Objective: To assess the occurrence of viral load greater than 50 copies/ml in patients on highly active antiretroviral therapy (HAART) having achieved less than 50 copies/ml and the chance of whether a viral load greater than 50 copies/ml would lead to a sustained and increasing viral load.
Design: A cohort of 553 patients on HAART with viral loads of less than 50 copies/ml were followed.
Results: Over a median of 56 weeks 35% of patients experienced a transient increase and 8% virological failure (two consecutive viral loads of > 400 copies/ml). Transient increases and virological failure were more common in those with greater drug experience, and those with initial raised viral load values of more than 400 copies/ml were more likely to have a sustained increase and become virological failures.
Conclusion: Transient increases in viral load are common, mainly in the 50–400 copies/ml range, and the majority of subsequent viral load estimations show a return to less than 50 copies/ml. A single raised viral load should lead to adherence support and intensified monitoring. Subsequent treatment decisions can then be based on evidence of true virological rebound and failure.
From the Department of Primary Care and Population Sciences and Centre for HIV Medicine, Royal Free and University College School of Medicine, Hampstead, London NW3 2PF, UK.
Correspondence and requests for reprints to: Antonia Moore, Department of Primary Care and Population Sciences and Centre for HIV Medicine, Royal Free and University College School of Medicine, Hampstead, London NW3 2PF, UK. E-mail: email@example.com
*See Appendix for all the members of the Royal Free Centre for HIV Medicine.
Received: 20 July 2001;
revised: 5 November 2001; accepted: 7 November 2001.
The goal of highly active antiretroviral therapy (HAART) is to reduce the viral load to undetectable levels and increase the CD4 cell count, so reducing the rate of disease progression and death. Viral load assays have become increasingly sensitive and sophisticated and vary in their accuracy and lower limits of detection [1,2]. In clinical practice the aim is to achieve and maintain a plasma viral load of less than 50 copies/ml, usually within 6 months of the commencement of therapy [3–5]. After the commencement of HAART the viral load undergoes a two-stage decline; an initial, rapid fall and a second slower phase. After the achievement of less than 50 copies/ml patients have their levels reassessed on approximately a 12 weekly basis . A proportion of patients will experience an increase in viral load above this level after this time-point. The significance of this is often unclear, the increase may be a ‘true’ increase, the consequence of reduced serum levels of antiretroviral agents resulting from poor adherence, altered absorption or metabolism of the drug, multi-drug interactions, the emergence of resistant virus or the effect of intercurrent infection or vaccination [4,6,7]. Assay variability could also result in an ‘apparent’ increase . It is often difficult for clinicians to know how to respond to a single value over 50 copies/ml. Whereas all possibilities should be considered and the importance of adherence re-emphasized , the cause of the first increase in viral load may remain uncertain.
Using data from the Royal Free Hospital we aimed to: (i) assess the occurrence of viral load values of over 50 copies/ml in patients with viral loads of less than 50 copies/ml being seen in routine clinical practice; and to (ii) assess the chances of whether a ‘rise’ is real and will lead to a sustained increase in viral load.
Subjects were part of the Royal Free Hospital cohort. All were on a HAART regime (most commonly a three-drug combination including two nucleoside reverse transcriptase inhibitors and either a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor), and all had viral load levels of less than 50 copies/ml. No patient was counted more than once. Subjects were followed from their first viral load of less than 50 copies/ml (in some, this was when the ultrasensitive viral load assay first became available in our clinic) until their last viral load measurement or until they had experienced virological failure (two consecutive viral loads of > 400 copies/ml, the time of the first retrospectively being identified as the cut-off point). The plasma HIV-1-RNA load was quantified using a reverse transcriptase polymerase chain reaction (Roche Amplicor Monitor, version 1.5: cut-off 50 copies/ml, 1.7 log copies/ml; Roche Diagnostics Corporation, Illinois, USA). An external quality assurance control (International Molecular Services: 10 000 copies/ml) was included in every run and was required to be within two standard deviations of the mean for 138 contributing European laboratories for the results to be clinically valid and reported. In line with recent work demonstrating that the Roche Amplicor Monitor assay shows increasing assay variability with decreasing viral loads , the standard deviation at levels of viral load of less than 50 copies/ml was assumed to be 0.3 log. Statistical analysis was performed using the SAS statistical software package. Survival curves were generated to compare the rates of a single increase in viral load to over 50 copies/ml, and the rates of virological failure according to baseline drug history and differences were tested for statistical significance using the log rank test. All data available up to summer 2000 were included.
Five hundred and fifty-three patients satisfied the inclusion criteria. Of these 207 (37%) had a viral load of less than 50 copies/ml measured within 48 weeks of starting their first HAART regimen, 79 (14%) between 48 and 72 weeks, 97 (18%) between 72 and 96 weeks, 86 (16%) between 96 and 120 weeks, and 84 (15%) after more than 120 weeks of treatment. The majority of patients, 362 (66%), were taking three-drug combinations, although 45 (8%) were on more than five drugs. Protease inhibitors were included in the regimes of 403 (73%) and non-nucleoside reverse transcriptase inhibitors in 215 (39%). The median follow-up time was 56 weeks (range 4–174 weeks), the median number of viral load measurements was four (range one to 19), and the median interval between measurements was 14 weeks. During follow-up 283 (12.7%) of the 2223 viral load values recorded were over 50 copies/ml. A total of 192 patients (35%) experienced at least one value over 50 copies/ml during the period of study, whereas 42 (8%) experienced virological failure (two consecutive viral loads of > 400 copies/ml). Of the 192 patients with a value of over 50 copies/ml recorded, 40% had values of over 200 copies/ml.
Those with more viral load measurements during follow-up were not those with more values of over 50 copies/ml; in the 210 patients with more than five measurements 145 (10.6%) of them were over 50 copies/ml. Approximately 12% of all viral load values were over 50 copies/ml in the first 6 months of follow-up after the original attainment of less than 50 copies/ml, and this proportion remained remarkably consistent over each 6 month period in the remainder of the study, see Table 1. The initial raised viral load value was less than 100 copies/ml in 39%, between 100 and 199 in 21%, between 200 and 299 in 7% and 300–399 in 5%. Twenty-nine per cent of first raised values were greater than 400 copies/ml, 45% of these being greater than 5000.
Those who had ever used more than four drugs were significantly more likely to experience both a transient increase in viral load and virological failure than those who had used four or fewer antiretroviral agents (P < 0.0001 and P = 0.004, respectively), see Fig. 1. Excluding those who experienced virological failure, the study endpoint thus becoming ‘transient rise', a similar relationship between drug experience and the risk of (benign) rebound was found.
During follow-up 131 (24%) of all patients added at least one drug to their antiretroviral regimen, including 71 (20%) of those who never had a viral load of more than 50 copies/ml. Among those failing therapy, 43% had new drugs initiated during follow-up, 19% adding three agents, whereas in the 150 patients with a transient increase but not experiencing virological failure 28% started a new drug but only 5% started three new agents. Examining the 154 patients who had had a single value of over 50 copies/ml and had not altered their therapy, the subsequent viral load returned to less than 50 copies/ml in 54% but was over 50 copies/ml in 46%, the median time between the original raised viral load and follow-up measurement being 8 weeks. Of the 42 patients with an initial raised value of over 400 copies/ml, 22 (51%) became virological failures (i.e. had a second, consecutive value of > 400 copies/ml), whereas among the 112 patients with an initial increase of less than 400 copies/ml only 8% went on to become virological failures. Twenty-two patients who had not changed therapy had a repeat viral load measurement after two consecutive values of 50–400 copies/ml; among them 68% returning to less than 50 copies/ml, 23% remaining between 50 and 400 copies/ml and 9% increasing to over 400 copies/ml.
Our results, in common with other studies [10,11], show that among those patients who have achieved viral load values of less than 50 copies/ml, subsequent values greater than this are common, mainly in the 50–400 copies/ml range, and that the majority of follow-up viral load estimations show a return to less than 50 copies/ml (i.e. transpire to be transient). However, in our cohort it appears that those with an initial raised value of over 400 copies/ml are more likely to have a sustained increase and become virological failures, whereas those with a first raised value of between 50 and 400 copies/ml have a follow-up of less than 50 copies/ml in the majority of cases and over 400 copies/ml in under 10% of cases. Both ‘blips’ and virological failure were found to be more common in those patients with greater drug experience.
The explanation behind the increase in viral load is often established only in retrospect, but a single raised value may be the result of intercurrent infection or vaccination, which may be associated with a transient increase in viral load [6,12]. Other transient increases may be the result of the emergence of drug resistance, but as shown in the study of Cohen Stuart et al. , in the short term, this does not necessarily translate into virological failure. Many such apparent increases could also be caused by assay variability. Assays used to estimate viral load may vary in accuracy, the variability increasing at lower viral loads, and this may lead to an artificial increase [9,14]. However, our results showed that 40% of values over 50 copies/ml were greater than 200 copies/ml (i.e. more than two standard deviations higher), and were therefore unlikely to be the result of assay variability and more likely to be ‘real’ increases.
As we and others have shown, two consecutive values of 50–400 copies/ml are likely to be followed by a value of less than 50 copies/ml in approximately half the cases , and by a value of over 400 copies/ml in less than 10% of cases.
In the context of any apparent increase in viral load, the importance of adherence must be reiterated , and the patient's drug combination must be assessed to evaluate the possibility of drug interactions leading to subtherapeutic levels of antiretroviral drugs and the possible emergence of resistant virus. As our results have shown, the majority of first viral load values over 50 copies/ml do not signal treatment failure, and are more likely to turn out to be ‘transient’ increases. This finding may not be the result of a spontaneous reduction to less than 50 copies/ml, and is possibly the result of improved patient adherence when faced with the potential threat of treatment failure. A single increased viral load measurement should lead to adherence support and intensified monitoring of the patient in order that subsequent treatment decisions can be based on evidence of true virological rebound and failure.
1. Ramratnam B, Mittler JE, Zhang L. et al
. The decay of the latent reservoir of replication-competent HIV-1 is inversley correlated with the extent of residual replication during prolonged antiretroviral therapy. Nat Med 2000, 6: 82–85.
2. Yerly S, Perneger TV, Vora S, Hirschel B, Permi L. Decay of cell associated HIV-1 DNA correlates with residual replication in patients treated during acute HIV-1 infection. AIDS 2000, 14: 2805–2812.
4. US Health and Human Services. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents.
] accessed October 2001. US Health and Human Services; April 2001.
5. Perelson AS, Essunger P, Cao Y. et al
. Decay characteristics of HIV-1 infected compartments during combination therapy. Nature 1997, 387: 188–191.
6. Gunthard HF, Wong JK, Spina CA. et al
. Effect of influenza vaccination on viral replication and immune response in persons infected with human immunodeficiency virus receiving potent antiretroviral therapy. J Infect Dis 2000, 181: 522–531.
7. Paterson DL, Swindells S, Mohr Swindells S, Mohr J. et al
. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med 2000, 133: 21–30.
8. Brambilla D, Reichelderfer PS, Bremer JW. et al
. The contribution of assay variation and biological variation to the total variability of plasma HIV-1 RNA measurements. The Women Infant Transmission Study Clinics. Virology Quality Assurance Program.
AIDS 1999, 13: 2269–2279.
9. Brambilla D, Granger S, Bremner J. Variation in HIV RNA assays at low RNA concentration.
In:7th Annual Conference on Retroviruses and Opportunistic Infections.
San Francisco, 30 January–2 February 2000 [Abstract 774].
10. Greub C, Cozzi-Lepri A, Staszewski S, et al
. Swiss and Frankfurt Cohorts. Low level HIV viral rebound and blips in patients receiving potent antiretroviral therapy.
In:8th Annual Conference on Retroviruses
. Chicago, 4–8 February 2000 [Abstract 522].
11. Havlir D, Bassett R, Levitan D. et al
. Prevalence and predictive value of intermittent viraemia in patients with previously undetectable HIV-1 RNA levels. JAMA 2001, 286: 171–179.
12. King JC Jr, Treanor J, Fast PE. et al
. Comparison of the safety, vaccine virus shedding, and immunogenicity of influenza virus vaccine, trivalent, types A and B, live cold-adapted, administered to human immunodeficiency virus (HIV)-infected and non-HI V-infected adults. J Infect Dis 2000, 81: 725–728.
13. Cohen Stuart JWT, Wensing AMJ, Kovacs C. et al
. Transient relapses (`blips') of plasma HIV RNA levels during HAART are associated with drug resistance. J Acquir Immune Defic Syndr 2001, 28: 105–113.
14. Sun R, Ku J, Jaykar H. et al
. Ultrasensitive reverse transcriptase PCR assay for quantification of human immunodeficiency virus type I RNA in plasma. J Clin Microbiol 1998, 36: 2964–2969.
The Royal Free Centre for HIV Medicine also includes: Dr Caroline A. Sabin, Dr Amanda Mocroft (Biostatistics/Epidemiology), Dr Simon Barry, Tony Drinkwater, Zuber Mitchla, Dr Jane Zuckerman (Clinical and Clinical Trials), Professor Christine Lee, Dr Thynn Thynn Yee (Haemophilia), Professor George Janossy, Dr Sabine Kinloch-de Loes, Dr Melvyn Kahan, Sandra Martins, Arabjan lqbal, Richard Tilling (Immunology), Professor Anne Johnson, Dr Andrew Hayward, Dr Stephen Conaty (Infectious Disease Epidemiology), Professor Ali Zumla, Professor John Stanford, Professor Graham Rook, Professor Chifumbe Chintu, Professor John Grange, Dr Gordon Cook (Infectious Diseases), Dr Stephen Gillespie (Medical Microbiology), Dr Jonathan Elford, Dr Lorraine Sherr, Graham Bolding, (Prevention and Behaviour Change), Louise Dann, Joanne Page, Fionna van Hooff, Emily Vrettou (Retrovirology), Professor Paul Griffiths, Professor Vince Emery, Dr Jane Deayton (Virology), Elaine Harris, Theresa Luxford (Administration). Cited Here...