Toxicity of antiretroviral nucleoside and nucleotide analogues: is mitochondrial toxicity the only mechanism?

G Moyle - Drug safety, 2000 - Springer
G Moyle
Drug safety, 2000Springer
Nucleoside analogues represent the cornerstones of antiretroviral regimens. A range of drug-
or tissue-specific toxicities, such as peripheral neuropathy, myopathy, pancreatitis and lactic
acidosis with hepatic steatosis, has been documented with these agents. The fat atrophy
seen on long term antiretroviral therapy may also be related to nucleoside analogues. The
mechanisms by which nucleoside analogues cause toxicity are not clearly established. In
vitro, the triphosphates of these agents are weak to modest substrates for human DNA …
Abstract
Nucleoside analogues represent the cornerstones of antiretroviral regimens. A range of drug- or tissue-specific toxicities, such as peripheral neuropathy, myopathy, pancreatitis and lactic acidosis with hepatic steatosis, has been documented with these agents. The fat atrophy seen on long term antiretroviral therapy may also be related to nucleoside analogues.
The mechanisms by which nucleoside analogues cause toxicity are not clearly established. In vitro, the triphosphates of these agents are weak to modest substrates for human DNA polymerases, showing the greatest affinity for mitochondrial DNA polymerase γ. Short term exposure in vitro to some nucleoside analogues has been demonstrated to cause increased lactate production or falls in mitochondrial DNA suggestive of mitochondrial toxicity. However, stavudine and to a lesser extent zidovudine are poor substrates for mitochondrial thymidine kinase type 2, the predominant form in cells that are not actively mitotic such as neurons, myocytes and adipocytes. These are the cell types where the proposed mitochondrial toxicities neuropathy, myopathy and lipoatrophy are observed. Thus, active concentrations of phosphorylated products of stavudine and zidovudine may not be present in mitochondria.
The familial mitochondrial diseases do not have identical presentations to nucleoside analogue toxicities. These disorders most commonly involve the CNS, typically with seizures or dementia, and occasionally the kidneys. Although nucleoside analogues are known to penetrate the CNS and are commonly renally excreted unchanged, mitochondrial toxicities at these sites have not been documented.
Furthermore, toxicity caused by nucleoside or nucleotide analogues does not always appear to arise through the mitochondrial route. Cidofovir appears to cause renal tubular dysfunction via a toxic intracellular metabolite, and zidovudine-related anaemia appears to be related to decreased globin RNA synthesis. In vitro or animal models suggest that zidovudine myopathy, stavudine-related (but not zalcitabine- or didanosine-related) neuropathy and didanosine-related pancreatitis may all be not related, or not exclusively related, to mitochondrial dysfunction.
The integration of nucleoside analogues into nuclear DNA, best documented with zidovudine but likely to occur with other agents, represents an alternative but potentially delayed pathway to cytotoxicity and cell apoptosis. This is the mechanism of cell death during therapy with antineoplastic nucleoside analogues, and may have contributed to the multisystem toxicities observed with the anti-hepatitis B drug fialuridine. New research evaluating the effects of long term exposure of cell lines is required to address the possibility that nuclear genotoxicity plays a role in long term nucleoside analogue toxicity.
Springer