Building upon previous work used to establish the Threshold of Toxicological Concern (TTC) approach, the feasibility of deriving new TTCs was investigated using estimated chronic human equivalent benchmark dose (eBMD_HED) values from EPA’s ToxValDB (v96_1). Chemical level values, i.e. those representative of the study level log eBMD_HED values, were derived by taking the 25^th percentile of the fitted normal distribution. Chemical structural information was extracted from the EPA CompTox Chemicals Dashboard resulting in a dataset of 4076 substances. Following processing through the Kroes et al. (2004) workflow, a set of 1269 substances were subsequently assigned a Cramer class using the Toxtree Cramer original module. The 5^th percentile of the empirical cumulative distribution function for each Cramer structural class was then derived. The Cramer I class 5^th percentile was estimated as 4.16 mg/kg-day resulting in a proposed TTC of 416 µg/kg-day. The TTCs for Cramer II and III were 169 µg/kg-day and 14 µg/kg-day respectively. In addition to deriving TTCs based on eBMD_HED values, a data-driven approach was investigated to derive TTCs by clustering the dataset into structure-based clusters for comparison with Munro Cramer classes. TTC values derived from the structural clusters were found to be in close to the Munro values showing that the chemical makeup of the categories themselves resulted in a more significant impact in the generated TTC values than the toxicity data itself. Nonetheless eBMD_HED values offer a more direct way of establishing exposure thresholds that are human-relevant and protective of human health.