The majority of people with HIV who control the virus without treatment appear likely to maintain that state for many years, follow-up of a French cohort of ‘HIV controllers’ shows.
The study of 302 people found that only 30% needed to start antiretroviral treatment during a median follow-up period of five-and-a-half years. Around one in six subsequently stopped treatment for periods ranging from two months to over three years. In almost all cases, those who stopped treatment maintained a viral load below 2,000 copies throughout their time off treatment.
People who can control HIV without antiretroviral treatment are extremely rare. Research suggests that somewhere between one in a hundred and one in two hundred people with HIV maintain an undetectable viral load and a high CD4 count for a long period after acquiring HIV.
HIV controllers are of great interest to HIV cure researchers for the clues that their immune systems can provide about mechanisms that ensure HIV control without treatment. They can also provide information about the long-term consequences of HIV control without treatment, especially the effects of immune activation caused by a low level of HIV replication on the risk of co-morbidities.
French researchers have been following a cohort of people with HIV recruited since 2007 who have controlled HIV without treatment. To be admitted into the CODEX cohort, people must have been diagnosed with HIV for at least five years, have never taken antiretroviral treatment and have maintained a viral load below 400 copies/ml on the last five occasions.
Léo Plaçais of the Hôpital Bicêtre and colleagues wanted to understand what proportion of HIV controllers eventually needed antiretroviral treatment, why they started treatment and the immunological outcomes of treatment.
The cohort consists of 302 people with HIV who have been followed for a median of five-and-a-half years, between 2007 and 2019. Participants had been living with HIV for a median of 14.8 years after diagnosis. One quarter of the cohort had maintained an undetectable viral load (below 50 copies/ml) since diagnosis (a group sometimes referred to as elite controllers). The remainder had experienced viral load blips below 2,000 copies/ml but had otherwise maintained a low or undetectable viral load (sometimes referred to as viraemic controllers).
HIV controllers who had maintained a consistently undetectable viral load were more likely to be women (1.8 women to each man).
During the follow-up period, 30% of participants (90 people) started antiretroviral treatment. Those who started treatment had been living with HIV longer (17.5 vs 13.9 years, p<0.001), were more likely to have had viral blips and were less likely to have the HLA B57/58 haplotype associated with HIV controllers. Haplotypes are inherited DNA variations that may confer protection against or vulnerability to diseases.
Of those who started treatment, 47% did so due to a confirmed or suspected fall in CD4 count of at least 200 cells, or below 350 cells, 26% due to confirmed or suspected loss of virological control (viral rebound above 2,000 copies/ml) and 13% (12 cases) because of non-AIDS-defining events including hepatitis B or C, shingles and cardiovascular disease. Other reasons included pregnancy (4 cases) and because of concern about possible transmission to an HIV-negative partner (3 cases).
CD4 count declines were associated with development of AIDS-defining and non-AIDS-defining cancers in three people. These cancers (one case of Kaposi’s sarcoma, one tonsillar cancer and one metastatic pancreatic cancer) were diagnosed either before or just after starting treatment, and the study authors say that development of cancer may upset immunological control of HIV. Another participant developed non-Hodgkin lymphoma shortly after starting treatment due to loss of virological control.
Although treatment did not have a signficant effect on CD4 cell counts or immune activation on people who had not experienced viral blips before starting treatment, those with a history of viral blips did experience a significant reduction in CD8 counts and the proportions of activated CD4 and CD8 cells.
Thirteen people chose to stop antiretroviral treatment, due to adverse events or personal decision, after a median of 64 days. However, 11 out of 13 resumed treatment after a median of 245 days (just under nine months). The time off treatment varied considerably; three people remained off treatment for at least three years, and one had not resumed treatment after three years of follow-up.
In all but one, viral load remained below 2,000 copies/ml during the time off treatment, and insofar as reasons for resumption of treatment were reported, a low CD4 count appears to have prompted resumption. However, CD4 counts either stabilised or increased in three-quarters of those who stopped treatment, say the study authors.
“Temporary ART could be a therapeutic option” for HIV controllers, say the study authors, but this strategy needs to be studied in a prospective trial.
Plaçais L et al. Antiretroviral therapy for HIV controllers: reasons for initiation and outcomes in the French ANRS-CO21 CODEX cohort. EClinical Medicine, 37: 100963, 2021 (open access).