An antibody from a Tanzanian woman discovered during screening for anti-HIV antibodies shows a strong therapeutic potential in a preclinical study. Named 04_A06, it neutralised (blocked) 97.3% of over 300 HIV strains tested, and blocked 77% of viruses resistant to other antibodies. In humanised mice – designed to have an immune system like ours – the antibody completely suppressed viral load after stopping treatment for more than a month.
Apart from the potential therapeutic promise, the broadly neutralising antibody could also be valuable for prevention as the same researchers modelled it to confer 93% protection when used as PrEP. Findings were published by Dr Lutz Gieselmann and colleagues in the journal Nature Immunology.
The place of antibodies in HIV treatment and prevention
Antibodies are protein molecules produced by the B-cells of the immune system that fulfil several key roles in our protection, one of which is to find and bind to pathogens (disease causing agents). When an antibody attaches to a virus, it can weaken or completely block the virus’ ability to infect cells. Some antibodies stand out for their capacity to neutralise a wide range of viral strains at once – they are called broadly neutralising antibodies, and are the ones sought after in HIV research.
However, antibodies, like all other treatment approaches, are not perfect. While they are effective at very low doses and can stay in the blood for weeks, allowing for injectable, long-acting formulations, the virus often develops resistance to them.
Because antibodies are large molecules and often carry an electrical charge, most broadly neutralising antibodies cannot enter cells and have to act outside – in the blood and other body fluids. This means, they cannot target post-entry stages of the viral lifecycle, which limits what antibody-based treatments can do. For this reason, most antibody strategies target the viral glycoprotein gp120, which is important for viral entry into the cell.
The gp120 glycoprotein binds to a protein receptor on the CD4 cells, tricking the cell into letting the virus inside. Although gp120 can mutate to evade most antibodies, it contains some highly conserved regions. These conserved molecular patterns on the core of gp120 have to remain the same, because mutations there would cripple the virus’s ability to bind CD4 to trigger a cellular entry.
This is where antibody 04_A06 shines, as it tightly binds to these conserved regions. The antibody has a tiny extra piece in its structure – just 11 building blocks of protein – that lets it reach a part of the virus most antibodies can’t touch. This extra reach allows it to grab the virus more firmly, locking onto a region that hardly ever changes. Because of that, the virus finds it much harder to mutate its way out of the antibody’s grip.
Antibody screening
The researchers at the University of Cologne, Germany, recruited participants from HIV clinics and hospitals. Most of the participants – 44%, were from Tanzania, followed by Germany (25%), Nepal (25%) and Cameroon (6%). Of the recruited participants, 47% were female and 66% were off treatment. Serum samples were collected from 2354 eligible participants and screened for neutralising antibodies. Thirty-two participants (3.7% of the cohort) were found to be elite neutralisers.
Elite neutralisers shouldn’t be confused with elite controllers. While conceptually similar, elite controllers are people with HIV who can keep their viral load suppressed without treatment, while elite neutralisers are people with HIV whose immune systems produce strong, broadly neutralising antibodies which may or may not achieve complete viral suppression clinically.
Testing against HIV strains
The antibodies and B-cells of the elite neutralisers were further analysed: 831 neutralising antibodies were identified and were then ran against a panel of six strains of HIV. Only seven of the antibodies were able to block all six strains and they all came from two of the elite neutralisers. However, the strongest ones came from participant EN02 – a Tanzanian woman.
Three of her antibodies were then used in head-to-head comparisons, including the highest-performer 04_A06.
Initially, antibody 04_A06 was tested against a panel of 12 HIV reference strains. Reference strains are a collection of laboratory versions of the virus chosen to represent the major global genetic variants. 04_A06 blocked all of the strains at lower doses than many antibodies that have been clinically investigated.
Following this, the antibody was tested against a very large reference of 337 lab strains of HIV, where only 9 strains showed resistance. While lab reference strains are very useful for overall screening, ‘real’ viruses from people with HIV are generally harder to neutralise. For this reason, next the antibody was tested against 50 replication-competent viral strains obtained from people with HIV, resulting in 88% neutralisation at relatively low doses.
The next step was to assess 04_A06’s neutralising capacity against viral strains resistant to VRC01. This broadly neutralising antibody belongs to the same group as 04_A06 as they both target the binding site of gp120 to the CD4 cell receptor. VRC01 is clinically advanced and has already undergone safety and efficacy trials in people, however has not been very promising when used as a single-drug therapy – it delays viral rebound but fails to maintain suppression and leads to the development of resistance.
04_A06 neutralised 77% of all strains resistant to VRC01 with a higher potency (lower dose) than all other tested broadly neutralising antibodies.
Testing on humanized mice
Before testing the antibody’s inhibitory performance in vivo (in living organisms, in this case mice), the researchers first ran it through a panel of viruses with mutations to the gp120 glycoprotein known to cause resistance or reduced susceptibility to other clinically advanced anti-gp120 antibodies. After confirming that none of these mutations was capable of conferring resistance to the virus, they moved on to the in vivo testing in humanised mice. These mice received transplants of human immune cells, giving them an immune system that responds to HIV more like a human body would. But they don’t perfectly replicate human immune responses.
In the live-animal tests, the researchers compared 04_A06 to VRC01 and its close relative VRC07. Some mice received these well-known antibodies, while others were put on 04_A06.
While VRC01 and VRC07 led to viral rebound and development of resistance after a temporary three-week reduction in viral load, 04_A06 suppressed viral load in all mice, maintained throughout 12 weeks of dosing. What’s more interesting is that none of the arising mutations during 04_A06 treatment reduced the virus’ susceptibility to the antibody.
The mice that were initially treated with VRC01 and experienced a treatment failure were then treated with 04_A06. These mice were all virally suppressed at eight weeks on 04_A06, despite having failed on VRC01. Upon stopping 04_A06, it took the mice on average 39 days to experience a viral rebound. Three of the 13 mice had no viral rebound at nine weeks after treatment with 04_A06 ended. Analysis of the gp120 mutations in these mice showed there was no viable selection of mutations that could decrease the antibody’s effectiveness.
This led the researchers to conclude that 04_A06 single-drug therapy was effective at suppressing the virus, even in mice resistant to VRC01, and lacked a clear tendency towards resistance selection. This is unusual for HIV and supports the thesis that the antibody binds to a highly conserved area on gp120 critical for the virus’ fitness. However, it’s important to note that the mice were infected with only a single viral strain which may not reflect real-world scenarios.
Testing its potential in HIV prevention
Treatment is not the only branch of HIV research that could potentially benefit from antibodies. Because some of these antibodies can stay in the blood for months following a single dose, they could be used as an effective way to passively immunise people against HIV.
One such effort were the two Antibody-Mediated Prevention (AMP) trials that tested VRC01’s prophylactic potential against HIV and ended in early 2021. Together, they are the largest study to test antibody-based HIV prevention, conducted across Africa, the Americas, and Europe.
The researchers in the present study tested the efficacy of 04_A06 against strains that were taken from the AMP trials; both from the placebo (untreated) arm where participants didn’t receive any prevention treatment and the treated group who had a breakthrough infection despite VRC01 prophylaxis. 04_A06 inhibited 98% of the viruses taken from the participants on the placebo arm and 94% of the viruses from the participants that were on VRC01 when the infection occurred.
Using this data and the available pharmacokinetic data for 04_A06, the researchers estimated that a version of the antibody modified to make it stay longer in the blood could provide 93% protection against HIV if used as PrEP.
Concluding thoughts
Interest in antibody-based HIV treatments declined in the early 2010s when most antibodies failed to cover enough viral strains and HIV quickly developed resistance to them. However, interest revived later in the decade following the discovery of more potent broadly neutralising antibodies and a shift toward using them in combination 'cocktails' – similar to antiretroviral therapy (ART) – for both prevention and treatment.
The diversity of antibodies that can be produced is practically limitless, meaning that there will be antibodies that target the most critical parts of the virus. In this context, 04_A06 appears close to an ideal antibody; it binds precisely to the CD4-binding site on HIV’s outer protein, gp120, and includes an unusual 11-amino-acid extension that lets it reach across a wider area of the viral surface. Its ability to fully suppress viruses resistant to other antibodies with the same target additionally reinforces its potential and the lack of the emergence of clear resistance patterns to 04_A06 makes it very promising.
Although the study is elaborate, these results need to be reproduced to confirm all the findings. It’s important to note that this is a laboratory study, in other words – pre-clinical. Even though the researchers used state-of-the-art methods to mimic the human organism in mice, these findings cannot yet be considered equivalent to human clinical trials. If future laboratory studies continue to show the same promising effects of 04_A06, only then will clinical trials be able to confirm its true potential.
Technical note: All the breadth percentages (the number of viral strains blocked out of the total tested) in this article refer to the antibody’s IC80 performance, which is the concentration at which it blocks 80% of the virus’ activity in lab. IC80 gives a better idea of how well an antibody might work in the body, compared to IC50 which measures only 50% blocking.
The only exception is the 77% figure for viruses resistant to VRC01 in the first paragraph, which is the antibody’s inhibitory breadth at IC50, since the study didn’t report an IC80.
Gieselmann L et al. Profiling of HIV-1 elite neutralizer cohort reveals a CD4bs bnAb for HIV-1 prevention and therapy. Nature Immunology, advance online publication, 6 October 2025 (open-access).
https://doi.org/10.1038/s41590-025-02286-5