When should you start HIV therapy?

The main measures of the stage of HIV infection are the CD4 cell count, the viral load (VL) and the clinical picture. It is on these that the strategies for commencement of antiretroviral therapy (ART) are based. But one has to take other factors into account as well.

Is the virologic response going to be durable? Are there any toxic effects? What will the effect be on the progression of the disease?

These topics were ventilated at a session at the XIV International AIDS Conference in Barcelona titled ?“When to start ART.?” The first presentation examined the prognosis and probability of disease progression in drug naïve patients starting ART.

Genevieve Chene of the ART cohort project in France said that prognosis varies widely, depending on the CD4 count as baseline. She showed from data taken from 13 cohort studies that in addition to viral load (VL) and CD4 count, history of IDU, age and stage of the disease were predictive of disease progression.

If ART was commenced at CDC stage C, the prognosis was poor. Baseline VL was not predictive unless it was >5log10. At six months into ART, however, VL is predictive at all levels. Chene said that she hoped that the results of this study would assist clinicians in deciding when to initiate HAART, and to formulate relevant guidelines.

Following this was a discussion of the effect of CD4 count on the durability of the virologic response. John Brooks of CDC presented a thought-provoking paper in which he asserted that HIV-infected persons with CD4 counts <350cells per mL experience significantly less durable reduction in VL at initiation of HAART.

The study did have limitations, however, as the subjects were restricted to those included in the Adult and Adolescent Spectrum of HIV Disease (ASD) Working Group, the subjects may have had prior exposure to ART and not disclosed this fact, and there was a short period of follow-up.

However, he said that starting ART with a CD4 count of less than 200 c/mL produces a less durable virologic response, and that achieving an undetectable viral load increases the durability of the response. Comments from the floor expressed concern about the lack of mention of compliance to ART, which they regarded as another limitation of the study.

It has been suggested that therapy with non-nucleoside reverse transcriptase inhibitors (NNRTIs) could be associated with increased mortality among HIV-infected patients. A paper presented by Robert Hogg from the British Columbia Center for Excellence attempted to debunk this theory. His trial which compared NNRTIs and protease inhibitors, resulted in the conclusion that it was the CD4 cell count and age, and not therapy with NNRTIs were associated with mortality. These results were consistent with those of recent clinical trials but they do contrast with some cohort studies.

If one is to defer ART, for how long can one wait, and what indicators does one use to decide when to start the ART? A possible answer to this question was provided by Alvaro Munoz, of the Johns Hopkins School of Public Health. He said that it was important, when conducting studies of this nature, to adjust for the lead time, and for fast progressors, in determining the time from HIV infection to AIDS.

Munoz said that there are stages when progression is unlikely, but that ART is efficacious at advanced stages. His conclusion was that deferring ART until the CD4 count is <200 c/mL is detrimental to the outcome.

Commencing ART when the CD4 count is between 200 and 350 seems preferable to initiating it when the CD4 count is between 350 and 500.

When to start ART in developing countries is influenced by the limited resources and lack of infrastructure in many of these countries.

In Africa, while millions need ART, only 4% of patients are actually receiving ART, according to Papa Salif Sow, of the Ministry of Health, Dakar, Senegal.

At the Dakar conference in October 2000, certain guidelines and recommendations were made.

It is crucial to do certain investigations before commencing therapy: blood count, liver and kidney function tests and CD4 count.

It is not necessary to measure viral load as, if this was done, many African countries would never start therapy. VL measurement is saved to monitor the efficacy of therapy.

Access to ART must be affordable, and integrated into programs which include counseling and testing, and prevention of opportunistic infections.

Good compliance and continuity of care must be assured. Patient education and appropriate ART prescription and monitoring, and education of health care personnel are vital.

With concomitant tuberculosis, ART is initiated according to the CD4 cell count, and is usually initiated when the CD4 count declines to 350 c/mL.

Whatever the determinants of initiation of therapy, Sow said that it is time to commit to action.

An overall review of the current thinking regarding commencement of ART was provided by Patrick Yeni from France. Many of the points he presented had been agreed by previous speakers, but he said that there is no need to treat patients with a CD4 count >500 c/mL. Studies suggest that there is no difference in outcome between starting treatment with a CD4 count of 2-300 and 3-500. The recommendation is therefore to defer treatment.

There is no uniform answer to the question of whether ART alters progression of the disease. Data on this are not clear, but they are the only data we have, even though the studies have many limitations.

The objectives of delayed therapy are: to maintain the quality of life for as long as possible; to save options and to prevent cross-resistance, to avoid compromising future therapies.

It has not yet been proved that early is better than deferred therapy.

In summary, Yeni said that one should start therapy when the CD4 count reaches 200 c/mL. If the CD4 count is >200, the decision is based on the rate of decline, the HIV-RNA level, patient interest in treatment and the potential to adhere to therapy, and individual risks of toxicity and drug-drug interaction based on the medical history of the patient.