Feasible approaches for arsenic speciation analysis in foods for dietary exposure assessment: a review.

Abstract

Arsenic (As) occurs naturally in different forms and oxidation states. Amongst them, inorganic arsenic (iAs) is classified as both genotoxic and carcinogenic whilst other organic arsenic species are considered less toxic. As in rice is mainly present in the form of iAs which therefore poses a health risk to populations that consume rice as a staple food. In 2011, the Joint Food and Agriculture Organisation/World Health Organisation Expert Committee on Food Additives determined the iAs benchmark dose lower confidence limit for a 0.5% increased incidence of lung cancer in humans (BMDL_0.5) which computed to be 3.0 μg/kg body weight (bw)/day. However, the European Food Safety Authority (EFSA) has recently lowered the BMDL_0.5 of iAs to 0.06 μg iAs/kg bw per day based on a low risk of bias case-control/cohort study on skin cancer as a Reference Point (RP). Subsequently, EFSA established a BMDL₁₀ of 18.2 mg monomethylarsenic (V) (MMA(V))/kg bw/day and 1.1 mg dimethylarsenic (V) (DMA(V))/kg bw/day as RPs with reference to studies on skin cancer and urinary bladder tumours in rats respectively. Therefore, As speciation is essential when conducting dietary exposure assessment. Recent studies showed thiolated counterparts of MMA and DMA were found in certain foodstuffs, especially grain. However, these thiolated As species were not recognised in acidic, basic or peroxide systems as they transformed to MMA and DMA in these extractants. Therefore, one of the objectives of this review was to assess whether published analytical methods are fit for As speciation analysis, especially for iAs, MMA and DMA, in foodstuffs. Besides, discussion was conducted on whether limits of detection are sufficiently low for dietary exposure assessment with respect to recently established RPs of iAs, MMA and DMA when an upper bound approach is applied. Moreover, possible future research gaps are identified based on current knowledge and existing literature.