Therapeutic HIV-1 Vaccines: A Cautious Perspective

1. Current Challenges of the HIV-1 Epidemic

1.1. Current HIV-1 Epidemic

By the end of 2025, 40.9 million people worldwide were living with HIV [1,2]. Since the beginning of the epidemic, approximately 83.5 million people have acquired HIV infection [2], resulting in an estimated 43.2 million deaths [1]. Due to unprecedented global efforts, the number of new HIV infections (1.3 million in 2024) declined by 40% between 2010 and 2024 and by 61% from its peak in 1995 [3]. The number of global AIDS-related deaths has declined by 70% since its peak in 2004. Importantly, the incidences of HIV infection have declined in regions bearing the greatest burden of HIV infection, including Central, Eastern, and Southern Africa and the regions of Asia and the Pacific. HIV incidences have stabilized in Western and Central Europe, North America, Latin America, and the Caribbean [2]. These achievements have been made possible largely through the widespread implementation of antiretroviral therapy (ART) programs, which have transformed HIV infection from a once-fatal disease into a manageable chronic condition. ART has also played an important role in HIV prevention by reducing viral loads in PLWH and thereby decreasing the risk of HIV transmission.

At the same time, the number of HIV cases continues to increase substantially across Eastern Europe and Central Asia (EECA) and the Middle East and North Africa. The HIV epidemic remains particularly poorly controlled in Eastern Europe. In 2024, an estimated 69,292 new HIV infections occurred in the region, corresponding to an incidence rate of 27.2 per 100,000 population. Most of these infections occurred in Russia (48,253 new cases; incidence rate, 33.5 per 100,000) and Ukraine (10,038 new cases; incidence rate, 26.5 per 100,000) [4]. The Russian Federal AIDS Centre reported an even higher number, with 58,740 new HIV cases in 2023 (incidence rate, 40.04 per 100,000 population)[5].

1.2. Factors Driving HIV Epidemics

Analysis of incidence rates reveals region-specific challenges, shortcomings, and gaps in HIV prevention and treatment policies. In Africa, key drivers of the HIV epidemic include social inequalities, increased vulnerability among adolescent girls and young women, uneven access to ART, and inadequate maternal health services [6]. In Asia and the Pacific, where HIV prevalence in the general population is relatively low, epidemics are primarily driven by sexual transmission, while legal and social barriers, including criminalization and discrimination, hinder access to prevention and treatment services among specific population groups [7]. The increasing number of HIV infections in Eastern Europe is driven by stigma, particularly affecting LGBTQ+ individuals, the criminalization of key affected populations, such as people who inject drugs and sex workers, and insufficient access to care and treatment. In particular, substantial delays in treatment initiation following HIV diagnosis, together with unsuppressed viral loads, contribute to ongoing transmission [8].

A common factor driving the HIV epidemics across these regions is conflict-related migration, which can disrupt HIV testing services, surveillance systems, and prevention programs and limit access to ART among displaced populations and refugees. These disruptions may lead to interruptions in treatment, potentially contributing to the emergence of drug-resistant HIV strains and treatment failure upon treatment re-initiation [9,10,11,12]. In addition, across these regions, strong cultural, social, and legal barriers affect communities at increased risk of HIV infection, resulting in limitations on harm reduction programs for key populations [8,13,14,15].

Another factor exacerbating the situation is the high cost of ART in LMICs [16,17]. As the number of people living with HIV continues to grow, the economic burden on national health budgets is substantial. Estimates of ART costs range from $120 [17] to $208 [18] per patient per year, including medications and related services. UNAIDS estimates total investments in HIV treatment and prevention in LMICs between 2016 and 2030 at approximately $350 billion [19].

ART is undergoing ongoing development toward long-acting treatment options [20]. Long-acting therapies can improve adherence, reduce healthcare utilization, and potentially prevent costly treatment failures. However, in the short term, they may increase direct drug costs compared with conventional generic oral ART. The high cost of novel ART drugs raises concerns about equitable access to treatment. For example, in June 2025, the FDA approved lenacapavir, developed by Gilead Sciences, as a novel long-acting HIV treatment [21]. ]. It is a novel long-acting therapy with reported 100% efficacy [22]. Research on ART generics indicates that lenacapavir could be produced at a cost of $40 per person per year, potentially falling to $25 within a year of rollout. However, Gilead has announced a price of $28,218 per person per year [23]. This kind of “gold” therapy is unlikely to change the global HIV pandemic because it will remain unaffordable.

The world has recently faced a substantial decline in public funding for ART. In the United States, constrained budgets are placing increasing pressure on HIV care and prevention programs, including the Ryan White HIV/AIDS Program, which provides grants to states, particularly through their AIDS Drug Assistance Programs (ADAPs), supporting HIV treatment and insurance assistance for PLWH. In response to budgetary pressures, state ADAPs have begun limiting enrollment and reducing the scope of services offered. Twenty states facing budget constraints have recently introduced cost-containment measures affecting ADAPs, including restrictions on eligibility and the scope of services, while some are considering implementing waiting lists in the future [24]. An analysis by KFF, a leading US health policy organization, found that 18 states have already adopted cost-cutting measures affecting their ADAPs, while five additional states are considering similar steps [24]. In the majority of EECA countries, national governments do fund ARV medicines. However, the broader network of services essential for ensuring access to HIV care and treatment—including community outreach, counseling, HIV prevention, peer-led support, and human rights advocacy—has never been fully integrated into national health systems and is currently facing substantial funding cuts due to reductions in contributions to the Global Fund to Fight AIDS, Tuberculosis and Malaria and cuts in US government funding, which also supports the Global Fund [25,26]. These changes are expected to result in PLWH losing access to care and treatment, potentially increasing morbidity and mortality and leading to new HIV infections, as an estimated four out of ten new HIV transmissions are associated with people living with HIV who are not engaged in care [25].

In addition to conflict-related migration and the high cost of ART, limited access to pre-exposure prophylaxis (PrEP) represents another major factor exacerbating the situation. As early as 2019, UNAIDS reported that, among 69 countries implementing PrEP, 56 reported barriers limiting access to PrEP, with cost and availability being among the main obstacles [27]. The U.S. Department of State announced that the President’s Emergency Plan for AIDS Relief (PEPFAR), which has provided PrEP, would scale back the program and provide PrEP medications only to pregnant and breastfeeding women as part of budget optimization measures [28]. Access to HIV treatment for many people in West and Central Africa has also been disrupted, as donors cover approximately 90% of the costs of ARV medicines. In Eastern and Southern Africa, this figure is approximately 38%. In Eswatini, which has the highest HIV prevalence in the world (23% among adults aged 15–49 years), the HIV program has lost 20% of its funding [29]. The impact of funding cuts in parts of Africa could be devastating, potentially resulting in 3.3 million new HIV infections by 2030 [30].

To conclude, the benefits of ART as both treatment and PrEP are being undermined not only by persistent challenges, such as unequal access, poor ART adherence in some social groups, stigma, and discrimination, but also by emerging threats, including: (i) conflict-related migration, resulting in disruptions to HIV testing services, surveillance systems, and prevention programs; (ii) rising ART costs; and (iii) declining government funding for ART and harm reduction programs in some countries. Together, these factors threaten to reduce the global effectiveness of HIV treatment and prevention. Failure to reach people with existing services could undermine the substantial progress achieved through ART and lead to devastating consequences.

1.3. An Effective Prophylactic HIV-1 Vaccine Remains Elusive

Given these challenges, alternative strategies and approaches for HIV-1 prevention remain a priority, particularly for people at high risk of infection. Vaccines remain the gold standard for preventing viral infections. Modeling studies have shown that the HIV pandemic could be brought under control only through the development and global distribution of an effective vaccine [19,31]. Yet, even a partially effective HIV-1 vaccine could drastically change the course of the pandemic [32]. Vaccine costs starts from as little as USD 0.50 per dose, with an average cost of USD 5–20 per dose [33], and even with repeated vaccination, the overall cost would remain substantially lower than that of lifelong ART.

The first Phase I trial of an HIV-1 candidate vaccine was conducted in the USA in 1987 [34]. Since then, more than 250 clinical trials have been conducted, although only a few have reached Phase III. Nearly four decades of preventive HIV-1 vaccine development have yielded limited results. The most promising results were obtained in the RV144 trial in Thailand. RV144 evaluated a canarypox-vector and protein prime–boost regimen in more than 16,000 community participants at relatively low risk of acquiring HIV-1 through heterosexual transmission and demonstrated a 60.5% reduction in the risk of HIV-1 infection at 12 months, with protection declining rapidly to 31% at 3.5 years [35]. Extensive analyses have examined the effects of immunogen design, vector backbone, and vaccination regimen on the efficacy of prophylactic HIV-1 vaccines in this and subsequent clinical trials [16,36,37]. The successes and pitfalls of prophylactic HIV-1 vaccine development have been reviewed elsewhere [38,39,40], and are beyond the scope of this review.

The challenges encountered in HIV-1 vaccine development are multifaceted and include: (1) extreme genetic diversity of the virus and its high mutation rate, allowing HIV to evolve rapidly and evade immune responses; (2) the lack of clearly defined immune correlates of protection; (3) the early establishment of latent viral reservoirs; (4) complex mechanisms of immune evasion; (5) the difficulty of inducing broadly neutralizing antibodies (bNAbs); and (6) limitations of animal models for evaluating vaccine efficacy [39,41]. In today’s world, where governments are seeking to optimize public spending, vaccine development programs and clinical trials are increasingly constrained by funding cuts rather than by a lack of scientific feasibility or efficacy [42,43]. This makes HIV-1 vaccine development an even greater challenge than before. The rapid success in developing COVID-19 vaccines raised expectations for progress in HIV-1 vaccine development, but these expectations have yet to be realized. COVID-19 vaccines benefited from decades of HIV research; however, applying these advanced approaches to HIV-1 vaccine development has not resulted in immediate success, owing to the profound biological differences between the two viruses, with HIV-1 presenting a much more complex target [44]. An effective prophylactic HIV-1 vaccine should not only be highly efficacious, ideally approaching 100%, as a partially effective vaccine may facilitate the emergence of vaccine-escape variants of HIV-1 [45], but should also be affordable to develop and deploy. To date, this goal remains elusive, prompting the development of alternative approaches. One such approach is therapeutic HIV-1 vaccine, which aims not to prevent HIV-1 infection but to treat the disease, achieve a functional cure, or at least slow disease progression.

In this review, we provide an overview of the background and history of therapeutic HIV-1 vaccine development and summarize the achievements to date, including the results of recent clinical trials. To demonstrate that therapeutic HIV-1 vaccination is safe and potentially effective, we review the outcomes of other commonly used vaccines administered to PLWH. We challenge the assumption that PLWH are too immunosuppressed to mount effective vaccine responses and address the concern that vaccination may activate CD4+ T cells, thereby increasing their susceptibility to HIV-1 infection.

2. Therapeutic HIV-1 Vaccines: Concept and Rationale

2.1. Entry of Therapeutic HIV-1 Vaccines into the Scene

Following repeated failures of prophylactic HIV-1 vaccine candidates, the time has come for therapeutic HIV-1 vaccines to take center stage.

Therapeutic HIV-1 vaccines were developed alongside prophylactic vaccines [46,47,48]. The initial goal of these vaccines was to replace ART and achieve immune control of HIV-1 infection, ideally leading to clearance of the virus. Overall, therapeutic HIV-1 vaccines have demonstrated safety and immunogenicity in PLWH [47,48], However, one Phase IIa study reported transient increases in plasma HIV-1 RNA (viral blips) following vaccination, which were interpreted as evidence of the reactivation of HIV-infected cells from latent reservoirs [49]. Alternatively, these findings could be interpreted as reflecting the infection of activated CD4+ T cells after the vaccination.Therapeutic HIV-1 vaccine development has progressed through three partially overlapping phases: (1) structured treatment interruptions (STIs), alone or in combination with immunization/vaccination; (2) therapeutic immunization with HIV-1 antigens delivered using various antigen forms and vehicles, including proteins, virus-like particles (VLPs), DNA, viral or bacterial vectors, and, more recently, RNA; and (3) therapeutic immunization using any of the above platforms in combination with ART or other therapies.

STIs use viral rebound as a source of immunizing antigens [50]. This early practice was abandoned because it failed to enhance viral control [51] and instead led to an increase in the viral reservoir through the infection of new target cells during viral rebound, as well as to the emergence of drug-resistant HIV-1 strains (see, for example, [46,47,52]). This approach is now being revisited, as studies in non-human primates (NHPs) indicate that multiple short analytical treatment interruptions (ATIs) can delay viral rebound and further reduce the viral set point through enhanced CD8+ T-cell responses [53].

The first therapeutic immunization approaches, developed in the early 1990s under the name Remune, used a formaldehyde-inactivated whole-virus vaccine developed by Jonas Salk and the Immune Response Corporation [54,55]. The underlying hypothesis was that augmenting HIV-1-specific immunity would slow disease progression independently of antiretroviral therapy. Initial trials showed that Remune recipients had decreased viral loads, increased CD4+ T-cell counts, and enhanced HIV-1-specific T-cell proliferation compared with placebo, with these benefits being additive to standard care [56]. However, a Phase III trial found no statistically significant differences in AIDS progression or mortality between the Remune and placebo groups, leading to early termination of the trial [57]. A critical assessment published in 2000 confirmed that, although CD4+ T-cell counts increased, there were no significant differences in HIV-1 RNA levels, CD4+ T-cell percentages, or body weight. Ultimately, Remune failed to improve progression-free survival when administered in addition to ART [58].

The concept of immune control of viremia was also investigated using immunizations with individual HIV-1 proteins. In the VaxSyn trial, the adjuvanted recombinant envelope glycoprotein rgp160 was evaluated for its ability to control HIV-1 infection and prevent the progression of immunodeficiency in asymptomatic HIV-1-infected participants [59]. However, the vaccine was found to be ineffective [59,60]. Other therapeutic approaches were also investigated, including DNA vaccines and peptide-based vaccines, which demonstrated safety and immunogenicity but provided no clinical benefit [61,62,63,64].

Overall, therapeutic immunization with HIV-1 antigens alone had no major effects on the course of HIV-1 infection, apart from transient reductions in viral load [47,48].

2.2. Therapeutic HIV-1 Vaccines in Combination with ART

The failure of therapeutic HIV-1 vaccines to alter the course of HIV-1 infection shifted the focus of their application from stand-alone treatment toward their use as an adjunct to ART. As early as 1998, in a Phase I clinical trial, 40 HIV-infected patients received recombinant gp160 in combination with zidovudine or placebo, while a matched group of untreated HIV-1-infected patients served as controls. All immunized patients developed a strong HIV-1-specific T-cell proliferative response, with no effect on viral load. In the control group, non-immunized individuals receiving ART demonstrated reduced viral loads but no HIV-1-specific immune responses, indicating that ART alone did not stimulate the development of anti-HIV-1 immunity. This study provided the first indication that combining ART with HIV-1-specific immunization could result in both control of viral load and enhanced HIV-1-specific immune responses [65].

In this context, therapeutic vaccination was considered as a single, potentially long-acting therapeutic intervention that could be used in combination with ART [66]. Such a combination was expected to reduce treatment costs, lessen the burden of regular ART administration on patients, and ultimately prevent the emergence of drug resistance. Driven by this rationale, an early clinical trial of therapeutic immunization with rgp160 in combination with contemporary standard ART was conducted in 835 HIV-1-seropositive patients at 20 centers in Sweden, Norway, and Finland (Nordic VAC-04). The trial demonstrated an increase in CD4+ T-cell counts compared with baseline and significantly lower mortality among vaccine recipients than in the control group at two years, but these effects did not translate into long-term clinical benefits [67].

Among the more recent trials with similar outcomes, the AELIX-002 trial evaluated a T-cell immunogen (HTI) delivered as a combination of DNA.HTI, MVA.HTI, and ChAdOx1.HTI in 45 individuals who had initiated ART during early HIV-1 infection [68]. All HTI vaccine components induced strong, broad, and polyfunctional CD4+ and CD8+ T-cell responses. In post hoc analyses, plasma viral load at the end of the ATIs and the duration of time off ART positively correlated with vaccine-induced HTI-specific T-cell responses measured at ART cessation [68]. However, although T-cell responses were induced, they were insufficient to prevent viral rebound during ATIs. Interestingly, subsequent studies demonstrated that these responses might be enhanced by increasing the number of anatomical sites at which vaccines were administered and by distributing vaccine components across multiple anatomical sites [69].

In parallel, combination strategies were also explored for prophylactic HIV-1 vaccination. Based on the rationale that ART could enhance the effects of vaccination, Excler et al. proposed the combined administration of a prophylactic HIV-1 vaccine and PrEP (VAXPREP), hypothesizing that the combination could have a synergistic effect and provide greater benefits than either vaccination or PrEP alone [70,71]. The most recent such study was the African-led, European-supported Phase IIb HIV-1 prevention trial conducted at four sites in three countries—Uganda, Tanzania, and South Africa—from 2018 to 2024 (NCT04066881, https://www.prepvacc.org/). The study evaluated two vaccine regimens combined with PrEP: a DNA vaccine followed by a protein-based vaccine (DNA-HIV-PT123 + monomeric AIDSVAX® B/E) and a DNA, MVA, and protein-based regimen (DNA-HIV-PT123 + trimeric CN54gp140, followed by poxvirus-based MVA CMDR + trimeric CN54gp140). In June 2023, the study announced completion of enrollment [72]. However, in December 2023, the vaccine arm of the trial was discontinued, with no additional protective effect of vaccination reported [73].

Overall, no major breakthroughs were achieved, not only with STIs and therapeutic HIV-1 vaccination alone, but also in the first therapeutic and prophylactic combination trials.

2.3. A New Paradigm: Therapeutic HIV-1 Vaccines as Adjuncts to ART

Since the early 2000s, virtually all vaccine recipients were on ART, and subsequent studies of therapeutic HIV-1 vaccination were largely conducted in combination with ART. Vaccination served as a complementary therapeutic modality, aimed at enhancing and prolonging the effectiveness of ART, preventing the emergence of drug resistance, and mitigating the consequences of non-adherence.

Such a development was strongly supported by the realities of the HIV-1 epidemic. First, the growing global population of children with vertically acquired HIV-1 who require lifelong ART is at high risk of ART non-adherence and virological failure during adolescence [74]. Second, limited drug availability, including stock-outs caused by supply-chain bottlenecks, funding shortages, and logistical and clinical barriers, can lead to unplanned ART interruptions [75,76]. During periods of ART non-adherence and/or unplanned treatment interruptions, the immune control induced by therapeutic HIV-1 vaccination was expected to help suppress viral replication.

3. Directions in Therapeutic HIV-1 Vaccine Development

Therapeutic HIV-1 vaccines have been developed along several distinct but partially overlapping approaches, including: (i) mimicking the immune responses of elite controllers; (ii) focusing on T-cell responses through the selection of favorable T-cell epitopes; (iii) enhancing antiviral T-cell responses through the use of latency-reversing agents (LRAs), followed by T-cell-focused immunization—the so-called “kick-and-kill” HIV-1 vaccine strategy; (iv) shifting toward alternative LRAs after limited success with the initial approaches; and (v), last but not least, the development and application of bNAbs.

3.1. Mimicking the Immune Responses of Elite Controllers

During the natural course of HIV-1 infection, some individuals are able to spontaneously control viral replication to below the limit of detection. These individuals are known as elite controllers [77]. Their existence demonstrates that HIV-1 infection can be controlled, presumably through effective immune responses. Characterization of the immune responses of elite controllers has revealed that they can develop populations of multifunctional T cells that produce multiple cytokines and are capable of suppressing viral replication in vitro [78,79,80].

Among the therapeutic vaccines aimed at mimicking the immune responses of elite controllers were subunit vaccines, DNA vaccines, and vaccines based on viral vector platforms [81]. Of these, DNA vaccines were particularly attractive because, at the time, they were considered the best available approach for reproducing the natural immune response to viral infection, as RNA vaccines were not yet available [82], and, potentially, the immune responses observed in elite controllers. Moreover, DNA-based vaccines had been shown to be effective in the treatment of other chronic viral infections, notably high-risk human papillomavirus (hrHPV) infections and associated cancers, providing therapeutic benefits in Phase II/III clinical trials [83].

Disappointingly, vaccination with a DNA vaccine encoding Gag, Pol, and Nef from HIV-1 clade B and Env from HIV-1 clades A, B, and C, followed by booster immunization with a replication-deficient adenovirus type 5-based vaccine encoding all viral antigens except Nef [84], as well as vaccination with the DNA vaccine VRC-HIVDNA 009-00-VP [85] resulted in neither a reduction in plasma viral load nor a decrease in the reservoir of latently infected resting CD4+ T cells. Similarly, DNA vaccination with the GTU®-multi-HIVB vaccine did not enhance pre-existing immune responses [86]. However, another study reported that vaccination with DNA immunogens encoding Env/Rev and Gag/Pol proteins reduced viremia compared with placebo [87]. Encouraging signals also emerged from prophylactic vaccine trials. Although the HVTN 505 and Step (HVTN 502) prophylactic vaccine trials failed to prevent HIV-1 infection, follow-up analyses revealed an association between reduced infection risk and high levels of vaccine-induced Env-specific CD8+ T-cell responses [88,89]. These findings were consistent with observations in elite controllers and, taken together, helped guide both prophylactic and therapeutic vaccine development toward the induction of CD8+ T-cell responses [90]. However, multiple subsequent clinical trials conducted in various populations of PLWH generated no clear confirmatory results. Although vaccines were able to elicit de novo or “shaped” immune responses of the desired specificity, none was able to alter the course of HIV-1 infection or provide viral control during ART interruption (reviewed in [48]).

Interestingly, in individuals who initiated ART during early HIV-1 infection, therapeutic vaccination was shown to redirect T-cell responses toward conserved regions of HIV-1 (BCN01 study) [91]. In a placebo-controlled Phase II randomized clinical trial evaluating the safety and immunogenicity of a therapeutic prime-boost vaccination strategy in HIV-1-infected patients receiving ART, a recombinant DNA vaccine (GTU-MultiHIV clade B) was followed by a booster vaccination with a lipopeptide vaccine (HIV-LIPO-5). Vaccination induced strong and polyfunctional HIV-1-specific CD4+ and CD8+ T-cell responses that significantly exceeded those observed in the placebo group [92]. These studies demonstrated that therapeutic vaccination may improve the specificity and potency of anti-HIV-1 immune responses compared with those induced by natural infection. At the same time, the GTU-MultiHIV clade B/HIV-LIPO-5 trial demonstrated that these immune responses were insufficient to control viremia following ATI [92]. Furthermore, vaccine recipients showed an increased proportion of memory CD8+ T cells co-expressing markers of exhaustion and senescence (PD-1/TIGIT and CD27/CD57) compared with the placebo group, indicating qualitative impairment of the vaccine-induced immune response [92].

These failures suggested that, to achieve sufficient immunogenicity and induce responses capable of controlling viremia, HIV-1 antigens may need to be modified by deleting regions containing so-called “decoy” or “exhaustion”-associated epitopes. In the antibody response, decoy epitopes are defined as highly immunodominant Env regions that induce the production of non-protective antibodies, thereby diverting the immune response away from vulnerable, conserved sites required for neutralization [93,94]. Similarly, for T-cell responses, there are “beneficial regions” that induce broad and potent CD8+ T-cell-mediated viral inhibitory activity against a panel of HIV-1 strains ex vivo. Such responses have been detected in viremic controllers but were not observed among participants in HIV Vaccine Trials Network (HVTN) studies [95]. In fact, these beneficial regions are infrequently targeted by CD8+ T cells elicited by vaccines encoding full-length HIV-1 proteins, either in healthy volunteers or in PLWH receiving ART. Effective anti-HIV-1 CD8+ T-cell responses may therefore be undermined by immunodominance hierarchies [95]. These findings point to the need to tailor HIV-1 immunogens through the systematic selection of empirically defined vulnerable regions within the viral proteome while excluding immunodominant decoy epitopes that are irrelevant to HIV-1 control.

3.2. HIV-1 Vaccines Directed Against T-Cell Epitopes

Epitope-based vaccines can focus the immune response on the most vulnerable “functional” regions of the virus that contain T-cell epitopes capable of mediating viral control.

As formulated more than a decade ago by Sanou et al. [96], an effective (protective) T-cell-based HIV-1 vaccine should stimulate CD4+ and/or CD8+ T cells with a broad T-cell receptor repertoire, preferably targeting Gag epitopes, thereby exerting immune pressure that drives mutations in Gag, which may impair viral fitness. Vaccine-induced T cells should express markers associated with central or effector memory functions, but not markers of exhaustion. They should proliferate, secrete multiple cytokines, and kill HIV-1-infected cells [96]. Possessing these properties, which mimic the T-cell responses observed in elite controllers, requires tailoring T-cell epitope-based vaccines. Additional constraints arise from the recognition of these T-cell epitopes by HLA types prevalent in the country where the vaccine trial is conducted [97] and from the characteristics of the HIV-1 strains circulating in that country (see, for example, [98]). There are two options for addressing these challenges: geographically or population-tailored vaccine candidates selected according to local HLA types and viral variants, or “universal” vaccines combining HIV-1 epitopes conserved across multiple viral clades and recognized by the most common HLA types in the target population. For HIV-1 vaccine developers, the latter approach was preferable. Promising results were obtained in animal models (reviewed in [40]).

Hundreds of multiepitope vaccines have been tested to date in newly or chronically infected individuals (see [99] for a comprehensive review). All vaccines were safe and well tolerated. The vaccines showing the most promising results are summarized in Table 1.

Early Phase I and IIa trials of multiepitope vaccines, using immunogens consisting of known T-cell epitopes conserved across HIV-1 clades and delivered on multipeptide, MVA, or DNA platforms, failed to induce potent and durable anti-HIV-1 cellular immune responses [96]. Subsequent trials demonstrated that epitope-based vaccines were immunogenic. Nevertheless, only a few of them, notably HIV-v and Vacc-4x, had an effect on viral loads or enabled control of viral loads during ATIs, thereby demonstrating some degree of control over viral replication [99].

Here, we provide a few examples. HIVconsvX, which utilizes a conserved mosaic strategy designed to target diverse HIV-1 variants, was evaluated in a series of Phase I clinical trials in healthy volunteers, including HIV-CORE 006 [100] in Africa and HIV-CORE 005.2 [101] in the UK. Mosaic immunogens were delivered using the ChAdOx1 and MVA vectors [100,101]. Specific T-cell responses peaked at a median of 4,433 IFN-γ spot-forming units per 10^6 PBMCs and were detected in all volunteers. Vaccine recipients recognized a median of 9 out of 10 peptide pools, indicating broad specificity, with each participant developing responses to at least nine HIV-1 epitopes (one per pool), including variant peptides. T cells proliferated upon re-exposure to the antigens, displayed multiple effector functions, and inhibited HIV-1 from the four major global clades A, B, C, and D [101]. Similar results were reported in the HIV-CORE 006 trial, with T-cell responses peaking at a median of 2,310 IFN-γ spot-forming units per 10^6 PBMCs in 70 of 71 volunteers [100].

Similarly, the therapeutic DNA vaccine p24CE (Conserved Element; NCT03560258, NCT03181789), encoding highly conserved regions of the HIV-1 Gag protein and delivered by electroporation, induced durable cellular immune responses directed against previously unrecognized conserved elements in nearly 50% of PLWH receiving ART when paired with a full-length p55 Gag DNA vaccine as a booster [102]. These responses were not detected in the placebo group [102]. Overall, these recent vaccine approaches were safe and well tolerated and induced high-frequency, cross-reactive T-cell responses, including responses targeting previously unrecognized epitopes.

The latest epitope-based vaccine candidate, the Gorilla Adenovirus Vectored HIV Networked Epitopes vaccine (GRAdHIVNE1), entered a Phase I clinical trial in July 2025 (NCT06617091). The trial, which plans to enroll approximately 120 adults aged 18–50 years, is evaluating the safety and immunogenicity of the vaccine candidate in PLWH and uninfected individuals, including 48 PLWH receiving successful ART. Participants receive either one or two doses of the candidate vaccine or placebo and are monitored for safety and immune responses for 19 months following immunization. The trial provides a unique opportunity to compare vaccine-induced immune responses between PLWH and uninfected individuals.

3.3. The “Kick-and-Kill” HIV-1 Vaccine Strategy

3.3.1. Introduction of Latency-Reversing Agents (LRAs)

The inability of therapeutic HIV-1 vaccines to achieve sustained control of viral replication was attributed, in part, to the inability of the immune response to eliminate latently infected CD4+ T cells. These cells are not readily recognized or eliminated by the immune system because they do not express HIV-1 antigens, yet they can serve as a source of viral rebound following ART interruption [103]. This provided a rationale for combining potent therapeutic immunization with interventions targeting the viral reservoir while maintaining ART. Inspiration for this approach was drawn from strategies used to target and control the proliferation of cancer cells [104]. The proposed strategy was to induce HIV-1 transcription and expression of viral antigens in infected cells using latency-reversing agents (LRAs), followed by elimination of HIV-1 antigen-expressing cells by CD8+ T cells activated by therapeutic HIV-1 vaccination—the so-called “kick-and-kill” approach [105]. The first LRAs targeted epigenetic mechanisms by disrupting the compact chromatin structure and thereby facilitating HIV transcription. These included histone deacetylase (HDAC) inhibitors, as well as agents that activate cellular signaling pathways, particularly those regulated by NF-κB, to induce HIV-1 transcription, such as protein kinase C (PKC) agonists.

The development of T-cell vaccines based on the “kick-and-kill” cure strategy has been extensively reviewed over the past decade [106,107,108], with a particular focus on whether therapeutic vaccination-induced immune responses following the “kick” can reduce the HIV-1 reservoir. The initial answer was “no”. The first trial to assess the effect of ART alone versus ART combined with a “kick-and-kill” intervention on markers of the HIV-1 reservoir was the Phase II, open-label, multicenter, randomized controlled RIVER trial, conducted at six clinical sites in the UK. Individuals newly diagnosed with HIV-1 were randomized to receive either ART alone (control) or ART combined with the HDAC inhibitor vorinostat to induce the “kick” and T-cell-inducing viral vector-based vaccines encoding conserved HIV-1 sequences, ChAdV63.HIVconsv as the prime and MVA.HIVconsv as the boost, to mediate the “kill” [109]. The intervention was well tolerated and caused no significant adverse effects. However, at weeks 16 and 18, mean total HIV-1 DNA levels did not differ significantly between the intervention and control groups, indicating that the “kick-and-kill” approach provided no significant benefit over ART alone in reducing the HIV-1 reservoir [109]. Subsequent studies similarly demonstrated that “kick-and-kill” interventions did not reduce the HIV-1 reservoir compared with ART alone, despite inducing potent virus-specific CD4+ and CD8+ T-cell responses [110].

More promising results were obtained in a clinical trial evaluating a combination of the HDAC inhibitor romidepsin and the MVA.HIVconsv vaccine (BCN02 study) [111]. Romidepsin treatment resulted in increased histone acetylation, higher levels of cell-associated HIV-1 RNA, and T-cell activation, and was associated with a marginally significant reduction in the viral reservoir. Vaccination induced robust and broad HIVconsv-specific T-cell responses that were strongly refocused toward conserved regions of the HIV-1 proteome. During ATIs, with a plasma viral load of >2,000 copies/mL as the criterion for ART resumption, 23% of participants maintained sustained suppression of viremia for up to 32 weeks without evidence of viral reservoir reseeding [111]. Interestingly, a later trial evaluating a combination of romidepsin and the broadly neutralizing anti-HIV-1 antibody 3BNC117 showed no effect [112]. These findings suggested a potentially important role for vaccine-induced CD8+ T-cell responses in achieving virological control, rather than an effect of LRA treatment alone. They therefore provided support for the functionality of the “kill” component of the “kick-and-kill” strategy in BCN02, although the study lacked a control group receiving LRA alone.

Thus, despite the conceptual elegance of the “kick-and-kill” strategy, its clinical efficacy has been limited by the dual failure of an insufficient “kick”, which fails to produce adequate levels of viral antigen for immune recognition, and a compromised “kill” response. The modest success of the BCN02 trial provides an important proof of concept that this barrier may be overcome. One possible approach to achieving clinically meaningful results is to use alternative LRAs, drawing on knowledge and experience gained from cancer therapy.

3.3.2. Alternative “Kick-and-Kill” Approaches: Using Different LRAs

Other approaches to “kick-and-kill” involve immunomodulation aimed first at reactivating dormant HIV-1-infected cells and then facilitating their elimination by strengthening antiviral immune responses.

3.3.2.1. Immune Checkpoint Inhibitors (ICIs)

ICIs block inhibitory receptors on immune cells that are upregulated during chronic HIV-1 infection, including programmed cell death protein 1 (PD-1), cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), lymphocyte-activation gene 3 (LAG-3), and T-cell immunoreceptor with Ig and ITIM domains (TIGIT). This can reverse T-cell exhaustion and enhance the ability of CD8+ T cells to recognize and eliminate HIV-1-infected cells following viral reactivation [113,114,115]. Interestingly, antibodies targeting PD-1, programmed death-ligand 1 (PD-L1), and CTLA-4 were shown not only to enhance anti-HIV-1 cytotoxic responses but also to affect HIV-1 latency [116]. To date, ICIs have not been associated with significant changes in total cell-associated HIV-1 DNA. However, they have been reported to increase the proportion of integrated proviruses containing frameshifts and stop codons and to cause a limited reduction in HIV-1 reservoir diversity [116].

An ICI strategy using anti-PD-1 antibodies was tested in SIV-infected macaques and resulted in an expansion of virus-specific CD8+ T cells and a decrease in plasma SIV RNA level [117]. However, not all ICI trials were successful. TIGIT blockade in nonhuman primates was evaluated in vivo as an approach to reverse T-cell and NK-cell exhaustion during SIV infection. It resulted in only a modest increase in T-cell proliferation and virus-specific T-cell responses in both magnitude and breadth, while plasma viral loads remained stable in treated animals. These findings indicated that anti-TIGIT monoclonal antibody treatment alone was insufficient to restore anti-SIV CD8+ T-cell function [118].

Early clinical trials of PD-L1 blockers demonstrated only a modest increase in virus-specific CD8+ T cells [119]. Further studies and optimization in the SIV model indicated that, in the setting of viremia, targeting a single inhibitory pathway may be insufficient, supporting the use of ICI combinations with immunotherapy. Indeed, a Phase I clinical trial in PLWH with cancer demonstrated that, while anti-PD-1 monotherapy did not affect HIV-1 latency or the size and composition of the latent reservoir, the combination of anti-PD-1 and anti-CTLA-4 resulted in a modest increase in HIV-1 RNA levels [120].

More recently, Marin et al. [121] conducted in vitro studies of PD-1 and TIM-3 blockade using PBMCs from PLWH enrolled in the Phase I, non-randomized, multicenter therapeutic HIV-1 vaccine trial BCN01. Participants with early HIV-1 infection receiving ART were either not vaccinated (Et) or received T-cell vaccination (Etvac), while PLWH receiving ART during chronic infection (Chro) served as a control group. The groups were compared with respect to the magnitude of vaccine-induced and HIV-1-specific CD8+ T-cell responses following in vitro peptide stimulation, assessed by flow cytometry and multiplex assays [121]. PD-1 blockade significantly increased vaccine-induced HIV-1-specific CD8+ T-cell responses in the Etvac group, whereas TIM-3 blockade alone had no effect. Responses in the Et group were not enhanced by either treatment. In the Chro group, PD-1 blockade increased the frequency of HIV-1-specific CD8+ T cells.

This study provided evidence supporting a potential beneficial effect of PD-1 blockade on the functionality of vaccine-induced HIV-1-specific CD8+ T-cell responses in PLWH receiving early ART, supporting further clinical evaluation of ICIs to improve the efficacy of therapeutic T-cell vaccination in PLWH, potentially using carefully selected ICI combinations.

3.3.2.2. Toll-Like Receptor (TLR) Agonists

An interesting alternative to conventional LRAs is the use of Toll-like receptor (TLR) agonists. These agents can act as indirect LRAs by reactivating latent HIV-1 through innate immune activation while simultaneously stimulating immune-mediated elimination of infected cells. TLR7/8 and TLR9 agonists act on dendritic cells (DCs) or plasmacytoid DCs (pDCs), stimulating the production of type I interferons, including IFN-α [122]. These cytokines can activate resting CD4+ T cells and promote HIV-1 latency reversal [123]. TLR1/2 agonists, such as Pam3CSK4 and SMU-Z1, can directly activate latent HIV-1 transcription in CD4+ T cells through the NF-κB pathway and promote NK cell-mediated suppression of HIV-1-infected autologous CD4+ T cells [124]. A combination of TLR2 and TLR7 agonists directly activated T cells while also inducing TNF-α production through pDCs, resulting in robust HIV-1 latency reversal [125].

In SIV-infected rhesus monkeys, a combination of the Ad26/MVA mosaic vaccine and the TLR7 agonist GS-9620 improved virologic control and delayed viral rebound following ART discontinuation [126]. However, the effect of this intervention on the viral reservoir remained unknown. Of note, ART was initiated during the early acute phase of infection in this study. Further experiments in SIV-infected monkeys initiating ART during the chronic phase of infection would be valuable, as this model would better represent the clinical setting of the majority of PLWH [126].

Two double-blind, placebo-controlled RCTs, AELIX-002 (Phase I) [68] and AELIX-003 (Phase II) [127], evaluated HIVACAT T-cell immunogen (HTI)-based vaccines in combination with the TLR7 agonist vesatolimod in 50 virally suppressed PLWH who had initiated ART early. Although vaccination induced strong, broad, HTI-focused T-cell responses, all vaccine recipients experienced viral rebound during ATIs [128]. The proportion of participants remaining off ART for 25 weeks did not differ between the vaccine and placebo groups [127]. The effect of the intervention on the viral reservoir was not reported.

Overall, TLR agonists enhanced vaccine-induced anti-HIV-1 responses but had little or no effect on proviral DNA levels. Thus, in contrast to classical LRAs, both ICIs and TLR agonists appeared effective in stimulating the “kill” component but had limited ability to induce the “kick”.

3.3.2.3. Challenges and Perspectives of LRAs and LRA Analogues

Extensive clinical evaluation of LRAs has revealed detectable perturbations in the HIV-1 reservoir but no measurable reduction in its size [129]. Several explanations may account for the limited efficacy of LRAs. First, LRAs need to induce HIV-1 latency reversal to enable subsequent recognition of infected cells by immune effector cells. However, HIV-1 latency is heterogeneous, and different mechanisms of transcriptional silencing may require different “kick” strategies or LRA agents [130]. This heterogeneity supports the use of combinations of LRAs rather than a single agent. In support of this approach, a combination of a histone deacetylase (HDAC) inhibitor and a poly(ADP-ribose) polymerase inhibitor reduced the reservoir size by 67% in HIVGR670-infected primary cells [131].

Furthermore, cells and tissues differ substantially in their susceptibility to LRAs depending on their activation status, differentiation state, and anatomical location, arguing against the feasibility of a single LRA treatment producing a sufficiently broad and substantial “kick” [132]. Moreover, HIV-1 reactivation is stochastic, and only a proportion of proviruses are intact and capable of being reactivated. This raises the possibility that currently used methods for reservoir quantification may inadequately capture the effects of LRA treatment [133]. Complicating the picture further, CD8+ T cells can suppress HIV-1 transcription and thereby blunt the activity of LRAs [134]. Together, these factors may result in an insufficient “kick”.

Illustrating the consequences of an inefficient “kick”, Ruiz et al. provided experimental evidence for a threshold of antigen production required for the rapid and efficient killing of HIV-1-reactivated, antigen-producing cells by specific CTLs [135]. Inefficient or partial latency reversal, resulting in an insufficient number of cells carrying reactivated proviruses and, consequently, insufficient levels of viral antigen production, would impair the speed of recognition and the magnitude of killing of HIV-1-reactivated cells, thereby undermining the “kill” effect.

Furthermore, effective clearance of reactivated cells depends on the functional characteristics of CD8+ T cells induced in recipients of “kick-and-kill” interventions. However, LRA agents have also been shown to affect CD8+ T-cell function [136]. An in-depth analysis of the cytotoxicity of LRAs from four different classes toward immune cells revealed that HDAC inhibitors are highly cytotoxic to both CD4+ and CD8+ T cells, whereas PKC agonists and inhibitors of the bromodomain and extra-terminal (BET) family exhibit lower toxicity [137]. The other side of the “kill” equation is the resistance of cells harboring reactivated latent HIV-1 to CTL-mediated killing. Notably, in PLWH who initiated ART during chronic infection, the viral reservoir consists almost entirely of CTL-resistant cells [48].

Exhaustion of HIV-1-specific T cells, T-cell receptor clonal replacement, and T-cell loss of stemness have been identified among the factors potentially explaining the failure or limited success of ICI trials [138]. The limited success of TLR agonists also has multiple possible explanations. First, TLR agonists are rapidly metabolized and degraded, posing challenges for their therapeutic application [139]. Second, exposure to agonists of TLR2, TLR3, TLR4, TLR5, TLR7, TLR8, and TLR9 mimics exposure to translocated microbial products in the damaged gut and can trigger activation-induced cell death of CD4+ T cells, providing a model for HIV-1-induced CD4+ T-cell dysfunction and loss [140]. Third, resting CD4+ T cells do not express physiological levels of either TLR7 or TLR9. Hence, TLR7/9 agonists cannot directly activate these cells to reverse HIV-1 latency [141]. Finally, like many classical LRAs, TLR agonists fail to reactivate the entire latent reservoir because of their indirect mechanism of action, the deep latency of some proviruses, and differences in cellular TLR expression [142]. TLR agonists also present limitations for clinical use because they induce potent immune activation, which may cause significant adverse effects [143].

This calls for innovative strategies to enhance the efficacy of LRA- and LRA analogue-based approaches in the context of HIV-1 infection. One such strategy is the use of combinations of LRAs with different mechanisms of action. An overview of the results obtained so far, mostly from in vitro studies, indicates that such combinations may be more effective than treatment with single agents [142]. Their evaluation in clinical trials involving PLWH is urgently needed.

3.4. Focus on Broadly Neutralizing Antibodies

An independent line of therapeutic HIV-1 vaccine development focuses on bNAbs. These antibodies target conserved epitopes on the HIV-1 Env glycoprotein, including the CD4 receptor-binding site (CD4bs), the interface between the gp120 and gp41 subunits, and the trimer apex quaternary epitope [144,145]. High levels of bNAbs have the potential to provide durable virologic control of HIV-1 replication in the absence of antiretroviral therapy. There are two approaches to achieving high levels of anti-HIV-1 bNAbs: passive and active immunization, analogous to passive and active immunization against rabies.

3.4.1. Passive Vaccination with Broadly Neutralizing Antibodies

Early efforts using passive vaccination with bNAbs focused predominantly on single-agent regimens. Although this approach could delay viral rebound, it was ultimately insufficient, as breakthrough viremia emerged due to the selection of viral variants resistant to the administered antibody (see reviews [146,147]).

Concerns about HIV-1 resistance to single bNAbs have led to the use of combinations of bNAbs with non-overlapping resistance profiles. While passive vaccination with a single bNAb offers a short-lived but significant reduction in HIV-1 viremia, combination therapy with ≥2 bNAbs can maintain viral suppression for a longer period and delay viral rebound after ATIs in PLWH harboring antibody-sensitive strains [148]. For example, combination therapy with the bNAb pairs 3BNC117 + VRC01 and 3BNC117 + 10-1074 significantly delayed viral rebound in several clinical trials [149,150]. The combination of 3BNC117 + 10-1074 also resulted in a reduction in the intact, but not defective, proviral reservoir [151]. Further-modified versions with prolonged half-lives (3BNC117-LS and 10-1074-LS, respectively; RIO study) were subsequently evaluated in a clinical trial and reduced the risk of viral rebound within 20 weeks by 91% compared with the placebo group [152]. Another clinical trial investigated a combination of three bNAbs—PGDM1400, PGT121, and VRC07-523LS. This combination resulted in a sharp and rapid decline in viral load, although viral rebound to pre-infusion levels occurred after a median of 20 days in all participants [153]. Some promising new results from combination bNAb therapy were also reported at the recent annual HIV/AIDS conference [154]. These clinical trials highlight the value of combining antibodies to increase the breadth and potency of antiviral coverage and prevent the emergence of resistance to neutralization. Importantly, they also support the possibility of achieving immunological control of HIV-1 replication if vaccination can induce sufficiently potent and broadly neutralizing antibodies.

An alternative next-generation approach involves engineered bispecific and trispecific antibodies targeting two or three independent viral epitopes [146]. Bispecific antibodies have entered clinical trials and have been shown to be safe and well tolerated [155]; however, their antiviral activity requires further confirmation.

3.4.2. In Vivo Expression of Broadly Neutralizing Antibodies

Passive immunization requires repeated administration of bNAbs. An attractive alternative is to express these antibodies in vivo. Until recently, there were no reliable methods for sustained production of bNAbs in humans [40]. A major breakthrough came with the introduction of adeno-associated virus (AAV)-mediated gene transfer. AAV vectors have since become a leading platform for in vivo gene therapy in humans and have demonstrated both safety and efficacy [156]. AAV-mediated gene transfer enables sustained in vivo production of desired anti-HIV-1 antibodies, an approach first described by Lewis et al. in 2002[157] and still actively pursued today. Recent studies have demonstrated its potential in mice [158,159], monkeys [160] and humans [161].

Despite its successful use in humans for the delivery of a wide range of therapeutic genes [156], results of AAV-mediated delivery of bNAbs in humans have been relatively modest. A Phase I trial of an AAV vector delivering the anti-HIV-1 antibody PG9 (rAAV1-PG9D) to healthy volunteers failed to induce sustained, high-level antibody expression, which was attributed to the development of an immune response against the AAV vector [162]. In contrast, a Phase I clinical trial in PLWH evaluating delivery of the VRC07 bNAb using an AAV8 vector (NCT03374202) demonstrated that intramuscular administration of AAV8-VRC07 was safe and well tolerated and induced production of VRC07 for at least one year and, in 4 of 8 participants, for up to 3 years [161]. However, AAV8-VRC07 treatment produced no significant changes in either CD4+ T-cell counts or viral load, indicating neither a detrimental nor a beneficial clinical effect. In addition, in several participants, treatment induced antibodies against the Fab portion of VRC07, which reduced VRC07 production [161]. This study provided proof of concept that AAV vectors can enable durable in vivo production of biologically active HIV-1-specific bNAbs that are otherwise difficult to induce in vivo, but also demonstrated that the clinical application of AAV-based delivery of anti-HIV-1 bNAbs remains challenging.

The latest approach is ex vivo B-cell gene therapy, which is aimed to generate B cells capable of expressing bNAbs. However, to date, this approach has been evaluated only in a proof-of-concept study in mice [163] and remains far from clinical application.

Overall, “saturation” of the bloodstream with appropriate bNAbs can allow ART-free control of viremia, at least temporarily. The challenge is therefore to achieve the sustained circulation of bNAbs, preferably through approaches that enable active induction rather than passive delivery.

3.4.3. Vaccines Designed to Induce Broadly Neutralizing Antibodies

Similar to T-cell epitope-based vaccines, bNAb-inducing vaccines have employed epitope-based strategies.

Overall, B-cell epitope-based immunogens were generally safe and immunogenic in animal models and induced HIV-1-specific immune responses in humans, similarly to T-cell epitope-based immunogens. However, despite their immunogenicity, they had limited ability to induce bNAb responses capable of controlling viral replication in humans.

The limited success of bNAb induction using epitope-based vaccines may be overcome by B-cell lineage- or germline-targeting vaccine strategies (see Table 2). Initial clinical trials of these approaches demonstrated the induction of bNAbs in non-human primates [164,165] and vaccine recipients [166], encouraging the development of additional heterologous boosters and improved boosting strategies to induce neutralizing antibodies with broader specificities. Overall, the results are promising, but robust proof of concept has yet to be established in larger clinical trials.

3.4.4. Combined Effects of Broadly Neutralizing Antibodies and CD8+ T-Cell Responses

A number of studies indicate that bNAbs could contribute to effective long-term treatment or a functional cure of HIV-1 through the induction of CD8+ T-cell responses, a phenomenon referred to as the “vaccinal effect”. The vaccinal effect is thought to occur when antibody–virus immune complexes enhance antigen processing and presentation, thereby promoting the activation and proliferation of HIV-1-specific T cells. In a study by Nishimura et al. [175], 46% of SHIV-infected macaques receiving bNAb therapy maintained post-treatment viral suppression during ATIs after bNAb levels had waned. Administration of anti-CD8+ monoclonal antibodies resulted in rapid viral rebound in all macaques, suggesting that viral suppression during ATI was mediated by HIV-1-specific CD8+ T cells [175]. This effect was actively discussed at the annual HIV/AIDS conference, although no definitive conclusion was reached [154]. Interestingly, a recent preprint reporting results from the RIO trial (NCT04319367), which evaluated a combination of two bNAbs, found that bNAb administration resulted in a significant increase in the proliferation of HIV-1 Gag-specific CD8+ T cells between weeks 0 and 36 [176].

In theory, the vaccinal effect could serve as a mechanism for achieving a functional cure. As this strategy has shown promise in NHP studies [147], results are awaited from clinical trials evaluating combinations of bNAbs with therapeutic vaccination (NCT04357821, NCT04983030, NCT05769569, NCT06071767, NCT03619278, NCT06484335, AbVax). Preliminary results from one of these trials (NCT04357821) have been presented [177]. This trial evaluated a combination treatment comprising the bNAbs 10-1074 and VRC07-523LS, the TLR9 agonist lefitolimod, and vaccination with an IL-12-adjuvanted, Gag (p24) conserved element (CE)-targeted DNA prime followed by an MVA/HIV62B boost. Seventy percent of participants receiving the combination therapy maintained at least partial virological control during ATI, with one participant maintaining complete viral control for more than 18 months off ART. The overall mean time to viral rebound was 15 weeks. Thus, passive immunotherapy with bNAbs, either alone or in combination with LRAs and therapeutic vaccines, has been shown to reduce viral load and delay viral rebound following ART interruption [81].

One should keep in mind that although the cited study (NCT04357821 [177]) showed promising results for a multiple-combination therapy for HIV-1 treatment, it was a single-arm study. Therefore, it was not possible to compare the combination treatment with either placebo or any of the individual components and to determine their respective contributions to the observed effects. Such contributions can, in part, be assessed based on other trials evaluating combinations of bNAbs and LRAs. Hsu et al. [147] and Nel and Frater [81] reviewed attempts to increase bNAb levels and the vaccinal effect by combining vaccination and immune-modulating LRAs. They concluded that, despite promising results in NHP and PLWH, the addition of LRAs to bNAbs provided no additional benefit compared with bNAbs alone [81,147]. This suggests that the main contribution to the reduced viral load and delayed viral rebound observed after ART interruption [81] was attributable to the combination of bNAbs and therapeutic vaccination.

Overall, the induction of the vaccinal effect is not consistent and appears to depend on both the virological status of the individual and the therapeutic strategy employed. This may explain why only 10–35% of treated PLWH achieve temporary control of viral load during ATIs [178]. Further research is needed to achieve more consistent and durable responses.

3.5. Addressing HIV-1 Drug Resistance

In addition to preventing progression toward AIDS, therapeutic vaccines could potentially prevent the development of drug resistance (DR) during HIV-1 infection, excluding cases in which individuals are infected with drug-resistant HIV-1. Vaccines targeting the development of drug resistance could hinder the evolution of HIV-1 toward resistance in PLWH who are receiving ART, reduce the incidence of DR and the need to switch to second- and third-line ART regimens, thereby prolonging the effectiveness of first-line ART and reducing treatment costs. Vaccination against DR in HIV-1 infection would be prophylactic in terms of preventing the emergence of resistance, yet therapeutic because it would be administered to individuals already living with HIV-1 [179].

To date, there are no vaccines specifically targeting DR HIV-1 variants. None of the currently investigated therapeutic vaccines specifically target DR mutations, despite the substantial changes in T-cell epitopes associated with these mutations and their potential recognition by the immune system (summarized in [179]). Importantly, several drug-resistance mutations have been shown to confer not only resistance to ART but also immune escape [180,181,182], potentially giving drug-resistant HIV-1 a dual advantage.

Several groups have conducted early preclinical studies [179]. Wahren et al. developed DNA, peptide, and fusion-protein prototype vaccines based on the HIV-1 clade B reverse transcriptase (RT) [183] and protease (PR) [184,185] carrying primary DR mutations, which were shown to be immunogenic in mice [186]. DNA vaccine candidates were also developed based on consensus sequences of PR [181], integrase (IN) [187] and RT [188,189] from HIV-1 clade A (FSU-A/IDU-A). All three proteins carrying primary DR mutations were shown to be immunogenic in mice. Experimental models were developed to assess the protective efficacy of these vaccine candidates using murine adenocarcinoma cells engineered to stably express HIV-1 enzymes carrying DR mutations (DR-RT, DR-IN, and DR-PR) [181,187,190]. DNA immunization with DR-PR was shown to protect mice against challenge with tumor cells expressing DR-PR, with protection mediated by CD8+ T cells [181]. DNA immunization with DR-IN provided partial protection, with tumor size inversely correlating with the magnitude of the IN-specific IFN-γ/IL-2 T-cell response [187]. DNA immunization with DR-RT protected mice against challenge with tumor cells expressing DR-RT, with tumor size inversely correlating with the magnitude of the RT-specific IFN-γ T-cell response [189].

Overall, therapeutic vaccines specifically targeting DR HIV-1 variants remain an area of active preclinical research [179]. Such vaccines could address an important unmet clinical need, analogous to vaccination strategies targeting drug resistance in other viral infections, although their path to clinical application is likely to be long [191]. Vaccination against DR HIV-1 would not be intended for all PLWH, nor for all individuals in the acute or chronic phases of infection, but rather for a relatively narrow group of PLWH at high risk of developing DR HIV-1 and its associated complications.

4. Lessons Learned from the Safety and Efficacy of Vaccines Against Other Pathogens in PLWH

A major concern regarding therapeutic HIV-1 vaccination is whether vaccination can be effective in PLWH at all. PLWH may be immunocompromised, and although ART restores immune function, recovery is often incomplete [192]. Another concern is that vaccination, by inducing T-cell activation, could increase the HIV-1 reservoir and accelerate disease progression. Experience with commonly used vaccines against other pathogens in PLWH helps to address these concerns.

PLWH are routinely and effectively vaccinated with a range of commonly used vaccines, although some require additional booster doses or higher vaccine doses. Clinical trials of common vaccines in PLWH provide valuable information on vaccine safety, optimal dosing, prime–boost regimens, and the durability of immune responses, including the need for revaccination. This knowledge could inform the development of therapeutic HIV-1 vaccines and potentially reduce the number of preclinical and clinical trials required.

The experience gained from the use of common vaccines in PLWH should be leveraged to accelerate progress toward successful therapeutic HIV-1 vaccine development and reduce development costs.

A range of commonly used vaccines is routinely administered to people living with HIV-1, as summarized in Table 3 and web resources describing routine vaccines recommended for use in PLWH [194,195].

With the exception of live attenuated viral vaccines, there are no major differences between PLWH and the general population with respect to vaccine safety, immunogenicity, and protective efficacy, although some vaccines require higher doses, additional doses, or booster doses (see [194,195], Table 3, and references therein). Others, such as pertussis, diphtheria, and tetanus vaccines, can be successfully administered using the same vaccination schedules as in the general population [193,194,195]. However, immune responses induced in PLWH may show shorter durability and lower antibody levels, as illustrated by the waning of protective IgG antibodies induced by childhood vaccination in vertically HIV-1-infected children compared with age-matched healthy controls (Figure 1 and original article [209]).

Vaccination did not induce significant changes in viral load or CD4+ T-cell counts [204,213,214]. Interestingly, for a number of vaccines, including HBV and SARS-CoV-2 vaccines, control of HIV-1 viremia, reflected by low HIV-1 RNA levels and restored CD4+ T-cell counts, predicts vaccine efficacy. Another important difference between PLWH and the general population which should be taken into consideration is the durability of immune responses, which is significantly shorter in PLWH [200,209], both following administration of common vaccines and in trials of therapeutic HIV-1 vaccines.

Taken together, studies of the efficacy of common vaccines in PLWH demonstrated, first and foremost, the safety of vaccination in this population. Second, these studies demonstrated that PLWH are capable of mounting effective immune responses that protect against multiple infections. Importantly, although research findings and clinical trials have demonstrated the need for higher doses, additional booster doses, or novel adjuvants [200,206,212,216], these findings are not reflected in current clinical guidelines [193,194]. Overall, these studies indicate that the fundamental challenge in developing therapeutic HIV-1 vaccines is not the inability of PLWH to mount protective immune responses in general, but rather the difficulty of inducing protective or therapeutic immune responses specifically against HIV-1. More broadly, the challenge may, at least for now, lie in our inability to design immunogens capable of inducing such responses.

5. Additional Issues to Address and Questions to Answer

A number of intriguing questions remain for which we do not have a complete or definitive answer yet (Figure 2). Here, we review these questions and briefly discuss plausible answers.

5.1. What Is the Duration of Immune Responses Induced by Therapeutic HIV-1 Vaccines?

There is no definitive answer even for common vaccines. Certainly, none provides lifelong immunity. The situation appears to be particularly unfavorable for hepatitis B and mumps vaccines, for which protection decreases by more than two-fold within five years (Figure 1). A similar time frame has been observed for the clinical effects of therapeutic HIV-1 vaccination. In an early study of therapeutic immunization with rgp160 combined with ART (Nordic VAC-04), an increase in CD4+ T-cell counts compared with baseline and significantly fewer deaths among vaccine recipients than in the control group were observed at 2 years (104 weeks), but not at later time points [67]. This suggests that the positive contribution of anti-gp160 immune responses to the clinical course of HIV-1 infection may persist for up to two years. In a much later and more complex trial combining the histone deacetylase inhibitor romidepsin as a latency-reversing agent with the MVA.HIVconsv vaccine (BCN02 study), 23% of participants undergoing a monitored ART interruptions maintained suppression of viremia for more than 32 weeks (less than one year) [111].

The duration of an immune response capable of restricting viral replication can, although indirectly, be assessed by the time from ART cessation to the end of ATI (time off ART), which marks the loss of immune control over viral replication. This approach provides an estimated duration of 10–30 weeks. However, this should be considered an indirect measure of immune response durability, as the time off ART depends strongly on the potency and specificity of both pre-existing and vaccine-induced anti-HIV-1 immune responses. During ATIs, vaccine recipients are continuously exposed to increasing numbers of HIV-1-infected cells, creating an ongoing challenge to immune control. Indeed, in the AELIX-002 and AELIX-003 clinical trials of HTI immunogen-based vaccines administered alone or in combination with the TLR7 agonist, Bailón et al. demonstrated that participants who remained off ART for more than 12 weeks (3 months) could be identified by a threshold of 835 HIV-1 vaccine-specific spot-forming cells per 10⁶ peripheral blood mononuclear cells [128].

It is also important to note that the durability of immune control induced by therapeutic HIV-1 vaccination is likely to depend strongly on the clinical characteristics of the individual. PLWH who initiate ART soon after the beginning of HIV-1 infection are likely to have better immune function and more durable immune responses than those with chronic HIV-1 infection who initiate ART later.

5.2. Can Therapeutic HIV-1 Vaccination Be Harmful by Increasing CD4+ T-Cell Susceptibility to HIV-1 Infection?

Viral blips in therapeutic HIV-1 vaccine recipients who remain on ART are relatively uncommon and have not been considered a consistent safety signal of therapeutic HIV-1 vaccines. However, if vaccination activates T cells generated earlier during infection, it could potentially promote the expansion of chronically activated CD4+ T cells, which are known to be more susceptible to HIV-1 infection and could thereby contribute to ongoing viral replication [123].

This may explain the viral blips reported in therapeutic HIV-1 vaccine recipients. The DNA vaccine “DNA-4,” targeting Nef, Gag, Pol (RT), and gp140 of HIV-1 clade A, was shown to be safe and immunogenic in Phase I and IIa clinical trials; however, vaccine recipients experienced spontaneous increases in viral load that were substantially greater in magnitude than the blips observed in the placebo group [49]. These blips were interpreted as evidence of viral reservoir reactivation, i.e., activation of dormant HIV-1-infected CD4+ T cells, which is an intended goal of some anti-HIV-1 therapeutic strategies. However, the findings could also be interpreted as manifestations of new HIV-1 infection of activated CD4+ T cells, occurring more frequently or to a greater extent in vaccine recipients than in the placebo group. Neither explanation can be confirmed, as the study did not assess the size of the replication-competent HIV-1 reservoir or levels of total and integrated proviral DNA.If vaccination activates T cells of new specificities, it should not enhance HIV-1 infection, as demonstrated by studies of common vaccines in PLWH. None of the clinical trials of common vaccines reported adverse effects of vaccination on the course of HIV-1 infection. Furthermore, MacGregor et al. [87] observed fewer viral blips among recipients of an HIV-1 DNA vaccine than among placebo recipients. In the placebo group, viral blips occurred in 3 of 5 participants, comprising 10 blip events among 49 viral load measurements, whereas in the vaccine group, blips occurred in only 1 of 13 participants, comprising a single event among 130 measurements. Overall, results from trial indicate that although viral blips have been observed among therapeutic HIV-1 vaccine recipients, vaccination, including therapeutic HIV-1 vaccination, has not been associated with a significant increase in the frequency of viral blips [87].

Based on the available data, therapeutic HIV-1 vaccination does not appear to increase the susceptibility of CD4+ T cells to HIV-1 infection. However, the immunogen itself may have detrimental effects.

A related question is whether viral blips could “naturally” boost the immune response induced by therapeutic vaccination through the reappearance of HIV-1 proteins. Our answer would be rather not, as a therapeutic vaccine should ideally induce immune responses with specificities distinct from those induced by the natural HIV-1 infection.

5.3. Can HIV-1 Vaccines Which Have Failed in Prophylactic Trials Be Repurposed as Therapeutic Vaccines If They Induce CD8+ T-Cell Responses?

Prophylactic vaccines that failed to prevent HIV-1 infection could potentially be repurposed as therapeutic vaccines if they induce sufficiently strong cell-mediated immune responses to control viral replication. Such an approach could substantially reduce early-stage development and production costs, as Phase I/II safety and T-cell immunogenicity data may already be available and scaled-up manufacturing processes may already be established. This concept is illustrated by the therapeutic potential of mosaic immunogens designed to induce strong T-cell responses, which failed to demonstrate efficacy in prophylactic trials, such as Mosaico and Imbokodo [48,90]. The Imbokodo mosaic vaccine is also being explored in a therapeutic setting. In the Mosaico Phase I trial, vaccination with Ad26.Mos4.HIV/MVA-Mosaic (Ad26/MVA) or Ad26.Mos4.HIV/Ad26.Mos4.HIV combined with adjuvanted gp140 protein (Ad26/Ad26+gp140) was shown to be safe, well tolerated, and highly immunogenic, with the protein boost resulting in increased binding and neutralizing antibody titers [219]. Of particular interest were the increases in T-cell responses [219]. Although these immune responses were insufficient to protect against HIV-1 infection, the authors suggested that they might nevertheless have therapeutic potential by reproducing or mimicking aspects of the immune control of viral replication observed in elite controllers. Other prophylactic vaccine candidates that failed to demonstrate protective efficacy but induced strong CD8+ T-cell responses could similarly be considered for evaluation in therapeutic trials.

5.4. Will Therapeutic Vaccines Based on HIV-1 Clade B Be Effective Against Other HIV-1 Clades?

Most therapeutic vaccines have targeted HIV-1 clade B, with a few targeting clade C and only one targeting clade A [49]. All major HIV-1 group M clades, including A, B, C, D, and CRF01_AE, share substantial portions of their genetic sequences, and many T-cell epitopes are conserved across clades. Therefore, the answer could potentially be “yes.” If a therapeutic vaccine induces strong CD8+ T-cell responses against conserved regions, these T cells may recognize cells infected with HIV-1 variants from non-B clades. This is the rationale behind vaccines such as HIVconsvX and HTI, which are specifically designed to focus immune responses on relatively conserved HIV-1 regions and thereby maximize cross-clade recognition. Particularly promising in this regard are mosaic immunogens, which are engineered to increase coverage of global HIV-1 sequence diversity and have demonstrated broader T-cell recognition in preclinical and clinical studies [100,101,219].

Still, the effect could be weaker against non-B clades than against HIV-1 clade B itself because of genetic differences between clades. Even single amino acid substitutions can alter T-cell recognition and the ability of T cells to kill infected cells. In addition, clade B is prevalent in Europe and North America, whereas other clades predominate in different geographical regions and populations. Differences in the distribution of HLA class I and class II haplotypes across populations may further affect recognition of vaccine-induced epitopes and the ability to generate beneficial immune responses.

Hence, although many therapeutic HIV-1 vaccine studies have demonstrated cross-reactive immune responses in laboratory assays, these responses have not yet translated into demonstrable clinical benefits.

5.5. Does Prior Participation in a Prophylactic HIV-1 Vaccine Trial Affect the Subsequent Course of HIV-1 Infection?

Despite several studies of breakthrough infections in HIV-1 vaccine trials [220,221,222,223,224], important questions remain regarding the subsequent course of HIV-1 infection in previously vaccinated individuals. Those who became infected despite vaccination generally did not develop an elite-controller phenotype, nor did they become rapid progressors. The course of infection following breakthrough infection was remarkably similar to that in unvaccinated individuals, although there were indications that some vaccines exerted modest early immune pressure on the virus [225].

The STEP trial (Merck Ad5 Gag/Pol/Nef vaccine) was one of the most important tests of a T-cell-based HIV-1 vaccine. Participants who became infected demonstrated (i) no durable reduction in viral load, (ii) no meaningful improvement in clinical outcomes, and (iii) no evidence of developing an elite-controller phenotype [224]. The STEP trial also provided no evidence that pre-existing HIV-1 immunity increased susceptibility to infection or altered the subsequent clinical course of disease.

There was some evidence of transient immune pressure on the founder virus and possibly modest short-term effects during acute infection, although none of these effects translated into long-term control of viral replication. One can speculate that the effects of vaccination on breakthrough HIV-1 infection may depend on the type of vaccine and differ between vaccines designed primarily to induce T-cell responses and those designed to induce antibody responses.

5.6. Can Immune Responses Against Drug-Resistance Mutations Be Induced to Prevent the Development of Drug Resistance?

Therapeutic HIV-1 vaccines designed to induce either B-cell- or T-cell-mediated responses could potentially serve as a backup to, or temporary substitute for, ART. An open area of research is their use to prevent or control the development of drug resistance[179]. The data described above in Section 3.5 point to the potential feasibility of this approach. However, none of the drug-resistance-targeting immunogens developed to date has entered clinical trials. Therefore, whether these effects can be achieved in PLWH remains to be determined.

5.7. If a Therapeutic HIV-1 Vaccine Is Developed, Who Should Receive It to Ensure Cost-Effectiveness?

In contrast to a preventive vaccine, which should be widely available to high-risk populations, a therapeutic vaccine is intended for use in populations in which it can complement standard ART and provide a safeguard in the event of treatment interruption. One such population is vertically HIV-1-infected children. Even moderately effective therapeutic HIV-1 vaccines may provide a modest but meaningful increase in life expectancy and may compare favorably with alternative uses of limited resources allocated to HIV-1 treatment. Moreover, if the total cost of a highly effective life-long vaccination course could be as low as USD 2,000, therapeutic vaccination may be cost-saving for public health care systems [226]. However, African populations have been shown to be significantly more sensitive to vaccine price than U.S. populations putting HIV-1 vaccination related cost-savings in these countries under question [227]. The inability of high-risk populations to afford HIV-1 vaccines in their todays (modern) format, creates a need for these vaccines to be provided either free of charge or at a reduced price. This, in turn, would create a heavy burden for health care systems and give danger signals to the developers, stimulating them to account for the potential financial losses in the overall cost of vaccine development, thereby increasing the final cost of HIV-1 vaccine production and application [228].

6. Conclusions and Perspectives

We would like to conclude this review by quoting a philosophical statement generated by ChatGPT: “From a scientific perspective, breakthrough infections in vaccine recipients taught researchers something surprising: inducing HIV-specific immunity is much easier than inducing HIV-1 control. The former has been achieved repeatedly; the latter remains one of the central challenges of HIV-1 vaccine development.”

Considerable progress has been made in the field of therapeutic HIV-1 vaccination. Although none of the therapeutic HIV-1 vaccines evaluated to date has achieved a sterilizing cure or even sustained ART-free remission, accumulating evidence demonstrates that they can substantially enhance HIV-1-specific immunity, have the potential to delay viral rebound following ART interruption, and remain compatible with a wide range of HIV-1 functional cure strategies. These features position therapeutic HIV-1 vaccines as a potentially important component of combinatorial approaches aimed at achieving a functional cure for HIV-1.

Furthermore, when debating which modality—antibody-based or CD8+ T-cell-based—is superior, or which approach holds greater promise for inducing bNAbs versus CD8+ T-cell responses, it is important to recognize that these debates parallel the question of which individual ART component is most effective. The history of HIV-1 treatment provides a clear lesson: no single antiretroviral agent can fully suppress HIV-1 replication, just as no single bNAb can neutralize all viral variants. What has proven effective—and is likely to remain effective—is the combination of complementary agents and immune mechanisms.

An additional consideration is the high cost of many proposed technologies, particularly when therapeutic vaccination is combined with latency-reversing agents and immune checkpoint inhibitors, as well as the need to identify the groups of PLWH for whom these expensive interventions would be cost-effective. For now, priority should be given to establishing whether therapeutic HIV-1 vaccination is feasible as a proof of concept. It is reasonable to anticipate that, once this question is answered affirmatively, the cost of such interventions, initially high, may eventually decrease, following a trajectory similar to that observed in cancer therapy. Thus, the path forward lies not in selecting a single approach, but in embracing synergy, pragmatism, and the long-term goal of an accessible functional cure.