Initiation of antiretroviral therapy leads to a rapid decline in cervical and vaginal HIV-1 shedding

Transcription

1 Initiation of antiretroviral therapy leads to a rapid decline in cervical and vaginal HIV-1 shedding Susan M. Graham a,g, Sarah E. Holte b,d, Norbert M. Peshu f, Barbra A. Richardson b, Dana D. Panteleeff e, Walter G. Jaoko g, Jeckoniah O. Ndinya-Achola g, Kishorchandra N. Mandaliya h, Julie M. Overbaugh e and R. Scott McClelland a,c,g Background: Antiretroviral therapy (ART) may decrease HIV-1 infectivity in women by reducing genital HIV-1 shedding. Objectives: To evaluate the time course and magnitude of decay in cervical and vaginal HIV-1 shedding as women initiate ART. Methods: This prospective, observational study of 20 antiretroviral-naive women initiating ART with stavudine, lamivudine, and nevirapine measured HIV-1 RNA in plasma, cervical secretions, and vaginal secretions. Qualitative polymerase chain reaction estimated HIV-1 DNA in cervical and vaginal samples. Perelson s two-phase viral decay model and non-linear random effects were used to compare RNA decay rates. Decreases in proviral DNA were evaluated using logistic regression and generalized estimating equations. Results: Significant decreases in the quantity of HIV-1 RNA were observed by day 2 in plasma (P < 0.001), day 2 in cervical secretions (P ¼ 0.001), and day 4 in vaginal secretions (P < 0.001). Modeled initial and subsequent RNA decay rates in plasma, cervical secretions, and vaginal secretions were 0.6, 0.8, and 1.2 log 10 virions/day, and 0.04, 0.05, and 0.06 log 10 virions/day, respectively. The initial decay rate for vaginal HIV-1 RNA was more rapid than for plasma RNA (P ¼ 0.02). Detection of HIV-1 DNA decreased significantly in vaginal secretions during the first week (P < 0.001). At day 28, 10 women had detectable HIV-1 RNA or proviral DNA in genital secretions. Conclusions: Genital HIV-1 shedding decreased rapidly after ART initiation, consistent with a rapid decrease in infectivity. However, incomplete viral suppression in half of these women may indicate an ongoing risk of transmission. ß 2007 Lippincott Williams & Wilkins AIDS 2007, 21: Keywords: AIDS, antiretroviral therapy, cervix, female, HIV-1, vagina, virus shedding Introduction Expanding access to antiretroviral therapy (ART) has begun to improve the health and lives of millions of women living with HIV-1 in resource-limited settings. In addition to providing direct individual benefits, it is hoped that ART expansion will reduce sexual transmission of HIV-1 in these areas [1,2]. While ART has From the Departments of a Medicine, the b Biostatistics, the c Epidemiology, University of Washington, Seattle, Washington, USA, the d Divisions of Public Health Sciences, the e Human Biology, Fred Hutchinson Cancer Research Center, Seattle, Washington, USA, the f Kenya Medical Research Institute, Kilifi, Kenya, the g Department of Medical Microbiology, University of Nairobi, Nairobi, Kenya, and the h Coast Provincial General Hospital, Mombasa, Kenya. Correspondence to Dr S. M. Graham, International AIDS Research and Training Program, University of Washington, Box , 325 Ninth Ave, Seattle, WA , USA. grahamsm@u.washington.edu Received: 6 July 2006; accepted: 13 November ISSN Q 2007 Lippincott Williams & Wilkins 501

2 502 AIDS 2007, Vol 21 No 4 been demonstrated to decrease HIV-1 transmission from mother to child significantly, less is known about the impact it will have on female-to-male transmission [3,4]. Although HIV-1-seropositive persons with low or undetectable plasma viral loads present a much lower risk of sexual transmission to their partners [5], continued genital HIV-1 shedding by woman treated with ART could represent a risk for transmission, diminishing the impact of ART on the spread of HIV-1. Cross-sectional studies have shown that the likelihood of genital HIV-1 shedding is significantly lower in treated women compared with women who are not receiving treatment [6 8]. However, treated women can have higher HIV-1 RNA levels in genital secretions than in plasma, and genital shedding may occur despite an undetectable plasma viral load. Prospective studies are needed to demonstrate the effectiveness of ART at suppressing HIV-1 shedding in the female genital tract. Rapid suppression of HIV-1 in the genital tract has been demonstrated after only 1 week of zidovudine monotherapy [9]. With more potent ART regimens, a small prospective study demonstrated a significant decrease in genital HIV-1 shedding over the first month of treatment [7]. The female genital tract may provide sanctuary sites for viral replication during ART [10], making the evaluation of genital HIV-1 shedding in treated women an important priority. ARTregimens containing non-nucleoside reverse transcriptase inhibitors (NNRTI) are the most commonly used worldwide [11]. With the rapid global expansion in access to ART, an evaluation of the effect of NNRTI-based regimens on HIV-1 infectivity is particularly timely. This study was designed to evaluate the time course and magnitude of genital HIV-1 suppression among female sex workers initiating and continuing NNRTI-based ART in Mombasa, Kenya. It was hypothesized that ART initiation would produce a rapid decrease in the quantity of HIV-1 in genital secretions during the first month of therapy, but that the time course and magnitude of decay might differ from those observed in plasma. Methods Study design The study was a prospective observational study in HIV- 1-seropositive Kenyan women with advanced disease who had multiple visits during the first month of ART. Study population and procedures HIV-1-seropositive, non-pregnant female sex workers attending a municipal clinic were invited to participate if they were eligible for ART according to the Kenyan National Guidelines (CD4 cell count < 200 cells/ml or AIDS-defining illness) and willing to undergo intensive follow-up. The initial ART regimen for all women was stavudine, lamivudine, and nevirapine, provided by the President s Emergency Program for AIDS Relief (PEPFAR) as three separate medications. Directly administered therapy was used as a means to observe one of the two daily ART doses on each weekday (i.e., five times weekly). Adherence was monitored by pill count at each visit. Eligible women were screened for genital infections and samples were collected to determine the baseline prevalence and quantity of genital HIV-1 prior to ART initiation. Treatment began a median of 13 days [interquartile range (IQR), 7 17] after screening. Study visits were conducted on days 0 (ART initiation), 1, 2, 4, 7, 14, and 28. At each visit, women were interviewed by nursing staff using standardized questionnaires about recent sexual behavior, contraceptive practices, and medication adherence. The physical examination performed included a pelvic speculum examination with screening for genital infections. Vaginal specimens for HIV-1 detection were collected by rolling a Dacron swab three full turns against the lateral vaginal wall. Cervical specimens were collected by inserting a Dacron swab gently into the cervical os and rotating two full turns. All participants gave informed consent to participate, and the ethical review committees of the Kenya Medical Research Institute, the University of Washington, and the Fred Hutchinson Cancer Research Center approved the study. Laboratory methods CD4 cell counts were determined using a manual method (Cytosphere, Coulter, Miami, Florida, USA). Plasma samples were frozen at 708C until they could be shipped to Seattle for testing. The Gen-Probe HIV-1 viral load assay (San Diego, California, USA) was used to quantify HIV-1 RNA. This assay has demonstrated accuracy and reproducibility for quantification of HIV-1 from subtypes found in Kenya [12]. The lower limit of detection was 50 copies/swab in genital secretions and 40 copies/ml in plasma. Proviral DNA (a marker for HIV-1-infected cells) was detected by a nested polymerase chain reaction (PCR) for the viral gag region. This assay has been shown to detect as little as one copy of HIV-1 DNA (equivalent to 250 copies/swab using the volumes tested in this study) [13]. Cervical secretions were collected for Gram stain, and the number of polymorphonuclear leukocytes in three non-adjacent high power fields on microscopy was quantified. Non-specific cervicitis was defined as the presence of 30 cells. A nucleic acid amplification assay (Aptima Combo 2, Gen-Probe, San Diego, California, USA) was used to detect Neisseria gonorrhoeae or Chlamydia trachomatis. The presence of yeast and Trichomonas vaginalis were detected by light microscopy of a saline wet preparation. Bacterial vaginosis was evaluated by microscopy of a vaginal Gram stain [14]. The

3 ART and female genital HIV-1 shedding Graham et al. 503 presence of sperm on Gram stain of cervical or vaginal secretions was recorded. Swabs of genital ulcers were cultured for Haemophilus ducreyi on an activated charcoal medium. Screening for syphilis was performed using rapid plasma reagin (RPR, Becton Dickinson, San Jose, California, USA) before and after (within 3 months) ART initiation. Reactive samples were confirmed using a Treponema pallidum hemagglutination assay (TPHA; Biotec, Ipswich, UK). Incident syphilis was defined as a fourfold or more increase in RPR titer with a positive TPHA result. Data analysis Data were analyzed using SPSS version 12.0 (SPSS, Chicago, Illinois, USA), Stata version 9.0 (Stata Corp., College Station, Texas, USA), and S-Plus version 7.0 (Insightful Corporation, Seattle, Washington, USA). The viral load was set at half the lower limit of detection (e.g., 25 copies/swab in genital secretions and 20 copies/ml in plasma) for samples with an HIV-1 RNA result below the lower limit of detection. For baseline HIV-1 RNA values, the results of specimens collected at study screening and at day 0 were averaged. Median HIV-1 RNA values at each time-point were compared using the Wilcoxon signed rank test. The Perelson model for twophase viral decay was used to describe the mean plasma and genital tract viral load over time [15]. Non-linear mixed effects methods with fixed effects for compartment (plasma, cervical, or vaginal) and random effect for initial viral load were used to evaluate decay rates in plasma and in genital secretions. Paired comparisons between DNA PCR results at day 0 and day 28 were made using McNemar s test. The decrease in detection of infected cells was also evaluated for the periods corresponding to the two phases seen in RNA decay (days 0 7, days 8 28) using generalized estimating equations with a logistic link function to describe the outcome detectable DNA yes/no over time. Results were analyzed with and without visits on which sperm was detected (three visits). Exclusion of these visits did not significantly alter the results, so the visits were included in the final analyses. A two-sided P value < 0.05 was considered statistically significant. Results Study participants and genital tract infections Twenty women were enrolled and completed this study, with only one visit missed. One additional woman completed visits through day 4 then withdrew owing to illness. Results from her visits are included in this analysis. Median adherence was 100% (range, ). Sociodemographic and clinical characteristics of participants are presented in Table 1. At baseline, none of the women had gonorrhea, chlamydial infection, or non-specific cervicitis. Although all participants denied abnormal vaginal discharge at baseline, 12 women had vaginal conditions Table 1. Baseline sociodemographic and clinical characteristics. Characteristic Median value or number Age [years (IQR)] 35 (29 38) Education [years (IQR)] 8 (7 11) Marital status (%) Never married Widowed or divorced 9 (429) 12 (571) WHO clinical stage (%) I 1 (48) II 4 (190) III 14 (667) IV 2 (95) Weight [kg (IQR)] 59 (51 68) CD4 cell count [cells/ml (IQR)] 121 (58 161) Use of hormonal contraception (%) a 5 (238) Sexually active in past week (%) b 3 (143) Vaginitis (%) 12 (571) Cervicitis (No) 0 Genital ulcer on examination (%) 1 (48) IQR, interquartile range; WHO, World Health Organization. a All five women used Depo-provera injections. b All three women reported sexual intercourse once with a single partner; two had used a condom. diagnosed by laboratory criteria (eight had bacterial vaginosis, two had vaginal yeast detected, one had both bacterial vaginosis and yeast, and one had trichomoniasis). The woman with vaginal trichomoniasis was treated with a single 2 g dose of oral metronidazole. Women with asymptomatic bacterial vaginosis or yeast colonization were not given additional medication during ART initiation, as treatment of these conditions in asymptomatic non-pregnant women is not medically indicated. Over the course of the study, no new cervical infections or cases of trichomoniasis occurred. One woman had a genital ulcer detected on examination at the baseline visit, and three additional women had genital ulcers detected over the first month of ART. There were no positive chancroid cultures and no incident syphilis cases. Genital HIV-1 RNA levels All participants had detectable cervical and vaginal HIV-1 RNA at baseline. Table 2 shows the decrease in median HIV-1 RNA after ART initiation in plasma and in genital secretions over the 28 days of therapy. The decrease in HIV-1 RNA was statistically significant by day 2 in plasma (P < 0.001), day 2 in cervical secretions (P ¼ 0.001), and day 4 in vaginal secretions (P < 0.001). Median HIV-1 RNA was below the limit of detection in cervical secretions from day 7 onwards and in vaginal secretions from day 14 onwards. At day 28, seven women (35%) had detectable genital HIV-1 RNA at one or both genital sites. HIV-1 RNA levels ranged from 52 to 1258 copies/swab in cervical secretions and from 135 to 1332 copies/swab in vaginal secretions. Modeling of HIV-1 RNA decay rates The Perelson model was used to evaluate HIV-1 RNA decay rates in plasma (Fig. 1a), cervical secretions

4 504 AIDS 2007, Vol 21 No 4 Table 2. Decrease in HIV-1 RNA during the first month of antiretroviral therapy. Time Median HIV-1 RNA Plasma [log 10 copies/ml (IQR)] a Cervical [log 10 copies/swab (IQR)] a Vaginal [log 10 copies/swab (IQR)] a Baseline 5.50 ( ) 3.70 ( ) 3.84 ( ) Day ( ) P ¼ ( ); P ¼ 0.12 Day ( ); P < ( ); P ¼ ( ); P ¼ 0.14 Day ( ); P < ( ); P < Day ( ); P < ( ); P < ( ); P < Day ( ); P < ( ); P < ( ); P < Day ( ); P < ( ); P < ( ); P < IQR, interquartile range. a P values given are in comparison to baseline levels for each sample type. (Fig. 1b), and vaginal secretions (Fig. 1c) [15]. For each compartment, estimates were derived for both the initial, rapid phase of decay (designated d) and the subsequent, slower phase of decay (designated m). For each estimate, a decay rate in log 10 virions/day and a half-life are presented. In cervical secretions, neither rate was significantly different from the decay rates seen in plasma. In contrast, the initial modeled decay rate was significantly more rapid in vaginal secretions than in plasma (P ¼ 0.02). The initial decay rate in vaginal secretions was slightly more rapid than the decay rate in cervical secretions, but this difference was not statistically significant (P ¼ 0.17). Detection of proviral DNA Nineteen women had genital specimens available for detection of HIV-1-infected cells at baseline and day 28. Between day 0 and day 28, the frequency of detection of HIV-1-infected cells decreased from 53% to 26% in cervical secretions (P ¼ 0.13) and from 47% to 5% in vaginal secretions (P ¼ 0.008). Decay rates were also analyzed for HIV-1-infected cells during the two time periods corresponding to the rapid, initial decline (days 0 7) and the subsequent, slower decline (days 8 28) in HIV-1 RNA (Fig. 2). The probability of detecting HIV-1-infected cells in vaginal secretions decreased significantly during the first week of therapy (P < 0.001). Anon-significant decrease in the probability of detecting HIV-1-infected cells in cervical secretions was also seen during this period (P ¼ 0.22). Detection of infected cells in cervical and vaginal secretions remained fairly constant after the first week of ART. These results did not change significantly when genital ulcer disease, use of hormonal contraception, and stage of menstrual cycle were entered in separate multivariate analyses. At the end of 1 month of ART, 10 women had either HIV-1 RNA, infected cells, or both present in cervical or vaginal secretions. Discussion This is the first prospective cohort study to offer a detailed quantitative assessment of the time course and magnitude of cervical and vaginal HIV-1 suppression among women initiating ART. Consequently, the study adds substantially to previous cross-sectional analyses, which demonstrated an association between ART and lower levels of genital virus but did not assess the temporal impact of ARTon this outcome measure [7,8]. Female genital HIV-1 shedding decreased rapidly after ART initiation, and the two-phase decline in HIV-1 RNA was similar to the pattern observed in plasma. Steep declines in genital viral levels are likely to represent a rapid reduction in infectivity. However, suppression of genital HIV-1 shedding was not complete, which may indicate an ongoing risk of transmission to sexual partners. In plasma, a two-phase decay in HIV-1 RNA has been reported after ART initiation, resulting in a decrease of approximately 99% over the first 2 weeks of treatment [15]. The rapid, initial decrease has been attributed to the elimination of free virus (half-life, 6 h) and loss of activated lymphocytes (half-life approximately 1.6 days). The loss of long-lived infected cells such as macrophages and dendritic cells (half-life, 1 4 weeks) is thought to be the major component of the second phase, with a minor contribution from an effect on latently infected lymphocytes (half-life, weeks) [15]. We found that the initial rate of HIV-1 RNA decline was more rapid in vaginal secretions than in plasma. In addition, the rate of detection of HIV-1-infected cells decreased significantly in vaginal secretions but not in cervical secretions over the first week of treatment. These findings could relate to either decreased production or increased clearance of the virus. The source of HIV-1 in female genital secretions is not clearly established, but most genital virus is thought to arise from the cervix and possibly the upper genital tract [4]. If vaginal HIV-1 RNA and infected cells are mostly produced higher in the genital tract, then more rapid clearance in this compartment could simply be a result of more rapid turnover of vaginal secretions compared with cervical secretions. However, HIV-1 can be detected in the vaginal secretions of women who have undergone hysterectomy [16], and independent risk factors for vaginal shedding have been identified [17]. Further

5 ART and female genital HIV-1 shedding Graham et al. 505 Fig. 1. Decrease in HIV-1 RNA in plasma (a), cervical secretions (b), and vaginal secretions (c) during the first month of antiretroviral therapy. Open, connected circles represent the data points for individual patients. The bold line represents the modeled estimate. t 1 / 2, half-life for each phase (equal to ln(2) divided by the decay rate in log 10 virions/day); d, estimate for the initial, rapid phase of decay; m, estimate for the subsequent, slower phase of decay. P values are presented for comparisons of the two decay rates in each compartment with those seen in plasma.

6 506 AIDS 2007, Vol 21 No 4 (a) (b) PCR positive (%) P < P = Days post treatment PCR positive (%) P = P = Days post treatment Fig. 2. Decrease in the probability of detecting HIV-1 DNA in vaginal (a) and cervical (b) secretions during the first month of antiretroviral therapy. Generalized estimating equations with a logistic link function were used to describe graphically the outcome using the polymerase chain reaction (PCR) for detectable DNA yes/no over time for days 0 7 and for days research is needed to delineate the sites and extent of ongoing HIV-1 replication in the female genital tract before and after ART initiation. There are several strengths of this study. To our knowledge, it is the largest and most intensive investigation of genital HIV-1 shedding in women starting ART to date. All women were on the same regimen, and the NNRTI-based regimen evaluated in this study is the most common ART combination currently used worldwide [11]. The study participants had very high adherence, limiting variability owing to regimen potency or missed doses. Shedding of both HIV-1 RNA and infected cells were measured, since it is not known which of these markers may more accurately predict the risk for sexual transmission. Finally, mathematical modeling of decay rates maximized the power of the analysis to evaluate HIV-1 shedding over time in genital secretions. This study also had limitations. First, highly sensitive assays may detect small quantities of HIV-1 RNA or DNA of unknown significance in terms of transmission risk. Nonetheless, the magnitude of reduction in genital HIV-1 RNA demonstrated in this study (over 100-fold in each compartment) reflects an important decrease in viral replication, which could considerably reduce the risk of sexual transmission. Second, because the need for intensive sampling limited the size of this study, it was designed to show the overall effect of ART initiation rather than the effects of other factors such as genital ulcer disease, use of hormonal contraception, and menstrual stage on the modeled decrease in HIV-1 RNA. Finally, we did not test for herpes simplex virus type 2, but the effect of non-ulcerative shedding of this virus on HIV-1 RNA levels is unlikely to have substantially altered our results compared with the magnitude of effect owing to ART initiation [18]. Genital HIV-1 suppression was achieved in this study despite the reported low penetration of stavudine into genital secretions [19]. Further research is needed to determine if regimens including other drugs that penetrate the genital compartment poorly, especially protease inhibitors, will result in similar decreases in genital HIV-1 shedding in women. With expanding global access to medications, ART will significantly reduce genital HIV-1 shedding among treated women. While this reduction in viral shedding may lower transmission risk, genital mucosal HIV-1 RNA and infected cells may persist in some women despite treatment. Acknowledgements We would like to thank the research staff in Mombasa for their hard work, the Mombasa Municipal Council for use of clinical space, Coast Provincial General Hospital for provision of laboratory space, and the Director, Kenya Medical Research Institute, for permission to publish this paper. Special thanks go to the women who participated in this study. This study was supported by National Institutes of Health (NIH) grant AI and Puget Sound Partners for Global Health grant S. Graham was supported by Fogarty International Center grant D43 TW S. Holte was supported by NIH grant R01-AI References 1. Wood E, Braitstein P, Montaner JS, Schechter MT, Tyndall MW, O Shaughnessy MV, et al. Extent to which low-level use of antiretroviral treatment could curb the AIDS epidemic in sub-saharan Africa. Lancet 2000; 355: Auvert BMS, Puren A, Taljaard D, Carael M, Williams B. Can highly active antiretroviral therapy reduce the spread of HIV?: A study in a township of South Africa. J Acquir Immune Defic Syndr 2004; 36: McIntyre J. Preventing mother-to-child transmission of HIV: successes and challenges. Br J Obstet Gynaecol 2005; 112: Coombs RW, Reichelderfer PS, Landay AL. Recent observations on HIV type-1 infection in the genital tract of men and women. AIDS 2003; 17: Quinn TC, Wawer MJ, Sewankambo N, Serwadda D, Li C, Wabwire-Mangen F, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med 2000; 342: Fiore JR, Suligoi B, Saracino A, Di Stefano M, Bugarini R, Lepera A, et al. Correlates of HIV-1 shedding in cervicovaginal secretions and effects of antiretroviral therapies. AIDS 2003; 17:

7 ART and female genital HIV-1 shedding Graham et al Cu-Uvin S, Caliendo AM, Reinert S, Chang A, Juliano-Remollino C, Flanigan TP, et al. Effect of highly active antiretroviral therapy on cervicovaginal HIV-1 RNA. AIDS 2000; 14: Kovacs A, Wasserman SS, Burns D, Wright DJ, Cohn J, Landay A, et al. Determinants of HIV-1 shedding in the genital tract of women. Lancet 2001; 358: Mbori-Ngacha D, Richardson BA, Overbaugh J, Panteleef DD, Nduati R, Steele M, et al. Short-term effect of zidovudine on plasma and genital human immunodeficiency virus type 1 and viral turnover in these compartments. J Virol 2003; 77: Tirado G, Jove G, Kumar R, Noel RJ, Reyes E, Sepulveda G, et al. Differential virus evolution in blood and genital tract of HIVinfected females: evidence for the involvement of drug and non-drug resistance-associated mutations. Virology 2004; 324: Ivers LC, Kendrick D, Doucette K. Efficacy of antiretroviral therapy programs in resource-poor settings: a meta-analysis of the published literature. Clin Infect Dis 2005; 41: Emery S, Bodrug S, Richardson BA, Giachetti C, Bott MA, Panteleeff D, et al. Evaluation of performance of the Gen-Probe human immunodeficiency virus type 1 viral load assay using primary subtype A, C, and D isolates from Kenya. J Clin Microbiol 2000; 38: Rousseau CM, Nduati RW, Richardson BA, John-Stewart GC, Mbori-Ngacha DA, Kriess JK, et al. Association of levels of HIV-1 infected breast milk cells and risk of mother-to-child transmission. J Infect Dis 2004; 190: Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial vaginosis is improved by a standard method of Gram stain interpretation. JClinMicrobiol1991; 29: Perelson AS, Essunger P, Cao Y, Vesanen M, Hurley A, Saksela K, et al. Decay characteristics of HIV-1-infected compartments during combination therapy. Nature 1997; 387: Farrar DJ, Cu Uvin S, Caliendo AM, Costello SF, Murphy DM, Flanigan TP, et al. Detection of HIV-1 RNA in vaginal secretions of HIV-1-seropositive women who have undergone hysterectomy. AIDS 1997; 11: Wang CC, McClelland RS, Reilly M, Overbaugh J, Emery SR, Mandaliya K, et al. The effect of treatment of vaginal infections on shedding of human immunodeficiency virus type 1. J Infect Dis 2001; 183: McClelland RS, Wang CC, Overbaugh J, Richardson BA, Corey L, Ashley RL, et al. Association between cervical shedding of herpes simplex virus and HIV-1. AIDS 2002; 16: Dumond J, Yeh R, Patterson K, Corbett A, Jung B, Rezk N, et al. First dose and steady-state genital tract pharmacokinetics of ten antiretroviral drugs in HIV-infected women: implications for pre- and post-exposure prophylaxis. Conference on Retroviruses and Opportunistic Infections, Denver, February 2006 [abstract 129].