The importance of targeting HIV reservoirs: preclinical insights on current and potential therapeutic opportunities

1. Introduction

Antiretrovirals for the treatment of HIV-1 are one of the most remarkable success stories of anti-infective drug development of the last 30 years. Advancements in antiretroviral pharmacology has altered disease progression to that of a manageable chronic condition for many individuals who start therapy in a timely manner, are not infected with virus resistant to multiple drugs, and who are adherent with their medications. A 2016 study showed significant improvements in life expectancy in people living with HIV/AIDS (PLWHA) who were receiving antiretroviral care, and a narrowing of the gap in life expectancy between PLWHA and people who did not have HIV as compared to the earliest days in the HIV epidemic [1]. Despite these remarkable improvements, to date a cure for HIV-1 has largely been elusive, with only one long-term functional cure for HIV-1, and a second case in remission with hopes of a functional cure [2,3]. While these two cases are extremely exciting, there are considerable costs, caveats, concerns, and risks associated with this approach, which limits the ability of this approach to be adopted wide scale.

Due to these risks and costs, a number of other strategies are under investigation as a means to combat HIV-1 replication. Broadly speaking, these strategies fall into two general approaches: antibody-based regimens to either prevent or control viral replication, and drug related strategies to either reactivate latent virus in cellular reservoirs, or to target viral replication in tissue reservoirs. While extremely interesting, antibody-based strategies are largely outside the focus of this review. Cellular reservoir strategies are based on the general premise that HIV-1 can reside in long-lived latently infected cells, then be reactivated when the cell is reactivated. Thus, the primary goal of these strategies is to either reactivate latent T cells, then kill the infected cells (shock and kill), or to prevent these cells from reactivating (block and lock). A second general strategy focuses on tissue sites where the virus can reside, and posits that these tissue sites represent a barrier preventing antiretrovirals from reaching cells in these sites. These sites can include the brain, lymph nodes, lungs, and secondary lymphoid tissue. Thus, there may be low-level viral replication occurring in these tissue sites, which once antiretroviral therapy has been interrupted can provide a location for the virus to then spread and reinfect the rest of the body. Both of these general strategies have their merits and weaknesses, and strategies are under development to combat HIV-1 replication through these disparate approaches.

2. Cellular reservoirs

As previously noted, cellular reservoir strategies are largely either focused on reactivating latently infected cells (shock and kill), or preventing latently activated cells from reactivating (block and lock). The first strategy to be investigated was shock and kill. Histone deacetylase inhibitors, including vorinostat, have been investigated as part of these strategies [4]. These drugs were shown in some early trials to be able to reactivate latently infected cells, allowing both the immune system to respond to these cells, and for antiretrovirals to work on the target cells as well. A concern with this strategy is that plasma viral titers would likely increase as latently infected cells are activated, although this increase would likely be transient. In the initial small clinical trial assessing this approach, it was well tolerated. There was also evidence that after vorinostat treatment there was a decline in HIV infected resting T cells. Subsequent trials, however, have not shown similar efficacy. Importantly, HIV-1 does not always integrate into the same location in the host cell genome, and depending on which promoter regions are near the integrated virus, it may be more difficult to fully reactivate viral replication [5]. Despite this initial excitement, a number of different drugs, either HDAC inhibitors or drugs in other classes, have also been investigated for potential effect in reactivating latently infected cells, but subsequent trials have not met the promise of the initial trial [5]. Indeed, it mainly appears that with current strategies neither the shock is sufficient to reactivate enough latently infected cells, nor the kill sufficient to eliminate all cells which have been reactivated. It is likely that going forward with this approach will require novel therapeutics with stronger activity, combinations of multiple agents which work through multiple mechanisms, or a combination of these strategies.

A second strategy, referred to as block and lock, instead is largely the inverse of shock and kill. Instead of activating latently infected cells in the hopes of reactivating and then killing the infected cells, it aims to permanently prevent the latently activated virus from reactivating. These strategies are largely focused on inhibiting or silencing the function of HIV transcription factors in infected cells, most commonly TAT [6]. In the first studies with the drug didehydro-Cortistatin A (dCA), the authors of a major study showed a delay in viral rebound in cells treated ex vivo with dCA from individuals, while not having an effect on T cell activation. A study in a humanized mouse model similarly showed in vivo that dCA treatment decreased HIV mRNA in cells, and viral rebound in the mice after antiretrovirals were interrupted. While still in their very early days, block and lock strategies have considerable promise, but extensive work is necessary to show the viability of this approach in humans.

3. Tissue reservoirs

A second viewpoint holds that one of the primary reasons why HIV cannot be eliminated from PLWHA, even in the presence of antiretroviral therapy is the existence of tissue sites in which antiretroviral concentrations may be too low to effectively inhibit viral replication. This is due to the presence of barriers in these tissue sites, including drug efflux transporters on the outside of the tissue, as well as the physicochemical properties of the drug [7]. Some of the early work in this line of inquiry has shown that antiretroviral concentrations in tissue sites including lymph nodes and gut/rectal associated lymphoid tissue are as a whole lower than concentrations in peripheral blood mononuclear cells [7]. This study further showed viral replication in lymphoid tissue even in the presence of an undetectable plasma viral load. They also observed a faster rate of viral decay in subjects who had higher concentrations of drugs in cells in their tissue sites. Subsequent studies have shown viral evolution and trafficking in tissue compartments, and suggested that the reason that resistance does not develop due to subtherapeutic antiretroviral concentrations in tissue reservoirs is that concentrations in these sites are too low to encourage viral resistance [8]. Additionally, there is considerable evidence that lymphoid tissue, both lymph nodes and gut associated lymphoid tissue, are the primary locations where infected cells reside after the initiation of antiretroviral therapy, further supporting the importance of tissue reservoirs [9]. An additional study showed that average concentrations in lymph nodes of efavirenz and raltegravir in lymph nodes and rectal tissue were less than the threshold concentrations to inhibit viral replication, with raltegravir concentrations <1/10^th of clinically relevant threshold concentrations [10]. Utilizing this theory, the critical need is then to increase intracellular antiretroviral concentrations past the range where there is a risk for resistance to develop in these tissue sites, and into the range where viral replication can be fully inhibited.

4. Expert Opinion

There are considerable strengths and weaknesses to both of these strategies, and a holistic approach addressing both cellular and tissue reservoir sites as therapeutic targets is necessary going forward in HIV cure research. With all of these approaches, novel therapeutics, rather than repurposing currently approved medications will be necessary to address these issues. Despite the early promise, the lack of success in subsequent studies further underscores the need for additional research and novel combinations of products, in order to be able to fully reactivate latently infected cells. While such research is ongoing, there is significant work necessary to develop these novel therapeutics. Block and lock has the advantage of not requiring all infected cells to be reactivated, instead focusing on preventing the reactivation of the virus in infected cells. Again, while there is considerable promise to this approach, significant research resources must be dedicated to this approach to see it into maturity. Similarly to shock and kill, currently available therapeutics may be able to provide a proof of concept, but are likely not capable of fully blocking viral replication in these sites.

Targeting tissue reservoirs has the advantage of not necessarily requiring novel therapeutics, but instead requires a need to develop a strategy to increase antiretroviral concentrations in these tissue reservoirs. Addressing this concern may involve one of multiple strategies. First, it may involve changing the therapeutic combinations of antiretrovirals which are utilized, to use combinations which have better penetration into tissue reservoirs. Considerable work has gone into determining which antiretrovirals have high penetration into the CNS reservoir has been done, and similar categorizing of antiretrovirals for penetration into other reservoirs may be necessary [11]. Second, developing secondary strategies to increase the combinations of currently approved drug combinations in tissue is also necessary. This may include small molecule inhibitors of drug transporters, inhibitors of drug metabolism in these sites, or targeted therapeutics (nanoformulated or otherwise) which can home to tissue reservoirs of HIV. Finally, novel therapeutics which have conducive physicochemical properties should be developed to increase antiretroviral concentrations in tissue reservoirs.

Another important question, however, is what level of “cure” is acceptable in individuals. In the realm of cure research, two types of cures are described. First, a sterilizing cure, in which all detectable HIV is eliminated from an individual. Second, a functional cure, which can also be described as remission, describes a situation in which the virus is undetectable for an extended amount of time, but it is conceivable that latently infected cells may still exist and may be able to reactivate at a later date. The two cures which have occurred to date, the Berlin and the London patients, represent functional cures, with non-replication competent viral particles still produced. The end result for some of these proposed approaches, notably shock and kill would result in a sterilizing cure. Block and lock, contrariwise, would result in a functional cure. With both of these approaches, determining what is acceptable in terms of the cure is of key importance.