Combination antiretroviral drug regimens have dramatically decreased morbidity and mortality in HIV-infected individuals [1,2]. Reductions in plasma viral load by antiretroviral therapy have been sustained over long periods of time in some patients [3,4]. This led to speculation that routine eradication of HIV may be achievable. Mathematical models based on the decline of plasma HIV RNA have been used to extrapolate the length of time necessary to suppress virus replication using antiretrovirals in order to achieve viral eradication . Complete viral clearance was initially estimated to require between 2.3 and 3.1 years, later extended to greater than 5 years  after the demonstration of the existence of latently HIV-infected cells [7-9].
There are numerous reports on the kinetics of viral load decreases upon starting therapy [e.g. 10-12] and on viral load increases during the development of drug resistance [e.g. 13,14]. However, there is little information available on the rate of increase of plasma HIV RNA after complete withdrawal from highly active antiretroviral therapy (HAART) following sustained viral suppression. Viral rebound occurred within 1 month in some patients after prolonged suppression of plasma HIV [15,16] but did not appear to occur in two patients who stopped therapy including hydroxyurea and didanosine . The study presented here investigates the initial kinetics of viral rebound in patients electing to withdraw from triple or quadruple combination therapies including at least one protease inhibitor.
The study involved six homosexual men infected with HIV-1 and having undetectable viral loads at multiple assessments who opted to withdraw from combination therapy. Patients elected to withdraw entirely from antiretroviral therapy (either temporarily or permanently) because of lipodystrophy, narcotic overdose, insomnia and/or high blood pressure.
Plasma viral load assay
Blood plasma samples were collected in Vacutainer CPT tubes (Becton Dickinson Canada Inc., Mississauga, Ontario, Canada) according to the manufacturer‚s instructions and were stored at -70∞C prior to testing. Plasma was collected at the time of stopping therapy and every 3-4 days for approximately 21 days. Plasma viral load assays were performed in triplicate for each sample (in order to increase the accuracy of the determinations without requiring patients to give blood more often) using the Roche Amplicor Monitor 1.5 assay (Roche Diagnostic Systems Inc., Branchburg, New Jersey, USA), adopting the Ultrasensitive protocol . Briefly, 0.5ml plasma was centrifuged for 1h at 24000¥g at 4∞C. The pellet was lysed with 600μl buffer containing guanidine thiocyanate, with subsequent steps being similar to the standard Amplicor assay. This protocol is able to quantify as little as 50 HIV RNA copies/ml plasma .
Linear regression analysis of the first four measurable data points was used to calculate the rate of viral rebound and to extrapolate to the amount of virus present at the time of withdrawing therapy. Circulating plasma virus in the whole body was estimated assuming 3000ml plasma per body.
HIV DNA and CD4 cell counts
In one patient, rebound in plasma HIV DNA was determined as described elsewhere . In another patient, CD4 and CD8 cell counts were determined over the time course of viral rebound by standard methods.
Patient characteristics and drug regimens are indicated in Table 1. Sequential plasma samples (three to six per patient) were obtained over 21 days from the six patients electing to stop HAART therapy. At the time of stopping therapy, all patients had plasma HIV RNA levels below 50copies/ml (Roche Ultrasensitive HIV RNA assay), and had sustained HIV RNA levels below 500copies/ml for a median of 390 days (range 39 to 542 days) (Table 1).
Plasma HIV RNA rebounded to detectable levels in all patients within days of stopping HAART therapy (Fig. 1). This rebound was very rapid (HIV doubling time ≊1.5 days) and approximated first-order kinetics before slowing as viral load approached pre-therapy levels after several weeks. The median rise in plasma HIV RNA corresponded to approximately a 2 log increase every 10 days and was similar for all six patients, with a range of 0.15 to 0.42 log/day (Table 2). Plasma HIV RNA levels returned to greater than 500copies/ml in a median of 10 days (range 6 to 15 days) (Table 2).
Linear regression was used to estimate the amount of virus still present in these patients with viral loads too low for detection at the time of withdrawing from therapy. Assuming 3000ml plasma per body, the median estimated residual total body plasma virus was 4.6log/body plasma (or approximately 10 HIV RNA copies/ml) with a range 0.7-5.3log (Table 2). In all but one patient, the lower limit of the 95% confidence interval was above zero. The extrapolation may not be reliable for this patient (number 4), as only three data points were available, resulting in an extremely wide confidence interval. This patient also died shortly after day 21. The baseline viral load of the other five patients correlated with the estimated amount of residual HIV (r2=0.88; P=0.02) and inversely with the time required to rebound to 500copies/ml (Pearson r2=0.85; P=0.03), suggesting that the net reduction in viral load achieved was fairly similar in all five patients. The estimated decrease in total body plasma HIV RNA during HAART therapy was 3.0 to 4.7log (excluding patient 4).
There was no obvious correlation between the rate of viral rebound and the baseline plasma viral load (r2=0.01), the baseline CD4 cell count (r2=0.5) or the duration of time patients had remained below 500 HIV RNA copies/ml (r2=0.01).
Total CD4 cell counts and helper:suppressor cell ratios were monitored for one individual (Patient 6) during the 21-day viral-rebound period and no apparent changes were observed during this time. In another individual (Patient 4), the change in peripheral blood HIV DNA in blood mononuclear cells was measured using a quantitative HIV DNA polymerase chain reaction (PCR) assay. HIV DNA levels increased approximately twofold in the 21 days following withdrawal from therapy (data not shown).
The rapid rates of viral rebound observed in these six patients clearly show that complete viral clearance had not occurred despite up to 18 months of viral suppression below 500copies/ml HIV RNA. As little as 1 week of withdrawal from HAART therapy resulted in plasma HIV RNA levels of greater than 500copies/ml, emphasizing the requirement for strict adherence to these drug regimens if HIV replication is to be minimized. This rapid viral rebound should be considered when interpreting sudden fluctuations in HIV RNA levels in clinical practice, as well as in clinical trials where the occurrence of ‚detectable‚ virus is an endpoint . Individuals should be very closely monitored between 7 and 10 days after stopping therapy to test for the possibility of HIV eradication.
The regression analysis estimates presented here suggest that thousands of copies of plasma HIV RNA may be present in the body at a time when even the most sensitive quantitative PCR assays yield undetectable results. Given the practical limits on sample size, only about 0.1% of the total body plasma can reasonably be analyzed; consequently, even the newest generation HIV RNA assays will not be sufficiently sensitive to demonstrate that the plasma (let alone the rest of the body) is free of HIV RNA. Other approaches for the routine monitoring of HIV infection in patients receiving HAART are, therefore, required.
There is surprisingly little information on the rebound of viral load upon stopping therapy after a sustained reduction in plasma viral load. Patients discontinuing 8-19 days of nucleoside analogue therapy  or longer periods of triple combination therapy  were reported to have a transient rise in plasma viral load above baseline. We previously reported a case of a patient with HIV RNA levels consistently below 50copies/ml for 28 months whose plasma viral load rebounded to levels far greater than that observed at baseline within 1 month of stopping therapy . Of interest, there is also one report of a complete absence of viral rebound in two patients for up to 1 year after withdrawing from combination therapy including didanosine and hydroxyurea .
The data presented here provide no evidence that the amount of residual plasma virus is decreasing with increasing time on therapy, as expected if viral eradication were to occur as a result of HAART. Latently infected or slowly cleared reservoirs of HIV [7-9] and/or residual HIV replication, particularly in tissue compartments other than plasma , could account for the lack of total viral clearance. It is feasible that current HAART simply is not sufficiently potent to inhibit viral replication completely and results in a new quasi-equilibrium of plasma viral load that is 3-5log lower than pre-therapy levels. This is consistent with evidence of viral evolution in both the plasma HIV RNA and DNA envelope sequences (though no evidence of evolution of drug resistance) in a patient over a 28-month period during which HAART had reduced plasma viral load below 50copies/ml  unpublished data].
The rates of viral load increase observed here are only slightly slower than the ≊0.5log/day reported during primary infection  or early interruption of triple combination therapy , and they are very similar to those observed for rebound of a resistant variant (V82A) selected during ritonavir monotherapy .
These results are limited by to the small number of patients analyzed. In addition, a limitation of the analyses is that the extrapolation to the moment of stopping therapy assumes a constant rate of viral increase and that this increase began at the time that patients stopped taking medication. Furthermore, patients choosing to withdraw from HAART therapy may not be representative of most patients. Nonetheless, these results suggest that many patients who choose to stop HAART after remaining below 500copies/ml HIV RNA for a considerable period of time will have increases in plasma viral HIV RNA of about 0.2log/day and reach easily detectable levels of HIV RNA within 1 or 2 weeks after stopping therapy.
We thank Cindy Christopherson, Shirley Kwok and John Sninsky at Roche Molecular Systems for performing the HIV DNA measurements.
1. Hogg SR, O‚Shaughnessy MV, Gataric N, et al
. Decline in deaths due to new antiretrovirals. Lancet
2. Pallela FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med
3. Gulick RM, Mellors JW, Havlir D, et al
. Simultaneous vs. sequential initiation of therapy with indinavir, zidovudine, and lamivudine for HIV-1 infection: 100-week follow-up. JAMA
4. Hammer SM, Squires KE, Hughes MD, et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. N Engl J Med
5. Perelson AS, Essunger P, Cao Y, et al. Decay characteristics of HIV-1 infected compartments during combination therapy.
Nature 1997, 387:
6. Ho D Toward HIV eradication or remission: the tasks ahead. Science
7. Wong JK, Hezareh M, Guthard HF, et al. Recovery of replication competent HIV despite prolonged suppression of plasma viremia. Science
8. Finzi D, Hermankova M, Pierson T, et al
. Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy. Science
9. Chun T-W, Stuyver L, Mizell SB, et al
. Presence of an inducible HIV-1 latent reservoir during highly active antiretroviral therapy. Proc Natl Acad Sci USA
10. Ho D, Neumann AU, Perelson AS, Chen W, et al
. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection
11. Wei X, Ghosh SK, Taylor ME, et al
. Viral dynamics in human immunodeficiency virus type 1 infection
12. Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho D HIV-1 dynamics in vivo: virion clearance rate, infected life-span, and viral generation time
. Science 1996, 271:
13. Schuurman R, Nijhuis M, van Leeuwen R, et al. Rapid changes in human immunodeficiency virus type 1 RNA load and appearance of drug-resistant virus populations in persons treated with lamivudine (3TC)
. J Infect Dis
14. Eastman PS, Mittler J, Kelso R, et al
. Genotypic changes in human immunodeficiency virus type 1 associated with loss of suppression of plasma viral RNA levels in subjects treated with ritonavir (Norvir) monotherapy
. J Virol
15. Montaner JS, Harris M, Mo T, Harrigan PR Rebound of plasma HIV viral load following prolonged suppression with combination therapy [letter]
16. Jubault V, Burgard M, Le Corfec E, et al. High rebound of plasma and cellular HIV load after discontinuation of triple combination therapy [letter]. AIDS
17. Vila J, Nugier F, Bargues G, et al. Absence of viral rebound after treatment of HIV infected patients with didanosine and hydracarbamide
18. Sun R, Ku J, Jayakar H, et al. Ultrasensitive reverse transcription-PCR assay for quantitation of human immunodeficiency virus type 1 RNA in plasma
. J Clin Microbiol
19. Kwok S, Sninsky J, Christopherson C, Krowka JF, Haynes S. Quantification of HIV-1 proviral DNA
. XII International Conference on AIDS
, Geneva, June 1998 [abstract 42172]
20. de Gruttola V, Hughes M, Gilbert P, Phillips A Trial design in the era of highly effective antiviral drug combinations for HIV infection. AIDS
21. de Jong MD, de Boer RJ, de Wolf F, et al. Overshoot of HIV-1 viraemia after early discontinuation of antiretroviral treatment
22. Lafeuillade A, Chollet L, Hittinger G, et al. Residual human immunodeficiency virus type 1 RNA in lymphoid tissue of patients with sustained plasma RNA of <200copies/ml
. J Infect Dis
23. Kaufmann GR, Cunningham P, Kelleher AD, et al
. Patterns of viral dynamics during primary human immunodeficiency virus type 1 infection. J Infect Dis
24. Neumann AU, Tubiana R, Calvez V, et al. Multi-phasic HIV decline following triple drug antiviral therapy is correlated with viral rebound dynamics during therapy interruption
. Fifth Conference on Retroviruses and Opportunistic Infections,
Chicago, February 1998 [abstract 517].