Every preclinical pain program begins with a species decision, which shapes everything downstream, including endpoint selection, translational confidence, timeline, cost, and regulatory strategy. The choice deserves more scrutiny than it typically receives.
Here is how the four most commonly used preclinical species compare, and where each one fits in a well-designed development program.
Rodent models remain the most widely used across nearly every therapeutic area. They are reproducible, well-characterized, and cost-effective. For early-stage mechanism-of-action studies, target engagement, and high-throughput screening, they are hard to beat.
In pain research, rodent models (CCI, SNL, SNI, Brennan plantar incision) dominate the landscape but carry a well-documented translational limitation. The primary endpoint in most studies is evoked withdrawal reflexes, which capture only one dimension of a multidimensional clinical experience. Rodents do not exhibit spontaneous pain behaviors, and their skin architecture and peripheral nerve density differ fundamentally from humans.
In other areas, rodents bring their own strengths. Rat cisplatin-induced and ischemia-reperfusion models are workhorses for acute kidney injury (AKI) research, with established biomarker panels (KIM-1, NGAL, creatinine, BUN). Rat myocardial infarction models provide reliable infarct size and cardiac function data. Rodent stroke models (MCAo, 4VO) are well-validated for both focal and global ischemia.
The limitation is consistent across therapeutic areas: rodent biology is often too far from human to predict clinical outcomes with confidence, particularly for endpoints that involve tissue architecture, wound biology, or complex behavioral readouts.
Canine models have a long history in toxicology and pharmacokinetics. They remain a standard non-rodent species for general safety assessment and are well-accepted by regulatory agencies for cardiovascular safety pharmacology, particularly hERG and hemodynamic studies.
For efficacy research, however, validated dog models are limited across most therapeutic areas. Behavioral scoring of pain in dogs is possible but less standardized than in pigs. Wound healing studies in dogs are uncommon due to fundamental differences in skin biology (dogs have loose, fur-covered skin that heals differently from human skin). Dogs are most useful when the primary question is systemic safety, PK, or cardiovascular risk rather than efficacy.
NHPs offer the closest neuroanatomical and neurophysiological match to humans. For CNS-penetrant compounds, neuroinflammation targets, immunology programs, or studies where central processing is a key variable, NHP models can provide data that no other species can.
The practical barriers are significant. NHP studies carry lead times of up to a year, per-animal costs that are orders of magnitude higher than other species, and increasing regulatory and ethical scrutiny. The 3Rs framework (Replacement, Reduction, Refinement) is driving the field toward alternative species where comparable translational data can be obtained without primate use.
For peripheral targets and tissue-level endpoints, which represent a large share of the current preclinical pipeline, NHP models are often unnecessary when a well-validated large animal alternative exists.
Porcine models occupy a unique position across multiple therapeutic areas precisely because pig anatomy, physiology, and tissue architecture closely mirror human across organ systems, not just the nervous system.
Pain
Pig skin is structurally and functionally identical to human skin across five key parameters: thickness, hair follicle content, pigmentation, collagen composition, and lipid content. A systematic review of porcine pain models (Meijs et al., 2021, Lab Animal) confirmed near-complete overlap between pig and human peripheral neurophysiology.
Pigs express spontaneous pain behaviors that parallel clinical presentations: guarding, weight shifting, altered gait, reduced social interaction, avoidance. These can be quantified using validated tools such as the distress behavior score, human approach test, and computerized open field analysis. Combined with electrophysiology endpoints (SNCV, cMAP, SNAP) and histological readouts (IENF density, CGRP, NaV1.7 expression), pigs enable multimodal assessment in a single animal.
The translational validation is concrete. Local anesthetic rankings in pig postoperative pain models match human clinical rankings exactly (Castel et al., 2017, J Pain Res). Pig POP studies contributed preclinical evidence for the pH-dependent mechanism underlying ZYNRELEF, subsequently confirmed in a human trial (Ottoboni et al., 2019, Reg Anesth Pain Med).
Wound Healing
Pig skin is the gold standard for wound healing research because it replicates human wound biology at a level no other species can match. Pig models are used for incisional wounds, excisional wounds, burn injuries, and chronic/diabetic wound models (streptozotocin-induced diabetes). Endpoints include wound closure kinetics, granulation tissue formation, scar architecture (Herovici staining), and angiogenesis (CD31 blood vessel quantification). Published data in pig diabetic wound models have demonstrated the efficacy of advanced wound care products across multiple product classes. Aged pig models are also available to mimic impaired healing in elderly populations, a clinically important variable that rodent wound models typically miss.
Cardiovascular Disease
While rodent MI models remain standard for early screening, pig cardiovascular anatomy (coronary artery size, distribution, and physiology) is much closer to human than either rodent or dog. Pig models are increasingly used for device testing, interventional cardiology, and pharmacology studies where cardiac anatomy and hemodynamics need to mirror the human condition. Göttingen minipigs in particular are accepted by FDA and EMA for cardiovascular safety and pharmacology assessments.
Kidney Injury
Rodent cisplatin and ischemia-reperfusion models remain the primary tools for AKI research. However, pig renal anatomy and physiology more closely resemble human, making porcine models relevant for programs targeting renal drug delivery, surgical interventions, or device-based therapies where organ size and vascular architecture matter.
Toxicology and Safety
This is where pigs, particularly Göttingen minipigs, have gained significant regulatory traction. Göttingen minipigs are pathogen-free, genetically defined, and accepted by FDA and EMA as a non-rodent species for GLP toxicology studies. Their slow weight gain allows chronic dosing studies in adult animals, reducing the confounding variable of rapid growth that affects domestic pig studies. For dermal toxicology, the minipig is considered the gold standard due to its skin similarity to human.
One of the practical advantages of building a program around pigs is that efficacy, safety, and toxicology data can come from the same species. Sponsors get a consistent dataset from discovery through IND-enabling studies, without the extrapolation risk that comes from testing efficacy in one animal and safety in another.
The most effective preclinical programs are built on a rational species progression, not a single model. Rodent models for early screening and target engagement. A translational large animal model for clinically predictive efficacy, safety, and PK data.
For peripheral pain targets (postoperative, neuropathic, inflammatory), wound healing, dermal toxicology, and programs requiring human-like tissue architecture, pigs provide the translational fidelity that rodent data alone cannot deliver. For CNS targets or programs requiring primate-specific neuroanatomy, NHPs remain the gold standard. Dogs fill a defined role in cardiovascular safety pharmacology and general toxicology but are rarely the right choice for efficacy assessment.
The species decision should be driven by the biology of your target, the clinical endpoints you need to predict, and whether your program benefits from generating efficacy and safety data in the same translational species. Getting that decision right early saves time, money, and translational risk downstream.
MD Biosciences operates translational pain models across rodent and pig platforms, with integrated behavioral, electrophysiological, and histological endpoints. To discuss species selection for your pain program, contact neuro@mdbiosciences.com.