Editorial 44.pmd

s seleccionados 11th Conference on Retroviruses and Opportunistic Infections February 8-11, 2004 San Francisco, CA, USA 23 EFFECT OF TREATMENT DURING VERSUS AFTER ACUTE RETROVIRAL SYNDROME (ARS) ON HIV VIRAL LOAD AND CD4 CELL COUNTS WITHIN 3 YEARS OF INFECTION N Voirin*1, D Smith2, J P Routy3, M Legault3, D Baratin4, C Trepo1,5, L Cotte5, J M Livrozet4, J L Touraine4, D A Cooper2, A Gayet-Ageron4, J Ritter4, J Fabry4, and P Vanhems1,4 1INSERM U271, Lyon, France; 2Univ. of New South Wales, Sydney, Australia; 3McGill Univ., Montreal, Canada; 4Edouard Herriot Hosp., Lyon, France; and 5Hotel Dieu Hosp., Lyon, France Background: Starting antiretroviral therapy (ART) during ARS might influence immediate immunological response and control of virological replication but it is unknown if this benefit persists over time. We therefore compared the HIV viral load and the CD4 cell counts at different time points from the initiation of ART among patients treated during or treated after ARS. Methods: Data from patients enrolled from 1995 to 2003 in 3 prospective primary HIV infection cohorts with similar follow-up (Lyon, Montreal, and Sydney) were analyzed. We compared mean HIV viral load and CD4 counts at initiation of therapy, 6 months and 1, 2, and 3 years after start of therapy between patients with documented ARS treated during or after ARS, using Kruskal-Wallis rank sum test. Results: Of a total of 203 patients, 188 (93%) were male, mean age 34.5 years. The presumed route of HIV infection was men who had sex with men for 148 (73%), heterosexual contact for 21 (10%), and injecting drug use for 33 (16%). ART was started after ARS for 117 (58%) (Gr1), during ARS for 51 (25%) (Gr2), and 35 (17%) received no therapy during the follow-up (Gr0). The mean time between onset of ARS and start of therapy was 164 days for Gr1 and 27 days (p = 0.001) for Gr2. No difference was observed between the 3 groups for gender (p = 0.65) and age (p = 0.47). The table shows the HIV viral load and CD4 count for each group at the various time points. Similar results persist after adjustment on ARS severity score. 38 HIV RESISTANCE AND TRANSMISSION FOLLOWING SINGLE-DOSE NEVIRAPINE IN A PMTCT COHORT N Martinson*1,2,3, L Morris1,2,3, G Gray1,2, D Moodley4, P Lupondwana1,2, C Chezzi1,2, S Cohen1, C Pillay1,2, A Puren1, M Ntsala1, J Sullivan5, J Steyn2, and J McIntyre2 1Natl. Inst. for Communicable Diseases, Johannesburg, South Africa; 2Perinatal HIV Res. Unit, Johannesburg, South Africa; 3Sch. of Med., Johns Hopkins Univ., Baltimore, MD, USA; 4King Edward Hosp., Durban, South Africa; and 5Univ. of Massachusetts Med. Sch., Worcester, USA Background: Nevirapine (NVP), given to mothers in labor and to neonates within 72 hours of delivery, is an effective, simple PMTCT regimen. Previous studies have shown that transient resistance develops in those exposed to this regimen. This study was designed to assess genotypic resistance induced by NVP (NVPR). Methods: We enrolled 623 HIV-infected mothers at 2 tertiary hospitals in South Africa from 32 to 38 weeks’ gestation. We report preliminary findings on baseline and follow-up visits scheduled at 6 weeks postpartum. High-level NVPR was defined as K103N, V106A/M, Y181C, YI88C, and G190A. Results: The median CD4 and viral load of pregnant women was 393 x106/l (IQR 266 to 585) and 28 700 copies/mL (IQR 6560 to 100,000). No drug naïve women had resistance at baseline; 456 were followed at a median time of 7.0 weeks postpartum (IQR 6.3 to 10.3); all but 4 were clade C; 38.8% (95%CI: 34 to 44) of mothers and 42.4% (95%CI: 30 to 55) of their infants were found to have NVPR. Maternal resistance in the first (4 to 6 weeks postpartum), second (6 to 7 weeks), third (7 to 10 weeks), and fourth quartiles (10 to 36 weeks) of the follow-up visit was 43%, 44%, 44%, and 24%, respectively (test for trend p = 0.006). Maternal mutations included K103N (31%), Y181C (12%), and Y188C (8.1%); 21% had a single mutation, 13% had 2, and 5%, 3 or 4 mutations. Infants’ mutations included Y181C (32%), K103N (12%) and Y188C (5%). The 10-week HIV MTCT rate was 8.6% (95%CI: 6.0 to 11.2). There was an association between postpartum NVPR and transmission (OR 2.9, 95%CI: 1.4 to 6.1) confounded by maternal viral load and CD4 count. Baseline CD4, log baseline viral load, the time from labor NVP dose to the postpartum blood draw and the total number of times a mother took NVP in her pregnancy were statistically significantly associated with development of resistance. Conclusions: Maternal and infant resistance induced by NVP is high. Its effect on subsequent pregnancies and choice of HAART regimen remains to be explored. Vol 12 No 44 Junio 2004 52 39 LOW-FREQUENCY NNRTI-RESISTANT VARIANTS CONTRIBUTE TO FAILURE OF EFAVIRENZ-CONTAINING REGIMENS J Mellors*1, S Palmer2, D Nissley2, M Kearney2, E Halvas1, C Bixby1, L Demeter3, S Eshleman4, K Bennett5, S Hart6, F Vaida5, M Wantman5, J Coffin2, and S Hammer for the ACTG 398 Study Group7 1Univ. of Pittsburgh, PA, USA; 2Drug Resistance Prgm., NCI, NIH, DHHS, Frederick, MD, USA; 3Univ. of Rochester, NY, USA; 4Johns Hopkins Univ., Baltimore, MD, USA; 5Harvard Sch. of Publ. Hlth., Boston, MA, USA; 6Frontier Sci. and Technology Res. Fndn., Amherst, NY, USA; and 7Columbia Univ., New York, NY, USA Background: The role of minor (low-frequency) drug-resistant variants in failure of antiretroviral therapy is not defined. In ACTG 398, 212 NNRTI-experienced and 269 NNRTI-naïve patients were randomized to efavirenz (EFV), abacavir, adefovir, and amprenavir plus a second PI or placebo. This allowed us to examine relations between NNRTI experience, NNRTI resistance, virologic response, and emergence of EFV resistance. Methods: Baseline genotypes were obtained in 452/481 patients by a standard (ViroSeq v2.0). Minor NNRTI-resistant variants were sought with single genome RTPCR and sequencing (SGS) and a Ty1/HIV-1 RT yeast system that detects EFV-resistant colonies. Phylogenetic analyses were performed with the neighbor joining method. Results: Virologic failure was associated with NNRTI experience (p <0.001), baseline NNRTI mutations (p <0.001), and development of EFV resistance (p <0.001). Standard genotype did not identify NNRTI mutations in baseline samples from 48 (22%) NNRTI-experienced patients. Virologic response in this group was no better than in the experienced group with baseline NNRTI mutations (n = 166), suggesting that standard genotype was insensitive for NNRTI-resistant variants. Baseline plasma from a random sample of 11 NNRTI-experienced and 12 NNRTI-naïve patients experiencing virologic failure despite negative standard genotypes for NNRTI mutations were assayed for NNRTI-resistant variants. Such variants were identified by SGS in 6/11 NNRTI-experienced patients with the following frequencies per positive patient: 181C and 190A (5/15 sequences); 181C (3/19); 181C (3/22); 108I (2/35); 103N (1/33); and 103N (1/34). By comparison, NNRTI-resistant variants were found in 2/12 NNRTI-naïve patients: 100I and 225L (1/33 sequences each) and 103N (1/41 sequences). The Ty1/HIV-1 RT assay detected EFV-resistant colonies in 8/10 NNRTIexperienced patients with the following frequencies: 7.2, 3.3, 2.8, 1.7, 1.7, 0.9, 0.8, and 0.6%. In NNRTI-naïve patients, resistant colonies were found in only 2/12 patients with frequencies of 0.6% and 0.4%. Phylogenetic analyses showed clustering of the baseline NNRTI-resistant variants identified by SGS with the genotype at virologic failure in 5/6 NNRTI-experienced patients. In the 2 NNRTI-naïve patients, the 103N variant clustered with the failure genotype but the 100I/225L variants did not. Conclusions: Prior NNRTI expsoure can select minor NNRTI-resistant variants that are missed by standard genotyping and can contribute to failure of EFV-based regimens. 41LB EXPOSURE TO INTRAPARTUM SINGLE-DOSE NEVIRAPINE AND SUBSEQUENT MATERNAL 6-MONTH RESPONSE TO NNRTI-BASED REGIMENS G Jourdain*1, N Ngo-Giang-Huong1, P Tungyai2, A Kummee2, C Bowonwatanuwong3, P Kantipong3, P Lechanachai4, S Hammer5, M Lallemant2, and Perinatal HIV Prevention Trial Group 1Harvard Sch. of Publ. Hlth., Boston, MA, USA; 2PHPT-Inst. de Recherche pour le Développement, France and Thailand; 3Ministry of Publ. Hlth., Bangkok, Thailand; 4Chiang Mai Univ., Faculty of Associated Med. Sci., Thailand; and 5Columbia Univ. Med. Ctr., New York, NY, USA Background: Single-dose nevirapine (SD-NVP) is efficacious in preventing HIV mother-to-child transmission (PMTCT) but resistance mutations to non-nucleoside reverse transcriptase inhibitors (NNRTI) are detectable in the postpartum period in a substantial proportion of mothers. Their clinical significance is unknown. Methods: Immunocompromized women participating in PHPT2, a randomized trial in Thailand, which showed an 80% decrease in perinatal HIV with the addition of intrapartum NVP to zidovudine prophylaxis, were offered antiretroviral therapy. A 10-day postpartum sample was assessed for RT mutations (ViroSeq HIV-1 Genotyping System; WWW.IASUSA.ORG), first in a random sample of 90 women stratified on CD4 and virus load; and then in all women who subsequently received an NNRTI-based regimen and for whom viral load could be assayed at 3 and/or 6 (+ 1.5) months (Cobas Amplicor HIV-1 v1.5 Roche). Results: NNRTI-resistance mutations were detectable in 18% of the random sample of women (K103N, G190A, or Y181C). At 3 months, 80% of the 66 women with at least one mutation, 87% of the 112 SD-NVP exposed women with no mutation and 88% of the 41 non exposed women had a viral load <400 copies/mL (viral load <50: 45%, 46%, and 54%). At 6 months, 68% of the 50 women with at least one mutation, 80% of the 92 exposed women without mutation and 85% of the 27 non exposed women had a viral load <400 (p for trend = 0.057) (viral load <50: 38%, 50%, 74%; p for trend = 0.0034). Of SD-NVP-exposed women without and with mutations who started therapy more than 6 months after delivery, 91% and 77%,