Treatment of primary HIV infection: new data for a new era

Initiation of antiretroviral therapy shortly after the acquisition of HIV infection was first suggested to be beneficial in 1995, when a 6-month course of zidovudine monotherapy was shown to decrease the prevalence of minor opportunistic infections over a 15-month period of observation [1]. The benefit was thought to relate to the preservation of a favorable antiviral immune response if virologic suppression is achieved and maintained while the immune system is still largely intact [2]. This hypothesis led to the conduct of a number of clinical trials to test the hypothesis that a time-limited course of therapy initiated within 12 months of the infection event could lead to functional control of HIV replication without ongoing treatment being required. We conducted one such study, in which patients were randomized to a 12-month course of antiretroviral therapy (followed by its discontinuation, with reinitiation as clinically indicated) or observation. There was no significant difference between the study arms, with permanent life-long therapy being indicated in all participants, on average within 18 months [3]. Two other studies essentially replicated these findings [4,5], confirming that, in the current model of antiretroviral therapy, once it is initiated, it will be for life. If it is ever stopped, virologic rebound and immunologic disease progression inevitably occur, and therapy must be reintroduced. In fact, the SMART study of strategic treatment interruptions clearly showed that uncontrolled HIV replication for any length of time is associated with a higher risk of clinical illness and death, even with a relatively preserved CD4+ cell count [6]. At this time, the permanent initiation of effective antiretroviral therapy is indicated for all individuals with confirmed HIV infection and a detectable plasma viral load, regardless of when infection occurred [7]. This clearly begs the question as to whether any advantage remains to starting treatment during primary HIV infection, or whether anything can be gained from doing so. In the current issue of AIDS[8], Abdi et al. report on a unique cohort of 20 individuals who initiated antiretroviral therapy within 5 days of a diagnosis of primary HIV infection. Therapy was highly effective, with complete virologic suppression being achieved within, on average, 95 days and maintained for 6–7 years. No blips in viral load of 20 copies/ml or greater were ever documented. However, in 13/20 patients, lower level detectable viral load measures were documented on at least one occasion. Total cell-associated DNA was remarkably low, below the limit of detection of the proviral load assay in 14/20 participants at the end of follow-up. Phylogenetic analyses showed remarkably little viral evolution over several years. However, by ultra-deep sequencing, the evolution of drug resistance associated mutations was detected in nine cases. This is, to be sure, a unique cohort. The authors are to be congratulated for finding as many as 20 individuals shortly after acquiring HIV infection, getting them on treatment within days and maintaining them in follow-up for several years with an optimal therapeutic response at all time points. This strategy seems to be associated with a reduced viral reservoir (as measured by peripheral DNA load, at least) and little viral evolution. The initial viral inoculum is known to be more homogeneous [9] and the strategy used in this case series seems to preserve this homogeneity, at least in the medium term. However, there is ongoing viral replication, albeit at a reduced level. Most patients had detectable viral load measures at one or more points in time. It is likely that such blips could have been documented in all cases had more measures been taken. Resistance mutations evolved as a result of drug pressure in almost half of the study cohort. This suggests that viral replication is, indeed, occurring at a low level. This is probably occurring in various tissue reservoirs in which there is insufficient drug penetration, despite the early and effective initiation of antiretroviral therapy. What do these new data teach us about treating primary HIV infection in this new era? Although it is difficult to identify such patients in large numbers, this study supports an immediate offer of treatment with a high likelihood of rapid virologic suppression and some reduction (as compared with initiation of therapy during chronic infection) in viral reservoirs and viral heterogeneity. It may be that this establishes a setting where immunologic interventions (including therapeutic vaccines) may be more effective, with less diverse isolates. It will be interesting to maintain cohorts of patients successfully treated during primary infection to test this hypothesis. The SMART study taught us that any period of uncontrolled viral replication may have deleterious effects. It is interesting to speculate that among individuals in whom viral replication is minimized over the lifetime of infection, premature aging and morbidity associated with HIV infection [10] may be reduced. If we can maintain cohorts of patients successfully treated during primary infection in long-term follow-up, this could finally help us demonstrate a tangible benefit of initiation of antiretroviral therapy in this setting when it is possible. Participants enrolled in some of the clinical trials conducted over 10–15 years ago may already provide information in this regard. The first studies of treatment of primary HIV infection were conducted over a quarter of a century ago. We have now learned that all patients, including those who are fortunate enough to be diagnosed shortly after being infected, should be offered therapy without delay. It may be that recently infected patients will benefit from such therapy in unique ways that should be the subject of ongoing observation and research. Acknowledgements Conflicts of interest There are no conflicts of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.