokinetics and metabolism of benzene in
24     Zymbal gland and other key target tissues after oral administration in rats. Environ Health
25     Perspect 82:215-222.

26     Low, LK; Lambert, C; Meeks, J; et al. (1995) Tissue-specific metabolism of benzene in Zymbal
27     gland and other solid tumor target tissues in rats. J Am Coll Toxicol 14:40-60.

28     Maltoni, C; Conti, B; Cotti, G. (1983) Benzene:  a multipotential carcinogen. Results of the long-
29     term bioassays performed at the Bologna Institute of Oncology. Am J Ind Med 4:589-630.

30     Maltoni, C; Conti, B; Perino, G; et al. (1988) Further evidence  of benzene carcinogenicity.
 31     Results on Wistar rats and Swiss mice treated by ingestion.  Ann NY Acad  Sci 534:412-426.

32     Marrazzini, A; Chelotti, L; Barrai, I; et al. (1994) In vivo genotoxic interactions among three
33     phenolic benzene metabolites.  Mutat Res 341:29-46.

        6/17/97                                   37        DRAFT-DO NOT CITE OR QUOTE

-------
  1     McDonald, TA; Waidyanatha, S; Rappaport, SM. (1993) Production of benzoquinone adducts
  2     with hemoglobin and bone-marrow proteins following administration of [13C6]benzene to rats.
  3     Carcinogenesis 14:1921-1925.

  4     McDonald, TA; Yeowell, OCK; Rappaport, SM. (1994) Comparison of protein adducts of
  5     benzene oxide and benzoquinone in the blood and bone marrow of rats and mice exposed to
  6     [14C/13C6]benzene. Cancer Res 54:4907-4914.

  7     Medinsky, MA; Schlosser, PM; Bond, JA. (1994) Critical issues in benzene toxicity and
  8     metabolism: the effect of interactions with other organic chemicals on risk assessment. Env
  9     Health Perspect 102 (Suppl 9): 119-124.

 10     Medinsky, MA; Kenyon, EM; Seaton, MJ; et al. (1996) Mechanistic considerations in benzene
  11     physiological model development. Environ Health Perspect 104(6): 1399-1404.

 12     Moolgavkar, S; Knudson, A. (1981) Mutation and cancer: a model for human carcinogenesis.  J
 13     Natl Cancer Inst 66:1037-1052.

 14     Mullin, AH; Rando, R; Esmundo, F; et al. (1995) Inhalation of benzene leads to an increase in the
 15     mutant frequencies of a lad transgene in lung and spleen tissues of mice.  Mutat Res 327:121-
 16     129.

 17     National Toxicology Program, Public Health Service, U.S.  Department of Health and Human
 18     Services. (1986) NTP technical report on the toxicology and carcinogenesis studies of benzene
 19     (CAS No. 71-43-2) in F344/N rats and B6C3Fj mice (gavage studies).  NTP TR 289.  Available
20     from: National Institutes of Health, Research Triangle Park, NC.

 21     Oberly, TJ; Bewsey, BJ; Probst, GS. (1984) An evaluation of the L5178Y TK+/- mouse
22     lymphoma forward mutation assay using 42 chemicals. Mutat Res 125:291-306.

23     Oberly, TJ; Rexroat, MA; Richardson, KK; et al. (1990) An evaluation of the CHO/HGPRT
24     mutation assay involving suspension cultures and soft agar  cloning: results for 33 chemicals.
25     Environ Mol Mutagen  16:260-271.

26     Occupational Safety and Health  Administration. (1987) Final rule on occupational exposure to
27     benzene. Federal Register 54:34660-34762.

28     Ott, MG; Townsend, JC; Fishbeck, WA; et al. (1977) Mortality among individuals occupationally
29     exposed to benzene. Exhibit 154, OSHA benzene hearings, July 9-August 19, 1977.

30     Ott, MG; Townsend, JC; Fishbeck, WA; et al. (1978) Mortality among individuals occupationally
 31     exposed to benzene. Arch Environ Health 33(1):3-10.

32     Parmley, R. (1988) Mammals. In: Vertebrate blood cells. Rowley, AF; Ratcliffe, NA, eds.
33     Cambridge: Cambridge University Press,  pp. 337-424.
        6/17/97                                   38        DRAFT-DO NOT CITE OR QUOTE

-------
  1     Pathak, DN; Levay, G; Bodell, WJ. (1995) DNA adduct formation in the bone marrow of
  2     B6C3FJ mice treated with benzene. Carcinogenesis 16:1803-1808.

  3     Paustenbach, DJ; Price, PS; Ollison, W; et al. (1992) Reevaluation of benzene exposure for the
  4     Pliofilm (rubber-worker) cohort (1936-1976).  J Toxicol Environ Health 36:177-231.

  5     Paustenbach, D; Bass, R; Price, P. (1993) Benzene toxicity and risk assessment, 1972-1992:
  6     implications for future regulation. Environ Health Perspect 101 (Suppl 6): 177-200.

  7     Paxton, MB. (1996) Leukemia risk associated with benzene exposure in the pliofilm cohort.
  8     Environ Health Perspect 104(6): 1431 -143 6.

  9     Paxton, MB; Chinchilli, V; Brett, SM; et al. (1992, Jan. 27) Reanalysis and update of the
 10     leukemogenic risk associated with occupational benzene exposure in pliofilm cohort. Environ
  11     Corporation, Arlington, Virginia.

 12     Paxton, MB; Chinchilli, V; Brett, SM; et al. (1994) Leukemia risk associated with benzene
 13     exposure in the pliofilm cohort. II. Risk estimate. Risk Anal 14(2):155-161.

 14     Pedersen-Bjergaard, J; Pedersen, M; Roulston, D; et al. (1995) Different genetic pathways in
 15     leukemogenesis for patients presenting with therapy-related myelodysplasia and therapy-related
 16     acute myeloid leukemia. Blood 86:3542-3552.

 17     Pellizzari, ED. (1982) Analysis for organic vapor emissions near industrial  and chemical waste
 18     disposal sites.  Environ Sci Technol 16:781-785.

 19     Plappert, U; Barthel, E; Raddatz, K; et al. (1994) Early effects of benzene exposure in mice.
20     Hematological versus genotoxic effects.  Arch Toxicol 68:284-290.

 21     Reddy, MV; Bleicher, WT; Blackburn, GR; et al. (1990) DNA adduction by phenol,
22     hydroquinone, or benzoquinone in vitro but not in vivo: nuclease Pi-enhanced 32P-postlabeling of
23     adducts as labeled nucleoside bisphosphates, dinucleotides and nucleoside monophosphates.
24     Carcinogenesis 11:1349-1357.

25     Reddy, MV; Schultz,  SC; Blackburn, GR; et al. (1994) Lack of DNA adduct formation in mice
26     treated with Benzene.  Mutat Res 325:149-155.

27     Rinsky, R; Young,  RJ; Smith, AB. (1981) Leukemia in benzene workers. Am J Ind Med 2:217-
28     245.

29     Rinsky, RA; Smith, AB; Hornung, R; et al. (1987) Benzene and leukemia: an epidemiologic risk
30     assessment.  NEnglJMed 316:1044-1050.

 31     Robertson, ML; Eastmond, DA; Smith, MT. (1991) Two benzene metabolites, catechol and
32     hydroquinone, produce a synergistic induction of micronuclei and toxicity in cultured human
33     lymphocytes. Mutat Res 249:201-209.

        6/17/97                                    39       DRAFT-DO NOT CITE OR QUOTE

-------
  1     Rothman, N; Haas, R; Hayes, RB; et al. (1995) Benzene induces gene duplicating but not gene
  2     inactivating mutations at the glycophorin A locus in exposed humans. Proc Natl Acad Sci USA
  3     92(4):4069-4073.

  4     Rothman, N; Li, GL; Dosemeci, M; et al. (1996) Hematotoxicity among Chinese workers heavily
  5     exposed to benzene. Am J Ind Med 29:236-246.

  6     Sabourin, PJ; Bechtold, WE; Griffith, WC; et al. (1989) Effect of exposure concentration,
  7     exposure rate, and route of administration on metabolism of benzene by F344 rats and B6C3FJ
  8     mice. Toxicol Appl Pharmacol 99:421-444.

  9     Sarto, F; Cominato, I; Pinton, AM; et al. (1984) A cytogenetic study on workers exposed to low
 10     concentrations of benzene. Carcinogenesis 5:827-832.

  11     Sasiadek, M. (1992) Nonrandom distribution of breakpoints in the karyotypes of workers
 12     occupationally exposed to benzene. Environ Health Perspect 97:255-257.

 13     Schlosser, PM; Bond, JA; Medinsky, MA. (1993) Benzene and phenol metabolism by mouse and
 14     rat liver microsomes.  Carcinogenesis 14(12):2477-2486.

 15     Schlosser, PM; Kenyon, EM;  Seaton, MJ; et al. (1995) Determinants of benzene metabolism and
 16     disposition. CUT Activities 15(6): 1-9.

 17     Schnatter, AR; Nicolich, MJ; Bird, MG. (1996) Determination of leukemogenic benzene exposure
 18     concentrations: refined analyses of the pliofilm cohort.  Risk Anal  16(6):833-840.

 19     Shah, JJ; Singh, HB. (1988) Distribution of volatile organic chemicals in outdoor and indoor air.
20     Environ Sci Technol 22:1381-1388.

 21     Smith, MT; Yager, JW; Steinmetz, KM; et al. (1989) Peroxidase-dependent metabolism of
22     benzene's phenolic metabolites and its potential role in benzene toxicity and carcinogenicity.
23     Environ Health Perspect 82:23-29.

24     Snyder, R; Kalf, GF. (1994) A perspective on benzene leukemogenesis. Crit Rev Toxicol 24:177-
25     209.

26     Snyder, CA; Goldstein, BD; Sellakumar, AR; et al. (1980) The inhalation toxicology of benzene:
27     incidence of hematopoietic neoplasms and hematotoxicity in AKR/J and C57BL/6J mice.  Toxicol
28     Appl Pharmacol 54:323-331.

29     Snyder, CA; Goldstein, BD; Sellakumar, A; et al. (1982) Toxicity of chronic benzene inhalation:
30     CD-I mice exposed to  300 ppm. Bull Environ Contam Toxicol 29:385-391.

 31     Snyder, CA; Goldstein, BD; Sellakumar, AR; et al. (1984) Evidence for hematotoxicity and
32     tumorigenesis in rats exposed to 100 ppm benzene.  Am J Ind Med 5:429-434.
        6/17/97                                   40        DRAFT-DO NOT CITE OR QUOTE

-------
  1     Subrahmanyam, VV; Ross, D; Eastmond, DA; et al. (1991) Potential role of free radicals in
  2     benzene-induced myelotoxicity and leukemia. Free Rad Biol Med 11:495-515.

  3     Thorslund, T. (1988) American Petroleum Institute (API) submission in response to EPA's
  4     proposed rule on the national standard for benzene emissions.

  5     Thorslund, T. (1993) Development and use of a dose response model for estimating the risk of
  6     acute myelogenous leukemia associated with inhalation of benzene. Prepared for Environmental
  7     Health Directorate, Health and Welfare Canada, Environmental Health Center.  2/5/93, George
  8     Mason University, Fairfax, Virginia.

  9     Tompa,  A; Major, J; Jakab, MG. (1994) Monitoring of benzene-exposed workers for genotoxic
 10     effects of benzene: improved-working-condition-related decrease in the frequencies of
  11     chromosomal aberrations in peripheral blood lymphocytes. Mutat Res 304:159-165.

 12     Turtletaub, KW; Creek, M; Mani, C; et al. (1996) Tissue distribution and macromolecular binding
 13     of [14C]-benzene in mice. Abstract presented at the 12th Health Effects Institute Annual
 14     Conference.  Asheville, NC, April 28-30.

 15     U.S. Environmental Protection Agency. (1979) Final report on population risk to ambient
 16     benzene exposures. Prepared by the Carcinogen Assessment Group, Research Triangle Park, NC.
 17     EPA/450/5-80-004.

 18     U.S. Environmental Protection Agency. (1985, Feb.  15) Interim quantitative cancer unit risk
 19     estimates due to inhalation of benzene. Prepared by the Carcinogen Assessment Group, Office of
20     Research and Development, Washington, DC. EPA/600/X-85-022.

 21     U.S. Environmental Protection Agency. (1986, Sept. 24) Guidelines for carcinogen risk
22     assessment.  Federal Register 51(185):33992-34003.

23     U.S. Environmental Protection Agency. (1987) June-September, 6-9 AM, ambient air benzene
24     concentrations in 39 U.S. cities, 1984-1986. Research Triangle Park, NC: U.S.  Environmental
25     Protection Agency, Atmospheric Sciences Research Lab. EPA/600/D-87/160.

26     U.S. Environmental Protection Agency. (1996, April 23) Proposed  guidelines for carcinogen risk
27     assessment.  Federal Register 61(79):17960-18011.

28     Van den Berghe, H; Louwagie, A; Broeckaert-Van Orshoven, A; et al.  (1979) Chromosome
29     analysis in two unusual malignant blood disorders presumably induced by benzene. Blood 53:558-
30     566.

 31     Vigliani, EC; Forni, A. (1976) Benzene and leukemia. Environ Res 11:122-127.

32     Vigliani, EC; Saita, G. (1964) Benzene and leukemia.  N Engl J Med 271:872-876.

33     Wallace, LA. (1989) Major sources of benzene exposure. Environ Health Perspect 82:165-169.

        6/17/97                                    41        DRAFT-DO NOT CITE OR QUOTE

-------
  1     Wallace, LA; Pellizzari, ED; Hartwell, TD; et al. (1987) Exposures to benzene and other volatile
  2     compounds from active and passive smoking. Arch Environ Health 42:272-279.

  3     Ward, E; Hornung, R; Morris, J; et al. (1996) Risk of low red or white cell count related to
  4     estimated benzene exposure in a rubberworker cohort (1940-1975).  Am J Ind Med 29:247-257.

  5     Ward, JB, Jr.; Ammenhauser, MM; Ramanujam, VM; et al. (1992) The mutagenic effects of low
  6     level sub-acute inhalation exposure to benzene in CD-I mice. Mutat Res 268:49-57.

  7     Weaver, VM; Davoli, CT; Heller, PJ; et al. (1996) Benzene exposure, assessed by urinary trans,
  8     trans-muconic acid, in urban children with elevated blood lead levels. Environ Health Perspect
  9     104(3):318-323.

 10     Wester, RC; Maibach, HI; Gruenke, LD; et al. (1986) Benzene levels in ambient air and breath of
  11     smokers and nonsmokers in urban and pristine environments. J Toxicol Environ Health 18:567-
 12     573.

 13     Wiltse, J; Dellarco, VL. (1996)  U.S. Environmental Protection Agency's guidelines for
 14     carcinogen risk assessment:  past and future.  Mutation Res 365:3-15.

 15     Witz, G; Maniara,  W; Mylavarapu, V; et al. (1990a) Comparative metabolism of benzene and
 16     trans,tmns-muconaldehyde to trans,trans-muconic acid  in DBA/2N and C57BL/6 mice. Biochem
 17     Pharmacol 40:1275-1280.

 18     Witz, G; Gad, SC; Tice, RR; et al. (1990b) Genetic toxicity of the benzene metabolite trans,
 19     fram'-muconaldehyde in mammalian and bacterial cells. Mutat Res 240:295-306.

20     Wong, O. (1987) An industry-wide study of chemical workers occupationally exposed to
 21     benzene. BrJ Ind Med 44:382-395.

22     Wong, O. (1995) Risk of acute myeloid leukemia and multiple myeloma in workers exposed to
23     benzene. Occup Environ Med 52:380-384.

24     Wong, O; Morgan, RW; Whorton, MD. (1983, Dec.  8) An industry-wide mortality study of
25     chemical workers occupationally exposed to benzene. Technical report submitted to Chemical
26     Manufacturers Association, by Environmental Health Associates, Berkeley, CA.

27     Wright, EC. (1995) The pathogenesis of leukemia. In:  Radiation toxicology: bone marrow and
28     leukemia. Hendry, JH; Lord, BI, eds. London: Taylor  and Francis,  pp. 245-274.

29     Yager, JW;  Eastmond,  DA; Robertson, ML; et al. (1990) Characterization of micronuclei induced
30     in human lymphocytes  by benzene metabolites. Cancer Res 50:393-399.

 31     Yin, SN; Li, GL; Tain, FD; et al. (1989) A retrospective cohort study of leukemia and other
32     cancers in benzene workers.  Environ Health Perspect 82:207-213.
        6/17/97                                   42        DRAFT-DO NOT CITE OR QUOTE

-------
  1     Young, BD; Saha, V. (1996) Chromosome abnormalities in leukaemia: the 1 Iq23 paradigm.
 2     Cancer surv 28:225-245.

 3     Zhang, L; Robertson, ML; Kolachana, P; et al. (1993) Benzene metabolite, 1,2,4-benzenetriol,
 4     induces micronuclei and oxidative DNA damage in human lymphocytes and HL60 cells. Environ
 5     MolMutagen 21:339-348.

 6     Zhang, L; Venkatesh, P; Creek, MLR; et al. (1994) Detection of 1,2,4-benzenetriol induced
 7     aneuploidy and microtubule disruption by fluorescence in situ hybridization and
 8     immunocytochemistry. Mutat Res 320:315-327.

 9     Zhang, L; Rothman, N; Wang, Y; et al. (1996a) Interphase cytogenetics of workers exposed to
10     benzene. Environ Health Perspect 104(6): 1325-1329.

 11     Zhang, L; Rothman, N; Wang, Y; et al. (1996b) Aneuploidy of chromosomes 7, 8, and 9 detected
12     by fluorescence in situ hybridization in workers exposed to benzene. Cancer Res (submitted).
       6/17/97                                   43        DRAFT-DO NOT CITE OR QUOTE

-------