Male erectile dysfunction is defined as the inability to attain or maintain a penile erection sufficient for satisfactory sexual performance. The aetiology of erectile dysfunction is usually multifactorial, although organic causes are implicated in the majority of cases. The point prevalence of erectile dysfunction is estimated to be 10%, and increases with advancing age. Two independent studies have reported that the prevalence of erectile dysfunction in HIV-infected homosexual men was 33%, and increased with more advanced HIV disease[4,5]. More than 50% of HIV-infected homosexual men rate enjoyable sexual functioning as an important aspect of quality of life. Many factors may contribute to HIV-associated sexual dysfunction including endocrine abnormalities, neurological disorders, adverse effects of medication and psychosocial issues[5-8].
Sildenafil citrate, at doses of 25 to 100 mg, is effective and well tolerated in the treatment of penile erectile dysfunction, acting as a potent and relatively specific inhibitor of the cyclic-guanosine monophosphate (cGMP)-specific phosphodiesterase type 5 in the corpus cavernosum[10,11]. Inhibition of cGMP metabolism in erectile tissue enhances the erectile response to sexual stimulation. Although a meta-analysis of over 3700 study participants, representing 1631 patient- treatment-years, showed no evidence of serious sildenafil-related adverse events, to date there are little published data on the use of sildenafil in HIV infection.
Concerns have been expressed about a possible interaction between sildenafil and combination antiretroviral therapy. Sildenafil is both metabolized by, and acts as a weak inhibitor of CYP3A4 and CYP2C9[12,15]. Therefore, co-administration of sildenafil with drugs which are substrates and/or inhibitors of cytochrome P450 may result in clinically significant drug interactions, with the potential of enhanced pharmacodynamic effects of sildenafil.
Interaction studies in healthy volunteers have shown that saquinavir increases sildenafil maximum plasma concentration (Cmax) by 140% and the area under the concentration curve (AUC) by 210%, and ritonavir increases Cmax and AUC by four- and 11-fold, respectively. Neither saquinavir nor ritonavir pharmacokinetics were affected by co-administration with sildenafil. On the basis of these data, the European Union‚s Committee for Proprietary Medicinal Products (CPMP) have indicated that sildenafil and ritonavir should not be co-administered and have recommended that a starting dose of sildenafil of 25 mg should be considered for patients receiving other protease inhibitor therapy.
Data from both in vitro  and in vivo  studies show that indinavir is a less potent CYP inhibitor than ritonavir, but more potent that saquinavir. Thus the position of indinavir among the protease inhibitors used with sildenafil needs to be addressed. Therefore, in response to the concerns raised by several of our HIV-positive patients who wished to avail themselves of sildenafil therapy, we evaluated the pharmacokinetics of combination therapy with sildenafil plus indinavir in HIV-positive individuals. Although this was a pharmacokinetic interaction study, we also noted all adverse events. In a parallel study the effect of indinavir, ritonavir, saquinavir and nelfinavir on the in vitro hepatic metabolism of sildenafil was assessed.
Six HIV-positive men with erectile dysfunction participated in this 2-day pharmacokinetic study. The mean age of the patients was 35 years (range 27-45years) and the mean weight was 66 kg (range 56-72kg). Four patients had Centers for Disease Control and Prevention (CDC) stage C3 disease, one stage B1 and one stage B3 (1993 CDC disease classification). The risk factors for acquisition of HIV infection were: injecting drug use in one patient, a blood transfusion in one patient and the remaining four patients were homosexual men. None of the patients were co-infected with either HTLV-I or II. The mean CD4 cell count was 333×106/l (range 184-614×106/l). All patients had an undetectable plasma HIV viral load measurement (bDNA, Chiron Corporation, Emeryville, California, USA) and were receiving triple antiretroviral therapy with dual nucleoside analogues (five patients zidovudine/lamivudine, one patient stavudine/didanosine) plus 800mg indinavir three times daily. There was no evidence of active peptic ulcer disease, a bleeding diathesis, ischaemic heart disease, cerebrovascular disease or penile deformities in any of the study participants (all contra-indications to sildenafil therapy). Routine biochemistry for hepatic and renal function were normal. No medication known to interfere with the metabolism of indinavir or sildenafil was prescribed in the 2 weeks prior to the study period.
On the first study day, patients attended the day ward following an overnight fast. Blood pressure was recorded and an indwelling intravenous cannula inserted into the antecubital fossa to facilitate blood sampling. At 0900h a fasting blood sample was taken (time 0 h). Patients then ingested their prescribed therapy including indinavir 800 mg. Blood samples were taken at 1, 2, 3, 4, 6 and 8h after dosing. Blood pressure readings were taken at intervals of 30min for 2h and hourly thereafter. The procedure on the second study day was identical to the first day except sildenafil (25 mg) was added to the prescribed therapy prior to sampling. Samples were centrifuged without delay and the separated plasma heated to 58°C for 30min to inactivate HIV. Plasma was stored at -80°C until analysis using high performance liquid chromatography (HPLC).
Plasma samples in duplicate (200 μl) were pipetted into glass tubes, and internal standard (ritonavir; a gift from Abbott Laboratories, Abbott Park, Illinois, USA) and 50 μl sodium hydroxide (0.5 M) added. The contents of the tube were mixed thoroughly. Standard curves were prepared containing blank plasma and indinavir at a concentration range of 0.5-5 μg/ml. Quality control samples for the assessment of precision and accuracy of the assay were prepared by adding known quantities of indinavir to blank plasma samples. Test samples, standards and quality control samples were extracted with dichloromethane (3 ml) for 10 min. After centrifuging for 5 min at 3292 g, the organic phase was transferred to clean tubes and evaporated to dryness. Extracts were reconstituted into the HPLC mobile phase (150 μl) and transferred to vials for injection into the HPLC. Indinavir and the internal standard were resolved on a Megellen 5C8 column (5μ: 250×4.6 mm; Phenomenex, Macclesfield, UK) with a mobile phase of acetonitrile far UV and water (57:43 v/v) at a flow rate of 1 ml/min. Absorbance was monitored at 205 nm. Peaks of interest, indinavir (retention time = 4.8 min) and internal standard (retention time = 9.7 min) were quantified using a Kontron MT2 data acquisition system (Kontron Instruments, Watford, Herts, UK). The limit of quantitation was 100ng/ml. Interassay variability was determined with two different control samples containing nominal concentrations of 1 and 10 μg/ml. The coefficients of variation were 9.3 and 5.6% respectively. Intra-assay precision was determined with samples containing 1 and 10 μg/ml. The coefficients of variation were 10.3 and 2.8% respectively.
Plasma samples in duplicate (500 μl) were pipetted into glass tubes and internal standard (50μl trihexyphexidyl; 100μg/ml; Sigma Chemical Company, Poole, Dorset, UK) and 100μl K2CO3 (1 M) were added. Tube contents were mixed thoroughly. Standard curves were prepared containing blank plasma and sildenafil at a concentration range of 20-500ng/ml. Quality control samples for the assessment of precision and accuracy of the assay were prepared by adding known quantities of sildenafil to blank plasma samples. Test samples, standards and quality control samples were extracted with dichloromethane (4ml) for 10 min. After centrifuging for 5 min at 3292 g, the organic phase was transferred to clean tubes and evaporated to dryness. Extracts were reconstituted into the HPLC mobile phase (150μl) and transferred to vials for injection into the HPLC. Sildenafil and the internal standard were resolved on a Hypersil Elite 5C18 column (5μ; 150×4.6 mm; Hypersil, Runcorn, UK) with a mobile phase of acetonitrile far UV and 50 mM sodium phosphate buffer pH 5.1 (32:68 v/v) at a flow rate of 1 ml/min. Absorbance was monitored at 220nm. Peaks of interest, sildenafil (retention time = 6.1min) and internal standard (retention time = 7.7min) were quantified using a Kontron MT2 data acquisition system. The limit of quantitation was 10ng/ml. Interassay variability was determined with a control sample containing a nominal concentration of 50ng/ml. The coefficient of variation was 5.8%. Intra-assay variability was determined at the same concentration with a coefficient of variation of 8.8%.
Sildenafil was extracted from a crushed tablet of Viagra by sequential additions of diethyl ether. Extracts were pooled and dried. Purity of the extracted material was determined by LC-MS on a Quattro II quadruple mass spectrometer (Micromass UK, Manchester, UK). Data was acquired by full-scanning acquisition between m/z 100-1050 with a scan duration of 4.91s and processed via MassLynx 2.1. Sildenafil was detected as the only intense peak in the ion chromatogram for m/z 475 representing the protonated molecule ([M+1]+).
In vitro hepatic metabolism of sildenafil
Human liver microsomes (n=4) were obtained from our human liver bank; histologically normal human livers having been obtained from kidney transplant donors with consent of both donor‚s relatives and the local Ethics Committee. Initial linearity studies were performed and revealed that the reaction was linear up to 1mg microsomal protein and an incubation time of 15 min. A 500μl incubation mixture containing 0.05mg protein was incubated with sildenafil (25μM) in the presence of MgCl2 (10mM) and NADPH (1mM) in phosphate buffer (0.067M; pH7.4). The reaction was terminated by the addition of trihexyphexidyl as internal standard and immediate extraction with dichloromethane (4ml; 15min). The organic layer was evaporated to dryness before reconstitution with mobile phase (200μl). Sildenafil metabolism was quantified by HPLC analysis. Five metabolites were separated from internal standard and sildenafil as described above. A range of protease inhibitor concentrations (saquinavir, nelfinavir to 10μM; indinavir to 5μM, ritonavir to 1μM) were screened against sildenafil metabolism. The values of concentration producing 50% inhibition of control enzyme activity (IC50) were determined using an iterative program GRAFIT 3.0 (Erithacus Software Ltd, Staines, Middlesex, UK).
Pharmacokinetic and statistical analysis
Indinavir and sildenafil concentrations were evaluated for maximum concentration (Cmax), time to Cmax (Tmax) and the area under the curve to 8h (AUC0-8h). Cmax and Tmax were obtained by inspection of the data. AUC and elimination half-life (T1/2 for sildenafil) values were determined by non-compartmental analysis using TOPFIT computer software (Gustav Fischer Verlag, Stuttgart, Germany). Cmax and AUC values were log-transformed prior to statistical analysis by the Mann-Whitney U test.
The indinavir levels are shown in Figure 1. The geometric mean AUC0-8h and Cmax for indinavir on the first study day were 19.69μg/mlh (range, 9.19-31.99 μg/mlh) and 7.02 μg/ml (range, 2.33-16.17μg/ml) respectively. In the presence of sildenafil the mean AUC0-8h was 22.37μg/mlh [range, 10.08-37.25μg/mlh; 95% confidence interval (CI) for difference between means -15 to 13.25μg/mlh] and Cmax was 9.11μg/ml (range, 3.41-22.78μg/ml; 95% CI -13 to 6.37μg/ml). Considerable interpatient variability in plasma indinavir levels was noted on both study days.
Figure 2 shows the plasma sildenafil levels from the second study day and, for comparison, sildenafil levels from phase I/II pharmacokinetic studies of 50 mg and 100 mg sildenafil which have been dose normalized to 25 mg[15,19]. The geometric mean AUC0-8h and Cmax for sildenafil were 1631ng/mlh (range, 643-2970ng/mlh) and 384ng/ml (range, 209-766ng/ml) respectively. Tmax occurred 2 h post ingestion of sildenafil. Data from the 50 and 100 mg studies were: AUC0-8h 735 and 1500 ng/mlh and Cmax 260 and 440 ng/ml respectively. The mean elimination half-life for sildenafil in the present study was 2.62 h (range 1.21-4.60 h).
Indinavir was a very potent inhibitor of sildenafil metabolism in vitro (IC50 = 0.39 ± 0.17 μM; mean±SD). For comparison the IC50 values for ritonavir, saquinavir and nelfinavir were 0.034±0.005, 2.86±1.34 and 1.16± 0.50 μM respectively, as shown in Figure 3.
Sildenafil, the first licensed oral drug treatment for erectile dysfunction, improves both erectile function and the number of successful attempts at sexual intercourse in men with erectile dysfunction of organic, psychogenic or mixed aetiology. Furthermore, pooled safety data from over 3700 study participants have indicated sildenafil to be safe and well tolerated. Although post-marketing reports and surveillance have revealed 69 deaths, none has been linked definitively to sildenafil itself.
Sildenafil is rapidly absorbed, with an absolute bioavailability of 40%. Maximal plasma concentrations occur 0.5 to 2h after dosing and this is delayed by co-administration with food. It is highly plasma protein-bound and is cleared non-renally with a terminal half-life of 4 h. Pharmacokinetics are linear over the recommended dosage range. Sildenafil is metabolized by CYP3A4 (major route) and CYP2C9 (minor route) with the major active N-desmethyl metabolite (UK 103,320) having comparable selectivity and approximately 50% of the potency of the parent compound. Sildenafil is also a weak inhibitor of CYPs 3A4, 2C9, 2C19, 1A2, 2D6 and 2E1.
Studies to date show that plasma sildenafil levels are increased by drugs which inhibit CYP3A4; for example, there is a 56% increase in the presence of cimetidine and 182% increase in the presence of erythromycin. Following the release of healthy volunteer interaction data showing marked increases in sildenafil Cmax and AUC when co-administered with saquinavir and ritonavir, the CPMP recommended that a starting dose of 25 mg of sildenafil should be used when co-administered with drugs which inhibit CYP3A4. However, the CPMP does not recommend the co-administration of sildenafil with ritonavir, and suggests that in circumstances where there is no acceptable alternative the maximum dose of sildenafil should not exceed 25 mg in 48 h. Although the dosing recommendations derived from these healthy volunteer pharmacokinetic studies serve as a useful guide, it is recognized that healthy volunteers may demonstrate significant differences in the disposition and handling of drugs when compared to HIV-infected individuals. Furthermore, since indinavir is an important component of triple combination therapy, pharmacokinetic data on the interaction with sildenafil are required.
On the first study day, marked interpatient variability in plasma indinavir levels was noted which is characteristic of drugs metabolized by CYP3A4 and has been previously reported with other protease inhibitors[21-23]. In the presence of sildenafil, there was a 11 and 47.6% increase in the AUC0-8h and Cmax of indinavir, respectively. However these increases in plasma indinavir levels are not statistically significant and will not be expected to result in an increased risk of nephrolithiasis.
All patients participating in this study were at steady state for indinavir. Concerns about the emergence of drug-resistant viral strains precluded the discontinuation of indinavir and therefore baseline sildenafil profiles were not included in the study design. Data from historical controls for doses of sildenafil of 50 and 100 mg which were dose normalized to 25 mg were used for comparison[15,19]. Co-administration of sildenafil 25 mg with indinavir resulted in plasma sildenafil levels that were 4.4 times higher than the dose-normalized data taken from the literature. In fact, co-administration of sildenafil 25 mg with indinavir resulted in plasma sildenafil levels in excess of those reached when sildenafil was administered at a dose of 100 mg in the absence of indinavir. The magnitude of this interaction is interesting as there is no improvement in efficacy with increasing the dose of sildenafil from 100 to 200 mg, but the incidence of adverse effects (i.e. headache, flushing, dyspepsia, nasal congestion, diarrhoea, dizziness) increased. The elevated sildenafil levels noted on the second study day were consistent with the clinical observation that all six study participants experienced the following sildenafil-related adverse effects: headache, flushing, dyspepsia and rhinitis. The mean maximal decrease in blood pressure noted in this study was 14/10mm Hg which is greater than the maximum decreases noted following 100mg oral dosing of sildenafil in other clinical studies. Interestingly, two of the six patients reported pharmacodynamic effects of sildenafil up to 72h after ingestion of a single dose. However, it is difficult to explain this observation solely on the basis of terminal half-life of the parent compound (2.62h) and may be due in part to the active metabolite (UK-103,320).
A potential mechanism for the increased sildenafil concentrations is provided by the in vitro data. Indinavir produced marked inhibition of the hepatic metabolism of sildenafil with an IC50 value of 0.39±0.17μM. Note that peak concentrations of indinavir in this study were approximately 15μM. Ritonavir was a more potent inhibitor in vitro (IC50 = 0.034±0.005μM) which is consistent with the pharmacokinetic data.
Despite the obvious limitations of this study which compares data in two very different groups of subjects, with the sildenafil concentrations being assayed in different laboratories and at different times, the marked increase in plasma sildenafil levels in the presence of indinavir, suggests that a lower starting dose of 12.5 mg may be more appropriate to minimize dose-related toxicity while preserving therapeutic efficacy. Furthermore, in view of the prolongation of the pharmacodynamic effect noted in this study, the possibility of drug accumulation with the attendant risk of priapism, dizziness and hypotension associated with daily dosing of sildenafil is a cause of concern. Once or twice weekly administration of sildenafil may perhaps be more appropriate in this clinical setting.
1. NIH. NIH Consensus Development Panel on Impotence.JAMA
2. Slag MF, Morley JE, Elson MK. Impotence in medical clinic outpatients.JAMA
3. Feldman HA, Goldstein I, Hatzichristou DG, Krane RJ, McKinlay JB. Impotence and its medical and psychosocial correlates: results of the Massachusetts male ageing study.J Urol
4. Tindall B, Forde S, Goldstein D, Ross MW, Cooper DA. Sexual dysfunction in advanced HIV disease. AIDS Care
5. Dobs AS, Dempsey MA, Ladenson PW, Polk BF. Endocrine disorders in men infected with human immunodeficiency virus.Am J Med
6. Chuang HT, Devins GM, Hunsley J, Gill MJ. Psychosocial distress and well-being among gay and bisexual men with human immunodeficiency virus infection.Am J Psychiatry
7. Croxson TS, Chapman WE, Miller LK, Levit CD, Senie R, Zumoff B. Changes in the hypothalamic-pituitary-gonadal axis in human immunodeficiency infected homosexual men.J Clin EndocrinolMetab
8. Martinez E, Collazos J, Mayo J, Blanco M. Sexual dysfunction with protease inhibitors.Lancet
9. Goldstein I, Lue TF, Padma-Nathan H, Rosen RC, Steers WD, Wicker PA. Oral sildenafil in the treatment of erectile dysfunction. New Engl J Med
10. Ballard SA, Gingell CJ, Tang K, Turner LA, Price ME, Naylor AM. Effects of sildenafil on the relaxation of human corpus cavernosum tissue in vitro and on the activities of cyclic nucleotide phosphodiesterase isozymes.J Urol
11. Terrett NK, Bell AS, Brown D, Ellis P. Sildenafil (Viagra®), a potent and selective inhibitor of type 5 cGMP phosphodiesterase with utility for the treatment of male erectile dysfunction.Bioorg Med Chem Lett
12. Goldenberg MM. Safety and efficacy of sildenafil citrate in the treatment of male erectile dysfunction. Clin Ther
13. Morales A, Gingell C, Collins M, Wicker PA, Osterloh IH. Clinical safety of oral sildenafil citrate (Viagra) in the treatment of erectile dysfunction.Int J Impot Res
14. Nandwani R, Gourlay Y. Possible interaction between sildenafil and HIV combination therapy.Lancet
15. Pfizer. Viagra US Prescribing Information.
Revised March 1999. Available from: URL: http://http://www.pfizer.com
16. National AIDS Treatment Advocacy Project. News.
Available from: URL: http://http://www.natap.org
17. Eagling VA, Back DJ, Barry MG. Differential inhibition of cytochrome P450 isoforms by the protease inhibitors, ritonavir, saquinavir and indinavir.Br J Clin Pharmacol
18. Barry M, Mulcahy F, Merry C, Gibbons S, Back DJ. Pharmacokinetics and potential interactions amongst antiretroviral agents used to treat patients with HIV infection.Clin Pharmacokinet
19. Cooper JD, Muirhead DC, Taylor JE, Baker PR. Development of an assay for the simultaneous determination of sildenafil (Viagra) and its metabolite (UK-103, 320) using automated sequential trace enrichment of diasylates and high- performance liquid chromatography.J Chromatogr B Biomed Sci Appl
20. Gregoire A. Viagra: on release. Evidence on the effectiveness of sildenafil is good.Br Med J
21. Merry C, Barry MG, Mulcahy FM, et al.Ritonavir pharmacokinetics alone and in combination with saquinavir in HIV-infected patients [letter]. AIDS
22. Merry C, Barry MG, Mulcahy FM, Halifax KL, Back DJ. Saquinavir pharmacokinetics alone and in combination with nelfinavir in HIV-infected patients.AIDS
23. Merry C, Barry MG, Mulcahy FM, et al
. Saquinavir pharmacokinetics alone and in combination with ritonavir in HIV-infected patients.AIDS