Most HIV Strains Can Escape the Immune System’s Best Antibodies With a Single Mutation

Seven thousand, seven hundred and seventy-six. That is how many parallel viral selection experiments researchers at Rockefeller University ran in order to build what is now the most comprehensive map of HIV-1’s escape routes from its most feared opponents. The antibodies in question are broadly neutralizing antibodies, or bNAbs, proteins so potent that a single injection can keep a patient’s viral load near undetectable for months at a stretch. The question the team was asking, through all those tiny wells and all that fluorescent cell media, was a straightforward but consequential one: how easily can the virus find its way around them?

The answer, it turns out, depends very much on which strain of HIV you are dealing with. And that variability is exactly the kind of knowledge that could determine whether bNAb therapies ever move from promising clinical novelty to genuine long-term cure.

BNAbs have generated real excitement in HIV research over the past decade or so. Unlike conventional antiretroviral drugs, which patients must take daily to suppress replication, bNAbs are monoclonal antibodies given by infusion, sometimes just once, that can neutralize a wide range of viral strains. They occur naturally in roughly 1 to 5 percent of HIV-positive individuals, people whose immune systems have stumbled onto something the other 95 percent haven’t. Rockefeller’s Michel Nussenzweig and Marina Caskey identified and developed two of the most clinically advanced of these antibodies, 3BNC117, which targets the site on the virus where it binds to immune cells, and 10-1074, which hits a different region of the viral surface altogether. Used together, the pair have suppressed viral loads for up to a year in some trial participants, a result that once would have seemed fantastical.

Resistance, though, is a persistent shadow. HIV-1 is among the fastest-mutating pathogens known to science, and it has been outwitting antibodies since long before humans started trying to turn that trick against it.

To understand the full landscape of possible escape, Paul Bieniasz and Theodora Hatziioannou’s lab developed a pipeline they called RISC, resistance identification via selection and cloning. First author Alex Stabell, an infectious disease physician working as a clinical scholar at Rockefeller, grew large batches of fifteen HIV-1 strains collected from patients across four genetic subtypes and four continents. Replication errors during growth naturally produced a genetically diverse population, some members of which might, by chance, already carry a mutation conferring resistance. Stabell then seeded thousands of tiny wells with these populations, each well containing a bNAb at one of three concentrations. Wells where virus kept spreading despite the antibody flagged possible escape mutants. Those viruses were isolated, sequenced, and tested again in standardised neutralization assays to confirm the resistance was real.

“We found that most viral strains can escape bNAb neutralization, but there’s substantial variation in the likelihood that they will and the mechanisms that enable it,” Stabell says. “Knowing how different strains of the virus respond to leading bNAb therapies will greatly improve our ability to anticipate whether a particular therapy will be effective for individual patients,” adds Bieniasz.

The headline finding is both impressive and, in a certain light, sobering. For twelve of the fifteen strains tested against 3BNC117, and for every strain tested against 10-1074, a single amino acid change was sufficient to confer meaningful resistance. One mutation. HIV-1 is a virus that generates millions of variants every day inside an infected person, so this is not a question of whether these escape variants will arise in a patient’s body but of when, and whether the antibody will be deployed before they take hold. “It was striking that it’s actually quite easy for most HIV strains to escape these special antibodies,” Bieniasz says. The finding may reframe how clinicians think about bNAb monotherapy: effective for short-term suppression, perhaps, but probably insufficient on its own as a durable strategy against a virus this genetically resourceful.

Not All Strains Are Equal

A handful of the viruses Stabell worked with defied the pattern. One, the TRO.11 strain, required the accumulation of several mutations before it could escape 3BNC117 at all, each additional substitution edging the virus further toward resistance. Another, CH607, found an altogether different route: rather than evolving resistance to free-floating virus particles, it developed mutations that made it better at passing directly from infected cell to uninfected cell, bypassing the antibody-rich environment entirely. Neither of these is easy for HIV to pull off. “The genetic barrier to resistance was higher for these viruses,” Stabell notes. “One of the goals of therapy these days is not simply to have therapies that are transiently effective, but to have this high genetic barrier.” Finding bNAb combinations that force HIV toward such difficult escape routes is now a concrete research target.

The study also turned up mutations in places no one expected them. For 3BNC117, which targets the CD4 binding site on the viral envelope, some resistance-conferring mutations arose well outside that binding site, in regions of the protein not previously known to affect antibody sensitivity. “These were quite prominent and unexpected,” says Hatziioannou. “No one would have predicted these would affect bNAb sensitivity.” This matters partly because clinical trials sometimes screen patients for pre-existing resistance by testing their virus against the antibody in the lab; if resistance can be hiding in novel locations, those screening assays may be missing more than researchers assumed.

The Case for Combination Therapy

The escape mutations were also, for the most part, highly strain-specific. A mutation that let one HIV strain dodge 3BNC117 rarely conferred the same benefit when transplanted into a different strain’s genetic background, suggesting that viral context shapes which escape routes are actually accessible. This has real implications for treatment design. Some mutations found to confer 3BNC117 resistance also rendered the virus more vulnerable to other bNAbs, a kind of evolutionary trade-off that combination therapy could exploit. Others, however, caused cross-resistance, meaning a single mutation could blunt the effect of multiple antibodies at once. One class of mutations, those affecting a structural element called the beta20 strand, appears to push the viral envelope protein into a more open configuration, which makes it resistant to a surprising range of antibodies targeting different epitopes. Understanding which mutations have these broad effects should help researchers design bNAb pairings that are harder to evade simultaneously.

The numbers underlying this work give some sense of what clinicians are up against. Based on the frequency of resistant variants the team measured in their lab populations, and extrapolated to the sizes of viral reservoirs known to persist in patients on antiretroviral therapy, it seems likely that most HIV-positive individuals already harbour viruses capable of resisting at least one bNAb. Resistance variants aren’t absent before treatment; they are simply outnumbered, waiting. “HIV-1 mutates so fast and the diversity in the population is already quite enormous, so we’ve long known that a multidrug approach is the best course of treatment,” Hatziioannou says. The new resistance map, she hopes, will let researchers move past that general principle and toward something more specific: “We hope to identify combinations that potentially raise the genetic barrier to resistance and are therefore more effective.” Which bNAb with which other bNAb, and against which range of viral backgrounds, is a question this pipeline is now equipped to answer, one exhaustive selection experiment at a time.

https://doi.org/10.1038/s41564-026-02347-x

Frequently Asked Questions

If broadly neutralizing antibodies are so powerful, why can HIV escape them so easily?

BNAb breadth comes from targeting regions of the virus that are structurally constrained, meaning HIV can’t easily change those sites without losing its ability to infect cells. But “constrained” isn’t the same as immutable: in most strains, the virus turns out to have at least one single amino acid change available that confers resistance without crippling its fitness. The antibodies are potent against the strains they encounter, but HIV’s sheer genetic diversity means resistant variants are probably already present in most infected individuals before treatment even begins.

Could doctors screen patients before giving bNAb therapy to check if their virus is already resistant?

This is already done in some clinical trials, but the new research suggests those screens may miss more than researchers thought. Resistance mutations can arise outside the antibody’s known binding site, in regions not typically tested, and standard assays may not sample enough of the viral reservoir to detect rare resistant variants reliably. A more useful approach might be to test representative viral isolates from each patient and measure directly how easily resistant mutants arise, rather than simply looking for known resistance mutations.

What makes a “high genetic barrier to resistance” and why does it matter for HIV treatment?

A high genetic barrier means the virus needs many simultaneous mutations, rather than just one, to escape a therapy. For most HIV strains, a single amino acid change is enough to overcome individual bNAbs, which is a low barrier. A handful of strains required multiple cumulative mutations or unconventional escape routes like enhanced cell-to-cell transmission, making them much harder to treat with single antibodies. Finding bNAb combinations that push every strain toward those harder escape paths is the central goal driving the next phase of this research.

Is this why HIV vaccines have been so difficult to develop?

Partly, yes. The same viral diversity that creates multiple escape routes for therapeutic antibodies also makes it nearly impossible to design a vaccine that trains the immune system against all circulating strains at once. The roughly 1 to 5 percent of people who naturally develop bNAbs do so only after years of evolving immune responses against a constantly shifting target. Understanding the escape landscape doesn’t immediately solve the vaccine problem, but it does clarify exactly how much genetic coverage any future vaccine or antibody therapy needs to provide.