Use of benchmark dose models in risk assessment for occupational handlers of eight pesticides used in pome fruit production

Journal Pre-proof Use of benchmark dose models in risk assessment for occupational handlers of eight pesticides used in pome fruit production Jane Gurnick Pouzou, John Kissel, Michael G. Yost, Richard A. Fenske, Alison C. Cullen PII:

S0273-2300(19)30268-5

DOI:

https://doi.org/10.1016/j.yrtph.2019.104504

Reference:

YRTPH 104504

To appear in:

Regulatory Toxicology and Pharmacology

Received Date: 21 June 2019 Revised Date:

12 October 2019

Accepted Date: 15 October 2019

Please cite this article as: Pouzou, J.G., Kissel, J., Yost, M.G., Fenske, R.A., Cullen, A.C., Use of benchmark dose models in risk assessment for occupational handlers of eight pesticides used in pome fruit production, Regulatory Toxicology and Pharmacology (2019), doi: https://doi.org/10.1016/ j.yrtph.2019.104504. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.

1

2

1. Introduction Despite a decades-old and often-echoed recommendation to adopt benchmark dose (BMD)

3 modeling as the default basis for regulatory limits1-6, the method has had slow acceptance. Use of the no 4 observable adverse effect level (NOAEL) is still the standard procedure for derivation of a regulatory 5 limit in many cases, including the US Environmental Protection Agency (EPA)7 and the Organisation for 6 Economic Cooperation and Development (OECD) Guidelines for the Testing of Chemicals8 (for example 7 the evaluation of neurotoxicity in rodents9). The limitations of the NOAEL approach have been 8 described3, 10. One of the most significant is that because selection of the NOAEL depends on the 9 identification of a statistical difference between groups, smaller group sizes are less likely to have 10 sufficient power to identify a small effect and will therefore produce a higher NOAEL. In contrast, the 11 BMDL method is based on uncertainty intervals – which are wider with a smaller n – and will produce a 12 lower BMDL when group sizes reduce statistical power11. Furthermore, the identified point of departure 13 necessarily must fall among the pre-selected dose groups used in the study when using the NOAEL 14 approach. Because of this constraint, there may be very little biological basis behind the specific dose 15 identified as a limit, despite the implications of regulatory meaning assigned and the use of the NOAEL 16 in quantification, particularly if the study was designed in the absence of existing toxicology studies. 17 Travis et. al10 identified a number of reasons why BMD methods have not been adopted with greater 18 speed and why the NOAEL should remain the predominant tool in determining the point of departure 19 (POD). For example, they suggest that the NOAEL is more intuitive and easier to verify and understand, 20 that quantal outcome BMDs cannot accurately reflect the same kind of outcome in human populations 21 since they are based on variability which may not correspond across species – also applicable to the 22 NOAEL – and that the BMD is too sensitive to the model type selected (a criticism now addressed 23 through model averaging methods). Despite these criticisms the BMD has become widely accepted as 24 the more advanced method in more recent years.12 The intuitive appeal of the NOAEL may lead to a 1

25 false sense of safety while the BMD more explicitly relies on an acceptable minimum effect of exposure, 26 and neither the NOAEL nor the BMD is guaranteed to provide zero risk. Furthermore, the applicability of 27 the NOAEL estimation may also suffer from differences in variability between human and animal 28 populations13. Simplicity of method is not a benefit if it limits the scientific basis and usefulness of the 29 result. The BMD method is easier to reconcile with uncertainties in dose estimates than deterministic 30 methods such as the NOAEL, and the resulting values are more useful in probabilistic assessments of 31 risk14. Wignall et al. discuss the lack of transparency and consistency in the BMD approach, and suggest 32 that a 1 standard deviation or 10% critical effect could serve as a standard basis for unified benchmark 33 dose modeling for large sets of toxicological data15. 34 More recently, Zeller et al.16 and Slob17 have suggested that endpoint-specific critical effect sizes allow 35 the most practical and most relevant assessments. Zeller and colleagues made the case that an 36 extended standard deviation-based approach using historical control data provides the best endpoint 37 which accounts for assay-specific and animal model-specific variability. Similarly, Slob suggests that the 38 critical effect size for a given endpoint should be adjusted based on the maximum value of the response 39 and the within-group variation. These measures would provide more biologically relevant points of 40 departure; however, the Zeller et al. method requires information on the historical values of these 41 endpoints in each animal model, which is not necessarily included in the typical study protocol. 42

Pesticide regulation in the United States is based on toxicological studies performed in

43 accordance with OECD guidelines, and is one area of chemical risk assessment where the NOAEL, along 44 with a set of uncertainty factors, is the basis of regulatory limits. Numerous tests are required for 45 registration of an active ingredient, assessing potential human and ecological impacts of any proposed 46 use. In the assessment of these studies, where a hazard is deemed present, one or more NOAELs is 47 chosen to pair with residential, dietary, and occupational doses to humans based on the length of the 48 exposure, sensitivity and specificity of the study and outcome, and the route of exposure. For 2

49 occupational exposures the NOAEL is divided by an estimate of the human dose over a single day to 50 calculate the Margin of Exposure (MOE)18, which must be above a Level of Concern (LOC) – 100 being 51 the most commonly used7. Substitution of a benchmark dose into this existing paradigm should be 52 feasible if the existing studies can be used to generate dose-response curves. A further advantage of this 53 method is that integration of new approaches involving physiologically-based pharmacokinetic (PBPK) 54 model development can be integrated with the dose-response models used in production of the BMD19. 55

The purpose of this analysis is to demonstrate the feasibility of using OECD guideline-compliant

56 toxicology studies carried out to generate NOAELs for production of dose-response models and 57 benchmark doses with associated lower confidence limits for a variety of pesticides that are or were 58 popular for the control of codling moth in tree fruit, and to determine whether the NOAEL or the BMD 59 approach is consistently more protective against acute exposures. The effect of using a variety of critical 60 effect sizes in continuous data on the BMD and the resulting Margin of Exposure for acute, one day 61 doses, is also explored. 62 2. Methods 63

Eight pesticides with a variety of potential acute or sub-acute health impacts as identified by

64 their respective EPA human health risk assessments were selected for this analysis: azinphos methyl, 65 acetamiprid, emamectin benzoate, methoxyfenozide, novaluron, phosmet , spinetoram, and 66 thiacloprid20-28. These pesticides are currently or were formerly used frequently in the production of 67 pome fruit, and have comparable task profiles for their application The one-day dose in mg/kg body 68 weight/day, which is the basis of occupational exposure regulation, of a pesticide handler mixing, 69 loading, and applying each pesticide using open cab airblast methods was calculated probabilistically as 70 described previously29, and is also described in the supplemental material. In brief, exposure data from 71 the Agricultural Handler Exposure Database (AHED) and the Pesticide Handler Exposure Database

3

72 (PHED) were used to construct probabilistic exposure rates similar to those deterministic rates used in 73 official calculations. The distributions of these rates were paired with distributions of other factors, 74 including clothing and PPE protection factors, application rates and areas, and anthropometric 75 variables30-32. In the EPA human health assessments, occupational exposures to methoxyfenozide and 76 spinetoram were calculated for the inhalation route only, as there was not considered to be evidence of 77 a hazard via the dermal route based on acute dermal studies27, 33. To remain consistent with the EPA 78 methods, probabilistic dose estimates were generated both with and without the dermal pathway for 79 spinetoram and methoxyfenozide. 80

Data from the same studies used to select a NOAEL for use in EPA occupational risk assessments

81 were used to construct multiple benchmark dose models for each pesticide. The studies, identified by 82 Master Record Identification (MRID) in Table 1, were obtained via Freedom of Information Act request. 83 Using the EPA Benchmark Dose Software (version 3.1.1.), available models were fit to the dataset, 84 according to whether the endpoint was quantal or continuous. Quantal models were fit using gamma, 85 logistic, log-logistic, log-probit, probit, Weibull, and quantal-linear models. Continuous models included 86 exponential, Hill, linear, polynomial, and power models. The BMD Software uses maximum likelihood 87 methods in the calculation of model parameters, and the confidence interval of the benchmark dose is 88 calculated using profile likelihood methods3. A 10% effect level was used for all quantal models to 89 permit the most comparability among models and among health outcomes. A benchmark response level 90 was chosen for each continuous outcome in several ways to compare the result of each method. 91

The response or critical effect size for the continuous models was first based on levels reported

92 by Dekkers et al. specified in a survey of experts on the commonly recommended effect size for a variety 93 of outcomes34. For all endpoints except cholinesterase depression, which was assigned 20% as a known 94 toxicologically-relevant effect size35, this level was 10% relative deviation from the control group. 95 Phosmet and azinphos methyl were modeled with both 10% and 20% depression as the critical effect 4

96 size. In addition, since phosmet risk assessment was based on a combination of data from an oral and 97 dermal toxicological study, the dermal dose was compared to the points of departure from the dermal 98 study without the inhalation dose. Second, based on the recommendation of EPA benchmark dose 99 modeling guidelines36, the 1 standard deviation effect size was also examined for continuous impacts. 100 Using the method described by Slob et al.17, the maximum value of the endpoint was used to derive a 101 more biologically-relevant effect size for each pesticide by dividing the natural log of the maximum by 102 eight, the Normalized Effect Size Benchmark Dose or BMDNES.The hybrid method5 was also applied using 103 a 10% effect level for all continuous outcomes for comparison against the relative effect BMDs. 104 105

2.1. Selection of toxicological outcomes for benchmark dose models 2.1.1. Acetamiprid

106 Symptoms of developmental neurotoxicity were recorded in rats through a functional observation 107 battery of neurotoxicity testing in offspring of orally-dosed dams (for 6 weeks perinatally). The most 108 relevant acetamiprid outcome recorded was change in the auditory startle reflex amplitude maximum in 109 males at post-natal days (PND) 20 and 60. Other outcomes (including reductions in pup viability and 110 alterations in weight gain) were non-specific to neurodevelopmental impacts. 111

2.1.2. Azinphos methyl

112 The outcome of interest for assessment of azinphos methyl neurotoxicity was identified a priori as 113 cholinesterase depression. Erythrocyte, plasma, and brain cholinesterase were measured at varying time 114 points during the 1-year feeding study performed with dogs, and at 13 weeks, significant depressions in 115 all three cholinesterases’ activity were recorded. Although acute studies assessing cholinesterase 116 inhibition in animals and in humans were available, this study was judged by the EPA to be the most 117 protective and appropriately conducted for comparison with biomonitoring dose measurements of 118 pesticide applicators. 119 2.1.3. Emamectin benzoate 5

120 The emamectin benzoate feeding study of acute (15-day) neurotoxicity in mice was used to find the 121 most protective NOAEL for occupational risk assessment. Several endpoints indicative of neurotoxicity 122 were recorded; tremors were the first frank symptom to be observed, followed by ptosis, gait and 123 posture abnormalities, decreased activity, urine staining, and labored breathing. At necropsy, some 124 animals in the highest dose groups had sciatic nerve degeneration. Since tremors appeared first and at 125 the lowest doses of all symptoms, they were regarded as the most sensitive indicator and selected for 126 this analysis and by the EPA for derivation of the NOAEL. 127 2.1.4. Methoxyfenozide 128 Occupational exposures to methoxyfenozide were assessed by the EPA only for inhalation, as acute 129 dermal toxicity studies did not indicate a hazard according to the EPA human health risk assessment for 130 this compound. Although various other outcomes were investigated, hematological impacts shown in a 131 two-week feeding study of dogs were selected to derive the NOAEL used in the occupational risk 132 assessment. This study included only two animals per sex per dose group; however, similar 133 hematological toxicity was observed at 3 months in the 1-year study with a sample size of 4 per group. 134 This study was therefore used to derive the benchmark dose model. 135 2.1.5. Novaluron 136 The NOAEL used for occupational risk assessment of Novaluron handlers was drawn from a 90-day 137 feeding study performed in rats which assessed a variety of hematologic parameters. Red blood count, 138 hematocrit, and hemoglobin were all influenced in the higher dose groups. In addition, spleen and liver 139 pigmentation and splenic hematopoiesis were observed, and a combination of all of these impacts led to 140 the derivation of the NOAEL from this study. 141 2.1.6. Phosmet 142 Like azinphos methyl, phosmet is known to act as a cholinesterase inhibitor, therefore the outcome of 143 cholinesterase activity was measured in acute toxicity studies. In the case of phosmet, a dermal and oral

6

144 study were both used to generate separate NOAELs for the routes of exposure, since there was no 145 human biomonitoring data as with the case of azinphos methyl. 146 2.1.7 Spinetoram 147 As with methoxyfenozide, insufficient evidence of dermal toxicity in the short or intermediate term was 148 found in the EPA human health risk assessment to warrant a complete risk assessment of that exposure 149 route beyond the hazard assessment stage. The outcomes used in the derivation of the NOAEL for the 150 inhalation route of exposure were also hematologic, drawn from the sub-chronic (90-day) feeding study 151 performed with dogs. Blood cell and hematocrit levels were significantly affected, and anemia, arteritis 152 and bone marrow necrosis were observed. The anemia and lowered platelet counts observed were 153 believed to be secondary to the bone marrow necrosis, so the arteritis and bone marrow necrosis were 154 both evaluated. 155 2.1.8. Thiacloprid 156 The normal battery of toxicological evaluations showed a number of potential impacts of thiacloprid 157 dosing. Occupational doses were evaluated by the EPA using a NOAEL derived from liver and thyroid 158 impacts observed in the subchronic inhalation and chronic feeding studies of rats. A variety of liver 159 impacts were recorded, including enzymatic induction (N-demethylase, O-demethylase, and CYP450), 160 and hepatocellular hypertrophy. Thyroid hypertrophy was also noted. Of the two organs, the liver 161 impacts were evaluated at lower doses. Thiacloprid is also classified as a likely human carcinogen, but as 162 the EPA occupational risk assessment was based on organ toxicities as the more protective outcomes, a 163 cancer risk assessment was not performed in this analysis. 164 2.2 Model assessment 165 Each possible model was assessed for goodness-of-fit using qualitative evaluation of the dose-response 166 graph and the p-value of the Χ2 goodness-of-fit test. Models were compared within each toxicological 167 endpoint using the qualitative fit of the curve, the Akaike Information Criterion (AIC), and the residuals. 7

168 The 95% confidence limit was calculated and the resulting BMD and the lower confidence limit of the 169 benchmark dose (BMDL) were compared with the NOAEL from the same study and the deterministic 170 dose used in the EPA human health assessment for the same pesticide application scenario. The 171 estimated dose distribution was also compared with all three points of departure (BMD, BMDL, and 172 NOAEL) using the calculated exceedance fraction, defined as the proportion of the dose distribution 173 which is above the POD, computed using the efraction.exact command of the R package STAND37. 174

3. Results

175 The toxicological studies and endpoints of interest for each 176 pesticide are described in Table 1. Although each compound was associated with multiple studies which 177 yielded toxicological endpoints potentially useful in creating a benchmark dose, the studies presented 178 here are those used in the EPA’s human health risk assessment for pesticide handlers to explore the 179 impact of use of a benchmark dose in place of a NOAEL. Goodness of fit values and the resulting BMD 180 and BMDL for each possible model fit to the eight pesticides’ outcomes can be found in Supplemental

8

181 Table B. The graphs for the selected models are shown in

182 183 Figure 1. Parameters for the chosen models are 184 listed in supplemental material. 185 3.1. Selection of toxicological outcomes and benchmark dose models 186 3.1.1. Acetamiprid 187 The outcomes not specific to neurodevelopmental toxicity observed in this study were not able to 188 produce a dose-response model due to variability in the control animals. PND 20 was selected for this 189 analysis as the dose-response effect was more evident than at PND 60. Of the models assessed, similar 190 results for goodness of fit and residuals were generated, and the Hill model had the lowest AIC. 191

3.1.2. Azinphos methyl 9

192 The erythrocyte cholinesterase outcome of the 1-year study provided the most protective result and an 193 advantage over the brain cholinesterase measurement in that it had been checked at 4 weeks after 194 baseline as well. This 4-week time point in males was used as the basis for the dose-response as it was 195 closer to the length of exposure expected in an occupational setting. However, the results from the 13196 week measurement produced a lower benchmark dose (at 10% effect size, 0.23 mg/kg/day compared 197 with the 13-week study’s 0.07 mg/kg/day), although the same NOAEL would be selected based on either 198 time point. 199 3.1.3. Emamectin benzoate 200 All dichotomous models were successfully fit to the dose-response of tremors in the 15-day study, 201 passing goodness of fit testing with satisfactorily low residuals (considered to be residuals less than 202 |2.0|). The AIC values were similar, with the lowest being the quantal-linear model. This model also was 203 the best fit after visual assessment of the curve. 204 3.1.4. Methoxyfenozide 205 The authors of the identified 2-week study noted symptoms but did not believe them to be treatment 206 related. However, dose-responsive patterns were found in the male treatment groups. The outcome for 207 which model fitting was successful was the three-month measurement of red blood cell count, in 208 contrast with platelet count, red blood cell count, hematocrit, and methemoglobin. All continuous 209 models showed satisfactory fit, and the exponential 4 model was selected based on a marginally lower 210 AIC. 211 3.1.5. Novaluron 212 With the exception of red blood cell counts and hematopoiesis, the dose-response for hematological 213 impacts was irregular and non-significant, resulting in poor model fits. The outcome of reduced blood 214 cell count was therefore used in the calculation of the benchmark dose. Models for both the red blood 215 cell count and spleen hematopoiesis passed the goodness of fit and variance tests and had comparable

10

216 AIC among them. Either endpoint could be used, but the dose-response is clearer in the RBC data. The 217 exponential 4 model of RBC was selected based on AIC and visual evaluation of the model fit. 218 3.1.6. Phosmet 219 Dose-response curves were constructed using both dermal and acute oral data, for plasma ChE in the 220 dermal study and red blood cell cholinesterase in the oral study, which proved to be the most sensitive 221 measures. For both the dermal and oral models, the Hill model provided the best fit and lowest AIC. The 222 model for the oral exposure was a better fit based on the results of the χ2 goodness-of-fit test as well as 223 providing a lower benchmark value, as expected. 224 3.1.7. Spinetoram 225 Since all animals in the control and lowest dose group were free of arteritis and all animals in the higher 226 dose groups developed it, the dose-responses for arteritis were of limited use, therefore the bone 227 marrow necrosis was used to derive the benchmark dose. All dose-response curves for necrosis were 228 similar overall with respect to AIC and residual values, but the logistic model offered the best fit 229 qualitatively for both the dose-response curve and its 95% confidence interval. 230 3.1.8. Thiacloprid 231 The liver enzyme induction impacts in general produced model fits which identified significant dose 232 responses and passed goodness-of-fit tests for the mean, but in many cases the dose-response was 233 inconsistent in the low-dose groups leading to a poorer model fit, particularly at the low doses. The 234 model which was selected was the log-logistic model of hepatocellular hypertrophy, which provided a 235 more consistent fit to the data at low doses (based on visual inspection of the curve), passed the 236 goodness-of-fit test, and had satisfactory residual values. 237 3.2 Benchmark doses and NOAELs in comparison to deterministic dose 238 The ratio of the 10% benchmark dose to the associated lower 95% confidence interval ranged from 1.02

11

239 to 5.76, with a mean of 3.3 (Table 2). The same ratio for the BMDNES ranged from 1.58-9.6, with a mean 240 of 3.6. Three of the benchmark doses of either type were lower than the NOAEL from the same study, 241 and five of the BMDL10 were lower than the NOAEL (Figure 2), 242 whereas all of the BMDLNES were lower than the NOAEL except phosmet which was similar but slightly 243 higher. The NOAELs for acetamiprid, novaluron, and phosmet were above the BMD for the same study, 244 indicating that in those cases the NOAEL is less sensitive than the BMD method. The NOAEL for azinphos 245 methyl, spinetoram, and emamectin benzoate fell below the BMDL, so that in those cases, the NOAEL 246 was more protective than the BMD. The BMD for emamectin benzoate falls above the range of the data 247 used to derive the model, and should therefore be interpreted with caution, as it represents an 248 extrapolation of the dose response curve which may or may not be supportable. In the remaining cases 249 of methoxyfenozide and thiacloprid, the NOAEL fell between the BMD and BMDL. The ratio of the 10% 250 BMD to the NOAEL ranged from 0.17 to 12.15 and averaged 2.8 (Table 3), and the BMDNES to NOAEL 251 ratio averaged 1.96 (ranging from 1.04 to 4.22). The Normalized Effect Sizes ranged from 20.75% to 252 30.86%, and so the associated BMDs were all higher than the 10% effect size BMDs, as were the BMDs 253 from the hybirid risk model. 254 The comparison of the NOAEL with the occupational handler doses used in the EPA risk assessments 255 using an MOE supports the findings of the EPA assessments (Table 4). The only ratio which falls below 256 the LOC, or level of concern (100 for all occupational doses except emamectin benzoate, which has an 257 LOC of 300), is that of azinphos methyl, as well as phosmet depending on the method of calculation and 258 data source. For the most part, the ratio of the BMD to the dose produces the same conclusion, except 259 in the case of acetamiprid. If the BMDL were used, azinphos methyl and acetamiprid would both 260 produce a ratio less than 100 and therefore of concern, but all other chemical exposures would still be 261 on average below the level of concern. The alternative of single standard deviation as the effect size for 262 the continuous effects found a higher benchmark dose in all cases except phosmet.

12

263 3.3 Probabilistic dose comparisons with BMD and NOAEL 264 Comparison of the probabilistic dose estimations with the various points of departure allows estimation 265 of an exceedance fraction, the proportion of the estimated potential doses which are above the 266 deterministic level of concern, and the margin of exposure (MOE) calculated by dividing the BMD 267 measures or the NOAEL by the dose. Table 4 shows this fraction for the NOAEL, BMD, and BMDL point 268 estimates divided by 100 (300 in the case of emamectin benzoate) to account for uncertainty factors, 269 and the MOE as well as the amount exceeding the level of concern (100 or 300 for emamectin 270 benzoate). Exceedance fractions and MOE varied, depending on chemical and point of departure 271 selected. Exceedance of the NOAEL ranged from 0.005 % in the case of methoxyfenozide to 72.2% for 272 azinphos methyl. Azinphos methyl doses had the highest fraction exceeding all of the points of 273 departure, and by extension, the highest fraction of MOE beyond the LOC. Exceedance fractions for 274 methoxyfenozide and spinetoram were increased by a minimum of 40 times by the addition of the 275 estimated dermal doses. The exceedance fractions for azinphos methyl and phosmet varied depending 276 on the choice of point of departure and the source of the dose-response data (oral vs dermal). The ratio 277 of the exceedance fractions of the BMDL and the BMD are a crude relative measure of the uncertainty in 278 the benchmark dose measurement. These ratios ranged from 1.1 to 8.4. By comparison, the ratio of the 279 exceedance fraction of the BMD to that of the NOAEL ranged from 0.1 to 5.0, and averaged at 1.2, 280 indicating the relative protective ability of the two points of departure. There was no correlation 281 between the two ratios. 282 1.

Discussion

283 This analysis shows that studies designed for the production of a NOAEL according to OECD guidelines 284 can be used to generate a dose-response curve and derive a benchmark dose and the associated 285 confidence interval, but that the existing standards could be built upon to improve the quality of data 286 for benchmark modeling. With the exception of azinphos methyl, there are not existing benchmark dose

13

287 models available for comparison of results. In the case of azinphos methyl, the 20% BMD and BMDL in 288 male rats found here (0.50 mg/kg/day and 0.35 mg/kg/day) are similar to the values reported by the 289 ATSDR benchmark analysis (0.48 and 0.30 mg/kg/day), which used the same data, although different 290 models were selected38. This analysis also showed that benchmark dose methods do not produce 291 inherently more (or less) conservative or protective dose limits than the NOAEL method, as the NOAEL 292 for many of the compounds was lower than the BMD, and in some cases less than the lower confidence 293 limit. In a few cases, the NOAEL is within the confidence limit of the BMD or BMDL, which may indicate 294 that they would produce indistinguishable results for regulation. However, the BMD and BMDL still 295 provide more information on the uncertainty of the outcome. 296 The relationship between the BMDSD, the BMDNES and the other points of departure is instructive as well, 297 as these alternative effect sizes represent the variability in continuous outcomes and therefore the 298 degree to which the NOAEL or BMD represents a change beyond that variability. The MOE results from 299 the BMD-SD of Methoxyfenozide, for example, is similar to the result of the BMD10, but in the case of 300 Phosmet, BMD-SD is closest to the NOAEL. In cases such as Azinphos methyl, Acetamiprid, and 301 Novaluron, the BMD-SD is higher than any other point of departure, which may suggest that the 10% 302 and 20% effects are within the outcome’s variability. The BMDNES effect size of 26% for azinphos methyl 303 and phosmet is close to the 20% value typically used as biologically relevant for cholinesterase 304 inhibitors. In all cases, however, the NES was higher than the standard effect sizes. This result might be 305 interpreted to suggest that the endpoints have too much variability to support use of a standard effect 306 size as low as 10%, but the NES is based on a scaling factor of 1/8, which is itself an arbitrary choice, 307 even though the overall method connects the effect size to the variation in the endpoint. A larger scaling 308 factor, for example 20, would produce an NES of closer to 10% in most cases. 309 The degree of uncertainty in the BMD estimates shown here varies, and in some instances health 310 outcomes were observed in the study which could not be used to build a dose-response model, either 14

311 due to infrequent observation, lack of significant dose-response, or, non-monotonic or inconsistent 312 dose-response. Although a successful dose-response model was fit in the case of each pesticide in this 313 study, it is likely that in other instances, particularly where the compound is of relatively low toxicity so 314 that responses will not be measured at lower doses, the OECD guidelines will not produce a study with 315 sufficient data to create a model, in which case the NOAEL could serve as a contingency method rather 316 than requiring an additional study. In several of the modeling attempts, some outcomes did not yield a 317 model fit and so were passed over in favor of others. This effect could bias the choice of outcome 318 towards those with less steep dose-responses, since the shorter curve would be less likely to show 319 impacts between the minimum and maximum at doses that are intermediate for endpoints with a 320 shallower curve. Quantal outcomes may be less likely to be selected for the same reason, especially 321 where group sizes are smaller. An increase in the standard required number of dose groups required 322 could help produce studies more amenable to a benchmark dose analysis and reduce the selection of 323 shallower dose-responses, as suggested by Slob in 200239. 324 It is important to recognize that a NOAEL is as likely as the BMD to be unreliable or impossible to 325 determine in cases where the dose-response is uncertain or variability in the controls is high40. The key 326 advantage of dose-response modeling and the benchmark dose, and a compelling reason to adopt these 327 methods as the status quo, is that some measure of the point of departure’s uncertainty is available and 328 expressed relatively simply as a confidence interval, whereas the uncertainty in estimation of a NOAEL is 329 potentially the same but left opaque if the number is taken at face value. The second advantage to the 330 BMD method in regulation is the incorporation of the entire dose response curve, which leaves the 331 outcome less vulnerable to biases introduced by the dose selection and study design. A further 332 advantage of the dose response method is also illustrated through the comparison of the results of the 333 cholinesterase inhibition models. The use of a continuous endpoint in the NOAEL paradigm may require 334 only a statistical difference in the outcome to determine the target value. No biological justification or 15

335 clinical significance is necessarily required. While this method may arguably be more sensitive to small 336 changes in a measure, it is not guaranteed to be more protective, and does not necessarily lead to a 337 result which is useful in risk management. Since the benchmark dose method requires that an effect size 338 be specified, the model is more flexible and as demonstrated here, can be scaled to the variability of the 339 endpoint of interest. In the case of azinphos methyl, use of 20% rather than 10% inhibition of 340 cholinesterase decreases the proportion of the workers predicted to receive doses over the level of 341 concern from 80% to 56%. However, for phosmet, the same decision changes the proportion from 14% 342 to 9%. The impact of this type of difference on a risk management decision is not clear, but the potential 343 for evaluating the sensitivity of the population’s level of concern to the effect size chosen has great 344 value in increasing the flexibility and transparency of risk management. Potentially, greater data 345 availability on the assessment of dichotomous outcomes and the effective critical effect size decision346 making they imply (as described by Slob and Pieter14) could allow for sensitivity analysis for the 347 designation of outcomes usually considered quantal, for example, cellular hypertrophy. 348 The potential dependence of the benchmark dose on the choice of critical effect size has been described 349 as a weakness of the method10. This study demonstrates that while effect size may have a large or small 350 impact on the benchmark dose, consistency of reporting the process and results of benchmark dose 351 modeling at these different possible effect sizes can provide transparency while quantifying uncertainty 352 in the data. Biological basis, natural variation in the endpoint for the animal model, and transparency in 353 the choice of critical effect size may be a more sound basis for developing a consistent BMD 354 methodology than choice of effect size as the basis for consistency, but these methods too may suffer 355 from arbitrariness in the selection of the cut-off value. 356 The methods used by the EPA to evaluate occupational exposures are based on single-day estimates for 357 non-carcinogenic outcomes, and therefore the dose estimates used in this analysis are also for one day. 358 However, in cases where doses may not clear from the body within 24 hours, or may have longer-term 16

359 sequelae, a longer period of exposure may be important even in occupational scenarios where the 360 assumption is that exposures are acute. The estimation of the dose over a longer period of time requires 361 additional data and modeling of the pharmacodynamics of the compound of interest as well as data on 362 the task patterns of applicators working for multiple days in a row using the same compound or 363 compounds with similar toxicological impacts, which likely is highly variable. 364 In summary, this analysis demonstrates the use of existing OECD guideline studies to build benchmark 365 dose models and derive points of departure for risk assessment. The use of the benchmark dose 366 compared to the NOAEL may or may not substantively impact the risk assessment outcome but is able 367 to provide a quantification of the uncertainty around the selected point of departure which is absent in 368 the reports of NOAELs from the same studies. Benchmark doses provide transparency and flexibility and 369 can be performed with existing study guidelines, despite room for improvement in study design. 370 Consistency in the process of modeling and reporting can provide the standardization necessary for the 371 adoption of these measures into standard operating procedures for official risk assessment. 372 5. Declaration of conflicts of interest 373 The authors report no conflicts of interest. Funding in support of this analysis was provided by the NIEHS 374 Environmental Toxicology and Pathology training grant and the CDC/NIOSH Cooperative agreement U54 375 OH007544.

17

376 377 378 379

Table 1: Study MRID and selected toxicological endpoints for each pesticide. RBC = red blood cell. The original studies which have been made public through the Freedom of Information Act request may be accessed through the EPA Office of Pesticide Programs Chemical Search at https://iaspub.epa.gov/apex/pesticides/f?p=chemicalsearch:1 Study MRID

Endpoint Used in BMD Model

Acetamiprid

4625561941

Changes in Perinatal (6 auditory startle week)

Azinphos methyl

100644 42

RBC Subchronic (13 Neurotoxicity Cholinesterase week)

Continuous Beagle

Emamectin benzoate

428515-04

Tremors

Acute (2 week) Neurotoxicity

Quantal

44

RBC count

Subchronic (10 Hemotoxicity week)

Continuous Beagle

Novaluron

4565150345

RBC count

Subchronic (13 Hemotoxicity week)

Continuous

Crl:CD(SD) BR rat

Mixed with food

Phosmet

4467330146

RBC Acute (one Cholinesterase dose)

Neurotoxicity

Continuous

Crl:CD(SD) IGS BR rat

Oral gavage

Phosmet (dermal)

4479680147

RBC Acute (3 week) Neurotoxicity Cholinesterase

Continuous

Crl:CD(SD) BR rat

Dermal (in water)

Spinetoram

4656850148

Bone marrow necrosis

Subchronic (13 Hemotoxicity week)

Quantal

Beagle

Thiacloprid

449277-15 Hepatocellular 49 hypertrophy

Subchronic inhalation (13 week)

Quantal

Hsd Cpb: WU (SPF) Rat

Pesticide

Methoxyfenozide

380 381

43

446177-28

Timing of Dose Class of Outcome

Type

Neurodevelopmental Continuous

Hepatotoxicity

Species Crl:CD(SD) IGS BR rat

Dose Route Oral gavage of the dams Mixed with food

Crl:CF-1 BR Mixed mouse with food Mixed with food

Mixed with food Directed flow noseonly aerosol

382 383 384 385

Table 2: Selected critical effect size, NOAEL from the investigated study, and Benchmark Dose in mg/kg/day with 95% Confidence limit for the critical effect size (CES) and alternate effect sizes (1 standard deviation for all continuous outcomes, and 10% inhibition for cholinesterase inhibitors all in mg/kg/day). The EPA-calculated dose in mg/kg/day for pesticide handlers using open cab airblast methods in pome fruit multiplied by two uncertainty factors of 10 is also compared*.

Pesticide Acetamiprid

1 SDNOAEL 10% 10% based (mg/kg/day) BMD BMDL BMD

1 SDbased BMDL

20% 20% BMD BMDL

Hybrid BMD

Hybrid BMDL

Normalized Effect Size (%)

NES BMD

NES BMDL

Maximum endpoint value

10

1.99

0.30

44.62

26.74

--

--

1.18

0.16

30.86

12.09

2.23

8.6

Azinphos methyl

0.15

0.23

0.17

4.42

3.51

0.50

0.35

0.75

0.4

25.99

0.633

0.469

0.09

Emamectin benzoate

0.075

0.91

0.19

--

--

--

--

--

--

--

--

--

Methoxyfenozide

16.8

27.69 9.56

28.93

10.13

--

--

83.11

10.16

20.75

36.21

5.84

4.52

Novaluron

4.38

2.00

0.90

7.66

2.92

--

--

21.28

2.34

25.44

6.24

2.75

6.13

Phosmet

4.5

2.75

0.58

4.37

3.97

4.60

4.53

7.36

5.89

25.99

4.68

4.62

188.20

Spinetoram

2.7

6.62

3.31

--

--

--

--

--

--

--

--

--

--

Thiacloprid

1.2

1.40

0.64

--

--

--

--

--

--

--

--

--

--

--

386 * The Maximum endpoint value represents the minimum point of departure in mg/kg/day which would, when combined with the deterministic 387 dose value used in the latest EPA exposure assessment, permit regulatory approval of the compound based on occupational risks. 388

389 390 391 392

Table 3: Ratios of points of departure and points of departure vs doseEPA, i.e., the EPA-derived deterministic doses used in the human health risk assessment for pesticide handlers. Two values are presented for phosmet: oral is calculated using the oral dosing study values compared with doses calculated using an adjustment factor for dermal absorption. The dermal study does not include adjustment for dermal absorption and uses the dermal toxicity testing data. BMD BMDL

BMDNES BMDLNES

NES

Acetamiprid

5.76

0.17

0.03

226

39

1302

7.44

1.21

Azinphos methyl

1.40

3.32

2.36

60

43

18

1.35

4.22

Emamectin benzoate

4.83

12.15

2.52

10171

2106

837

Methoxyfenozide

2.90

1.65

0.57

8073

2787

4898

2.99

0.84

Novaluron

2.22

0.46

0.21

249

112

545

2.27

1.42

Oral

1.02

1.53

1.01

153

151

92

1.01

1.04

Dermal

3.86

0.90

0.23

175

45

270

2.19

1.39

Spinetoram

2.00

2.45

1.23

23629

11820

9643

Thiacloprid

2.20

1.17

0.53

368

167

314

Phosmet

393 394

395 Table 4: Exceedance fractions of probabilistically-estimated doses for each compound and the associated points of departure estimated from the 396 dose-response studies, and the ratio of these exceedance fractions, demonstrating the difference in protective ability of each selected point of 397 departure as a regulatory limit Exceedance fractions

Acetamiprid Azinphos methyl Emamectin benzoate Methoxyfenozide Methoxyfenozide (dermal + inhalation routes) Novaluron Phosmet (oral) Phosmet (dermal)

Ratio of Exceedance Fractions

NOAEL

BMD

BMDL

BMDL/ BMD

BMD/ NOAEL

5.0 72.2 19.6

24.8 55.7 2.0

61.2 63.5 9.7

2.5 1.1 4.8

5.0 0.8 0.1

0.005

0.002

0.015

8.4

0.4

2.7

1.5

4.7

3.1

0.6

5.09 8.94 12.62

9.3 13.5 13.8

15.7 36.9 34.9

1.7 2.7 2.5

1.8 1.5 1.1

Margin of Exposure: mean (% over LOC)

NOAEL

BMD

BMDL

SD-BMD

1502983 (5) 28352 (25) 343 (62) 51011 (0.7) 432 (54) 446 (44) 330 (60) 8571 (5) 53581 (20) 650122 (2) 135740 (10) 16573165 27316127 9430920 28539384 (0.002) (0.0006) (0.006) (0.0005) 19248 (0.6) 31725 (0.25) 10953 (2) 33146 (0.23) 36101 (1) 12945 (4) 13620 (2) 8542041 (0.53) 45282 (22)

16485 (3) 111025 (15) 8323 (3) 20943820 (0.15) 1283 (4)

7418 (8) 63136 (0.5) 1668 (26) 12571 (4) 1756 (20) 13227 (1.7) 10471910 (0.4) 55512 (20) -

0.6 0.2 0.4 2.7 0.3 Spinetoram Spinetoram (dermal + inhalation 22.6 15.3 20.8 1.4 0.7 routes) 138 (88) 161 (87) 74 (94) 32.0 30.1 40.2 1.3 0.9 Thiacloprid 398 *Exceedance fractions are the proportion of the dose distribution above the deterministic point of comparison of the NOAEL, BMD, or BMDL. In 399 the MOE, exceedance fraction is the proportion of the distribution below the LOC (100 for all cases but Emamectin Benzoate, which was 300). 400

401 402 403 404 405

Figure 1: Graphs of dose-response modls for the selected outcome for each pesticide. The dashed line represents the benchmark dose and confidence interval associated with an alternative critical effect size of 1 standard deviation from the control. The solid line benchmark dose is associated with the selected critical effect size listed in table 2 A) Hill model for acetamiprid-induced decreased maximum amplitude of auditory startle B) Exponential model of erythrocyte acetylcholinesterase activity and oral dose of azinphos methyl C) Quantal-linear model of

406 407 408 409

emamectin benzoate-induced tremors D) Exponential model of decreased red blood cell count associated with methoxyfenozide dosing E) Exponential model of decreased red blood cell count associated with Novaluron dosing F) Hill model of erythrocyte acetylcholinesterase activity and oral dose of phosmet G) Logistic model of fraction of population with bone-marrow necrosis induced with spinetoram dosing H) Log-logistic model of hepatocellular hypertrophy associated with thiacloprid dosing

410

411 412 413 414 415 416 417 418

Figure 2: NOAEL, EPA-calculated daily dose x 100 of active ingredient to a mixer/loader/applicator in pome fruit using open cab application as published in the EPA HHR risk assessments, and BMDL-BMD range for selected critical effect size, and alternative effect size for continuous endpoints. Selected effect size was 10% for all quantal impacts, 10% for all continuous except azinphos methyl and phosmet, and 20% for azinphos methyl and phosmet (cholinesterase inhibitors). The alternate effect size is 1 standard deviation from baseline. The dose x 100 represents the EPA-estimated human dose combined with the uncertainty factors used to adjust the points of departure from animal studies for comparison. The dose x 100 therefore represents the minimum toxicological POD that would produce an acceptable occupational margin of exposure under the current regulatory system in the United States, based on the doses predicted by the EPA to occur in one day of work.

419

References

420 1. Filipsson AF, Sand S, Nilsson J, Victorin K. The benchmark dose method--review of available models, 421 and recommendations for application in health risk assessment. Crit Rev Toxicol 2003;33(5):505-42. 422 2. Gaylor D, Ryan L, Krewski D, Zhu Y. Procedures for calculating benchmark doses for health risk 423 assessment. Regulatory Toxicology and Pharmacology 1998;28(2):150 164. 424 3. Davis JA, Gift JS, Zhao QJ. Introduction to benchmark dose methods and U.S. EPA's benchmark dose 425 software (BMDS) version 2.1.1. Toxicol Appl Pharmacol 2011 Jul 15;254(2):181-91. 426 4. Crump K. A new method for determining allowable daily intakes*1. Fundamental and Applied 427 Toxicology 1984;4(5):854, 871. 428 5. U.S. EPA. Benchmark dose technical guidance. Washington, D.C.: U. S. Environmental Protection 429 Agency; 2012 June 2012. Report nr EPA/100/R-12/001. 430 6. EFSA. Guidance of the scientific committee on use of the benchmark dose approach in risk 431 assessment. EFSA Journal 2017 24 January, 2017;15(1):4658. 432 7. U.S. EPA. Revised risk assessment methods for workers, children of workers in agricultural fields, and 433 pesticides with no food uses. Washington, D.C.20460: Office of Pesticide Programs U.S. 434 Environmental Protection Agency; 2009 December 7, 2014. Report nr EPA-HQ-OPP-2009-0889435 0002. 436 8. OECD. Organisation for economic cooperation and development guidelines for the testing of 437 chemicals, section 4. 2018. 438 9. OECD. Test no. 424: Neurotoxicity study in rodents 439 . In: OECD guidelines for the testing of chemicals, section 4. Paris: OECD Publishing; 1997. . 440 10. Travis KZ, Pate I, Welsh ZK. The role of the benchmark dose in a regulatory context. Regul Toxicol 441 Pharmacol 2005 Dec;43(3):280-91. 442 11. Oberg M. Benchmark dose approaches in chemical health risk assessment in relation to number and 443 distress of laboratory animals. Regul Toxicol Pharmacol 2010 December 2010;58(3):451. 444 12. Benford D, Halldorsson T, Hardy A, Jeger MJ, Knutsen KH, More S, Mortensen A, Naegeli H, Noteborn 445 H, Ockleford C, et al. Update: Use of the benchmark dose approach in risk assessment. EFSA Journal 446 2017 24 January 2017;15(1). 447 13. Allen B. Dose-response assessment for developmental toxicity II. comparison of generic benchmark 448 dose estimates with no observed adverse effect levels. Fundamental and Applied Toxicology 449 1994;23(4):487 495.

450 14. Slob W, Pieters MN. A probabilistic approach for deriving acceptable human intake limits and human 451 health risks from toxicological studies: General framework. Risk Analysis 1998;18(6):787 798. 452 15. Wignall JA, Shapiro AJ, Wright FA, Woodruff TJ, Chiu WA, Guyton KZ, Rusyn I. Standardizing 453 benchmark dose calculations to improve science-based decisions in human health assessments. 454 Environ Health Perspect 2014 May;122(5):499-505. 455 16. Zeller A, Duran-Pacheco G, Guerard M. An appraisal of critical effect sizes for the benchmark dose 456 approach to assess dose-response relationships in genetic toxicology. . Arch Toxicol 2017 Dec 457 2017;91(12):3799-807. 458 17. Slob W. A general theory of effect size, and its consequences for defining the benchmark response 459 (BMR) for continuous endpoints. Crit Rev Toxicol 2017 Apr 2017;47(4):342-51. 460 18. OECD. WHO OECD ILSI/HESI international workshop on risk assessment of combined exposures to 461 multiple chemicals. Paris: OECD; 2011 February 15, 2011. Report nr Series on Testing & Assessment 462 No. 140. 463 19. Guidance for Applying Quantitative Data to Develop Data-Derived Extrapolation Factors for 464 Interspecies and Intraspecies Extrapolation [Internet]: U.S. EPA; c2015 [cited 2016 6/20/2016]. Available from: https://www.epa.gov/risk/guidance-applying-quantitative-data-develop-data465 466 derived-extrapolation-factors-interspecies-and. 467 20. US EPA. Emamectin benzoate human health assessment scoping document in support of registration 468 review. Office of Chemical Safety and Pollution Prevention: US Environmental Protection Agency; 469 2011 June 16 2011:Available at: https://www.regulations.gov/document?D=EPA-HQ-OPP-2011470 0483-0004. 471 21. US EPA. Human health risk assessment for the section 3 registration of acetamiprid on cotton, leafy 472 vegetables, brassica vegetables, fruiting vegetables, citrus, pome fruits, grapes, and canola and 473 mustard seed. US Environmental Protection Agency; 2005. Report nr EPA-HQ-OPP-2005-0190474 0011:Available at: https://www.regulations.gov/document?D=EPA-HQ-OPP-2005-0190-0011. 475 22. US EPA. Revised occupational exposure and risk assessment for azinphos methyl (reflecting 476 recommendations from the human studies review board). US Environmental Protection Agency 477 Office of Prevention, Pesticides, and Toxic Substances; 2006. Report nr EPA-HQ-OPP-2005-0061478 0136. 479 23. US EPA. Novaluron: Human health risk assessment for proposed section 3 uses on bushberry crop 480 subgroup 13-07B; brassica, leafy greens, crop subgroup 5B; turnip, greens; and fruit, stone crop 481 group 12. Office of Prevention, Pesticides, and Toxic Substances: US Environmental Protection 482 Agency; 2009 Sept 9, 2009. 483 24. US EPA. Methoxyfenozide human health risk assessment to support proposed new uses on herbs, 484 caneberries, dates and sorghum; to establish rotational crop tolerances in the rapeseed and 485 sunflower oilseed subgroups; as well as to extend and update crop group tolerances on multiple

486 487

commodities. Office of Chemical Safety and Pollution Protection: US Environmental Protection Agency; 2013 August 6, 2013.

488 25. US EPA. Phosmet (chemical ID 059201). HED human health risk assessment and supporting 489 documentation for the reregistration eligibility decision document (RED). Office of Prevention, 490 Pesticides, and Toxic Substances: US Environmental Protection Agency; 1998 October 30, 1998. 491 26. US EPA. Pyriproxyfen: Human health risk assessment for the request to add uses on herb subgroup 492 19A, and the expansions of existing crop group uses to numerous crop subgroups. Office of 493 Chemical Safety and Pollution Prevention: US Environmental Protection Agency; 2011 September 8, 494 2011. 495 27. U.S. EPA. Spinosad and spinetoram: Human-health assessment scoping document in support of 496 registration review. US Environmental Protection Agency 2011 August 18, 2011. 497 28. US EPA. Thiacloprid in/on pome fruits and cotton. health effects division risk assessment. US 498 Environmental Protection Agency Office of Prevention, Pesticides, and Toxic Substances 2003 July 499 23, 2003. 500 29. Pouzou JG, Cullen AC, Yost MG, Kissel JC, Fenske RA. Comparative probabilistic assessment of 501 occupational pesticide exposures based on regulatory assessments: Comparative probabilistic 502 assessment of occupational pesticide exposures. Risk Analysis 2017 November 8, 2017. 503 30. AHETF. Agricultural handlers exposure task force (AHETF) volume IV: Standard operating 504 procedures. In: ; April 7, 2008. . 505 31. US EPA. U.S. EPA. exposure factors handbook 2011 edition (final). Washington, DC: US 506 Environmental Protection Agency Office of Research and Development; 2011. Report nr 507 EPA/600/R-09/052F. 508 32. U.S. EPA. Development of statistical distributions or ranges of standard factors used in exposure 509 assessments. Exposure Assessment Group, Office of Health and Environmental Assessment, US 510 Environmental Protection Agency 1985 August, 1985;PB85-242667. 511 33. U.S. EPA. Pyriproxyfen human health risk assessment for the request to add uses on herb subgroup 512 A, and the expansions of exisiting crop group uses to numerous crop subgroups. Office of Chemical 513 Safety and Pollution Prevention; 2012 August 22, 2012. 514 34. Dekkers S, de Heer C, Rennen MAJ. Critical effect sizes in toxicological risk assessment: A 515 comprehensive and critical evaluation. Environ Toxicol Pharmacol 2001;10(1-2):33 52. 516 35. Buist HE, von Bölcsházy G, Dammann M, .Telman J, Rennen MAJ. Derivation of the minimal 517 magnitude of the critical effect size for continuous toxicological parameters from within-animal 518 variation in control group data. Reg Toxicol Pharmacol 2009 Nov 2009;55(2):139-50. 519 36. US EPA. Benchmark dose technical guidance. Risk Assessment Forum U.S. Environmental Protection 520 Agency; 2012 June 2012. Report nr EPA/100/R-12/001.

521 37. Frome EL, Frome DP. STAND: Statistical analysis of non-detects. R package version 2.0. http://CRAN.R-project.org/package=STAND. 2015. 522 523 38. ATSDR. Toxicological profile for guthion. U.S. department of health and human services, public 524 health service, agency for toxic substances and disease registry. 2008 September 2008;tp188. 525 39. Slob W. Dose-response modeling of continuous endpoints. Toxicol Sci 2002 Apr;66(2):298-312. 526 40. WHO Task Group on Environmental Health Criteria on Principles for Modelling Dose-Response for 527 the Risk Assessment of Chemicals., United Nations Environment Programme., International Labour 528 Organisation.,Inter-Organization Programme for the Sound Management of Chemicals.,. Principles 529 for modelling dose-response for the risk assessment of chemicals. Geneva, Switzerland: World 530 Health Organization; 2009. ID: 264018462. 531 41. Nemec M. Acetamiprid: An oral developmental neurotoxicity study in rats. laboratory project ID WIL532 21193 . Ashland, Ohio: WIL Research Laboratories, Inc.,; 2003 November 21, 2003. Report nr 533 462556-19. 534 42. Allen TR. 52-week oral toxicity (feeding) study with azinphos-methyl (E 1582) in the dog. Itingen, 535 Switzerland: Mobay Corporation (Bayer AG); 1990 May 31, 1990. Report nr 100644. 536 43. Gerson RJ. L-660,599: Fifteen day dietary neurotoxicity study in CF-1 mice. MRID 428515-03. Merck 537 & Co.; 1993 April 2, 1993. Report nr 92-049-0. 538 44. Morrison RD, Shuey DL. RH-2485 technical: One-year dietary toxicity study in dogs. Report 94R-257. 539 Spring House, PA: Rohm and Haas Company; 1997 May 21, 1997. Report nr 446177-28. 540 45. Kirk SJ. GR 572 (technical) toxicity to rats by dietary administration for 13 weeks. Lab. project no.: 541 AGR 50/90386. Cambridgeshire, England: Makhteshim Chemical Works Ltd.; 1990 July 2, 1990. 542 Report nr 45615-03. 543 46. Cappon GD. An acute neurotoxicity study of phosmet in rats. WIL-331004. Yuma, AZ: Gowan 544 Company; 1998 October 8, 1998. Report nr 44673301. 545 47. Hilaski RJ. A 21-day dermal toxicity study of imidan in rats. Mattawan, MI: Gowan Company; 1999 546 March 18, 1999. Report nr 447958-01. 547 48. Stebbins KE, Brooks KJ. XDE-175: 90-day dietary toxicity study in beagle dogs. Midland, MI: Dow 548 AgroSciences, LLC; 2005 May 31, 2005. Report nr 465685-01. 549 49. Pauluhn J. YRC 2894: Subacute inhalation toxicity on rats (exposure 5 x 6 hours/week for 4 weeks). 550 Wuppertal, Germany: Bayer AG; 1998 June 17, 1998. Report nr 449211-15. 551 552 Supplemental Materials:

553 Table A: Mean and confidence intervals for parameters generated for the dose-response models 554 selected for each pesticide (the oral dosing models are presented here for phosmet and azinphos 555 methyl) Pesticide

Acetamiprid

Azinphos methyl

Emamectin benzoate

Methoxyfenozide

Novaluron

Phosmet

Spinetoram

Thiacloprid

Model

Variable intercept maximum change Hill power dose w/ half-max change background response Exponential slope asymptote parameter background Quantal-linear slope background response Exponential slope asymptote parameter background response Exponential slope asymptote parameter intercept maximum change Hill power dose w/ half-max change intercept Logistic slope background Log-logistic intercept slope

Mean 214.1 -101.6 1 6.5 2.0 0.9 0.1 0 0.1 6.5 0.02 0.8 7.5 0.1 0.9 3730.7 -2805.2 18 4.87 -2.72 0.16 0 -2.54 1

95% CL 163.8 264.4 -168.9 -34.3 ---8.2 21.2 1.8 2.3 0.3 1.4 -0.01 0.2 ---0.2 0.4 6.3 6.8 -0.002 0.04 0.8 0.9 7.3 7.7 -0.03 0.2 0.8 1.0 3624.0 3837.3 -2944.4 -2666.1 --4.8 5.0 -5.0 -0.4 0.01 0.3 ---3.5 -1.6 ---

556 Supplemental Table B: Fit results and information for continuous endpoints

Compound

Model

Specified Effect Risk Type

Hill Exponential2 Exponential3 Exponential4 Acetamiprid

0.1 Exponential5

Relative deviation

p-value Test 4 AIC

Scaled residual for control group

BMD

BMDL

p-value Test 1

p-value Test 2

p-value Test 3

1.99

0.301

0.082

0.322

0.873

0.960

389.3

-0.129

0.008

9.965

5.860

0.082

0.322

0.873

0.341

389.4

-0.724

0.786

9.965

5.860

0.082

0.322

0.873

0.341

389.4

-0.724

0.786

2.363

0.575

0.082

0.322

0.873

0.804

389.3

-0.260

0.102

2.363

0.575

0.082

0.322

0.873

0.804

389.3

-0.260

0.102

Linear

12.289

8.347

0.082

0.322

0.873

0.291

389.8

-0.786

0.871

Polynomial

12.289

8.347

0.082

0.322

0.873

0.291

389.8

-0.786

0.871

Power

12.289

8.347

0.082

0.322

0.873

0.291

389.8

-0.786

0.871

Hill

0.674537

0.273635

<.0001

0.4671

0.3746

NA

-53.6

-0.000423

1.17

Exponential2

0.485982

0.38887

< 0.0001

0.4671

0.3746

0.8392

-60.4

-0.01687

0.3692

Exponential3

59.2109

< 0.0001

0.4671

0.3746

< 0.0001

-11.1

-3.198

1.824

Exponential4

0.468416

0.345456

< 0.0001

0.4671

0.3746

0.5882

-58.5

0.1456

0.2833

Azinphos methyl

0.2 Exponential5

Relative deviation

0.468416

0.345456

< 0.0001

0.4671

0.3746

0.5882

-58.5

0.1456

0.2833

Linear

0.926849

0.808682

<.0001

0.4671

0.3746

0.03643

-54.2

-1.74

1.4

Polynomial

0.926849

0.808682

<.0001

0.4671

0.3746

0.03643

-54.2

-1.74

1.4

Power

0.926849

0.808682

Hill

87.819

<.0001

0.4671

0.3746

0.03643

-54.2

-1.74

1.4

0.001306

0.3105

0.1972

0.1378

-10.01

-0.0155

0.353

Exponential2

796.353

442.99

0.001306

0.3105

0.1972

0.005095

-3.41

0.1533

1.024

Exponential3

796.352

442.99

0.001306

0.3105

0.1972

0.005095

-3.41

0.1533

1.024

28.9341

10.1306

0.001306

0.3105

0.1972

0.1649

-10.6

1.303

0.05501

38.1091

10.1441

0.001306

0.3105

0.1972

0.1356

-9.98

0.7769

0.3532

Linear

1005.03

658.856

0.001306

0.3105

0.1972

0.004971

-3.36

0.138

1.03

Polynomial

1005.03

658.856

0.001306

0.3105

0.1972

0.004971

-3.36

0.138

1.03

1005.03

658.856

0.001306

0.3105

0.1972

0.004971

-3.36

0.138

1.03

0.001483

0.3181

0.2707

0.4768

-41.88

-0.201

-0.107

Exponential4 Methoxyfenozide

0.1 Exponential5

Relative deviation

Power Novaluron

Scaled residual for dose group nearest the BMD

Hill

0.1

Relative

44.2534

deviation

Exponential2

36.6296

26.3435

0.001483

0.3181

0.2707

0.4619

-43.81

0.5171

0.7673

Exponential3

36.6296

26.3435

0.001483

0.3181

0.2707

0.4619

-43.81

0.5171

0.7673

Exponential4

44.9443

17.1943

0.001483

0.3181

0.2707

0.7553

-43.82

-0.1604

-0.1007

Exponential5

44.9443

15.7542

0.001483

0.3181

0.2707

0.7553

-43.82

-0.1604

-0.1007

Linear

36.8442

26.9298

0.001483

0.3181

0.2707

0.4315

-43.63

0.514

0.82

Polynomial

36.8442

26.9298

0.001483

0.3181

0.2707

0.4315

-43.63

0.514

0.82

Power

36.8442

26.9298

0.001483

0.3181

0.2707

0.4315

-43.63

0.514

0.82

Hill

4.59891

4.52582

<.0001

<.0001

<.0001

0.5139

1544.7

-0.23

-0.451

Exponential2

3.9221

3.58418

< 0.0001

< 0.0001

< 0.0001

< 0.0001

1599.0

0.6277

-3.787

Exponential3

6.67631

5.60923

< 0.0001

< 0.0001

< 0.0001

< 0.0001

1569.7

-2.38

-1.046

3.9221

3.58418

< 0.0001

< 0.0001

< 0.0001

< 0.0001

1599.0

0.6277

-3.787

Exponential4 Phosmet

0.2 Exponential5

Relative deviation

4.5755

4.51643

< 0.0001

< 0.0001

< 0.0001

N/A

1553.0

-0.0005

-0.6

Linear

5.92562

5.77355

<.0001

<.0001

<.0001

<.0001

1561.5

-1.49

-1.98

Polynomial

7.76135

5.86397

<.0001

<.0001

<.0001

<.0001

1561.2

-2.46

-0.86

Power

7.60666

6.04661

<.0001

<.0001

<.0001

<.0001

1560.2

-2.48

-0.733

Hill

37.3124

0.1147

0.7014

0.7162

NA

-68.08

0.0294

-0.106

Exponential2

57.1192

35.7982

0.1147

0.7014

0.7162

0.5937

-71.03

0.2606

0.3879

Exponential3

57.1193

35.7982

0.1147

0.7014

0.7162

0.5937

-71.03

0.2606

0.3879

45.5845

15.4466

0.1147

0.7014

0.7162

0.6557

-69.88

-0.04784

-0.1582

Exponential4 Phosmet (dermal)

0.2 Exponential5

Relative deviation

32.769

16.0394

0.1147

0.7014

0.7162

N/A

-68.08

0.02941

-0.1061

Linear

58.7912

39.1498

0.1147

0.7014

0.7162

0.5409

-70.8534

0.242

0.478

Polynomial

58.7913

39.1498

0.1147

0.7014

0.7162

0.5409

-70.8534

0.242

0.478

Power

58.7913

39.1498

0.1147

0.7014

0.7162

0.5409

-70.8534

0.242

0.478

557 Supplemental Table C: Fit results and information for quantal endpoints

Compound

Model

Chisquare

DF

Specified Effect

Gamma

0.47 3

0.1

Logistic

0.73 3

0.1

LogLogistic

0.47 3

0.1

0.92 3

0.1

Probit

0.72 3

0.1

Weibull QuantalLinear

0.47 3

0.1

0.45 4

0.1

Gamma

6.91 3

0.1

Logistic

9.86 2

0.1

2.34 3

0.1

LogProbit

9.49 3

0.1

Probit

9.83 2

0.1

Emamectin benzoate LogProbit

Spinetoram

Specified Risk Type Effect

LogLogistic

0.1

0.1

Relative deviation

Relative deviation

Risk Type

Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk

Scaled Scaled residual residual for dose for group control nearest group the BMD

BMD

BMDL P-value

0.815361

0.185655

0.925

9.89146

-0.094

0

0.464364

0.236773

0.8653

10.0738

-0.077

-0.204

0.834977

0.184284

0.9255

9.89084

-0.093

0

0.536346

0.179528

0.82

10.1647

-0.113

-0.238

0.488477

0.225902

0.868

10.0581

-0.091

-0.193

0.818901

0.186618

0.9253

9.89166

-0.092

0

0.911319

0.188685

0.978

7.89297

-0.071

0

2.99629

1.84813

0.0747

33.9807

-0.499

0

10.7616

6.68818

0.0072

41.164

2.564

-1.255

1.40433

0.638499

0.5048

30.8165

-0.744

0

3.66405

2.2077

0.0235

34.9573

1.883

0

10.1096

6.6502

0.0073

40.9651

2.588

-1.221

AIC

Thiacloprid

558

Weibull QuantalLinear

6.91 3

0.1

6.91 3

0.1

Gamma

6.91 3

0.1

Logistic

9.86 2

0.1

LogLogistic

2.34 3

0.1

9.49 3

0.1

Probit

9.83 2

0.1

Weibull QuantalLinear

6.91 3

0.1

6.91 3

0.1

LogProbit

0.1

Relative deviation

Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk Extra risk

2.99629

1.84813

0.0747

33.9807

-0.499

0

2.99629

1.84813

0.0747

33.9807

-0.499

0

2.99629

1.84813

0.0747

33.9807

-0.499

0

10.7616

6.68818

0.0072

41.164

2.564

-1.255

1.40433

0.638499

0.5048

30.8165

-0.744

0

3.66405

2.2077

0.0235

34.9573

1.883

0

10.1096

6.6502

0.0073

40.9651

2.588

-1.221

2.99629

1.84813

0.0747

33.9807

-0.499

0

2.99629

1.84813

0.0747

33.9807

-0.499

0

Highlights 1) Benchmark doses were estimated for eight pesticides using toxicological studies completed for registration. 2) The Benchmark doses/confidence limits were compared with the NOAEL as the regulatory POD; none were consistently lowest. 3) Occupational doses were compared with each POD as MOEs with exceedance fractions to compare the protection provided. 4) Current guidelines may produce BMDs as well as NOAELs, with the advantage of quantifying the uncertainty about the POD.

Funding Body Information

The authors report no conflicts of interest. Funding in support of this analysis was provided by the NIEHS Environmental Toxicology and Pathology training grant T32ES007032 and the CDC/NIOSH Cooperative agreement U54 OH007544. The funding organizations had no influence on the planning, data collection, analysis, or manuscript preparation.

Use of benchmark dose models in risk assessment for occupational handlers of eight pesticides used in pome fruit production. Corresponding Author: Jane Gurnick Pouzou Corresponding Author's Institution: EpiX Analytics, LLC Order of Authors: Jane Gurnick Pouzou1; John Kissel2; Michael G Yost2; Richard A Fenske2; Alison C Cullen2 1 2

EpiX Analytics, LLC University of Washington

Declaration of interests ☐ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. X The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Dr. Pouzou reports previous consulting engagements with National Cattlemen’s Beef Association and Tillotts Pharmaceuticals which were conducted after this study but prior to this publication, and unrelated to the study. Dr. Pouzou, Dr. Yost, and Dr. Fenske were supported from a grant from NIOSH/CDC cooperative agreement U54 OH007544 during the conduct of the study.