Within two to three years—at most—an array of HIV vaccines currently in the pipeline will be moving into large-scale Phase III clinical trials involving thousands and thousands of volunteers. The ability to distinguish vaccine-generated antibodies from actual early HIV infection among these multitudes will be imperative, says Hana Golding.

It will also be a reality.

Thanks to a Bench-to-Bedside award, Golding, chief of the Laboratory of Retrovirus Research at the Center for Biologics Evaluation and Research, FDA, and her team (postdoc Surender Khurana and fellow James Needham) have developed an alternative assay that, unlike currently licensed HIV-1 detection kits, is designed to yield a negative result in the face of vaccine-generated HIV-1 antibodies.

And on a more global level, as the numbers of trial participants balloon, false-positive test results can become a major problem for blood and plasma collection centers, as well as a source of discrimination in employment, insurance, and other facets of life.

"We have identified to our satisfaction two epitopes that will generate antibodies only in the presence of true infection," says Golding, noting that the validity of large-scale HIV vaccine trials involving high-risk participants will depend on being able to identify quickly those who get infected. The vaccine regimen would then be halted, and they would be transferred to another arm, where any effect of the vaccine could be followed and where they might be eligible for HIV therapy, according to the standard of care at the site of the trial.

Without her Bench-to-Bedside collaboration with Barney Graham, chief of the Viral Pathogenesis Laboratory and director of clinical studies at the Vaccine Research Center, NIAID, this work could not have been accomplished, she says. Further support will come from OAR, DAIDS, and NHLBI.

The epitopes had to meet quite specific requirements:

To find these epitopes, Khurana constructed a gene-fragment phage-display library expressing segments from the entire open reading frame of the HIV-1 genome. The library was used to screen sera drawn from individuals within three months of initial HIV infection to establish the early repertoire of HIV-1 antibodies. Many were already components of existing HIV detection kits.

Gag-p6 and gp41 peptides met all criteria. The gp41 peptide identified is located in the cytoplasmic tail of the envelope glycoprotein and, unlike extracellular gp41, has no neutralizing epitopes. Similarly, Gag-p6, a small part of the much larger Gag-Pol protein, shows no evidence of raising neutralizing antibodies or protective cellular immunity.

"We identified these two sequences, which to our surprise had not been described in the literature as potent inducers of antibodies," Golding says. "Somehow during acute infection, enough of these portions of the proteins is exposed to the immune system to generate antibodies, and they were recognized at very high rates by samples from very early seroconverters."

Using the peptides, the team got to work producing a new immunosorbent assay that they subjected to testing with a variety of serum panels that include all known serotypes—from many countries, including Australia, Cameroon, Uganda, and the United States. Results compared favorably with currently licensed HIV-detection kits.

The team is now working on fine-tuning the assay, testing a second-generation of gp41 and Gag-p6 consensus peptides with even higher homology to all clades.

The assay, Golding says, is simple, rapid, and cheap, and requires no special expertise to carry out and no more than a small regular refrigerator to store the samples.

The team now has two tasks: to complete its screening of seroconverters around the world and to launch the screening of vaccinated individuals and seroconverters in HIV vaccine trials. The latter samples will come from ongoing VRC trials involving complex candidate vaccines and from already completed studies from the NIAID-funded HIV Vaccine Trials Network (HVTN).

"Here is where Barney Graham, who has been a source of advice all along, will be instrumental in the coming stages of the project and beyond the formal ending of the Bench-to-Bedside grant—in identifying and providing the best samples for us to test and to work on third-generation shorter peptides," Golding says.

For his part, Graham maintains that "this is Dr. Golding’s project—and we’re just trying to help." His role, he says, is mainly as a conduit to thousands of blood samples from the NIAID-funded HVTN.

Graham says that for the assay to have utility, it will eventually need to be licensed as a commercial detection kit and that its accuracy and sensitivity may help shape the content of future candidate HIV vaccines. "When Dr. Golding completes her testing and publishes her data, the scientific community and vaccine manufacturers will become aware of the need to design a vaccine to complement her assay—to avoid triggering a response detected by the new diagnostic test," he says, noting that candidate vaccines in VRC trials do not contain either the gp41 or Gag-p6 epitopes.

by Karen Ross

As a result of advances in treatment, perinatal HIV infection is now a rare event in the United States, and pediatric HIV/AIDS is now a chronic disease, observes Lauren Wood, who over the past 12 years at NCI has watched her HIV-positive patients move from childhood to adolescence and on to young adulthood, the beneficiaries of ever-improving antiretroviral strategies.

The problem now is drug resistance. Some of Wood’s "babies"—"now driving, graduating from high school, and getting tattoos"—are also starting to fail on highly active antiretroviral therapy (HAART).

"We need salvage regimens that will stave off progression to AIDS, invasive opportunistic infections, or death," says Wood, of the HIV & AIDS Malignancy Branch, NCI, and co-director of the NCI Pediatric Outpatient Clinic. She thinks the anticancer agent 5-fluorouracil (5-FU) may augment the effectiveness of certain HAART regimens and, pending bench data in the making, is designing a clinical protocol to demonstrate proof of principle.

Chad Womack, a senior research fellow at the VRC and the bench partner in this Bench-to-Bedside project, has been studying the resistance patterns in clinical isolates from Wood’s patients—who are on a variety of antiretro-viral agents—and measuring the effects of 5-FU in vitro.

The virus mutates rapidly, quickly developing resistance to the antiviral drugs it encounters, but that ability to mutate and survive in an antiviral drug environment, Womack says, may come at a price—a compromised ability to replicate.

The virus, he says, is "dancing on the edge of chaos," balancing survival through mutation against extinction through loss of replication capacity or fitness.

There is reason to believe that 5-FU will throw the virus over the edge, the researchers say.

Traditionally used as an anticancer drug, 5-FU is at the benign end of the toxicity spectrum of such agents, Wood says. As an anti-HIV agent, "it appears to alter the availability of the genetic building blocks" the virus uses to make more of itself.

Specifically, 5-FU inhibits thymidylate synthetase, an enzyme that generates thymine nucleotides, natural building blocks that HIV requires for replication. Wood nots that there are data suggesting that 5-FU can improve the intra-cellular pharmacokinetics of two of the nucleoside analogs commonly used in HAART regimens—AZT (zidovudine) and d4T (stavudine). By lowering levels of natural building blocks, 5-FU causes a preferential incorporation of defective building blocks supplied by the phosphorylation of drugs like AZT and d4T.

Womack believes it may also push the virus to hypermutate and severely cripple the virus’ replication machinery. "It’s a strategy called lethal mutagenesis, and there are other in vitro data suggesting it can be successful against HIV," he says.

Thus, the antiretroviral activity of AZT and d4T is optimized at the intracellular level without incurring the increased systemic toxicity caused by higher drug dosages, Wood notes. This point is critical, Womack adds, because many HIV-infected patients who are experiencing therapeutic failure are running out of options.

The clinical study will look at low-dose 5-FU in combination with a HAART regimen that includes either AZT or d4T in heavily treatment-experienced adolescents and preadolescents, most of whom were born with HIV, have been on therapy for 10 to 12 years, have undergone several rounds of HAART, and are failing clinically, immunologically, and/or virologically. The target enrollment is 30 patients and is anticipated to begin early next year.

A basic scientific question, says Wood, is whether virus grown out from different cellular compartments in the same individual displays a uniform resistance profile.

"We’re looking at plasma, peripheral blood mononuclear cells, CD4 cells and subsets including naïve and memory cells, and monocytes/macrophages. We need to know what’s happening to the virus in these different compartments as we give these drug combinations—that’s why these preclinical bench studies are so critical."

This project, headed by Michele Di Mascio, applies nuclear medicine technology to virology with the goal of developing a noninvasive way of tracking HIV in the body. Di Mascio, a biological systems modeler, is based at the Biostatistics Research Branch, NIAID.

Currently, the only way to identify HIV-infected tissues is through biopsies, procedures that are difficult for patients and yield only a limited view of the scope of infection.

The ability to see HIV throughout the body would clearly provide greater understanding of how HIV acts upon organs and how its profile changes over time. It would also help physicians find and target therapy to pools of virus that have evaded antiretroviral drugs, says Sunil Pandit, one of the CC originators of the study and now a leader in the bioengineering and genomic applications group, Heart and Vascular Diseases Division, NHLBI.

The project is highly challenging, says Di Mascio, but the initial results are quite promising. http://www.cc.nih.gov/ldrr/staff.html

The team set out to introduce into the body a radioactively labeled molecule that binds specifically to a protein of interest. Doctors then use imaging equipment such as a gamma camera or PET to detect the radioactive signal. The first hurdle is to choose the optimal molecule and then label it without compromising its ability to bind to its target, in this case, sites of active HIV infection.

A team from the CC Laboratory of Diagnostic Radiology Research (King Li, director, and Narasimhan Danthi, staff scientist) and NIAID (Tomozumi Imamichi, senior scientist, NIAID/SAIC-Frederick Laboratory of Human Retrovirology, and Cliff Lane, NIAID clinical director) has focused on a class of anti-HIV drugs called fusion inhibitors, the first of which was recently approved by the FDA.

These drugs block HIV infection of new cells by preventing the merger of the viral and cell membranes.

In the presence of fusion inhibitors, HIV becomes trapped, stuck to the cells it was on the verge of infecting with the drug bound to the junction between cell and virus. A radiolabeled fusion inhibitor, therefore, has the potential to be an excellent tool to monitor HIV infection in a patient.

So far the group has successfully attached a radioactive molecule to a fusion inhibitor and has shown that the drug is still effective at blocking viral infection in cultured cells, suggesting that the label did not radically alter its binding properties.

The next step is to test the strategy in monkeys infected with SHIV, a hybrid of HIV and its monkey counterpart, SIV. The CC’s Esther Lim and NIAID’s Malcolm Martin and Tatsuhiko Igarashi, are heading up this effort.

This model strongly predicts that an untreated HIV-infected individual harbors a genetically diverse population of virus. But just how diverse is the viral population, and how quickly does its genetic makeup change over time?

Frank Maldarelli, staff clinician in the DRP Host-Virus Interaction Unit, and his colleagues in the NIH Clinical Center—the bedside team—collected a series of samples over the course of 18 months from HIV-positive people who had not yet been treated with antiretroviral drugs. (Patients with low viral loads and relatively healthy immune systems often elect to delay treatment, Maldarelli observes.)

Sarah Palmer, manager of the DRP Virology Core Facility, and her group—the bench team—developed a new assay, known as single-genome sequencing (SGS), to analyze these samples. Unlike older methods, which can detect only the predominant viral genotypes within a sample, SGS provides an accurate picture of the full range of viral diversity.

The investigators found that viral populations were indeed diverse and that the most frequently occurring genotypes changed slowly over time, probably in response to pressure from the patient’s immune system.

They were also able to verify a laboratory prediction—that HIV genomes change not only through acquiring mutations during replication but also through exchanging parts with each other, a process known as recombination.

For viruses to recombine, at least two must infect the same cell. Based on the frequency of recombination, the researchers concluded that at least 20 percent of infected cells were invaded by multiple copies of the virus, a surprisingly high number, Maldarelli says, that suggests that HIV-infected people carry large populations of virus.

The team expects these new insights into HIV genetics to help guide therapy in the future. Knowing the speed of viral diversification can inform the decision of when to initiate anti-HIV treatment, and drug combination choices can be optimized by tracking virus recombination patterns.

Stifling recombination could become a therapeutic strategy, Maldarelli says.

Dean Hamer (NCI), Michael Root (Thomas Jefferson University in Philadelphia), and a team of clinical investigators from the Clinical Center and NIAID led by Joe Kovacs are working on two related anti-HIV drugs called 5-Helix and 5-Helix-PE (Pseudomonas exotoxin).

These drugs are designed to combat two of the lingering problems that defy attempts to cure HIV infection: drug resistance and latent infection.

5-Helix is a fusion inhibitor, a new class of drugs that interferes with the progression of HIV infection by preventing the virus from entering new host cells. Root developed 5-Helix as part of his research into HIV-host cell interactions.

HIV, Hamer explains, uses a "protein machine" that consists of components on both the virus and cell membranes to enter cells. Once the virus has made contact with receptors on cells, the machine must snap closed "like a spring trap" in order for the virus and cell membranes to fuse. Fusion inhibitors are fragments of protein that bind to the trap and "gum it up." HIV gets stuck in an intermediate state, attached to the cell, but unable to go further.

The first and only fusion inhibitor to receive FDA approval so far, T20, is helping beat back disease in some patients whose virus is resistant to all other treatments. Unfortunately, Hamer says, after as little as one month of therapy, T20-resistant strains of HIV are already appearing. 5-Helix, however, may elude this pitfall, Hamer says. It will be difficult for HIV to acquire resistance to it because the part of the virus recognized by 5-Helix is absolutely critical for HIV function.

Mutations in this region disable the virus so severely that strains with changes significant enough to prevent 5-Helix from binding will probably not be able to survive and multiply, he explains.

This year, Hamer’s group has shown that 5-Helix is indeed effective in vitro against a wide range of existing HIV variants. They are now moving into animal studies, testing 5-Helix in SCID-hu mice infected with HIV. Studies with macaque monkeys infected with SHIV, a human-simian hybrid immunodeficiency virus, are also planned.

In an effort to seek out and destroy latently infected cells, Hamer and his group have created 5-Helix-PE, which is a form of 5-Helix with an attached toxin. Because the membranes of HIV-infected cells contain the viral protein recognized by 5-Helix, the 5-Helix-PE should bind tightly and specifically to infected cells and then kill them with the toxin.

By the end of next year, the group hopes to test this approach using cells taken from infected patients.

Meanwhile, Kovacs watches from the wings. "If the preclinical findings continue to be as encouraging as they are now, we would be very excited to bring this class of drugs into clinical trials," he says.

Hamer is grateful for the Bench-to-Bedside Award program for providing the resources for the preclinical work. 5-Helix and other fusion inhibitors, he notes, are proteins, which are much more difficult and expensive to make than traditional small-molecule drugs. The award has also supported the animal experiments.