aPHLS Antiviral Susceptibility Reference Unit, Division of Immunity and Infection, University of Birmingham, Birmingham, UK; bDepartment of Sexual Medicine, Birmingham Heartlands Hospital, Birmingham, UK; cDepartment of Pharmacology and Therapeutics, University of Liverpool, Liverpool, UK; and dDepartment of Primary Care and Population Sciences, Royal Free and University College Medical School, London, UK.
Sponsorship: S.T. is supported by a West Midlands NHSE Sheldon Clinical Research Fellowship. Financial support is acknowledged from Dupont Pharmaceuticals.
Received: 14 December 2000;
revised: 31 May 2001; accepted: 5 June 2001.
Efavirenz is a potent non-nucleoside reverse transcriptase inhibitor, licensed for the treatment of HIV-1. Data on sanctuary site penetration are limited. Therefore, we measured efavirenz concentrations in the blood and semen of 19 HIV-1-positive men and found concentrations in seminal plasma averaged 10% of those in blood plasma. Furthermore, seminal plasma viral loads were suppressed by 24 weeks of therapy in all patients. These data suggest that efavirenz-containing regimens have antiviral activity within the male genital tract.
Nineteen HIV-1-positive men attending Birmingham Heartlands Hospital were recruited to study efavirenz drug concentrations and antiviral activity in semen. Local ethics committee approval and informed consent were obtained. All patients took efavirenz 600 mg at night and were at steady state before the study; all were screened for urethritis.
Before commencing drug regimens, we obtained baseline (week 0) samples for the measurement of blood plasma viral load (BPVL) and seminal plasma viral load (SPVL). Subsequently, blood was drawn at 12 and 24 weeks and semen samples were obtained after 24 weeks for viral load analysis. Matched blood and semen samples produced between 4 and 24 weeks were used for efavirenz drug concentration determination.
On study day 1 patients provided matched blood and semen samples approximately 24 h post-efavirenz ingestion (before taking their next dose) One week later patients provided further matched semen and blood samples 12 h post-drug ingestion. Samples were processed as previously described . Concentrations of efavirenz in plasma and semen were determined by high performance liquid chromatography (HPLC).
Drug concentrations were compared using unpaired t-tests or Mann–Whitney U tests when appropriate. Ratios of blood plasma to seminal plasma efavirenz were calculated for patients in whom samples had been produced within one hour of each other. These were analysed using the Mann–Whitney U test. Mann–Whitney U tests or Kruskal–Wallis tests were used for the comparison of viral loads. All correlations were performed using Spearman’s rank correlation coefficient.
Nine of the patients were antiretroviral naive at the time of starting efavirenz (group A), eight were antiretroviral experienced but had started efavirenz after taking a short treatment interruption, median duration 11.5 months, range 1–17 months (group B), and the remaining two patients switched onto efavirenz-containing regimens directly from a successful non-efavirenz-containing regimen (group C). One patient tested positive for chlamydia at baseline, but was treated and has been included in all analyses. Before starting efavirenz, the eight patients in group B had been exposed to a median of four (range three to nine) antiretroviral drugs over a total period of 73 (range 8–240) weeks. The two patients from group C had previously received a total of seven and eight drugs over a total period of 278 and 354 weeks, respectively. At the time of starting efavirenz, patients started a median of two (range two to five) additional drugs. Among patients in group A, the most common treatment combination was efavirenz in combination with zidovudine and lamivudine (n = 6), with two patients receiving efavirenz in combination with lamivudine and stavudine. Among patients in groups B and C treatment combinations varied widely, and included nine different treatment combinations among the 10 patients.
Baseline blood plasma viral loads were significantly lower in group A than in group B (P = 0.04, Mann–Whitney U test). However, there were no significant differences in the baseline CD4 cell counts or SPVL in the two groups (P = 0.70 and 0.14, respectively), although SPVL were lower in group A than in group B.
Sixty-nine blood plasma and 69 time-matched seminal plasma samples were produced over the study period. In group A, median week 0 BPVL was 4.99 log10 copies/ml (range 4.1–5.87) and week 0 SPVL was 3.48 log10 copies/ml (range 2.6–4.48). In group B, week 0 BPVL was 5.88 log10 copies/ml (range 4.56–5.88) and week 0 SPVL was 4.51 log10 copies/ml (range 2.6–6.3). After 24 weeks on treatment, BPVL in group A was less than 1.69 log10 copies/ml in seven out of seven patients and SPVL was less than 2.6 log10 copies/ml in seven out of seven patient. In group B, seven out of eight patients had BPVL less than 1.69 log10 copies/ml (range 1.69–3.58), whereas all eight patients had SPVL less than 2.6 log10 copies/ml.
One patient in group C remained suppressed at 24 weeks in both blood plasma and seminal plasma, whereas one patient had viral rebound in blood plasma but not seminal plasma.
Efavirenz concentrations were measured in a total of 34 blood and 34 time-matched semen samples. Of these matched samples, three (8.8%) were taken in the first 8 h, 17 (50.0%) at approximately 12 h (median 11 h 30 min, range 8–14 h 30 min), and 14 (41.2%) at approximately 24 h (median 23 h 55 min, range 22–24 h 15 min).
The median efavirenz concentration in blood plasma at 12 and 24 h post-dosing was 2184 ng/ml (843–14 356) and 1785 ng/ml (694–8096), respectively. The median efavirenz concentration in seminal plasma 12 and 24 h post-dosing was 215 ng/ml (62–622) and 238 ng/ml (49–1256), respectively. The median seminal : blood plasma ratio at 12 h was 0.08 (0.04–0.16), whereas the median seminal : blood plasma ratio at 24 h was 0.09 (0.03–0.43).
Drug concentrations measured at different times from starting efavirenz were also compared. There were no significant differences between the concentrations in blood (P = 0.38), semen (P = 0.66) or the ratio of the two values (P = 0.92) according to the length of time since starting efavirenz (Kruskal–Wallis test). Although median concentrations of efavirenz in blood plasma appeared to decrease over time (2362, 2184 and 1785 ng/ml at 8, 12 and 24 h, respectively), these differences were not significant (P = 0.85). Similarly, neither semen efavirenz concentrations (medians of 333, 216 and 238 ng/ml, P = 0.83), nor the ratio of the two values (medians of 0.14, 0.08 and 0.09, P = 0.78) differed over the three time periods. Therefore, for all subsequent analyses, all samples have been combined (Fig. 1).
Trough blood plasma efavirenz concentrations in this study were comparable with values derived from population pharmacokinetic studies of efavirenz . Furthermore the median blood plasma trough efavirenz concentrations of 1758 ng/ml (5.6 nM) were above a recently proposed minimum trough concentration for efavirenz of 1000 ng/ml  and approximately 20 times an estimated blood plasma protein corrected EC90 of 92.8 ng/ml (294 nM) (data on file Dupont Pharma). Currently, it is unknown to what degree efavirenz is protein bound in the genital tract. If it is similar to protein binding in blood plasma, then we estimate that absolute efavirenz concentrations should exceed this value by more than twofold. Whereas the efficacy of a drug regimen cannot be attributed to any single component, it is reassuring to note that SPVL were suppressed in all 18 patients reaching 24 weeks of therapy on efavirenz-containing combinations in this study.
Our findings are in agreement with those of Kim et al., who also demonstrated that efavirenz reached the genital tract at concentrations that exceeded the estimated protein corrected EC90 for efavirenz.
In conclusion, we have successfully measured efavirenz concentrations in the semen of 19 HIV-1-infected men and shown that efavirenz-containing regimens appear to be effective in reducing SPVL. Further understanding of the ability of individual drugs to access sanctuary sites is important, because the antiviral activity of a drug combination will be a function of the level of penetration of each of its components . Drug levels in semen may therefore influence the development and sexual transmission of resistant HIV-1.
The authors would particularly like to thank all patients and staff involved with the semen studies.
Caroline A. Sabind
Susan M. Drakeb
David J. Whiteb
David J. Backc
1. Taylor S, van Heeswijk RP, Hoetelmans RM. et al
. Concentrations of nevirapine, lamivudine and stavudine in semen of HIV-1-infected men. AIDS 2000, 14: 1979–1984.
2. Joshi AS, Barrett JS, Fiske WD, et al. Population pharmacokinetics of efavirenz in phase II studies and relationship with efficacy. 39th Interscience Conference on Antimicrobial Agents and Chemotherapy
. San Franciso, September 1999 [Poster 1201].
3. Marzolini C, Telenti A, Decosterd LA. et al
. Efavirenz plasma levels can predict treatment failure and central nervous system side effects in HIV-1 infected patients. AIDS 2001, 15: 71–75.
4. Kim J, Gotzkowsky K, Eron J, et al. Penetration of efavirenz into the male genital tract as measured by seminal plasma to blood plasma concentration and AUC 24 h ratios. 40th Interscience Conference on Antimicrobial Agents and Chemotherapy
. Toronto, September 2000 [Abstract 1171].
5. Taylor S, Pereira A. Antiretroviral drug concentrations in semen of HIV-1 infected men. Sex Transm Infect 2000, 77: 4–11.