The most important factor limiting the success of an antiretroviral therapy regimes is toxicity. Toxicity can depend on a number of factors; some of these are intrinsic to the host and may not only affect the latter's outward appearance, but also determine the intensity these toxic effects may reach. The former is exemplified by idiosyncratic or hypersensitivity reactions, whereas the latter is usually appreciated in metabolic disturbances or fat redistribution syndromes. Some of the determinants of antiretroviral toxicity are genetic in origin and have been the subject of intense study in recent years. Some of these are linked to a single nucleotide polymorphism (SNP), whereas others depend on a complex interaction between multiple genes variations. One of these tests (HLA B*5701) is now being applied in clinical practice and widely used to prevent the risk of hypersensitivity reactions to abacavir. Many other genetic determinants of antiretroviral drug toxicity have been suggested as an explanation for nucleoside analogue toxicity; these include lactic acidosis, peripheral neuropathy and pancreatitis, and have also been suggested as a potential basis for the non-nucleoside toxicity derived from immunogenetic factors involved in nevirapine hypersensitivity to SNPs in efavirenz enzyme metabolism, amongst other things. Metabolic toxicity, mainly due to protease inhibitors (PIs) is far more complex and depends on the interaction of various genes. The same seems to be true for fat redistribution syndromes and atherosclerosis, although a clear picture of the genetic factors operating in these syndromes is yet to emerge. The ultimate goal of pharmacogenetics is to customize antiretroviral therapy by identifying the genes that can maximize efficacy whilst helping avoid known side effects of antiretroviral drugs.