When to switch

Current good practice in antiretroviral treatment is to switch to a new regimen when viral load rises above 400-500 copies/ml, in order to minimise the development of resistance. In regimens containing 3TC (lamivudine) and NNRTIs (efavirenz or nevirapine) high-level resistance to these agents is likely to emerge more quickly because only one mutation is required for the virus to develop resistance to the agent. In individuals receiving protease inhibitor treatment as a second or third line regimen, there is evidence that clinical benefit persists despite viral load rebound, in some cases for up to four years, and is associated with a very strong initial viral load reduction when commencing protease inhibitor therapy (see Immune recovery in Anti-HIV therapy: Choosing your treatment strategy for further discussion of the evidence).

In resource-limited settings where viral load testing is not available, WHO recommends that the triggers for switching should be:

• Intolerance of one or more drugs in the combination

• Decline in CD4 cell count of more than 30% from peak, or decline in total lymphocyte count, indicating that the drug combination is no longer controlling viral replication

• New clinical symptoms of immune deficiency (likely to be the first available trigger for changing treatment in the absence of laboratory tests such as CD4 cell counting)

• New AIDS-defining illness in people with prior AIDS diagnosis

Choice of second-line treatment

The failure of a first-line regimen which contains an NNRTI (nevirapine or efavirenz) means that cross-resistance to the other agent in that class is likely to be present, so it will not be possible to use an NNRTI in the second regimen.

In these circumstances it is unlikely that Trizivir (AZT/3TC/abacavir) will have a substantial effect, because most first line regimens that use NNRTIs will also use 3TC and AZT, leading to multiple NRTI mutations known to compromise the response to abacavir. This is why Trizivir is recommended by WHO as a first-line regimen.

The failure of Trizivir may allow the use of d4T (stavudine), ddI and an NNRTI, although it is possible that some nucleoside analogue mutations could have been selected by prior AZT and abacavir treatment. The use of d4T and ddI together increases the risk of peripheral neuropathy compared with the use of either agent alone.

Second-line combinations are likely to be more expensive, since fixed dose co-formulations are likely to have formed the first line regimen in many cases.

AZT and ddI may be a suitable nucleoside analogue backbone after the failure of d4T/3TC in first-line treatment, together with either nelfinavir or a ritonavir-boosted protease inhibitor (either saquinavir, indinavir or lopinavir). However, the use of ddI with nelfinavir is problematic due to restrictions on food intake. Didanosine must be taken on an empty stomach, while nelfinavir must be taken with food. A ritonavir-boosted protease inhibitor is better tolerated if taken with food, but can be taken on an empty stomach with the exception of indinavir, which must not be dosed at the same time as ddI if the ddI formulation used is the ddI tablet rather than the ddI capsule. This is because the antacid buffer in the ddI tablet will interfere with the initial absorption of indinavir as it dissolves in the stomach, prior to the boosting effect of ritonavir, which takes place in the wall of the intestine (check this) and the liver.

After the failure of Trizivir, ddI may be active but d4T is less likely to be active if there are multiple NRTI resistance mutations. It is possible that the abacavir component of Trizivir may still be active if early failure is spotted, since this is associated with the emergence of resistance to 3TC , but the combination of ddI and abacavir has not been studied, and is more expensive in most regions than d4T and ddI.

Once daily ritonavir/saquinavir (100/1600mg) has been studied in Thailand and proved well tolerated and effective. The cost of the saquinavir component of this regimen had fallen to approximately $600 per year in May 2002.

Ritonavir/indinavir may be more problematic due to the increased risk of nephrolithiasis (kidney stones and indinavir crystals in urine) caused by high peak levels of indinavir. The risk of nephrolithiasis may be reduced if an adequate fluid intake is maintained, but in hot climates maintaining adequate levels of hydration may be difficult, especially where clean water supplies are a problem. People taking indinavir are advised to drink at least 1.5 litres of water over and above their normal daily intake to ensure that a build-up of indinavir crystals does not lead to kidney damage.

Ritonavir/lopinavir has recently become available at a reduced price in Africa, and is now comparable in cost to ritonavir/saquinavir. Further price reduction is unlikely in 2002 owing to a decision at an April 2002 Abbott shareholders meeting, at which shareholders voted not to sanction a price cut to around $400.

See Anti-HIV therapy: Changing Treatment for further discussion of choosing second-line regimens.

Resistance testing

Where laboratory support for resistance testing is available, clinicians report that patients have been willing to pay for it if they have the means. In Thailand for example, resistance tests cost approximately $70, and have been invaluable due to the history of dual nucleoside analogue therapy prior to the introduction of affordable triple regimens. In 145 patients who had received AZT/ddI for up to six years, Thai researchers found 42 individuals had nucleoside analogue resistance mutations. One third of those had at least four mutations (indicating high level resistance), and three had mutations associated with broad resistance to the nucleoside analogue class.

A network of laboratories has been established to carry out genotypic viral resistance testing. Twelve labs have now been accredited and 60 physicians have been trained to provide an expert-led reference service. The Ministry of Health has acquired 6000 test kits which will be used in HIV+ people experiencing their first treatment failure, with a viral load above 5000, on ARV combinations that include at least one protease inhibitor. It is clear that such a programme is a commitment not to be undertaken lightly, but it is argued that it can still be justified, even in a country with very limited economic resources (Dantas).

However, in most countries resistance testing will be unaffordable and laboratories will not be equipped to carry out the tests. WHO nonetheless recommends that arrangements to use resistance tests - perhaps on stored samples, shipped abroad - should be considered as a surveillance tool. For example, to track the acquisition of drug-resistant virus among people who have recently acquired HIV.

Cautious prescribing based on treatment history is the only realistic alternative, and whilst trials have shown that this approach is inferior to resistance testing, it is nevertheless possible to make informed guesses about the likely cross-resistance profiles of drugs that have been in use for some years, since extensive information has accumulated regarding these drugs.

See Anti-HIV therapy: Resistance for more information on resistance testing and drug resistance patterns.

Realism about what is achievable

Where treatment is switched due to clinical progression, the development of drug resistance and the limited range of affordable agents may limit the chances of a sustained response to a second-line regimen. The durability of first-line antiretroviral treatment in resource-limited settings is still unknown, since only limited numbers of patients have been followed. Even less is known about responses to second-line treatment.

Given all the difficulties inherent in rolling out large scale access to antiretroviral therapy in resource-limited settings, it should not be surprising if there is a relatively high failure rate for first line therapy. Antiretroviral therapy may not provide such sustained gains in terms of life and health as seen in the North.

References

Dantas MCS et al. Building up a national network for HIV genotyping test in Brazil. XIV International AIDS Conference, Barcelona, abstract B10188, 2002.

Weidle PJ et al. HIV/AIDS treatment and HIV vaccines for Africa. The Lancet 359: 2261-67, 2002.

World Health Organisation. Scaling up antiretroviral therapy in resource limited settings: guidelines for a public health approach (draft). WHO, 2002.