Applications for the addition of 12 antiretroviral (ARV) medicines to the Model List were received from the WHO Department of HIV/ AIDS. All of the proposed medicines are recommended for use as antiretroviral medicine combinations in the new WHO clinical guidelines for antiretroviral treatment in resource-poor settings (18).

In accordance with the new procedures (see section 3.3), the Committee reviewed written submissions containing information about the 12 medicines; these submissions comprised summaries of available evidence relating to clinical benefits and adverse effects, practical details of treatment, and comparisons with other members of their therapeutic group. The clinical evidence had been assembled from comprehensive literature reviews for each medicine and medicine combination, several of which had been conducted by staff working for the Cochrane Collaboration. The written submissions also provided background information on the public health impact of HIV infection worldwide, the overall impact of ARV therapy on the course of illness in HIV-infected subjects, the value of surrogate markers as measures of treatment response, and a summary of the experience and impacts of delivering ARV therapy in resource-limited settings.¹

¹ The full application is available from the WHO Essential Medicines Library via the Internet at http://mednet3.who.int/mf/.

The content of the written submissions was summarized for members of the Committee in a series of oral presentations made by WHO staff from the Departments of HIV/AIDS and Essential Drugs and Medicines Policy. Presentations were also given on the history of the development of WHO guidelines for the use of ARVs in resource-limited settings and on recent experience with ARVs in selected developing countries. Issues such as the rationale for the selection of first-line and alternative regimens, the development of appropriate criteria for case selection, and minimum standards for monitoring treatment response and toxicity were of particular interest and concern to the Committee.

Having considered the data before it, the Expert Committee agreed that there was substantive evidence to support the claims of efficacy of ARV combinations comprising three or four medicines. Such combinations typically comprise two nucleoside reverse transcriptase inhibitors (NRTIs) given in conjunction with either a non-nucleoside reverse transcriptase inhibitor (NNRTI), or a third NRTI, or a protease inhibitor. The Committee accepted the validity of surrogate markers of efficacy (i.e. CD4 cell counts and viral load estimates), which had been used in the majority of clinical trials of these agents. The Committee also agreed that both meta-analyses of randomized clinical trials and large, well conducted cohort studies provided convincing evidence that combination ARV treatment substantially reduces AIDS mortality. Furthermore, survival gains were not seen to be offset by severe adverse effects. It was thus concluded that effective ARV treatment leads to large absolute reductions in mortality and the restoration of a worthwhile quality of life. However, highly active ARV therapy is not a cure for HIV/AIDS and long-term suppressive therapy is necessary.

Before considering each individual application, the Committee raised a number of issues that were of general concern. These were:

- the need for listing what was seen to be a large number of medicines;

- the safety and efficacy of therapy when delivered with the minimal levels of laboratory monitoring advocated in the WHO clinical guidelines;

- the advantages and disadvantages of fixed-dose combinations.

The first, the large number of medicines proposed for listing, was addressed by staff from the Department of HIV/AIDS. The point was made that while there are many circumstances where one essential medicine may substitute readily for other members of its class, thus allowing the placement of a single agent on the Model List (with appropriate advice about substitution), this is not possible in the case of HIV treatment. Effective therapy requires commencement of at least three medicines simultaneously, and alternative regimens are necessary to meet specific requirements at start-up, to substitute for first-line regimens in the case of toxicity, or to replace failing regimens. The Committee thus agreed that if the ARV agents were to be listed, then all of the proposed medicines should be added to the Model List.

The availability of adequate laboratory monitoring of ARV therapy was a particular concern of several members of the Committee. The current WHO ARV guidelines allow a relatively low level of monitoring, i.e. a level that can be provided by basically equipped health facilities (18). This raises questions about the safety of ARV therapy, in particular, concerns that an inability to recognize medicine toxicity and/or failing regimens could, in turn, lead to viral resistance. It was noted that not many large field studies have been conducted on the efficacy and safety of ARV therapy provided under such circumstances. Following extensive debate, the Expert Committee agreed to accept the current WHO clinical guidelines regarding case selection and monitoring. However, the Committee recommended that these issues should be reviewed at its next meeting, that suitable footnotes should be added to specific medicines in the Model List, and that appropriate organizations should be strongly encouraged to fund field trials of low-level monitoring of ARV treatments. The Committee also supported the proposal that laboratory facilities for CD4 counts should be made available more widely and encouraged the Global Fund to Fight AIDS, Tuberculosis and Malaria to support the provision of such services.

The advantages and disadvantages of prescribing fixed-dose combinations of ARV medicines were also discussed. The principal advantage of fixed-dose combinations is the improved adherence to treatment due to simplification of regimens, which in turn should result in higher levels of efficacy and lower rates of viral resistance. The main disadvantages are inflexibility in dosing, and doubts about the pharmaceutical quality of fixed-dose combination products that are produced in the absence of strict regulatory and quality standards. In principle, the Committee was in favour of assured quality fixed-dose combination products that incorporate suitable doses of appropriate medicine combinations. It was noted that, at present, only a limited number of such combinations are available internationally. However, it is likely that the availability of fixed-dose combinations will increase in the future and it is hoped that the relevant products will be tested under the WHO’s pre-qualification programme as soon as is practicable.

When considering each individual medicine proposed for listing, the Committee took the following materials into account:

- the written application for each medicine (some of which had been available on the WHO web site prior to the meeting);

- any additional written material received from external stakeholders;

- oral presentations (also provided in hard copy) by staff from WHO and the Joint United Nations Programme on HIV/AIDS (UNAIDS).

Generally speaking, the evidence contained in the applications was classified as follows:

• Level 1. Evidence from relevant high quality systematic reviews of unbiased randomized comparative clinical trials.

• Level 2. Evidence from at least one relevant unbiased randomized comparative clinical trial.

• Level 3. Evidence from relevant controlled observational studies.

A summary of the supporting evidence for each application is provided in Annex 2.

The Committee also debated whether the medicines should be included in the core or in the complementary section of the Model List. It was argued that complementary listing could be used to signal that, to date, experience of the use of ARV therapy in resource-poor settings was limited. On the other hand, it was felt that complementary listing would be inconsistent with previous decisions regarding medicines that require monitoring and, moreover, the decision might be used inappropriately as an argument against wider access to ARV medicines. On balance, concerns about the problems of limited laboratory monitoring notwithstanding, the Committee decided that core listing was appropriate, with the addition of suitable footnotes.

In sum, the Committee recommended that the following medicines be added to the core section of the Model List:

- the nucleoside reverse transcriptase inhibitors, zidovudine (ZDV or AZT), lamivudine (3TC), stavudine (d4T), didanosine (ddI) and abacavir (ABC);

- the non-nucleoside reverse transcriptase inhibitors, nevirapine (NVP) and efavirenz (EFV or EFZ);

- the protease inhibitors, nelfinavir (NFV), indinavir (IDV), lopinavir/low-dose ritonavir (LPV/r), ritonavir(r) and saquinavir (SQV).

The Committee also requested that the following notes be added to the Model List:

The antiretroviral medicines do not cure the HIV infection, they only temporarily suppress viral replication and improve symptoms. They have various adverse effects and patients receiving these medicines require careful monitoring by adequately trained health professionals. For these reasons, continued rigorous promotion of measures to prevent new infections is essential and the need for this has not been diminished in any way by the addition of antiretroviral medicines to the Model List. Adequate resources and trained health professionals are a prerequisite for the introduction of this class of medicines. Effective therapy requires commencement of three or four medicines simultaneously, and alternative regimens are necessary to meet specific requirements at start-up, to substitute for first-line regimens in the case of toxicity, or to replace failing regimens. The Committee strongly recommends the use of three- or four-medicine combinations as specifically recommended in the WHO clinical guidelines (18). The use of fixed-dose preparations for these combinations is also recommended, with assured pharmaceutical quality and interchangeability with the single products as approved by the relevant drug regulatory authority.

Selection of two or three protease inhibitors from the Model List will need to be determined by each country after consideration of local clinical guidelines and experience, as well as the comparative costs of available products. Ritonavir is recommended for use in combination with indinavir, lopinavir and saquinavir as a booster, and not as a medicine in its own right.

It was noted that the current WHO ARV guidelines recommend the use of the total lymphocyte count as a surrogate for a CD4 count only for HIV-infected individuals who were, or had been, symptomatic (18). It was requested that the permanent copies of the submissions for the ARV medicines be amended to reflect this.