A novel testosterone gel formulation normalizes androgen levels in hypogonadal men, with improvements in body composition and sexual function

Transcription

1 Original Article TESTOSTERONE GEL NORMALIZES ANDROGEN LEVELS T.A. McNICHOLAS et al. Men s health has become so important that it is now virtually a political issue. With advancing age, testosterone levels decrease and this has an effect on body composition, mood, and sexual function. There have been many attempts to counteract what has now become known as the andropause. Many questions need to be answered. For example, does the serum testosterone level need to be restored to normal? Is it always abnormal in older men? Will this restoration to normal levels enhance well being without incurring other risks? Are gels more effective than patches? Not all of these questions can be answered in a single paper, but some of them are. However, the big question remains: will men s health continue to be of major importance, and will testosterone replacement be generally accepted? A novel testosterone gel formulation normalizes androgen levels in hypogonadal men, with improvements in body composition and sexual function T.A. McNICHOLAS, J.D. DEAN*, H. MULDER, C. CARNEGIE and N.A. JONES Department of Urology, Lister Hospital, Stevenage, Herts, *The Salisbury Clinic, Plymouth, Devon, Osteosupport, Rotterdam, the Netherlands and Auxilium UK Ltd, Brinkworth, Wiltshire, UK Accepted for publication 3 October 2002 OBJECTIVE To compare the safety and efficacy of two doses of a new testosterone gel formulation (Testim, Auxilium Pharmaceuticals, Inc., Norristown, PA, USA) to a permeation-enhanced testosterone patch (Andropatch, GlaxoSmithKline, UK) for treating men with confirmed low serum testosterone levels, and associated signs and symptoms of hypogonadism. PATIENTS AND METHODS In all, 208 men were randomized and treated at 29 centres in Denmark, Germany, Netherlands, Sweden and the UK. The men were treated for 90 days, and the pharmacokinetics and treatment effectiveness of Testim at two doses (50 and 100 mg/day, delivering a daily dose of 5 and 10 mg testosterone, respectively) and Andropatch (2 2.5 mg patches, each delivering 2.5 mg testosterone and containing 12.2 mg of testosterone) were compared. Pharmacokinetic profiles were obtained, body composition measured, and mood and sexual function data recorded. mean increases from baseline to 90 days in testosterone were 12.41, 6.54 and 3.82 nmol/l for Testim 100 and 50 mg/day and the Andropatch, respectively. Both doses of Testim significantly improved positive and negative mood over baseline; Andropatch did not. All three treatments increased lean body mass, and the higher dose of Testim produced a significant decrease in percentage body fat. At all sample times both doses of Testim significantly improved sexual performance, sexual motivation, sexual desire and spontaneous erections. Andropatch provided insignificant improvements from baseline at all sample times for sexual desire, an inconsistent improvement in sexual motivation, but no effect on spontaneous erections. These results are similar to those previously reported for testosterone replacement therapy in hypogonadal men, suggesting that normalization of serum testosterone restores sexual function. However, the present data suggest that higher serum testosterone levels may further improve sexual function. Gel treatment was well tolerated, while patch treatment produced higher rates of application-site reactions and study discontinuation. RESULTS Testim produced dose-dependent improvements in all pharmacokinetic variables compared with Andropatch. The CONCLUSION The favourable pharmacokinetic profile and treatment outcome, combined with the enhanced tolerability of Testim, suggest that 2003 BJU INTERNATIONAL 91, doi: /j x 69

2 T.A. McNICHOLAS ET AL. this new gel formulation is a safe and effective treatment in men with low serum testosterone levels and associated signs and symptoms of hypogonadism. KEYWORDS levels of testosterone with less fluctuation than with Androderm and had significant clinical benefits over placebo for muscle mass and sexual function variables. Testim was also better tolerated than Androderm because it produced fewer adverse skin reactions [9]. testosterone (i.e. fatigue, decreased muscle mass, reduced libido, and reduced sexual functioning that was not mechanical). Except for hypogonadism, the men were generally in good health. All men signed an informed consent form previously approved by an Independent Ethics Committee. testosterone, clinical trial, male, gel formulation, hypogonadism, sexual function, body composition INTRODUCTION It has been previously shown that testosterone levels decrease with ageing and that the decrease in testosterone below levels considered physiologically eugonadal for young mature men has a deleterious effect on sexual characteristics, sexual behaviour, energy levels, mood, muscle development and possibly bone density. Restoring testosterone levels ameliorates these deleterious affects [1 5]. The therapeutic methods currently available in Europe for replacing testosterone in the hypogonadal state are intramuscular injections, oral formulations or the transdermal patch. Each of these methods of testosterone supplementation have deficiencies; intramuscular injections are painful, require repeated office visits for injection, and produces high levels of serum testosterone initially that decline below physiological levels towards the end of the injection cycle. The oral formulations of testosterone have a high first-pass liver metabolism, and certain oral preparations have significant liver toxicity. The transdermal patches, while providing relatively consistent testosterone levels, produce variability in therapeutic testosterone levels over the length of the day, and produce significant skin irritation with prolonged use [6 8]. A unique topical testosterone gel (Testim, Auxilium Pharmaceuticals, Inc., Norristown, PA, USA) was developed to circumvent the shortcomings of the other available forms of testosterone delivery, and to provide consistent therapeutic levels of testosterone therapy. In a multiple-dose, active (2 2.5 mg of testosterone in transdermal patches, Androderm, GlaxoSmithKline, USA) and placebo-controlled study in 406 hypogonadal males, conducted in the USA, Testim at doses of 50 and 100 mg/day provided higher The present study comprised ageing men with low serum testosterone levels ( 10.4 nmol/l) and associated signs and symptoms of hypogonadism. We compared three parallel groups treated with Testim at two doses or Andropatch (GlaxoSmithKline). The study was designed to evaluate and compare the efficacy of transdermal testosterone preparations in normalizing serum testosterone, and ameliorating the signs and symptoms of hypogonadism. PATIENTS AND METHODS In all, 208 men were randomized and treated at 29 centres in Denmark, Germany, the Netherlands, Sweden and the UK, with 70 in each treatment group (Table 1). The men were aged years and had a morning serum testosterone level of 10.4 nmol/l at screening (measured at a central laboratory) and one or more symptoms of low TABLE 1 The characteristics of the men The study was designed as a randomized, multidose, multicentre, active controlled study. The three daily treatments under study were Testim 50 and 100 mg (delivering a daily dose of 5 mg and 10 mg testosterone, respectively), and a transdermal testosterone patch (Andropatch, two patches each delivering 2.5 mg testosterone and containing 12.2 mg). The Testim treatments were doubleblinded by using a matched placebo gel but was open-label for the Andropatch group. All study drug treatments were applied in the morning; repeat applications were administered at the same time of day for the duration of the study. On the day before starting treatment the men had a baseline 24-h profile for serum testosterone, free testosterone and dihydrotestosterone (DHT), consisting of serum samples taken at 08.00, 10.00, and hours, the 24 h sample being taken at hours on the first day, immediately Variable Testim, mg/day Andropatch Total Demographics Number of men Mean (SD): Age, year 59.0 (9.5) 56.7 (10.3) 57.9 (10.8) 57.9 (10.2) Height, cm 177 (6) 178 (7) 178 (7) 177 (7) Weight, kg 85.9 (12.1) 88.1 (11.6) 89.2 (12.3) 87.7 (12.0) Body mass index 27.7 (3.6) 27.9 (4.0) 28.3 (3.6) 28.0 (3.7) Testosterone, nmol/l 7.95 (2.17) 7.92 (2.45) 7.90 (2.23) 7.92 (2.28) IPSS 4.8 (4.9) 5.3 (4.7) 4.8 (5.4) 5.0 (5.0) PSA, ng/ml 1.17 (1.23) 1.18 (1.22) 1.40 (1.37) 1.25 (1.27) % of men by PSA level: > Cause of hypogonadism, n Primary Secondary *Ageing, % *Normogonadotrophic, % *% of total by treatment group; distribution by cause is shown only if it occurred in >5% of men BJU INTERNATIONAL

3 TESTOSTERONE GEL NORMALIZES ANDROGEN LEVELS before the first dose of study drug. At 30, 60 and 90 days, the 24-h profile was repeated for the three forms before dosing and at 2, 4, 8 and 24 h after the study drug was administered. The prostate status was evaluated at screening (PSA levels only), and the day before and 90 days after treatment by PSA levels, the IPSS and a DRE. Body composition, i.e. lean body mass, fat mass and percentage of body fat (%fat), and bone mineral density of the L1 L4 section of the lumbar spine were measured by dual-energy X-ray absorptiometry at 1, 60 and 90 days. The sexual function and mood questionnaires were recorded daily for 14 days before starting the drug and then daily for 7 days before the 30, 60 and 90 days assessment. Skin irritation was examined using a standardized, discrete scoring system before dosing (day 1) and at 30, 60 and 90 days afterward. The medical history and physical examinations were conducted and all adverse events recorded at each visit. The skin irritation scoring at the application site was as follows: 0, no erythema; 1, minimal erythema; 2, moderate erythema with sharply defined borders; 3, intense erythema with or with no oedema; 4, intense erythema with oedema and blistering. Sexual functioning and mood assessments were based on a questionnaire that had previously been validated for assessing sexual function and mood, and used previously for evaluating the effects of testosterone gel on sexual function and mood [1]. The questionnaire elicited information on sexual function, i.e. performance, motivation, spontaneous erections, desire, enjoyment (with and without a partner), and satisfaction with erection duration and percentage of full erection. The men also rated positive mood (alert, full of energy, friendly, well, or good) and negative mood (angry, irritable, sad or depressed, tired, nervous). Serum testosterone and derivatives were all measured using validated radioimmunoassay kits (Diagnostic Products Corp., Los Angeles, CA for testosterone and free testosterone, and Diagnostic Systems Laboratories, Webster, TX for DHT). STATISTICS The 24-h pharmacokinetic profiles for testosterone and DHT were summarized by average, minimum and maximum concentrations (C avg, C min and C max ) after dosing. The changes from baseline to 30, 60 and 90 days in C min, C avg, and C max were analysed using an analysis of covariance (ANCOVA) with the baseline value as the covariate and treatment group as the factor. Similar analyses were used for the change from baseline in sexual function, mood and body composition, and for the clinical laboratory variables at 30, 60 and 90 days. Treatment-emergent adverse events were compared using Fisher's exact test. Skin irritation at each sample times was analysed using the Wilcoxon rank-sum test. At 30, 60 and 90 days both the 50 and 100 mg Testim groups were compared separately with the Andropatch group using an α level of 0.05; all comparisons for safety variables also used this level. The changes from baseline in sexual function, body composition and mood were also analysed for non-zero differences within each treatment group, based on the adjusted least squares means from the ANCOVA model. RESULTS In all, 208 men were randomized, with 68, 72 and 68 in the Testim 50, Testim 100 mg/day and Andropatch groups, respectively (Table 1). The baseline characteristics were comparable across all three groups. The causes of hypogonadism were primarily attributed to the secondary causes of ageing and normogonadotrophic hypogonadism changes (Table 1). A significant proportion of enrolled men completed the 90-day study (93% and 97% in the Testim 50 and 100 mg/day groups, respectively, and 79% in the Andropatch group). The primary reason for the higher rate of discontinuation in the Andropatch group was adverse events (13%) with most being related to skin irritation at the patch site. Dosing compliance over the completed dosing periods ranged from 96% (Andropatch) to 97% (Testim 100 mg/day). ANDROGEN LEVELS Testim increased serum testosterone levels more than did Andropatch and generally in a dose-dependent manner (Table 2). Consistently throughout the study, Testim 100 mg/day increased C avg, C min and C max more than 50 mg/day and Andropatch. At 90 days the mean change from baseline with the 100 mg/day Testim in testosterone C avg was over three times greater than that with Andropatch (P < 0.001) and about double the increase seen with the 50 mg/day Testim (P < 0.05). The results at 30 and 60 days were consistent with those at 90 days. The mean changes in DHT (C avg, C min and C max ) from baseline at 90 days are also shown in Table 2. Testim had similar effects on DHT as on testosterone, with significant consistent increases from baseline throughout the study and in a dose-dependent manner. Conversely, treatment with Andropatch produced a relatively small increase in all three variables for DHT. At 90 days the change in DHT C avg was greater with 100 and 50 mg/day Testim than with Andropatch (P < for both comparisons, Table 2). The results from 30 and 60 days were consistent with those at 90 days. The data for free testosterone are also shown in Table 2 and were comparable with those for testosterone. Testim increased free testosterone levels more than did the Andropatch and in a dose-dependent manner; the mean change in C avg at 90 days with 100 mg/day Testim was more than twice that with 50 mg/day Testim (P < 0.001) and more than three times that with Andropatch (P < 0.001). The results at 30 and 60 days were consistent with those at 90 days. BODY COMPOSITION For lean body mass, at 90 days, as at 60, there were significant within-treatment group changes from baseline for all three treatments. The only other significant withintreatment group change from baseline was in the 100 mg/day Testim group for %fat (P < 0.01), where there was a reduction of 0.12%. The increase in lean body mass in the 100 mg/day Testim group was 2.5% (1.5 kg). The 90 day body composition results are also shown in Table 2. Not surprisingly in a 90-day study, there were no statistically significant changes in bone mineral density within any of the three treatment groups. MOOD Although there were no statistically significant treatment differences among the three groups the 100 mg/day Testim group had significant increases from baseline in positive mood at all samples times, while the 50 mg/day Testim group had significant increases at 30 and 90 days. The Andropatch group had no statistically significant changes from baseline at any time. The alleviation of negative mood, as shown by a reduction in negative mood scores, was consistent in the 50 mg/day Testim group at all times from baseline and for the 100 mg/ day group at 60 and 90 days. In contrast there 2003 BJU INTERNATIONAL 71

4 T.A. McNICHOLAS ET AL. was no statistically significant within-group change for Andropatch at any time. There were significant differences between treatment groups at 90 days in favour of Testim over the Andropatch (P < 0.05). SEXUAL FUNCTION For those sexual function variables assessed the results showed that all treatment groups had a significant improvement in sexual function throughout the study, although there were few significant differences between treatment groups at any time. At 90 days there were significant improvements within treatments from baseline for all three groups in sexual motivation, sexual desire and sexual performance (Table 2). While there were no statistically significant differences among treatment groups, both Testim groups had a statistically significant within-treatment improvement in spontaneous erections at all times from baseline (P < 0.05). Andropatch produced no statistically significant within-treatment changes from baseline at any time. Men treated with Testim had about twice the incidence of spontaneous erections at 30 and 60 days than at baseline. Men treated with Testim had a consistent improvement in sexual motivation throughout the 90-day period, with significant differences from baseline at all times, but the Andropatch group had a less consistent improvement, with no differences from baseline at 60 days. Men in all three groups had a consistent and significant increase in sexual desire throughout the study. Men treated with Testim generally had a greater and more consistent improvement in sexual performance than those treated with Andropatch. At 30 days there was a significant increase with 100 mg/day Testim group vs the Andropatch (P < 0.05). The change from baseline at 90 days is shown in Table 2, with significant and clinically meaningful improvements of 50%, 38% and 33% in the 100, 50 mg/day Testim and Andropatch groups, respectively (P < 0.001, < 0.05 and < 0.05). SAFETY Overall, the incidence of experiencing one or more treatment-emergent adverse events was similar in the Testim groups, but was about twice as high in the Andropatch group, at 35% for 50 and 29% for 100 mg/day TABLE 2 The baseline values and change ( ) at 90 days for serum testosterone, DHT, sexual function scores and body composition variables Mean (SD) variable Testim, mg/day Andropatch Testosterone, nmol/l C avg 7.84 (3.25) 7.80 (3.35) 8.73 (3.17) C min 6.34 (2.85) 6.42 (3.00) 6.90 (2.55) C max 9.98 (3.85) 9.66 (4.12) (3.78) C avg 6.54 (8.62) (12.56) 3.82 (4.39) C min 2.98 (5.64)* 7.14 (8.64) 0.35 (3.55) C max (22.85) (22.27) * 8.23 (7.76) DHT, nmol/l C avg 0.50 (0.23) 0.49 (0.27) 0.52 (0.22) C min 0.37 (0.20) 0.35 (0.20) 0.38 (0.18) C max 0.66 (0.28) 0.63 (0.33) 0.67 (0.30) C avg 0.91 (0.94) 1.39 (1.16) 0.03 (0.25) C min 0.67 (0.76) 1.04 (0.93) < 0.01 (0.20) C max 1.17 (1.35) 1.85 (1.59) 0.05 (0.34) Free testosterone, pmol/l C avg (13.41) (13.41) (12.34) C min (11.08) (11.61) (9.02) C max (17.01) (17.68) (14.38) C avg (32.09) (48.50) (18.21) C min 9.14 (17.48) (33.57) 2.06 (13.42) C max (79.19) (81.92)* (33.54) Sexual function scores: Spontaneous erection (mean/week) 0.8 (1.0) 0.8 (1.1) 0.9 (1.0) 0.6 (1.4) 0.5 (1.4) 0.3 (1.2) Motivation (mean/week) 1.5 (1.4) 1.5 (1.3) 1.3 (1.1) 0.4 (1.2) 0.4 (1.3) 0.5 (1.4) Desire (mean/day) 1.6 (1.1) 1.9 (1.4) 1.8 (1.2) 0.8 (1.0) 0.7 (1.4) 0.5 (1.2) Performance (mean/week) 0.8 (0.9) 0.8 (1.0) 0.9 (1.0) 0.3 (1.0) 0.4 (0.9) 0.3 (1.0) Body composition: Lean body mass 57.4 (11.3) 57.2 (11.4) 58.5 (11.7) 0.9 (1.8) 1.5 (1.7) 1.0 (1.9) Fat mass 24.1 (10.9) 26.2 (11.3) 26.6 (13.2) 0.1 (1.8) 0.2 (1.5) 0.1 (2.0) Percentage fat 28.2 (11.9) 30.0 (11.6) 29.5 (12.5) 0.4 (2.0) 0.7 (1.3) 0.3 (1.9) Significant vs Andropatch *P < 0.05, < 0.001; significant for Testim 50 vs 100 mg/day, P <0.05, < 0.001; significant within-treatment group change from baseline P <0.05, < Testim, and 63% for the Andropatch. Whilst treatment in the Testim groups was relatively well-tolerated over the 90-day period, the Andropatch treated men had a substantially higher rate of adverse events. Those most commonly reported were erythema, irritation and reactions at the application site. In the Andropatch group there was a higher frequency of withdrawal from the study because of adverse events, i.e. 13% with Andropatch and 4% in the 50 and none in the 100 mg/day Testim groups. All of the discontinuations for adverse events in the Andropatch group were caused by local tolerability problems associated with the patch application site. LABORATORY ANALYSES The incidence of clinically notable laboratory values was low for all treated men. There were small decreases in total cholesterol and highdensity lipoprotein (HDL) which were doserelated; there were significant differences BJU INTERNATIONAL

5 TESTOSTERONE GEL NORMALIZES ANDROGEN LEVELS FIG. 1. The distribution of men with positive application-site irritation scores at 30 (open bars), 60 (green bars) and 90 days (red stippled bars). The scoring system is detailed in the text. There were significant differences (P < 0.001) for each of the Testim vs Andropatch comparisons at each time. between the 100 and the 50 mg/day Testim dose (P < 0.05) but not compared with Andropatch. There was no significant effect on the total cholesterol/hdl ratio for any of the three treatments. Increases in haemoglobin, haematocrit, and red blood cells (RBCs) are known pharmacological effects of testosterone [1,5]. Consistent with this, men in the 100 mg/day Testim group had statistically significant increases from baseline to 90 days in haematocrit and haemoglobin, compared with both the 50 mg/day Testim and Andropatch groups (P < for haematocrit and RBC count, P < 0.01 for haemoglobin, both group comparisons). The effects with Andropatch treatment were consistently smaller than those with Testim, reflecting the lower mean serum testosterone levels with Andropatch treatment. There were also statistically significant mean changes for eosinophils and basophils, but the changes were minor and of no apparent clinical significance. The mean changes in PSA at 90 days were relatively small; they were >4.0 ng/ml at least once during treatment in 3%, 4% and 3% of men in the 50 and 100 mg/day Testim and Andropatch groups, respectively. These PSA changes were consistent with the known effects of testosterone therapy [1,5,9]. Most increases in PSA were <1 ng/ml and levels stayed below that considered to be abnormal (4.0 ng/ml). The changes from baseline IPSS were very small and the incidence of men having worse findings on a DRE was low. No variables provided evidence of clinically relevant treatment-related effects or differences. SKIN IRRITATION Number of men Figure 1 shows the frequency distribution of men with positive skin irritation scores. As Score Testim Testim Andropatch 50 mg/day 100 mg/day 5 mg/day noted, the scoring system was based on a five-point categorical series; overall, the Testim-treated groups had a very low incidence of skin irritation. Doubling the dose of Testim (to 100 mg/day) did not increase the incidence of skin irritation events. In comparison, the Andropatch group had a substantially higher incidence of skin irritation, with 47% of men reporting irritation at 30 days and 53% reporting it at 90 days. Furthermore, the reported irritation was more severe, with many men categorizing it as moderate to intense. DISCUSSION This study showed that testosterone gel was superior to the marketed comparison Andropatch; Testim 100 mg/day was better than Andropatch for normalizing serum testosterone levels. Furthermore, there was a clear dose response, in that Testim 100 mg/day was better than 50 mg/day for normalizing serum testosterone levels. Comparable results were obtained with serum levels of DHT and free testosterone. While both doses of Testim produced similar statistically significant increases over baseline in lean body mass, the higher dose of Testim also produced a statistically significant decrease over baseline in %fat. Both doses of Testim gave statistically significant improvements in positive and negative mood over baseline; there were no similar effects on mood with Andropatch. The therapeutic effect was further substantiated because both doses of Testim improved the sexual variables, improving sexual performance throughout the study at statistically significant levels, whereas Andropatch had no effect at 30 days. At all three times during the study, both doses of Testim produced significant improvements from baseline in sexual performance, sexual motivation, sexual desire, and spontaneous erections. Such improvements after restoring serum testosterone to normal levels are likely to be clinically meaningful in older men with low testosterone levels. Andropatch produced inconsistent improvements over baseline in sexual motivation, significant improvements at all times for sexual desire, but no effect at any time on spontaneous erections. The men had potentially beneficial and known pharmacological class effects of androgens, i.e. raising the levels of haematocrit, haemoglobin and RBCs. There were small mean increases in PSA in all groups, but no effect on the IPSS. There were small but notable dose-related decreases in total cholesterol and HDL with Testim, and no significant changes in the total cholesterol/ HDL ratio. There were more application-site reactions with Andropatch than with Testim; more men treated with Andropatch discontinued treatment because of these reactions than with Testim. Both doses of Testim improved mood, lean body mass, sexual performance, sexual desire, sexual motivation and spontaneous erection. Finally, men complied better with Testim as a result of the lower incidence of dermal irritation, which is an important advantage for a potentially longterm therapy. This study confirms that Testim is a viable and better alternative to the currently available therapeutic methods in Europe for testosterone replacement. ACKNOWLEDGEMENTS This study was sponsored by Auxilium UK Limited. Auxilium UK Limited thanks the following investigators for their participation and contribution to this clinical study. Dr M. Callander, Dr J. Dean, Dr G. Hackett, Dr M. Kirby, Dr A. Levy, Mr T. McNicholas, Dr D. Price, Dr R. Quinton, Dr T. Robinson, Dr J. Tomlinson (UK); Dr H. Bulk, Dr B. Jan de Boer, Dr J. Jonker, Dr H. Mulder, Dr P. van der Graaff (Netherlands); Dr U. Ekstrom, Dr M. Haggmann, Dr J. Oldbring (Sweden); Dr T. Gerstenberg, Dr J. Otto Jorgensen, Dr J. Poulsen (Denmark); Dr P. Malsy-Mink, Prof H. Porst, Dr D. Quast, Dr B. Rosenkranz, Prof W. B. Schill, Dr A. von Keitz, Dr P. Weitz, Dr R. Zillmann (Germany). REFERENCES 1 Wang C, Swerdloff R, Iranmanesh A et al. Transdermal testosterone gel improves 2003 BJU INTERNATIONAL 73

6 T.A. McNICHOLAS ET AL. sexual function, mood, muscle strength, and body composition parameters in hypogonadal men. J Clin Endocrinol Metab 2000; 85: Arver S, Dobs AS, Meikle AW et al. Longterm efficacy and safety of a permeationenhanced testosterone transdermal system in hypogonadal men. Clin Endocrinol 1997; 47: Snyder PJ, Peachey H, Hannoush P et al. Effect of testosterone treatment on bone mineral density in men over 65 years of age. J Clin Endocrinol Metab 1999; 84: Snyder PJ, Peachey H, Hannoush P et al. Effect of testosterone treatment on body composition and muscle strength in men over 65 years of age. J Clin Endocrinol Metab 1999; 84: Tenover JL. Testosterone and the aging male. J Androl 1997; 18: Jordan WP, Atkinson LE, Lai C. Comparison of the skin irritation potential of two testosterone transdermal systems: an investigational system and a marketed product. Clin Ther 1998; 20: Jordan WP. Allergy and topical irritation associated with transdermal testosterone administration: a comparison of scrotal and non-scrotal transdermal systems. Am J Contact Dermatol 1997; 8: Dobs AS, Meikle AW, Arver S, Sanders SW, Caramelli KE, Mazer NA. Pharmacokinetics, efficacy, and safety of a permeation-enhanced testosterone transdermal system in comparison with biweekly injections of testosterone enanthanate for the treatment of hypogonadal men. J Clin Endocrinol Metab 1999; 84: Steidle C, Schwartz S, Jacoby K, Sebree T, Smith T, Bachand R. Testim normalizes androgen levels in aging males with improvements in body composition and sexual function. J Clin Endocrinol Metab 2003; in press Correspondence: J.D. Dean, The Salisbury Clinic, 43 Salisbury Road, Plymouth, Devon PL4 8QU, UK johndean@netcomuk.co.uk Abbreviations: DHT, dihydrotestosterone; %fat, percentage of body fat; ANCOVA, analysis of covariance; HDL, high-density lipoprotein; RBCs, red blood cells BJU INTERNATIONAL