Effect of antiretroviral drugs on the quality of semen^†

INTRODUCTION

Potent combination antiretroviral therapy has increased the life expectancy and quality of life of patients infected with HIV-1. Currently, some couples request medical assistance to achieve conception while reducing to a minimum the risk of HIV-1 transmission. This technical requirement for Assisted Reproductive Technology program raises the question of semen characteristics in men infected with HIV-1 under antiretroviral treatment. Several studies have assessed semen quality in men infected with HIV-1 [Dulioust et al., 2002; Nicopoullos et al., 2004; Bujan et al., 2007]. However, the use of antiretroviral drugs (ARVs) did not appear to be correlated with any sperm characteristics while most of the patients were receiving antiretroviral therapy. A practical method was developed for quantifying the penetration in genital tract of the many different ARV drug regimens currently prescribed, as estimated by the penetration score. This method uses publicly available information about ARV drug characteristics, measured semen concentrations, and effectiveness in the genital tract to rank drugs relative to one another. ARVs showed different penetration score in the male genital tract and may therefore influence differently spermatogenesis, or semen quality which is a key factor for reproductive success.

The aim of this cross-sectional cohort study was undertaken to determine which ARVs were associated with changes in semen characteristics and the impact of these ARVs according to their penetration score into the male genital compartment.

PATIENTS AND METHODS

One hundred and forty four men infected with HIV-1 attending the Pitié-Salpêtrière Hospital in the multidisciplinary assisted reproductive technology program between January 2002 and September 2007 were studied retrospectively. HIV-1 RNA quantitation was determined by using the Amplicor Monitor assay (Cobas 1.5; Roche Diagnostics, Basel, Switzerland), which has a detection limit of 200 copies/ml in seminal plasma.

All semen samples were tested in the Unit of Reproductive Biology according to standardized methods throughout the study period and according to the World Health Organization (WHO) guidelines [WHO, 1992].

Semen samples were collected in the laboratory by masturbation into a sterile graduated container after 3–5 days of sexual abstinence. After 30 min at 35°C, each sample was examined. The volume of the seminal fluid, semen pH and viscosity were determined. Motility was graded (a) rapid progressive motility, (b) slow progressive motility, (c) non-progressive motility, and (d) no motility according the WHO guidelines. The percentage of motile spermatozoa was determined at T0 (30–45 min after semen collection) and at T1 (4–5 hr after semen collection). Sperm and round cell concentrations were determined using a Kova Slide (HYCOR Biomedical, Garden Grove, CA). Sperm vitality was assessed after staining with Eosin–Nigrosin. The multiple abnormalities index represents the ratio of the number of sperm abnormalities, over the number of abnormal spermatozoa. Teratospermia is the opposite of the percentage of morphologically normal spermatozoa.

The present study was carried out in accordance with the declaration of Helsinki and was approved by ANRS AC11 Resistance Study Group scientific committee. All patients gave written informed consent.

Penetration of ARVs was characterized using a hierarchical approach based on the best available evidence. Data on chemical and pharmacokinetic characteristics were reviewed for ARVs approved by national agencies using package inserts, drug references, published papers, and international conference abstracts [Taylor et al., 2001; Pereira et al., 2002; Solas et al., 2003]. Using this approach, ARVs were classified into three categories. Individual ARVs were assigned a rank based on penetration category (0 = lowest, 0.5 = intermediate and 1 = highest penetration, Table I). The final rank was then determined by adding the individual penetration scores for each ARV drug in a regimen.

Continuous variables were compared using the Kruskal–Wallis test while categorical variables were compared using chi-square test. All analyses used a two-sided α of 0.05. Statistical analyses were performed with SAS release 9.1 for Windows XP Pro.

RESULTS

The main characteristics of the study population are described in Table II. At the time of semen collection, 124 men (86%) were receiving ARVs (dual therapy: three men; triple therapy: 109 men; ≥4 drugs: 12 men). The median number of drugs used was two nucleosides reverse transcriptase inhibitors NRTIS (range, 0–4), 1 protease inhibitor PI (range, 0–1) and 0 non nucleoside reverse transcriptase inhibitors NNRTI (range, 0–1).

Semen characteristics are shown in Table II. The characteristics of seminal fluid ranged from 0.13 to 8.50 ml (mean 3.5) for volume and 7.4–9.0 (mean 8.0) for pH. Majority of ejaculates 83.9% (115/137) had a normal viscosity. The concentration of spermatozoa ranged from 2.61 to 676 million/ml (mean 116). The concentration of round cells ranged from 0 to 36 10⁶/ml (mean 2.79). The total sperm count was ranged from 1.3 to 2,000 10⁶ per ejaculate (mean 385.3). The mean percentage of rapid progressively motile spermatozoa was 16.6 (range 0–60) and the mean percentage of slow progressively motile spermatozoa was 27.3 (range 5–75). The mean percentage of vitality was 78.6 (range 35–98). The mean of multiple abnormality index was 1.70 (range 1.17–2.86). The percentage of abnormal sperm morphology was ranged from 30 to 99 (mean 69). Twenty-four % of patient (35 out of 144) had all their sperm characteristics strictly normal according to WHO standards.

Studying the different variables of the quality of semen, no relationship between semen quality and the penetration score was found. There was no difference in the characteristics of sperm between patients receiving NRTI, NNRTI, or PI regimen. In patients receiving NRTIs or PIs no differences were observed between the ARVs of these classes. No association was shown between thymidine analogs (D4T or AZT) or other NRTIs and semen characteristics. However, considering only the NNRTI containing regimen, it was shown that nevirapine was associated with a higher percentage of both progressively motile spermatozoa [at T0 (immediately after liquefaction) and at T1 (4 hr after liquefaction)] and vitality versus efavirenz. Regarding the mean percentage of progressively motile spermatozoa both at T0 and T1, nevirapine (22.05 and 15.24) and PI (17.31 and 12.71) provided similar values while efavirenz (7.92 and 3.81) led to smaller values. The same trend was observed in the percentage of vitality with a mean of 85.3 for nevirapine, 72.2 for efavirenz, and 80.1 for PI. The difference observed between nevirapine and efavirenz does not seem to be related to the NRTI-associated treatment. Indeed, thymidine analogues were distributed equally between nevirapine and efavirenz groups. The known characteristics of patients were similar between nevirapine and efavirenz groups.

DISCUSSION

In the present study, the characteristics of sperm were the same as those found in other studies [Dulioust et al., 2002; Nicopoullos et al., 2004; Bujan et al., 2007; van Leeuwen et al., 2008b] except for sperm concentration which was higher. The same trend was observed for the total sperm count. Several studies have assessed semen quality in men infected with HIV-1 in comparison to HIV-1 seronegative men [Dulioust et al., 2002; Nicopoullos et al., 2004; Bujan et al., 2007]. The study by Bujan et al. [2007] demonstrated decreases in the semen volume, spermatozoa motility, and total motile sperm count and increases in the pH values and multiple anomaly indices of patients infected with HIV. In the study by Dulioust et al. [2002], the most significant semen alterations were a decrease of the rapidly progressive motile spermatozoa (motility a) and an increase of less rapidly progressive spermatozoa (motility b) in the patients infected with HIV. In the study by Nicopoullos et al. [2004], ejaculate volume, sperm concentration, total count, progressive motility, and normal morphology were decreased. However, none of these studies related these semen alterations to HIV infection or ARV therapy. Another study explored the characteristics of semen before and after the start of ARV with different results and demonstrated a statistically significant reduction in the percentage of progressively motile spermatozoa in patients with HIV-1 infection during treatment with ARV therapy [van Leeuwen et al., 2008b]. They demonstrated previously that there was no detectable change in semen quality in patients during a period of untreated asymptomatic HIV-1 infection with similar proportion of progressively motile spermatozoa [van Leeuwen et al., 2008a]. These observations suggest that the reduction in progressively motile spermatozoa, which occurred in the current study during treatment, was related to the use of ARV and not to HIV infection. In these studies, it is not clear if some semen alterations could be the result of antiretroviral treatment or the result of HIV infection. Indeed, various studies have demonstrated a relationship between mitochondria and sperm motility [Donnelly et al., 2000; May-Panloup et al., 2003]. Since several ARVs have mitochondrial toxicity, the observed changes in motility could be due to the ARV itself. However, no association was found in the current study between quality of semen and the thymidine analog containing treatment.

However, it was found that nevirapine was associated with a higher percentage of progressively motile spermatozoa at T0 and at T1 and a better vitality in comparaison with efavirenz. The proportion of progressively motile spermatozoa under nevirapine or PI were normal according to WHO standards, whereas in the efavirenz group this percentage was decreased. Concerning vitality, nevirapine, PI, or efavirenz remained according to WHO standards although the efavirenz group presented vitality significantly inferior to nevirapine or the PIs groups. The percentage of progressively motile spermatozoa and vitality are related measures of semen quality. Nevirapine has a better penetration score than efavirenz and is a priori more efficient in this compartment, a better control of the virus may explain a better quality of semen. Another explanation could be a direct toxicity of efavirenz on cells producing spermatozoa. Indeed, there is no study on sperm characteristics before and after infection. Despite no relationship with the penetration score it would be interesting to improve this score and to try to validate it in a prospective study.

In conclusion, NRTIs and PIs were not associated with any semen changes. However, in patients receiving NNRTIs, nevirapine was associated with a better quality of semen versus efavirenz. This study confirms the fact that semen alterations are frequent in patients infected with HIV and suggests that these alterations could be related to ARV treatment.