Drug transporters, which can pump molecules out of cells or potential sanctuary sites such as the central nervous system, have gained increasing attention as potential mechanisms of drug failure of HAART regimens. The Drug Transporter session at the 8th Conference on Retroviruses and Opportunistic Infections (8th CROI) brought together four of the leading experts in drug transporters and HIV disease to discuss the potential clinical implications of these molecules.

Dr. Richard Kim of Vanderbilt University discussed the current understanding of P–glycoprotein (P–gp), a protein encoded by the MDR–1 gene (Abstract S1) [1]. P–glycoprotein is a transmembrane protein which transports molecules such as digoxin and HIV protease inhibitors out of cells and certain sites in the body, such as the central nervous system (CNS) and the testes. P–glycoprotein uses ATP for energy and is a member of the ABC transporter family. It is inhibited by a number of drugs such as ketoconazole, quinidine, verapamil and cyclosporin (many of which also block the hepatic P450 CYP 3A enzyme). P–glycoprotein is known to transport indinavir, nelfinavir, saquinavir, and amprenavir, and is inhibited by nelfinavir, saquinavir and ritonavir. Some drugs, such as rifampin and St John's Wort, can induce the enzyme as well. In addition, there is significant genetic variability of P–gp in patients. It is not known what impact P–gp levels have on the activity of PI in individual patients at this time.

Of interest, P–gp, acting as part of the blood brain barrier, transports drugs out of the CNS. This can result in low levels of HIV protease inhibitors in the CNS. In an MDR–1 knockout mouse model, there was no effect on plasma PI drug levels but significant increases in CNS levels. Newer P–gp inhibitors have been developed which selectively block P–gp but have little impact on the hepatic P450 system (for example, Lilly's LY335979). These drugs offer the possibility of obtaining much higher CNS levels of PIs without increasing the doses given to HIV–infected patients with AIDS dementia or cognitive difficulties.

Dr. Arnold Fridland of Gilead Sciences discussed a second class of drug transporters, the MRP family, which can transport nucleoside drugs like AZT and ddI (Abstract S2) [2]. Six MRP variants have been detected, of which MRP–4 and MRP–5 appear to transport nucleoside analogs. The role of a drug transporters in nucleoside drug resistance was first suspected when CEM cells exposed to AZT developed a multi–nucleoside drug resistance profile while still growing drug sensitive virus. The cells were found to pump out the nucleoside monophosphate form of the drugs (AZT–MP, PMEA) in an ATP–dependent process. Subsequently, the drug transport was found to be caused by increased expression of MRP–4. The MRP transporters can be blocked by dipyridamole and prostaglandin A1. The search for specific MRP inhibitors is currently in progress. Notably, PMPA is much less affected by MRP than PMEA in vitro. The role of MRP in nucleoside drug failure remains unknown.

David Back discussed the clinical relevance of drug transporters (Abstract S3) [3]. A major concern was that P–gp, especially if overexpressed, could lead to suboptimal PI drug levels in the brain or individual cells. In this regard, P–glycoprotein levels in untreated HIV–infected patients were inversely related to viral load (higher viral load, lower P–gp levels). Effective antiretroviral therapy returned P–gp levels to normal, however. Different polymorphisms of P–gp in patients were also found to have a significant effect on drug levels of both PI and NNRTI agents.

Having made these observations, Dr. Back proceeded to present data from oncology studies which showed that remission rates in cancer patients were lower in patients whose blast cells expressed increased P–gp levels. With respect to this, the newer generation of P–gp inhibitors is being aggressively developed by the oncologists to combat drug resistance and increase CNS penetration. Of note, some drugs which cross the blood–brain barrier poorly show increased CNS toxicity in the presence of P–gp inhibitors. The impact of drug transporters on the success of HAART and the utility of these agents in targeting HIV sanctuary sites remain to be determined. Dr. Back suggested that patient profiling using DNA arrays to evaluate polymorphisms in genes for P–450 and P–gp may lead to the use of pharamacogenetics to individualize drugs and dosing for each patient.

Lastly, Dr. Charles Flexner from Johns Hopkins University presented data relating drug transporter expression to HIV replication (Abstract S4) [4]. In vitro, increased P–gp levels were associated with decreased HIV expression which was corrected by adding verapamil. The mechanism for this effect was unclear, particularly since elevated MRP–1 levels were correlated with increased HIV expression (an effect which was partially reversed with the addition of an MRP–inhibitor). This presentation lent a cautionary note that the net impact of altering P–gp and MRP activities in individual patients is unknown and that the newer generation of P–gp inhibitors should only be given in the context of controlled clinical trials.

In conclusion, the potential role of drug transporters in the loss of activity of individual drugs remains uncertain but very intriguing (especially for drugs that fail with low levels of resistance such as ddI, d4T and ddC). P–glycoprotein may extrude PI from sanctuary sites like the brain and testes. Newer generation P–gp inhibitors are actively being investigated in cancer trials and MRP inhibitors are currently being sought. Clinical data pertaining to the impact of drug transporters on clinical outcomes and the use of drug transporter inhibitors are likely to emerge quickly over the next year. HIVresistanceWeb will report on these events as they develop.

References