Animal models have misled scientists in the past and this has resulted in human deaths.
Penicillin stayed on the shelf for over a decade because the rabbits Fleming tested it on led him to believe it would be ineffective in humans. Scientists were misled about how HIV enters the human cell because of studies on monkeys. The polio vaccine was delayed by decades because the way monkeys responded turned out to be very different from the way humans reacted. The cardiopulmonary bypass machine killed the first patients it was used on and it was only after human data was used that the machine was made safe. Studying strokes and brain hemorrhage in animals has led to multiple medical treatments that worked in animals but that resulted in harm to human patients. HIV vaccines that protected monkeys have actually increased the risk of contracting HIV in the volunteers that took the vaccine. The flip side of all this is the fact that society has also lost cures and treatments because scientists believed the results from animals: the National Cancer Institute has said that we have lost cures for cancer because studies in rodents have been believed.
Animals simply do not react the same as humans in a reliable manner. More examples:
Cancers in mice have been cured but the cures did not work in humans. Humans respond to tobacco and asbestos by suffering from cancer while most animals do not. Smoking leads to heart disease in humans not animals. High chol leads to heart disease in humans not animals. Babies of mothers who took thalidomide in the late 1950s early 1960s suffered severe birth defects but animals for the most part did not. Rabbits reacted to the penicillin PCN administered by Fleming in 1929 very differently than humans. HRT was administered to women based on animal studies. Every drug that kills people tested safe on animals. (Melanoma in dogs is malignant in nailbed, eye, and mouth.)
Plavix is an anticlotting drugs that is not effective in some people. Plavix is converted by a CYP enzyme in the liver to another chemical that actually does the work of preventing blood clots. If the patient has 2 copies of a variant of the gene coding for this particular CYP enzyme then the drug will not be converted into the active chemical and about 14% of Chinese patients have this variant. However, even patients that have only 1 copy of the variant can also be affected.
Other drugs that are metabolized or processed differently in some way by the body include Iressa, methotrexate, 6-mercaptopurine, codeine, tamoxifen, warfarin aka Coumadin, and succinylcholine. We also know that drugs like 5-FU-based chemotherapies, aspirin, and opiates or other drugs that act on Kappa receptors have effects that vary between the sexes. All of this at least in part explains why 90% of drugs work in only 30 to 50% of the people.
Physicians have realized for decades that people respond differently to drugs. Because of the Human Genome Project and various spinoffs we now have a lot more data about this.
Men are affected differently than women by diseases like cardiovascular diseases and myocardial infarction. But there are a lot more examples.
Caucasians and African-Americans have a similar prevalence of early age-related macular degeneration. However, the progression to the late form of this disease is very rare for African-Americans while being common in Caucasians. Similarly, infantile hemangiomas of the skin are commonly seen in Caucasians but are rare in African-Americans. Certain breast cancers are less common in young black women but usually much more lethal than in young white women even when socioeconomic factors are taken into account. Among cigarette smokers, African Americans and Native Hawaiians are more susceptible to lung cancer than whites, Japanese Americans, and Latinos.
Breast cancer is a good example of a disease that is now treated based on the genotype of the patient and the tumor. It seems like every week I see another study that links a gene to a disease.