Interactions between antiretrovirals and antineoplastic drug therapy

Despite the established impact of highly active antiretroviral therapy (HAART) in reducing HIV-related morbidity and mortality, malignancy remains an important cause of death. Patients who receive the combination of cancer chemotherapy and HAART may achieve better response rates and higher rates of survival than patients who receive antineoplastic therapy alone. However, the likelihood of drug interactions with combined therapy is high, since protease inhibitors (PIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs) are substrates and potent inhibitors or inducers of the cytochrome P450 (CYP) system. Since many antineoplastic drugs are also metabolised by the CYP system, coadministration with HAART could result in either drug accumulation and possible toxicity, or decreased efficacy of one or both classes of drugs. Although formal, prospective pharmacokinetic interaction studies are not available in most instances, it is possible to infer the nature of drug interactions based on the metabolic fates of these agents. Paclitaxel and docetaxel are both metabolised by the CYP system, although differences exist in the nature of the isoenzymes involved. Case reports describing adverse consequences of concomitant taxane-antiretroviral therapy exist. Although other confounding factors may have been present, these cases serve as reminders of the vigilant monitoring necessary when taxanes and HAART are coadministered. Similarly, vinca alkaloids are substrates of CYP3A4 and are, thus, vulnerable to PI- or NNRTI-mediated changes in their pharmacokinetics. Interactions with the alkylating agents cyclophosphamide and ifosfamide are complicated as a result of the involvement of the CYP3A4 and CYP2B6 isoenzymes in both the metabolic activation of these drugs and the generation of potentially neurotoxic metabolites. Existing data regarding the metabolic fate of the anthracyclines doxorubicin and daunorubicin suggest that clinically detrimental interactions would not be expected with coadministered HAART. Commonly used endocrine therapies are largely substrates of the CYP system and may, therefore, be amenable to modulation by concomitant HAART. In addition, tamoxifen itself has been associated with reduced concentrations of both anastrozole and letrozole, raising the concern that similar inducing properties may adversely affect the outcome of PI- or NNRTI-based therapy. Similarly, dexamethasone is both a substrate and concentration-dependent inducer of CYP3A4; enhanced corticosteroid pharmacodynamics may result with CYP3A4 inhibitors, while the efficacy of concomitant HAART may be compromised with prolonged dexamethasone coadministration. Since PIs and NNRTIs may also induce or inhibit the expression of P-glycoprotein, the potential for additional interactions to arise via modulation of this transporter also exists. Further research delineating the combined safety and pharmacokinetics of antiretrovirals and antineoplastic therapy is necessary.