BHIVA: Annual CD4 counts possible for stable patients, SMART analysis finds

HIV-positive individuals with CD4 counts above 350 cells/mm³ and with stable undetectable viral loads are at very low risk of disease progression and do not require frequent CD4 monitoring, according to a new analysis from the SMART study presented to the 14^th Annual BHIVA Conference in Belfast last week.

The investigators therefore recommend that annual CD4 count monitoring in patients who maintain viral suppression and a CD4 count above 350 cells/mm³ is viable and likely to save both money and time.

There are very few data to guide recommendations on frequency of CD4 count monitoring in patients with undetectable viral loads (below 50 copies/ml). Although current US guidelines suggest three-monthly CD4 counts in all patients, there is no firm UK guidance, and currently policies vary across UK centres.

Previously, a 2002 study in 166 patients from London’s Royal Free Hospital found that CD4 decline was rare and transient where viral load remained undetectable and CD4 counts were over 500 cells/mm³. Only five individuals experienced a decline to below 350 cells/mm³, and these were typically isolated low values with CD4 cell percentages remaining high.

The authors of that study suggested “that it may be possible to reduce the frequency of CD4 cell count monitoring in individuals with a value greater than 500 cells/mm³ so long as regular viral load monitoring indicates a value less than 50 copies/ml.” (Phillips, 2002)

In Belfast last week, Dr Daniella Chilton, of London’s Mortimer Market Centre, presented a new analysis of data from the SMART study examining CD4 count declines, clinical events and time spent with CD4 counts below 350 cells/mm³in the almost 2700 individuals who participated in the viral suppression (VS, or continuous treatment) arm of the study with at least twelve months of follow-up.

The aim of this analysis was to investigate the clinical utility of CD4 count monitoring in patients with undetectable viral loads and good CD4 recovery (which they defined as CD4 counts above 350 cells/mm³).

They chose the SMART study for their analysis because of its well-characterised population of patients with CD4 counts above 350 cells/mm³ and viral suppression; its peer-reviewed clinical outcome data, large numbers of patients, and robust data.

There were a total of 2752 patients in the VS arm, of whom 2696 were included in the initial analysis. The 41 participants who were lost to follow-up and the 15 who died before the end of the first year were excluded to rule out length of time bias.

Data were available reporting all scheduled viral load measurements in the first year for 1471 participants where viral load was recorded below 400 copies/ml, and of these 1471, 688 had viral loads below 50 copies/ml. The remaining 1225 participants had at least one viral load measurement that was above 400 copies/ml in the first twelve months of the study.

Overall there were very few AIDS-defining (11 vs. 6) and non-AIDS defining (6 vs. 9) events in those with viral loads above or below 400 copies/ml, respectively, and the differences were not statistically significant (p=0.12).

The same was true when the investigators analysed the rate of AIDS-defining (0.9 vs. 0.4) and non-AIDS defining (0.5 vs. 0.6) events per 100 person-years in those with viral loads above or below 400 copies/ml, respectively, and again the differences were not statistically significant (p=0.67).

When the investigators examined the rate of CD4 decline (to below 350 cells/mm³) per 100 person-years, they found a statistically significantly higher rate of decline in individuals with a viral load above 400 copies/ml compared with those with a viral load below 400 copies/ml (30.9 vs. 13.4; p <0.0001).

In the second year of follow-up, the rate of CD4 decline was reduced in both groups, but was still significantly higher in individuals with a viral load above 400 copies/ml compared with those with a viral load below 400 copies/ml (25.2 vs. 7.6; p <0.0001).

For both time periods, the rate of CD4 decline in individuals with a viral load below 50 copies/ml was only marginally lower than in those with a viral load below 400 copies/ml.

In addition, the amount of follow-up time that anyone with a viral load below 400 copies/ml spent with CD4 counts below 350 cells/mm³ was 3.2% in the first twelve months, and 3% in the second year of follow-up.

In total, there were 15 clinical events in patients with a viral load below 400 copies/ml during the first year of follow-up. However, the lowest proximal CD4 count of any patient experiencing an event was 360 cells/mm³ and the median proximal CD4 count was 609 cells/mm³ for AIDS-defining events and 678 cells/mm³ for non-AIDS-defining events.

“Crucially,” noted Dr Chilton, “no patient had a CD4 count below 350 at the time of a clinical event, so measuring CD4 counts would not have helped predict those events.”

The investigators used multivariate analysis to look for predictors of events in individuals with a viral load below 400 copies/ml. However, the numbers of events were small and, consequently, they found no clear predictors of clinical events based on age, previous AIDS diagnosis, nadir or latest CD4 count, or viral load.

In contrast, using the same parameters, multivariate analysis found that older age, and having a prior AIDS diagnosis were associated with clinical events individuals with a viral load above 400 copies/ml.

The investigators concluded that:

• Patients with undetectable viral loads and CD4 counts above 350 cells/mm³ have low rates of CD4 decline.
• The time spent with CD4 counts below 350 cells/mm³ is transient.
• In those patients with clinical events, proximal CD4 counts were not below 350 cells/mm³.
• There were no clear predictors of clinical events in multivariate analysis

“These are all really reassuring data,” Dr Chilton said, although there were some caveats to their study. This randomised, controlled study may not reflect clinical practice; follow-up time was short; there were a small number of events; and if CD4 counts are not measured as often in stable patients there will be a loss of data for future analyses.

“Nonetheless,” she continued, “we do support less frequent CD4 monitoring in those who maintain optimal viral suppression and a CD4 count above 350.” She said that annual CD4 monitoring was now recommended and that this was already being put into practice at her large central London HIV clinic.

During the question and answer sessions that followed, Dr Chilton said that annual CD4 counts had become acceptable to most patients once they had been reassured.

Speaking from the audience, Garry Brough, Mortimer Market’s Patient Representative, added that the strategy had worked following a consultation with patients. “What the clinic considered cost saving,” he said, “we also considered time saving.”

The result, he said was that in clinically stable patients with undetectable viral loads and CD4 counts above 350 cells/mm³ viral load testing was now done every four months, and CD4 counts done annually, resulting in just three visits a year for the patient, and considerable time and cost-saving for the clinic.