Antiretroviral Therapy Effects on Genetic and ...

127

Antiretroviral Therapy Effects on Genetic and Morphologic End Points in Lymphocytes and Sperm of Men with Human Immunodeficiency Virus Infection Wendie A. Robbins,1,a Kristine L. Witt,2 Joseph K. Haseman,3 David B. Dunson,3 Luigi Troiani,5 Myron S. Cohen,5 Carol D. Hamilton,8 Sally D. Perreault,4 Bishara Libbus,7 Stan A. Beyler,6 Douglas J. Raburn,9 Shelia T. Tedder,8 Michael D. Shelby,1,a and Jack B. Bishop1

1

National Institute of Environmental Health Sciences, Laboratory of Toxicology, 2ILS, Information Sciences Division, 3National Institute of Environmental Health Sciences, Biostatistics Branch, and 4US Environmental Protection Agency, Reproductive Toxicology Division, Research Triangle Park; Departments of 5Medicine and of 6Obstetrics and Gynecology, University of North Carolina School of Medicine, and 7Genetics Research, Chapel Hill; Departments of 8Medicine and of 9Obstetrics and Gynecology, Duke University Medical Center, Durham, North Carolina

Nucleoside reverse-transcriptase inhibitors such as zidovudine (Zdv; a.k.a. azidothymidine) represent a key component of highly active antiretroviral therapy for human immunodeficiency virus (HIV) infection [1–3]. Use of these and other agents in combination has led to dramatic improvement in the health and survival of patients with HIV [4–6]. In addition, nucleoside antiretrovirals are administered prophylactically to healthy persons after occupational exposure to HIV [7, 8]. Nucleoside antiretrovirals are designed to integrate into replicating DNA and thus may have cytogenetic and reproductive effects that require study. Evaluation of their toxic effects is particularly important, because of the necessity for life-long therapy with these agents [4] and because many people with HIV infection are now pursuing options for conceiving healthy off-

Received 11 September 2000; revised 3 April 2000; electronically published 25 June 2001. a Present affiliations: UCLA Center for Occupational and Environmental Health, Los Angeles, California (W.A.B.); National Institute of Environmental Health Sciences, Center for the Evaluation of Risks to Human Reproduction, Research Triangle Park, North Carolina (M.D.S.). Reprints or correspondence: Dr. Jack B. Bishop, NIEHS, PO Box 12233, Research Triangle Park, NC 27709 ([email protected]). The Journal of Infectious Diseases 2001; 184:127–35 䉷 2001 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2001/18402-0002$02.00

spring [9, 10]. Early studies with high doses of Zdv demonstrated cytogenetic damage and adverse male reproductive effects in both animals and humans [11–19]. To investigate these issues further, we examined the effects of nucleoside reversetranscriptase inhibitors on lymphocyte and sperm chromosomal material, semen quality, and sperm function in men beginning therapy for HIV infection. Materials and Methods Study Cohort

Study subjects were recruited through the National Institute of Environmental Health Sciences Clinical Center contracts with InPresented in part: Environmental Mutagen Society annual meeting, Washington, DC, March 1999 (abstract 235); Environmental Mutagen Society annual meeting, New Orleans, March 2000 (abstract 166). Informed consent was obtained from all study participants, and research was performed within the ethical guidelines and with the written approval of the internal review boards of the National Institute of Environmental Health Sciences, University of North Carolina, and Duke University Medical Center. The US Environmental Protection Agency (EPA) National Health and Environmental Effects Research Laboratory reviewed the study and approved this publication. Approval does not signify that the contents reflect the views of the EPA, nor does mention of trade names or commercial productsconstitute endorsement or recommendation for use. Financial support: National Institutes of Health (contracts 273-97-C0058, 273-98-C-0072, N01-ES-35356, N01-ES-35357, and N01-ES-65401; grant NIDDK DM RO1); EPA.

Downloaded from http://jid.oxfordjournals.org/ by guest on February 2, 2016

Many human immunodeficiency virus (HIV)–infected persons receive prolonged treatment with DNA-reactive antiretroviral drugs. A prospective study was conducted of 26 HIV-infected men who provided samples before treatment and at multiple times after beginning treatment, to investigate effects of antiretrovirals on lymphocyte and sperm chromosomes and semen quality. Several antiretroviral regimens, all including a nucleoside component, were used. Lymphocyte metaphase analysis and sperm fluorescence in situ hybridization were used for cytogenetic studies. Semen analyses included conventional parameters (volume, concentration, viability, motility, and morphology). No significant effects on cytogenetic parameters, semen volume, or sperm concentration were detected. However, there were significant improvements in sperm motility for men with study entry CD4 cell counts 1200 cells/mm3, sperm morphology for men with entry CD4 cell counts ⭐200 cells/mm3, and the percentage of viable sperm in both groups. These findings suggest that nucleoside-containing antiretrovirals administered via recommended protocols do not induce chromosomal changes in lymphocytes or sperm but may produce improvements in semen quality.

128

Robbins et al.

Table 1. Selected demographic and clinical variables for 26 human immunodeficiency virus (HIV)–infected men receiving antiretroviral therapy (n p 26) Variable Demographics Race Black White Mixed black/white Age, years 20–29 30–39 40–49 a Smoke cigarettes Clinical characteristics b CD4 cell count, cells/mm3 At study entry At end of study c HIV load, cells/mL At study entry At end of study

Outcome

14 (53.8) 11 (42.3) 1 (3.8) 6 15 5 13

(23.1) (57.7) (19.2) (59.1)

280; 45 (10–815) 331; 37 (34–676) 10.3; 10.7; 0.34 (8.0–14.3) 6.0; 6.1; 0.27 (3.8–10.6)

NOTE. Data are no. (%) unless otherwise noted. a Smoking status was known for only 22 men. b Data are mean; SD (range). c Data are median; mean; SE log scale (range).

Table 2. Antiretroviral drugs administered to study subjects and specific treatment regimens represented in this study and no. of subjects receiving each regimem Specific drug combinations given Zdv, Zdv, Zdv, Zdv, Zdv, Zdv, d4T, d4T,

3TC 3TC, 3TC, 3TC, 3TC, 3TC, 3TC, 3TC,

indinavir nelfinavir abacavir, amprenavir delavirdine, indinavir lopinavir, ritonavir, lopinavir, ritonavir, indinavir

No. of subjects 6 6 2 3 1 a 2 a 5 2

NOTE. Drug regimens given in this study included nonnucleoside reverse-transcriptase inhibitors (delavirdine), nucleoside reverse-transcriptase inhibitors (zidovudine [Zdv], lamivudine [3TC], stavudine [d4T], and abacavir), and protease inhibitors (indinavir, nelfinavir, lopinavir, ritonavir, and amprenavir). a One patient switched from d4T, 3TC, lopinavir, and ritonavir to Zdv, 3TC, lopinavir, and ritonavir; both treatments are indicated, and total no. of patients equals 27, rather than 26.

Study Design

The longitudinal study design included repeated collection of biologic, clinical, and questionnaire data over 12 weeks. Sampling times coincided with routine clinic visits as much as possible. There were 4 sampling times: baseline before initiation of drug therapy and at 4, 6–8, and 12 weeks after initiation of drug therapy. These specific time points were chosen to target sensitive periods of spermatogenesis. The 6–8 week time period was chosen to assess effects during meiosis, and the 12-week (or greater) exposure period permitted assessment of stem cell effects. In addition, the 4-week point for peripheral blood and semen sampling was chosen to assess potential effects from the typical treatment duration of postexposure prophylactic treatment that would be administered for occupational exposures [4].

Questionnaires, Semen, and Blood Collection

Before antiretroviral drug treatment was initiated, study managers administered standardized questionnaires and reviewed medical records to obtain data that included age, race, occupation, therapeutic drug history (both for treatment of HIV infection and for unrelated conditions), CD4 cell count, and information on current or recent illnesses. In addition, before the start of treatment and at each subsequent visit, a self-administered questionnaire was used to collect data on lifestyle factors that might affect semen quality, such as recent caffeine consumption (e.g., sodas, coffee, and/or tea), cigarette or pipe use, alcohol consumption, and exposure to high temperatures via sauna or hot tub. For the lymphocyte cytogenetic assays, whole blood samples were drawn into an 8-mL sodium heparin green top Vacutainer tube during the clinic visit and were transported at room temperature via same-day courier service to LabCorp (Research Triangle Park, NC) for lymphocyte culturing and slide preparation. Semen samples were obtained by masturbation, either in the clinic or at the subject’s home, and were delivered to the andrology laboratory at either the University of North Carolina or Duke University for routine semen analysis [21]. An abstinence interval of at least 2


fectious Disease Clinics at Duke University Medical Center Hospital and University of North Carolina Hospital at Chapel Hill from July 1997 through December 1998. Study subjects were otherwise healthy patients diagnosed with HIV infection and were about to start antiretroviral therapy with a regimen that included at least 1 nucleoside analogue. Furthermore, subjects must not have received nucleoside analogue therapy in the past year, must not have undergone a vasectomy, and may not have had any history of sperm or reproductive problems. They needed to be able to read and understand English and be willing to donate blood and semen samples repeatedly over the 12-week research period. Sixty men met these criteria and were asked to participate. Of these, 26 (43%) agreed, and 34 (57%) refused. The major reasons given for refusal to participate were embarrassment regarding semen donation, transportation difficulties, and/or scheduling problems. The 26 men who agreed to participate did not differ significantly from nonparticipants for age, race, or range of virus load and CD4 cell counts reported by the recruiting clinics during the recruitment period. Selected demographic and clinical characteristics of the study subjects are shown in table 1. None of the men had known occupational exposures to potential mutagens or chemicals that might alter their sperm. Various antiretroviral treatment therapies were administered (table 2). However, all study subjects received Zdv or lamivudine (3TC), or both, as a component of their therapy. Selfreported adherence to prescribed drug therapy averaged 95.3% (SD, 2.1%). Drug levels were measured in a subset of subjects [20] providing further evidence of adherence to prescribed treatment at time of semen collection. Drug toxicities were closely monitored, and no research subjects discontinued nucleoside analogue therapy because of severe drug toxicity during the study. One subject substituted Zdv for stavudine (d4T) several weeks into treatment because of side effects.

JID 2001;184 (15 July)

JID 2001;184 (15 July)

HIV Antiretroviral Therapy Effects

days was recommended before collection of the study sample [21], although not all men adhered to this recommendation. Because of the requested minimum 2-day abstinence period between semen sample collections, the need for subjects not to delay the start of their medications, as well as the need to avoid requesting multiple additional clinic visits that might affect participation rates, we constrained our sampling to a single semen specimen per time point of interest. Delivery of specimens to the andrology laboratory for motility analysis within 2 h of collection was specified but was not possible in all cases.

129

morphology assessment by using World Health Organization criteria [21] and strict criteria [26]. After scoring 200 sperm cells/ sample, the percentage of morphologically normal sperm cells in each sample was calculated for each method. The remaining semen was aliquoted into cryovials and was frozen without cryopreservative at ⫺70⬚C. Samples were later shipped by overnight delivery on dry ice to University of California, Los Angeles, for sperm cytogenetic analyses.

Statistical Analyses Laboratory Analyses

Lymphocyte cytogenetics. Analysis of variance procedures were used to assess the significance of differences in chromosomal aberration frequencies among sample time groups, differences among recruitment sites, and subject-to-subject variability. Pairwise comparisons were made by the Dunnett test. We applied the variance stabilizing arc sine transformation to these percentages before statistical analysis. In addition, regression analyses were carried out to assess possible linear or quadratic trends in response as a function of time. Within individual subjects, time-related changes in lymphocyte chromosomal aberration frequencies were assessed by Cochran-Armitage trend tests. Significance was set at P ! .05. With this study design and scoring 600 cells/sample, there was 90% power to detect a 3-fold change in aberration frequency per subject over time. Sperm aneuploidy and sperm chromosome breakage. Analysis of variance procedures were used to assess the significance of differences in sample time groups, differences among recruitment sites, and subject-to-subject variability. The variance-stabilizing arc sine transformation was used in these analyses. Pairwise comparisons among the different sample times were made by the Dunnett test. With this study design and significance set at P ! .05, power calculations indicated at least a 75% probability of detecting a doubling for any one of the pretreatment values. For most of the variables, including total aneuploidy and total breakage, the power to detect a doubling was 195%. Conventional semen analyses. As with the lymphocyte and sperm data, semen end points were analyzed for treatment-related changes. In addition, the semen data were further analyzed to determine whether any observed changes were correlated with disease stage of the study subjects at the time of enrollment. Disease stage was classified by CD4 cell counts as follows: early disease (1200 cells/mm3) or later disease (⭐200 cells/mm3). The statistical package used for the conventional semen analysis was S-PLUS version 5.1 (Mathsoft). To simplify the analyses and to reduce the number of degrees of freedom of the tests for differences in the semen end points, all during-treatment groups were combined into 1 group. This combination was motivated by the relatively few men under study, by differences among subjects in the timing of the different visits after treatment initiation, and by the presence of missing observations. Since the during-treatment groups were combined before any analyses, there is no inflation of the type I error rate. Because of the longitudinal study design, we expect good power to detect moderate-sized differences in end points, such as morphology, which do not exhibit much variability within a subject. However, our


Lymphocyte cytogenetic assays. Lymphocyte culturing and slide preparation were done by means of published cytogenetic techniques [22, 23]. A 48-h culturing period was used. In most cases, 600 wellspread metaphases from each blood sample were scored for frequency of chromosomal aberrations. Categories of aberrations scored included chromatid and chromosome breaks, double minutes, triradials, quadriradials, dicentrics, rings, complex rearrangements, cells with 110 aberrations, pulverized cells, and pulverized chromosomes. Gaps were scored and recorded but were not included in the analysis of cells with chromosomal aberrance. The mitotic index was determined from at least 1000 cells. Genetic Research (Chapel Hill, NC) scored the lymphocyte spreads. Sperm aneuploidy and chromosome breakage assays. Sperm fluorescence in situ hybridization (FISH) analyses for aneuploidy were performed on thawed aliquots of semen, according to the method described by Robbins et al. [24]. Direct-labeled, high-intensity fluorophore a satellite probes for chromosomes X (labeled with SpectrumGreen) and 18 (labeled with SpectrumAqua), as well as satellite III DNA for the Y chromosome (labeled with Spectrum Orange), were used. Probes were purchased from Vysis. About 10,000 sperm cells/sample (range, 9951–10,056 sperm cells/sample) were scored for aneuploidy. Sperm cytogenetic analyses for chromosome breakage followed the chromatin preparation protocol, as described by Robbins et al. [24]. Hybridization was adapted for sperm from the method for buccal cells described by Rupa and Eastmond [25]. The methods differed as follows: probe mixture contained 7.7 mL of hybrizol hybridization buffer (Oncore), 1 mL of direct fluorescein isothiocyanate–labeled a satellite probe targeting a small centromeric region of chromosome 1 (Vysis), and 1.3 mL of direct Cy3-labeled probe targeting adjacent classical satellite II region within the chromosome 1 heterochromatin (provided by D. A. Eastmond, University of California, Riverside). Posthybridization washes were in 50% formamide–2⫻ standard saline citrate (SSC) at 45⬚C for 10 min, which was followed by 1⫻ SSC at 45⬚C for 10 min. Slides were counterstained with 0.025 mg of 4,6-diamidino-2-phenylindole (DAPI) in VectaShield mounting medium. A total of 3000 sperm cells/sample was scored for chromosome 1 breakage. Conventional semen analysis. Semen samples were evaluated for semen volume, sperm concentration (by means of hemocytometer), percentage of motile sperm, and percentage of viable sperm (eosin B dye exclusion). Air-dried semen smears were fixed in 95% ethanol and were mailed to Fertility Solutions (Cleveland), where they were stained with Papanicolau stain and were subjected to

130

Robbins et al.

Table 3. Mean lymphocyte chromosomal aberration frequencies over time for 26 human immunodeficiency virus–infected men receiving antiretroviral therapies. Sample time Baseline Midstudy (22–33 days) End of study (34–175 days) 1-Year follow-up

No. of samples a evaluated 25 14 32 16

Percentage of aberrant cells, mean Ⳳ SD 1.62 1.73 1.76 1.43

Ⳳ Ⳳ Ⳳ Ⳳ

0.93 0.76 0.93 0.74

a

One baseline sample failed to yield data; not all subjects provided samples for each time period, and some were sampled more than once during the end of study period.

viding a pretreatment blood sample, chromosomal aberration frequencies could be monitored throughout the course of the study for each subject. No biologically relevant changes in lymphocyte chromosomal aberration frequencies were observed for any of the 26 subjects. None of the values obtained during treatment exceeded the upper range established by the pretreatment measurements, and all values were within the normal range (mean, 1.98; range, 0–6) previously reported for healthy people [28, 29]. Suggestions of treatment-related changes in aberrations were noted for 6 study subjects during the first 12 weeks of treatment (4 increases and 2 decreases) but these were generally not confirmed in long-term follow-up measurements taken after at least 1 year of antiretroviral drug therapy (table 4). For 2 of the 4 subjects showing increases in aberration frequencies during the course of the main study, follow-up samples did not confirm the trend, and long-term data were unavailable for the other 2 subjects. One of the 2 decreasing trends was confirmed at 1 year. Sperm aneuploidy and chromosome breakage end points. Table 4. Lymphocyte chromosomal aberration frequencies for individual subjects who demonstrated a statistically significant (P ! .05) change over the initial 12-week monitoring period. Percentage of aberrant cells

No. of aberrant cells/total

Baseline 1 2 1-Year follow-up

2.00 3.00 0.50 a 0.67

12/600 17/600 3/600 4/600


1.67 1.33 0.16 0.83

10/600 8/600 1/600 5/600

Baseline 1 2

0.83 2.17 2.50

5/600 13/600 15/600

Baseline b 1

0.30 2.00

2/600 12/600


0.67 0.50 1.83 0.67

4/600 3/600 11/600 4/600


1.01 4.00 3.83 1.67

6/594 10/250 3/78 10/600

Subject, sample

Results Lymphocyte cytogenetic end points. A total of 26 men donated blood for the lymphocyte cytogenetic evaluation before treatment and ∼4 and 12 weeks after the initiation of treatment. Most study subjects could schedule blood draws at the intervals designated in the study protocol; however, allowances were made for subjects to continue their participation in the study, despite occasional scheduling difficulties. Analysis of pooled data for all 26 men showed no statistically significant association between chromosomal aberration frequencies and length of treatment (P p .174; table 3). None of the mean values of the treatment groups differed significantly from the pretreatment value, and there were no significant trends when time was used as an independent variable. In addition to treatment time, aberration frequencies also were analyzed with respect to source clinic (Duke University vs. University of North Carolina at Chapel Hill) and treatment regimen (presence or absence of Zdv and effects associated with any of the 10 individual drugs in the 8 treatment regimens used for this group of 26 men). No associations among any of these factors and chromosomal aberration frequencies were identified. Chromosomal aberration frequencies also were measured in 16 of the original 26 study subjects after at least 1 year of continuous antiretroviral therapy. The mean frequency at 1 year was slightly lower than the mean pretreatment value, but the difference was not statistically significant (table 3). Because study subjects served as their own controls by pro-

A

B

C

D

E

F

NOTE. For each subject, data are pretreatment values followed by 2 sequential during-treatment values (about 4 and 12 weeks after initiation of therapy). Additional samples obtained for 4 subjects were after at least 1 year of continuous therapy; the data did not confirm the shorter-term trends, except for subject A. Subjects A and B had significant decreases, whereas subjects C–F had significant increases. Significance was determined by Cochran Armitage trend test. a Trend also significant (P ! .05) with 1-year follow-up data. b No further samples were obtained from this subject.


power to detect treatment effects on volume and concentration, which are highly variable, is low. Analysis of variance methods, after log transformation, were used to assess changes in semen volume and concentration. Logistic regression analyses were used to assess changes in the proportion of motile, viable, and morphologically normal sperm. Variability among subjects was accounted for by including the subject as a blocking factor in each model. Since human sperm motility and viability have been reported to decline with time after ejaculation and would be expected to decrease by as much as 25% in 5 h [27], the statistical analysis for motility and viability was controlled for sample aging. We included indicators of !2, 2–4, and 1 4 h in the relevant models. Analyses of volume and concentration were adjusted for abstinence intervals of !2 versus ⭓2 days.

JID 2001;184 (15 July)

JID 2001;184 (15 July)


Table 5. Sperm–fluorescence in situ hybridization cytogenetic data (ⳲSD) for chromosomes X, Y, and 18 aneuploidy and XY18-18 diploidy based on ∼10,000 sperm cells scored and for chromosome 1 breakage and diploidy/aneuploidy based on ∼3000 sperm cells/sample scored. Sperm category Aneuploidy XX18 YY18 XY18 18-18 Diploidy XY18-18 Chromosome 1 Breakage b Diploidy/aneuploidy

Pretreatment sample

Midstudy sample

End of study sample

n p5 1.3 Ⳳ 1.3 3.1 Ⳳ 2.2 23.1 Ⳳ 13.9 3.3 Ⳳ 2.2

n p5 1.0 Ⳳ 1.0 2.4 Ⳳ 1.5 24.4 Ⳳ 16.1 1.8 Ⳳ 0.8

n p 26 1.8 Ⳳ 2.0 4.5 Ⳳ 3.5 23.8 Ⳳ 14.2 2.8 Ⳳ 2.7

Ⳳ p Ⳳ Ⳳ

4.9 Ⳳ 2.5 a n p 21 3.4 Ⳳ 1.3 3.5 Ⳳ 2.1

4.7 Ⳳ 2.7 n p 13 2.8 Ⳳ 1.8 3.1 Ⳳ 1.2

8.0 n 2.7 3.0

2.2 3 1.5 1.0

a

a

Because of clinic scheduling, many men were late donating samples at midstudy. To decrease misclassification for analysis of meiotic effects without DNA replication effects, only samples within a tight window of 34–56 days were included in the midstudy category. Late samples were analyzed with the end of study grouping, since they would include premeiotic effects. b Chromosome 1 was hybridized alone; therefore, this category represents an abnormal event representing either a diploid or an aneuploid sperm.

and with amorphous shape decreased with treatment, whereas the percentage with tail abnormalities increased with treatment.

Discussion Although advances in antiretroviral therapy have brought many benefits, the long-term effects of these therapies are not known and could be of considerable importance. For example, protease inhibitors are now recognized to have a variety of important long-term metabolic effects [30–33]. The majority of patients with HIV receive at least 1 nucleoside reverse-transcriptase inhibitor as a component of drug therapy [4]. Because nucleoside reverse-transcriptase inhibitors are designed specifically to incorporate into DNA [3], it is reasonable to ask whether these drugs can affect chromosomal and genetic integrity. Early studies of humans treated with high doses of the nucleoside reverse-transcriptase inhibitor Zdv demonstrated cytogenetic damage. A 16-fold increase in chromosomal aberrations in blood lymphocytes from 7 patients with AIDS treated with Zdv (1200 mg/day for 1–7 months), compared with that in blood lymphocytes from 4 patients with AIDS without exposure to Zdv or other nucleoside analogues, was reported by Shafik et al. [12]. Increased micronucleus frequencies in human B and T cells, telomere shortening, and large gene deletion mutations were found in in vitro studies by using cell lines treated with Zdv [34–37]. Multidosing animal studies also have found cytogenetic effects in the form of micronucleus induction and telomere shortening [11, 13–16, 38]. In our study, adverse effects of nucleoside reverse-transcriptase inhibitors on lymphocyte chromosome aberrations could not be demonstrated by using aggregate data across all men. On an individual basis, 6 men showed statistically significant changes (4 increases and 2 decreases) in aberration frequency related to


Fifteen men had adequate semen sample remaining for the sperm aneuploidy assay, and 13 had adequate sample for the additional sperm chromosome breakage assay after conventional semen analyses were complete. Sperm-FISH was used to detect and measure aneuploidy involving chromosomes X, Y, 1, 18, and diploidy XY18-18 and breakage in chromosome 1. Hybridization efficiency for both the aneuploidy and breakage assays exceeded 99.98%. No statistically significant treatmentrelated effects were found in the sperm cytogenetic assays across the sampling times (table 5), except XY18-18 at midstudy (P ! .05). In addition, the ratio of X:Y-bearing sperm did not differ from the expected 50:50 ratio at pretreatment or at any follow-up treatment visit. Semen analysis end points. Twenty men donated semen samples at the initial (drug naive) visit and at least 1 follow-up visit. The semen analyses were based on these 20 study subjects, 9 with a CD4 cell count !200 cells/mm3 on entry into the study (categorized as late disease) and 11 with a CD4 cell count 1200 cells/ mm3 at study entry (categorized as early disease). The mean values for the percentages of sperm that were motile, viable, and morphologically normal and for the semen volume and concentration are presented in table 6. When more than 1 during-treatment sample was available for a given man, the values for these samples were pooled before calculating the means across men. On the basis of linear regression analyses that accounted for abstinence interval (!2 days) and intersubject variability, there were no significant treatment-related differences in semen volume or concentration (P p .79 and P p .77, respectively). On the basis of logistic regression analyses that were stratified by patients and that accounted for variable time from semen donation until semen analysis (!2, 2–4, and 14 h), there were highly significant (P ! .01) differences between pre- and during-treatment semen samples in 3 areas: the percentage of motile sperm from men with early disease; the percentage of viable sperm, regardless of CD4 cell count; and the percentage of sperm with normal morphology from men with late disease (figure 1). These differences all reflect improvements in semen quality with treatment. There were no significant treatment-related differences in either the percentage of motile sperm from men with late disease or the percentage of sperm with normal morphology from men with early disease. In general, treatment-related differences in sperm viability and morphology were greater for samples from men with late disease, whereas differences in sperm motility were greater for men with early disease. However, even without stratification based on CD4 cell count, there were significant treatment-related improvements in sperm motility, viability, and morphology (P ! .01, P ! .01, and P p .015, respectively). We also tested separately for treatment-related differences within morphologic categories (table 6) by using logistic regression with patient as a blocking factor. There were significant treatment-related differences in the percentage of sperm with tapered heads (P p .04), amorphous shape (P p .01), and tail abnormality (P ! .01). The percentages of sperm with tapered heads

131

132

Robbins et al.

JID 2001;184 (15 July)

Table 6. Mean semen end points (SE) before and during treatment for men categorized by CD4 cell count as early disease (1200 CD4 cells/mm3) vs. late disease (⭐200 CD4 cells/mm3). CD4 cell count ⭐200 cells/mm3 (n p 9) Before treatment

Semen end point a

10.6 44.6 22.1 3.2 3.5 70.0 25.6 16.9 1.5 52.3

(1.2) (1.4) (1.4) (0.4) (0.6) (1.9) (1.1) (0.9) (0.1) (4.5)

11.5 66.5 29.4 1.6 3.4 64.6 24.3 20.8 2.3 128.5

(0.7) b (1.1) b (1.0) (0.1) (0.5) (0.6) (1.4) (1.0) (0.2) (12.0)

(n p 11) Before treatment

During treatment

20.9 72.6 32.8 1.0 3.1 71.5 20.8 17.7 2.0 80.4

34.1 79.0 33.8 0.7 3.0 67.2 23.9 22.5 1.8 61.3

(1.0) (1.8) (0.7) (0.1) (0.1) (0.8) (0.5) (0.5) (0.1) (7.4)

b

(2.4) b (1.0) (0.9) (0.1) (0.2) (0.8) (0.8) (0.7) (0.1) (6.2)

a

Values for percentage of motile and percentage of viable are adjusted for sample aging. Statistically significant difference from pretreatment value. c World Health Organization criteria [26] were used for morphology assessment. Sperm with 11 defect are counted in each appropriate abnormal category; thus, categories do not sum to 100%. b

treatment. However, no values exceeded the highest value defined by the pretreatment frequencies (range of aberrant cells in pretreatment samples, 0.3%–4.5%), no values fell outside the range reported previously for healthy people [28, 29], and the direction of observed changes was inconsistent among the 6 subjects. This strongly suggests that the individual findings are not biologically significant or treatment related. Indeed, follow-up blood samples provided by some of these subjects at 11 year showed a return toward pretreatment aberration frequencies, and tests for trends recalculated to include the long-term follow-up values were generally nonsignificant (table 4). Virus load values and CD4 cell levels indicated good response to antiretroviral therapy for each of these 6 patients, and mitotic indices measured in the stimulated lymphocyte cultures were considerably improved at the time of long term follow-up. Thus, blood parameters are consistent with good response to treatment. Regarding germ cell cytogenetic damage induced by exposure to nucleoside reverse-transcriptase inhibitors, no human sperm aneuploidy or chromosome breakage studies have been published. However, the US National Toxicology Program (NTP) conducted tests for induction of chromosome damage by both Zdv and d4T in male mouse germ cells by using the dominant lethal test. After acute treatment by intraperitoneal injection, neither Zdv nor d4T induced such damage (authors’ unpublished data). In the current study, we did not detect significant treatmentassociated cytogenetic changes in sperm aneuploidy for chromosomes X, Y, 1, and 18 or breakage for chromosome 1. A finding of diploid sperm at midstudy (indicated by XY18-18; P ! .05) is biologically unexpected, since meiotic errors (e.g., XY18 aneuploidy) would be more likely. The finding of a statistically significant result is consistent with chance alone because

of the multiple sperm cytogenetic outcomes evaluated. The finding of diploid sperm midstudy resolved by the final collection time point. In previous studies, sperm aneuploidy was useful for detecting cytogenetic damage associated with chemotherapeutic, occupational, environmental, and lifestyle exposures [39–42]. Most of the previous studies used cross-sectional designs. However, longitudinal sampling would be expected to provide more power to detect induced effects. We sampled longitudinally at time points expected to capture stem cell and meiotic effects and scored 10,000 sperm cells/sample. Aggregate frequencies for sperm aneuploidy for the men in the current study fell within the ranges reported for unexposed cohorts in other published research studies for all abnormalities, except XY18 [24, 39, 40]. On the basis of ∼10,000 sperm cells scored per man, the mean value for XY18 aneuploidy was 23.9 (SD, 13.9; range, 7–61). This is twice the mean value reported for healthy men [24, 39, 41]. Nuovo et al. [43] found that viral DNA, in the testes of men infected with HIV had localized primarily to spermatogonia and spermatocytes and, to a lesser extent, was associated only with postmeiotic spermatids. Because XY18 aneuploidy represents a meiosis I error, the high values detected in our study are of interest and deserve further investigation. In the current study, we adapted the tandem-labeling technique to detect chromosome breakage in sperm cells of the study subjects. Tandem labeling has proven useful in detecting chromosome breakage in irradiated human lymphocytes in vitro and in buccal mucosal cells in vivo [25, 44, 45]. No baseline chromosome breakage frequencies for sperm by using tandem labeling have been published; however, our results for sperm (∼1 breakage event per chromosome 1 per 1000 sperm cells scored) fall within the range reported for buccal cells and lym-


Percentage motile a Percentage viable c Percentage normal morphology Percentage tapered Percentage abnormal size Percentage amorphous Percentage abnormal midpiece Percentage abnormal tail Semen volume, mL Concentration, cells ⫻106/mL

During treatment

1200 cells/mm3

JID 2001;184 (15 July)


133

phocytes [24, 44]. No differences in breakage frequencies were detected among sampling times. Thus, we did not detect an effect of nucleoside reverse-transcriptase inhibitors on sperm aneuploidy or breakage over the 4-month study period. The majority of recent semen studies done with men on antiretroviral therapy have looked at effects of treatment on semen infectivity for HIV [20, 46–49]. However, several investigators have examined the effects of antiretroviral therapy on semen quality [20, 50–53]. Krieger et al. [50] found no significant effects of Zdv on semen parameters in HIV-infected men on antiretroviral therapy for an average of 4 months (range, 1–11 months). Crittenden et al. [51] found no significant effects in men sampled before and as much as 12 months after starting Zdv therapy (doses of 600–1200 mg/day). In a study conducted by Politch et al. [52], HIV-1 infected men with a CD4 cell count 1200 cells/mm3 had semen parameters consistent with those of fertility, whereas men with a CD4 count !200 cells/mm3 initially had lower counts that subsequently improved with Zdv treatment (100–200 mg every 4 h). Muller et al. [53] examined differences in semen quality between 250 HIV-seropositive men and 38 healthy fertile control subjects in a cross-sectional design. A subgroup of men receiving antiretroviral drugs was examined for treatment-related semen effects; only small changes in volume and pH were found. By use of computeraided motility and morphology assessment, they observed cor-

relations between CD4 cell count and semen quality, noting that men with fewer CD4 cells had significantly decreased motility and poorer strict criteria morphology; sperm counts were not affected in these men. The NTP evaluated effects of Zdv on male reproductive parameters in mice and rats [18]. At the highest dose of 1000 mg/ kg, modest decreases were observed in mouse testis and rat epididymal weights; epididymal sperm density was reduced in both rats and mice. None of these changes were considered to be significant. Additional NTP tests evaluated the potential reproductive effects of Zdv in CD1 mice by using the Reproductive Assessment by Continuous Breeding protocol, as described by Chapin and Sloane [17]. After twice daily oral administration of 50–200 mg of Zdv/kg of body weight for up to 90 days, dose-related decreases for all treatment groups in the F0 generation were observed for several end points, including the number of pups/litter; sperm motility, velocity, linearity, and density; and spermatid head counts. At the highest dose in the F1 generation, slight decreases were noted in the number of pups per litter (authors’ unpublished data). In the current study, we found that antiretroviral therapy led to improved sperm motility, viability, and morphology. The improvement in sperm motility was most pronounced among subjects with early disease (CD4 cell count at study entry, 1200 cells/mm3). Improvement in viability and morphology was most


Figure 1. Treatment-related changes in semen end points. Analyses were based on logistic regression that was stratified by patients and that accounted for variable time from semen donation until semen analysis (!2, 2–4, and 1 4 h). Disease stage is classified by CD4 cell count as early disease (1200 cells/mm3) or late disease (⭐200 cells/mm3). *Highly significant (P ! .01 ) differences between pre- and during-treatment semen samples. NS, nonsignificant values.

134

Robbins et al.

3. 4.

5. 6.

7.

8.

9. 10.

11.

12.

13. 14.

15.

16.

17.

18.

19.

Acknowledgments We acknowledge Patrick Vojta and Perry Blackshear for valuable suggestions on the manuscript and recognize the exceptional quality and efficiency of the work in database management and support of Richard Morris, David Johndrow, and Erika Schuett of Analytical Sciences.

20.

21.

22. References 23. 1. Blower S, Gershengorn H, Grant R. A tale of two futures: HIV and antiretroviral therapy in San Francisco. Science 2000; 287:650–4. 2. Sackoff J, McFarland J, Shin S. Trends in prescriptions for highly active

24.

antiretroviral therapy in four New York City HIV clinics. J Acquir Immune Defic Syndr 2000; 23:178–83. Brenner G. Pharmacology. Philadelphia: WB Saunders, 2000. Carpenter C, Cooper D, Fischl M, Gatell J, Gazzard B, Hammer S. Antiretroviral therapy in adults: updated recommendations of the International AIDS Society–USA Panel. JAMA 2000; 283:381–90. Moore R, Chaisson R. Natural history of HIV infection in the era of combination antiretroviral therapy. AIDS 1999; 13:1933–42. Vittinghoff E, Scheer S, O’Malley P, Colfax G, Holmberg S, Buchbinder S. Combination antiretroviral therapy and recent declines in AIDS incidence and mortality. J Infect Dis 1999; 179:717–20. Centers for Disease Control and Prevention. Public Health Service guidelines for the management of health-care worker exposures to HIV and recommendations for postexposure prophylaxis. MMWR Morb Mortal Wkly Rep 1998; 47(RR-7):1–33. Pallela F, Delaney K, Moorman A, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338:853–60. Kahn J. Risky business? Helping the HIV-infected have babies. J Androl 2000; 21:8–9. Semprini A, Levi-Setti P, Bozzo M, et al. Insemination of HIV-negative women with processed semen of HIV-positive partners. Lancet 1992; 340: 1317–9. Phillips M, Nascimbeni B, Tice R, Shelby M. Induction of micronuclei in mouse bone marrow cells: an evaluation of nucleoside analogues used in the treatment of AIDS. Environ Mol Mutagen 1991; 18:168–83. Shafik H, Nokta M, Pollard R. Recombinant human interferon beta ser protects against zidovudine-induced genetic damage in AIDS patients. Antiviral Res 1991; 16:205–12. Gonzalez, Cid M, Larripa I. Genotoxic activity of azidothymidine (AZT) in in vitro systems. Mutat Res 1994; 321:113–8. Olivero O, Anderson L, Diwan B, et al. Transplacental effects of 3-azido-2, 3-dideoxythymidine (AZT): tumorigenicity in mice and genotoxicity in mice and monkeys. J Natl Cancer Inst 1997; 89:1602–8. Olivero O, Shearer G, Chougnet C, et al. Incorporation of zidovudine into leukocyte DNA from HIV-1–positive adults and pregnant women, and cord blood from infants exposed in utero. AIDS 1999; 13:919–25. Ayers KM, Clive D, Tucker WE Jr, Hajian G, de Miranda P. Nonclinical toxicology studies with zidovudine: genetic toxicity tests and carcinogenicity bioassays in mice and rats. Fundam Appl Toxicol 1996; 32:148–58. Chapin R, Sloane R. Reproductive assessment by continuous breeding: evolving study design and summaries of ninety studies. Environ Health Perspect 1997; 105(Suppl 1):199–395. National Toxicology Program. Toxicology and carcinogenesis studies of AZT (CAS no. 30516-87-1) and AZT/alpha-interferon A/D in B6C3F1 mice (gavage studies). Appendix L. Technical report series no. 469. National Institutes of Health no. 96-3959. Research Triangle Park, NC: US Departments of Health and Human Services, Public Health Service, National Institutes of Health, 1999. Sikka S, Gogu S, Agarwal K. Effect of zidovudine (AZT) on reproductive and hematopoietic systems in the male rat. Biochem Pharmacol 1991;42:1293–7. Pereira A, Kashuba A, Fiscus S, et al. Nucleoside analogues achieve high concentrations in seminal plasma: relationship between drug concentration and virus burden. J Infect Dis 1999; 180:2039–43. World Health Organization. WHO laboratory manual for the examination of human semen and sperm-cervical mucus interaction. 3d ed. Cambridge, UK: Cambridge University Press, 1992. Hungerford DA. Leukocytes cultured from small inocula of whole blood and the preparation of metaphase chromosomes by treatment with hypotonic KCl. Stain Technol 1965; 40:333–8. Morehead P, Nowell P, Mellman W, Battips D, Hungerford D. Chromosome preparations of leukocytes cultured from human peripheral blood. Exp Cell Res 1960; 20:613–6. Robbins W, Vine M, Young Truong K, Everson R. Use of fluorescence in


pronounced among subjects with later disease (CD4 cell count at study entry, !200 cells/mm3). Nucleoside antiretroviral drugs appear to concentrate in semen and may gain access to sperm cells during their formation [47]. These results suggest that antiretroviral therapy could lead to improved fertility in men with HIV infection. There are limitations to the current study. First, the observational nature of the study did not permit control over the nucleoside antiretroviral therapies used. Thus, a wide variety of drug regimens were studied in a relatively small group of men. An uninfected control group was not enrolled because of concerns about previous animal and human study data showing adverse effects of nucleoside analogues on cytogenetic end points. The inclusion of an untreated HIV-infected group was initially considered. However, no study subjects opted to defer drug therapy; therefore, an untreated HIV-infected group was not enrolled. Adherence to drug therapy might have been overestimated, as this information was self-reported; actual drug levels were measured only in a subset of subjects at the time of semen donation. Finally, the time period for follow-up was chosen to maximize detection of cytogenetic effects in blood and semen, including stem cell effects, and to span a complete cycle of spermatogenesis. However, chronic long-term effects could be missed by this sampling scheme. In conclusion, this longitudinal study provided us with the opportunity to examine sperm and lymphocyte cytogenetic end points plus semen quality parameters in men receiving antiretroviral therapy. We failed to detect any significant chromosomal aberrations either in lymphocytes or sperm cells that would suggest increased risk for neoplasms or cytogenetic problems in offspring of patients with HIV receiving antiretroviral therapy. Further longitudinal studies and clinical studies that include men, women, and their children will be required to confirm these results. The use of antiretroviral therapies has led to dramatic improvement in health and survival of patients with HIV. Along with increased life expectancy, people with HIV infection are pursuing options for conceiving healthy offspring. For our study cohort, improvements in semen quality were noted for sperm motility, viability, and morphology. Given the current need for a lifelong therapy with nucleoside analogues, these results can be viewed as reassuring.

JID 2001;184 (15 July)

JID 2001;184 (15 July)

25.

26.

27.

28.

29.

31.

32.

33.

34.

35. 36.

37.

situ hybridization (FISH) to assess effects of smoking, caffeine, and alcohol on aneuploidy load in sperm of healthy men. Environ Mol Mutagen 1997; 30:175–83. Rupa D, Eastmond D. Chromosomal alterations affecting the 1cen-1q12 region in buccal mucosal cells of betel quid chewers detected using multicolor fluorescence in situ hybridization. Carcinogenesis 1997; 18:2347–51. Menkveld R, Stander F, Kotze T, Kruger T, van Zyl J. The evaluation of morphological characteristics of human spermatozoa according to stricter criteria. Hum Reprod 1990; 15:586–92. Makler A, Zaidise I, Paldi E, Brandes J. Factors affecting sperm motility. I. In vitro change in motility with time after ejaculation. Fertil Steril 1979;31: 147–54. Bender M, Preston R, Leonard R, Pyatt B, Gooch P. Chromosomal aberration and sister-chromatid exchange frequencies in peripheral blood lymphocytes of a large human population sample. II. Extension of age range. Mutat Res 1989; 212:149–54. Bender MA, Preston RJ, Leonard RC, Pyatt BE, Gooch PC, Shelby MD. Chromosomal aberration and sister-chromatid exchange frequencies in peripheral blood lymphocytes of a large human population sample. Mutat Res 1988; 204:421–33. Mulligan K, Grunfeld C, Tai V, et al. Hyperlipidemia and insulin resistance are induced by protease inhibitors independent of changes in body composition in patients with HIV infection. J Acquir Immune Defic Syndr 2000; 23:35–43. Hadigan C, Corcoran C, Stanley T, Piecuch S, Klibanski A, Grinspoon S. Fasting hyperinsulinemia in human immunodeficiency virus–infected men: relationship to body composition, gonadal function, and protease inhibitor use. J Clin Endocrinol Metab 2000; 85:35–41. Martinez E, Conget I, Lozano L, Casamitjana R, Gatell J. Reversion of metabolic abnormalities after switching from HIV-1 protease inhibitors to nevirapine. AIDS 1999; 13:805–10. Carr A, Samaras K, Burton S, et al. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998; 12:F51–8. Agarwal RP, Olivero OA. Genotoxicity and mitochondrial damage in human lymphocytic cells chronically exposed to 3-azido-2,3-dideoxythymidine. Mutat Res 1997; 390:223–31. Stern M, Cid M, Larripa I, Slavutsky I. AZT-induction of micronuclei in human lymphocyte subpopulations. Toxicol Lett 1994; 70:235–42. Gomez D, Tejera A, Olivero O. Irreversible telomere shortening by 3-azido2, 3-dideoxythymidine (AZT) treatment. Biochem Biophys Res Commun 1998; 246:107–10. Sussman H, Olivero O, Meng Q, et al. Genotoxicity of 3-azido-3-deoxythymidine in the human lymphoblastoid cell line, TK6: relationships between DNA incorporation, mutant frequency, and spectrum of deletion mutations in HPRT. Mutat Res 1999; 429:249–59.

135

38. Dertinger S, Torous D, Tometsko K. Induction of micronuclei by low doses of azidothymidine (AZT). Mutat Res 1996; 368:301–7. 39. Padungtod C, Hassold T, Millie E, et al. Sperm aneuploidy among Chinese pesticide factory workers: scoring by the FISH method. Am J Ind Med 1999; 36:230–8. 40. Robbins W, Rubes J, Selevan S, Perreault S. Air pollution and sperm aneuploidy in healthy young men. Environ Epi Toxicol 1999; 1:125–31. 41. Rubes J, Lowe X, Moore D II, et al. Smoking cigarettes is associated with increased sperm disomy in teenage men. Fertil Steril 1998; 70:715–23. 42. Robbins W, Meistrich M, Moore D, et al. Chemotherapy induces transient sex chromosomal and autosomal aneuploidy in human sperm. Nat Genet 1997; 16:74–8. 43. Nuovo G, Becker J, Simsir A, Margiotta M, Khalife G, Shevchuk M. HIV1 nucleic acids localize to the spermatogonia and their progeny: a study by polymerase chain reaction in situ hybridization. Am J Pathol 1994; 144:1142–8. 44. Rupa D, Hasegawa L, Eastmond D. Detection of chromosomal alterations offering the 1cen-1q12 region in irradiated granulocytes and lymphocytes by multicolour FISH with tandem DNA probes. Mutagenesis 1997; 12: 195–200. 45. Rupa D, Hasegawa L, Eastmond D. Detection of chromosomal breakage in the 1 can-1q12 region of interphase human lymphocytes using multicolor fluorescence in situ hybridization with tandem DNA probes. Cancer Res 1995; 55:640–5. 46. Vernazza P, Troiani L, Flepp M, et al. Potent antiretroviral treatment of HIVinfection results in suppression of the seminal shedding of HIV. AIDS 2000; 14:117–21. 47. Kashuba A, Dyer J, Kramer L, Raasch R, Eron J, Cohen M. Antiretroviraldrug concentrations in semen: implications for sexual transmission of human immunodeficiency virus type 1. Antimicrob Agents Chemother 1999;43: 1817–26. 48. Mayer K, Goldstein R, Lo W, et al. Persistence of human immunodeficiency virus in semen after adding indinavir to combination antiretroviral therapy. Clin Infect Dis 1999; 28:1252–9. 49. Zhang H, Dornadula G, Beumont M, et al. Human immunodeficiency virus type 1 in the semen of men receiving highly active antiretroviral therapy. N Engl J Med 1998; 339:1803–9. 50. Krieger J, Coombs R, Collier A, et al. Fertility parameters in men infected with human immunodeficiency virus. J Infect Dis 1991; 164:464–9. 51. Crittenden J, Handelson D, Stewart G. Semen analysis in human immunodeficiency virus infection. Fertil Steril 1992; 57:1294–9. 52. Politch J, Mayer K, Abbott A, Anderson D. The effects of disease progression and zidovudine therapy on semen quality in human immunodeficiency virus type 1 seropositive men. Fertil Steril 1994; 61:922–8. 53. Muller C, Coombs R, Krieger J. Effects of clinical stage and immunological status on semen analysis results in human immunodeficiency virus type 1–seropositive men. Andrologia 1998; 30(Suppl 1):15–22.


30.
