Skip navigation

HIV-1 virus: finding it and stopping it

Bandolier is amazed at the rapid changes taking place in HIV monitoring and treatment in recent months. Developments - a test and a treatment - have come together to revolutionise things, with much agonising for purchasers and providers as the speed of the science overwhelms ponderous bureaucracy. This brief review covers only the main points; a fuller version may appear on Bandolier's Internet pages.

Background to HIV infection

When HIV viruses enter cells (lymphocyte or macrophage), viral RNA undergoes reverse transcription to produce double-stranded viral DNA which is integrated into the host DNA. Transcription and translation by cellular enzymes produces large, non-functional, polypeptide chains called polyproteins which are assembled and packaged at the cell surface to produce immature virions to be released.

These immature virions have their polyproteins cleaved into smaller, functional, proteins by HIV proteases, allowing the virions to mature into new active viruses. In 1988 the HIV-1 protease was crystallised and its three-dimensional structure determined. Computer models found chemicals to fit the cleavage site and inhibit protease activity. These inhibitors have been tested in man, and have exciting efficacy in reducing the number of HIV viruses in the body.

Clinical course

The average time between infection and development of AIDS is about 11 years, but about 20% progress rapidly to AIDS within five years. Another 12% of infected individuals remain free of AIDS for up to 20 years. Viral replication in lymphocytes, some 100 million or so virus particles a day, is associated with the defeat of the immune system. It kill the cells, which is why high levels of virus in blood is associated with low CD4 T-cell levels. Falling CD4 counts herald advanced immunosuppression and AIDS. This defeat of the immune system is associated with the development of conditions like cytomegalovirus infection of the retina, Pneumocystis carinii infection of the lungs, or tumours like Karposi's sarcoma or non-Hodgkin's lymphoma.

Knowing levels of both viral load and CD4 cells is useful in the management of HIV infected patients. The interaction between CD4 count and viral load in the blood has been likened to an impending train crash, where the viral load indicates the speed of the train and the CD4 cell count the distance to the crash site.

HIV viral load

Various forms of viral nucleic acid can act as markers for disease progression or response to antiretroviral therapy. Tests have detection limits as low as 500 molecules of viral RNA per mL, but can measure levels above 1,000,000 molecules/mL, covering the range of concentrations of viral RNA seen in patients with HIV and AIDS.

Viral load, CD4, and prognosis

Two studies have demonstrated that viral load is an excellent predictor of progression to AIDS [1,2].

The first [1] studied all 209 HIV-1 seropositive men enrolled in a Pittsburgh clinic in 1984 and 1985. Clinical status, CD4 T-cell count and blood samples for laboratory studies were obtained at baseline and every six months for up to 11 years. Results demonstrated a clear relationship between viral load at entry and progression to AIDS and death. Those in the lowest quartile of viral load had only a small chance of progressing to AIDS, and the median time to progression or death (>10 years) was limited by the duration of the study. Those in the highest quartile had a high chance of developing AIDS (62% within 5 years) and half died within 5 years.
Prognosis and viral load in 209 men
Viral load quartiles (molecules/mL) Progression to AIDS (%) by 5 years Median time to development of AIDS (years) Proportion who died within 5 years (%) Median survival time (years)
<4,500 8 10 5 10
4,500 - 13,000 26 7.7 10 9.5
13,000 - 36,300 49 5.3 25 7.4
36,300 62 3.5 49 5.1

A larger study in 1604 men infected with HIV-1 concurred [2]. A wide range of markers were compared for their ability to predict progression to AIDS and death over ten years. Five risk categories were defined by plasma viral load - from under 500 molecules/mL to more than 30,000 molecules/mL.

Prognosis and viral load in 1604 men
AIDS-free survival (%)
CD4 count (cells/µL) Viral load category N 3 years 6 years 9 years
500 I 110 94 82 70
II 180 96 81 56
III 237 92 70 42
IV 202 84 50 25
V 141 66 28 14
351-500 II 47 96 74 40
III 105 90 57 31
IV 121 83 40 16
V 121 50 20 5
201-350 II 27 93 74 59
III 44 91 52 27
IV 53 62 26 11
V 104 34 9 6
<200 IV 20 50 25 10
V 70 14 1 0
30,000 molecules/mL

Those in the highest viral load categories progressed most rapidly. But this larger study also demonstrated the effect of plasma viral load on the reduction in CD4 lymphocyte count over time. The higher the HIV-1 RNA concentration, the greater the rate of decline in the CD4 lymphocyte count.

Viral load and disease progression in infants

Viral load is prognostic of disease progression in infants [3]. In 106 infants infected with HIV-1 at birth, plasma samples were obtained up to 24 months. Plasma viral load increased rapidly after birth before falling. Infants with a rapid progression of disease had a higher peak viral load in the first two months of life than those without rapid progression.

Viral load and mother-to-baby transmission

Evidence on mother-to-infant transmission depending on maternal viral load continues to accumulate [4]. Higher maternal viral load gave more infected babies. The study had information on other factors that may affect transmission, like use of abused drugs, unprotected vaginal intercourse during pregnancy, and duration of ruptured membranes, all of which were positively associated with transmission of infection to the infant. Caesarean section may protect from mother-to-infant transmission.

Viral load predicts therapeutic response

Two papers have shown the usefulness of viral load measurement and CD4 counts in predicting success and failure with antiretroviral therapy [5,6]. In randomised trials a reduction in plasma viral load about eight weeks after starting treatment reduced the risk of disease progression by about half. But return to baseline viral load within six months was associated with progression to AIDS [6].

Viral load models

Mathematical modelling of viral load and disease progression is possible [7]. In the absence of antiretroviral treatment, patients with a viral load of 100,000 molecules/mL are at risk of progression to AIDS in fewer than three years. Those with a viral load of about 300,000 molecules/mL are at risk in less than one year. But with lower viral load, the time to progressing to AIDS extends, so that at 10,000 molecules/mL patients have at least 2.8 years and up to 19 years.

HIV-1 protease inhibitors

A systematic review of HIV-1 protease inhibitors up to September 1996 has been published [8], as well as a review which puts protease inhibitors into perspective in HIV [9]. There are some important practical points about the use of protease inhibitors, of which there are four - saquinavir, ritonavir, indinavir and nelfinavir (available on a named-patient basis in the UK):

  • They may have limited oral bioavailability; saquinavir is only about 4% available, but others, like indinavir, are up to 60% available orally. Patient acceptability of ritonavir may be poor because of the need to keep it cool.
  • Drugs that induce cytochrome P450 activity may reduce the availability of some protease inhibitors by increasing first pass metabolism. Drugs that inhibit P450 may increase oral availability.
  • There will be complex drug interactions in HIV or AIDS patients on a variety of different drugs, including antiretrovirals.
  • Protease inhibitors are associated with a number of adverse effects, including gastrointestinal disturbances and rashes.
  • Other adverse effects are more esoteric. For instance, indinavir precipitates in the renal collection system leading to obstruction and symptoms of renal colic, so high fluid intake is recommended.

Efficacy of protease inhibitors

Studies have yet to appear as full peer-reviewed papers to allow data abstraction. The ACTG 320 trial with indinavir has attracted much media attention because it was stopped early because of good results. Preliminary outcome results are available from the Internet [10].

ACTG 320 trial

This enrolled 1,156 HIV infected people with fewer than 200 CD4/µL, with over three months experience of AZT and less than seven days experience of 3TC (another antiretroviral drug), and no protease inhibitor experience. Patients were randomly assigned to dual therapy with nucleosides or triple therapy with additional indinavir for about 38 weeks. Baseline stratification was CD4 count of above or below 50/µL.

The rate of progression to AIDS or death was reduced by half in a mean of just 38 weeks of treatment. The number-needed-to-treat (NNT) was 11 for first clinical event for all patients, and as low as 6 for those with very low CD4 T-cell counts of fewer than 50/µL.
Main outcomes of ACTG 320 trial
Percent with event on
Outcome Dual therapy Triple therapy Hazard ratio NNT
All patients
First clinical event (AIDS or death) 18 9 0.50 (0.33-0.76) 11
Death 5 2 0.43 (0.19-0.99) 33
CD4 T-cell <50µ/L
First clinical event (AIDS or death) 34 16 0.49 (0.30-0.82) 6
Death 9 3 0.37 (0.13-1.04) 17
First clinical event (AIDS or death) 9 4 0.51 (0.24-1.10) 20
Death 2 1 0.59 (0.14-2.5) 100
NNT was calculated by: 100/(Event rate dual therapy-Event rate triple therapy)

Treatment Guidelines

A consensus statement by the British HIV Association has outlined some broad principles for HIV-treating physicians [11], along with a Drug and Therapeutics Bulletin [12], and recommendations about treatment from the US panel of the International AIDS Society [13]. These were based on the current available evidence, and are suggesting earlier and more aggressive treatment of HIV infection.

More good news

Three new reports [14-16] bring more good news. The issue is one of reservoirs of HIV virus, where latent infection of CD4 cells may occur despite apparently low viral loads in blood, or lack of symptoms.

As few as 5 and 7 cells per million may be infected in lymph nodes and blood respectively. The mean frequency of macrophages containing integrated HIV-1 DNA was 54 cells per million. So what happens when triple therapy apparently clears the virus from the blood? In eight patients starting treatment, HIV-1 in plasma dropped by more than 99% in the first two weeks due to rapid elimination of the free virus and loss of productively infected cells. About 2.3 to 3.1 years of treatment with a completely inhibitory regimen would be needed to completely eliminate HIV-1 from these longer-lived compartments.

Serial tonsil biopsies from ten patients treated with triple therapy showed that the amount of HIV-1 virus dropped rapidly. After 24 weeks more than 99.9% of virus had been cleared from the lymphoid tissue reservoir.


All of this is exciting stuff. If the fight against HIV and AIDS has seemed in the past to be like trench warfare, with every yard being hard fought, the new combination of effective tests for viral load and the advent of protease inhibitors seems like a cavalry charge. Ground is being gained, but the battle, let alone the war, is far from over.

Many challenges remain. Perhaps the most important is drug resistance in viruses, and the need to ensure compliance with therapy. But adverse effects, and the complex pharmacokinetic and pharmacodynamic interactions of these complicated therapies will challenge the clinical skills of those treating individuals with HIV and AIDS.

The other is cost. In the UK a patient going from no treatment to triple therapy would increase costs of drugs and tests by about [sterling]10,000 a year. Will the savings offset the costs? We have to wait for more information. But there are signposts that can help us look ahead. For instance, the balance of costs to society of a young man or woman in employment paying taxes becoming ill and unemployed, not paying taxes, and consuming healthcare will likely favour keeping them healthy. Treating some AIDS-related disorders, like cytomegalus infection of the retina, is hugely expensive.


  1. JW Mellors, CR Rinaldo, P Gupta, et al. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 1996 272: 1167-70.
  2. JW Mellors, A Munoz, JV Gorgini, et al. Plasma viral load and CD4 lymphocytes as prognostic markers of HIV-1 infection. Annals of Internal Medicine 1997 126: 946-54.
  3. WT Shearer, TC Quinn, P LaRussa et al. Viral load and disease progression in infants infected with human immunodeficiency virus type 1. New England Journal of Medicine 1997 336: 1337-42.
  4. DN Burns, S Landesman, DJ Wright et al. Influence of other maternal variables on the relationship between maternal virus load and mother-to-infant transmission of human immunodeficiency virus type 1. Journal of Infectious Diseases 1997 175: 1206-10.
  5. MD Hughes, VA Johnson, MS Hirsch et al. Monitoring plasma HIV-1 RNA levels in addition to CD4 lymphocyte count improves assessment of antiretroviral therapeutic response. Annals of Internal Medicine 1997 126: 929-38. Annals of Internal Medicine 1997 126: 929-38.
  6. WA O'Brien, PM Hartigan, ES Daar et al. Changes in plasma HIV RNA levels and CD4 lymphocyte counts predict both response to antiretroviral therapy and therapeutic failure. Annals of Internal Medicine 1997 126: 939-45.
  7. JPA Ionnidis, JC Cappelleri, J Lau, HS Sacks, PR Skolnik. Predictive value of viral load measurements in asymptomatic untreated HIV-1 infection: a mathematical model. AIDS 1996 10: 255-62.
  8. SG Deeks, M Smith, M Holodniy, JO Kahn. HIV-1 protease inhibitors: a review for clinicians. Journal of the American Medical Association 1997 277: 145-53.
  9. MB Feinberg. Changing the natural history of HIV disease. Lancet 1996 348: 239-46.
  10. M Harrington. It's time to change the standard of care for people with AIDS.
  11. BHIVA Guidelines Coordinating Committee. British HIV Association guidelines for antiretrovial treatment of HIV seropositive individual. Lancet 1997 349: 1086-92.
  12. Drug and Therapeutics Bulletin. Major advances in the treatment of HIV-1 infections. 35: April 1997.
  13. CCJ Carpenter, MA Fischi, CM Hammer et al. Antiretroviral therapy for HIV infection in 1997. Journal of the American Medical Association 1997 277: 1962-9.
  14. T-W Chun, L Carruth, D Finzi et al. Quantification of latent tissue reservoirs and total body viral load in HIV-1 infection. Nature 1997 387: 183-8.
  15. AS Perelson, P Essunger, Y Cao et al. Decay characteristics of HIV-1-infected compartments during combination therapy. Nature 1997 387: 188-91.
  16. W Cavert, DW Notermans, K Staskus et al. Kinetics of response in lymphoid tissues to antiretroviral therapy of HIV-1 infections. Science 1997 276: 960-4.

previous or next story in this issue