ARV Therapy and HIV Elimination

I'm waiting for my train in Ernakulum in an internet cafe that must be 40 degrees C. While sitting here sweating, I noticed a January 2009 Lancet article that has generated a lot of debate. The article (vol 373, issue 9657: 48-57) uses a theoretical model to asses whether testing everybody for HIV, then starting all positives on immediate ARV would eliminate HIV. [Note that the word "eliminate" does not mean the same as eradicate. They define eliminate as incidence of transmission less than 1 case per 1000 per year.] The idea of treatment as prevention is not new, of course, but this article takes it to the extreme. From the accompanying editorial:

In The Lancet today, Reuben Granich and colleagues (including two of us, KMDC and CFG) use mathematical modelling to assess the impact of expanded HIV testing and earlier antiretroviral therapy (ART) on HIV transmission.1 These researchers evaluated a theoretical programme of annual universal HIV testing and immediate treatment on HIV diagnosis, irrespective of CD4+ cell count, in an HIV epidemic with southern African population dynamics. The exercise suggested that HIV transmission could be substantially reduced within a few years. Elimination of HIV transmission, defined as an incidence below one case per 1000 population per year, could be achieved within a decade, and the overall prevalence of HIV infection reduced to below 1% before the middle of the century. Compared with current practice of starting ART at a specific CD4+ count, deaths would be halved between now and 2050.

The article generated a lot of interesting correspondence, some of which is published in vol 373, 9669. For me, the most important issue was raised in the letter by Jaffe et al:

In their important and provocative article,1 Reuben Granich and colleagues argue that universal voluntary HIV testing and immediate antiretroviral therapy, irrespective of the degree of immune suppression, could eliminate HIV from countries where the infection is highly prevalent. However, we agree with Geoffrey Garnett and Rebecca Baggaley2 that this approach could strongly shift the benefits of treatment from the individual to the population.

Although current HIV treatment guidelines favour earlier treatment, the risks and benefits of treatment for people with CD4+ cell counts above 350 per μL are unknown. Trials of therapy for patients with higher counts are yet to begin.

Within the field of communicable diseases, we are aware of little precedent for the approach of “treating for the common good”. Treatment of diseases such as tuberculosis might have the effect of decreasing transmission, but the primary goal is to decrease morbidity and mortality for the affected person. A better analogy might be found in immunisation programmes—eg, rubella vaccination of infants and children aims to reduce exposure among pregnant women. However, there is still a clear benefit and minimal risk for the individual vaccinee.

The World Medical Association international code of medical ethics states that “A physician shall act in the patient's best interest when providing medical care.”3 If we are to deviate from this basic principle, we will need a robust ethical model for balancing individual and societal benefits.