TDM and adherence in HIV patients

teaser

Dominique Breilh
PhD PharmD
Department Head Researcher
Laboratoire de Pharmacocinétique Clinique EA 525
Université Victor Segalen Bordeaux 2
Pharmacie Centrale Hôpital Haut-Lévêque
CHU de Bordeaux Pessac
France
E:[email protected]

Several factors have been shown to determine response of HIV-infected patients to ­antiretroviral therapy. These include:

• Susceptibility of the strain of the virus to treatments.
• Adherence to the planned treatment.
• Effective concentrations in body compartments where HIV replication is active.

Highly active antiretroviral therapy (HAART) has proven effective in adequately controlling replication,(1) and treatment failure has been shown to arise in those with resistant virus,(2) in those less adherent to therapy(3) or in cases where sustained effective concentrations are not achieved.(4–7) Therapeutic drug monitoring (TDM) has led to an increased awareness of concentration–response relationships and the mapping of the pharmacokinetic characteristics of HAART. Many physicians evaluated the effects of TDM on surrogate endpoints such as increased percentage of patients achieving target concentrations. The ultimate goal of TDM is to personalise HAART – in other words, to determine the most effective and safest dosage of drugs.(8) But response and toxicity are likely to be complex, depending on the influence of many genes interacting with environmental and behavioural factors. Interactions between drugs and lack of adherence to treatment are insufficiently acknowledged as causes of therapeutic failure or adverse effects.(9,10)

Implications of nonadherence and exposure–response relationship
The rapid replication and mutation rates of HIV require that adherence to HAART be extremely precise to achieve durable suppression of HIV and prevent resistance.(11,12) Missing even a single dose in a 28-day reporting period has been shown to predict treatment failure.(13) In treating HIV infection, early adherence is essential as it contributes to a good initial response, which can determine long-term success or failure of the regimen.(14–16) Relationships between plasma concentration of protease inhibitors (PIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs) and response to therapy have been the most extensively evaluated.(17,18) Virological responses associated with higher versus lower PI and/or NNRTI plasma or intracellular (PBMCs) concentrations include:

• Presence of undetectable HIV-1 RNA.
• Faster rate of decline in plasma HIV-1 RNA.
• Larger absolute decreases in plasma HIV-1 RNA.
• Larger increases in CD4+ cell count.
• Lower incidence of resistance mutations.
• Longer duration of response.

Achieving target exposure is often a complex matter and, in the case of PIs, it relies on the effective use of ritonavir to boost or enhance pharmacokinetic markers. Ritonavir is a highly potent inhibitor of the CYP3A4 isoenzyme, the use of which results in delayed metabolism of the primary PI.(18,19) The clinician is frequently faced with the question of defining the desired PI concentration for a given patient. Clearly, patients harbouring drug-resistant virus will require higher PI concentrations than those with wild-type virus alone.(18) Variability in pharmacokinetic profiles makes drug concentrations unpredictable, and the greater the variability, the greater the magnitude of the problem.(20) The cause of large interpatient pharmacokinetic ­variability is multifactorial and includes differences in drug absorption, metabolism or distribution and complex drug–drug or drug–food interactions. Variability in drug concentrations in a treated population of HIV patients is inevitable and reflects the different rates at which patients eliminate the same dose of a drug. Another important factor contributing to pharmacokinetic variability is interindividual differences in hepatic drug metabolism. The cytochrome CYP3A4 enzyme shows a wide range of activities in different patients.(21) One of the most ubiquitous and best-studied drug transport proteins is P-glycoprotein (P-gp). Epithelial cells in the intestinal mucosa represent a major site for expression of P-gp. Intestinal P-gp pumps some drugs out of the cell interior and back into the intestinal lumen, thus decreasing the overall absorption of the drug. In the liver, P-gp can pump drug into the bile and facilitate elimination.(22)

Integrating pharmacokinetics and pharmacogenetics into HAART
Two fundamental pharmacological applications to clinical practice have shaped our current therapeutic approach. Ritonavir boosting of PIs and once-daily dosing of nucleoside reverse transcriptase inhibitors (NRTIs) have improved patient outcomes and allowed the simplification of many regimens. The use of ritonavir-enhanced PI regimens has resulted in more durable viral suppression and reduced rates of emergent drug-resistant viral variants. The ability to dose many of the NRTI agents once daily derives directly from the availability of assays to measure NRTI-triphosphate levels in the intracellular compartment. Intracellular rather than plasma drug exposures have provided the rationale for this change. Future advances in molecular biology, including a greater understanding of drug-metabolising enzyme function and the role of drug ­transporters, will further shape our therapeutic approaches. Pharmacogenetic-oriented TDM allows us to go beyond the pharmacokinetic interpretation. In future, pharmacogenetic information could be applied a priori for initial dose stratification and identification of cases where certain drugs would be ineffective. Pharmacogenomics is a powerful tool to investigate variable responses to antiretroviral therapeutics. HIV management is characterised by differing response rates and adverse effects. To date, few agents appear to have a clear and causal genotype–phenotype correlation. Such correlations have, however, been demonstrated for abacavir hypersensitivity and atazanavir hyperbilirubinaemia, and more work is needed to explore the relationship between CYP enzyme polymorphisms, particularly at CYP3A5 and CYP2B6, and subsequent pharmacokinetic and pharmacodynamic outcomes.(23–27) The major candidate genes should be well documented as being functionally relevant, and should cover all aspects of the pharmacology and toxicology (receptors, enzymes, transporters and immunomodulators).(28) Demographic and environmental factors (eg, drug interactions) should also be considered as covariates. Drug exposure and adherence to treatment should be assessed rigorously to reduce noise in the data.

Conclusion
HAART is complex and the management of HIV/AIDS
is challenging, yet many patients succeed in adequately controlling it. The routine use of TDM (beyond its use as a monitoring tool for recent adherence) will depend on the accumulation and modelling of drug level data from large cohorts of patients at various stages of the disease. The primary goal of therapy for all HIV-infected patients should always be complete viral suppression. If this is not possible, maintaining immunological function and preventing clinical deterioration is the most important goal. Boosted PIs are a key component of regimens administered to drug- experienced patients. Adherence is crucial to the success of antiretroviral therapy. Patients must be aware that, once drug resistance develops, treatment options become limited, as does the ability of the provider to make the HAART regimen lifestyle-friendly. A positive relationship between patient and physician is likely to enhance adherence, as will discussing side-effects and their management proactively. The physician will then be able to dose-adjust confidently, in order to maximise antiretroviral effect in patients with suboptimal initial responses to therapy or to minimise toxicity while still maintaining effective viral inhibition. In this respect, consultations with virology and pharmacology experts could lead to optimal use.

Given that genetic factors need to be put into perspective, the challenge is now to carry out large, prospective, multidisciplinary, multicentre projects to assess the real clinical, and perhaps economic, value of predictive genetic testing in antiretroviral therapy.

References

1. N Engl J Med 1998;338:853-60.
2. J Am Med Assoc 2000;283:229-34.
3. Ann Itern Med 2000;133:21-30.
4. AIDS 2001;15 71-5.
5. Ther Drug Monit 2001;23:332-40.
6. AIDS 2002;16:1331-40.
7. AIDS 2004;18:1305-10.
8. J Clin Virol 2003;26:223-7.
9. BMJ 2000;320 987-90.
10. Clin Pharmacokinet 2001;40:783-802.
11. AIDS Patient Care 2003;17:170-7.
12. JAMA 1999;282:1135-41.
13. N Engl J Med 1998;339:307-11.
14. JAMA 1998;279:930-7.
15. AIDS 2001;15:47-54.
16. Clin Infect Dis 2001;33:466-72.
17. Antivir Ther 1999;3:29-41.
18. AIDS Read 2001;11:87-98.
19. Clin Pharmacokinet 2005;44:279-304.
20. AIDS 1997;11:29-33.
21. Eur J Clin Pharmacol 2005;60:865-73.
22. Top HIV Med 2003;11:136-9.
23. J Infect Dis 2005;192:1381-6.
24. Drug Metab Dispos 2005;33:1729-39.
25. Pharmacogenet Genomics 2005;15:1-5.
26. AIDS 2004;18:2391-400.
27. Eur J Med Res 2003;8:531-4.
28. Top HIV Med 2005;13:90-5.