No differences in clinical outcomes with the addition of viral load testing to CD4 cell count monitoring among HIV infected participants receiving ART in rural Uganda: Long-term results from the Home Based AIDS Care Project

In this extension of the HBAC study as a two-arm trial, we found no statistically significant differences in clinical outcomes associated with the addition of quarterly VL monitoring to quarterly CD4 cell count monitoring after over 5 years of follow-up. Furthermore, we did not find any differences in terms of the proportion of participants with unsuppressed VL or rate of switching to second-line therapy between these two strategies. Our analysis again suggests that the addition of VL monitoring to CD4 cell count monitoring may not result in improved clinical outcomes for HIV positive patients receiving ART in resource limited settings. This conclusion is the same as that of the original HBAC study and the only other direct comparison of VL and CD4 cell count monitoring, another RCT conducted in Thailand [12, 17]. The latter study reported that a CD4 switching strategy was non-inferior in terms of clinical outcomes among HIV-positive adults, 3 years after beginning ART when compared to a VL -based switching strategy [12]. The authors found that there was also no difference between the strategies in terms of virologic suppression and immune restoration. Importantly, however, even though patients in the CD4 arm spent longer with a high viral load than patients in the VL arm, the emergence of HIV mutants resistant to antiretroviral drugs was similar in the two arms [12]. Unfortunately, we do not have any resistance data in order to make comparisons in this regard.

These findings differ somewhat from the results of an analysis of mortality of patients on ART in Southern Africa from the International epidemiological Databases to Evaluate AIDS in Southern Africa (IeDEA-SA) [8]. Participants from programs which did not have access to VL testing, namely those in Zambia and Malawi reported higher rates of death and loss to follow up, in comparison to participants from South Africa where VL measurement was accessible and readily available. However, it is unlikely that the only differences between these programs related to the provision of VL testing and differences in health care systems and living environments of these patients likely also influenced the differences in outcomes observed. Studies which compared the effect of routine VL testing to the standard of care where VL was used sparingly to adjudicate discrepancies between CD4 and clinical assessments, found that VL monitoring did not reduce death over the first 36 months of ART but did result in earlier ART regimen change [8, 9, 18].

A similar exploratory study by AIDS Clinical Trials Group A5115 that followed up participants for three years and compared a treatment switching strategy based on CD4-only monitoring versus VL thresholds in 21 public hospitals throughout Thailand reported no significant differences in activated or total CD4 cells at study end [19, 20]. Despite the lack of evidence of clinical benefit to support the use of routine VL testing, there may be other reasons to promote increase use of VL testing. Routine monitoring of participants with VL may result in reduction in the time a patient takes a failing regimen and potentially reducing the frequency of developing drug resistant mutations [21]. However, to date, there is very little evidence that the drug resistance mutations which develop while patients are failing their first-line regimens have much effect on the success second-line therapy. A study from Malawi found that virologic responses to a second line regimen among 109 participants with immunologically-defined treatment failure and a measured VL ≥1000 copies/mL was quite good (85 % VL < 400 copies/mL among those with VL measurements at 12 months after switching), although mortality was quite high at 9 %. All patients in this study had viruses with at least one resistance mutation, and 56 % of patients had viruses with thymidine analogue mutations, but the authors did not find an association with these mutations and virologic suppression at one-year after treatment switching [22]. Furthermore, the Thai RCT described above, did not find differences in the accumulation of virologics resistance mutations. More evidence from larger studies are needed to determine whether virologic monitoring can improve outcomes for individuals diagnosed with treatment failure in resource-limited settings. In the interim, designing HIV programmes that maximize retention of patients in the continuum of care and support adherence counselling to treatment should remain the focus of HIV treatment programmes. [23‐25] Many programmes in Sub-Saharan Africa have reported a loss to follow up among patients on ART of 20 % or more suggesting potential for improvement [26, 27].

This study has a number of limitations; firstly, the generalizability of our study findings to routine care settings may be limited as participants in this trial were seen and counseled more frequently than is routine in most settings. In the first phase of the HBAC study, participants received weekly home delivery of ART and clinical monitoring by field officers. However, in this phase of the study we extended the interval between home visits to once every 2 months over a 4 month period, in order to reflect standard care models. The intensity of the follow-up likely contributed to the low overall rates of virologic failure and loss to follow-up in comparison to those reported in most other settings. It is also important to note that laboratory evaluations were performed every 3 months, rather than every 6 months that is recommended by WHO. Furthermore, the rates of virological failure in our study were generally lower than most reported programmes from the region, as surveyed in a recent systematic review [27‐30].