Animals used in testing undergo a wide range of procedures depending on the type of research, from single-dose chemical exposures to surgical operations to tumor implantation. Most test animals are mice and rats, though rabbits, dogs, primates, and fish are also used. The experience varies enormously by study type, but in the majority of cases, the animals are euthanized at the end of the experiment so their tissues can be examined.
Toxicity Testing
One of the most common categories of animal testing involves determining whether a substance is poisonous and at what dose. In acute toxicity tests, each animal receives a single, relatively high dose of a chemical, either by mouth, through the skin, or by inhalation. The animals are then watched for up to 14 days for signs of illness or recovery. Researchers record symptoms like tremors, labored breathing, loss of appetite, lethargy, or organ failure. At the end of the observation period, the animals are killed and their organs are examined under a microscope.
The most well-known version of this is the LD50 test, which aims to find the dose that kills 50% of the animals in a group. A starting dose of 5 grams per kilogram of body weight is typically administered. If animals die at that level, lower doses are tried on additional groups, often around five animals each, until researchers can estimate the lethal threshold. While regulatory agencies now emphasize observing symptoms and recovery rather than simply counting deaths, animals in these studies still experience the full toxic effects of the substance being tested.
Eye and Skin Irritation Tests
The Draize test, developed in the 1940s and still used in modified form, involves applying a chemical directly to the eye of an albino rabbit. The rabbit’s eye is observed for up to 21 days to track swelling, redness, discharge, ulceration, or blindness. Substances classified as serious irritants cause irreversible eye damage that persists through the entire 21-day observation period. Rabbits are used because they produce fewer tears than other animals, which means the chemical stays in contact with the eye longer. A parallel version of the test applies chemicals to shaved patches of skin to measure burning, rash, or tissue destruction.
Cancer Research Models
In cancer studies, mice are the primary subjects. Researchers implant human tumor cells into mice whose immune systems have been suppressed or genetically altered so their bodies won’t reject the foreign tissue. These are called xenograft models. The tumors are allowed to grow over several weeks, measured externally with calipers at regular intervals. A tumor larger than 100 cubic millimeters is typically scored as a positive result.
Different mouse strains respond differently to the same cancer cells. The same implanted tissue can produce vastly different tumor sizes, immune responses, and growth patterns depending on the genetic background of the mouse. As tumors grow, mice can experience weight loss, reduced mobility, visible masses under the skin, and organ disruption if the cancer spreads internally. Studies define “humane endpoints,” size or health thresholds at which the animal is euthanized rather than allowed to deteriorate further, though the specific cutoffs vary by institution.
Surgical and Neurological Research
Some of the most invasive procedures involve primates and are designed to study conditions like spinal cord injury, brain function, or neurological disease. In a spinal cord injury study, for example, a primate is placed under general anesthesia, and surgeons cut through skin, muscle, and bone to expose the spinal cord. Small incisions are then made directly into the cord at precise intervals to simulate an injury. In one documented approach, cuts were made at 3-millimeter intervals, 3 millimeters wide and about 4 millimeters deep. After surgery, the animal’s recovery and movement are tracked over weeks or months.
Behavioral testing often follows these surgeries. Primates recovering from spinal cord procedures may be placed in narrow corridors that force them to walk in specific ways so researchers can score their limb function on standardized scales. These corridors constrain the animal’s natural movement to isolate hindlimb activity from the compensating use of their arms. Electrode implantation in the brain is used in other neurological studies, where primates may be fitted with head restraints or recording chambers that remain in place for the duration of the research.
What Happens After the Study Ends
The vast majority of laboratory animals are euthanized when a study concludes. This is partly because their tissues need to be examined for internal damage, tumor growth, or organ changes that aren’t visible from the outside. The American Veterinary Medical Association recognizes a long list of approved methods. For lab rodents, the most common approaches include injectable sedatives that cause unconsciousness followed by death, carbon dioxide exposure, and physical methods like cervical dislocation for small mice and rats. The method used depends on the species, the size of the animal, and what tissue samples researchers need to preserve.
Rehoming is possible but far less common. The NIH supports institutions that choose to adopt out research animals after studies end, but the decision rests entirely with each facility. Successful programs must evaluate whether the animal is healthy enough, whether it can be socialized to live in a home, and whether the adoption meets federal and state laws. Dogs, particularly beagles (the breed most commonly used in pharmaceutical testing), and retired chimpanzees are the animals most frequently placed in adoption programs or sanctuaries. For the millions of mice and rats used each year, rehoming is essentially nonexistent.
Oversight and Legal Protections
In the United States, the Animal Welfare Act requires every research facility to maintain an Institutional Animal Care and Use Committee. This committee inspects all animal facilities at least every six months, reviews every research protocol involving animals, and investigates complaints from the public or lab staff. If deficiencies are found, the committee must produce a specific plan with deadlines for correcting them. It also has the authority to suspend any ongoing animal study.
The guiding ethical framework in animal research is known as the 3Rs: Replacement, Reduction, and Refinement. Replacement means using alternatives whenever possible, such as human cell cultures, computer models, or organ-on-a-chip systems that mimic human tissue. Reduction involves designing experiments to use fewer animals while still producing valid data, through better statistical methods or sharing animals between studies. Refinement focuses on minimizing suffering in the animals that are used, through pain management, environmental enrichment like toys or social housing, and humane endpoints that prevent prolonged suffering.
Shifting Regulatory Requirements
For decades, the FDA required animal testing data before any new drug could enter human trials. That changed with the FDA Modernization Act 2.0 and a 2025 announcement that the agency would begin phasing out animal testing requirements for certain drug categories, starting with monoclonal antibodies. The FDA now encourages companies to submit safety data from non-animal methods, including AI-based toxicity models, lab-grown human cell systems, and real-world safety data from countries where a drug has already been tested in people.
Companies that submit strong non-animal safety data may receive streamlined review, creating a financial incentive to move away from animal models. The FDA estimates that thousands of animals, including dogs and primates, could be spared each year as these alternative methods become standard. Implementation has already begun for new drug applications, though a full transition across all drug categories will take years.