Abstract
A variety of interventions to induce a functional cure of HIV are being explored, with the aim being to allow patients to cease antiretroviral therapy (ART) for prolonged periods of time or for life. These interventions share the goal of inducing ART-free remission from HIV pathogenesis and disease progression but achieve this in quite different ways, by reducing the size of the latent reservoir (for example, small-molecule stimulation of latently infected cells), reducing the number of target cells available for the virus (for example, gene therapy) or improving immune responses (for example, active or passive immunotherapy). Here, we consider a number of these alternative strategies for inducing post-treatment control of HIV and use mathematical modelling to predict the scale of the challenge inherent in these different approaches. For many approaches, over 99.9% efficacy will likely be required to induce durable ART-free remissions. The efficacy of individual approaches is currently far below what we predict will be necessary, and new technologies to achieve lifelong functional cure are needed.
Similar content being viewed by others
Introduction
Current antiretroviral therapy (ART) is highly effective. It suppresses plasma viral RNA to below the standard level of detection for prolonged periods of time and halts viral evolution. However, in most patients, the virus rapidly rebounds within weeks of ART interruption1,2 (Fig. 1a). Several therapeutic approaches that aim to prevent or delay viral rebound after treatment interruption, producing a post-treatment remission or functional cure of HIV, are being investigated. These would allow subjects to cease ART for prolonged periods, reducing side effects, cost, the emergence of drug resistance and stigmatization. Current interventions primarily focus on driving the reactivation of latent virus, the targeting and destruction of highly infected cell subsets, or the use of gene therapy to either target the latent provirus or make cells refractory to viral replication. In addition, immunotherapy and vaccination are being explored to prevent HIV reactivation or to reduce post-reactivation viraemia and keep it at very low levels (Fig. 1b–d). Inherent to each approach is the requirement to remove a sizeable proportion of latently infected cells or control the virus at a particular level. Achieving some level of remission seems feasible, as there are a number of case studies in the literature of patients who have experienced prolonged post-treatment remission or viral control3,4,5. However, in almost all these cases, the virus rebounded in a shorter time frame than would be acceptable as part of a functional cure for HIV. Therefore, it is important to consider what level of efficacy would be required from each treatment methodology in order to achieve an acceptable functional cure. Here, we use a mathematical modelling framework to analyse and compare the predicted impacts of different potential interventions (outlined in Supplementary methods). For each intervention, we estimate the magnitude of the therapeutic effect that would be required to produce a functional cure of HIV of sufficient scale and durability to contribute to global goals for reducing HIV transmission and disease.
Shrinking the latent reservoir
One approach to producing prolonged post-treatment control is to either eliminate or shrink the HIV reservoir down to a size where successful viral reactivation from latency occurs rarely, if ever6. This approach appears to be the mechanism of delayed viral rebound observed in the so-called Boston patients, who underwent bone marrow transplantation while on ART. The pretransplant conditioning removed the majority of T cells (including latently infected cells) and replaced them with (uninfected) donor cells3. However, a small residual HIV reservoir persisted, leading to a very delayed production of replication-competent virus and viral rebound after only 3–7 months of remission3. A similar phenomenon was observed in the so-called Mississippi baby case, in which it is thought that very early treatment and a very low initial viral reservoir size allowed a prolonged period of nearly 2 years of treatment-free remission before viral reactivation led to rapid viral replication4. Extremely early treatment of adult infection has also been associated with a very small reservoir size; however, it correlates with a much more modest prolongation of ART-free remission of only days or weeks7,8.
These case studies suggest that shrinking the reservoir can induce prolonged remission. But how much of a reduction in reservoir size is required to achieve a clinically useful period of remission (see Box 1)? Successful HIV reactivation from latency after treatment interruption is estimated to occur with a frequency of approximately once a week in most patients treated during chronic infection1. This estimation gives us a scale against which to assess how much we need to reduce the normal frequency of HIV reactivation to produce a post-treatment remission of different duration6. For example, a 20-fold reduction in the frequency of reactivation should result in an average remission duration of 20 weeks (similar to that observed with bone marrow transplantation3); a 100-fold reduction should result in a 2-year remission (similar to that seen in the Mississippi baby case report4); and a 1,000-fold reduction would be required for an average remission duration of 20 years. Thus, a reduction of somewhere in the order of 1,000-fold in the frequency of reactivation would be necessary to produce a meaningful period of remission, in which the threat of viral rebound, immunodeficiency and possible onward transmission would not occur in the majority of patients for many years (discussed in more detail in ref.6). Importantly, the key metric we discuss here is the frequency of reactivation, not the number of latently infected cells per se (see Box 1). For approaches aimed at reducing the size of the reservoir, an average 20-year remission equates to at least an average 99.9% reduction in the frequency of reactivation.
Inducing activation and death of latent virus
One approach to reducing the frequency of HIV reactivation is to induce the activation of latent virus, leading to the spontaneous or immune-mediated removal of infected cells9. To date, the effects of short-term treatment have been modest10,11,12, with modelling suggesting that application of a single treatment or a combination of a few current-generation treatments may at best remove a few per cent of the reservoir13. Although it seems unlikely that a single treatment could reduce the frequency by >99% (as would be required), multiple rounds of treatment could have a more profound effect over time, assuming that re-treatments cumulatively reduce the frequency of reactivation (Fig. 2a,b). Thus, for example, a treatment that was capable of reducing the reactivation frequency by 50% upon each treatment could potentially be administered twice to reduce the frequency to one-quarter of its original rate, once again to reduce it to one-eighth, and so on. Such a treatment would need to be administered 10 times to reduce the reactivation frequency 1,000-fold, thus achieving the average 20-year remission discussed above. If, however, the effect of the treatment was more modest, with a 20% reduction per treatment cycle, this would mean that the treatment would need to be given ~30 times, while an intervention that reduced reactivation by 10% per cycle would need to be administered ~65 times. It quickly becomes clear that not only the potency but also the tolerability of the treatment is important.
Another major consideration is whether an intervention continues to retain its efficacy over multiple rounds of treatment. Indeed, if we have a treatment whose efficacy wanes by 5% with each round of treatment, and if the aim is to achieve a 99.9% reduction in reactivation in less than 100 rounds of treatment, this would be achievable only if the efficacy in the first round was at least 32%. Additionally, the maximum level of reduction in frequency ever achievable by the treatment is also limited. For example, if a treatment that began with an ability to reduce reactivation size by 30% became 5% less effective upon each subsequent treatment (that is, it achieves a 30% reduction in reactivation upon the first treatment, 28.5% upon the second, 27.08% upon the third, and so on), then we find that the maximum remission that could ever be achieved is 13.3 years (see Fig. 2c). Thus, three aspects are crucial to predicting the effectiveness of therapies aimed at reducing reservoir size and the frequency of HIV reactivation: potency, tolerability and sustainability of the treatment effect over multiple rounds.
Targeting cell subsets
A minimally toxic approach to reservoir reduction would be to selectively eliminate latently infected cells, leaving the uninfected cells intact. However, the silent nature of latent virus makes it difficult to identify these cells. A number of studies have compared the levels of provirus or reactivation-competent virus in different cell subsets and identified subsets that have higher levels of infection14. However, critically, it is not the increased level of infection in the targeted subset that is important but the proportion of total latent virus (or more specifically, total number of reactivations; see Box 1) that occurs within the subset. For example, if a subset is 10 times more infected than other cells but comprises only 5% of the total number of infected cells, then it still represents only approximately 35% of the total number of infected cells. Even if we could target and entirely remove this subset, only 35% of the latently infected cells would be removed, and a durable remission would not be achieved. We would still need to deal with the remaining 65% of latent virus that is resident in non-targeted cells. The frequency of reactivation in the non-targeted subsets would still need to be reduced by approximately 99.8% to achieve an overall 1,000-fold reduction in reservoir size. Even complete eradication of a subset can have a relatively small benefit in reducing the overall frequency of reactivation or inducing notable remission unless this is also combined with a concomitant reduction in the frequency of reactivation in the general (non-subset) HIV reservoir.
Targeted gene silencing
A prerequisite for latent cell reactivation is the presence of an intact HIV genome, direct targeting of these genomes is one avenue to reducing the frequency of reactivation. The use of CRISPR–Cas9-based gene therapy to cleave provirus and thus eliminate latent HIV or to silence HIV reactivation has been suggested. Currently, these technologies can reduce viral replication by 10–20-fold in vitro and by 10-fold in vivo in animal models15,16,17,18. However, the problem of scale is also an issue here. By extension from the analysis above, if, for example, a gene therapy aims to enter and cleave or silence HIV proviral DNA, it must aim to enter 99.9% of infected cells to silence a meaningful proportion of the reservoir (Fig. 3a,b). This level of gene delivery is currently not achievable. For example, ex vivo transduction of human CD4+ T cells with an adeno-associated virus encoding zinc-finger nucleases targeted at CCR5, a co-receptor used by HIV for cell entry, achieved efficacy rates of less than 30%19. While higher levels of transduction of CD34+ haematopoietic stem cells have been reported (>50%), there are issues with achieving efficient engraftment of substantial numbers of these cells.
Removing susceptible HIV targets
Another example of prolonged antiretroviral-free remission from HIV infection is the Berlin patient20. In this case study, the patient received a bone marrow transplant with CCR5-deficient bone marrow. Thus, the transplant conditioning acted to remove latently infected recipient cells (reducing the size of the reservoir), and the transplant itself replaced susceptible recipient (CCR5+) CD4+ T cells with resistant (CCR5−) donor cells, thus reducing the ability of the virus to replicate. It is thought that transplantation with CCR5− bone marrow effectively removes susceptible CD4+ T cells, and so even if the virus reactivates from latency, it has very few susceptible CCR5+ CD4+ T cells in which to replicate. This transplantation has provided more durable control than a bone marrow transplant alone, as no rebound has been observed in more than 10 years.
This example raises an alternative use for gene therapy: if the HIV-suppressive effect of gene therapy is permanent (making the transduced cells and subsequent daughter cells refractory to future infection), this may not only reduce the frequency of reactivation but also have the additional effect of removing susceptible target cells for future viral replication (Fig. 3c). Importantly, the removal of susceptible cells may actually present a greater benefit from gene therapy than the reduction in the viral reservoir. Modelling suggests that if it were possible to protect 50% of cells from infection (that is, 50% chimerism with gene-modified resistant cells), then set point viral loads would be reduced by approximately 60%21,22 (see Supplementary methods), which is consistent with anecdotal evidence in human and animal infection19,23,24. However, if 75% of cells were protected, then the viral load would be reduced to 14.3% of its normal level. Importantly, we can show that there is a minimum number of susceptible cells required for the virus to grow at all (once the number of target cells is below a threshold, each currently infected cell is able to infect only less than one additional cell, and the virus essentially peters out; Fig. 3c). Our modelling suggests that if >87.5% of cells were made refractory to infection, then the virus would be unable to grow successfully (assuming a basal reproductive ratio of HIV R0 = 8)22 (Fig. 3d).
Thus, three major factors to consider in any gene therapy approach are the transduction efficiency (proportion of cells that can be reached), the proportion of these cells that then engraft and are long lived or capable of generating progeny and to what extent the treatment makes cells durably resistant to future infection.
Immune control of HIV
Immune recognition of the virus or infected cells may also contribute to long-term, post-treatment HIV remission. This may occur in a variety of ways: immune recognition may help to shrink the viral reservoir, or immunity may be ‘reactivation-blocking’ in that it acts to block or control a proportion of reactivation events, reducing the frequency of successful HIV reactivation events and delaying the time to viral recrudescence. Alternatively, initial viral reactivation from latency may occur, but immunity may be ‘reactivation-controlling’, and viral replication may subsequently be controlled such that it remains at low levels. Studies of cohorts of patients who received ART very early in primary HIV infection have shown that a subset of patients exhibits varying degrees of post-treatment control25,26. Analyses of the viral dynamics after treatment interruption suggest that both reactivation-blocking and replication-controlling effects may have occurred in these cohorts, as most patients did show early evidence of viral recrudescence, followed by a period of immune control25,26. This control was attributed to an alteration in the viral–immune balance as a result of early treatment, in which host immunity may be induced in primary infection and is not impaired by ongoing viral replication if primary HIV infection is rapidly controlled by ART27,28.
However, other studies of patients treated early in infection have not always shown similar control of HIV replication29,30, perhaps because of the importance of the timing of initial treatment. If the treatment is initiated too early, the very small amount of antigen exposure means that little host immunity will be induced31,32. If the treatment is started too late, then host immunity may be exhausted33 or eliminated by the infection34,35, or viral escape mutation may have occurred36. As the majority of patients undergoing ART are not identified and treated in early infection, early initiation of ART will be of limited value for the immune control of HIV in most subjects.
Immune targeting of the reservoir
The latent reservoir of HIV is generally thought to be invisible to immune surveillance, as truly quiescent latent cells are not thought to express notable amounts of viral antigen. However, a proportion of latently infected cells may transiently express viral antigens periodically (as a result of cell division or activation), rendering them susceptible to immune killing by CD8+ T cells or antibody-dependent mechanisms. Immunization or passive immunotherapy during ART might induce responses capable of killing cells during these periods of antigen expression, allowing the elimination of a proportion of the reservoir each year (depending on the frequency of antigen expression)37,38 (Fig. 4a). If, for example, 10% of latently infected cells expressed antigen each year, and an effective immune response could remove these cells, then the reservoir could be reduced by 10% per year, and we would see a 90% reduction over 20 years (producing an average 10-week remission). However, if 50% of cells expressed antigen each year, then after 10 years the reservoir would be reduced to 0.1% of its starting value, achieving an average 20-year remission (Fig. 4b). It is clear that this requires both a fairly high proportion of cells to express antigen regularly and the efficient killing of these cells in order to contribute to notable remission.
Therapeutic vaccination to prevent HIV reactivation
Vaccination or immunotherapy during ART offers the potential of enhanced immune control during treatment interruption either to prevent viral recrudescence or to control the rebound virus and limit it to low levels. HIV reactivation from latency after treatment interruption might be thought of as similar to a serial low-dose challenge by a reactivating virus. Studies suggest that, on average, a latently infected cell successfully initiates viral recrudescence approximately once every week following treatment interruption1. Thus, if we think of vaccine efficacy as the proportion of these reactivation events that are successfully blocked, then we can estimate the expected delay in reactivation we would obtain for different levels of vaccine efficacy (Fig. 4c,d). Vaccination studies in macaques infected with simian immunodeficiency virus (SIV) demonstrated significant protection against serial challenge, blocking as much as 80% of challenges39. However, the most successful current HIV vaccine trial showed only approximately 30% efficacy against sexual transmission40. Passive immunotherapy in macaques with SIV has also been shown to prevent multiple low-dose challenges41,42, and passive immunotherapy during HIV treatment interruption has shown evidence for delays in the time to detection of a virus43,44. Figure 4e shows the relationship between vaccine efficacy (that is, the proportion of HIV challenges to HIV reactivations from latency that are blocked) and the predicted average duration of post-treatment remission. We see that even a vaccine that is capable of blocking 80% of reactivation events would still allow most patients to have a breakthrough reactivation event within 5 weeks. Figure 4e illustrates that vaccines or immunotherapies capable of blocking >99% of reactivation events would be required to produce clinically significant delays in the time to detection of a virus.
Immune control of HIV following viral recrudescence
An alternative strategy to achieve HIV remission after ART interruption may be to induce post-treatment immune control of HIV after the virus has reactivated. That is, even if the immune system is unable to block HIV recrudescence, it may still be able to control viral replication and keep it at low levels, as sometimes observed in patients treated early after infection25,26. It seems likely that if sufficient immune protection is present at the time of HIV reactivation, this may switch the host–viral balance in favour of host control28,44,45,46. Inducing such disease-modifying immunity has proved challenging in the context of prophylactic HIV vaccine development, in which the only successful prophylactic vaccine did not alter set point viral loads after infection40. However, this process may be easier in the context of therapeutic vaccination and treatment interruption, where the timing of treatment interruption can be manipulated to coincide with peak immunity or the presence of passive antibodies44,45,46. The effects of early ART on long-term viral control suggest that shifting the balance towards host control of viral replication is possible in this context26,47. Such post-recrudescence control offers the potential for long-term ART-free remission. However, it is not known whether such immune control can be reliably induced in patients with HIV treated during chronic infection or whether this control can be maintained for prolonged periods or might be undermined by factors such as immune escape47,48,49. Another major question is what level of viral control would be deemed sufficient. That is, at what level of ongoing viral replication would the trade-off between ongoing ART and the risks of disease progression and transmission balance out? Recent studies have suggested that even in patients with very low viral levels (>200 copies per ml), the risk of disease progression is increased compared with that in patients with an undetectable virus (<50 copies per ml)50. Transmission risk is thought to be eliminated at viral loads <200 copies per ml51,52. Together, this suggests that a decision to withhold ART would likely require post-treatment control of viral loads to low levels (at least <200 copies per ml).
Combinations and synergies
Given the diversity of strategies discussed above, combinations of two or more therapies clearly have the potential to increase the overall effect of treatment. An obvious opportunity for combination therapy lies in combining agents that act in a similar way to achieve greater potency (Fig. 5a, left panel). Thus, for example, treatments inducing the activation and death of latently infected cells, strategies targeting subsets of infected cells, gene therapy to remove provirus and inducing immunity to block HIV reactivation all act to reduce the frequency of reactivation, and we might expect their effects to be additive (Fig. 5). However, each agent in a combination would still require high efficacy to contribute to the overall reduction required. For example, for approaches targeting a subset of highly infected cells, the ability to reduce the frequency of reactivation by 50% might be regarded as a considerable achievement. Similarly, if gene therapy could achieve gene silencing of HIV in 50% of latently infected cells, or if the immune response could block 50% of reactivation events, these would be major successes for these approaches. However, these would take us only a short way towards the level of reduction in reactivation that is needed. For example, Fig. 2 illustrates that an activation-and-death treatment that reduces the reactivation frequency by 10% per cycle would need to be given more than 65 times to achieve a 20-year remission. If the same activation-and-death treatment was included in a combination strategy involving a subset-targeting therapy that removes 50% of the reservoir, gene silencing that prevents HIV reactivation in 50% of the reservoir and a vaccine inducing immunity that blocked 50% of HIV reactivations, this combination therapy would reduce the number of treatment cycles needed from 65 to 45 cycles (Fig. 5b), unless major synergies can be achieved.
Strategies to reduce viral replication after initial viral rebound, such as inducing immune control or making target cells refractory to infection, would be expected to synergize to reduce post-reactivation viral loads (Fig. 5a, right panel). What is less clear is the extent to which synergy might occur between the frequency-reducing and post-reactivation-control strategies. For example, the Boston patients and Mississippi baby cases involved long delays between treatment interruption and viral reactivation from latency (indicating a greatly reduced frequency of reactivation), but high levels of viral replication occurred following recrudescence, suggesting that a low frequency of reactivation does not predict subsequent viral control3,4. This makes sense if we realize that frequency-reducing strategies might delay the time at which the first post-treatment viral reactivation event occurs (acting as the first infectious challenge to the host) (Fig. 1b), whereas strategies for post-reactivation viral control determine the trajectory of replication after viral growth has been initiated (Fig. 1c).
Conclusion
The potential long-term benefits of an HIV cure or durable remission have driven the development of several strategies to reduce the frequency of HIV reactivation from latency or to control the virus after reactivation. These are currently at an early stage and have yet to demonstrate notable post-treatment remission. However, a number of case studies of post-treatment control following early treatment or transplantation provide encouragement that remission is possible, given the right circumstances. In prioritizing the development of different agents, it is useful to consider the magnitude of effect that will be required using these different approaches (Fig. 6). For example, therapies aimed at reducing reservoir size must have a combination of primary efficacy (upon the first round of administration), durability (over multiple rounds) and tolerability to have any potential to achieve meaningful remission. Similarly, strategies to target CD4+ T cell subsets are unlikely to succeed unless the subset contains the majority of infected cells. Strategies to induce post-rebound control of viral loads, either via making cells refractory to infection (via gene therapy) or by inducing host immune control, appear attractive. However, the mechanisms and potential of either gene therapy to produce refractory cells or boosted host immune control remain largely theoretical, and further work is required to demonstrate feasibility in vivo.
This review is based on the modelling of our current understanding of HIV reactivation and the effects of different therapeutic approaches. The baseline frequency of HIV reactivation from latency (that is, without any anti-latency interventions) in our Analysis was assumed to be on average once a week on the basis of a detailed analysis and the modelling of multiple clinical studies of patients undergoing treatment interruption1 (see Box 1). The estimate of the baseline frequency of reactivation we used is lower than the estimate used in some earlier modelling studies, which have assumed either continuous viral production or frequencies of reactivation as high as four times a day53,54. If the baseline frequency of reactivation was higher than once a week, then more effective anti-latency treatments or a greater number of treatment cycles would be required to produce the same length of remission. Further work is clearly required to better characterize the basic determinants of HIV recrudescence from latency and subsequent HIV replication, which will allow a refinement in predictions of the effects of different interventions.
The number and diversity of strategies being pursued to tackle HIV latency speak to the importance and potential of this approach. Clearly, the challenges are large in many cases, and the scale of effect needed seems beyond the reach of current interventions. Modelling of treatment efficacy, scheduling and combination therapy has been used to propose optimal treatment strategies in oncology and to maximize remission and cure55,56. Quantitative considerations of the scale of the problem and the strength of interventions required can provide a rational guide for the development of agents to enable post-treatment remission of HIV.
References
Pinkevych, M. et al. HIV reactivation from latency after treatment interruption occurs on average every 5–8 days—implications for HIV remission. PLOS Pathog. 11, e1005000 (2015).
Bloch, M. T. et al. The role of hydroxyurea in enhancing the virologic control achieved through structured treatment interruption in primary HIV infection: final results from a randomized clinical trial (Pulse). J. Acquir. Immune Def. Syndr. 42, 192–202 (2006).
Henrich, T. J. et al. Antiretroviral-rree HIV-1 remission and viral rebound after allogeneic stem cell transplantation. Ann. Intern. Med. 161, 319 (2014).
Luzuriaga, K. et al. Viremic relapse after HIV-1 remission in a perinatally infected child. N. Engl. J. Med. 372, 786–788 (2015).
Lisziewicz, J. et al. Control of HIV despite the discontinuation of antiretroviral therapy. N. Engl. J. Med. 340, 1683–1683 (1999).
Cromer, D. et al. Modeling of antilatency treatment in HIV: what is the optimal duration of antiretroviral therapy-free HIV remission? J. Virol. 91, e01395-17 (2017).
Henrich, T. J. et al. HIV-1 persistence following extremely early initiation of antiretroviral therapy (ART) during acute HIV-1 infection: an observational study. PLOS Med. 14, e1002417 (2017).
Colby, D. J. et al. Rapid HIV RNA rebound after antiretroviral treatment interruption in persons durably suppressed in Fiebig I acute HIV infection. Nat. Med. 24, 923–926 (2018).
Deeks, S. G. Shock and kill. Nature 487, 439 (2012).
Elliott, J. H. et al. Activation of HIV transcription with short-course vorinostat in HIV-infected patients on suppressive antiretroviral therapy. PLOS Pathog. 10, e1004473 (2014).
Sogaard, O. S. et al. The depsipeptide romidepsin reverses HIV-1 latency in vivo. PLOS Pathog. 11, e1005142 (2015).
Rasmussen, T. A. et al. Panobinostat, a histone deacetylase inhibitor, for latent- virus reactivation in HIV-infected patients on suppressive antiretroviral therapy: a phase 1/2, single group, clinical trial. Lancet HIV 1, e13–e21 (2014).
Petravic, J., Rasmussen, T. A., Lewin, S. R., Kent, S. J. & Davenport, M. P. Relationship between measures of HIV reactivation and decline of the latent reservoir under latency-reversing agents. J. Virol. 91, e02092-16 (2017).
Hiener, B. et al. Identification of genetically intact HIV-1 proviruses in specific CD4+ T cells from effectively treated participants. Cell Rep 21, 813–822 (2017).
Kaminski, R. et al. Excision of HIV-1 DNA by gene editing: a proof-of-concept in vivo study. Gene Ther. 23, 690–695 (2016).
Kaminski, R. et al. Elimination of HIV-1 genomes from human T-lymphoid cells by CRISPR/Cas9 gene editing. Sci. Rep. 6, 22555 (2016).
Wang, G., Zhao, N., Berkhout, B. & Das, A. T. CRISPR-Cas9 can inhibit HIV-1 replication but NHEJ repair facilitates virus escape. Mol. Ther. 24, 522–526 (2016).
Zhu, W. et al. The CRISPR/Cas9 system inactivates latent HIV-1 proviral DNA. Retrovirology 12, 22 (2015).
Tebas, P. et al. Gene editing of CCR5 in autologous CD4 T cells of persons infected with HIV. N. Engl. J. Med. 370, 901–910 (2014).
Hütter, G. et al. CCR5 targeted cell therapy for HIV and prevention of viral escape. Viruses 7, 4186–4203 (2015).
Nowak, M. & May, R. M. in Virus Dynamics: Mathematical Principles of Immunology and Virology (Oxford Univ. Press, 2000).
Ribeiro, R. M. et al. Estimation of the initial viral growth rate and basic reproductive number during acute HIV-1 infection. J. Virol. 84, 6096–6102 (2010).
Matrajt, L., Younan, P. M., Kiem, H. P. & Schiffer, J. T. The majority of CD4+ T-cell depletion during acute simian-human immunodeficiency virus SHIV89.6P infection occurs in uninfected cells. J. Virol. 88, 3202–3212 (2014).
Perez, E. E. et al. Establishment of HIV-1 resistance in CD4+ T cells by genome editing using zinc-finger nucleases. Nat. Biotechnol. 26, 808 (2008).
Sáez-Cirión, A. et al. Post-treatment HIV-1 controllers with a long-term virological remission after the interruption of early initiated antiretroviral therapy ANRS VISCONTI Study. PLOS Pathog. 9, e1003211 (2013).
Sneller, M. C. et al. A randomized controlled safety/efficacy trial of therapeutic vaccination in HIV-infected individuals who initiated antiretroviral therapy early in infection. Sci. Transl Med. 9, eaan8848 (2017).
Saez-Cirion, A., Jacquelin, B., Barré-Sinoussi, F. & Müller-Trutwin, M. Immune responses during spontaneous control of HIV and AIDS: what is the hope for a cure? Phil. Trans. R. Soc. Lond, B, Biol. Sci. 369, 20130436 (2014).
Conway, J. M. & Perelson, A. S. Post-treatment control of HIV infection. Proc. Natl Acad. Sci. USA 112, 5467–5472 (2015).
Gianella, S., Anderson, C. M., Richman, D. D., Smith, D. M. & Little, S. J. No evidence of post treatment control after early initiation of antiretroviral therapy in the San Diego primary infection cohort. AIDS 29, 2093–2097 (2015).
Calin, R. et al. Treatment interruption in chronically HIV-infected patients with an ultralow HIV reservoir. AIDS 30, 761–769 (2016).
de Souza, M. S. et al. Initiation of antiretroviral therapy during acute HIV-1 infection leads to a high rate of nonreactive HIV serology. Clin. Infect. Dis. 63, 555–561 (2016).
Dong, K. L. et al. Detection and treatment of Fiebig stage I HIV-1 infection in young at-risk women in South Africa: a prospective cohort study. Lancet HIV 5, e35–e44 (2018).
Day, C. L. et al. PD-1 expression on HIV-specific T cells is associated with T cell exhaustion and disease progression. Nature 443, 350 (2006).
Mattapallil, J. J. et al. Massive infection and loss of memory CD4+ T cells in multiple tissues during acute SIV infection. Nature 434, 1093–1097 (2005).
Douek, D. C. et al. HIV preferentially infects HIV-specific CD4+ T cells. Nature 417, 95–98 (2002).
Deng, K. et al. Broad CTL response is required to clear latent HIV-1 due to dominance of escape mutations. Nature 517, 381–385 (2015).
Borducchi, E. N. et al. Ad26/MVA therapeutic vaccination with TLR7 stimulation in SIV-infected rhesus monkeys. Nature 540, 284–287 (2016).
Barouch, D. H. et al. Therapeutic efficacy of potent neutralizing HIV-1-specific monoclonal antibodies in SHIV-infected rhesus monkeys. Nature 503, 224–228 (2014).
Burton, S. L. et al. Breakthrough of SIV strain smE660 challenge in SIV strain mac239-vaccinated rhesus macaques despite potent autologous neutralizing antibody responses. Proc. Natl Acad. Sci. USA 112, 10780–10785 (2015).
Rerks-Ngarm, S. et al. Vaccination with ALVAC and AIDSVAX to prevent HIV-1 infection in Thailand. N. Engl. J. Med. 361, 2209–2220 (2009).
Hessell, A. J. et al. Effective, low-titer antibody protection against low-dose repeated mucosal SHIV challenge in macaques. Nat. Med. 15, 951–954 (2009).
Gautam, R. et al. A single injection of crystallizable fragment domain-modified antibodies elicits durable protection from SHIV infection. Nat. Med. 24, 610–616 (2018).
Bar, K. J., Sneller, M. & Harrison, L. J. Effect of HIV-specific antibody VRC01 on viral rebound after treatment interruption. N. Engl. J. Med. 375, 2037–2050 (2016).
Scheid, J. F. et al. HIV-1 antibody 3BNC117 suppresses viral rebound in humans during treatment interruption. Nature 535, 556–560 (2016).
Nishimura, Y. et al. Early antibody therapy can induce long-lasting immunity to SHIV. Nature 543, 559–563 (2017).
Byrareddy, S. N. et al. Sustained virologic control in SIV+macaques after antiretroviral and α4β7 antibody therapy. Science 354, 197–202 (2016).
Grijsen, M. L. et al. No treatment versus 24 or 60 weeks of antiretroviral treatment during primary HIV infection: the randomized Primo-SHM trial. PLOS Med. 9, e1001196 (2012).
Davenport, M. P., Loh, L., Petravic, J. & Kent, S. J. Rates of HIV immune escape and reversion: implications for vaccination. Trends Microbiol. 16, 561–566 (2008).
Namazi, G. et al. The control of HIV after antiretroviral medication pause (CHAMP) study: post-treatment controllers identified from 14 clinical studies. J. Infect. Dis. https://doi.org/10.1093/infdis/jiy479 (2018).
Bernal, E. et al. Low-level viremia is associated with clinical progression in HIV-infected patients receiving antiretroviral treatment. J. Acquir. Immune Def. Syndr. 78, 329–337 (2018).
Bavinton, B. R. et al. Viral suppression and HIV transmission in serodiscordant male couples: an international, prospective, observational, cohort study. Lancet HIV 5, e438–e447 (2018).
Rodger, A. J. et al. Sexual activity without condoms and risk of HIV transmission in serodifferent couples when the HIV-positive partner is using suppressive antiretroviral therapy. JAMA 316, 171–181 (2016).
Hill, A. L., Rosenbloom, D. I. S., Fu, F., Nowak, M. A. & Siliciano, R. F. Predicting the outcomes of treatment to eradicate the latent reservoir for HIV-1. Proc. Natl Acad. Sci. USA 111, 13475–13480 (2014).
Rong, L. & Perelson, A. S. Modeling HIV persistence, the latent reservoir, and viral blips. J. Theor. Biol. 260, 308–331 (2009).
Michor, F. & Beal, K. Improving cancer treatment via mathematical modeling: surmounting the challenges is worth the effort. Cell 163, 1059–1063 (2015).
Bozic, I. et al. Evolutionary dynamics of cancer in response to targeted combination therapy. eLife 2, e00747 (2013).
Sharaf, R. R. & Li, J. Z. The alphabet soup of HIV reservoir markers. Curr. HIV/AIDS Rep. 14, 72–81 (2017).
Eriksson, S. et al. Comparative analysis of measures of viral reservoirs in HIV-1 eradication studies. PLOS Pathog. 9, e1003174 (2013).
Williams, J. P. et al. HIV-1 DNA predicts disease progression and post-treatment virological control. eLife 3, e03821 (2014).
Li, J. Z. et al. The size of the expressed HIV reservoir predicts timing of viral rebound after treatment interruption. AIDS 30, 343–353 (2016).
Reeves, D. B. et al. Anti-proliferative therapy for HIV cure: a compound interest approach. Sci. Rep. 7, 4011 (2017).
Fennessey, C. M. et al. Genetically-barcoded SIV facilitates enumeration of rebound variants and estimation of reactivation rates in nonhuman primates following interruption of suppressive antiretroviral therapy. PLOS Pathog. 13, e1006359 (2017).
Pennings, P. S. Standing genetic variation and the evolution of drug resistance in HIV. PLOS Comput. Biol. 8, e1002527 (2012).
Pinkevych, M. et al. Modeling of experimental data supports HIV reactivation from latency after treatment interruption on average once every 5–8 days. PLOS Pathog. 12, e1005740 (2016).
Acknowledgements
The authors receive funding support in the form of research grants and fellowships from the National Health and Medical Research Council (NHMRC) of Australia, the Australian Research Council and the National Institutes of Health (NIH). S.R.L. receives additional research funding support from The American Foundation for AIDS Research (amFAR).
Competing interests
The authors declare potential competing interests: S.R.L. receives investigator-initiated grant funding from ViiV, Merck and Gilead Sciences. She has provided paid consultancies to AbbVie and received honoraria for chairing meetings from Gilead, ViiV, Merck and BMS. A.D.K. receives non-financial support from Merck and Cytognos for the conduct of clinical trials outside this work and grants from Calimmune and Celgene. S.J.K. reports grants and personal fees from ViiV HealthCare and Gilead Sciences and grants from Johnson and Johnson and Sanofi Pasteur. M.P.D., D.S.K. and D.C. declare no competing interests.
Author information
Authors and Affiliations
Contributions
M.P.D., D.S.K. and D.C. researched data for the article. M.P.D., D.S.K., D.C., S.R.L., A.D.K. and S.J.K. reviewed and edited the manuscript before submission, made substantial contributions to the discussion of content and wrote the manuscript.
Corresponding author
Ethics declarations
Reviewer information
Nature Reviews Immunology thanks N. Goonetilleke and other anonymous reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
About this article
Cite this article
Davenport, M.P., Khoury, D.S., Cromer, D. et al. Functional cure of HIV: the scale of the challenge. Nat Rev Immunol 19, 45–54 (2019). https://doi.org/10.1038/s41577-018-0085-4
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41577-018-0085-4
This article is cited by
-
Cocaine regulates antiretroviral therapy CNS access through pregnane-x receptor-mediated drug transporter and metabolizing enzyme modulation at the blood brain barrier
Fluids and Barriers of the CNS (2024)
-
Impact of misclassified defective proviruses on HIV reservoir measurements
Nature Communications (2023)
-
Complex Dynamics for an Immunosuppressive Infection Model with Virus Stimulation Delay and Nonlinear Immune Expansion
Qualitative Theory of Dynamical Systems (2023)
-
Awareness of HIV functional cure and willingness in participating in related clinical trials: comparison between antiretroviral naïve and experienced men who have sex with men living with HIV
BMC Infectious Diseases (2022)
-
Addressing an HIV cure in LMIC
Retrovirology (2021)