 |
|
|
| |
Clinical Pharmacology at CROI 2008: Drug Delivery and Disposition, Drug Interactions, Pharmacogenetics and Applications to Patient Care
|
| |
| |
Courtney V. Fletcher, Pharm.D.
Dean and Professor
College of Pharmacy
University of Nebraska Medical Center
986000 Nebraska Medical Center
Omaha, NE 68198
cfletcher@unmc.edu
The 15th Conference on Retroviruses and Opportunistic Infections (CROI) was held in Boston, MA from February 3-6, 2008. CROI continues to be the premier HIV-focused scientific meeting. The 15th CROI had many abstracts related to clinical pharmacology. In this report I will highlight those I think are of broad interest or might benefit from some expert clarification. I will discuss abstracts in four broad categories: drug delivery and disposition, drug interactions, pharmacogenetics, and the application of pharmacologic information to patient care.
Abbreviations
ACTG, adult AIDS clinical trials group
ARV, antiretroviral drug
ATV, atazanavir
AUC, area under the concentration-time curve
Cmin, minimum drug concentration
CNS, central nervous system
CSF, cerebrospinal fluid
DRV, darunavir
EFV, efavirenz
f-APV, fosamprenavir
IC50, concentration of drug required to inhibit viral replication in vitro by 50%
IM, intramuscular
IQ, inhibitory quotient
LPV, lopinavir
MVC, maraviroc
NRTI, nucleoside reverse transcriptase inhibitor
NNRTI, non-nucleoside reverse transcriptase inhibitor
PACTG, pediatric AIDS clinical trials group
PBMCs, peripheral blood mononuclear cells
PI, inhibitor of HIV protease
r or RTV, ritonavir
SC, subcutaneous
TDM, therapeutic drug monitoring
TFV, tenofovir
TPV, tipranavir
I. Drug Delivery and Disposition
This category of abstracts is concerned with the delivery of drug into the body, whether by oral administration or by some other route, such as intramuscular administration. Drug disposition refers to what happens to the drug, such as what spaces in the body the drug penetrates into and the processes responsible for removing drug from the body.
Drug delivery research holds the promise of improving the way drugs can be given to patients or frequency of drug administration. One recent example of improved drug delivery is the new meltrex formulation of LPV/r, which overcame the need for refrigeration, reduced the number of doses that needed to be taken and improved gastrointestinal tolerance. Studies of drug disposition are important to an understanding of whether a drug penetrates into the spaces that are desired, such as the CNS or the genital tract, and the rate and mechanisms of drug removal from the body, such as via the kidneys or liver, which informs as to the frequency of drug administration and implications for dosing in patients with reduced organ function.
Abstract 134, discussed the intramuscular (IM) or subcutaneous (SC) administration of a depot (or prolonged release) formulation of the investigational NNRTI TMC278. These investigators demonstrated that the release of TMC278 from a nanosuspension (nano=very small) formulation gave a prolonged release over 3 to 6 months in rats and dogs. In 48 human volunteers, this formulation administered SC or IM had an elimination half-life of approximately 5 weeks and declined to 10 ng/mL in 12-26 weeks. 10 ng/mL is the threshold, or lowest concentration in the body that these investigators believed would still be associated with a therapeutic effect. The IM formulation was better tolerated than the SC. Development of this product will require a considerable amount of additional work - but I found this work to be exciting in its implications for reducing the burden, frequency and variability associated with oral drug administration.
Abstract 131 presented data on 117 participants in a study of the effects of ARV therapy on HIV-associated neurological disease that included measurement of the ARV drug in plasma and CSF. The focus of this abstract was on the penetration of tenofovir into the CSF. As expected, very low concentrations of tenofovir were found in the CSF: the average CSF concentration was 96 ng/mL, which was approximately 4% of plasma concentrations. This concentration is less than the IC50 of wild type virus to tenofovir of approximately 201 ng/mL and suggests that tenofovir may not contribute to the inhibition of HIV replication in the CSF because its concentrations in this compartment are too low. This information should be a consideration for physicians constructing ARV regimens for patients who need treatment for HIV-associated neurologic disease.
Maraviroc (MVC) pharmacokinetics in blood plasma, and in genital tract fluid and tissue from 12 female volunteers were presented in abstract 135LB. In these women, MVC was detectable in cervicovaginal fluid with concentrations reaching or exceeding plasma concentrations. The ratio for the AUC in cervicovaginal fluid to plasma was 3.9 on study day 7. Also on day 7, MVC was detectable in vaginal tissue biopsy samples with a ratio of 1.9 to that in plasma. This was a very well done study and demonstrated that MVC is present in cervicovaginal fluid and vaginal tissue in higher concentrations than in plasma, and provides a pharmacologic basis for MVC to be studied for HIV prophylaxis.
NRTIs remain the backbone of virtually all ARV regimens. NRTIs, like all ARVs, have to penetrate into their site of action (in this case the HIV-infected cell) but unlike other ARVs, must be phosphorylated to their active form. The contribution of pharmacologic characteristics such as influx, efflux and phosphorylation for NRTIs in terms of the amount of drug in reservoir compartments is unknown. Abstract 754 investigated these issues by measuring the concentrations of tenofovir-diphosphate (TFV-DP, the pharmacologically-active drug) in PBMCs and in lymphocytes from lymph node tissue in HIV-infected persons. In 7 individuals, the median TFV-DP in lymph nodes was 16 (range, 6 to 41) fmol/106 cells, and the median ratio of TFV-DP concentrations in lymph nodes to PBMCs ratio was 0.55. This means that concentrations in lymph nodes were half of those in the PBMCs. What are the potential implications of these observations? First, these data demonstrate that TFV-DP (the active form of TFV) was present in lymphocytes from lymph node tissue. This tells us that active drug is present in this compartment, and that is important news. Second, the observation that concentrations in the lymph nodes were less than in PBMCs raises the question as to whether the amount that gets in is sufficient to inhibit viral replication in this compartment. This latter question is critical, and answering it would tell us whether HIV replication persists in reservoir sites, despite suppression of replication in blood, because there is inadequate concentration of drug in this compartment. This is an important area for more work.
The disposition of ARVs in pregnancy has been reported at several past CROIs. Previously, it was shown that LPV/r concentrations in pregnant women receiving the standard dose of 400/100 mg twice daily of the capsule formulation had significantly reduced LPV concentrations in the 3rd trimester. Abstract 629 reported the results of study 1026s conducted by the Pediatric AIDS Clinical Trials Group (PACTG) that examined concentrations of LPV/r in pregnant women taking a higher LPV/r dose of 600/150 twice daily of the tablet formulation in the 3rd trimester. LPV concentrations in the 3rd trimester as a result of this higher dose met the specified desired value for AUC in 19 of 21 women; concentrations, however, were still lower than values postpartum when women had returned to the standard 400/100 twice-daily doses. These data provide support that the higher dose of LPV/r achieved the target concentrations and should be recommended for pregnant women in the 3rd trimester. This study also examined LPV/r concentrations with the tablet dose of 400/100 twice daily in pregnant women in the 2nd trimester. Pharmacokinetic data were only presented on 6 women. LPV concentrations in these women were approximately half of those in women taking the same dose postpartum. Thus, these data raise the question as to when is the appropriate time in pregnancy to increase the dose of LPV/r. They may indicate that a higher dose should be given in the 2nd trimester, but further research into this question is needed.
Very consistent with the LPV/r story in pregnant women, Abstract 624 presented the pharmacokinetics of ATV in pregnant women in the 3rd trimester and then postpartum. ATV was dosed at 300 mg plus 100 mg of RTV both given once daily. The AUC and Cmin of ATV were lower than historical controls and in these women postpartum. Because the AUC value was less than the protocol-specified desired value, an increased dose of ATV to 400/100 once daily is now being evaluated.
II. Drug Interactions
The study of interactions among ARV drugs, and among ARVs and other commonly used drugs remains an area of considerable study. Primarily, these interactions arise because of the concomitant use of drugs that are substrates, inhibitors and inducers of drug metabolizing enzymes, especially those of the cytochrome P450 system (CYP).
Efavirenz (EFV) is a well-known inducer of CYP, and when it is coadministered with many other drugs metabolized by CYP a reduction in concentrations of the other agents results. Such is the case when EFV is co-administered with LPV; LPV concentrations are reduced. Previously this interaction had been managed by increasing the LPV dose from 400/100 mg twice daily to 533/133 mg twice daily, of the capsule formulation. The replacement of the capsule formulation with the meltrex tablet formulation in the strength of 200/50 mg per tablet precluded the use of this dose increase approach. Clinicians were then left with the option of not increasing the LPV/r dose, or increasing to 600/150 mg twice daily, which would result in LPV concentrations approximately 35% higher than those of the standard dose without EFV. Abstract 765 reported the results of a drug-drug interaction study with LPV/r and EFV, where the standard dose of LPV was compared with an increased dose of LPV/r to 500/125 mg twice daily given with EFV. This dose was made possible by the recent introduction of a new strength of the LPV/r meltrex tablet of 100/25 mg of each. This study demonstrated that the 500/125 dose of LPV/r when given with EFV approximated the concentration of LPV in the standard dose without EFV. The good news for clinicians using LPV and EFV in combination is that a dose increase of LPV/r to 500/125 twice daily looks to be adequate to manage this drug-drug interaction. For patients, however, this approach will be problematic in that it will require prescriptions for two different strengths of the LPV meltrex tablets, the 200/50 and the 100/25.
At the 14th CROI last year the surprising interaction between LPV/r and rosuvastatin was reported, which showed a 2 to 4-fold increase in rosuvastatin concentrations in plasma but evidence for attenuation of the cholesterol and triglyceride lowering properties of rosuvastatin. At CROI this year, the results of an interaction study between rosuvastatin and TPV/RTV were reported in Abstract 767. Rosuvastatin AUC was increased 37% in the presence of TPV/RTV, which is less of an increase seen with LPV/r. For patient management, if rosuvastatin must be used with TPV/RTV or LPV/r, it should be done so very cautiously, starting at the lowest rosuvastatin dose (5 mg/day) and titrating to response with very close monitoring for safety and tolerance.
There remains a real unmet need for effective strategies for concomitant treatment of TB and HIV, and in particular whether rifampin can be given with any ritonavir-boosted PI. This latter question needs to be addressed in healthy volunteer studies because of the risk of suboptimal PI concentrations as a result of the potent CYP inducing properties of rifampin. These studies have been complicated by the significant hepatotoxicity that has been seen in healthy volunteers receiving rifampin and PIs. Once again this hepatotoxicity was seen as presented in Abstract 766b. Investigators from the ACTG had previously shown that when rifampin was given with ATV (non RTV boosted) that adequate ATV concentrations could not be maintained, but that the combination was well tolerated. This year, this same protocol team evaluated the effects of rifampin on ATV/RTV. When ATV/RTV was added to rifampin in healthy volunteers, within 12 hours all developed nausea and vomiting and significant elevation of hepatic transaminases. After only 3 of the 14 planned subjects were enrolled, this study was stopped because all developed significant toxicity. The mechanism(s) for the development of severe hepatotoxicity with the combination of rifampin plus ritonavir-boosted PIs remains unclear, as does exactly how to proceed to determine whether such a combination is possible to be used in TB and HIV co-infected persons.
Malaria infection results in 1.5 to 2.7 million deaths annually. Similar to the discussion above regarding the need to determine if anti-TB agents can be administered with HIV PI, the same is true for anti-malarial agents. Abstract 132 evaluated the pharmacokinetics of the combination of Coartem (artemether/lumefantrine) and LPV/r and represent an important effort to address therapy for the patient co-infected with malaria and HIV. The AUC of lumefantrine was almost doubled when coadministered with LPV/r, most likely as a result of inhibition of CYP metabolism. These authors concluded that because of the safety profile of lumefantrine, a dose adjustment was not necessary. However, both lumefantrine and artemether are metabolized via CYP. I believe it is most prudent to hold off on any dose recommendations until artemether concentration data are available and pilot studies with the combination are done in patients vs. healthy volunteers.
III. Pharmacogenetics
CROI has provided an important venue for information related to the contribution of naturally occurring genetic variability to the variability seen in the pharmacokinetics of ARVs among patients and the variability seen in the response (pharmacodynamics) to ARV therapy.
This year, I thought the most important pharmacogenetic abstract was #133, because of what it taught us about the relationship between pharmacogenetics and pharmacokinetics. Efavirenz (EFV) is known to be primarily metabolized by a specific CYP enzyme known as CYP2B6. It is also known that natural variations occur that result, in simple terms, in certain patients being poorer metabolizers of EFV, and thus having higher concentrations in plasma, or better metabolizers and thus having lower EFV concentrations in plasma. These investigators examined what happens to EFV metabolism when CYP2B6 is impaired: do other metabolic pathways contribute to its metabolism, and is genetic variability in these other metabolic pathways important? The answer to both questions was yes. It was shown that patients with a genotype for impaired CYP2B6 EFV metabolism who had extremely high EFV concentrations also had an impaired ability to metabolize EFV via two other CYP pathways (3A4 and 2A6). What this abstracts helps us to appreciate is that if one metabolic pathway is impaired, a secondary or accessory pathway may fill in, and if these secondary pathways are impaired, such as by naturally occurring genetic variability as shown in this work, it can result in these patients having very high drug concentrations. This work then helps to advance our understanding of why patients have the drug concentrations that they have.
IV. Application of Pharmacologic Information to Patient Care
This final category of abstracts is concerned with the application of pharmacologic information - that is pharmacokinetic, pharmacodynamic or pharmacogenetic data to the use of ARV in patient care. Therapeutic drug monitoring (TDM) is one example, where pharmacokinetic data are used to guide the dosing of a drug in a patient in order to maximize the probability of the desired response and minimize the risk of adverse drug reactions.
Abstract 35 presented the results of ACTG A5146, a prospective, randomized, controlled, open-label trial evaluating the effect of TDM and PI dose escalation on viral load (VL) responses in ARV-experienced patients. 92 pts were randomized to TDM and 91 to standard of care (SOC). Median trough conc increased more in TDM vs SOC for all PIs with the exception of fosamprenavir (f-APV). For f-APV, concentrations in the TDM arm were lower than in the SOC arm despite dose adjustment. Overall, TDM and SOC did not differ in the primary endpoint, which was VL change between randomization and 20 wks later, or other outcome measures, or toxicities. Hispanic and black pts (49% of subjects), however, did benefit from TDM (p=0.035, 0.05, respectively). Given this, the lack of an overall effect is not understood, but the inability to increase f-APV concentrations in the TDM arm may have contributed to this, especially given that nearly half of the patients in the TDM arm were receiving f-APV.
So, what are the implications of this study for TDM - is it dead? First, let me mention three other abstracts at CROI-2008 that speak to the potential value of TDM in patient care.
Abstract 770 used TDM to guide the switch from ATV/RTV to unboosted ATV in 56 patients with HIV-RNA < 50 copies/mL. All patients were complaining of ATV and/or RTV related side effects. Among other criteria necessary for the switch, a Cmin concentration of ATV ≥ 150 ng/mL was documented. After the switch to ATV, while the median Cmin declined (as expected with the discontinuation of RTV), Cmin was still above the threshold value in all but 4 patients, three of who were taking tenofovir, which is known to lower ATV concentrations. In these patients virologic suppression was maintain in all but one. The rate of hyperbilirubinemia was 52% for the combination of ATV/RTV and declined to 12% after the switch to ATV.
Abstract 768 evaluated the inhibitory quotient (IQ) as a predictor of virologic response to darunavir (DRV)-based salvage regimen. 32 patients received DRV/RTV, 600/100 twice daily as part of a salvage regimen. The IQ for DRV was calculated as the ratio of the viral genotype and phenotype to the measured DRV Cmin. After 24 weeks of therapy, the DRV IQ, whether based on viral genotype or phenotype, was the only independent predictor of virologic success. Patients who had a DRV IQ based on viral phenotype that was ≥ 1.5 had a 8-fold greater chance of virologic response to < 50 copies/mL than those with an IQ less than this value.
Abstract 772 investigated the association between viral blips (intermittent plasma viremia) and the plasma concentrations of ARVs in a total of 273 HIV-infected persons (85 cases with blips and 188 controls without). Subtherapeutic drug concentrations were found in 49% of those subjects who had viral blips compared with 17% of those who did not (P<0.01), and this difference was seen for all ARV drugs.
In summary then, these three abstracts showed the value of TDM to simplify an ARV regimen and manage drug-related side effects, while preserving virologic suppression (#770); demonstrated that knowing how much drug was in the body (Cmin) compared with a measure of virus susceptibility (genotype or phenotype) was an independent predictor of virologic success (#768); and finally that persons who were receiving ARV therapy and had viral blips had subtherapeutic ARV concentrations, whether because of individual patient pharmacokinetic characteristics or medication adherence, more often than those who did not have viral blips (#772).
So, is TDM dead - I'll answer with an unequivocal no. It was disappointing that the prospective A5146 trial did not show a benefit of TDM. However, given that the desired experimental conditions were not achieved for all drugs (that is, concentrations of f-APV were lower in the TDM arm than in the SOC) this was not, unfortunately, a valid comparison of the benefit of TDM. Abstracts 768, 770 and 772 each showed a benefit of knowing the concentration of drug in the body: as a predictor of virologic success, as a strategy to simplify therapy and reduce side effects, and as a correlate with virologic suppression. I believe these and other clinical scenarios are settings where knowledge of drug concentrations can contribute to patient management. I do acknowledge the lack of widespread availability of laboratories to measure ARV concentrations in the United States is a limiting factor to the use of TDM. But, if there is a demand among clinicians for these services, the current limited ability to measure ARV drug concentrations will be overcome. To me, the most compelling case for clinicians is that if we are going to spend fairly large amounts of money to know a patient's viral genotype or phenotype in order to guide the selection and monitoring of an ARV regimen, then why not spend a relatively small amount of money to know whether the patient has ARV concentrations in the body that are necessary to ensure therapeutic success.
|
| |
|
|
|
|
|
