An accurate collection of journal articles from 2020 and 2021 on preclinical safety pharmacology gathered relevant peer reviewed papers focused on the challenges that are redefining preclinical and toxicology milestones in drug evaluation (Preclinical safety pharmacology in a new era; Drug Discovery Today Volume 27, 1 Jan. 2022)
In this blog, we will discuss new advanced models to assess immunotherapy candidates (Breous-Nystrom et al. Current Opinion in Toxicology 2020, 23-24:39–45).
The immunotherapy field has led to important therapy discoveries. However, they have unveiled the little relevance of the used animal models. Even though these entities have undergone all the expected preclinical safety assessments, up to 90% of the candidates had failed due to either low efficacy or unmanageable toxicity.
Because the most used animal models have become irrelevant and even misleading in those cases, there is an urgent need for new models that can better predict the reactions to be expected at the clinical setting.
In search for alternatives, new humanized micro-physiological systems and in silico models have made possible animal free and human/patient centric developments.
In non-human primates (NHP) settings, low predictability can be explained due to disease presentation, acute and chronic effects that are not displayed in studies that conventionally last 4 weeks. It has also been discerned that some candidate targets are different depending on the species. As an example, in humans CD16 antigen recognizes IGg1 and 3 of its targets (FcYRc) while in NHP the targets are IGg1 and 2 of these receptors. It was also a surprise to find out that NHP do not express the target in granulocytes (as it does in humans). These factors together with the generation of antibodies against the humanized candidate tends to discard NHP as predictable model.
2D cell cultures have been extensively used to measure cytotoxicity. However, some evidence has alerted that this model could miss to measure some safety concerns. 3D models have been used in parallel to evaluate 2D data credibility. As a good example of this discrepancy, when beating cardiomyocytes 3D models were exposed to a cell therapy agent (MAGE A3), an important cytotoxicity feature became apparent.
The authors call for more research on different cell types and in vitro models that could represent in vivo expression patterns.
Recent advances in cell culture technology, stem cell biology, biomaterials, and microfluidics have allowed the creation of complex, reproducible, physiology mimicking in vitro models, with improved predictability of adverse events of biologics, these models are broadly divided into:
- Human organoids: collection of organ-specific cell types from stem cells or organ progenitors. They self-organize via cell sorting. Many organoids have been developed from different organs
- Human micro-physiological systems (MPS): offered as organ (liver- or lung)-on-a-chip that recreates physiology of the prepared organs.
These systems emulate better the biology of human tissue and organs when using human cells.
Even though the use of these new models is encouraged, a major challenge is to build trust in users: regulators, clinicians, toxicologists and even researchers.
To agree on common grounds requirements, the authors suggest the following points
- A more systematic approach when selecting cells and emphasize on those key vital organs.
- Immune cells (from both tissue resident and peripheral blood) must be incorporated in the systems
- Parallel assessment of 2D and 3D models is needed to demonstrate the relevance of both
- Already marketed drugs must be used in parallel as a benchmark for reactions that could be expected with the immunotherapy agents.
After reviewing the subject, it is possible to believe that many candidates could have been missed in previous more conventional settings due to their low predictability.
Human derived models might complement the preclinical area information that we need to ensure safety and efficacy of the drug/immunotherapy agents that should be characterized before they are available to the patients.