Role of Risk of Bias in Systematic Review for Chemical Risk Assessment: A Case Study in Understanding the Relationship Between Congenital Heart Defects and Exposures to Trichloroethylene.

The National Academy of Science has recommended that a risk of bias (RoB; credibility of the link between exposure and outcome) assessment be conducted on studies that are used as primary data sources for hazard identification and dose-response assessment. Few applications of such have been conducted. Using trichloroethylene and congenital heart defects (CHDs) as a case study, we explore the role of RoB in chemical risk assessment using the National Toxicology Program's Office of Health Assessment and Translation RoB tool. Selected questions were tailored to evaluation of CHD and then applied to 12 experimental animal studies and 9 epidemiological studies. Results demonstrated that the inconsistent findings of a single animal study were likely explained by the limitations in study design assessed via RoB (eg, lack of concurrent controls, unvalidated method for assessing outcome, unreliable statistical methods, etc). Such limitations considered in the context of the body of evidence render the study not sufficiently reliable for the development of toxicity reference values. The case study highlights the utility of RoB as part of a robust risk assessment process and specifically demonstrates the role RoB can play in objectively selecting candidate data sets to develop toxicity values.

Integration of mechanistic and pharmacokinetic information to derive oral reference dose and margin-of-exposure values for hexavalent chromium.

The current US Environmental Protection Agency (EPA) reference dose (RfD) for oral exposure to chromium, 0.003 mg kg-1  day-1 , is based on a no-observable-adverse-effect-level from a 1958 bioassay of rats exposed to ≤25 ppm hexavalent chromium [Cr(VI)] in drinking water. EPA characterizes the confidence in this RfD as "low." A more recent cancer bioassay indicates that Cr(VI) in drinking water is carcinogenic to mice at ≥30 ppm. To assess whether the existing RfD is health protective, neoplastic and non-neoplastic lesions from the 2 year cancer bioassay were modeled in a three-step process. First, a rodent physiological-based pharmacokinetic (PBPK) model was used to estimate internal dose metrics relevant to each lesion. Second, benchmark dose modeling was conducted on each lesion using the internal dose metrics. Third, a human PBPK model was used to estimate the daily mg kg-1 dose that would produce the same internal dose metric in both normal and susceptible humans. Mechanistic research into the mode of action for Cr(VI)-induced intestinal tumors in mice supports a threshold mechanism involving intestinal wounding and chronic regenerative hyperplasia. As such, an RfD was developed using incidence data for the precursor lesion diffuse epithelial hyperplasia. This RfD was compared to RfDs for other non-cancer endpoints; all RfD values ranged 0.003-0.02 mg kg-1  day-1 . The lowest of these values is identical to EPA's existing RfD value. Although the RfD value remains 0.003 mg kg-1  day-1 , the confidence is greatly improved due to the use of a 2-year bioassay, mechanistic data, PBPK models and benchmark dose modeling.

Systematic review of the potential adverse effects of caffeine consumption in healthy adults, pregnant women, adolescents, and children.

To date, one of the most heavily cited assessments of caffeine safety in the peer-reviewed literature is that issued by Health Canada (Nawrot et al., 2003). Since then, >10,000 papers have been published related to caffeine, including hundreds of reviews on specific human health effects; however, to date, none have compared the wide range of topics evaluated by Nawrot et al. (2003). Thus, as an update to this foundational publication, we conducted a systematic review of data on potential adverse effects of caffeine published from 2001 to June 2015. Subject matter experts and research team participants developed five PECO (population, exposure, comparator, and outcome) questions to address five types of outcomes (acute toxicity, cardiovascular toxicity, bone and calcium effects, behavior, and development and reproduction) in four healthy populations (adults, pregnant women, adolescents, and children) relative to caffeine intake doses determined not to be associated with adverse effects by Health Canada (comparators: 400 mg/day for adults [10 g for lethality], 300 mg/day for pregnant women, and 2.5 mg/kg/day for children and adolescents). The a priori search strategy identified >5000 articles that were screened, with 381 meeting inclusion/exclusion criteria for the five outcomes (pharmacokinetics was addressed contextually, adding 46 more studies). Data were extracted by the research team and rated for risk of bias and indirectness (internal and external validity). Selected no- and low-effect intakes were assessed relative to the population-specific comparator. Conclusions were drawn for the body of evidence for each outcome, as well as endpoints within an outcome, using a weight of evidence approach. When the total body of evidence was evaluated and when study quality, consistency, level of adversity, and magnitude of response were considered, the evidence generally supports that consumption of up to 400 mg caffeine/day in healthy adults is not associated with overt, adverse cardiovascular effects, behavioral effects, reproductive and developmental effects, acute effects, or bone status. Evidence also supports consumption of up to 300 mg caffeine/day in healthy pregnant women as an intake that is generally not associated with adverse reproductive and developmental effects. Limited data were identified for child and adolescent populations; the available evidence suggests that 2.5 mg caffeine/kg body weight/day remains an appropriate recommendation. The results of this systematic review support a shift in caffeine research to focus on characterizing effects in sensitive populations and establishing better quantitative characterization of interindividual variability (e.g., epigenetic trends), subpopulations (e.g., unhealthy populations, individuals with preexisting conditions), conditions (e.g., coexposures), and outcomes (e.g., exacerbation of risk-taking behavior) that could render individuals to be at greater risk relative to healthy adults and healthy pregnant women. This review, being one of the first to apply systematic review methodologies to toxicological assessments, also highlights the need for refined guidance and frameworks unique to the conduct of systematic review in this field.