There is no dispute that the major goal of HIV treatment should be to maintain good health and quality of life, but there is considerable debate amongst doctors, researchers and people with HIV about the best way of achieving this goal.

The following sections look at the many issues involved in choosing an HIV treatment strategy, starting with the link between suppressing viral load and improving prognosis.

The evidence of a link between low viral load and a good prognosis among untreated people, and between a reduction of viral load and an improved prognosis in those taking treatments is discussed in Viral load, CD4 cell counts and other tests.

The case for hitting hard

Advocates of 'hitting hard' propose that this approach is the only logical one, on the basis of recent scientific studies. They make three main arguments:

• When treatment is started, you should always choose a combination that is likely to suppress viral load to below the limit of detection.

• Dual combinations consisting of two nucleoside reverse transcriptase inhibitors (NRTIs) are not strong enough to achieve this and the likelihood that you will subsequently become resistant to those drugs will limit your future choices. Although a minority of people do achieve undetectable viral load on two NRTIs, this result is rarely sustained. This is why triple combination therapy is the minimum standard of care.

• Viral load ideally needs to be suppressed below the limit of the viral load test, which is 50 copies/ml. When viral load is only partially suppressed, suppression will not be sustained for long, because resistance to the drugs will emerge. Studies suggest that we see the greatest durability of suppression if viral load is suppressed to below 50 copies/ml. Some experts believe that no viral evolution occurs when HIV is suppressed to very low levels, making the development of drug resistance unlikely (Blankson 2002; Kieffer 2004). However, resistance has been shown to emerge in people with viral load in the range of 50 to 400 copies/ml. See Viral load blips in Anti-HIV therapy: Changing treatment for further information.

We do not know how long it may take for resistance to develop among people with viral load between 50 and 400 copies/ml. In some cases it might take many months. One study found that 52 of 168 people with viral load between 50 and 400 copies/ml had viral rebound above 400 copies/ml during an average of 25 weeks follow-up (Pilcher 1999). For some people, particularly those who have few remaining treatment options, just getting below 400 copies/ml may be good enough to buy the additional time needed for new treatments to come along, in the opinion of some doctors.

Despite these concerns about the development of resistance, a recent analysis of over 3000 patients in the United States found that patients with viral loads between 400 and 20,000 copies/ml for at least six months, but could not achieve lower viral loads, had similar rates of death and development of AIDS-defining illnesses to those with viral loads below 400 copies/ml (D'Aquila 2004). While this may be encouraging for patients unable to fully suppress HIV in the blood, the study did not examine the effect of viral load measurements below 50 copies/ml, the limit of detection with most current viral load tests.

Viral nadir

There have been recent suggestions that the rate at which viral load falls during the first few weeks of treatment influences the nadir (lowest viral load attained). This in turn predicts how long a treatment will work. According to this theory, the harder you hit, the faster viral load falls, and the longer the treatments work. While this remains a theoretical position, research shows that a four-drug combination including a non-nucleoside reverse transcriptase inhibitor (NNRTI), a protease inhibitor and two NRTIs causes a more rapid decline in viral load than triple combination therapy. Viral decay is particularly rapid in the first three or four days of treatment (Polis 1998).

However, an analysis of data from the INCAS and AVANTI studies showed that adherence was more important in predicting sustained suppression of viral load than viral nadir (Raboud 2002).

Implications of aiming for below 50 copies/ml

According to the 2003 British HIV Association (BHIVA) guidelines on antiretroviral treatment, the aim of anti-HIV therapy should be to suppress viral load to below 50 copies/ml within 24 weeks of starting treatment, particularly when taking treatment for the first time. BHIVA also advise that viral load should be monitored a month or so after starting treatment, and again at 12 weeks after starting. Alterations to a regimen may be made on the basis of these early results.

For example, if the 12-week viral load test result is in the 50 to 400 copies/ml range, a patient may face a dilemma. On the one hand, it might be unwise to change therapy during this period solely because viral load remains above 50 copies/ml, since given a little longer it might fall to below this value without using up any more of the available drugs. On the other hand, by doing nothing and waiting to see what happens, the patient is running the risk that resistance may start to develop, which would also limit long-term treatment options. This is why advocates of 'hitting hard' argue that the most cautious response is to monitor viral load closely and regularly, with rapid access to the test results, and to modify treatment as soon as there is any sign of an increase.

Different doctors are likely to offer different advice about this dilemma. Some will suggest that people whose viral load is between 50 and 400 copies/ml should wait a few weeks and have a second test done, to see whether it just needs a little longer to get below 50 copies/ml. The chance that resistance will develop in the meantime is very low. But if it remains above 50 copies/ml, there is a case for promptly adding an extra drug, to try to drive it that little bit further down. When the viral load is so low, you can probably get away with just adding a single extra drug. If you wait longer and the viral load rises into the hundreds or thousands, you may need to change all your drugs, they argue. See Reasons to change in Anti-HIV therapy: Changing treatment for further discussion of intensifying therapy.

However, with this strategy, you may end up changing treatment (and using up drugs) unnecessarily. Many doctors have observed patients whose viral load increased from 'undetectable' into the few hundreds or thousands, but after a couple of months but which returns to 'undetectable' without any change in treatment. These increases may have been stimulated by an infection or vaccination, or may be due to statistical fluctuations in the viral load test (Nettles 2005). For this reason, other doctors would not recommend changing treatment unless repeated tests over several months indicated that the rise in viral load was real and persistent. At that point, most would be worried that adding or changing only one drug might lead to resistance, and instead they would switch at least two components of the combination. See Viral load blips in Anti-HIV therapy: Changing treatment for more information.

How achievable is 'below 50 copies/ml'?

For people who have never taken anti-HIV drugs before, there is now a reasonable chance of suppressing HIV below 50 copies/ml with their first regimen. Studies show that between 60 and 70% of people starting combination treatment achieve a viral load below 50 copies/ml at six months. People who have previously taken anti-HIV drugs have less chance of achieving a viral load below 50 copies/ml in the medium term, and people with high viral load (above 100,000 copies/ml) may take longer.

How long does it take to go below 50 copies/ml?

Another practical problem is that it is unclear how quickly someone can reasonably expect their viral load to fall below 50 copies/ml.

We know that with the standard viral load tests, most people who are going to go below 400 copies/ml will have done so by the time they have received 16 to 24 weeks of triple therapy. That makes it easy to say that if viral load remains above 400 copies/ml after 16 to 24 weeks, a patient should consider changing therapy.

Throughout the first year of triple therapy with AZT (zidovudine, Retrovir), 3TC (lamivudine, Epivir) and indinavir (Crixivan) in the Merck 035 trial, participants' viral load fluctuated around the limit of detection of the sensitive test. For instance, an individual would have below 50 copies/ml one month, but the next month it would be detectable at around 60 to 70 copies/ml, before dropping back below the limit of detection after another month (Gulick 1998).

However, after around a year of treatment, most people do turn out to have viral load below 50 copies/ml when tested with the sensitive test. Only a very small proportion fall into the 50 to 400 copies/ml zone.

A review of the DMP-006 study found that people who started treatment with viral load above 300,000 copies/ml were less likely to have viral load below 50 copies/ml by week 24, and took up to 36 weeks to achieve viral load below 50 copies (Moyle 2000).

More recently, a review of the M98-863 study, which compared ritonavir-boosted lopinavir (Kaletra) with nelfinavir (Viracept) in treatment-naive patients found that a substantial proportion of patients took more than 24 weeks to reach viral below 50 copies/ml, and that baseline viral load and the choice of regimen was significantly associated with the time taken to achieve viral load below 50 copies/ml. Those who achieved viral load below 50 copies/ml by week 24 had an average baseline viral load of 77,000 copies/ml, while those who achieved viral load below 50 copies/ml after week 24 had an average viral load of 177,000 copies/ml (Walmsley 2002).

References

Blankson JN et al. The challenge of viral reservoirs in HIV-1 infection. Ann Rev Med 53: 557-593, 2002.

D'Aquila R et al. Effect of persistent moderate viremia on disease progression during HIV therapy. J Acquir Immune Defic Syndr 37: 1147-1154, 2004.

Gulick RM et al. Simultaneous vs sequential initiation of therapy with indinavir, zidovudine, and lamivudine for HIV-1 infection: 100-week follow-up. JAMA 280: 35-41, 1998.

Kieffer T et al. Genotypic analysis of HIV-1 drug resistance at the limits of detection: virus production without evolution in treated adults with <50 copies/ml of plasma HIV-1 RNA. Eleventh Conference on Retroviruses and Opportunistic Infections, San Francisco, abstract 650, 2004.

Moyle G et al. Time to treatment success: 24 weeks is not enough in patients starting with high viral load in DMP-006. 40th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, abstract 547, 2000.

Nettles RE et al. Intermittent HIV-1 viremia (blips) and drug resistance in patients receiving HAART. JAMA 293: 817-829, 2005.

Pilcher CD et al. Detectable HIV-1 RNA at levels below quantifiable limits by Amplicor HIV-1 monitor is associated with virological relapse on antiretroviral therapy. AIDS 13: 1337-1342, 1999.

Polis MA et al. Increased initial rate of HIV-1 elimination with a four-drug antiretroviral regimen in treatment-naive patients. 38th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego, abstract LB-3, 1998.

Raboud JM et al. Impact of adherence on duration of virological suppression among patients receiving combination antiretroviral therapy. HIV Med 3: 118-124, 2002.

Walmsley S et al. Lopinavir-ritonavir versus nelfinavir for the initial treatment of HIV infection. N Engl J Med 346: 2039-2046, 2002.