This article was written by Keith Alcorn and Theo Smart, with input from HATIP advisory panel members Chris Green (Indonesia), Francois Venter, Catherine Orrell, Norman Nyazema (South Africa), Brian Gazzard (UK), Rob Camp (USA).

• Many countries that are planning national treatment programmes are now making decisions about first-line ARV combinations.
• There are conflicting opinions about which drug combinations to use in first-line treatment and whether fixed dose combinations that combine all the drugs in one tablet should be prioritised.
• The World Health Organisation, strongly supported by the experience of Medecins San Frontieres, has recommended that the first-line regimen should be a fixed dose combination of stavudine (d4T), lamivudine (3TC) and nevirapine. Other options also preferred by WHO are fixed dose combinations of AZT/3TC/nevirapine (not yet prequalified by WHO) and efavirenz-containing combinations that include AZT/3TC or d4T/3TC as a fixed dose tablet.
• The ability to recognise clinical symptoms of liver toxicity is an important skill for any health care worker managing patients receiving fixed dose combinations containing nevirapine. Instructions on how to recognise the signs must also be given to patients in language they understand when starting treatment, and patients should receive clear information on the risks and benefits of the treatment they will receive.
• It is unclear now if the US government will fund generic fixed dose combinations prequalified only by WHO. It seems likely to apply different criteria to its choice of drugs.
• This article looks at some of the issues being debated at the moment and also covers the key toxicities that must be watched for in patients taking the main combinations.

Since WHOs 3x5 initiative was first announced, a growing number of nations have taken the first steps towards scaling up the health care infrastructure needed to administer ARV treatment to at least 3 million people with HIV by the end of 2005. As of February 2004, forty-two countries had formally requested participation in the 3 by 5 Initiative. Some including India, South Africa, Zimbabwe, Malawi and Mozambique, to name a few, have announced plans to rollout ARV treatment in the public sector. Some will rely upon caregivers with only minimal training to administer simplified and standardised ARV regimens.

Taking a page out of Botswanas book, many countries that lack that infrastructure are forging new strategic partnerships (like ACHAP). An alliance with an experienced and/or well-resourced partner such as Harvard AIDS Institute can speed the development and implementation of effective and sustainable medical/public health interventions. Other partnering organisations offer primarily technical expertise, such as the Clinton Foundation, which played a leading role in designing at least eight national treatment programmes in Africa and the Caribbean and also negotiated much cheaper prices on a generic ARV regimen that those programmes may use.

Sufficient financial resources have not yet been mobilised to fully support the 3 x 5 initiative and national HIV programmes. However, this year both the Global Fund and Presidents Bushs Emergency Plan for AIDS Relief (PEPFAR) will begin disbursing billions of dollars, some of which will be used to support ARV rollout.

The Global Fund to Fight AIDS, Tuberculosis and Malaria was the first international fund to support the purchase of ARVs. Since 2001, the Global Fund has committed US $ 2.1 billion dollars in grant money to combat the three pandemics. However, only US $ 245 million has been disbursed so far. Disbursement is performance-based so that recipients will only receive grant money once they can demonstrate the capacity to use it effectively.

However, according to the Global Fund projections, grant disbursements will rise to a total of almost US $ 1 billion by the end of 2004. Some of this money will be used to purchase ARV drugs and other essential medicines to support the 3 x 5 initiative.

In its fourth round of grant proposals (due early April) the Global Fund acknowledged this round of proposals came in the wake of the of the 3 by 5 initiative [and that] the Fund expects most applications to be related to furthering this goal.

This year US$2 billion dollars (US) have been allocated to Presidents Bushs Emergency Plan for AIDS Relief (PEPFAR). PEPFAR is a five-year, $15 billion initiative to combat the HIV/AIDS pandemic. The plan includes nearly $9 billion in new money to specifically target 14 heavily affected countries in Africa and the Caribbean. But some think that PEPFAR grants may come with strings attached, particularly when it comes to selection of ARV regimens (particularly where they can be bought from).

Like the Global Fund, PEPFAR grants can be used to fund the purchase of ARV drugs in a national or provincial HIV treatment programme. However, there is a great deal of confusion as to whether PEPFAR monies can be used to purchase cheaper generic antiretroviral drugs or whether the US plan will insist that only name branded pharmaceuticals can be purchased and used even in those countries where no patents exist (see below).

The 3 x 5 Initiative clearly favours the use of simplified generic coformulated antiretroviral regimens. WHO believes that the price of the generics is what makes 3 x 5 affordable, and that generic coformulations so simplify administration that minimally trained paraprofessionals can be relied upon to treat people, if need be. The only branded triple drug coformulation currently marked is Trizivir which data show is less effective than NNRTI or PI-based ARV regimens.

Some national and/or provincial HIV programmes have already begun procuring and treating patients with ARV drugs but deciding which ARV regimen should be used as the first line option and where the programme should buy it from can be difficult. A number of factors must be considered in this choice. Last year, participants at a workshop held in Nairobi by Medecins sans Frontieres (MSF) concluded that an ideal first-line ARV regimen should be:

• Effective and well tolerated, with minimal side effects
• Potent, even in advanced patients, and robust (favourable resistance profile)
• No interactions or contra-indications
• Appropriate for use in TB patients and in pregnant or lactating women
• Available in a fixed-dose combination (FDC) once or twice a day
• Stable in tropical conditions
• Doesnt require laboratory monitoring
• Affordable.

Unfortunately no existing first-line regimen fulfils all these criteria. On the basis of present knowledge and evidence, there is no ideal combination that can be immediately used in all patients with TB (using rifampicin) and in pregnant women. However, in an emergency, one must make do with whatever is available.

Fixed-dose combinations (FDC) are seen as important tools for scaling up in resource-poor, high-prevalence settings. They are easier to take and easier to distribute and usually less expensive than combinations in which each drug must be dosed as a separate tablet.

To see more information on the potential advantages of fixed dose combinations, published by MSF, click here

One regimen was deemed to be preferable to the others by the MSF workshop, and was also endorsed as the first choice by a WHO-sponsored meeting in Zambia last November that sought to make operational recommendations on how to carry forward the 3 x 5 goals. That regimen is nevirapine/d4T/3TC, taken as bi-daily (BID) fixed-dose combination (FDC). Efavirenz-based regimens are the preferred option in people who also require TB treatment alongside ARVs.

To see more information on the WHO operational recommendations for implementing the 3 x 5 plan click here.

WHO has selected generic FDC nevirapine/d4T/3TC as a preferred regimen for use in the 3 x 5 initiative because of its simplicity and price. Recently, it added two generic FDC formulations of nevirapine/d4T/3TC to its list of prequalified drugs deemed safe, efficacious and of high quality to be distributed in developing and transitional countries. Prequalification entails a rigorous review process and ongoing quality monitoring and is meant to prevent the entry of substandard and counterfeit medicines into the supply channels.

More information on the prequalification process and the drugs that have been prequalified can be found by following this link.

As part of the 3 by 5 strategy, WHO and its partners have set up the AIDS Medicines and Diagnostics Service (AMDS), a new mechanism created to ensure that safe, effective and affordable medicines of good quality are more easily accessible. The AMDS will not directly purchase medicine but will assist national authorities and programme implementers to procure and maintain a reliable supply of the drugs. Furthermore, the AMDS should eventually help countries secure the best price possible.

More information on the AMDS can be found by following this link.

WHOs treatment guidelines also emphasise the merits of choosing fixed dose combinations to assist in pill taking and simplify drug procurement and distribution. Click here to view the guidelines.

Although fixed dose combinations carry important advantages in terms of adherence, drug supply management and cost, the currently available fixed-dose combinations also have some drawbacks in terms of side-effects that need to be clearly understood when introducing the treatment. At the very least, the introduction of nevirapine-based combinations needs careful training of health care workers to ensure that they are used safely.

Nevirapine carries the risk of several life-threatening toxicities, including hepatitis. US and European drug regulators recently insisted on new warnings about the increased risks of hepatitis accompanied by rash in women with CD4 counts above 250 cells/mm³. The agencies also insisted that anyone who develops a rash after starting nevirapine treatment should receive a liver function test immediately. Liver function tests should be carried out every 14 days for the first two months of treatment, again one month later and regularly thereafter. Anyone found to have liver enzyme elevations of greater than five times the upper limit of normal should switch from nevirapine to another drug, the European agency says. Although the risk of hepatitis is greatest during the first six weeks of treatment, they recommend that frequent monitoring of liver enzymes should continue throughout nevirapine treatment.

These changes occurred after the WHO guidelines were issued. They reflect increasing caution about the use of nevirapine as clinical experience accumulates and the numbers treated with the drug grow.

In the absence of laboratory monitoring patients should be warned that loss of appetite, nausea, jaundice, pale stools, dark-coloured urine and tenderness or swelling in the area of the liver may be signs of liver toxicity, and are reasons to seek immediate medical advice from the person managing their HIV treatment.

In people with viral hepatitis, the risk of liver damage is greater. One recent Thai study showed that nevirapine-treated patients with hepatitis B coinfection were 57 times more likely to develop severe liver toxicity than those without hepatitis who were treated with a protease inhibitor or efavirenz (Law). The Thai findings suggest that in a setting where a primary route of transmission is injecting drug use, nevirapine carries a high level of risk. Similar findings have been reported in Spain, where an incidence of 13.1 cases of severe liver toxicity per 100 patient years of treatment was observed in nevirapine-treated patients (Martinez). Hepatitis B is endemic in many resource-limited settings and coinfection is common.

Nevirapine-based treatment is complicated by the increasing use of single dose nevirapine as prophylaxis against mother to child transmission. Mothers with HIV may develop resistance to nevirapine as a result of a single dose and suffer a poorer response to treatment. This issue was discussed in detail in the previous edition of HIV & AIDS Treatment in Practice.

Nevirapines introduction also has to be carefully managed to avoid severe rash. The key risk factors for developing serious rash include failure to follow the initial dosing of 200 mg daily during the 14-day lead-in period and delay in stopping the drug treatment after the onset of the initial symptoms.

Finally, nevirapine may not be compatible with rifampicin, a common component of the four drug phase of tuberculosis treatment. Rifampicin reduces nevirapine trough levels by about 40%, although some argue that that may not be enough to compromise nevirapines performance. For people who require TB treatment and ARV treatment at the same time, WHO recommends the use of efavirenz or abacavir instead of nevirapine for adults, saquinavir/ritonavir or abacavir for pregnant women, women of childbearing age without adequate contraception and children, and abacavir for low body weight children.

In fixed dose combinations nevirapine is being paired with nucleoside analogue backbones that contain either zidovudine or stavudine. Both drugs present some potential problems.

Zidovudine (AZT) can cause anaemia and neutropenia. Anaemia, a shortage of oxygen-carrying red blood cells, results in severe fatigue. Neutropenia, a shortage of white blood cells called neutrophils, results in a reduced ability to fight bacterial infections. Both conditions are caused by AZTs effect on the bone marrow. Although dose reduction has been tested and found effective, fixed dose combinations of AZT/3TC do not permit a reduction from 300mg of AZT to 250 mg or 200mg twice daily.

Patients in resource-poor settings appear to be more vulnerable to AZT-related anaemia, partly because of high levels of pre-existing anaemia in the population due to malnutrition and/or malaria. In Botswana 7.8% of 288 patients who started treatment with AZT-based therapy developed grade 3 or 4 anaemia (haemoglobin below 6.9) and four patients died as a consequence, according to a report at last years International AIDS Society conference in Paris (Wester) Colleagues subsequently reported that whilst baseline haemoglobin testing might help predict the majority of cases of anaemia, 39% of those who developed severe anaemia had baseline haemoglobin above 10g/dL, suggesting the need for ongoing testing (Johnson).

Stavudine can also cause neutropenia, but its other side effects are potentially more serious. Lactic acidosis, a build up of lactate in the blood, appears most likely to result from treatment with d4T and ddI together, but d4T alone is the second most frequently associated drug. Nausea, malaise, fatigue and muscle weakness indicate the onset of a potentially life threatening syndrome. Nucleoside analogues should not be resumed until lactate levels normalise easier said than done in patients with advanced disease, few available drug options and limited ability to monitor lactate levels. Laboratory monitoring of lactate appears to be of little value in predicting who will develop lactic acidosis.

Another difficult side effect is peripheral neuropathy, or nerve damage, particularly in the legs and feet. Around 15-20% of people who start d4T treatment will develop this condition The risk of this condition is greatest in people with more advanced HIV disease, precisely the people who will be treated with stavudine. The risk may also be increased in people taking isoniazid for TB prophylaxis or treatment alongside d4T. If peripheral neuropathy develops, stavudine should be stopped and not reintroduced unless no replacement is available in the forseeable future. If it must be reintroduced, the daily dose should be halved.

Some of these problems might be avoided if doctors and pharmacists paid proper attention to the need for stavudine to be dosed by weight. People weighing less than 60kg should receive a 30mg dose twice daily. The prequalified generic fixed dose combinations that combine stavudine with another drug are all available in either 30mg or 40mg forms.

Scant attention is paid [in Southern Africa] to differing dosage forms for Triomune so that a number of patients are overdosed with the 40mg d4T dosage form and the risk of drug-induced neuropathy is increased. d4T-containing fixed-dose regimens in the presence of treatment for tuberculosis [with isonaizid] will lead to increased occurrence of neuropathy, Dr Desmond Martin told HIV & AIDS Treatment in Practice last April when we first reviewed the use of nevirapine-based fixed dose combinations.

Click on this link to view HATIP # 3.

MSF argues that its approach is paying dividends in terms of high levels of adherence, and that rates of of toxicity are lower than anticipated. Dr. Alexandra Calmy, MSFs AIDS Advisor, told HIV & AIDS Treatment in Practice that the organisation is in the process of collating data on nearly 10,000 people who have now been treated through MSF antiretroviral projects.

Cutaneous rash is rarely reported after people start treatment with nevirapine. Perhaps it is underdiagnosed, or perhaps it is because existing pruritic skin conditions are so common that people are little troubled by it. We have seen that around 1% of patients have discontinued nevirapine-based combinations because of rash, she said.

Liver toxicity is similarly infrequent, according to MSF, with grade 3 and 4 liver enzyme elevations occurring in around 1% of patients. Almost all MSF projects have been measuring liver enzymes every two weeks but its doctors are now discussing whether liver enzyme measurement can be abandoned in view of such low incidence. Until now MSFs policy in settings such as Mozambique and Cameroon has been to monitor twice in the first month, to increase the nevirapine dose after 14 days only in patients with grade 1 liver enzymes whilst continuing to monitor for a further two weeks in patients with grade 2 elevations before increasing the dose. Monitoring continues in this group for another month. Patients with grade 3 or 4 liver enzyme elevation at any point are switched to efavirenz.

We found in Cameroon that liver enzyme elevations of ten times the upper limit of normal were invariably accompanied by clinical signs, Dr Calmy told HIV & AIDS Treatment in Practice. Clinical signs of liver toxicity include nausea, vomiting pain and tenderness in the right upper quadrant of the torso and/or the abdomen, and jaundice. Nevirapine treatment should be stopped and should not be resumed until the symptoms go away. An alternative approach is to switch from nevirapine to efavirenz where available.

In comparison Botswanas Princess Marina Hospital has reported that 3.4% of the first 146 patients who started nevirapine-based HAART developed grade 3 or 4 liver enzyme elevations. Two deaths occurred (Johnson).

Neuropathy also appears to be infrequent. MSFs project in Cambodia has observed that only 1% of patients developed peripheral neuropathy after six months of treatment and after 18 months the proportion that has developed neuropathy (grade unspecified) is only 7%

Overall, 20% of the patients in our projects receiving first-line regimens have stopped one component of the regimen, half because of intolerance, the other because of clinical issues like TB that require a change of medication due to drug interactions, or else because of poor adherence.

Overall, she said, 2.6% of patients had stopped nevirapine due to severe toxicity, 1.3% had stopped efavirenz due to severe toxicity and 5% of AZT-treated patients had to stop treatment due to severe anaemia.

These rates of treatment switching due to poor tolerance of medication, although difficult to compare directly to other studies due to lack of data, nevertheless suggest that the fixed dose regimens advocated by WHO and MSF are not as poorly tolerated as studies in the developed world might suggest.

Of course, it is possible that patients in Europe are quicker to complain, and that if you dont ask the question you dont collect the data, but our impression is that these regimens are well tolerated in the vast majority of patients, Dr Calmy said.

Where the alternative option is no treatment and early death, its amazing what people will tolerate, says Chris Green, a treatment advocate in Indonesia. Should they have to? That's a different question.

Lipoatrophy, a side-effect associated chiefly with d4T or AZT treatment in the developed world, is rarely seen in African patients, Dr Calmy told us. In India however, a prospective cohort of 107 patients recently reported at the Eleventh Conference on Retroviruses and Opportunistic Infections found that 21% of patients lost weight in the first six months of HAART treatment (a mean of 3kg, of which 1.3kg was fat), but no association was found with d4T-based treatment, low body mass index or active tuberculosis (Saghayam). Skinfold measurements showed clear fat loss from the limbs. This pattern is in contrast to the natural history of fat loss seen in North American cohorts, where fat loss rarely becomes evident in the first year of treatment, and may point to different nutritional, metabolic or genetic drivers of HAART-induced fat loss in Asian people. Lipoatrophy has also been reported in Thai patients.

We need much better reporting systems to identify side effect profiles in all of the hundreds of major ethnic groups concerned" says Chris Green. "I think we also need to start setting up `sample banks` to store blood samples of patients for later examination as more sophisticated tools become available.

In South Africa, Dr Catherine Orrell has seen similar levels of adverse events to those reported by MSF. Less than 10% of patients in the Hannan Crusaid Treatment Cohort in Gugulethu are receiving nevirapine, and only one has discontinued treatment due to an adverse event.

However Dr Francois Venter, who treats patients at Johannesburg General Hospital, is more cautious. It is difficult to argue with the MSF experience, as they have documented things well, but I have had lots of problems with nevirapine. Obviously, if nothing is available, or it means treating large numbers of additional people, I'd use it, but I've seen several near-deaths with nevirapine. I saw a large percentage get side effects, and I have effectively stopped using it, unless someone cannot tolerate efavirenz.

Dr Catherine Orrell believes that concern over adverse events must not be allowed to get in the way of viable treatment. We have to be careful not to discount drugs due to a specific, manageable adverse event (either AZT, d4T or nevirapine). We must expect a few problematic patients in any programme, she said.

Chris Green agrees. We should be aware of the risks. We MUST ensure that patients are treated with informed consent, particularly regarding the known and unknown risks. But we must press on, ensuring that we are not discouraged or unduly hindered by the concerns!

The main alternative to use of nevirapine-based fixed dose combinations is the use of an efavirenz-based combination. The main advantage of efavirenz is that it can be used alongside rifampicin. The risks of serious rash and liver toxicity are low too.

The most common side effect seen with efavirenz is a spectrum of central nervous system problems that range from mild dizziness and vivid dreams at one end of the scale to depression and psychosis in more vulnerable individuals. While there has not been a high rate of CNS problems reported in the developing world, there may not be culturally appropriate tools to monitor this type of side effect.

Recent US research suggests that the side effects may be related to drug levels and that there may be ethnic differences in drug metabolism, with Americans of African origin significantly slower to clear the drug and therefore at potentially higher risk of its side effects than Caucasians due to a genetic variation in drug metabolism that is much more common in Americans with African ancestry (Haas).

Not all studies support a link between drug levels, ethnicity and toxicity however, so more research is urgently needed both on tolerability and predictors of tolerability across ethnic groups.

Another disadvantage of efavirenz is that it can harm the unborn child and all guidelines stress that it should not be prescribed to women without adequate contraception (which should include barrier protection). In Botswana guidelines recommended that efavirenz be prescribed to men and nevirapine to women in order to guard against this hazard. Some activists who believe efavirenz to be the better drug contend that this is second class treatment of women, and essentially constitutes putting the health of a non-existent though possible foetus ahead of the womans health.

Furthermore the single 600mg efavirenz tablet available in rich nations is still unregistered in some key countries planning to scale up ARV treatment, including South Africa, so an efavirenz-based combination is less convenient than a nevirapine-based fixed dose tablet due to a larger pill burden. WHO has not yet prequalified a 600mg tablet manufactured by Cipla.

WHO guidelines recommend tenofovir as a component of second-line treatment, preferably with didanosine, because both drugs have some activity after the failure of AZT/3TC or d4T/3TC.

But some prominent clinicians disagree with this positioning. Professor Brian Gazzard of the Chelsea and Westminster Hospital, London, argues that the best possible fixed dose combination would group drugs with similar half-lives, so that if a dose is missed, drug levels would decay at similar rates, lessening the risk of resistance.

In his view, the combination of tenofovir, 3TC and efavirenz would be most suitable for once daily dosing. If this regimen failed, second line treatment might be able to recycle tenofovir

Speaking at the Eleventh Conference on Retroviruses and Opportunistic Infections last month Prof. Robert Schooley of the University of Colorado urged that treatment programmes should use the best possible regimens now available and spend more now to avoid spending much more on salvage of failed treatment later.

You may treat more people in year 1 but what happens in year 3 when a third of patients have failed and need treatment with regimens that cost three times as much? he asked.

Tenofovir, 3TC and efavirenz or d4T/3TC/efavirenz were his examples of regimens that achieved viral suppression in close to 90% of patients after a year of treatment, and which resulted in very low rates of virologic failure leading to drug resistance. Just 12% in each arm experienced virologic failure by week 96, of whom just over a third in the tenofovir group and half the d4T group had no evidence of drug resistance.

A number of other regimens, both efavirenz- and protease inhibitor-based regimens have produced similar results in recent studies.

Comparisons with cohorts in resource-limited settings are difficult because patients in resource-limited settings have higher viral load and more advanced disease, probably predisposing them to a poorer response to treatment. Until longer-term studies can be carried out in resource-limited settings it is impossible to say whether the regimens that are being recommended by WHO are likely to result in a higher rate of failure, or whether comparable results can be achieved with nevirapine-based combinations. The 2NN study reported last year showed that treatment with nevirapine/d4T/3TC was equivalent to treatment with d4T/3TC/efavirenz when judged by treatment failure at week 48, with no significant difference in the rate of clinical adverse events between the nevirapine twice daily and efavirenz groups (van Leth).

In some cases, a strategic partner or funder may be the one to choose a national programmes firstline regimen. Where Merck is a partner, free efavirenz will provide an obvious incentive to use efavirenz-based regimens, particularly in men.

Nations that partner with the Clinton Foundation not only receive expert assistance in designing their national ARV programmes, they also reap the benefits of the Clinton Foundations negotiating skills which have secured the lowest price yet on a generic nevirapine-based FDC regimen: $132 a year, half the price offered to other countries. Countries now able to benefit from a fixed dose combination of nevirapine and 3TC plus either d4T or AZT include Rwanda, Mozambique, Tanzania, Haiti, The Bahamas, the Dominican Republic, the Eastern Caribbean States and at one time, South Africa.

Now, however, South Africa has begun its own prolonged drug procurement process, which could be yet another stalling tactic to further postpone the ARV rollout there. Or perhaps the South African government simply prefers not to have the choice of first line regimen dictated to it and who can blame it. But since it was the assurance of a predictable volume of drug orders that clinched the deal for Clinton (because it convinced manufacturers to take the risk of scaling up large scale production), it is not clear what will become of the arrangement now that the country that represented the lions share of the order has bolted.

The current president of the United States may wind up having more of an influence at least in the fourteen PEPFAR beneficiary countries in Africa and the Caribbean. These are Botswana, Cote dIvoire, Ethiopia, Guyana, Haiti, Kenya, Mozambique, Namibia, Nigeria, Rwanda, South Africa, Tanzania, Uganda and Zambia. US administration officials privately expect that US funds will pay for the treatment of two-thirds of those who receive antiretrovirals over the next two years, so the US governments purchasing rules will have just as big an influence on which drugs are used as the WHOs guidelines.

HIV & AIDS Treatment in Practice has learnt that the US government met with US pharmaceutical companies in January and asked them to work on developing fixed dose versions of their products or blister packed drug combinations in partnership with each other.

Meanwhile some members of the administration have said it will not allow US government money to be spent on fixed dose combinations that have not been approved by a stringent regulatory authority in the developed world (i.e. the US, Canada, the European Union or Japan.). A subsequent Centers for Disease Control briefing paper has reiterated this position.

Yet when launching his initiative President Bush specifically mentioned using drug combinations that cost US$300, implying generic products, a position confirmed by Bushs budget director to consumer advocate Ralph Nader. Furthermore, one of the first recipients of PEPFAR funding, the Harvard AIDS Institute, is operating in the belief that it can purchase generic antiretrovirals with PEPFAR money for treatment programmes in Nigeria and Botswana. The same is true for one yet-to-be-announced PEPFAR recipient in South Africa, who has openly discussed plans to purchase ARV generics with those administering PEPFAR.

Pharmaceutical companies and the US administration have preferred to highlight issues of quality rather than safety when talking about fixed dose combinations, possibly because safety is an inescapable talking point whoever makes the drugs.

A meeting will now take place in Botswana on March 29/30 to agree standards for approval of fixed dose combinations. The Botswana meeting will consider proposals for approval of fixed dose combinations that would theoretically allow the better studied drug combinations to be coformulated without the need for expensive efficacy studies. This would not discriminate against generic manufacturers provided that their drug production methods and facilities met European or US standards. However the proposals also contain language implying that the post-marketing surveillance burden will fall on generic manufacturers if they wish to get fixed dose ARVs licensed a questionable requirement given that they are producing generic products under compulsory license at the request of national governments.

Norman Nyazema of South Africa's University of the North argues that fixed dose combinations do present regulators with difficult issues, but that what is needed is the support to develop national regulatory systems and quality assurance systems. "The issue is not just about adherence and cost. What happens if people experience an adverse drug reaction to an FDC? How is causality going to be assessed? What regulatory decisison is going to be taken?"

This issue is particularly important as new drug combinations are assembled and introduced in settings where the rates of toxicity may differ because of population factors like hepatitis.

A common question about any potential first line ARV treatment is whether it is suitable for administration with only barebones lab monitoring. How safe is it to skimp on laboratory monitoring? Despite MSFs extremely low reported rate of adverse events, the organisation has on many occasions acknowledged that it selects highly motivated patients who are deemed likely to succeed on therapy. Thus, their experience may be a best case scenario that cannot be extrapolated to all those in need of treatment.

Treating with ARVs without any laboratory monitoring is a controversial strategy. Any of these regimens may be unsafe if used at full dose in patients with renal impairment, elevated liver enzymes or hepatitis, regardless of who manufactured the drugs. Can clinical symptoms identify all or most of the patients with significant kidney or liver problems that could complicate ARV therapy? How? Will every health care worker involved in ARV administration be adequately prepared to recognise such signs in time to do no harm?

At present no one knows the answer to those questions, and in the words of the WHO mantra, we are learning by doing. Unfortunately in the case of adverse events this can mean learning by doing harm. As Chris Green noted earlier in this article, We MUST ensure that patients are treated with informed consent, particularly regarding the known and unknown risks.

Surveillance systems need to be part of national monitoring and evaluation programmes, and in turn these need to be properly funded by the international community if it is serious about bringing treatment to resource-limited settings.

To some extent the choice of first-line regimen will lie outside the power of individual clinicians or national treatment programmes.

However, what lies within the power of everyone reading this article is to ensure proper preparation of health care workers and patients before starting treatment.

Training programmes in settings where laboratory monitoring for all potential toxicities is not available should focus on clear instructions about dosing by weight and when dosing should be increased if stavudine and nevirapine are being used, clinical recognition of drug toxicities, instructions on how to manage drug toxicities and how to substitute drugs if specific toxicities occur.

A set of training slide sets on each drug combination recommended for first-line therapy by WHO has been developed by the editors of this newsletter in consultation with doctors carrying out training in resource limited settings. These slide sets provide clear guidance on the monitoring of fixed dose ARV combinations where resources are limited.

To download the slide sets follow this link.

Johnson O et al. Preliminary analysis of toxicity and tolerability among the first ARV-treatment naïve HIV-1C infected persons in the public sector at Princess Marina Hospital. NHASORC, Gabarone, December 2003.

Law WP et al. Risk of severe hepatotoxicity associated with antiretroviral therapy in the HIV-NAT Cohort, Thailand, 1996-2001. AIDS 17: 2191-2199, 2003.

Martinez E et al. Hepatotoxicity in HIV-1-infected patients receiving nevirapine-containing antiretroviral therapy. AIDS. 2001 Jul 6;15(10):1261-8.

Saghayam S et al. Weight and Body Shape Changes Associated with Antiretroviral Therapy in South India: Initial Prospective Analysis. Eleventh Conference on Retroviruses and Opportunistic Infections, San Francisco, abstract 586, 2004.

Van Leth F et al. Results of the 2NN study: a randomized comparative trial of first-line antiretroviral therapy with regimens containing either nevirapine alone, efavirenz alone or both drugs combined, together with stavudine and lamivudine. Tenth Conference on Retroviruses and Opportunistic Infections, abstract 176, 2003.

Wester W et al. Preliminary analysis of toxicity and tolerability among the first ARV-treatment naïve HIV-1C infected persons of the Botswana national ARV treatment program. Second International AIDS Conference on HIV Pathogenesis and Treatment, Paris, abstract 1213, 2003.