Persistent low-level HIV-1 RNA between 20 and 50 copies/ml in antiretroviral-treated patients: associated factors and virological outcome

J Antimicrob Chemother 2012; 67: 2231 2235 doi:10.1093/jac/dks191 Advance Access publication 29 May 2012 Persistent low-level HIV-1 RNA between 20 and 50 copies/ml in antiretroviral-treated patients: associated factors and virological outcome Charlotte Charpentier 1 *, Roland Landman 2,Cédric Laouénan 3,Véronique Joly 2, Gwenn Hamet 2, Florence Damond 1, Françoise Brun-Vézinet 1, France Mentré 3, Diane Descamps 1 and Patrick Yeni 2 1 Laboratoire de Virologie, Assistance Publique-Hôpitaux de Paris (AP-HP), Groupe Hospitalier Bichat-Claude Bernard, HUPNVS, Université Paris Diderot, Paris 7, Sorbonne Paris Cité, EA4409, 75018 Paris, France; 2 Service des Maladies Infectieuses et Tropicales, AP-HP, Groupe Hospitalier Bichat-Claude Bernard, HUPNVS, Université Paris Diderot, Paris 7, Sorbonne Paris Cité, EA4409, 75018 Paris, France; 3 AP-HP, Groupe Hospitalier Bichat-Claude Bernard, Service de Biostatistiques, Université Paris Diderot, Sorbonne Paris Cité, UMR 738, INSERM, UMR 738, 75018 Paris, France *Corresponding author. Hôpital Bichat-Claude Bernard, Laboratoire de Virologie, 46 Rue Henri Huchard, 75018 Paris, France. Tel: +33-1-40-25-61-50; Fax: +33-1-40-25-67-69; E-mail: charlotte.charpentier@bch.aphp.fr Received 28 February 2012; returned 22 March 2012; revised 6 April 2012; accepted 19 April 2012 Objectives: The aim of our study was to identify factors associated with persistent low-level viraemia (LLV) in HIV-infected patients under suppressive antiretroviral therapy and to assess the virological outcome of these patients. Methods: LLV was defined as at least two HIV-1 RNA values between 20 and 50 copies/ml during 1 year of follow-up. We compared patients with all values,20 copies/ml (LLV2) and patients with LLV (LLV+). The blip ratio was defined as (number of HIV-1 RNA values.50 copies/ml)/(number of HIV-1 RNA determinations) before study inclusion. Results: Among the 656 patients included, 5.8% were in group LLV+. CDC stage B/C at study inclusion and a higher blip ratio before the study period were the only factors independently associated with LLV. During the 1 year follow-up, the proportion of patients experiencing virological failure was not different between the LLV2 and LLV+ groups, and 40% of patients shifted from LLV+ to LLV2 status. Conclusions: LLV was infrequent in our series and the follow-up did not evidence a higher rate of virological failure than in fully suppressed patients. LLV seems to be a transient phenomenon that might be driven by residual ongoing viral replication and/or viral release and/or accuracy of viral load assay at lower values. Keywords: HIV, viral load assay, low-level viraemia Introduction The objective of combined antiretroviral therapy (cart) is to reach and maintain plasma virological suppression,50 copies/ml. 1 Several studies using ultra-sensitive techniques evidenced residual low-level viraemia (LLV) in a high proportion of virologically suppressed patients under cart. 2 6 In clinical trials and international recommendations, the level of 50 copies/ml is the current accepted threshold for virological suppression. Upgraded commercial assays are now able to detect viral loads,50 copies/ml, such as the COBAS AmpliPrep/COBAS TaqMan HIV-1 v2.0 assay with a limit of detection of 20 copies/ml. Thus, it is expected that some patients display HIV-1 RNA values between 20 and 50 copies/ml. However, the relevance in the clinical setting of LLV between 20 and 50 copies/ml is still uncertain. The aim of our study was to identify factors associated with the detection of persistent LLV, defined as several HIV-1 RNA values between 20 and 50 copies/ml, in patients receiving stable cart and to assess the virological outcome of these patients. Patients and methods Study patients This was a longitudinal study among the 4820 patients followed in the clinical unit of infectious diseases of Hôpital Bichat-Claude Bernard, # The Author 2012. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com 2231

Charpentier et al. Paris, France. Among the 4820 patients, 3836 were receiving an antiretroviral-based treatment. The first period of the study was retrospective from May 2009 to June 2010 (Period 1). The second period of the study, from July 2010 to June 2011 (Period 2), was a prospective follow-up of patients included during Period 1. Patients were eligible for inclusion in Period 1 if they met the following inclusion criteria: (i) stable cart for at least 6 months before study inclusion; (ii) all HIV-1 RNA values,50 copies/ml during Period 1; and (iii) at least three HIV-1 RNA determinations during Period 1. Two groups of patients were defined: group LLV2 included patients with all HIV-1 RNA values,20 copies/ml during Period 1; and group LLV+ included patients with at least two HIV-1 RNA values between 20 and 50 copies/ml, consecutive or not, during Period 1. Patients with an isolated HIV-1 RNA value between 20 and 50 copies/ml during Period 1 were excluded from the study. The study protocol was approved by an institutional ethics review board (Saint-Germain en Laye, France) and all patients provided their written informed consent. Measurement of plasma HIV-1 RNA level HIV-1 RNA quantification assay was performed using the new COBAS AmpliPrep COBAS TaqMan HIV-1 test, version 2.0 (CAP/CTM v2.0) (Roche Molecular Systems, Branchburg, NJ, USA) with the lower quantification limit of 20 copies/ml. No repeat of the HIV-1 RNA quantification assay was performed. In order to assess the occurrence of transient detectable viraemia.50 copies/ml and,1000 copies/ml a blip ratio was defined as (number of HIV-1 RNA values.50 copies/ml)/(number of HIV-1 RNA determinations during period of interest). The blip ratio was calculated: (i) before study inclusion during all the time the patient received ongoing cart; and (ii) during Period 2. By definition, no detectable viraemia.50 copies/ml was allowed during Period 1. Genotypic resistance tests Bulk sequencing of protease and reverse transcriptase was performed according to the complete sequencing procedures and primer sequences described at www.hivfrenchresistance.org. Resistance mutations were identified according to the IAS-USA list. 7 Virological outcome To assess the impact of LLV on virological outcome, all patients included in Period 1 were further followed up during Period 2. All HIV-1 RNA values available during Period 2 were collected. As in Period 1, only patients with at least three HIV-1 RNA determinations during this period were included in Period 2. Virological failure was defined as two consecutive HIV-1 RNA values.50 copies/ml. Several outcomes were assessed during Period 2: (i) proportion of patients with all HIV-1 RNA values,20 copies/ml; (ii) proportion of patients with LLV; and (iii) proportion of patients with virological failure. Statistical analysis Continuous variables were expressed as medians and IQRs, and categorical variables were expressed as numbers and percentages. To compare clinical and virological characteristics, at study inclusion or before, of patients in group LLV2 or LLV+ of Period 1, the Wilcoxon test was used for continuous variables and Fisher s exact test was used for categorical variables in the univariate analysis. Variables with P,0.2 in univariate analysis were included in a multivariate logistic regression model using a backward elimination procedure and a significance level of P¼0.05. Two-way interactions were studied between significant variables in multivariate analysis; when interactions were significant at the level of P¼0.05 they were retained in the final multivariate model. For Period 2, the occurrence of virological failure was compared between group LLV2 and group LLV+ using Fisher s exact test and the blip ratio using the Wilcoxon test. All tests were two-sided at the 0.05 significance level. Analyses were performed with SAS statistical software (version 9.2; SAS Institute, Cary, NC, USA). Results Factors associated with persistent LLV During Period 1, 656 patients fulfilled the inclusion criteria. Among them, 618 patients (94.2%) displayed all HIV-1 RNA values,20 copies/ml and were included in group LLV2. The 38 remaining patients (5.8%) had at least two HIV-1 RNA values between 20 and 50 copies/ml and were included in group LLV+. The median number of HIV-1 RNA determinations was similar in both groups (n¼4). The characteristics at study inclusion of the patients in both groups and the results of univariate analysis are reported in Table 1. No differences between the LLV2 and LLV+ groups could be observed in the therapeutic history of the patients (Table 1). Neither the duration of virological suppression nor the nature of the ongoing cart regimen differed significantly between the two groups. A protease inhibitor (PI)-based regimen was received by 49% and 50% of patients in groups LLV2 and LLV+, respectively. In the univariate analysis, the following factors were associated with being in group LLV+: age (P¼0.04); CDC clinical stage at study inclusion (P¼0.004); CDC clinical stage at time of initiation of the first cart regimen (P¼0.006); HIV-1 RNA value at initiation of the ongoing cart (P¼0.02); CD4 cell count at initiation of the ongoing cart (P¼0.02); duration of ongoing cart (P¼0.047); and a higher blip ratio during ongoing cart before study inclusion (P¼0.007). In the multivariate analysis, only CDC stage B/C at study inclusion (OR 2.9; 95% CI 1.4 5.9; P¼0.003) and a higher blip ratio (OR 0.9; 95% CI 0.9 1.0; P¼0.001) were independently associated with persistent LLV. No significant difference in CD4 cell count variations during Period 1 was observed between LLV2 and LLV+ patients (median change in CD4 during Period 1: 11 versus 53 cells/mm 3 in groups LLV2 and LLV+, respectively). Virological characteristics in study groups Virological analysis was performed in order to evidence possible differences in viral characteristics between viruses from patients of groups LLV2 and LLV+. Genotypic resistance tests were available before study inclusion in 135 patients in group LLV2 and 24 patients in group LLV+. No significant difference was evidenced in the proportion of HIV-1 non-b subtypes between the two groups: 66% in group LLV2 versus 67% in group LLV+ (P¼1.0). The median number of resistance-associated mutations was similar in both groups (two versus one in groups LLV2 and LLV+, respectively; P¼0.7), with no differences among the antiretroviral drug classes (data not shown). 2232

HIV-1 low-level viraemia JAC Table 1. Characteristics of study patients included in group LLV2 (all HIV-1 RNA values,50 copies/ml) and in group LLV+ (at least two HIV-1 RNA values between 20 and 50 copies/ml) Group LLV2 (n¼618) Group LLV+ (n¼38) Univariate analysis P value Multivariate analysis OR (95% CI) P value At inclusion age (years) 44 (38 50) 47 (43 53) 0.04 male (%) 61 74 0.1 CD4 cell count (/mm 3 ) 552 (405 738) 612 (403 734) 0.7 CD4 cell count nadir (/mm 3 ) 168 (69 255) 124 (26 252) 0.2 time since HIV diagnosis (years) 11 (7 17) 14 (6 18) 0.29 duration of any ARV therapy (months) 92 (44 144) 127 (31 156) 0.6 duration of ongoing ARV therapy (months) 28 (16 41) 21 (10 36) 0.054 total number of ARV drugs received 7 (4 9) 7 (4 11) 0.4 number of virological failures a 1 (0 2) 1 (0 2) 0.32 ongoing ARV-based treatment 2 NRTIs+1 PI/ritonavir 302 (49) 19 (50) 2 NRTIs+1 NNRTI 211 (34) 9 (24) 0.23 ARV-based regimen without INI 29 (5) 2 (5) ARV-based regimen with INI 37 (6) 4 (10) other triple combination ARV regimen 23 (4) 1 (3) monotherapy or bitherapy ARV regimen 16 (2) 3 (8) duration of HIV-1 RNA values,50 copies/ml 9 (3 23) 7 (2 17) 0.047 before inclusion (months) CDC stage (%) A 56 32 0.004 2.9 (1.4 5.9) 0.003 B/C 44 68 At initiation of ongoing ARV therapy HIV-1 RNA level (log 10 copies/ml) 1.7 (1.7 3.4) 2.0 (1.7 4.5) 0.02 CD4 cell count (/mm 3 ) 402 (262 587) 254 (142 583) 0.02 blip ratio (%) 0 (0 8) 6 (0 13) 0.007 0.9 (0.9 1.0) 0.001 At initiation of first-line ARV therapy HIV-1 RNA level (log 10 copies/ml) 4.6 (3.7 5.3) 4.8 (4.5 5.1) 0.7 CDC stage (%) A 70 47 0.006 B/C 30 53 CD4 cell count (/mm 3 ) 228 (122 346) 204 (48 310) 0.3 ARV, antiretroviral; INI, integrase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor. Continuous variables are expressed as medians and IQRs, and categorical variables are expressed as numbers and percentages (unless otherwise stated). Blip ratio [(number of HIV-1 RNA values.50 copies/ml)/(number of HIV-1 RNA determinations before study inclusion)]. a A virological failure was defined as two consecutive HIV-1 RNA values.200 copies/ml. Virological outcome of patients with persistent LLV During Period 2, 413 (66.8%) patients from group LLV2 and 25 (65.8%) from group LLV+ were assessed. The proportion of patients experiencing virological failure was not significantly different between groups LLV2 and LLV+ (4% versus 8%, respectively; P¼0.32) (Table 2). The median HIV-1 RNA value at the time of failure was 250 copies/ml (IQR 94 1342) in group LLV2 and 88 copies/ml in group LLV+ (IQR 84 123). During Period 2, the blip ratio was not different between the two groups. The follow-up showed that 267 (65%) and 11 (44%) patients displayed all HIV-1 RNA values,20 copies/ml in groups LLV2 and LLV+, respectively (Table 2). The proportion of patients experiencing LLV was higher in group LLV+ (16%, n¼4) when compared with group LLV2 (2%, n¼7) (P¼0.002). Discussion The new commercial HIV-1 RNA quantification assays led to the identification of patients with HIV-1 RNA values between 20 and 2233

Charpentier et al. Table 2. Virological outcome of patients during Period 2 of the study Group LLV2 (n¼413) Group LLV+ (n¼25) P value All HIV-1 RNA values 65 44 0.053,20 copies/ml (%) Two HIV-1 RNA values between 2 16 0.002 20 and 50 copies/ml (%) Two consecutive HIV-1 RNA values 4 8 0.320.50 copies/ml (%) Blip ratio, % (IQR) 0 (0 0) 0 (0 25) 0.070 Blip ratio¼(number of HIV-1 RNA values.50 copies/ml)/(number of HIV-1 RNA determinations during Period 2). 50 copies/ml, but there is no standard therapeutic monitoring for this situation. In the present study, based on the follow-up of 656 patients under stable suppressive cart, detection of persistent LLV between 20 and 50 copies/ml was infrequent, observed in 5.8% of our population. The detection of persistent LLV was not significantly linked to the nature of the antiretroviralbased regimen or to CD4 cell count or HIV-1 RNA value at initiation of the first-line regimen or ongoing treatment. However, CDC clinical stage B/C was associated with higher occurrence of LLV. In addition, patients displaying persistent LLV more frequently had a past history of detectable viraemia.50 copies/ml before study inclusion than patients with fully suppressed replication. Regarding HIV-1 subtype and the number of resistance-associated mutations, no significant difference could be observed between viruses from patients with LLV and those from fully suppressed patients. Finally, LLV does not seem to have a deleterious effect on immune status. Taken together, these findings likely suggest that the detection of persistent LLV between 20 and 50 copies/ml seems not to be a random biological phenomenon. A higher prevalence of LLV (25.3%) was observed in a previous study assessing the prevalence of HIV-1 RNA values between 20 and 50 copies/ml in virologically suppressed patients (n¼91). 8 This difference might be explained by the study design: a transverse study with only one HIV-1 RNA measure in the study of Pascual-Pareja et al. 8 compared with a longitudinal study with at least three HIV-1 RNA determinations over a 1 year period in our study. In the study of Pascual-Pareja et al., 8 factors associated with the detection of HIV-1 RNA between 20 and 50 copies/ml were HIV-1 RNA level before initial cart and time on cart, different from those observed in our study. Previous studies have reported an association between the level of residual viraemia and the nature of the antiretroviral-based regimen. Thus, two studies showed that the use of non-nucleoside reverse transcriptase inhibitors was associated with lower levels of residual viraemia when compared with the use of PIs, 4 6 and another study showed a higher efficacy of nevirapine than efavirenz in achieving an HIV-1 RNA level,1 copy/ml. 5 In the present study, we did not demonstrate an association between the nature of cart and the detection of LLV. However, we assessed a different stratum of viral replication, between 20 and 50 copies/ml, probably reflecting a distinct physiopathological mechanism from viral replication at the level of,3 copies/ml, which might explain the differences in results. The phenomenon of LLV can result from ongoing residual viral replication or from viral release mechanisms. 9 Furthermore, quantification assays might also drive this phenomenon, by the significant widening of the coefficients of variation at the lower limits of the quantitative range. 10 Thus, the Department of Health and Human Services (DHHS, USA) guidelines recently reported that for the purposes of clinical trials the AIDS Clinical Trials Group (ACTG) currently defines virological failure as a confirmed viral load.200 copies/ml, which eliminates most cases of apparent viraemia caused by blips or assay variability. 11 We assessed the virological outcome of the 38 patients displaying persistent LLV during the year following the detection of LLV between 20 and 50 copies/ml. This small sample size might lead to a lack of statistical power; however, we showed no difference in the frequency of virological failure during the year of follow-up in patients with LLV compared with patients with sustained HIV-1 RNA values,20 copies/ml. In addition, 44% of the patients with LLV during the inclusion period displayed all HIV-1 RNA values,20 copies/ml during the follow-up period and 16% still displayed LLV. A previous study, assessing the significance of LLV between 12 and 40 copies/ml (n¼69 patients), similarly showed no significant association with the development of virological failure when compared with patients with undetectable HIV-1 RNA values (i.e.,12 copies/ml). 12 However, patients with detectable LLV in this latter study were significantly less likely to achieve subsequent virological suppression. 12 Conversely, a recent study reported the association between a very low level of viral replication (.3 copies/ml) and the risk of virological failure. 13 One limitation of our study could be the duration of follow-up, which was 1 year. We can hypothesize that a longer period of follow-up might have led to evidence of a higher number of virological failures in the LLV+ group. In conclusion, persistent LLV was infrequent in our series and the 1 year follow-up did not evidence a higher rate of virological failure than in patients with all HIV-1 RNA values,20 copies/ml. LLV seems to be a transient and dynamic phenomenon, with 40% of patients shifting from LLV status to complete viral load suppression. For the moment, no selection of resistance mutations has been described in the literature at LLV,50 copies/ml. Indeed, selection of resistance mutations has been described only in case of persistent viraemia.50 copies/ml. 14,15 These results suggest that LLV, in patients with no viral blip.50 copies/ml, should not drive therapeutic modifications. However, clinical trials to assess the potential benefit of therapeutic interventions in patients exhibiting persistent LLV between 20 and 50 copies/ml are needed to establish the management of these patients. Acknowledgements Presented in part at the Eighteenth Conference on Retroviruses and Opportunistic Infections, Boston, MA, USA, 2011 (Abstract 577) and at the Nineteenth Conference on Retroviruses and Opportunistic Infections, Seattle, WA, USA, 2012 (Abstract 349). 2234

HIV-1 low-level viraemia JAC Funding This work was supported by the French National Agency for Research on AIDS and Viral Hepatitis (ANRS) and the European Community s Seventh Framework Programme (FP7/2007 2013) under the project Collaborative HIV and Anti-HIV Drug Resistance Network (CHAIN) (grant no. 223131). Transparency declarations None to declare. References 1 Thompson MA, Aberg JA, Cahn P et al. Antiretroviral treatment of adult HIV infection: 2010 recommendations of the International AIDS Society-USA panel. JAMA 2010; 304: 321 33. 2 Palmer S, Maldarelli F, Wiegand A et al. Low-level viremia persists for at least 7 years in patients on suppressive antiretroviral therapy. Proc Natl Acad Sci USA 2008; 105: 3879 84. 3 Hatano H, Delwart EL, Norris PJ et al. Evidence of persistent low-level viremia in long-term HAART-suppressed, HIV-infected individuals. AIDS 2010; 24: 2535 9. 4 Bonora S, Nicastri E, Calcagno A et al. Ultrasensitive assessment of residual HIV viraemia in HAART-treated patients with persistently undetectable plasma HIV-RNA: a cross-sectional evaluation. J Med Virol 2009; 81: 400 5. 5 Haïm-Boukobza S, Morand-Joubert L, Flandre P et al. Higher efficacy of nevirapine than efavirenz to achieve HIV-1 plasma viral load below 1 copy/ml. AIDS 2011; 25: 341 4. 6 Palmisano L, Giuliano M, Nicastri E et al. Residual viraemia in subjects with chronic HIV infection and viral load,50 copies/ml: the impact of highly active antiretroviral therapy. AIDS 2005; 19: 1843 7. 7 Johnson VA, Brun-Vézinet F, Clotet B et al. Update of the drug resistance mutations in HIV-1: December 2010. Top HIV Med 2010; 18: 156 63. 8 Pascual-Pareja JF, Martínez-Prats L, Luczkowiak J et al. Detection of HIV-1 at between 20 and 49 copies per milliliter by the Cobas TaqMan HIV-1 v2.0 assay is associated with higher pretherapy viral load and less time on antiretroviral therapy. J Clin Microbiol 2010; 48: 1911 2. 9 Richman DD, Margolis DM, Delaney M et al. The challenge of finding a cure for HIV infection. Science 2009; 323: 1304 7. 10 Montaner JS, Richman DD, Hammer SM. Poor agreement between 2 assays for measuring low levels of HIV-1 viral load. Clin Infect Dis 2009; 49: 1283 4. 11 Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. Department of Health and Human Services, 2011. http://www.aidsinfo.nih.gov/contentfiles/adultandadolescentgl.pdf (16 November 2011, date last accessed). 12 Widdrington J, Payne B, Medhi M et al. The significance of very low-level viraemia detected by sensitive viral load assays in HIV infected patients on HAART. J Infect 2011; 62: 87 92. 13 Maggiolo F, Callegaro A, Cologni G et al. Residual low-level HIV viremia in HAART-treated patients and risk of virological failure: a new standard of care? In: Abstracts of the Sixth IAS Conference on HIV Pathogenesis, Treatment and Prevention, Rome, Italy, 2011. Abstract TULBPE030. 14 Nettles RE, Kieffer TL, Kwon P et al. Intermittent HIV-1 viremia (Blips) and drug resistance in patients receiving HAART. JAMA 2005; 293: 817 29. 15 Taiwo B, Gallien S, Aga E et al. Antiretroviral drug resistance in HIV-1-infected patients experiencing persistent low-level viremia during first-line therapy. J Infect Dis 2011; 204: 515 20. 2235