It has been a while, but if you recall in the previous (and first) blog in this series, we gave the potted history of the main TTC benchmarks that can used to assess an untested chemical. For any exuding a structural suspicion of genotoxicity, the critical value was 0.15 µg/day, exposures below this being assumed to pose no significant carcinogenic risk even if experienced daily for life. For compounds with no such DNA-reactive structural baggage, the magic numbers are 90 µg/day, to judge those with the toxicological nastier molecular appearance, and 1800 µg/day for their more innocuous-looking peers. If the estimated exposure to an untested Cramer III compound is, for example, below 90 µg/day (again, daily for ever) then a TTC enthusiast will be concluding that the toxicological risks are acceptably low. Given that EFSA has previously stated that “in principle, the science supports the application of the TTC approach in any area of chemical risk assessment for which human exposures are low, whether exposure is from deliberate addition or due to contamination” (EFSA, 2012), does the risk assessor only now need to remember three numbers to ply their trade in the land of the untested chemical?

Well, sadly no.

A lack of representation in the original compound database means that the TTC approach is not appropriate for inorganics including metals (Munro et al., 1996), or for polymers (EFSA, 2012). It is also inappropriate for proteins because of the possibility of allergenicity from low exposures (Kroes et al., 2004), for substances predicted to have local effects on the gastro-intestinal tract, for nanomaterials, and for substances with a high potential for bioaccumulation (EFSA, 2012).

Any other exclusions?

Although the 1.5 µg/day threshold of regulation (TOR) was widely accepted in the US, scientists immediately started looking at ways to develop and improve upon it. The impact of chemical structure on carcinogenic potential was explored within the FDA, with a view to raising the safety margin of the TOR (Cheeseman et al., 1999). Of the 709 carcinogens that they examined, the same structural features kept cropping up in those that were the most potent: the N‑nitroso group; strained heteronuclear rings; α-nitrofuryl compounds; hydrazines, triazenes, azides and azoxy compounds; organophosphates; and polycyclic amines. Certain heavy metals and some endocrine-active substances weren’t looking too good either. Later, an International Life Sciences Institute (ILSI) task force (Kroes et al., 2004) expanded the dataset to 730 chemicals, which they split into 18 structural groups (classes). Five of the classes contained a significant number of members for which the carcinogenic risk would likely be greater than 1 in a million. These five classes (aflatoxin-like, azoxy and N-nitroso compounds, steroids, and tetrahalogenated dibenzodioxins and dibenzofurans) were termed “the Cohorts of Concern”. Neither the TTC approach nor the cancer TTC value of 0.15 µg/day (0.0025 µg/kg bw/day) can be applied to chemicals belonging to any of these cohorts.

Any other “special cases” or problematic groups of chemicals?

The original dataset employed to set the non-cancer/toxicity TTC values (Munro et al., 1996) was expanded and re-evaluated by ILSI to establish whether effects on the central nervous, immune or endocrine systems or on development (including developmental neurotoxicity) were more sensitive than the TTC value for Cramer class III compounds; and, if so, whether they were adequately covered by the TOR of 1.5 µg/day (0.025 µg/kg bw/day) (Kroes et al., 2000). Cumulative distribution of NOELs was consistent with previous evaluations suggesting that these endpoints, apart from neurotoxicity, were adequately covered by the existing TTC values. In addition, none of the endpoints (including neurotoxicity) were more sensitive than carcinogenicity, the cancer NOELs still being 2-3 orders of magnitude lower than the NOELs for neurotoxicity (after application of the safety factor of 100 for intra- and inter-species differences).

The problematic neurotoxicity data prompted a further ILSI re-evaluation, which found that the 5th percentile NOEL for organophosphate compounds in isolation was an order of magnitude lower than that of other neurotoxins in the dataset (Kroes et al., 2004). The organophosphates and carbamates were therefore removed from Cramer class III and given a group-specific TTC value of 18 µg/day (0.3 µg/kg bw/day). Subsequent re-assessment of the remaining neurotoxins showed that the TTC value for Cramer class III compounds was suitably health-protective for this endpoint.


EFSA Scientific Committee (2012). Scientific Opinion on Exploring options for providing advice about possible human health risks based on the concept of Threshold of Toxicological Concern (TTC). EFSA Journal 2012;10(7):2750 [103 pp.]

Kroes R, Galli C, Munro I, Schilter B, Tran L, Walker R and Würtzen G (2000). Threshold of toxicological concern for chemical substances present in the diet: a practical tool for assessing the need for toxicity testing. Food and Chemical Toxicology, 38, 255-312.

Kroes R, Renwick AG, Cheeseman M, Kleiner J, Mangelsdorf I, Piersma A, Schilter B, Schlatter J, van Schothorst F, Vos JG and Wurtzen G (2004). Structure-based Thresholds of Toxicological Concern (TTC): Guidance for application to substances present at low levels in the diet. Food and Chemical Toxicology, 42, 65-83.

Munro IC, Ford RA, Kennepohl E and Sprenger JG (1996). Correlation of a structural class with no‑observed-effect levels: a proposal for establishing a threshold of concern. Food and Chemical Toxicology, 34, 829-867.



TTC More recent work News Meet our team