Human toxicity predictions

Toxicity Amelioration. Cancer researchers traditionally have not focused their attention on the question of toxicity amehoration. This is partiy attributed to the lack of predictive animal models for human toxicities. For example, the preclinical rat model, used as a predictor of myelosuppression, has failed to predict myelosuppression in humans in clinical trials. In addition, reduction of one toxicity may result in the emergence of another, more serious problem. Research efforts to address the problem of toxicity amelioration has progressed in several directions. The three most prominent areas are analogue synthesis, chemoprotection, and dmg targeting. 

In a first attempt to derive characterization factors with QSARs, the entire dataset of plastics additives was included, and aquatic ecotoxicity was predicted for two different trophic levels. This generated characterization factors that did not correspond well with the ones derived from experimental data [30]. Hardly surprising, but a clear indication that two trophic levels are unsufficient. A second attempt to derive characterization factors with QSARs are currently being performed [31]. In this second attempt, substances that are difficult to model in QSAR models have been removed from the dataset and the ecotoxicity has been predicted for three different trophic levels instead of two. However, results have not yet been obtained from this second attempt. If the results show that it is possible to derive reliable characterization factors by the use of QSARs, the current data gap regarding characterization factors for human toxicity and ecotoxicity could be... 

Grieshaber, C. (1991). Prediction of human toxicity of new antineoplastic drugs from studies in animals. In The Toxicity of Anticancer Drugs. Powis, G. and Hacker, M. Eds. Pergamon Press, New York, pp. 10-24. 

Use of in vivo Tests. In vivo tests are more relevant indicators than are in vitro tests of immunotoxicity since the dynamic interactions between the various immuno-components, as well as the pertinent pharmacokinetic (absorption, distribution, plasma concentrations) and metabolic factors, are taken into consideration. However, it is important to select the appropriate animal model and to design the protocol such that it will accurately reflect drug (or relevant metabolite) exposure to humans. For example, one should consider species variability when selecting the animal model, since biological diversity may further obscure the ability to accurately predict human toxicity. 

Use of Reference Databases to Interpret Toxicogenomic Results and Predict Potential Human Toxicity... 

Cytotoxicity Evaluated for confounding interpretation of in vitro efficacy assays, for predicting potential for human toxicity especially in liver but also if warranted by other safety assessments in bone marrow, kidney, neurons, immu-nocytes and so on. Also used for developing understanding of biochemical mechanisms of toxicity. HCA has been repeatedly demonstrated to be an effective tool in predictive toxicology. May also be used for certain translational safety biomarkers of toxicity [37]... 

Historically, in vitro cytotoxicity tests have not been effective in predicting human toxicity potential [11]. This has been attributable largely to insufficiency of duration of... 

There are also several important features of the cell type used that determine its effectiveness in predicting human toxicity. It should be of the same species as is... 

The goals of preclinical toxicity studies include identifying potential human toxicities, designing tests to further define the toxic mechanisms, and predicting the specific and the most relevant toxicities to be monitored in clinical trials. In addition to the studies shown in Table 5-1, several quantitative estimates are desirable. These include the no-effect dose—the maximum dose at which a specified toxic effect is not seen the minimum lethal dose—the smallest dose that is observed to kill any experimental animal and, if necessary, the median lethal dose (LD50)—the dose that kills... 

As the use of metabonomics advances, there are several challenges facing scientists using this tool that must be addressed in order to make it more mainstream and more relevant to predicting toxicity, and useful for hazard identification, human risk assessment and clinical medicine. First, advancing the use of metabonomics to identify mechanisms of toxicity is essential, and such efforts should help to increase the overall usefulness, validity, and relevance of toxicity prediction and biomarker development. Second, the use of metabonomic evaluations in the course of chronic toxicity rather than the heretofore emphasis on acute studies will help to establish its place in following the... 

In other areas of predictive toxicology and fate, progress has been steady, and spurred on in recent years by many of the legislative and commercial pressures mentioned in Table 1.1. Progress and interest in the prediction of human effects and pharmacokinetics has been complemented by advances in chemo-informatics. This has resulted in a large number of commercially available expert system approaches to toxicity prediction (see Chapter 9) and algorithms for the prediction of absorption, distribution, metabolism, and excretion (ADME see Chapters 10 and 11). 

Approximately 100,000 separate chemicals may be released into the environment annually it is frightening to consider that reliable toxicity data exist for only a tiny proportion of these chemicals, probably less than 5%. The percentage of chemicals with a complete set of reliable toxicity data (i.e., across a broad spectrum of environmental and human health effects) is considerably less than 5%. Computer-aided prediction of toxicity has the capability to assist in the prioritisation of chemicals for testing, and for predicting specific toxicities to allow for labeling. Chapter 19 describes these activities in more detail. As the reliability of models for toxicity prediction increases, there will undoubtedly be increased use for the filling of data gaps. 

The most successful application of structure-based predictive modeling in the future may be to specific endpoints. As described in Chapter 8 the prediction of something as broad as carcinogenicity is very difficult. If one isolates individual effects and endpoints within this category, then more success will ensue. Such an approach is likely to require the development of tiered strategies for toxicity prediction. It will also allow for the integration of test data from other assays, and where possible, human knowledge. 

BMD models are used to estimate human health guidance values for environmental substances. QSARs are used to provide data estimates for chemicals that lack adequate experimental documentation. The ATSDR uses two commercial computational toxicology models to make toxicity predictions based on QSARs. To increase confidence in the models predictions, ATSDR used the models similarity search features and established a minimum threshold similarity distance value of 0.25 to increase the probability that predicted toxicity values are close to nearest analog chemicals. 

The Draize eye irritancy test, in which unanesthetized rabbits have irritant substances applied to their eyes, yields results that are inherently unreliable in predicting human toxicity. Humans and rabbits differ in the structure of their eyelids and corneas as well as in their abilities to produce tears. When comparing rabbit to human data on the duration of inflammation after exposure to 14 household products, they differed by a factor of 18 to 250. 

Exposure to toxic fire effluents can lead to a combination of physiological and behavioral effects of which physical incapacitation, loss of motor coordination, disorientation are only a few. Furthermore, survivors of a fire may experience postexposure effects, complications, and burn injuries, leading to death or long-term impairment. The major effects, such as incapacitation or death, may be predicted using existing rat lethality data, as described in ISO 1334431 or more recently, based on the best available estimates of human toxicity thresholds as described in ISO 13571,5 by quantifying the fire effluents in different fire conditions in small-scale tests, using only chemical analysis, without animal exposure. 

Ekwall, B. et al. 1998. MEIC evaluation of acute systemic toxicity VI prediction of human toxicity by rodent LDj0 values and results from 61 in vitro methods. ATLA 26, 617-658. 

Computers are now used in the design and development of new chemicals, and their employment in toxicity prediction could lead to improved products that present a reduced hazard to humans. Although computers are useful for performing routine calculations, they do not usually possess insight or rationalization. Therefore, they should represent only one of a number of test procedures used to formulate a full safety evaluation in a given chemical. Where they are used, their results should be interpreted by a panel of expert toxicologists capable of providing an overall view of the likely toxic risk in the human environment. 

Quantum Pharmaceuticals recently proposed a new method for toxicity prediction based on computation of small molecules affinity to about 500 human proteins. The analysis of binding profiles for about 1000 known pharmaceutical agents led to establishment of a relation between the toxicological properties of a molecule and its activity against the selected representatives of approximately 50 protein families. This activity profile was further used as a natural set of descriptors for various toxicological endpoints predictions, including human-MRDD, human-MRTD, human-TDLo, mouse-LDso (oral, intravenous, subcutaneous), rat-LDso (oral, intravenous, subcutaneous, intra-peritoneal), etc. ... 

---