This method can be used to support the identification of substances that likely do not require classification for acute oral toxicity through the use of animal-based testing methods.
Building on the results of a previous validation study, we ran a follow-up validation study for this method to assess its predictive capacity.
The results and our recommendation can be found on TSAR, the Tracking System for Alternative methods towards Regulatory acceptance.
Read more about the 3T3 Neutral Red Uptake (NRU) Cytotoxicity assay on TSAR
Acute oral toxicity/cytotoxicty
[collapsed]Acute oral toxicity refers to the general adverse effects induced by a substance following administration of single or multiple doses over a short period (24 hours). For a substance to be classified as acutely toxic, adverse effects should occur in a short period of time after administration (at least 14 days).
Acute oral toxicity testing usually involves studies with rats or other rodents. The major regulatory driver for using such tests is for classification and labelling purposes (according to the provisions of Regulation 1272/2008 on Classification, Labelling and Packaging.
All the accepted methods for determining acute oral toxicity are based on in vivo experiments that estimate the LD50 value (describes the lethal dose for 50% of a given population). The endpoint measured in the standard assays is either animal morbidity or death (OECD TG 423 and 425) or evident signs of toxicity (OECD TG 420).
Substances with an oral LD50 value greater than 2000 mg/kg body weight do not require classification as acute oral toxicants under EU CLP Regulation.
Although currently there are only in vivo methods approved for acute oral toxicity testing, efforts are underway to develop non-animal approaches to replace these.
The use of cell-based methods to predict acute oral toxicity has been extensively investigated. In this regard, in vitro cytotoxicity assays have been developed and evaluated against in vivo oral LD50data.
Cytotoxicity simply refers to whether a substance is toxic for a cell through general non-specific mechanisms of toxicity common to most cell types.
Previous research efforts have uncovered evidence of correlations between LD50 values in animals and IC50 cytotoxicity values in vitro.
On that basis, efforts continue to evaluate cell-based models for cytotoxicity as alternatives for acute oral toxicity testing in animals.[/collapse]
The 3T3 Neutral Red Uptake (NRU) Cytotoxicity assay
[collapsed]The 3T3 NRU method is a cell-based approach that assesses the cellular uptake of a dye in the presence of a particular substance. The 3T3 refers to the BALB/c3T3 mouse embryonic fibroblast cell line and NRU refers to Neutral Red Uptake (i.e. uptake of the dye).
The method is used to assess basal cytotoxicity. Assays that assess this type of toxicity look at how chemicals exert effects by disrupting structures and functions that are universal to all cells (e.g. cell membrane integrity, energy production and metabolism, transportation of molecules, secretion of molecules). This is opposed to assays that assess the effects of chemicals on specific molecular targets (such as receptors, ion channels etc.).
The 3T3 NRU method specifically targets basal cytotoxicity because the fibroblast cells lack any mechanisms relating to these specific molecular targets and also lack any metabolic competence in terms of phase I and II metabolism (i.e. they cannot biotransform parent compounds into metabolites).
Basal cytotoxicity assays quantify cytotoxicity through IC50 values - the concentration of test substance needed to decrease cell viability by 50% in the cell culture. This is measured by assessing the uptake of the dye, which is assessed via spectrophotometry.[/collapse]
Animal testing replacement
[collapsed]Several studies by us and others suggested that the 3T3 NRU method could identify chemicals that exerted effects through basal cytotoxicity and that correlation rates between in vitro IC50 values and oral rat LD50 values were high at around 60 to 70% (Ekwall, 1999; Halle, 2003; NIH, 2006; Prieto et al., 2013).
However, despite these good correlation rates, our previous NICEATM/ECVAM validation study (NIH, 2006) suggested that the method was not able to predict with sufficient accuracy the acute oral toxicity categories as defined by the UN global harmonised system (GHS).
While the peer review of that validation study (see below) suggested that the approach may be useful to predict starting dose for acute oral toxicity tests in animals, other data indicated specifically that the precision of prediction of low systemic toxicity from in vitro cytotoxicity test data is much better than the prediction of high systemic toxicity. This suggested that the 3T3 NRU test method may allow discrimination of a large fraction of the EU CLP unclassified compounds (LD50 > 2000 mg/kg body weight) with low false negative rate.
To assess this aspect more fully, we conducted a further validation study (see below) that had only two potential outcomes: either a compound falls under the classified category, (i.e. LD50 ≤ 2000 mg/kg) or under the unclassified group (i.e.LD50 > 2000 mg/kg). The validated test method protocol and the IC50–LD50 regression models from the previous NICEATM/ECVAM validation study were used.
This was done on the basis that if this test could be used to rule out compounds completely prior to animal-based testing (i.e. they were negative in the test, and so, they would not be classified at all as acutely toxic after oral exposure in animal studies), it could reduce the need for in vivo testing for acute oral toxicity when used as a building block in a wider tiered approach.
Currently there are only in vivo standardised test methods available for assessing acute oral toxicity for regulatory purposes. These are:
- The Fixed Dose Procedure (FDP in OECD TG 420)
- The Acute Toxic Class Method (ATC in OECD TG 423)
- The Up-and-Down Procedure (UDP in OECD TG 425)[/collapse]
[collapsed]The EURL-ECVAM follow-up study on the 3T3 NRU method used the same protocol and the IC50–LD50 millimole and weight regression models as the previous validation study. Fifty-six coded industrial chemicals were tested in three laboratories.
This follow-up study did not assess transferability or within/between laboratory reproducibility of the method since these were validated in the previous validation study.
There were also secondary aims to assess whether two variants of the previously validated protocol, an abbreviated version of the protocol and the protocol adapted for an automation platform give comparable results and therefore be useful for identifying unclassified chemicals.[/collapse]
Validation study outcomes
[collapsed]On the basis of the study we performed, as well as data from previous work, the 3T3 NRU test method appears to have a high sensitivity (92–96%) and consequently a low false negative rate (8-4%), when employed in conjunction with a prediction model, to distinguish potentially toxic versus non-toxic compounds - i.e. compounds that likely require classification or not. As such, substances found negative in the test would likely not require classification for acute oral toxicity
Assuming that most industrial chemicals are not likely to be acutely toxic, this test method could prove to be a valuable component of an integrated testing strategy, a read-across argument, or weight-of-evidence approach to identify non-toxic chemicals (LD50 > 2000 mg/kg). However, drawing conclusions on classification should only be done after consideration of a number of points. See below for these considerations.[/collapse]
EURL ECVAM recommendations
[collapsed]Our study suggested that the 3T3 NRU method may have considerable value in predicting whether substances likely do NOT have cytotoxic effects. This is significant for a number of reasons, including the potential for reducing (considerably) the need for animal based studies when assessing acute oral toxicity.
However, there are some limitations that need to be considered in relation to the negative results that this assay produces. These include the following points:
- Firstly the test can only be used to assess chemicals that act through basal cytotoxicity. This means that there is the potential for chemicals to be classified as negative but still be toxic via other mechanisms.
- Moreover, chemicals requiring some sort of activation or transformation prior to becoming toxic will also go undetected with this method.
Care is therefore needed in interpreting negative outcomes, despite the low false negative rate that we report above.
Additionally, this validation study suggests that the false positive rate is far too high to be of use for regulatory purposes. However, based on the previous validation study the method may be of use in terms of identifying the likely starting point for acute oral toxicity testing in animals. This may contribute to additional reductions in such animal usage, although this has been questioned (Schrage et al., 2011).
As such, the method should always be used in combination with other data to build confidence in the decision to not classify a substance for acute oral toxicity. Possible sources of information may include chemical analogues, physico-chemical properties, structural alerts, structure-activity relationships, and toxicokinetic data. The 3T3 NRU method may fit well within a weight of evidence (WoE) approach or as a component of integrated approaches for testing and assessment (IATA).
The method appears to be highly relevant for chemicals that are not specifically designed to have cellular toxic effects or targets (e.g. industrial chemicals).
As the method can give crucial information relating to basal cytotoxicity, a key event in many toxicological pathways, identification of chemicals that likely do not have this effect is important.
On the basis of this study and previous research, we recommended that data from the 3T3 NRU assay should be considered before embarking on animal experiments. In particular, the 3T3 NRU test method could prove a valuable component of a WoE or IATA approach for supporting hazard identification and safety assessment in agreement with the EU CLP Regulation and international regulatory schemes implementing the upper threshold of UN GHS Category 4 as the cut-off for non-classification of substances.
In particular, data from the 3T3 NRU assay may constitute an information source within a WoE approach under the provisions of the REACH regulation (Annex XI, 1.2) potentially supporting conclusions on absence of acute oral toxicity of industrial chemicals.
The possibility to waive the acute oral toxicity study based on the results of alternative tests that predict low acute oral toxicity has been included in the OECD Guidance Document on Considerations for Waiving or Bridging of Mammalian Acute Toxicity Tests (OECD GD 237).
A Weight-of-Evidence (WoE) based adaptation to the standard information requirement for an acute oral toxicity study has been included in the revised version of the ECHA guidance on the conduct of acute systemic toxicity studies, which offers the possibility to avoid an in vivo acute oral toxicity study for substances of low toxicity (i.e. LD50 > 2000 mg/kg).
Data from an oral repeated-dose toxicity study together with other pieces of information, such as results from in vitro cytotoxicity testing, physico-chemical properties, structural analysis and toxicokinetic assessment could be used by the registrant (ECHA, 2016).[/collapse]