Selection of some current topics in preclinical safety (PCS)

PCS testing has developed tremendously since its beginning in the early 1960s. Development is continuing. A few aspects are presented in Toxicology in the 21^st century.

A selection of topics is listed below. Please refer to textbooks, regulations and publications for further and more updated information.

The fields of -omics (genomics/transcriptomics/proteomics with metabolomics and toxicogenomics, etc.) has seen much development during the recent decades. They are useful e.g. for screening of compounds, once it is known for what particular change one has to look for. However, similar to in vitro testing, -omics investigations do generally not substitute for in vivo studies.

New approaches include also the use e.g. of the CRISPR/Cas especially for transgenes and the next generation of sequencing techniques, new types of formulations e.g. produced by nanotechnologies, etc. Epigenetic data are increasingly assessed. It has recently been shown that epigenetic changes (e.g. methylation, histone modification, effects on non-coding RNA, etc.) can be inherited over several generations, though such changes were once considered as being reversible.

Also, in the lay press, the discussion is continuing since many decades about the value of animal testing for prediction of adverse findings in humans.

General aspects are summarized in the Introduction to ‘Concordance between preclinical and clinical findings’. No doubt, there are findings e.g. in general and reproductive toxicity, which are typical animal findings without (much) meaning for humans.

The presentation Non-proliferative lesions shows that adverse non-neoplastic effects are not uncommon. Standard studies with multiple endpoints are often sufficient, if undertaken by expert study directors and undergoing good histopathological examination. Confounding factors are e.g. immature test animals, spontaneous diseases including infestation, etc. Identification of the primary target might help to establish a mode of action (MoA), which may need investigations of early time points and time-course studies. Overall, a sound and comprehensive scientific assessment is needed, taking into account also safety ratios, reversibility, severity of the treated disease including therapeutic alternatives and monitorability in man. Hormonally mediated effects are generally reversible and often species-specific, that is without relevance to man, but the ultimate proof is often possible only in man with early markers.

The presentation Genotoxicity – Carcinogenicity underlines that tumors in animals are relevant for human risk assessment, if not shown otherwise, e.g. by the proof that the MoA is without relevance to human beings. Tumors with doubtful relevance for the risk assessment in humans are summarized.

The conclusion of the presentation Reproductive toxicity says that embryotoxicity or teratogenicity does not result in termination of the development of a drug candidate, but leads to modifications of its use (e.g. indication, restriction). Adverse fertility effects and delayed, but reversible effects on fetal development may be handled similar to general toxicity findings, namely with an assessment based upon risk management strategies.

First in humans administered doses are estimated based on no observed adverse effect levels (or other key values for anti-cancer drug candidates) in the most sensitive and representative species. The resulting values are compared with the minimally anticipated biologic effect level in humans and the pharmacologically active dose in animals. Animal doses might have to be converted to human equivalent doses according to the respective FDA guidance and safety factors have to be used. As summarized in the presentation Conclusions a 1:1 concordance between test animals and humans cannot be expected, as this is also true of different human studies. This means that often a proper risk assessment has to be conducted.

Today, many toxic phenomena can be screened in silico, e.g. by using databases for structures often associated with genotoxicity and certain types of other toxicities, such as hepatotoxicity. Screening for further toxic phenomena is possible with in vitro methods, in particular cell cultures often used for screening for hepatotoxicity. Some in vitro methods are firmly established such as the Ames test for genotoxicity. However, in the experience of the author of this website, the majority of toxic effects are not known a priori and therefore too many in vitro studies would be needed for screening. The only reasonable way is by testing animals, knowing – as said above – that sometimes results cannot be translated easily to humans. Even for testing of genotoxicity one in vivo test is necessary for registration of a new drug as explained in Type, duration and dose of regulatory studies.

Challenging for the interpretation of findings and the risk assessment of compounds are also the differentiation of adverse and non-adverse findings. The latter can be due e.g. to exaggerated pharmacological effects or to adaptation to increased chemical exposure. The controversy continues and often no clear distinction is possible.

Another topic under discussion, also for experiments at large, is reproducibility of findings. It is known that biological effects show much variation and depend on factors only artificially controllable in experiments. This was one of the reasons why the determination of the lethal dose (LD₅₀ testing) was abandoned. It is therefore important to always judge a finding in the light of its potential biological significance.

The MoA and the mechanism of action (MeA) have received much attention, too. Both terms are often used interchangeably, though theoretically they differ somewhat: the MoA is characterized by the physical, anatomical or functional (‘physiological’) changes that take place (e.g. mitochondrial toxicity), while the MeA refers to the biochemical process bringing about the specific action of the administered substance (e.g. binding of the toxin to a receptor). Thus, the continuum starts with biochemical modifications by toxins, goes on to subcellular (partly biochemical, partly ‘physiological’) and cellular (often ‘physiological’) alterations, results in toxic (generally ‘physiological’) changes in organs and may finally lead to toxic changes in the whole organism. The term adverse outcome pathways (AOP) is another key word frequently used in this context. An AOP is a structured way to see chemically induced biological events leading to adverse outcomes in (parts of) organs and potentially in whole organisms. The AOP is considered relevant to risk assessment.

Finally, though the enumeration of issues is far from complete, the microbiome is increasingly recognized as a contributing factor to health in general and toxicity in particular. The microbiome is defined as the totality of microorganisms including e.g. bacteria, fungi and viruses that inhabit a particular environment, especially the gastrointestinal system. It is not difficult to imagine that the microbiome is influenced by toxins, which may result in organ changes and/or sickness of the whole organism.