Brown, Kevin C. PharmD*; Patterson, Kristine B. MD†; Jennings, Steven H. BS*; Malone, Stephanie A. MA*; Shaheen, Nicholas J. MD‡; Asher Prince, Heather M. PA-C†; Spacek, Melissa BSN‡; Cohen, Myron S. MD†; Kashuba, Angela D. M. BScPhm, PharmD, DABCP*
The Joint United Nations Program on HIV/AIDS (UNAIDS)1 strategy aims to reduce the sexual transmission of HIV by one-half by 2015. Disproportionate to the burden of disease, men who have sex with men (MSM) have limited access to prevention services other than barrier methods. As such, the number of new HIV infections in MSM in the US continues to increase.2 To achieve this goal, a comprehensive prevention program using antiretroviral therapy (ART) is being incorporated into this strategy.
Recently, the HTPN 052 trial in HIV serodiscordant heterosexual couples demonstrated that immediate initiation of antiretrovirals in the HIV-positive partner at a CD4+ T-cell count of >350, compared with delayed therapy at a CD4+ cell count of <250, reduced HIV transmission to the uninfected sexual partner by 96%.3 HIV transmission modeling has correlated increasing concentrations of HIV RNA in semen to an increasing probability of infection.4 Despite suppression of HIV RNA in blood, HIV RNA can still be detected in genital secretions of up to 8% of HIV-infected men on ART.5 Using selected antiretrovirals to eliminate HIV replication in the genital secretions of HIV-infected men has implications in preventing transmission to sexual partners. Antiretrovirals that achieve high concentrations in colorectal tissues may also reduce HIV replication, and limit infectiousness. Additionally, selecting antiretrovirals that target vulnerable mucosal tissues may offer optimal protection from HIV acquisition when used in a postexposure prophylaxis (PEP) regimen. Characterizing genital tract and colorectal tissue antiretroviral pharmacokinetics is critical for informing the development of future prevention strategies. The present study was designed to define the exposure of darunavir (DRV) plus ritonavir (RTV) and etravirine (ETR) in the seminal fluid and rectal tissue (RT) after single and multiple doses.
Study Design and Subject Selection
This 8-day, open-label, pharmacokinetic (PK) study in healthy HIV-negative male volunteers was conducted between June 2009 and January 2010 at the University of North Carolina at Chapel Hill (UNC). DRV and ETR tablets and RTV capsules were provided by Janssen Therapeutics. UNC Biomedical Institutional Review Board approved the study. All visits were conducted in the UNC Clinical and Translational Research Center (CTRC). The study was registered with the NIH clinical trial registry (NCT00855088). All subjects provided written informed consent before any procedures were performed.
Screening procedures occurred within 42 days of study drug dosing. Subjects were eligible to participate if they were healthy males 18–49 years of age having a body mass index (BMI) between 18 and 30 kg/m2, with intact genital and gastrointestinal tracts. Subjects were excluded if they had a history of regular alcohol consumption, were currently smoking more than 5 cigarettes per day, had a positive urine drug screen, or had a currently active sexually transmitted disease. Subjects were screened for gonorrhea, chlamydia, trichomonas, syphilis, herpes simplex virus-2, hepatitis B and C, and HIV. All testing was performed in the UNC Health Care McLendon Laboratories and in the UNC Sexually Transmitted Diseases Cooperative Research Center Microbiology Core Laboratory.
Subjects were excluded for any clinically significant abnormality in the laboratory results or physical examination deemed by the study physician to increase subject risk or compromise study results. Twelve-lead electrocardiogram testing was performed on subjects who were older than 35 years of age, and subjects were excluded if they exhibited a QTc >450 ms. All prescription and nonprescription medications and supplements, with the exception of acetaminophen (up to 1 g/d), were required to be discontinued at least 7 days before study drug dosing until study completion. Subjects were instructed to abstain from all sexual activity and use of intrarectal products 72 hours before dosing until study discharge.
Safety laboratory monitoring was performed on the day before dosing, 6 days after initial dosing, and at follow-up. A full physical examination was performed at screening and at follow-up, and brief physical examinations were performed on the day before dosing and day 6. Urine toxicology screening was performed at screening, on the day before dosing, and on day 6.
Subjects received 600 mg of DRV, 100 mg of RTV, and 200 mg of ETR, orally twice daily on days 1–7 and a single dose on the morning of day 8. Subjects followed a low fiber diet for 3 days and a clear liquids diet the afternoon before the flexible sigmoidoscopy to collect RT. Subjects were admitted the evening before day 1 to the UNC CTRC and provided a baseline seminal plasma (SP) sample. Subjects fasted for 2 hours before and 4 hours after dosing on days 1, 7, and 8 and were provided a standardized meal with each dose (500 kcal, 20% fat). On day 1, blood plasma (BP) was obtained immediately predose and then 1, 2, 3, 6, 8, and 12 hours after first dose. Each subject collected 2 semen specimens that were time matched with 2 of the 6 postdose BP samples. A total of 4 subjects were assigned to each time point. A single RT biopsy was obtained that was time matched with one postdose BP sample. Two subjects were assigned to each RT biopsy time point. Subjects recorded dosing times on days 2–6 while at home and were instructed to take doses with meals. Subjects were readmitted to the CTRC in the evening of day 6. On days 7 and 8, BP and RT PK sampling identical to day 1 were performed. However, each subject collected 6 semen specimens over days 7–8 that were time matched with BP samples. Subjects were discharged after the 12-hour PK sample collection on day 8 and returned for safety evaluations 7–10 days after the last dose of study medications.
Sample Collection and Processing
Whole blood was obtained using K2EDTA collection tubes (BD Diagnostics, Franklin Lakes, NJ) and centrifuged at 1700g at 5°C for 10 minutes. Whole semen samples were allowed to liquefy at room temperature for at least 45 minutes, and then centrifuged at 2500g at 10°C for 15 minutes. Before collection, rectal biopsy sites were rinsed with a solution containing simethicone 40 mg (simethicone oral suspension 40 mg/0.6 mL, Major Pharmaceuticals, Livonia, MI) diluted in 500 mL of sterile water for irrigation. Ten single RT biopsies were collected using Radial Jaw 4 Large Capacity Forceps (Boston Scientific, Natick, MA), pooled into a single cryovial, and snap frozen. All specimens were stored at −80°C until analysis.
DRV, RTV, and ETR in BP and SP were analyzed using a previously published method.6 For the analysis of RT, approximately 25 mg of blank RT was spiked with 100 μL of a known concentration of DRV, ETR, and RTV to make a calibration range of 10–10,000 ng/mL (40–40,000 ng/g). Quality controls were prepared at 30, 750, and 7500 ng/mL (120, 3,000 and 30,000 ng/g). The samples were placed in 2.8 mm of Precellys metal bead kit vials (P/N 03961-1-008; Bertin Technologies, Villeurbanne, France) and homogenized (Precellys 24, Bertin Technologies, Villeurbanne, France). The subsequent sample preparation extraction steps were identical to the published method.6
All 3 matrix concentrations were analyzed using validated methods on an Agilent 1100 series High Performance Liquid Chromatography System and an Agilent 1100 MSD (Agilent Technologies, New Castle, DE). The Agilent 1100 MSD instrument was used in positive electrospray ionization mode, with a source temperature of 350°C. Analytes were separated on an Agilent Zorbax XDB C-8 (3.0 × 50 mm, 1.8 m) PN#927975-306 with a frit [Agilent (4.6 mm, 0.2 μm) PN 5067-1562] using a gradient mobile phase method. Analysis was performed in secondary ion mass mode with the mass:charge ratio (m/z) being 309.0 for alprazolam (internal standard), 548.2 for DRV, 721.3 for RTV, and 435.0 for ETR. The quantification ranges of the assays were 2–2000 ng/mL for BP and SP and 40–40,000 ng/g for RT. Intraday and interday accuracy and precision were ≤15% and ≤10%, respectively, for all matrices.
SP protein binding was determined by incubating 300 μL of SP pooled by subject from PK2 in duplicate at 37°C, 300 rpm for 18 hours in rapid equilibrium dialysis cartridges (Rapid Equilibrium Dialysis Device System, Thermo Scientific, Pittsburg, PA; Thermo Scientific RED Device Inserts, Thermo Scientific Part No: 89810; reusable Teflon base plate, Thermo Part No: 89811), followed by liquid extraction using methyl-tert-butyl-ether. Concentrations were analyzed using the same equipment and settings as previously stated. The quantification range of the assay was 2–10,000 ng/mL for SP binding. Intraday and interday accuracy and precision were ≤15% for semen protein binding. All methods were validated as mandated by the industry guidance set by the US Department of Health and Human Services, Food and Drug Administration, and Center for Drug Evaluation and Research.7
BP, SP, and RT pharmacokinetic parameters were estimated using noncompartmental methods (Phoenix WinNonlin; Pharsight, Cary, NC). The maximum concentration (Cmax) was determined visually, and time to maximum concentration (Tmax) was defined at Cmax. Exact sample collection times were used in the analysis. The area under the plasma concentration–time curve within the dosing interval (AUC12h) was estimated using the log-linear trapezoidal method, and visual curve stripping was performed to estimate terminal elimination slopes. Blood plasma concentrations 12 hours postdose (C12h) were obtained from the intermediate output calculating AUC12h. For PK2, individual time concentration profiles were created using the 6 samples collected on days 7 and 8, and by supposition, the concentration at 12 hours postdose was used at time zero. Previous investigations have determined that sampling frequency does not affect SP concentrations of antiretrovirals.8 To estimate SP PK parameters for PK1, and RT PK parameters for PK 1 and PK2, a composite approach was used by analyzing geometric mean concentrations. Composite profiles for PK1 SP, PK1 RT, and PK2 RT were created using geometric mean concentrations and times at each time point, and samples were grouped using the closest nominal time. An RT density of 1.04 g/mL was used to convert ng/g to ng/mL.9 To compare SP and RT exposure to BP, SP:BP and RT:BP AUC12h ratios were calculated for days 1 and 7/8. To describe multidose accumulation in BP, SP, and RT, PK2:PK1 AUC12h ratios were calculated.
Descriptive statistics were generated by SAS Institute, Inc, software version 9.1.3 (Cary, NC). Demographic data and pharmacokinetic parameters are presented as median (range). Geometric mean ratios (GMRs) with 90% confidence intervals are presented for PK2 SP versus BP. For the ratios that included a composite profile (PK1 SP vs. BP, PK1 RT vs. BP, and PK2 RT vs. BP), the composite parameter value was divided by the corresponding geometric mean BP.
The percent protein unbound was calculated by subtracting the percent protein bound from 100%. The unbound exposure and trough concentration of drug in SP was calculated by multiplying the total exposure and trough concentration by the percent protein-unbound derived from the RED cartridge analytical method. The protein-unbound geometric mean ratios were calculated similarly by using the reported unbound fraction in BP (DRV: 5%, RTV: 3%, ETR: 0.1%).
Subject Demographics, Disposition, and Safety
Eighteen men screened for this study: 13 were enrolled and 12 completed. Of the 13 subjects, one subject withdrew because of schedule conflicts. This subject did not contribute demographic or pharmacokinetic data. Median (range) age of the 12 participants was 27 (21–36) years, weight was 78.6 (68.4–108.1) kg, and body mass index was 25.5 (20–29.9) kg/m2. Six (50%) subjects were Caucasian, 1 (8%) was Hispanic Caucasian, and 5 (42%) were African–American.
Subjects tolerated the study medications well. Most adverse events (AEs) were mild, and no serious AEs were reported. The most frequently reported AE was GI disturbance, including loose stools or diarrhea (69%). Other AEs included fatigue (15%), decreased concentration (15%), and headache (15%). Grade 2 rash was experienced by one participant approximately 1–2 hours after leaving his PK2 study visit and resolved in 7 days. One subject reported sustained sensations of urinary urgency; resolution could not be assessed because subject was lost to follow-up. RT sampling was well tolerated. A single subject reported spotting on toilet tissue and a small amount of blood on stool immediately following the procedure, which resolved within a few hours.
BP, Semen, and RT Pharmacokinetics
Figures 1–3 depict the BP, SP, and RT concentrations for all 12 subjects at day 1 (PK1) and days 7/8 (PK2), respectively. After the first dose, DRV (Fig. 1A), RTV (Fig. 2A), and ETR (Fig. 3A) were detected in all biological matrices. Table 1 summarizes the PK parameters for each matrix. After single and multiple doses, median AUC12h and C12h were highest in RT and lowest in SP for all 3 drugs. Table 2 summarizes the relative exposures of each drug in SP and RT versus BP and the PK2:PK1 accumulation ratios.
After the first dose, DRV exposures in SP were 82%–92% lower than in BP and were detected 1 hour postdose. After multiple doses, DRV exposures in SP were 80%–85% lower than in BP. Using C12h and AUC12h as measures of exposure, DRV accumulated (PK2:PK1) in SP by 2- to 2.8-fold upon multiple dosing.
RTV was not detected in SP until 2 hours after the first dose, and peak exposures were not reached until 8 hours postdose. After the first dose, RTV exposures in SP were 89%–95% lower than in BP. After multiple doses, RTV exposures in SP were 93% lower than in BP. RTV accumulated (PK2:PK1) in SP by 1.4- to 2.3-fold with multiple dosing.
ETR was detected in SP 2 hours after the first dose. ETR exposures in SP were 83%–87% lower than in BP after the first dose, and 85%–88% lower than in BP after multiple dosing. ETR accumulated (PK2:PK1) in SP by 3.6- to 5.2-fold with multiple dosing.
SP Protein Binding
The median [interquartile range (IQR)] protein binding in SP was 14.0% (10.1%–18.4%) for DRV. This is much lower than protein binding of DRV in BP, which is approximately 95%.10 Based on these data, the protein-free AUC12h and C12h in SP were 3.4- and 2.5-fold higher than in BP, respectively. For RTV, the median (IQR) protein binding in SP was 70.3% (66.9%–73.3%). Protein binding of RTV in BP is reported to be 98%.11 Based on these data, the protein-free AUC12h and C12h in SP were similar (within 2%) to BP. For ETR, the median (IQR) protein binding in SP was 96.7% (95.2%–99.0%). Protein binding of RTV in BP is reported to be 99.9%.12 Based on these data, the protein-free AUC12h and C12h in SP were 4.8- and 3.8-fold higher than in BP, respectively.
After the first dose, DRV exposures in RT were 1.1- to 1.2-fold higher than in BP and were detected 1 hour postdose. Tmax occurred 2.4 hours after peak plasma concentrations. After multiple doses, DRV exposures in RT were 2.3- to 2.7-fold higher than in BP. DRV accumulated (PK2:PK1) in RT by 3.3- to 4-fold with multiple dosing. After the first dose, RTV exposures in RT were 5.8- to 12-fold higher than in BP, and were detected 1 hour postdose. Tmax occurred 4.9 hours after peak plasma concentrations. After multiple doses, RTV exposures in RT were 13- to 27-fold higher than in BP. RTV accumulated in RT by 3.7- to 5.2-fold with multiple dosing. After the first dose, ETR exposures in RT were 15- to 16-fold higher than in BP and were detected 1 hour postdose. Tmax occurred 4.9 hours after peak plasma concentrations. After multiple doses, ETR exposures in RT were 7.5- to 9.7-fold higher than in BP. ETR accumulated in RT by 2- to 3.6-fold with multiple dosing.
Despite strong implementation of behavioral interventions, the incidence of HIV in the United States has not substantially declined over the past 15 years.2 Therefore, combination prevention, the use of antiretroviral agents to prevent HIV transmission in conjunction with behavior modification and barrier methods, seems to be the most effective strategy to date. For example, HPTN 052 demonstrated that transmission to serodiscordant heterosexual partners was reduced by 96% by offering potent ART to the HIV-infected partner.3 The most likely means by which transmission was prevented in this study was through the reduction in systemic, and thus genital tract, viremia.4 This study was performed to characterize the pharmacokinetics of darunavir (DRV) and ETR in SP and RT, informing their role in reducing HIV replication at these sites, as well as being used in a PEP regimen where colorectal tissue was exposed to HIV.
Whether specific antiretroviral regimens are more effective in offering protection from HIV acquisition or preventing HIV transmission is unknown. Drugs that reach high concentrations quickly at sites of transmission and infection would be favorable in antiretroviral-based postexposure prophylaxis. Antiretrovirals that achieve high exposures in genital secretions and can limit local HIV replication may be favorable for decreasing the infectivity of the HIV-infected person. Therefore, defining the antiretroviral exposures in biological compartments that are vulnerable to acquisition and are sources of infection, such as RT and semen, could assist in selecting regimens for HIV prevention.
The total drug concentrations of DRV, RTV, and ETR in SP were 80–93% lower than in BP. This is consistent with other antiretrovirals, in which higher BP protein binding results in lower SP concentrations.13 As BP protein binding decreases, there is a greater amount of protein-unbound drug available to cross cellular membranes and distribute into physiological compartments. However, the drug-binding proteins albumin and alpha-1-acid glycoprotein (AAG) are approximately 97% lower in SP than in BP.14 We measured the protein-unbound concentrations in SP and confirmed that lower protein binding exists for all 3 antiretrovirals in this compartment. DRV binds primarily to alpha-1-acid glycoprotein. Unbound DRV concentrations in SP were approximately 2.5-fold higher than BP concentrations. Our measured protein-unbound trough concentration (230 ng/mL) was 150-fold higher than the unbound EC90 measured in vitro (1.5 ng/mL or 2.7 nM15). The relatively low fraction of protein-bound DRV in SP is likely because of the very low concentrations of albumin and alpha-1-acid glycoprotein in SP, and the resulting high protein-unbound concentration of DRV relative to its EC90 could be favorable in suppressing viral replication in the male genital tract and in reducing infectivity. ETR binds to both albumin and alpha-1-acid glycoprotein. This investigation determined that unbound ETR concentrations in SP were approximately 3.8-fold higher than BP concentrations, and the unbound ETR trough concentration (1.2 ng/mL) was similar to the protein-free EC90 (1.3 ng/mL or 2.9 nM16).
Understanding antiretroviral RT pharmacokinetics is essential to developing appropriate strategies to reduce viral replication in this highly vascularized compartment rich in lymphoid tissue and to select optimal antiretrovirals for postexposure prophylaxis regimens. Compared with BP, RT exposures of DRV, RTV, and ETR were 1.3, 5.8, and 15.7 times higher after a single dose and 2.7, 12.8, and 7.5 times higher after multiple doses, respectively. All drugs were detected in the rectal mucosa within 1 hour after dosing. The quick penetration of these drugs into the RT may be because of a highly vascularized mucosa and protein binding in interstitial fluid.17
The multiple-dose accumulation of DRV and RTV in RT was approximately 2-fold greater than in BP (4.0 vs. 1.8 for DRV, 5.2 vs. 2.3 for RTV). These high exposures may in part be because of the elimination pathway of these drugs. Forty-one percent of DRV and 34% of RTV is fecally eliminated as unchanged drug, allowing for local colorectal drug exposure. Mucus trapping of drug is another potential factor in the accumulation in RT.18 Conversely, the accumulation of ETR in RT after multiple dosing was only 50% than that in BP (2.0 vs. 4.2). The long BP half-life of ETR explains the high BP accumulation ratio. Additionally, more ETR (86%) is fecally eliminated as unchanged drug than DRV or ritonavir (RTV).19,20 Because we found ETR to reach multiple-dose concentrations in the RT by 6 hours after the first dose, it may be possible that the relatively larger amount of the fecally eliminated ETR “saturates” the tissue after the first dose. Therefore, any accumulation of ETR in BP contributes less to the RT concentrations than the fecal concentrations.
The RT pharmacokinetic profiles of these drugs are similar to previous data that we have reported for maraviroc. Maraviroc also had detectable concentrations 1 hour after the first dose, and the accumulation ratio with multiple dosing was 4.0. The relative exposure of maraviroc in RT compared with BP was 26-fold, which is higher than what was seen with DRV (2.7), RTV (12.8), and ETR (7.5).21 Rectal concentrations of tenofovir and emtricitibine were measured 24 hours after a single dose by Patterson et al. Tenofovir rectal concentrations at the end of the dosing interval were 46-fold higher than BP, which is higher than DRV (1.1), RTV (12.3), and ETR (15.4). Emtricitibine rectal concentrations were 2.6-fold higher than BP, which is lower than RTV and ETR, but higher than DRV.22 Furthermore, these high exposures in RT indicate that these drugs are penetrating a suspected HIV tissue reservoir, thus minimizing the possibility of HIV replication.23 Data from Kelley et al and Lampinen et al demonstrating rectal HIV RNA shedding is rare, even in the face of sexually transmitted infections, when HIV RNA is below quantification in BP, provides indirect evidence that high rectal antiretroviral concentrations are providing strong activity at this site.24,25
One potential limitation of this study is that it characterized the pharmacokinetics of these antiretrovirals in unperturbed rectal mucosa. Bowel preparations were not used before the biopsy procedure because hyperosmolar enemas can shift a significant amount of water into the lumen of the colon and cause epithelial sloughing.26 Future studies assessing the impact of bowel preparations may be important, as they are commonly (up to 60% of MSM) used before anal intercourse27 and may increase the risk of HIV transmission.28
In summary, we evaluated combination antiretroviral exposure in RT and defined exposure and protein binding in SP. These data provide pharmacologic plausibility for the use of DRV plus RTV and ETR in secondary HIV prevention, in both infected and uninfected individuals. The quick penetration and sustained concentrations of DRV and ETR in the rectal mucosa are desirable characteristics for prevention of HIV acquisition. The unbound concentrations in semen are higher than in blood and could be effective in suppressing HIV replication in the male genital tract. Future investigations will determine if these concentrations in RT and semen can fully suppress viral shedding. Despite intensive sampling and scheduling challenges for multiple precisely timed rectal biopsies and semen samples, our data also demonstrate that these studies can be performed efficiently and safely.
3. Cohen MS, Chen YQ, McCauley M, et al.. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505.
4. Chakraborty H, Sen PK, Helms RW, et al.. Viral burden in genital secretions determines male-to-female sexual transmission of HIV-1: a probabilistic empiric model. AIDS. 2001;15:621–627.
5. Bujan L, Daudin M, Matsuda T, et al.. Factors of intermittent HIV-1 excretion in semen and efficiency of sperm processing in obtaining spermatozoa without HIV-1 genomes. AIDS. 2004;18:757–766.
6. Rezk NL, White NR, Jennings SH, et al.. A novel LC-ESI-MS method for the simultaneous determination of etravirine, darunavir and ritonavir in human blood plasma. Talanta. 2009;79:1372–1378.
7. US DHHS, FDA and CDER. Guidance for Industry: Bioanalytical Method Validation. US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research, Center for Veterinary Medicine; Rockville, MD; 2001. Available from: http://www/fda.gov/cder/guidance/index.htm
. Accessed November 1, 2010.
8. Cao YJ, Ndovi TT, Parsons TL, et al.. Effect of semen sampling frequency on seminal antiretroviral drug concentration. Clin Pharmacol Ther. 2008;83:848–856.
9. Mardirossian G, Tagesson M, Blanco P, et al.. A new rectal model for dosimetry applications. J Nucl Med. 1999;40:1524–1531.
13. Nicol MR, Kashuba AD. Pharmacologic opportunities for HIV prevention. Clin Pharmacol Ther. 2010;88:598–609.
14. Cao YJ, Hendrix CW. Male genital tract pharmacology: developments in quantitative methods to better understand a complex peripheral compartment. Clin Pharmacol Ther. 2008;83:401–412.
15. De Meyer S, Azijn H, Surleraux D, et al.. TMC114, a novel human immunodeficiency virus type 1 protease inhibitor active against protease inhibitor-resistant viruses, including a broad range of clinical isolates. Antimicrob Agents Chemother. 2005;49:2314–2321.
16. DeJesus E, Lalezari JP, Osiyemi OO, et al.. Pharmacokinetics of once-daily etravirine without and with once-daily darunavir/ritonavir in antiretroviral-naive HIV type-1-infected adults. Antivir Ther. 2010;15:711–720.
17. Faed EM. Protein binding of drugs in plasma, interstitial fluid and tissues: effect on pharmacokinetics. Eur J Clin Pharmacol. 1981;21:77–81.
18. McConnell EL, Fadda HM, Basit AW. Gut instincts: explorations in intestinal physiology and drug delivery. Int J Pharm. 2008;364:213–226.
19. Kakuda TN, Scholler-Gyure M, Workman C, et al.. Single- and multiple-dose pharmacokinetics of etravirine administered as two different formulations in HIV-1-infected patients. Antivir Ther. 2008;13:655–661.
20. Scholler-Gyure M, Kakuda TN, Raoof A, et al.. Clinical pharmacokinetics and pharmacodynamics of etravirine. Clin Pharmacokinet. 2009;48:561–574.
21. Brown KC, Patterson KB, Malone SA, et al.. Single and multiple dose pharmacokinetics of maraviroc in saliva, semen, and rectal tissue of healthy HIV-negative men. J Infect Dis. 2011;203:1484–1490.
22. Patterson KB, Prince HA, Kraft E, et al.. Penetration of tenofovir and emtricitabine in mucosal tissues: implications for prevention of HIV-1 transmission. Sci Transl Med. 2011;3:112–114.
23. Cohen J. Tissue says blood is misleading, confusing HIV cure efforts. Science. 2011;334:1614.
24. Kelley CF, Haaland RE, Patel P, et al.. HIV-1 RNA rectal shedding is reduced in men with low plasma HIV-1 RNA viral loads and is not enhanced by sexually transmitted bacterial infections of the rectum. J Infect Dis. 2011;204:761–767.
25. Lampinen TM, Critchlow CW, Kuypers JM, et al.. Association of antiretroviral therapy with detection of HIV-1 RNA and DNA in the anorectal mucosa of homosexual men. AIDS. 2000;14:F69–F75.
26. Schmelzer M, Schiller LR, Meyer R, et al.. Safety and effectiveness of large-volume enema solutions. Appl Nurs Res. 2004;17:265–274.
27. Carballo-Dieguez A, Bauermeister JA, Ventuneac A, et al.. The use of rectal douches among HIV-uninfected and infected men who have unprotected receptive anal intercourse: implications for rectal microbicides. AIDS Behav. 2008;12:860–866.