Module 3 - Strategic case studies in practice

ICH Q3D(R1) Guideline

ARSENIC

Summary of PDE for Arsenic

Arsenic (As)

Oral

Parenteral

Inhalation

PDE (µg/day)

15

15

1.9

Introduction

Arsenic (As) is ubiquitous in the environment and present in food, soil, drinking water and in air. Inorganic arsenic occurs in trivalent (e.g., arsenic trioxide, sodium arsenite) or pentavalent (e.g., sodium arsenate, arsenic pentoxide, arsenic acid) forms. Arsenic has no known useful biological function in human or mammalian organisms. This assessment focuses on inorganic arsenic, because this is most relevant for drug products.

Safety Limiting Toxicity

Inorganic arsenic has shown to be genotoxic, but not mutagenic and has been acknowledged as a human carcinogen (Group 1; IARC, 2012). Due to its ubiquitous nature and toxicity profile, there have been many risk assessments conducted of arsenic and arsenic compounds, which utilize non-threshold, linear dose response approaches (Meharg and Raab, 2010). For the most part the effects of arsenic in humans have not been reproduced in animals, so the risk assessments have to rely heavily upon epidemiology data in populations with high exposure concentrations (Schuhmacher-Wolz et al. , 2009). In humans, both cancer and non-cancer effects have been linked to arsenic exposure. Oral exposure has been linked to cancers of the skin, liver, lung, kidney and bladder. Following inhalation exposure there is evidence for an increased risk of lung cancer (ATSDR, 2007; IARC, 2012; EU EFSA, 2009; WHO, 2011; US EPA, 2010). The skin (dyspigmentation, palmoplantar keratosis) and gastrointestinal tract (e.g., nausea) appear to be the most sensitive targets for non-cancer adverse effects after oral ingestion while vascular disease, reproductive effects and neurological effects are also reported as non-cancer endpoints (IARC, 2012; Schuhmacher-Wolz et al. , 2009; US EPA, 2007). Oral exposure studies suggest that skin lesions may appear at levels above 0.02 mg As/kg/day; no effects were generally seen at levels from 0.0004 to 0.01 mg As/kg/day (ATSDR, 2007). There are insufficient epidemiological data to set a LOEL or NOEL for other endpoints. The regions of hyperkeratosis may evolve into skin cancers (ATSDR, 2007) and can possibly be considered predictive of skin and internal cancers and the non- cancer long-term adverse health effects (Chen et al , 2005; Hsu et al. , 2013; Ahsan and Steinmaus, 2013). Studies of large populations (~40,000) exposed to arsenic concentrations in well water at 1000 µg/L and higher in southwestern Chinese Taipei have been the basis of risk assessments of skin cancer, and more recently of bladder and lung cancer (US EPA, 2010). Recent meta-analyses of cancer risk have indicated no additional bladder cancer risk at low dose exposure (<100–200 µg/L) (Chu and Crawford-Brown, 2006, 2007; Mink et al. , 2008). This is consistent with the work of Schuhmacher-Wolz et al. , (2009). An inhalation unit risk for cancer of 0.0043 per µg/m 3 has been established by the US EPA based on data from two US smelters (US EPA, 2007). The Texas Commission on Environmental Quality provided an update to the US EPA Unit Risk Factor (URF), incorporating additional years of follow-up to the US EPA data and additional data on workers from the United Kingdom and Sweden. The Commission calculated a URF of 0.0015 per µg/m 3 . This URF translates to an air concentration of 0.067 µg/m 3 at a risk of 1 in 100,000 excess lung cancer mortality (Erraguntla et al. , 2012).

30