Draft Assessment of Bisphenol A for Use In Food Contact Applications by dla17169

VIEWS: 8 PAGES: 25

More Info
									  Draft Assessment of
 Bisphenol A for Use In
Food Contact Applications
         Update
             FDA Science Board
             August 17th, 2009

        Mitchell Cheeseman, Ph.D.
      U.S. Food and Drug Administration
  Center for Food Safety and Applied Nutrition
    Science Board Comments on the
           Draft Assessment
•   Expand/Update exposure assessment
•   Consider dose modeling for point of departure
•   Clarify uncertainty factors used
•   Reevaluate toxicity studies deemed “adequate”
•   Consider the value of meta-analysis
•   Expand discussion of biomonitoring data
•   Carefully consider planned toxicological studies
    to ensure they may clarify remaining
    uncertainties
Toxicology Assessment
       Update
        Biomonitoring Data
• Comment: lack of evaluation of
  biomonitoring data (towards cumulative
  exposure).
• Response: completed evaluation of
  biomonitoring data.
          Biomonitoring Review
• Analysis and comparison of analytical techniques
   – ELISA
   – GC with MS
   – HPLC
       • HPLC/FLD, HPLC/ECD, LC/MS
• Critique of variability and uncertainty in analysis
• Review includes recently published data
• Results
   – Mean daily intake for adults is < 1.6 – 7.7 µg/day
   – Mean daily intake for children (6-12 y) is 2.4 – 4.32 µg/day
• Next step: compare to updated dietary concentration
  values.
   Ongoing Low Dose Studies
          Evaluation
• Comment - criteria used in evaluating
  toxicology studies were not clearly stated.
• FDA reassessment to include:
  – Review of CERHR criteria and clear FDA
    assessment criteria
  – Review of CERHR studies
  – Update with studies published through 6/09
  – Update discussion of epigenetics and
    metabolism/obesity
Point of Departure (POD) Analysis
• Comment: suggested a Benchmark Dose
  analysis using available software as basis for
  POD.
• Response: FDA has evaluated previous BMDL
  analysis of Tyl et al. studies.
     Point of Departure (POD) Analysis
                                FDA Update
•   BMDL analyses of BPA data already published:

     – Willhite et al. BMDL10           75 mg/kg bw/day in female mice
                                             ovarian cysts (Tyl 2008)

     – CERHR


    Source       Finding                     BMD10   BMDL10 BMD1SD BMDL1SD
                                                            mg/kg bw/day

    Rat          Terminal body weight        46      35          41        31
    (Tyl 2002)   (F2♂)
    Mouse        Hepatocyte centrilobular    20.7    11.8
    (Tyl 2008)   hypertrophy incidence
                 (F0 ♂)
    Results of BMDL Analysis
• FDA previously approached POD using NOAELs
  from Tyl et al. (2002, 2008).
• For Tyl et al. studies, FDA has confidence in
  relying on the published analysis
  – Standard approach using EPA software
  – CERHR BMDL analysis was comprehensive

• BDML10 approach does not indicate a more
  protective POD.
Exposure Assessment

       Update
           Infant Formula Exposure
                -Existing Data-
•   Probabilistic analysis on the existing formula data
     – Distributions used in place of point estimates
     – Provides an intake distribution
•   BPA concentrations in liquid formula
     – FDA, Canada and EWG studies resulted in a total of 36 liquid
       samples representative of the US market
     – 0.04-8.6 μg/kg range, 3.1 μg/kg average (at use levels)
•   Formula consumption
     – Food intakes derived from the 1994-96, 98 USDA Continuing
       Survey of Food Intakes by Individuals (CSFII)
     – 2-Day survey; ~1200 infants (<1 year)
        Infant Formula Exposure
             - Existing Data-
• Estimated Daily Intake (EDI) of BPA from
  existing literature data on liquid infant formula
   – Mean : 0.4 μg/kg-bw/day
   – 90th Percentile : 0.8 μg/kg-bw/day
• Assumptions
   – 2-day average intake data reflects usual intake
   – Literature BPA concentration data set is
     representative of all liquid infant formula
        Infant Formula Exposure
               -FDA Data-
• Jan. 2009 our labs analyzed several formula
  products
  – Brand, formula type, container size, location
• Final results on analysis
  – All major formula manufactures
  – 35 Unique product/container combinations with 100
    individual samples for analysis
  – Liquid (13 RTF, 9 Concentrates) with 85 individual
    analyses
  – Powder (13) with 15 individual analyses
       Infant Formula Exposure
              -FDA Data-
• Update of our 1997 method to allow for rapid
  analysis
  – Liquid chromatography tandem mass spectrometry
    (LC-MS/MS)
     • Reverse phase with C18 column
     • Methanol/water (mobile phase) with gradient
     • EPA method (40 CFR Ch 1, Part 136) used to determine a
       method LOD of 0.15 ng/g
  – Modified sample preparation
     • All samples analyzed “as consumed”
     • Aqueous extract (with D6 BPA) subjected to solid phase
       extraction
     • D16 BPA used as an internal standard
       Infant Formula Exposure
              -FDA Data-
• Liquid formula
  – 0.5-10.6 μg/kg range; 3.4 μg/kg average
  – Average for RTF higher
• Powder formula
  – Non-detect at <0.15 μg/kg (except 1 sample)
• Observations
  – BPA concentrations for liquids consistent with what
    FDA previously reported
  – No difference in BPA concentrations based on
    geography
  – Manufacturer and can surface area-to-mass ratio
    were the only two important factors
       Infant Formula Exposure
              -FDA Data-
• BPA concentration data
  – Individual results from all 100 data points considered
    to be independent of each other
  – <LOD taken as ½ LOD
• Formula Consumption
  – Food intakes from 1999-2006 National Health and
    Nutrition Examination Survey (NHANES)
  – 1-Day survey, ~2000 infants (<1 year)
• Both segregated by RTF, concentrates &
  powder
       Infant Formula Exposure
              -FDA Data-
• EDIs for BPA for 0-12 months of age
                Mean         90th         N
                µg/kg-bw/d   µg/kg-bw/d   infants
  All Formula   0.13         0.45         1450
  RTF           0.31         0.85         176
  Concentrate   0.39         0.65         291
  Powder        0.01         0.01         1077
         Infant Formula Exposure
                -FDA Data-
• EDIs for BPA by month of age & formula type
            PC Bottle Exposure
• Original assumptions
  – <10 ppb (sterilization, as high as 100ºC), months 1-2
  – <1 ppb (normal, RT conditions), months 3-12
• New considerations
  – Liquid is sterile, powder is not
  – If used at all, boiled water may be added to powdered
    formula anywhere from RT to boiling (100ºC).
    FAO/WHO recommends no less than 70ºC
• Expanded analysis on testing conditions
  – Segregated according to recommended use, current
    practices and extreme (non-typical) use
  – Focused on BPA migration to water for short time
    frames at temperatures ranging from RT to 100ºC
           PC Bottle Exposure
• Recommended use (WHO, CFSAN, AAP)
  – Boiled & cooled, no microwave heating, refrigerated
    or used within 2 hours; <1 ppb BPA
• Current practices (microwave oven, dishwasher)
  – Reheating <1 ppb, dishwasher <2.5 ppb
• Data generally support the same or lower levels
  than our previous assumption
• Next Step: Probabilistic exposure model to
  consider migrations data and use patterns from
  Infant Feeding Practices Study
          Toddler Food Exposure
             -Existing Data-
• Toddler food is packed in glass jars with metal
  closures
  – Metal lid contains a primer coating not directly contact food
• Recent study from Canada (Cao 2008)
  – Analyzed >120 toddler food products (7 manufacturers)
  – Fruits, vegetables, meats and desserts
  – BPA not confirmed in 23 of 122 products
  – Of the 99 products, 15% non-detect (<0.18 ppb) and 70% <1
    ppb. Average BPA concentration was 1.1 ppb
  – One brand showed higher BPA concentrations in all foods
  – Highest BPA concentration was 7.2 ppb in 2 products
          Ongoing Activities
        -Existing Toddler Data-
• Currently conducting a probabilistic
  analysis using Canadian data on BPA in
  toddler food
  – Individual results from all 100 samples
    considered to be independent of each other
  – Rely on food intakes from NHANES 1999-
    2006
• Preliminary EDIs for 0-12 months
  – Mean : 0.02 μg/kg-bw/day
  – 90th Percentile : 0.06 μg/kg-bw/day
         Ongoing Activities
      -New Toddler Food Data-
• Plans for FDA study on BPA in toddler
  food for the US market
  – Sampling plan prepared focusing on
    • Five major brands
    • Three food stages (Stage 1, 2 and 3) which
      represent a range of foods, consistencies, ages
      and locations
  – GC-MS methodology will need to be modified
    and validated for “solid” foods
             Ongoing Activities
             -Adult Exposure-
• Original deterministic approach
  – 22 ppb BPA average all canned fruits & vegetables
  – ~17% of an individual’s daily diet is canned food
• Probabilistic approach
  – Updated survey of the open literature
  – Over 10 publications containing BPA concentrations
    in a variety of food & beverages from several
    countries
  – Publication details, including sample identities,
    country of origin, and BPA concentrations, tabulated
    and further categorized by food class.
           Ongoing Activities
           -Adult Exposure-
• Canned food consumption
  – Food intakes from NHANES 1999-2006
  – Foods categorized by class or type (e.g.,
    fruits)
• Individual BPA concentration data and
  canned food intakes for each class used
  for probabilistic analysis

								
To top