4.9.11 General Toxicological Screening Models

Acute Oral Toxicity — LD50 Screening Model Beyond the disease-specific efficacy models described above, every candidate pharmacological agent must...

Acute Oral Toxicity — LD50 Screening Model

Beyond the disease-specific efficacy models described above, every candidate pharmacological agent must additionally undergo an acute toxicological screen to establish a preliminary safety margin before further in-vivo efficacy testing proceeds, most commonly using the OECD 423 (Acute Toxic Class) or OECD 425 (Up-and-Down Procedure) protocols described in full in Phase 5. Because this screening step logically precedes and safeguards every disease-specific efficacy study presented in this section, acute toxicity assessment is properly regarded as a foundational, cross-cutting element of in-vivo study design rather than a disease model in its own right, and the interested reader is directed to the detailed treatment of acute, sub-chronic, and chronic toxicology testing in Phase 5.

Selecting the Appropriate Disease Model

With such a broad catalogue of validated models available, selecting the model most appropriate to a specific research question requires the investigator to weigh several factors: the mechanistic fidelity of the model to the human disease process under investigation, the specific pharmacological hypothesis being tested (a model probing GABAergic anticonvulsant mechanisms, for example, is not interchangeable with one probing sodium-channel mechanisms), the practical resources and technical expertise available, the study duration that can be ethically and practically justified, and the specific regulatory or publication context the resulting data are intended to support. As a general principle, acute chemically or pharmacologically induced models are best suited to early-stage screening of large numbers of candidate compounds, while genetic, surgical, or chronic dietary models — though more resource-intensive — offer superior translational fidelity for late-stage candidate characterisation intended to support an eventual regulatory submission.

Key Point

No single animal model perfectly reproduces the full complexity of a human disease; each model represents a deliberate simplification that isolates and reproduces specific, experimentally tractable features of the human condition. Sound experimental practice therefore requires triangulating findings across multiple complementary models — for example, both an acute chemically induced and a chronic or genetic model of the same disease — before a candidate compound's efficacy signal is considered sufficiently robust to justify progression to the toxicological and, eventually, clinical evaluation stages described in the remainder of this text.