MDH’s TK maternal/infant model approach for deriving HBGVs

MDH developed an Excel-based TK model to predict serum levels from birth through adulthood. MDH chose to develop its model in Excel to maximize the transparency and accessibility of the model. In addition, the relationship between intake (dose) and serum concentration can adequately be described by a single-compartment model [13]. This type of model has been used by others to describe the relationship between dose and serum levels [14]. Two exposure scenarios were evaluated (Fig. 2): (1) an infant fed with formula reconstituted with contaminated water starting at birth, followed by a lifetime of drinking contaminated water; and (2) an infant breastfed for 12 months, followed by a lifetime of drinking contaminated water. In both scenarios, infants began life with a transgenerational body burden calculated from the mother’s serum concentration using a placental transfer factor. Exposure was simulated through consumption of breastmilk or formula reconstituted with contaminated water. Daily intake, elimination, and serum concentrations were calculated over a simulated period of 20,000 days (about 55 years) to ensure attainment of steady state (See Table 1).

Conceptual representation of the toxicokinetic model for the two exposure scenarios evaluated

Exposure and chemical-specific toxicokinetic parameters used in modeling PFOA serum concentrations

Mean paired ratios ranging from 0.68 [38] to 1.26 [4] have been reported. A more comprehensive literature review has been conducted since MDH selected 0.87 during its expedited review [16]. See Supplemental Table S2. Preference was given to paired ratios over ratios based on summary statistics.

Maximum ratios for individual mother–infant pairs range from 1.52 [7] to 2.16 [6]. 95th percentile values (mean + 2SD) were calculated from author-reported mean and SD values, and range from 1.11 [7] to 1.69 [4].

Mean paired ratios ranging from 0.025 [39] to 0.12 [40] have been reported. A more comprehensive literature review has been conducted since MDH selected 0.052 during its expedited review [16]. See Supplemental Table S3.

Maximum individual pair ratios were not reported by study authors. In the absence of maximum individual values, the maximum mean value of 0.12 is used to represent an upper percentile value.

Values for exclusively^a breastfed infants (Table 15-1 [36]). Body weight at birth was set at 3.38 kg, the mean birth weight for singleton births at 37–41 weeks of gestation [41]. Body weights (kg) were calculated from data presented in Table 15-1 for each age group (i.e., mL/day ÷ mL/kg per day):

The midpoint in time for each age group was set equal to age group value. Daily intake rates and body weights between one midpoint and the next were calculated by linear interpolation to avoid abrupt changes in values.

American Academy of Pediatrics [42] recommends exclusively breastfeeding for the first 6 months, with continued breastfeeding alongside introduction of complementary food for at least 12 months. The Center for Disease Control (CDC) Breastfeeding Report Card for 2016 [43] reports nearly 66% of mothers in Minnesota report breastfeeding at 6 months, with 31.4% exclusively breastfeeding. At 12 months, 41% of mothers reported breastfeeding.

Central tendency: exclusively^a breastfed intake rates used from birth to 6 months of age. From 6 to 12 months, breastfeeding is phased out and water intake is phased in. Upper percentile: exclusively breastfed intake rates used from birth to 12 months of age. At 12 months, breastfeeding ends and water intake begins.

Values for consumers only. (Table 3-1 [36]). Body weights (kg) were calculated from data presented in Table 3-1 for each age group (i.e., mL/day ÷ mL/kg per day):

The midpoint in time for each age group was set equal to age group value. Daily intake rates and body weights between one midpoint and the next were calculated by linear interpolation to avoid abrupt changes in values.

For calculation of maternal serum concentration at time of delivery, a time-weighted average water intake rate was calculated from birth to 30–35 years of age, resulting in a mean and 95th percentile water intake rate of 18 and 47 mL/kg per day, respectively.

Friis-Hansen [15] (and consistent with Felter et al. [45]). Early life stages have higher body water content per unit weight than adults. Adjustment factor is designed to account for this difference. This is an area of uncertainty since the precise nature of the V_d is not known. Use of the V_d AF reduces serum concentration estimates, and increases model accuracy compared with empirical data.

The midpoint in time for each age group was set equal to age group value. Daily V_d AF between one midpoint and the next were calculated by linear interpolation to avoid abrupt changes in values.

Final model parameters for calculation of the PFOA HBGVs shown in bold

^aExclusively breastfed as defined by USEPA [36] refers to infants whose sole source of milk is breastmilk and not formula. Exclusively breastfed infants in the studies underlying these USEPA estimates were not excluded from other foods, typically after six months. This definition differs from other sources, which may define exclusive breastfeeding as the only source of nourishment (solid or liquid) as breastmilk

Because PFOA is well absorbed and not metabolized, the dynamic relationship between serum concentrations and intake (dose) can be calculated using Eq. 1:

Where:

for water ingestion—

Dosemgkg⋅day= WaterintakerateLkg⋅day×WaterconcentrationmgL

for breastmilk—

Dosemgkg⋅day= BreastmilkintakerateLkg⋅day×BreastmilkconcentrationmgL

and

ClearancerateLkg⋅day=Vd×k

Vd=VolumeofdistributionLkg

An annotated list of model exposure and chemical parameter values is presented in Table 1.

The model assumes that maternal exposure began prior to pregnancy, so that steady-state serum concentration was achieved by the time of delivery. The infant’s serum concentration at birth was calculated using Eq. 2:

For all subsequent days, the infant’s final daily post-elimination serum concentration was calculated using Eq. 3:

The V_d parameter, assumed to be extracellular water, is both chemical specific and age specific. In order to account for age-specific differences in extracellular water volume during early childhood, V_d was multiplied by an adjustment factor (AF) starting at 2.1 at birth and declining to 1.0 by 10 years of age [15].

To maintain mass balance, daily maternal serum concentrations incorporated loss of chemical via transfer to the infant during breastfeeding, as well as excretion represented by the clearance rate. The infant’s daily intake (and thus the mother’s loss) was calculated from the breastmilk intake rate and the breastmilk concentration, derived using Eq. 4: