HIV Remission News From IAS 2017
The 9th International AIDS Society Conference on HIV Science (IAS 2017) began on Sunday in Paris, and the topic of HIV remission has been the focus of several high profile presentations.
One of the first major news stories to emerge from the meeting involves a newly described example of prolonged HIV remission in a nine-year-old South African (see the conference abstract and poster discussion slides presented today by Avy Violari). The case was identified because the child was a participant in the Children with HIV Early Antiretroviral Therapy (CHER) trial, which took place from 2005-2011 in South Africa and evaluated immediate, time-limited (40- or 96-week) courses of antiretroviral therapy (ART) in perinatally infected infants compared to the strategy of deferring initiation until clinical or immunological signs of disease progression were evident (preliminary results from the study, demonstrating the benefits of immediate ART, were published in the New England Journal of Medicine in 2008).
The child was diagnosed by HIV DNA PCR 32 days after birth, and displayed pre-treatment viral load levels of over 750,000 copies/ml and 151,000 copies/ml on days 39 and 60, respectively. ART was initiated after 61 days, with viral load declining to below 50 copies/ml at week 24 of on-treatment follow up. ART was subsequently stopped at week 40. The CHER protocol mandated restarting ART if certain immunological and/or clinical criteria were met, but the child has remained healthy and maintained a high CD4 count throughout follow up.
When stored samples were later accessed in order to measure viral load, the levels were found to be persistently below 20 copies/ml. The first post-interruption sample was taken eight weeks after stopping ART, so it's not known if viral load failed to rebound or if there was a transient increase that was rapidly controlled. Viral load has now remained undetectable for 8.75 years and counting, making this one of the longest reported cases of HIV remission (there is a French teenager first reported in 2015 and described in a paper in The Lancet HIV last year who has contained viral load to undetectable levels for over 12 years, after receiving a more prolonged course of ART in early life).
The researchers have conducted multiple analyses to try and gain insight into the factors that may have contributed to the outcome. Measurement of the HIV reservoir using a sensitive test for total HIV DNA revealed similar levels after one year and at 9.5 years of age: around 5 copies per million peripheral blood mononuclear cells (PBMC). However, no replication-competent virus could be detected using two different culture methods. Virus sequencing showed that the child was infected with HIV-1 from clade C, the most prevalent strain in South Africa.
The ELISA HIV antibody test was negative but weak reactivity to Gag p40 and p24 was detectable by Western blot. One exception to the generally weak HIV-specific antibody responses was a high IgA2 (a type of mucosal antibody) response to the gp41 protein. In studies of cellular immunity, a low-level CD4 T cell response to the HIV Gag protein was discernible with a whole blood intracellular cytokine assay, but no HIV-specific CD8 T cells could be detected.
Additional parameters that were investigated included the density of CCR5 molecules on CD4 T cells, which was found to be lower than HIV-negative controls -- it is perhaps possible that lower CCR5 density contributed to the maintenance of an HIV-specific CD4 T cell response by reducing the susceptibility of these cells to HIV infection. Encouragingly, levels of immune activation were similar to HIV-negative individuals and lower than those observed in elite controllers (in the latter group, elevated immune activation has been associated with disease progression despite persistently low viral loads).
PD-1 expression was found to be high on both CD4 and CD8 cells compared to HIV- individuals and elite controllers; the significance of this finding remains to be elucidated. None of the beneficial HLA alleles that have been associated with elite control were present, although the child was heterozygous at all HLA loci and this has been linked to a better prognosis compared to homozygosity.
The researchers note that other factors beyond the temporary course of ART likely contributed to the salutary outcome -- the majority of participants in both the 40- and 96-week immediate ART arms of the CHER trial had met the criteria for restarting treatment by the time the study ended, and no other cases of HIV remission have yet been identified.
The hope is that additional analyses will uncover novel correlates of immune control and aid the design of interventions aiming to induce HIV remission in larger numbers of HIV-positive individuals.
The news coverage of the case -- which has been extensive -- appears to have generally been accurate, although the BBC erred by describing the child as "virtually cured" in their headline -- it is not known if viral load might rebound at some point in the future, as has occurred in some other examples of HIV remission. There are articles describing the child as either a girl or a boy, but Amy Green from the South African Health News Service interviewed Avy Violari and reports that gender is not being disclosed in order to protect the individual's privacy.
The New York Times article, authored by Donald McNeil Jr., has some poor mistakes: the opening paragraph erroneously states that the ART was given at "high doses" when the CHER trial employed normal pediatric dosing. Later in the piece, individuals who are homozygous for the CCR5delta32 mutation, which confers a high degree of resistance to HIV infection, are mistakenly referred to as elite controllers (elite controllers are HIV-positive individuals who naturally contain HIV viral load to low levels in the absence of ART). Timothy Brown, the one individual considered cured of HIV infection, is said to have received a bone marrow transplant from an elite controller, which is not the case - his bone marrow donor was HIV-negative and homozygous for the CCR5delta32 mutation. One of these errors has since been corrected, but two remain at the time of writing.
The first mention of the South African HIV remission case at the IAS conference came yesterday in a presentation by Anthony Fauci, Director of the National Institute of Allergy and Infectious Diseases (NIAID) during the "Challenges and Opportunities in HIV Science" session -- this appears to explain the timing of the NIAID news release about the study. Fauci's talk was titled: "Sustained ART-free HIV remission: opportunities and obstacles" and he offered some additional nuggets of news, as well as discussing issues to be considered when evaluating the remission cases that have been reported to date (slides are available for download, and video was posted to the IAS conference youtube channel on July 26).
In particular, he outlined two potential contributors to HIV remission: in some cases, the size of the HIV reservoir may be so small that after an ART interruption the few latently infected CD4 T cells that are present remain dormant for an extended period, before one or more cells is activated and starts producing HIV (triggering viral load rebound). Fauci referred to this phenomenon as "stochastic reactivation," and evidence suggests it may explain the period of remission documented in the Mississippi baby who was started on ART prior to developing detectable immune responses to HIV (it has also been posited to explain the adult examples of the Boston patients and an individual treated within days of infection at UCSF -- see below).
In other instances -- such as the South African child, the French teenager and the VISCONTI cohort participants -- evidence suggests immunological control of HIV replication is contributing, although the exact mechanisms are unclear.
Fauci noted that both of these phenomena could also be operational simultaneously, but presently the short answer as to precisely what explains a given case of HIV remission is: "we don't know."
On the topic of interventions aiming to induce HIV remission, Fauci offered a glimpse of unpublished data from two studies. The first was a clinical trial conducted at the National Institutes of Health (NIH) Clinical Center that evaluated the effect of therapeutic vaccination in HIV-positive individuals started on ART within 12 weeks of a diagnosis of acute or early infection, who had maintained undetectable viral loads for over a year.
A total of 30 participants enrolled, with 15 randomized to receive a prime-boost HIV vaccine regimen developed by Profectus Biosciences (comprising DNA and vesicular stomatitis virus vectors) and 15 randomized to receive placebo immunizations. The final vaccination was administered at week 48 of the study and, eight weeks later, all participants underwent a 16-week analytical ART interruption.
The results demonstrated that there were no significant differences in time to viral load rebound to over 40 copies/ml (or to over 400 copies/ml) between the vaccine and placebo groups. Fauci emphasized the importance of the placebo group by showing that at the end of the ATI period, 20% of controls had viral loads less than 20 copies/ml, 27% were below 400 copies/ml, and 40% were below 2,000 copies/ml.
This finding may temper excitement about the single-arm study of therapeutic vaccination plus romidepsin that drew so much attention at CROI earlier this year -- in that trial, the fact that 38% of participants maintained viral loads below 2,000 copies/ml at 4-6 months after an ATI was presented as evidence that the interventions had enhanced control of HIV replication, compared to historical studies of ART alone. Fauci suggested that preliminary clinical trials and animal studies using broadly neutralizing antibodies (bNAbs) offer evidence that they may have more potential for enhancing post-treatment control of viral load than therapeutic vaccines.
Fauci's own laboratory has reported that administering an antibody against the α4β7 integrin to SIV-infected macaques led to surprisingly robust control of SIV replication after an ART interruption, and in his presentation he showed the unpublished results of recent experiments aiming to shed light on the mechanisms involved. The approach that the researchers took was to deplete different types of immune cells in the animals controlling SIV viral load, then assess whether this led to an increase in viral replication. The experiments compared:
- Antibodies to the CD8 receptor alpha chain, which deplete CD8 T cells, natural killer T cells (NKTs) and natural killer (NK) cells
- Antibodies to the CD8 receptor beta chain, which deplete CD8 T cells
- Antibodies to CD20, which deplete B cells
A transient rebound in viral load was only seen in recipients of antibodies to the CD8 receptor alpha chain, indicating that NKTs and NK cells are making an important contribution to the observed control of SIV replication. A clinical trial investigating the anti-HIV effects of the anti-α4β7 integrin antibody vedolizumab, which is FDA-approved for the treatment of ulcerative colitis and Crohn's disease, is ongoing at the NIH Clinical Center, and another is due to start soon at the Ottawa Hospital Research Institute in Canada.
Timothy Henrich from UCSF gave another presentation involving HIV remission today, in the same poster discussion session during which Avy Violari described the South African case (video of most of the session is posted on the IAS youtube channel, but Violari's talk is not included). Henrich reported details relating to an individual who was mentioned several times during CROI earlier this year (see the BSVC blog conference report). Both Henrich's abstract and slide presentation are available on the IAS conference website.
The individual in question was diagnosed within an estimated 10 days after acquiring HIV infection, because the exposure occurred in a short window between screening for a pre-exposure prophylaxis (PrEP) program and the baseline visit at which Truvada PrEP was first administered. When the baseline test results came back showing the presence of HIV, Truvada was quickly switched to a four-drug ART regimen including Truvada, darunavir and raltegravir. No HIV could be detected in blood or tissue samples during ART, although a low level of viremia was briefly detected in one out of 10 humanized mice administered a large volume of cells sampled from the individual (this novel test is known as the murine viral outgrowth assay).
After 34 months of continuous ART, an analytical treatment interruption was undertaken, and no viral load rebound occurred for 224 days, during which time no HIV DNA or RNA was detectable in plasma or sampled CD4 T cells. After this prolonged period of HIV remission, viral load rebounded and was initially detected at a level of 36 copies/ml. Six days later viral load had increased to 59,805 copies/ml and ART was immediately restarted. Sequencing of the HIV genome demonstrated that the rebounding virus was identical to that sampled at the time of acute infection.
In a collaboration with Alison Hill, mathematical modeling studies estimated the size of the HIV reservoir to have been approximately 200 latently infected cells prior to the ART interruption. Based on Hill's prior work, a latent HIV reservoir of this size confers only a small (~1%) chance of achieving a lifelong cure, due to the risk of stochastic reactivation of one or more of the latently infected cells.
Henrich and colleagues continue to evaluate enrollees in PrEP projects for evidence of recent HIV infection, and he showed a slide documenting that another individual has been started on ART very early as a result of the effort. Follow up of this second individual is ongoing.
In addition to several sessions and presentations related to HIV cure research at the main conference (see here for a roadmap), IAS hosted a pre-conference HIV Cure & Cancer Forum at the Institut Curie. The abstracts and many of the presentations have already been made available online, and a brief report from the meeting will follow in a separate blog post.