Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
15 Methylene Chloride Raghupathy Ramanathan, Ph.D. Toxicology Group Habitability and Environmental Factors Division Johnson Space Center National Aeronautics and Space Administration Houston, Texas Spacecraft maximum allowable concentration (SMAC) values for methyl- ene chloride (also called 1,2-dichloromethane, DCM) were previously published in volume 2 of the series, Spacecraft Maximum Allowable Concentrations (SMAC) for Selected Airborne Contaminants, for exposure durations of 1 h, 24 h, 7 d, 30 d, and 180 d (Wong 1996). With NASAâs current focus on exploration missions beyond low Earth orbit to the moon and Mars, there is a need to derive acceptable concentrations (ACs) for long-duration missions, such as for 1,000 d. GENERAL APPROACH GUIDELINES The effort consists of identifying new toxicology literature on DCM toxic- ity since the previous SMAC document was prepared (Wong 1996) and suitable studies that may have been missed during the preparation of that earlier docu- ment, in order to derive ACs for 1,000 d and other durations. Another objective is to determine if the previous SMAC (approved by the previous SMAC com- mittee) needs to be updated based on new data that have become available or novel approaches have been developed, such as the benchmark dose (BMD) approach, which can be used on the data from the principal studies that were used before. GENERAL PROPERTIES AND OCCURRENCE Physical and chemical properties and occurrence have already been dis- cussed (Wong 1996). The formula weight of DCM is 84.9 and conversion fac- tors for DCM are 1 part per million (ppm) = 3.47 milligrams per cubic meter (mg/m3) and 1 mg/m3 = 0.29 ppm. DCM has been detected in the Shuttle space- craft atmosphere in 28 of 33 missions at concentrations ranging from 0.029 to 289
290 SMACs for Selected Airborne Contaminants 0.29 ppm (0.1-1 mg/m3) and also has been detected in the International Space Station atmosphere at about 0.5 mg/m3. The odor threshold for DCM in air is 250 ppm. BACKGROUND AND SUMMARY OF ORIGINAL APPROACH Studies in human volunteers show that DCM is well absorbed (up to 70%) by resting subjects during inhalation exposures, and exercise changes the ab- sorption (DiVincenzo et al. 1972, Astrand et al. 1975, DiVincenzo and Kaplan 1981). Animal studies indicate that inhaled DCM is distributed in the liver, kid- neys, lungs, brain, muscle, adipose tissues, and adrenals about 1 h after inhala- tion exposure, with the highest concentrations found in white adipose tissue and the next highest in liver (McKenna et al. 1982). Systemically absorbed DCM is metabolized by two pathways. One path- way is via the microsomal mixed function oxidase (MFO) in the cytochrome P- 450 system (cytochrome P-450 2E1 or CYP 2E1) (Gargas et al. 1986, Guengerich et al. 1991). The oxidative dehalogenation yields hydrogen chloride, carbon monoxide (CO), and carbon dioxide, with formyl chloride as an interme- diate. At low exposures, this pathway predominates, and it is saturable at about 300 to 500 ppm (Gargas et al. 1986). The CO from this pathway binds reversibly to hemoglobin, forming carboxyhemoglobin (COHb). COHb reduces the oxy- gen-carrying capacity of the blood and also impairs the release of O2 from oxy- hemoglobin, thus leading to tissue oxygen deficiency. In six sedentary human subjects exposed to DCM at 50, 100, 150, or 200 ppm for 7.5 h on 5 consecutive days, concentrations of COHb in blood were 1.9%, 3.4%, 5.3%, and 6.8%, re- spectively (DiVincenzo and Kaplan 1981). Numerous investigations have shown that CO is toxic to the cardiovascular system (changes heart rate and minute volume) and also to the central nervous system (CNS), where it has adverse ef- fects such as impairing vigilance and performance in addition to causing head- ache, decreased vision, and other symptoms. The second pathway is the glutathione (GSH)-dependent cytosolic path- way via glutathione S-transferase theta 1 (GSTT1) (Kubic and Anders 1975, Ahmed and Anders 1976, Andersen et al. 1987, Reitz et al. 1989). This pathway is a low-affinity first-order pathway that metabolizes DCM to hydrogen chlo- ride, formaldehyde, and carbon dioxide. In the GSTT1 pathway, the haloalkane is metabolized to produce the reactive S-chloromethylglutathione intermediate, which has the capacity to interact with cellular DNA. The chloromethyl glu- tathione is short-lived; it undergoes rapid hydrolysis to yield formaldehyde. This GSH pathway is not saturable and is linear up to 10,000 ppm (Gargas et al. 1986). Carcinogenicity of DCM in long-term inhalation exposure of rodents has been attributed to metabolism of the compound via the GST-dependent pathway. Andersen et al. (1987) reported that large quantities of GSH-DCM conjugates in vivo may increase the frequency of lung and liver tumors that develop in some species of animals (such as B6C3F1 mice). DCM metabolism via the GSH
Methylene Chloride 291 pathway in the target tissue has been the subject of several studies as the basis for species sensitivity to DCM-induced tumor incidence. It has also been sug- gested that formaldehyde produced from DCM metabolism via the same GST pathway may be responsible for the observed tumors due to the formation of DNA-protein crosslinks mediated by formaldehyde (Casanova et al. 1997). De- scribing and discussing numerous studies that attempted to explore the mecha- nism of DCM-induced tumors and its specificity in organs and species are far beyond the scope of this document. Several reports have appeared on physiologically based pharmacokinetic (PBPK) modeling of these various metabolic pathways using in vivo and in vitro metabolic rates obtained from animal and human tissue samples and validating the kinetic data from human subjects exposed to DCM (Andersen et al. 1987, Reitz et al. 1989, Andersen et al. 1991, Clewell 1995, and many others). These earlier PBPK models for DCM have gone through several refinements and de- velopments, including integrating the statistical models of the parameters for uncertainty and population distributions with the toxicokinetic models. In addi- tion, to obtain target concentrations of DCM, PBPK models have been described for DCM uptake and distribution during rest and exercise (Dankovic and Bailer 1994, Jonsson et al. 2001). These models use blood flow and perfusion changes that occur during rest and exercise, accounting for changes in ventilation rate and cardiac output. Using statistical methods and PBPK models, investigators have attempted to estimate interindividual and population variability in the rate of metabolizing DCM. To obtain a more accurate assessment of human health risk from synthetic halomethanes in the last few years, investigators have attempted to correlate and explain the interindividual variations and species sensitivities to DCM-induced carcinogenicity by the existence of polymorphisms in theta-class isoforms of GST (GSTT1). In humans, GSH-dependent conjugation of halomethane is po- lymorphic, with 60% of the population classed as conjugators and 40% classed as nonconjugators, implying that conjugators will be more sensitive to DCM than nonconjugators. Pemble et al. (1994), Hallier et al. (1994), Nelson et al. (1995), Katoh et al.(1996), Thier et al. (1998), and El-Masri et al. (1999) have discussed the importance of GSST1 polymorphism in many different ethnic groups in the risk assessment of haloalkanes such as DCM. GSTT1 can catalyze the GSH conjugation of DCM via a metabolic pathway that has been shown to be mutagenic in Salmonella typhimurium mutagenicity tester strains and was believed to be responsible for the carcinogenicity of DCM reported in the NTP (1986) DCM inhalation bioassay study in mice. Thus, concerns have been raised that this polymorphism is an important factor that will affect the risk estimates for DCM (El-Masri et al. 1999, Jonsson and Johanson 2001). SUMMARY OF ORIGINAL APPROACH AND ACs The SMACs for exposure durations of 1 h to 7 d were based on CNS de-
292 SMACs for Selected Airborne Contaminants pression. The 30- and 180-d SMACs were based on hepatotoxicity. Wong et al. (1996) also estimated ACs based on the end point of carcinogenicity. Wong (1996) derived a 1-h AC based on the adverse CNS effects of DCM vapors reported by many investigators in humans, both in an occupational set- ting and in controlled studies. He based the critical effect on reports by Peterson (1978), Putz et al. (1979), Winneke (1974, 1981), Winneke and Fodor (1976), and Stewart et al. (1972) in which impaired hand-eye coordination, increased tracking error, and impaired vigilance were reported in human volunteers ex- posed to different concentrations of DCM for different lengths of time. Many of these studies had measured COHb, and the effects correlated with the concentra- tions of serum COHb resulting from DCM oxidative metabolism. Wong adopted a strategy of using the previously established NASA SMACs for CO (Wong 1994) as a basis for deriving some ACs for DCM. On that basis, the 1- and 24-h ACs for blood COHb level were set at 3% (Wong 1994). Winneke (1974) stated that the observed CNS effects are directly related to DCM and not COHb formed from its metabolism, because there were no CNS effects at 100 ppm, in spite of the formation of COHb in blood. If one assumes a threshold concentra- tion of COHb that will not produce any such effects, Wongâs approach (Wong 1996) is reasonable. He collated the data from the human volunteer studies in which concentrations of COHb were measured after various concentrations and durations of DCM exposures and derived a linear regression of the total dose of DCM versus the percent increase of COHb concentrations. From the slope of the fitted regression line, a concentration of DCM was calculated that will produce an increase of 2.4% COHb over the background nominal concentration of 0.6% COHb, which is produced by endogenous CO production in the human body. This corresponded to a DCM concentration of 100 ppm as the 1- and 24-h no- observed-adverse-effect levels NOAELs. Wong (1996) used another method to derive a 24-h AC; the COHb con- centration was used as a surrogate variable for CO formation from DCM. He used the data for CO and COHb formation computed from the PBPK model developed by Andersen et al. (1991), which modeled the parent compound and its metabolites, CO and COHb, in rats and humans. The human model was vali- dated with human volunteers exposed to DCM at 100 or 350 ppm for 6 h. Wong (1996) calculated that an exposure of 35 ppm for 24 h would produce a final COHb of 3%. It was indirectly implied that neurotoxicity (CNS depression, vis- ual performance, and perception of time) was the adverse end point used, as that is based on the levels of COHb. For deriving the 7-d AC, Wong considered neurotoxicity and hepatotoxic- ity to be critical effects. For ACs for 7 d and longer, Wong adopted the 7-, 30-, and 180-d AC he had derived for CO, with a target COHb concentration of 1.6%. For CNS effects, he followed the approach mentioned above using the Andersen et al. (1991) PBPK model and computed an AC of 14 ppm for DCM, which will lead to increased COHb concentrations from the background 0.6% to 1.6%. Wong (1996) rounded the 14 ppm to 15 ppm in the AC summary table (see Table 15-1 for a summary 1996 SMACs).
Methylene Chloride 293 TABLE 15-1 Summary of Previously Published SMACS for DCM (Wong 1996) Duration ppm mg/m3 Critical effect Principal studies 1h 100 350 CNS depression Various human data 24 h 35 120 CNS depression Andersen et al. 1991(PBPK) 7d 15 50 CNS depression Andersen et al. 1991 30 d 5 20 Hepatotoxicity Burek et al. 1984 180 d 3 10 Hepatotoxicity Burek et al. 1984 The 7-d AC was also derived using hepatotoxicity as the end point. In a 13-wk study (NTP 1986) in which groups of 10 rats per gender (F344/N) and 10 mice per gender (B6C3F1) were exposed to air containing 0, 525, 1,050, 2,100, 4,200, or 8,400 ppm cytoplasmic vacuolization and necrosis of the liver as well as hemosiderosis and focal granulomatous inflammation were noted in mice after repetitive exposures to DCM greater than 2,100 ppm for 6 h/d, 5 d/wk for 13 wk. Using the NOAEL of 2,100 ppm, a 7-d AC of 210 ppm was derived after a species factor of 10 was applied. For a 30-d AC derivation, Wong used the 2-y NTP (1986) study (in which rats and mice were exposed to 0, 1,000, 2,000, or 4,000 ppm of DCM for 6 h/d, 5 d/wk for 102 wk) for cytoplasmic vacuolization, hemosiderosis, and focal granulomatous inflammation in liver. The Burek et al. (1984) 2-y study looked for similar effects at a lower dose of 500 ppm. Using the LOAEL of 500 ppm, and after applying factors of 10 for LOAEL to NOAEL and species, each author arrived at a 30-d AC of 5 ppm (see Wong 1996 for details). For a 180-d AC derivation, Wong used the same hepatotoxicity as the end point reported in the Burek et al. (1984) study in which 500 ppm was identified as the LOAEL. The author derived an AC of 3.6 ppm as a 180-d AC after ad- justing for the LOAEL for discontinuous to continuous exposure and applying factors of 10 for LOAEL to NOAEL and for species extrapolation. The author rounded the value of 3.6 ppm to 3.0 in the AC summary table (see Table 15-1). For deriving a 180-d AC for carcinogenicity risk, the 2-y NTP carcino- genesis bioassay data, as summarized by Mennear et al. (1988), were used. The results of the 2-y carcinogenicity bioassay conducted in various species exposed to DCM by inhalation are summarized in Table 15-2. Several epidemiologic studies have been conducted of workers exposed to DCM in the manufacturing of triacetate fibers (Lanes et al. 1990), photographic film, and paint and varnish. The collected data do not demonstrate a strong, sta- tistically significant excess cancer risk associated with occupational exposures to DCM below 500 ppm (Ott et al. 1983; Hearne et al. 1987, 1990 [Kodak workers study]; Lanes et al. 1993). However, positive results from the animal carcinogenicity tests have driven some regulatory agencies to declare that DCM may be carcinogenic to humans.
294 SMACs for Selected Airborne Contaminants TABLE 15-2 Summary of Rodent Carcinogenicity Bioassays for Exposure to DCM by Inhalation Route and Dosage Species/ Comments Reference Dosing (Number of Strain Animals) Inhalation 0, 2,000, B6C3F1 Dose-related increases in both NTP 1986 6 h/d, 5 4,000 ppm; 50 mouse hepatocellular adenomas and d/wk, 2 y mice/gender/ hepatocellular carcinomas; dose increased incidence of alveolar/bronchiolar adenomas in lungs of both genders at both doses; also, increases in the incidence of animals bearing multiple lung tumors. Inhalation 0, 1,000, 2,000, F344 rat Mammary and integumentary NTP 1986 6 h/d, 5 4,000 ppm; 50 fibromas and fibrosarcomas in d/wk, 2 y rats/gender/ both genders; increased dose incidence of leukemia in femalesâthus, clear evidence of carcinogenicity in females; some evidence of carcinogenicity in males. Inhalation 0, 500, 1500, Sprague- Number of female rats with a Burek et al. 6 h/d, 5 3,500 ppm; 95 Dawley rat benign tumor did not increase, 1984 d/wk, 2 y rats/gender/ but total number of these dose tumors increased; in male rats, the number of sarcomas near the salivary gland increased. Inhalation 0, 500, 1500, Syrian No malignant tumors Burek et al. 6 h/d, 5 3,500 ppm; golden observed. 1984 d/wk, 2 y 90 hamsters/ hamster gender/dose Inhalation 0, 50, 200, Sprague- No increase in malignant Nitschke et 6 h/d, 5 500 ppm; 70 Dawley rat tumors even at 500 ppm. al. 1988b d/wk, 2 y rats/gender/ dose In the NTP study (1986), exposure to DCM for 2 y at 0, 2,000, and 4,000 ppm produced 3 of 50, 30 of 48, and 41 of 48 cases of lung tumors and 3 of 50, 16 of 48, and 40 of 48 cases of liver tumors, respectively, in female B6C3F1 mice. Instead of using the airborne DCM concentrations to calculate the 1 in 10,000 tumor risk, Wong used the equivalent concentration of active metabolite produced by the GST pathway (dose metrics) in the lung and liver, as estimated by a PBPK model (Andersen et al. 1987). The exposure concentrations of 2,000 and 4,000 ppm were substituted by the corresponding values of the metabolites produced in the liver and lung in the multistage linearized model as the doses against tumor incidence. According to the Andersen et al. (1987) PBPK model,
Methylene Chloride 295 DCM exposure concentrations of 6 and 12 ppm for lung and liver, respectively, for humans will result in a lifetime excess tumor risk of 1 in 10,000. The lower concentration of 6 ppm (for lung) was then used in the final risk assessment. After adjusting the 6 ppm for the discontinuous to continuous exposure, the au- thor arrived at a risk value of 1.1 ppm for a lifetime excess lung tumor incidence of 1 in 10,000. For calculating the 180-d cancer risk, Wong used the NRC (1992) recommended formula for calculating a cancer risk from a lifetime expo- sure to less than lifetime durations that resulted in a factor of 146.7. By multi- plying 1.1 ppm with this factor, a value of 160 ppm was established as the expo- sure that would produce an excess risk of lung tumor incidence of 1 in 10,000 after 180 d of continuous exposure to DCM (see Wong 1996 for details). CHANGES IN FUNDAMENTAL NRC- RECOMMENDED APPROACHES The original SMACs were published in 1996 before the current NRC ap- proaches to data analysis (BMD and ten Berge approach) were commonly used. Values were derived using the approach of identifying and modifying LOAEL and NOAEL by default safety factors. New Data Since 1995 and Data Not Discussed in the Wong (1996) Document A survey of the literature around and after 1995 for any DCM toxicity studies yielded no new experimental data that could change the previous SMACs or that could be used to develop a 1,000-d AC. However, numerous papers have been published on the PBPK modeling and simulation of DCM me- tabolism to active intermediates, which have refined the model or addressed the variability of the parameters used, their distribution in the population, and the uncertainty associated with the distribution of model parameters. Some publica- tions recommended factors that should be considered and incorporated in the PBPK model simulation to derive meaningful risk estimates; for example, in- cluding the effect of exercise (Dankovic and Bailer 1994, Jonsson et al. 2001) and changes in target tissue kinetics because of aging (Thomas et al. 1996a,b). Additional Studies Not Discussed by Wong (1996) Nitschke et al. (1988a) conducted a 2-y inhalation toxicity and oncogenic- ity study in which they exposed male and female Sprague-Dawley rats to 0, 50, 200, or 500 ppm of DCM for 6 h/d, 5 d/wk for 2 y. These doses were chosen to identify a NOAEL, because the Burek et al. (1984) study determined only a LOAEL of 500 ppm for adverse hepatic effects. Furthermore, because the MFO system, and thus COHb formation, becomes saturated at 500 ppm, the authors
296 SMACs for Selected Airborne Contaminants chose these doses to be close to the saturation region to obtain a monotonic dose-response relationship. Liver lesions included increased incidence of hepa- tocellular vacuolization in male and female rats exposed to DCM at 500 ppm and an increased incidence of multinucleated hepatocytes in female rats. These effects were not seen in the 200-ppm DCM exposure treatment group. In the 2-y study (Nitschke et al.1988a), these authors identified 200 ppm of DCM as a NOAEL and 500 ppm as a LOAEL for hepatotoxicity in Sprague-Dawley rats. Gross pathology in all tissues and detailed histopathology were examined. Histopathologic lesions in DCM-treated female rats (up to 500 ppm) indicated that the observed incidences of benign mammary tumors (adenomas, fibromas, and fibroadenomas) with no progression to malignancy were comparable to his- torical control values. The authors also measured DNA synthesis (using [3H]thymidine incorporation) in the livers of female rats exposed to 200 or 500 ppm of DCM for 6 or 12 mo. The results were comparable to those for the con- trol groups. COHb was also measured during the interim durations of 6 and 12 mo. Though the levels increased as a function of dose, the rate of increase was less than linear with dose. Another study not discussed in the 1996 document on DCM was a 1988 study by Nitschke et al. (1988b), who conducted a two-generation DCM inhalation reproductive study in F344 rats. Male and female rats (30 each), approximately 7 wk old, were exposed to DCM at 0, 100, 500, or 1,500 ppm for 6 h/d, 5 d/wk for 14 wks and then mated (within the same treatment group) to produce F1 litters. The F0 rats continued to be exposed. Fertility, litter size and neonatal growth, and survival were determined as reproductive indices, and these were done for two successive generations. Also, after weaning, 30 F1 pups per gender per group (ran- domly selected) were exposed to DCM for 17 wk under the same schedule. They were mated to produce F2 litters. All animals were examined for visible lesions, and tissues were examined histopathologically. No changes were reported in any of the reproductive parameters measured and no abnormal tissue histopathology was observed in any of the F0, F1, or F2 weanlings. So, one might identify at least 1,500 ppm as the NOAEL for reproductive effects for up to 31 wk. As the NOAEL dose is rather high, NASA decided not to derive an AC for this end point. RATIONALE FOR THE 1,000-d SMAC In general, the ACs were determined according to the NRC guidelines for developing SMACs for space station contaminants (NRC 1992). The exposure limits set or recommended by other organizations are pre- sented in Table 15-3. Occupational Safety and Health Administration (OSHA) reduced the 8-h time-weighted average (TWA) permissible exposure level (PEL) and the short-term exposure limit in 1997, as it believed that strong evi- dence existed for a risk of human cancer incidence from occupational exposure to DCM.
Methylene Chloride 297 TABLE 15-3 Exposure Limits Recommended or Set by Other Organizations Concentration, Concentration, Organization, Standard ppm mg/m3 Reference ACGIH ACGIH 1986 TLV TWA, 8 ha 50 174 OSHA OSHA (62 Fed. PEL TWA, 8 ha, 25 87 Reg.1491 ) Action level 12.5 44 EPA carcinogenicity risk EPA 1995 1 in 10,000 0.06 0.2 1 in 100,000 0.006 0.02 1 in 1,000,000 0.0006 0.002 ATSDR ATSDR 2000 Acute duration inhalation 0.6 2.1 MRL Intermediate duration 0.3 1.0 inhalation MRL (150-364 d) Chronic duration inhalation 0.3 1.0 MRL (â¥365 d) a OSHA reduced the PEL from 500 ppm to 25 ppm.This is based on an upper 95th percen- tile of human internal dose distribution; if a mean of this distribution of this analysis were used, then the maximum likelihood estimate of extra cancer risk would be 1.24/1,000 for 25 ppm of DCM of an occupational lifetime exposure (8 h/d, 5 d/wk, for 45 y). The OSHA estimated risk at the previous PEL of 500 ppm is 126 excess cancer deaths per 1,000 workers; the revised standard of 25 ppm will effect a substantial reduction to a risk of 3.62 deaths per 1,000 workers occupationally exposed to DCM for a working lifetime. Abbreviations: ACGIH, American Conference of Governmental Industrial Hygienists; ATSDR, Agency for Toxic Substances and Disease Registry; EPA, U.S. Environmental Protection Agency; MRL, minimal risk level; TLV, Threshold Limit Value; TWA, time- weighted average. STATUS OF CARCINOGENICITY CLASSIFICATION FOR DCM â¢ International Agency for Research on Cancer (IARC) classification: Group 2B. Possibly carcinogenic to humans (inadequate evidence for the car- cinogenicity of DCM in humans and sufficient evidence in experimental ani- mals) (IARC 1999). â¢ U.S. Environmental Protection Agency (EPA) classification: B2 car- cinogen (a probable human carcinogen) (EPA 1995). â¢ Health Canada classification: Group II (probably carcinogenic to hu- mans) (Health Canada 1993). The Agency for Toxic Substances and Disease Registry (ATSDR) acute inhalation minimal risk level (MRL) was derived from the behavioral and per- formance effects of DCM in a human volunteer study by Winneke (1974). Audi- tory and visual performance and psychomotor task performance were decreased.
298 SMACs for Selected Airborne Contaminants The inhalation MRL for the intermediate duration was based on cytoplasmic vacuolization and nonspecific tubular degeneration changes in the kidney re- ported in rats in a 14-wk DCM exposure study by Haun et al. (1972). The MRL for the chronic exposure duration was derived from the 2-y inhalation toxicity study in rats, in which hepatocellular cytoplasmic vacuolization with fatty changes and multinuclear hepatocytes were reported (Nitschke et al. 1988b). DERIVATION OF 1,000-d ACs A toxicity summary for neoplastic and non-neoplastic adverse end points from studies used for deriving the 1,000-d AC are shown in Tables 15-2 and 15-4. The 1,000-d AC for DCM by inhalation is based on liver toxicity reported in three studies that used different ranges of concentrations. The NTP (1986) study reported cytoplasmic vacuolization, hemosiderosis, and focal granuloma- tous inflammation in liver in rats exposed to DCM at 1,000, 2,000, or 4,000 ppm, 6 h/d, 5 d/wk, for 2 y. Burek et al. (1984), who exposed male and female rats (90 each) to DCM at 0, 500, 1,500, or 3,500 ppm for 6 h/d, 5 d/wk, for 2 y, similarly reported cytoplasmic vacuolization, indicative of fatty liver, at concen- trations as low as 500 ppm. This study identified a LOAEL of 500 ppm for hepatotoxicity. A NOAEL concentration was not identified in either of the above studies. In the third study, a 2-y inhalation toxicity and oncogenicity study by Nitschke et al. (1988a), male and female Sprague-Dawley rats were exposed to 0, 50, 200, or 500 ppm of DCM for 6 h/d, 5 d/wk for 2 y. These doses were chosen to identify a NOAEL and to possibly obtain a monotonic dose-response relationship. Liver lesions observed included increased incidence of hepatocellu- lar vacuolization with fatty liver in male and female rats exposed to 500 ppm of DCM as well as increased incidence of multinucleated hepatocytes in female rats. These effects were not observed in the 200-ppm DCM exposure group. A NOAEL for hepatotoxic effects could be identified as 200 ppm in this study. A 1,000-d AC was derived from this study with supporting observations of similar liver toxicity end points from the NTP and Burek studies described above. Nitschke et al. (1988b) used three exposure concentrations. The incidence of hepatic vacuolization data was used in the EPA BMD software (EPA 2007) to derive the benchmark concentration (BMC) and the 95% lower confidence boundary on the BMC (the BMCL) for benchmark responses (BMR) of 10% (BMCL10), 5% (BMCL05), and 1% (BMCL01). Upon review of the results, it was decided to use a BMR of 10% as the most appropriate point of departure, on the advice of the statistician expert of the NRC committee who reviewed this docu- ment. The choice of a BMR of 10% was primarily based on the calculated added risk and biological plausibility of detecting such a change without being too conservative. Six different dose-response models offered in the EPA software were used, and BMC10 and BMCL10 were summarized for all models. No single
TABLE 15-4 Summary of Noncancer Effects of Chronic Inhalation Exposures to DCM Dosage (Number. Route and Dosing of Animals) Species/Strain Description of Effects Reference Inhalation 6 h/d, 5 0, 2,000, 4,000 ppm; B6C3F1 mouse 2,000 ppm: increased cytoplasmic vacuolization NTP 1986, Mennear d/wk, 2 y 50 mice/gender/dose in liver, consistent with fatty liver; 4,000 ppm: et al. 1988 hepatic cytologic degeneration in both genders; for controls, 2,000, and 4,000 ppm dose groups, respectively, testicular atrophy in male mice (0/50, 4/50, 31/50) and ovarian atrophy (6/50, 28/47, 32/43) and atrophy of the uterus (0/50, 1/48, 8/47) in female mice. Inhalation 6 h/d, 5 0, 1,000, 2,000, 4,000 F344 rat 1,000 and 2,000 ppm: hemosiderosis, NTP 1986; data d/wk, 2 y ppm; 50 rats/gender/ hepatomegaly, cytoplasmic vacuolization, focal summarized by dose necrosis, focal granulomatous inflammation in Mennear et al. 1988 liver (female rats) and bile ducts (male rats); 4,000 ppm: all the above plus reduced survival, squamous metaplasia of nasal cavity, benign papillary mesothelioma of tunica vaginalis in males, mononuclear cell leukemia in females. Inhalation 6 h/d, 5 0, 1,000, 2,000, 4,000 F344/N rat 1,000-4,000 ppm: renal tubular cell Mennear et al. 1988, d/wk, 2 y ppm; 50 degeneration in female rats. NTP 1986 rats/gender/dose Inhalation 6 h/d, 5 0, 500, 1,500, 3,500 Sprague-Dawley rat 500 ppm: cytoplasmic vacuolization in liver Burek et al. 1984 d/wk ppm; 95 rat/gender/ cells, multinucleated hepatocytes in female rats, dose 13% increase in COHb. 1,500 ppm: in addition to the above, necrosis of hepatocytes and chronic glomerulonephropathy were seen in male rats. (Continued) 299
TABLE 15-4 Continued 300 Dosage (Number. Route and Dosing of Animals) Species/Strain Description of Effects Reference Inhalation 6 h/d, 5 0, 500, 1,500, 3,500 Syrian golden 500-3,500 ppm: no injuries in hamsters; 27% Burek et al. 1984 d/wk, 2 y ppm; 90 hamster increase in COHb. hamsters/gender/dose Inhalation 6 h/d, 5 0, 50, 200, 500 ppm; Sprague-Dawley rat 500 ppm: (at terminal sacrifice) histopathologic Nitschke et al. 1988b d/wk, 2 y 70 rats/gender/dose; lesions in liver (multinucleated hepatocytes in end points were also females) and mammary tissues of rats evaluated at 6, 12, 15, (increased spontaneous benign mammary and 18 mo. tumors). NOAEL = 200 ppm; COHb concentrations increased (less than linear rate) with the DCM concentrations but leveled off after 6 mo of the 2-y study.
Methylene Chloride 301 model or detailed algorithm to choose a particular one was used to derive the final BMC or BMCL. A summary of the data from Nitschke et al. (1988b) used for obtaining the BMCs are shown in Table 15-5. Some of the output values from using the BMD software on this data are summarized in Table 15-6. Of the two end points, cytoplasmic vacuolization and multinucleated hepatocytes, the former was used because of its extensive use as an adverse end point in the toxi- cology literature. NASA calculated the BMC for both end points, but only one is shown here. The model-averaged BMC was calculated as a weighted average of the in- dividual BMCs using the posterior probabilities as weights to account for dose- response model uncertainty. However, the NASA statistician advised NASA that this weighted-average method should not be applied to lower confidence bounds for BMCsâthat is, to BMCLsâas BMCL does not represent model uncertainty (A. Feiveson, NASA, personal communication, 2004). Therefore, it was decided to estimate the BMCL from the model averaged BMC and the weighted mean of the BMC/BMCL ratio. Similar calculations done using the incidences of multinucleated hepato- cytes gave an estimated BMCL10 comparable to that of the BMCL10 for hepatic vacuolization; the data are not included here. TABLE 15-5 Incidence of Hepatic Vacuolization in Rats from DCM Inhalation DCM, ppm Number of Rats Incidence % affected 0 70 41 59 50 70 42 60 200 70 41 59 500 70 53 76 Source: Data from Nitschke et al. 1988b. Reprinted with permission; copyright 1988, Applied Toxicology. TABLE 15-6 Summary of BMC and BMCL for Hepatic Vacuolization for Various Models BMC10, BMCL10, BMC10/BMCL10 BMD model AIC P value ppm ppm ratio Gamma 365.80 0.980 383 69.7 5.49 Weibull 367.82 0.843 442 69.7 6.34 Multistage 366.12 0.945 292 69.0 4.23 Probit 366.92 0.559 119 74.2 1.60 Probit (log) 367.82 0.843 400 121.7 3.29 Logistic 366.90 0.551 117 72.3 1.63 Log-logistic 367.80 0.843 434 58.2 7.46 Abbreviation: AIC, Akaike information criterion.
302 SMACs for Selected Airborne Contaminants The models were averaged using weights derived from the Akaike infor- mation criterion. Weighted means were calculated for BMC10 and the ratio, which were 307 ppm and 4.10, respectively. The weighted mean for BMC10 was divided by the weighted ratio to obtain the estimated BMCL10 of 75 ppm. This value was used as the point of departure DCM concentration for calculating the AC for liver toxicity. The 1,000-d AC for hepatotoxicity will use the BMCL10 concentration, which will be adjusted from a discontinuous to a continuous duration. The NRC Committee on Spacecraft Exposure Guidelines suggested that, although adjust- ment factors for daily exposure and number of daily exposures per week should be used to arrive at a dose rate for continuous exposures, an additional correc- tion of 728 d (104 wk) to 1,000 d is not necessary, because 2 y is a greater frac- tion of a ratâs lifetime than 1,000 d is of a human lifetime. Thus, the value de- rived without a time extrapolation factor will still be protective for a long duration. The BMCL10 concentration will be adjusted for discontinuous to con- tinuous exposure by using the adjustment factors of (adjusted) 6h/24h and 5d/7d. BMCL10(adjusted) = 75 ppm(estimated BMCL10) Ã [6 h/24 h Ã 5 d/7 d] (discont. to contin.) = 13.39, rounded to13.5 ppm 1,000-d AC(hepatotoxicity) = BMCL10 (adjusted) Ã (1/10) (species factor) = 1.35 ppm, rounded to 1.4 ppm 1,000-d AC for Nephrotoxicity Using Data from NTP (1986) Another end point was used to derive a 1,000-d AC for a non-neoplastic effect reported in the NTP (1986) study. This study reported that female F344/N rats showed treatment-related squamous metaplasia of the nasal cavity, degen- eration of kidney tubules, and fibrosis of the spleen when exposed to 1,000, 2,000, or 4,000 ppm of DCM by inhalation for 6 h/d, 5 d/wk for 102 wk (Table 15-7). The LOAEL appears to be 1,000 ppm and no NOAEL could be identified. As the data showed a reasonable dose-response profile, they were processed by the BMD method and the LOAEL-NOAEL method was, therefore, not used. TABLE 15-7 Non-neoplastic Changes in Female F344/N Rats Exposed to DCM for 2 y Incidence At 1,000 At 2,000 At 4,000 Lesions Control ppm ppm ppm Renal tubular cell degeneration 14/50 20/50 22/50 25/49 Splenic fibrosis 0/50 2/50 4/50 4/49 Nasal cavity squamous 1/50 2/50 3/50 9/50 metaplasia Sources: Data from NTP 1986 and Mennear et al. 1988.
Methylene Chloride 303 The authors did not say whether these data were statistically analyzed. NASA assumes, based on the trend, that 1,000 ppm is the LOAEL. NASA re- viewed the data and, judging by the excess risk at 1,000 ppm for splenic fibrosis and the effect on the nasal cavity, it was clear that the AC that would be calcu- lated for renal tubular degeneration would drive the 1,000-d AC for the non- neoplastic lesions reported for the NTP study. A BMC was derived using the BMD method with the EPA BMD soft- ware. For this derivation, a BMR was set at 1% excess risk. The BMC01 and BMCL01 and the ratio summary data are presented in Table 15-8. The probit (log) model and the quantal-quadratic model BMDs were omit- ted as the estimated curve did not fit the data well. The weighted BMCL01 was computed from the weighted mean of the BMC01 and the weighted mean of the ratios of BMC01 and BMCL01 as described previously. The model weighted mean for BMC01 was 94 ppm and the weighted mean of the BMC/BMCL ratio was 1.485. The estimated BMCL01 is 63 ppm. A 1,000-d AC for renal tubular cell degeneration was calculated after as- certaining a BMCL01 (adjusted), which are dose estimates obtained for a con- tinuous exposure from a discontinuous exposure by multiplying BMCL01 by 6 h/24 h and 5 d/7 d. As explained earlier, no factor is used to account for the ex- trapolation for 1,000 d from 2 y (730 d). BMCL01(adjusted) = 63 ppm (estimated BMCL01) Ã [6 h/24 h Ã 5 d/7 d] (discontin. to contin.) = 11.25 ppm 1,000-d AC(nephrotoxicity) = BMCL01 (adjusted) Ã 1/10 (species factor) = 1.12 ppm, rounded to 1 ppm Thus, the 1,000-d AC for nephrotoxicity is 1.0 ppm. 1,000-d AC for Carcinogenicity Risk In the NTP (1986) study, the incidence of lung neoplasms in B6C3F1 mice exposed to DCM for 2 y was as follows: in the males, the incidence of adenomas and carcinomas combined was 5/50 in controls, 27/50 in the 2,000- ppm group, and 40/50 in the 4,000-ppm group; in the females, the incidence of adenomas and carcinomas combined was 3/50 in controls, 30/48 in the 2,000- ppm group, and 41/48 in the 4,000-ppm group. The incidence of hepatocellular neoplasms in B6C3F1 male mice exposed to DCM for 2 y was as follows: in male mice, adenomas and carcinomas com- bined were 22/50 in controls, 24/49 in the 2,000-ppm group, and 33/49 in the 4,000-ppm group; in female mice, they were 3/50 in controls, 16/48 in the 2,000-ppm group, and 40/48 in the 4,000-ppm group. According to the NTP report, the historical controls for this end point in this strain of mice are âMale:
304 SMACs for Selected Airborne Contaminants TABLE 15-8 DCM and Renal Tubular Degeneration (NTP 1986): Summary of Results from the BMD Method BMC01/BMC Modela P value AIC BMC01 BMCL01 L01 ratio Gamma 0.799 267.5 102.58 58.55 1.75 Weibull 0.799 267.5 102.58 58.55 1.75 Probit (no log) 0.706 267.8 138.15 91.99 1.50 Probit (log) 0.482 268.6 646.22 393.30 1.64 Logistic (no log) 0.699 267.8 141.33 94.67 1.49 Log-logistic 0.868 267.4 82.50 41.98 1.97 Multistage 2 0.799 267.5 102.58 58.55 1.75 Quantal linear 0.799 267.5 102.58 58.55 1.75 Quantal quadratic 0.413 268.9 699.57 500.45 1.40 a Both multistage degrees 2 and 3 gave the same values. Data from quantal linear and quantal quadratic were not included in the model averaging, as these two models are a variation of the multistage model. Data from the probit (log) model were also not in- cluded in the model averaging. Abbreviation: AIC, Akaike information criterion. 33% Â± 8% and for females it is 2.7% Â± 2.99%,â which are not very different from the numbers reported in this study. There were only two treatment groups in this study. Initially, NASA con- templated doing BMD modeling for all these data. However, it was decided to adopt the carcinogenicity excess risk determination with suitable factors for NASA extended-duration and exploration missions, because it is based on the target tissue dose metrics. The advanced PBPK model that was developed for these data from NTP (1986) should be preferable to values that could be derived from BMD analysis based on exposure concentrations. Data were collected in carcinogenicity bioassays using DCM exposure to three different species (rat, mouse, and hamster) by two routes of administra- tionâoral and inhalation. NTP concluded that there was some evidence of car- cinogenicity in male rats and there was clear evidence of lung and liver tumors in male and female mice as a result of exposure to DCM. OSHA believed that there was a significant risk of carcinogenicity to humans in an occupational set- ting and that the current PEL (time-weighted average PEL) is too high. OSHA (62 Fed. Reg.1491 ) based its risk analysis on two PBPK models that represented substantial refinement over the conventional risk esti- mates based on applied dose. While incorporating animal and human metabolic parameters in their risk analysis, OSHA extensively addressed the concepts of uncertainty, variability, and sensitivity of the model parameters used. The choice of OSHAâs risk estimate value is very sound, because OSHA (62 Fed. Reg.1491) used state-of-the-art advanced computational methods for PBPK modeling of the NTP lung carcinogenesis data with extrapolation to
Methylene Chloride 305 humans for an occupational environment. This was the result of a collaborative effort of several experts in toxicology, pharmacokinetics, and mathematics. The derivation was extensively scrutinized and commented on by the scientific community. OSHA (62Fed. Reg.1491 ) developed two PBPK models using the 95th percentile of the distribution of GST metabolites from the Bayesian analy- sis as the input to the multistage model, instead of using the 95th percentile of the Monte Carlo simulation distribution of GST metabolites, as the input rec- ommended by Clewell et al. (1993). The PBPK analysis showed a final estimate of risk of 3.62 deaths per 1,000 workers occupationally exposed to 25 ppm DCM for a working lifetime. An occupational lifetime here means exposure to DCM for 8 h/d, 5 d/wk for 45 y (as opposed to 70 y of human lifetime in EPA risk estimation procedures). Because of high confidence in the overall proce- dures that OSHA followed in developing and applying the PBPK model, NASA decided to use the new PEL value and modify it with factors that are applicable to NASA SMAC derivations. NASA assumed 25 ppm as the acceptable risk concentration (based on OSHA PEL) for the exposure conditions. This AC is adjusted to reflect space- craft exposure duration as follows: AC (adjusted) = 25 ppm (OSHA PEL Ã [8 h/24 h Ã 5 d/7 d] (discontin. to contin.) = 5.95, rounded to 6 ppm NASA accepts a cancer risk factor of 1 in 10,000. OSHAâs excess risk at 25 ppm is 3.62 deaths in 1,000, which may also mean 36.2 deaths in 10,000. Assuming a linear response relationship, the calculated exposure was divided by 36.2 to give a 1 in 10,000 risk, which is equal to 0.165 ppm if exposure contin- ues for 45 y. This approach is very conservative and may overestimate the can- cer risk. In 1992, the NRC SMAC committee recommended that NASA use time- compression factors to derive carcinogenicity risk for durations shorter than a lifetime. In this case, the duration is 1,000 d and the time that OSHA calculated is 45 y. Using the formula and approach that NRC (1992) provided to NASA, this factor can be calculated in the following way. According to NRC (1992), setting k = 3 (the number of stages in the car- cinogenic process affected by DCM) and t = 16,425 d (occupational lifetime of 45 y) to 10,950 d (an initial exposure age of 30 y), the adjustment factor for 1,000 d can be calculated to be 61.92 d (NRC 1992). Thus, the 1,000-d exposure that would produce a 1 in 10,000 excess can- cer risk is as follows: 0.165 ppm Ã 61.92 = 10 ppm If NASA uses 70 y as a lifetime, then the time-compression factor for 1,000 d becomes 27.953. Then, the 1,000-d AC is as follows:
306 SMACs for Selected Airborne Contaminants 1,000-d AC (carcinogenicity risk 1/10,000) = 0.165 ppm Ã 27.953 (time extrapolation, 70 y) = 4.6 ppm, rounded to 5 ppm Therefore, the 1,000-d AC for carcinogenicity risk at 1 in 10,000 is 5 ppm. As carcinogenicity to humans has not been conclusively ascertained, the GST activity in the mouse lung appears to be greater than in rats and humans, and since the exposure time used here is only 1,000 d, there is a large enough margin of safety at the 1,000-d AC of 5 ppm. Along the same lines, the compression factor for 180 d can be calculated as 146.7 (based on NRC 1992 guidelines), and the 180-d AC based on OSHA estimated values is 0.165 ppm Ã 146.7 = 24 ppm. Therefore, the 180-d AC for carcinogenicity risk for 1 in 10,000 will be 24 ppm. A summary of the 1,000-d ACs derived is shown in Table 15-9. REDERIVATION OF 30-d AND 180-d ACs A summary of SMACs (Wong, 1996) is shown in Table 15-1. One- hour, 24-h, 7-d, and 30-d ACs were determined on the basis of CO generation and formation of COHb, and the values from different investigations were fitted into a regression curve to calculate the ACs for DCM. Also, Wong (1996) used the PBPK model developed by Andersen et al. (1991) to compute the AC for 7, 30, and 180 d. Rederivation of 30-d AC Based on Hepatotoxicity The 30-d AC was based on the hepatotoxicity of DCM. Wong (1996) used the same study that he had used for 180 d (the NTP and the Burek et al. study), basing the calculation on the LOAEL. This document uses the Nitschke et al. (1988b) study in which a NOAEL is identified. Using BMD methodology on the dose-response data for hepatotoxicity, a BMCL10 of 75 ppm was derived, as described in the section on 1,000-d AC derivation, which was used to derive a 30-d AC. First, this concentration was adjusted for discontinuous to continuous exposure as follows. BMCL10(adjusted) = 75 ppm(BMCL10) Ã [6 h/ 24 h Ã 5 d/7 d] (discontin. to contin.) = 13.39 ppm, rounded to 13.4 ppm rounded NASA initially considered that it may not be appropriate to use a ten Berge interpolation factor from 2 y to 30 d and hence had derived the 30-d AC after using a species factor of 10 on the 2-y BMCL value of 75 ppm for hepato- toxicity without adjusting for discontinuous to continuous exposure. NASA thought that there might be enough margin of safety in this approach. However,
Methylene Chloride 307 TABLE 15-9 Summary of 1,000-d ACs Critical effect AC, ppm Principal Studies Hepatotoxicity 1.4 Nitschke et al. 1988a Nephrotoxicity 1.0 NTP 1986, Mennear et al. 1988 Carcinogenicity 5.0 NTP 1986, OSHA (62 Fed. Reg. 1491 ) PBPK model extrapolation the NRC Committee on spacecraft exposure guidelines (SEG) disagreed with NASA and concluded that NASA should use the ten Berge approach. As the committee recommended, ten Bergeâs time conversion method was used (NRC guideline document, NRC 2000) to derive a concentration for 30 d from the 730-d data. The ten Berge approach (ten Berge et al. 1986) for time interpolation (from longer duration to shorter duration) uses a default exponent of 2, as suggested by the National Advisory Committee for Acute Exposure Guideline Levels for Hazardous Substances in 1997 when no relevant duration versus response data were available to calculate the value for the exponent. The ten Berge approach is as follows. The ten Berge equation is CN Ã T = K where C = concentration,13.4 ppm, N = exponent, default factor of 2, T = exposure days, 730 d, and K = (13.4)2 Ã 730 d C (for 30 d) = (K/30)Â½ = approximately 66 ppm Thus, the ten Berge adjusted concentration is 66 ppm. 30-d AC(hepatotoxicity) = 66 ppm Ã 1/10 (species factor) = 6.6 ppm, rounded to 7 ppm Thus, 30-d AC = 7 ppm NASA decided to use only hepatotoxicity as the adverse end point for derivation of the 30-d AC and for the 180-d AC and not the nephrotoxicity data reported in the NTP study (NTP 1986), because, in the 13-wk rat study con- ducted as a part of the 2-y study, NTP did not observe any nephrotoxicity even at a dose as high as 8,400 ppm. Rederivation of 180-d AC Based on Hepatotoxicity Wong derived the 180-d AC based on the NOAEL-LOAEL method using the NTP (1986) study and the Burek et al. (1984) study in which a NOAEL was not identified for hepatotoxicity. Wong applied a LOAEL-to-NOAEL factor and
308 SMACs for Selected Airborne Contaminants a species factor and obtained a 180-d AC of 3.6 ppm for hepatotoxicity. In this document, the AC is updated using the Nitschke et al. (1988b) data, which indi- cated a NOAEL of 200 ppm for hepatic effects. Furthermore, the dose-response data is processed by the benchmark dose method and an AC is derived from the BMCL10 using a BMR of 10% excess risk. For calculating the 180-d AC, the BMCL10 of 75 ppm for hepatic vacuoli- zation derived earlier is adjusted for discontinuous to continuous exposure. BMCL10(adjusted) = 75 ppm (estimated BMCL10) Ã [6 h/24 h Ã 5 d/7 d] (discontin. to contin.) = 13.39 ppm, rounded to 13.4 ppm rounded. The ten Berge approach (ten Berge et al. 1986), as mentioned above for time interpolation, was used as follows. CN Ã T = K where N is the exponent, T is exposure days (730 d), and K is a constant. After calculating K, it was used to derive C with T = 180 d (target exposure duration). 180-d ten Berge time adjusted BMCL10 = 28 ppm 180-d AC(hepatotoxicity) = 28 ppm Ã 1/10 (species factor) = 2.8 ppm, rounded to 3 ppm Thus, the 180-d AC for hepatotoxicity is 3 ppm. A final summary of all SMACs for DCM is shown in Tables 15-10 and 15-11, with updated values incorporated as discussed in this document. ADDITIONAL INFORMATION During the development of this document, several articles were published on the risk assessment using PBPK modeling, especially for DCM. Most of these focused on using advanced statistical tools to address the probability dis- tributions of variability and uncertainty of the input parameters used in the PBPK modeling so one can use the data of the outcome to estimate the effective dose (predict the target dose) in humans more accurately. For example, Marino et al. (2006) refined the mouse PBPK model used to characterize the dose re- sponse of the tumor incidence in lung and liver of male and female B6C3F1 mice noted in the NTP (1986) carcinogen bioassay study for DCM. The authors used the Bayesian Markov chain Monte Carlo (MCMC) analysis. They showed that the internal dosimetrics (the target organ dose, the reactive GSH metabolite, S-(chloromethyl)glutathione) was 3- to 4-fold higher than doses that support the EPA cancer risk assessment, meaning a decrease in the magnitude of the dose response (that is, EPA was 4 times more conservative). Similarly, using the same methodology, David et al. (2006) refined and calibrated the human PBPK model for human DCM exposure (using human
Methylene Chloride 309 TABLE 15-10 Summary of Spacecraft Maximum Allowable Concentration Duration ppm mg/m3 Critical effect Principal studies 1h 100 350 CNS depression Various human data 24 h 35 120 CNS depression Andersen et al. 1991 7 da 14 49 CNS depression Andersen et al. 1991 30 d 7 24 Hepatotoxicity Nitschke et al. 1988b 180 d 3 10 Hepatotoxicity Nitschke et al. 1988b 1,000 d 1 3.5 Nephrotoxicity NTP 1986 a Wong (1996) rounded the 7-d SMAC of 14 to 15. This committee insisted that NASA not round this value to 15. TABLE 15-11 Acceptable Concentrations for Cancer Risk of 1 in 10,000 Critical Duration ppm mg/m3 effect Principal studies 180 d 24 84 Cancer NTP 1986, OSHA (62 Fed. Reg.1491 ) 1,000 d 5 18 Cancer NTP 1986, OSHA (62 Fed. Reg.1491 ) DCM exposure studies composed of 43 subjects from five published studies) by incorporating the human genetic GSTT1 polymorphism (20% nonconjugators in the U.S. population) into an MCMC PBPK model. The authors concluded that the unit risk for cancer from DCM (for lung and liver together) is only 9.33 Ã 10â10 (50th percentile), which is about 500 times lower than the current EPA unit risk of 4.7 Ã 10â7. That means EPA is 500 times more conservative than what these advanced probabilistic PBPK modeling methodologies can estimate for human carcinogenic risk for inhaled DCM. Note that unit risk is defined as the risk of cancer from exposure to DCM at 1 microgram/m3 over a lifetime. These studies and others were discussed in a forum on the reassessment of the cancer risk of DCM in humans (Starr et al. 2006). REFERENCES ACGIH (American Conference of Governmental Industrial Hygienists). 1986. Methylene chloride. Documentation of the Threshold Limit Values and Biologic Exposure In- dices. American Conference of Governmental Industrial Hygienists, Cincinnati, Ohio. Ahmed, A.E., and M.W. Anders. 1976. Metabolism of dihalomethanes to formaldehyde and inorganic halide. I. In vitro studies. Drug Metab. Dispos. 4(4):357-361.
310 SMACs for Selected Airborne Contaminants Andersen, M.E., H.J. Clewell III, M.L. Gargas, F.A. Smith, and R.H. Reitz. 1987. Physiologically based pharmacokinetics and the risk assessment process for me- thylene chloride. Toxicol. Appl. Pharmacol. 87(2):185-205. Andersen, M.E., H.J. Clewell III, M.L. Gargas, M.G. MacNaughton, R.H. Reitz, R.J. Nolan, and M.J. McKenna. 1991. Physiologically based pharmacokinetic modeling with dichloromethane, its metabolite, carbon monoxide, and blood carboxyhemo- globin in rats and humans. Toxicol. Appl. Pharmacol. 108(1):14-27. Astrand, I., P. Ovrum, and A. Carlsson. 1975. Exposure to methylene chloride. I. Its con- centration in alveolar air and blood during rest and exercise and its metabolism. Scand. J. Work Environ. Health 1(2):78-94. ATSDR (Agency for Toxic Substances and Disease Registry). 2000. Toxicological Pro- file for Methylene Chloride. U.S. Department of Health and Human Services, Pub- lic Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, GA. September 2000 [online]. Available: http://www.atsdr.cdc.gov/toxprofiles/ tp14.pdf [accessed Jan. 17, 2008]. Burek, J.D., K.D. Nitschke, T.J. Bell, D.L. Wackerle, R.C. Childs, J.E. Beyer, D.A. Dit- tenber, L.W. Rampy, and M.J. McKenna. 1984. Methylene chloride: A two year inhalation toxicity and oncogenicity study in rats and hamsters. Fundam. Appl. Toxicol. 4(1):30-47. Casanova, M., D.A. Bell, and H.D. Heck. 1997. Dichloromethane metabolism to formal- dehyde and reaction of formaldehyde with nucleic acids in hepatocytes of rodents and humans with and without glutathione S-transferase T1 and M1 genes. Fundam. Appl. Toxicol. 37(2):168-180. Clewell, H.J. 1995. Incorporating biological information in quantitative risk assessment: An example with methylene chloride. Toxicology 102(1-2):83-94. Clewell, H.J., J.M. Gearhart, and M.E. Andersen. 1993. Analysis of the Metabolism of Methylene Chloride in the B6C3F1 Mouse and its Implications for Human Car- cinogenic Risk. Submission to OSHA Docket No. H-071, Exhibit No. 96. January 15, 1993. Dankovic, D.A, and A.J. Bailer. 1994. The impact of exercise and intersubject variability on dose estimates for dichloromethane derived from a physiologically based phar- macokinetic model. Fundam. Appl. Toxicol. 22(1):20-25. David, R.M., H.J. Clewell, P.R. Gentry, T.R. Covington, D.A. Morgott, and D.J. Marino. 2006. Revised assessment of cancer risk to dichloromethane. II. Application of probabilistic methods to cancer risk determinations. Regul. Toxicol. Pharmacol. 45(1):55-65. DiVincenzo, G.D., and C.J. Kaplan. 1981. Uptake, metabolism, and elimination of me- thylene chloride vapor by humans. Toxicol. Appl. Pharmacol. 59(1):130-140. DiVincenzo, G.D., F.J. Yanno, and B.D. Astill. 1972. Human and canine exposure to methylene chloride vapor. Am. Ind. Hyg. Assoc. J. 33(3):125-135. El-Masri, H.A, D.A. Bell, and C.J. Portier. 1999. Effects of glutathione transferase theta polymorphism on the risk estimates of dichloromethane to humans. Toxicol. Appl. Pharmacol. 158(3):221-230. EPA (U.S. Environmental Protection Agency). 1995. Dichloromethane (CASNR 75-09- 02). Integrated Risk Information System, U.S. Environmental Protection Agency [online]. Available: http://www.epa.gov/NCEA/iris/subst/0070.htm [accessed Jan. 17, 2008]. EPA (U.S. Environmental Protection Agency). 2007. Benchmark Dose Software (BMDS) Version 1.4.1c. National Center for Environmental Assessment, U.S. En-
Methylene Chloride 311 vironmental Protection Agency [online]. Available: http://www.epa.gov/ncea/ bmds/ [accessed July 22, 2008]. Gargas, M.L., H.J. Clewell III, and M.E. Anderson. 1986. Metabolism of inhaled diha- lomethanes in vivo: Differentiation of kinetic constants for two independent path- ways. Toxicol. Appl. Pharmacol. 82(2):211-223. Guengerich, F.P., D.H. Kim, and M. Iwasaki. 1991. Role of human cytochrome P-450 IIE1 in the oxidation of many low molecular weight cancer suspects. Chem. Res. Toxicol. 14(2):168-179. Hallier, E., K.R. Schroder, K. Asmuth, A. Dommermuth, B. Aust, and H.W. Goergens. 1994. Metabolism of dichloromethane (methylene chloride) to formaldehyde in human erythrocytes: Influence of polymorphism of glutathione transferase theta (GST T1-1). Arch. Toxicol. 68(7):423-427. Haun, C.C., E.H. Vernot, K.I. Darmer, and S.S. Diamond. 1972. Continuous animal ex- posure to low levels of dichloromethane. Paper No. 12. Pp. 199-208 in Proceed- ings of the Third Annual Conference on Environmental Toxicology. AMRL-TR- 72-130. Wright-Patterson Air Force Base, Dayton, OH. Health Canada. 1993. Canadian Environmental Protection Act (CEPA). Priority Sub- stances List (PSL) Assessment Report: Dichloromethane. Ottawa: Ministry of Pub- lic Works and Government Services [online]. Available: http://www.hc- sc.gc.ca/ewh-semt/alt_formats/hecs-sesc/pdf/pubs/contaminants/psl1-lsp1/dichloro methane/dichloromethane-eng.pdf [accessed July 14, 2008]. Hearne, F.T., F. Grose, J.W. Pifer, B.R. Friedlander, and R.L. Raleigh. 1987. Methylene chloride mortality study: Dose-response characterization and animal model com- parison. J. Occup. Med. 29(3):217-228. Hearne, F.T., J.W. Pifer, and F. Grose. 1990. Absence of adverse mortality effects in workers exposed to methylene chloride: An update. J. Occup. Med. 32(3):234-240. IARC (International Agency for Research on Cancer). 1999. Dichloromethane. Pp. 251- 315 in Re-Evaluation of Some Organic Chemicals, Hydrazine, and Hydrogen Per- oxide Part 1. IARC Monographs on the Evaluation of Carcinogenic Risks to Hu- mans. Lyon, France: IARC [online]. Available: http://www.inchem.org/ documents/iarc/vol71/004-dichloromethane.html [accessed Jan. 18, 2008]. Jonsson, F., and G. Johanson. 2001. A Bayesian analysis of the influence of GSTT1 polymorphism on the cancer risk estimate for dichloromethane. Toxicol. Appl. Pharmacol. 174(2):99-112. Jonsson, F., F. Bois, and G. Johanson. 2001. Physiologically based pharmacokinetic modeling of inhalation exposure of humans to dichloromethane during moderate to heavy exercise. Toxicol. Sci. 59(2):209-218. Katoh, T., N. Nagata, Y. Kuroda, H. Itoh, A. Kawahara, N. Kuroki, R. Ookuma, and D.A. Bell. 1996. Glutathione S-transferase M1 (GSTM1) and T1 (GSTT1) genetic polymorphism and susceptibility to gastric and colorectal adenocarcinoma. Car- cinogenesis 17(9):1855-1859. Kubic, V.L., and M.W. Anders. 1975. Metabolism of dihalomethanes to carbon monox- ide. II. In vitro studies. Drug Metab. Dispos. 3(2):104-112. Lanes, S.F., A. Cohen, K.J. Rothman, N.A. Dreyer, and K.J. Soden. 1990. Mortality of cellulose fiber production workers. Scand. J. Work Environ. Health 16(4):247-251. Lanes, S.F., K.J. Rothman, N.A. Dreyer, and K.J. Soden. 1993. Mortality update of cellu- lose fiber production workers. Scand. J. Work Environ. Health 19(6):426-428. Marino, D.J, H.J. Clewell, P.R. Gentry, T.R. Covington, C.E. Hack, R.M. David, and D.A. Morgott. 2006. Revised assessment of cancer risk to dichloromethane: Part I
312 SMACs for Selected Airborne Contaminants Bayesian PBPK and dose-response modeling in mice. Regul. Toxicol. Pharmacol. 45(1):44-54. McKenna, M.J., J.A. Zempel, and W.H. Braun. 1982. The pharmacokinetics of inhaled methylene chloride in rats. Toxicol. Appl. Pharmacol. 65(1):1-10. Mennear, J.H., E.E. McConnell, J.E. Huff, R.A. Renne, and E. Giddens. 1988. Inhalation toxicity and carcinogenesis studies of methylene chloride (dichloromethane) in F344/N rats and B6C3F1 mice. Ann. N.Y. Acad. Sci. 534:343-351. Nelson, H.H., J.K. Wiencke, D.C. Christiani, T.J. Cheng, Z.F. Zuo, B.S. Schwartz, B.K. Lee, M.R. Spitz, M. Wang, X.P. Xu, and K.T. Kelsey. 1995. Ethnic differences in the prevalence of the homozygous deleted genotype of glutathione S-transferase theta. Carcinogenesis 16(5):1243-1245. Nitschke, K.D., D.L. Eisenbrandt, L.G. Lomax, and K.S. Rao. 1988a. Methylene chlo- ride: Two-generation inhalation reproductive study in rats. Fundam. Appl. Toxicol. 11(1):60-67. Nitschke, K.D., J.D. Burek, T.J. Bell, R.J. Kociba, L.W. Rampy, and M.J. McKenna. 1988b. Methylene chloride: A 2-year inhalation toxicity and oncogenicity study in rats. Fundam. Appl. Toxicol. 11(1):48-59. NRC (National Research Council). 1992. Guidelines for Developing Spacecraft Maxi- mum Allowable Concentrations for Space Station Contaminants. Washington, DC: National Academy Press. NRC (National Research Council). 2000. Methods for Developing Spacecraft Water Ex- posure Guidelines. Washington, DC: National Academy Press. NTP (National Toxicology Program). 1986. Toxicology and Carcinogenesis Studies of Dichloromethane (Methylene Chloride) in F344/N Rats and B6C3F1 Mice (Inhala- tion Studies). NTP Technical Report 306. NIH Publication No. 86-2562. U.S. De- partment of Health and Human Services, Public Health Services, National Institute of Health, National Toxicology Program, Research Triangle Park, NC. Ott, M.G., L.K. Skory, B.B. Holder, J.M. Bronson, and P.R. Williams. 1983. Health evaluation of employees occupationally exposed to methylene chloride. Scand. J. Work Environ. Health 9(Suppl. 1):8-16. Pemble, S., K.R. Schroeder, S.R. Spencer, D.J. Meyer, E. Hallier, H.M. Bolt, B. Ketterer, and J.B. Taylor. 1994. Human glutathione S-transferase theta (GSST1): cDNA cloning and the characterization of a genetic polymorphism. Biochem. J. 300(Pt.1): 271-276. Peterson, J.E. 1978. Modeling the uptake, metabolism and excretion of dichloromethane by man. Am. Ind. Hyg. Assoc. J. 39(1):41-47. Putz, V.R., B.L. Johnson, and J.V. Setzer. 1979. A comparative study of the effects of carbon monoxide and methylene chloride on human performance. J. Environ. Pathol. Toxicol. 2(5):97-112. Reitz, R.H., A.L Mendrala, and F.P. Guengerich. 1989. In vitro metabolism of methylene chloride in human and animal tissues: use in physiologically based pharmacoki- netic models. Toxicol. Appl. Pharmacol. 97(2):230-246. Starr, T.B., G. Matanoski, M.W. Anders, and M.E. Andersen. 2006. Workshop overview: Reassessment of the cancer risk of dichloromethane in humans. Toxicol. Sci. 91(1):20-28. Stewart, R.D., T.N. Fisher, M.J. Hosko, J.E. Peterson, E.D. Baretta, and H.C. Dodd. 1972. Experimental human exposure to methylene chloride. Arch. Environ. Health 25(5):342-348.
Methylene Chloride 313 ten Berge, W.F., A. Zwart, and L.M. Appelman. 1986. Concentration-time mortality response relationship of irritant and systemically acting vapours and gases. J. Haz- ard. Mater. 13(3):301-309. Their, R., F.A. Wiebel, A. Hinkel, A. Burger, T. Bruning, K. Morgenroth, T. Senge, M. Wilhelm, and T.G. Schulz. 1998. Species differences in the glutathione transferase GSTT1-1 activity towards the model substrates methyl chloride and dichloro- methane in liver and kidney. Arch. Toxicol. 72(10):622-629. Thomas, R.S., R.S. Yang, D.G. Morgan, M.P. Moorman, H.R. Kermani, R.A. Sloane, R.W. OâConnor, B. Adkins Jr., M.L. Gargas, and M.E. Andersen. 1996a. PBPK modeling/Monte Carlo simulation of methylene chloride kinetic changes in mice in relation to age and acute, subchronic, and chronic inhalation exposure. Environ. Health Perspect. 104(8):858-865. Thomas, R.S., P.L. Bigelow, T.J. Keefe, and R.S. Yang. 1996b. Variability in biological exposure indices using physiologically based pharmacokinetic modeling and Monte Carlo simulation. Am. Ind. Hyg. Assoc. J. 57(1):23-32. Winneke, G. 1974. Behavioral effects of methylene chloride and carbon monoxide as assessed by sensory and psychomotor performance. Pp. 130-144 in Behavioral Toxicology, C. Xinitaras, B.L. Johnson, and I. deGroot, eds. Washington DC: U.S. Government Printing Office. Winneke, G. 1981. The neurotoxicity of dichloromethane. Neurobehav. Toxicol. Teratol. 3(4):391-395. Winneke, G., and G.G. Fodor. 1976. Dichloromethane produces narcotic effect. Occup. Health Saf. 45(2):34-49. Wong, K.L. 1994. Carbon monoxide. Pp. 61-90 in Spacecraft Maximum Allowable Con- centrations for Selected Airborne Contaminants, Vol. 1. Washington, DC: National Academy Press. Wong, K.L. 1996. Methylene chloride. Pp. 277-305 in Spacecraft Maximum Allowable Concentrations for Selected Airborne Contaminants, Vol. 2. Washington, DC: Na- tional Academy Press.