Emerging Resistance to Nonnucleoside Reverse Transcriptase Inhibitors: Title and subTitle BreakA Warning and a Challenge

Two decades after the recognition of the human immunodeficiency virus (HIV) epidemic, the potential for widespread treatment of HIV infection is now at a pivotal, yet precarious time. In the United States and Europe, patients with HIV infection are living longer without acquired immunodeficiency syndrome (AIDS)–defining illnesses, new highly active antiretroviral therapy (HAART) strategies are being developed and tested, and resistance testing is becoming a new standard for managing therapy after failure with HAART. Yet in much of the developing world, most notably sub-Saharan Africa, HIV and AIDS are devastating innumerable communities. At a time when the introduction of HAART in Africa is in its infancy, 2 articles in this issue of THE JOURNAL illustrate the crucial role the nonnucleoside reverse transcriptase inhibitors (NNRTIs), as well as the implications of NNRTI resistance, will play in the treatment of HIV in both developed and developing countries.

The study by Hammer et al¹ is a randomized controlled trial comparing dual vs single protease inhibitor (PI) therapy in 481 patients who failed HAART treatment. The investigators demonstrated that the addition of a second PI to an amprenavir (PI)/efavirenz (NNRTI)–based regimen decreased the HIV-1 RNA level to less than 200 copies/mL in 35% of patients, compared with 23% in the placebo plus an amprenavir/efavirenz-based regimen group at 24 weeks (P = .002), and 30% vs 22% (P = .04), respectively, at 48 weeks. There was a modest increase in the CD4 cell count in the second PI group vs the placebo group at 24 weeks (+34/µL vs +13/µL; P = .048), but the 2 groups had similar increases at 48 weeks. However, a more intriguing finding was the association of virologic failure with NNRTI experience and efavirenz resistance. Forty-three percent of NNRTI-naive patients achieved an HIV-1 RNA level of less than 200 copies/mL at 24 weeks compared with 16% of NNRTI-experienced patients (P<.001). Moreover, reduced susceptibility to efavirenz was associated with decreased virologic suppression, and efavirenz hypersusceptibility was associated with virologic response to treatment. This is a striking finding given that all 4 classes of antiretrovirals (ARVs) were used in this study.

The effectiveness of the NNRTI class is impressive but not surprising. Patients treated with an NNRTI-based regimen were significantly more likely to achieve an undetectable viral load and had better outcomes at 6 months when compared with PI-based regimens in 888 ARV-naive patients.² In a study of 450 patients comparing efavirenz plus 2 nucleoside reverse-transcriptase inhibitors (NRTIs) vs indinavir (a PI) plus 2 NRTIs, the efavirenz-based regimen achieved undetectable HIV-1 RNA levels in 70% of the patients, compared with 48% with the indinavir-based regimen (P<.001).³ The NNRTIs have been used successfully both in combination with PI-based regimens⁴ and in multidrug salvage therapy for patients who did not respond to PI-based regimens.^{5 - 6} Numerous studies also have shown that switching from a PI-based regimen to an NNRTI regimen results in decreased rates of virologic failure,^{7 - 8} decreased rates of ARV adverse effects,⁹ increased rates of adherence,^{8 - 9} and increased quality of life.^{8 - 9}

However, emerging resistance to the NNRTI class of drugs may be increasing, especially when the drugs are administered in nonsuppressive regimens or as single-dose prophylaxis. In another study in this issue of THE JOURNAL, Grant et al¹⁰ performed genotype and phenotype analysis on HIV isolates obtained from 225 patients in San Francisco with acute HIV infection from 1996 through 2001. The salient observation in this study was the marked increase in NNRTI resistance in HIV isolates even before ARV therapy was initiated. Genotype mutations conferring resistance increased from 0% in 1996-1997 to 13.2% in 2000-2001 (P = .01), and phenotypic resistance increased from 0% to 9.9% (P = .02). In contrast, PI genotypic resistance increased less dramatically and phenotypic resistance to NRTIs actually decreased. A similar study from the United Kingdom demonstrated that primary NNRTI resistance increased from 0 of 22 individuals in 1997-1999 to 3 of 26 (11.5%) individuals in 2001.¹¹ Additionally, the HIVNET 012 trial in Uganda that used single-dose nevirapine (an NNRTI) intrapartum and then neonatally as prophylaxis to prevent transmission of HIV from mother to infant, resulted in single-point mutations that were associated with NNRTI resistance in 19% of mothers and 46% of the 16% of infants infected with HIV despite nevirapine prophylaxis.^{12 - 13}

The articles by Hammer et al¹ and Grant et al¹⁰ illustrate the importance of the NNRTI class and its pivotal role in the future management of HIV infection. The HIV susceptibility to this class of drugs may be an important predictor of achieving therapeutic success, but resistance may be developing rapidly. Moreover, virus with mutations reflecting PI resistance and multidrug resistance is often less robust than wild-type virus, with the viral load often not returning to pretreatment levels after "virologic failure."¹⁴ However, some studies suggest that "viral escape" from NNRTI-based regimens may occur, resulting in viral load levels that may return to baseline, or even higher, with an associated reduction in the CD4 cell count.^{15 - 17}

As important as these findings are for care of patients with HIV infection in the United States and Europe, the implications may be even more important for the developing world, such as sub-Saharan Africa, where there are almost 25 million individuals infected with HIV.¹⁸ Until now, HAART has not been available to the vast majority of these HIV-infected individuals, due to the prohibitive cost and limited expertise in administering ARVs. However, ARVs are becoming more universally available. The Global Fund has recently distributed its first cycle of funds to resource-limited nations to fight HIV, tuberculosis, and malaria. The US government has also invested resources through the US Agency for International Development, Centers for Disease Control and Prevention, and National Institutes of Health, among others, to facilitate AIDS care and to find solutions to control the HIV epidemic in Africa. Joint public and private coordinated efforts to support HIV care in sub-Saharan Africa include, among many others, the Merck/Bill and Melinda Gates Foundation project in Botswana, the Doris Duke Foundation/Harvard University project in South Africa, and the Pfizer/Academic Alliance for AIDS Care and Prevention in Africa, in association with Makerere University in Kampala, Uganda. The missions of these large initiatives are, in part, to train African health care practitioners in the evidence-based use of ARVs and to build up the infrastructure to facilitate large HIV treatment campaigns and research programs.

However, because drug cost is a critical factor in the availability of ARVs, the least expensive drugs will dominate any treatment program. While the NNRTIs and NRTIs are relatively inexpensive to produce, the PIs are more difficult to manufacture and currently considerably more expensive. Without the ready availability of PIs, preservation of NNRTI susceptibility is clearly a critical factor to ensure sustained, widespread treatment success in sub-Saharan Africa and other resource-limited areas.

Maintaining widespread susceptibility may not be an easy task, especially if there is not a reliable supply of inexpensive ARVs. A study of resistance patterns from 1998-1999 in 68 individuals from the Ivory Coast, where an ARV pilot program had been initiated, revealed phenotypic resistance to at least one NRTI to be 39.7%; for NNRTIs, 8%; and for PIs, 6% of patients.¹⁹ A major factor leading to virologic failure is believed to be unplanned treatment interruptions.²⁰ In Africa, the major barrier to optimal drug adherence may be the lack of resources to purchase the ARV drugs. If an uninterrupted supply of ARVs can be guaranteed, drug adherence may be good, or even better, than in the United States or Europe. For example, in a study of 270 treatment-naive patients receiving a 3-drug ARV regimen in Cape Town, South Africa, it was found that drug adherence in Xhosa, Afrikaans, and English-speaking patients was greater than 90% at 12 and 48 weeks.²¹

While ARV resistance is a concern, the clear benefit of treatment should drive the dissemination of HAART into Africa and other resource-limited continents. The struggling nations of sub-Saharan Africa and those in other resource-limited regions are desperate for immediate action to control the HIV epidemic. Thus, HAART is essential to preserve human life, maintain the labor force and social stability, reduce HIV transmission, and provide optimal management of HIV infection. The time to introduce HAART is now. The issues to be addressed are how to use HAART and whether HAART will be reliably available to everyone.

Furthermore, the medical community and governments must adapt as fast as the HIV virus can mutate. The effectiveness of NNRTIs is due not only to their intrinsic antiretroviral activity, but also to their relatively recent arrival as a new class of ARV drugs. Emerging NNRTI resistance illustrates the continuing need for newer classes of drugs, and while western governments need to support patents that encourage the development of new medications, they must also subsidize the costs for drug acquisition in resource-limited regions.

The challenges for the medical and research community are to establish strategies to treat HIV in the individual patient and to maintain potent HIV drugs as viable treatment options. Virologic, immunologic, resistance, and adherence monitoring, new drug development, and education of clinicians in the proper use of ARVs are the keys to the continued success of HAART. Resource-limited areas, such as Africa, also require the most affordable drug regimens possible. Safeguards must be in place to ensure drug availability when money is scarce, so adherence will not falter, and to minimize the emergence of resistant HIV strains. Effective and appropriate drug delivery, linked with careful clinical and laboratory monitoring, will be critical to the long-term success of ARV programs in resource-limited areas, as it has been in the West. Wide distribution of ARVs without development of adequate infrastructure is a recipe for the emergence of drug-resistant virus, especially to the NNRTI class, which would lead to both dramatically diminished benefits and only transient success of current efforts.

The success of HAART, the developing story of the potent and relatively inexpensive NNRTI class of HIV drugs, and the threat of emerging resistance to this class of drugs serve as both a warning and a challenge. While there is the potential to make a substantial, worldwide difference in the lives of HIV-infected individuals, care must be taken to do it right.