A PrEP Rally

Key updates on pre-exposure prophylaxis and HIV-specific antibodies were among the main attractions at the annual Conference on Retroviruses and Opportunistic Infections

By Kristen Jill Kresge and Regina McEnery

The 18th Conference on Retroviruses and Opportunistic Infections (CROI), which took place from February 27 to March 2 in Boston, turned into somewhat of a PrEP rally.

Pre-exposure prophylaxis (PrEP), which refers to the administration of antiretrovirals (ARVs) either orally or topically prior to HIV exposure in an effort to protect against infection, dominated much of the scientific discussion and hallway conversation. This was not all that surprising, given that two recently completed trials have found this strategy is about 40% effective in blocking HIV acquisition. Three of the six plenary talks dealt with aspects of PrEP, and new data was presented from both clinical and preclinical studies.

With regards to HIV vaccine research, the focus was on antibodies. Updates were provided on the role of antibody-mediated protection in nonhuman primates (NHPs) and the ongoing efforts to understand the crop of recently isolated broadly neutralizing antibodies (bNAbs) against HIV. “The good news,” said Michel Nussenzweig, the Sherman Fairchild Professor of molecular immunology at Rockefeller University, “is that these things [broadly neutralizing antibodies] can be made by people and they have lots of targets on the virus.” The outstanding challenge facing vaccine researchers now is translating the antibody discoveries into vaccine candidates.

New PrEP data

Momentum for PrEP began building at last summer’s International AIDS Conference in Vienna, where researchers reported results of the Centre for the AIDS Programme of Research in South Africa (CAPRISA) 004 trial that showed a vaginal microbicide candidate consisting of a gel containing 1% of the antiretroviral tenofovir (TDF) was able to reduce HIV incidence by 39% in 889 South African women at high risk of infection (see Microbicides Finally Gel, Securing Spotlight at the International AIDS Conference, IAVI Report, July-Aug. 2010).

This was followed in November by results of the iPrEx trial of nearly 2,500 men and transgendered women who have sex with men, which found that daily administration of oral Truvada, TDF combined with the ARV emtricitabine (FTC), was 44% effective in preventing HIV infection (see Vaccine Briefs, IAVI Report, Nov.-Dec. 2010).

Much of the new PrEP data presented at CROI was derived from sub-analyses of the iPrEx trial. Robert Grant, principal investigator of the trial, provided an additional seven months of data from iPrEx that showed that the overall efficacy was a slightly lower, but still statistically significant 42%.

Grant also presented results of a case-controlled study that showed the levels of TDF/FTC in blood strongly correlated with protection among volunteers in iPrEx. The sub-analysis showed that 91% of individuals in the PrEP arm who became HIV-infected had no detectable drug levels, while 9% had extremely low levels of the drugs in their blood. This was despite the fact that daily dosing was not required for there to be detectable drug levels. “Probably four pills in two weeks would have been enough to have a detectable drug level,” said Grant.

Importantly, Grant said, no one with a drug level that would have been expected with daily dosing became HIV infected. “It was protective when used and not protective when not used,” he said. “Our data shows that 50% [of volunteers] took the pill almost every day. That’s remarkable. Adherence was outstanding in iPrEx for half the people.”

Now, researchers are shifting their focus to the other half of the volunteers. To better understand the factors that influence adherence, the iPrEx investigators have initiated an open-label study to see if adherence improves when participants are given TDF/FTC in an un-blinded setting. “Now, we can say this is a pill that will protect you but you have to use it consistently,” said Grant. The TDF-FTC combination will be offered to every HIV-uninfected participant from the iPrEx trial as part of the open-label study, according to Grant. This study will provide more long-term safety data, as well as the opportunity to study whether sexual practices change when people knowingly take a pill that can prevent the risk of HIV acquisition.

One of the difficulties with all behaviorally dependent prevention modalities, such as PrEP, is how to monitor adherence. One study presented at CROI by Rivet Amico, a research scientist with the University of Connecticut Center for Health, Intervention and Prevention, evaluated self-reported daily pill use and other subjective indicators of adherence to see how accurately they reflected actual pill use among participants in the iPrEx trial.

Amico analyzed the intra-cellular drug level of TDF/FTC in blood specimens taken 24 weeks post-enrollment in the iPrEx trial in a random sample of 179 participants from the PrEP arm. Her analysis of this subset of volunteers showed that only 68% of those who reported taking their pills every day were actually 100% compliant. And weekly pill counts (the number of pills dispensed per week minus those returned) reflected actual adherence just 59% of the time.

In another study based on data from the iPrEx trial, Peter Anderson, an associate professor of pharmacology at the University of Colorado, analyzed frozen peripheral blood mononuclear cells (PBMCs) from iPrEx study volunteers and found that only 50% of those who received PrEP had detectable levels of the active form of TDF (tenofovir-diphosphate) in PBMCs, while 62% had detectable levels of FTC. Anderson said FTC is present at relatively high concentrations after a single dose, which likely explains the differences in drug levels. Anderson said geography, age, and recent report of sexual activity were predictors of high drug levels. Either TDF or FTC was detected in PBMCs from 97% of the US participants in the iPrEx trial, but in only 50% of the samples from non-US participants. Adherence was also higher among men 25 years or older (with 73% of samples from men in this age group having detectable drug levels in PBMCs versus 43% in younger men), and among men who reported having had receptive anal intercourse without a condom in the previous 12 weeks (71% of these volunteers had detectable drug levels versus 30% of volunteers who did not report this behavior).

But not everyone is so concerned about adherence. At a community forum held in conjunction with CROI, co-sponsored by the Boston-based Fenway Health and the New York-based prevention research advocacy organization AVAC, Grant challenged the notion that adherence will always be higher in a clinical trial. He thinks adherence will likely be higher when volunteers are told they are receiving Truvada, and that it was shown to be effective in protecting against HIV only when taken reliably. “Adherence is a solvable problem,” said Grant.


Interim Guidelines for PrEP  

Following the results of the iPrEx trial, the US Centers for Disease Control and Prevention (CDC) issued interim guidance regarding the use of oral antiretrovirals for pre-exposure prophylaxis (PrEP) among men who have sex with men (MSM), available at www.cdc.gov/hiv/prep. There is no conclusive data yet on whether oral PrEP is effective in heterosexual men or women (see Vaccine Briefs).

Side effects

While adherence is considered one of the biggest hurdles to PrEP efficacy, its long-term safety in HIV-uninfected individuals is also of concern. Specifically, researchers are monitoring whether TDF decreases bone mineral density (BMD), a marker for fractures, in HIV-uninfected individuals in the same manner it does in those taking the drug for treatment. Tenofovir’s manufacturer, the pharmaceutical company Gilead Sciences, now recommends regular bone scans for HIV-infected individuals on TDF.

Two studies presented at CROI showed that use of TDF or Truvada for prevention have resulted in small, but statistically significant decreases in BMD among HIV-uninfected individuals. The first study was a sub-analysis of a trial that evaluated the safety and acceptability of daily oral TDF in 400 HIV-uninfected men who have sex with men (MSM) in the US. Albert Liu, a co-investigator of that trial, said dual energy X-ray absorptiometry (DEXA) scans performed on 184 men at the start of the trial and then nine, 12, and 24 months later showed a 1% net decrease in the mean BMD at the neck, and a 0.8% decline in BMD at the hip among TDF users compared to the placebo arm or to their baseline measurement. The drop in BMD was most prominent during the first 12-15 months of taking the drug. Despite the drop in BMD, Liu said there were no differences observed in fracture rates among the volunteers in the two groups.

In a separate sub-analysis of 503 volunteers from the iPrEx trial, Kathleen Mulligan, a researcher in the Division of Endocrinology and Metabolism at the University of California-San Francisco, also found a small but significant decline in BMD in the Truvada arm, as compared to placebo. Mulligan said DEXA scans of the spine and hip were taken at 24-week intervals over a 72-week period and showed that BMD tended to decrease in the PrEP group and increase in the placebo group, resulting in a modest (0.7% - 1%), but statistically significant difference in BMD between the two groups by week 24.

Another concern related to PrEP use is the development of drug-resistant virus in individuals who become HIV infected despite using ARVs for prevention. But at CROI, concerns over resistance were largely allayed based on observations from the iPrEx trial. A sub analysis of the iPrEx data showed that none of the participants randomized to the PrEP arm who were not HIV infected at the start of the trial developed drug-resistant virus mutations, including three participants with low, but detectable levels of FTC or TDF in their blood. “We looked much more closely for drug resistance and didn’t find it,” said Grant.

In another encouraging development related to resistance, researchers from the CDC showed that oral Truvada protected all five rhesus macaques rectally challenged with a simian immunodeficiency virus (SIV)/HIV hybrid known as SHIV that contained the M184V mutation in reverse transcriptase that most commonly confers resistance to FTC. The monkeys received Truvada three days before and two hours after SHIV challenge, and remained uninfected after 14 challenges. In contrast, all five of the control animals became infected after a median of three SHIV exposures.

Gerardo Garcia-Lerma, who presented these data, said that while the SHIV strain is almost completely resistant to FTC, it is also two to four times more susceptible to TDF than wild-type SHIV strains. Additionally, he said the peak viral load in the untreated macaques was considerably lower than it would have been had they been infected with a non-mutated strain, with fitness assays showing that the SHIV strain replicated less efficiently and was four times less infectious than the non-resistant virus, which helps explain the results.

Cost effectiveness

In addition to issues of resistance, safety, and adherence, researchers are also evaluating another obstacle to PrEP implementation: cost.

In low- and middle-income countries only a third of the estimated 15 million HIV-infected individuals who qualify for ARVs are currently receiving them. Should the World Health Organization (WHO) eventually decide to include PrEP in its list of recommended strategies for the prevention of HIV in certain high-risk populations, many of these poor countries will need to confront the ethical conundrum of providing ARVs to uninfected individuals when so many infected people are going without.

Even in the US, where Gilead intends to file a supplemental investigational new drug application with the US Food and Drug Administration for the use of Truvada for prevention, there is no guarantee that public or private health insurance programs will agree to cover the costs of PrEP for MSM or other high-risk populations.

Timothy Hallett, a research fellow at Imperial College London’s School of Public Health, used mathematical models to determine whether providing PrEP to the uninfected partners of serodiscordant couples or providing ARVs to the HIV-infected partners was a more cost-effective way to ensure that the HIV-uninfected partner remained alive and HIV uninfected at age 50.

His model showed PrEP is a more cost-effective approach than highly active antiretroviral therapy (HAART) when the HIV-infected partner’s T-cell count is 500 cells per milliliter (ml) of blood. However, when the HIV-infected partner’s T-cell count reaches 350, the point at which the WHO’s guidelines recommend ARV treatment begin, PrEP is no longer the most cost-effective option unless PrEP efficacy is at least 70% and the cost of PrEP is 40% less than the cost of ARVs. It is important to note that this model only evaluated the cost effectiveness of the two approaches and did not evaluate which approach led to a better health outcome for the HIV-infected partner. It is also still unclear whether oral PrEP will be effective in preventing heterosexual transmission (see Vaccine Briefs, this issue).

Hot topic(als)

Adding to the excitement about PrEP, Charles Dobard, a research scientist at the CDC’s Division of HIV/AIDS Prevention, presented data from a NHP study showing that administration of a vaginal gel containing 1% raltegravir, the only licensed ARV that works by inhibiting the virus’s integrase enzyme, prevented HIV infection in five of six pigtailed macaques when applied topically three hours after the animals were challenged vaginally with a low-dose SHIV-SF162P3. Monkeys that received the gel remained uninfected after 20 SHIV challenges, while all four animals that received a placebo gel became infected after an average of 10 SHIV exposures.

Before challenging the macaques, the CDC team conducted in vitro experiments to determine how long after exposure to HIV a microbicide could be applied and still be effective. They found that reverse transcriptase inhibitors like TDF began to lose efficacy if applied more than two hours after exposure, whereas integrase inhibitors could be applied up to eight hours after exposure before losing efficacy.

Dobard said this marks the first time raltegravir has been tested as a microbicide. The study was also the first time a microbicide was tested solely as post-exposure prophylaxis (PEP). Dobard said because the raltegravir gel was able to protect the monkeys when applied right after SHIV exposure, it may also be able to block infection when used a short time before or after exposure to HIV. “With any interventions, the more options, the better,” said Dobard. “If effective, this may create more flexibility for women and may also help improve adherence.”

Researchers also reported results from a Phase I study evaluating the safety and acceptability of a 1% TDF microbicide gel applied rectally. The US study of 14 HIV-uninfected men and four HIV-uninfected women was only the second clinical trial to have evaluated a rectal microbicide candidate.

Trial participants were first given a single dose of oral TDF. Two weeks later, they were randomized to receive either a rectal dose of 1% TDF or placebo gel. Two weeks after that they were asked to use either the TDF or placebo gel once a day for seven days. During the course of the trial, researchers collected more than 2,000 colon and rectal tissue samples. When these tissue samples were exposed to HIV ex vivo, researchers found that the samples from participants who received the TDF gel showed significant inhibition of HIV, suggesting this approach may be useful in blocking rectal transmission.

However, principal investigator Peter Anton, a professor of medicine at the University of California, Los Angeles, said three of the 18 participants experienced eight grade-three adverse events, which are considered to be severe, after using the rectal microbicide. The adverse events were primarily diarrhea, cramps, and gastrointestinal discomfort. Acceptability was also a factor. Only 25% of those who received the TDF gel said they liked using it compared to 50% in the placebo group, although 75% in the TDF group and 67% in the placebo group said they would likely use a rectal microbicide if one became available.


Gel vs. Pill

To gauge whether women prefer taking antiretrovirals (ARVs) for HIV prevention in a pill or gel, researchers at Johns Hopkins University polled 144 women in Africa and the US enrolled in a Phase II study that compared oral administration of the ARV tenofovir (TDF) with topical application of a TDF microbicide gel. The researchers found that while 100% of African women reported they would use either the pill or gel form if it was found to be effective, the African women reported greater sexual satisfaction when using the gel. In contrast, 72% of US women said they preferred the pill.

“For many people it makes more sense to put it where the action is,” said Robert Grant, principal investigator of the iPrEx trial that found orally administered TDF and emtricitabine was effective at blocking HIV infection in men and transgender women who have sex with men. It is still unclear whether oral Truvada will protect women (see Vaccine Briefs). —R.M.

HIV antibodies

The discovery of a score of new bNAbs, some of which are much more potent and broad than those previously identified, has provided researchers with plenty to study about antibody development in HIV-infected individuals and how these antibodies neutralize HIV. “There are lots of ways to neutralize the virus. I think antibodies are boring, but in this disease they’re actually quite exciting,” said Nussenzweig, adding that some of the recently identified antibodies, “are just the beginning.”

One common attribute of these new antibodies is the high level of somatic hypermutation they have undergone (see Vaccines to Antibodies: Grow Up!, IAVI Report, July-Aug. 2010). These mutations are known to be important because reverting the antibodies to their inferred germline sequence results in much weaker binding, or loss of binding altogether in some cases. “Hypermutation is a prominent feature of these antibodies,” Nussenzweig said. To better understand the hypermutation process, he is studying what happens to the B cells that produce these antibodies in the germinal center, where multiple cycles of division and selection lead to the development of better antibodies.

To do this, Nussenzweig and colleagues have developed a new tool that allows them to watch labeled cells in situ in a lymph node using a photo-activatable transgenic mouse. This allows them to see how these photo-activated cells get redistributed to different parts of the lymph node over a period of a few hours—movement that is indicative of the best cells getting selected.

But just how this selection process occurs is unclear. Nussenzweig said previous research has shown that T cells help to govern germinal center selection of B cells by driving inter-zonal migration, clonal expansion, and affinity maturation of B cells within the lymph node. However, too much of a good thing can be deleterious. “If you flood the B cell with T-cell help you kill affinity maturation,” said Nussenzweig, who noted that research by Herman Eisen, one of the first researchers to describe the process of affinity maturation, “showed that low levels of antigen were really optimal for affinity maturation.”

While many researchers are focused on the new crop of bNAbs that were isolated from chronically HIV-infected individuals, others are focused on learning more about vaccine-induced antibodies, which develop under very different circumstances. Gunilla Karlsson Hedestam, an associate professor in the department of microbiology, tumor and cell biology at the Karolinska Institute, and colleagues are using the NHP model to study the quality of B-cell responses stimulated by soluble HIV Env proteins. “Subunit Env vaccines may elicit some degree of protection,” said Hedestam, referring to the results of the RV144 trial in Thailand, “but it’s unclear if this is so solid and it’s definitely not sufficient to be satisfied with the response.”

Hedestam and colleagues conducted a SHIV protection study involving six NHPs that received five vaccinations of a soluble trimeric HIV Env immunogen along with adjuvant at one-month intervals. They then sorted the Env-specific memory B cells from the animals. “The frequency of Env-specific antibodies in the vaccinated animals is about 10-fold higher than in chronically [HIV] infected patients,” said Hedestam. So far, the researchers have cloned and characterized eight functional Env-specific antibodies from the vaccinated animals and determined their binding properties using an ELISA assay. Six of the antibodies are directed toward the CD4-binding site, while two are directed at a still-to-be determined spot on the virus.

As for how well these antibodies neutralize, Hedestam said they are capable of neutralizing the usual, easy to neutralize viruses. “We know we’re far away from eliciting a response that’s anywhere near as potent as you achieve in chronic infection, but we want to understand what we are eliciting,” she said, adding that this work should help guide the selection of Env immunogens for future vaccine trials.

What types of antibodies protect?

While researchers are just beginning case-controlled studies to try to identify possible immune correlates of protection from the RV144 trial—the first to show any evidence of vaccine-induced protection against HIV—there has been some speculation that the modest 31% efficacy demonstrated in this trial may be due, at least in part, to non-neutralizing antibody activity. As a result, “various vaccine design projects are now based on the notion that non-neutralizing antibodies might be protective,” said John Moore, professor of microbiology and immunology at Weill Cornell Medical College. “If you’re going to design vaccine antigens, I’d rather see them be based on something that is proven to be effective.”

To see whether non-neutralizing antibodies are in fact protective, Moore and colleagues conducted a set of studies in NHPs. In the first study, they evaluated the protective effects of b12 (one of the first bNAbs to be identified) and b6 (an antibody that is closely related to b12 and binds to a similar spot on gp120, but is not broadly neutralizing), as compared to controls. The antibodies were delivered systemically at a dose of 25 mg/kg. The animals were then treated with progesterone and challenged with a single vaginal administration of SHIV-162P3.

Following challenge, all four of the monkeys that received b6 were infected, compared to only one of the four monkeys that received the bNAb b12 and only one of two control animals. “The results suggest b12 protects against this virus but the non-neutralizing antibody did absolutely nothing,” said Moore.

In a study conducted in 2003, Moore and colleagues showed that vaginal administration of b12 protected rhesus macaques against vaginal challenge with SHIV-162P4, with 12 of 13 control animals becoming infected compared to only three of 12 b12-treated animals.

To see whether non-neutralizing antibodies would impart any protection against this challenge virus, Moore and colleagues decided to repeat their 2003 study and test the non-neutralizing antibodies b6 and F240, which targets gp41, as well as b12. The antibodies in this study were delivered in a single vaginal administration. The animals were challenged 30 minutes later, following progesterone administration, with a vaginally delivered SHIV-162P4.

Results from this study showed that while none of the seven b12-treated animals became infected, all five animals in the b6 group were infected, a statistically significant difference. Meanwhile, the non-neutralizing antibody F240 appeared to provide partial protection, with three of five animals becoming infected, though this difference was not statistically significant. In this study, all five of the control animals were infected.

When the researchers investigated the founder virus populations that established infection in the different groups, they found that a median of one founder virus established infection in the b12- and F240-immunized macaques, while in the b6-immunized macaques, the median number of founder viruses was three (a statistically significant difference). Moore called this finding “intriguing,” and said it suggests that “non-neutralizing antibodies might enhance mucosal transmission under some circumstances.” But Moore also noted that the concentrations of antibodies given to the macaques in this study were much higher than would be expected in a vaccine setting. “I’m not saying non-neutralizing antibodies are dangerous but there’s something to be investigated,” he said.

In another antibody-related study, Nancy Haigwood, director and senior scientist at the Oregon National Primate Research Center, evaluated the potential of passively administered monoclonal antibodies to limit mother-to-child transmission (MTCT) of HIV. In this study, Haigwood compared the efficacy of three different antibody regimens to see which was best at protecting against or controlling viremia following SHIV challenge. The three antibody preparations included a mixture of polyclonal antibodies that would mimic the Immunoglobulin (Ig) G that is transferred from a mother to her infant across the placenta, the monoclonal bNAb b12 alone, and a combination of an IgG purified from animals infected with the SHIV-SF162P3 strain along with b12. All newborn macaques were infused with antibodies at a dose of 200 mg/kg and then challenged orally a day later with SHIV-SF162P3.

Four of the six newborn macaques that received the combination of SHIV IgG and b12 were completely protected against challenge, while the other two had low plasma viral loads. However, none of the six macaques that received IgG or b12 alone were protected. In fact, the viral loads were significantly and persistently higher in the group of newborn macaques that received the b12 bNAb by itself, compared to both of the other groups. “This is a very small study but it suggests single monoclonal therapy might be risky,” said Haigwood, though she noted that the quantity of b12 given to these animals was five-fold lower than the levels that have been shown to protect against low-dose challenge in rhesus macaques.

Haigwood said these data may indicate that b12 puts selective pressure on the virus that allowed only more pathogenic viruses to get through and establish infection in the b12-treated macaques. “We are actively looking at those animals to try to find out,” she said.

Haigwood also plans to study other bNAbs in this model, including VRC01, which was isolated recently by scientists at the Vaccine Research Center at the US National Institute of Allergy and Infectious Diseases, and is more broad and potent than b12. “I would suggest that passive immunotherapy be modeled carefully to avoid and understand unanticipated outcomes that might be seen in expanded studies,” said Haigwood. However, she thinks that the protection afforded by the SHIV IgG and b12 does warrant the development and use of monoclonal antibodies in addition to drug therapy to further limit MTCT and enhance immunity during breastfeeding.


Giving HIV the Zinc Finger

Gene therapy has a checkered history and has shown limited efficacy in treating HIV. But at CROI, results from a Phase I gene therapy trial and other preclinical studies suggest this approach might hold promise.

The gene therapy approach that headlined CROI involves modifying human cells to disable expression of CCR5, the preferred co-receptor HIV utilizes to gain entry into cells. For some time, researchers have known that individuals who are homozygous for the CCR5∆32 allele—a truncation in the CCR5 gene that prevents the receptor from being expressed on a cell’s surface—are protected against infection with CCR5-tropic HIV. This polymorphism is thought to have developed in response to the Black Death, and is estimated to be present in approximately 1% of the population. These naturally CCR5-deficient individuals demonstrate that knocking out the CCR5 receptor has few deleterious effects, though these individuals are more susceptible to certain viruses and diseases, including West Nile virus and rheumatoid arthritis.

The first reports of the so-called Berlin patient, an HIV-infected man who received a hematopoietic stem cell (HSC) transplant from a donor homozygous for the CCR5∆32 mutation, further confirmed the protective role this mutation confers against HIV (1). In 2009, doctors reported that the Berlin patient had sustained an undetectable viral load without resuming ARV therapy for 20 months after the transplant. More recent reports indicate that the patient has now gone three and a half years without any evidence of HIV infection, leading the doctors who conducted the transplant to say, “It is reasonable to conclude that cure of HIV infection has been achieved in this patient,” (2).

The success of the Berlin patient has reinvigorated research into finding an HIV cure, a four-letter word that researchers shied away from in the past. And according to Jay Lalezari, director of Quest Clinical Research and an investigator of a gene therapy trial in HIV-infected individuals, it also validates that CCR5 is a relevant target for gene therapy approaches to HIV treatment.

Lalezari’s trial is one of two that are evaluating whether a zinc finger nuclease (ZFN) technology developed by California-based Sangamo BioSciences can be used to permanently knock out CCR5 expression on CD4+ T cells ex vivo, making them resistant to CCR5-tropic HIV. The ZFN is a small peptide that specifically targets and binds to a region just upstream of the ∆32 mutation in the DNA of the CCR5 gene, where it clips the DNA, introducing a double-stranded break. This triggers an error-prone repair mechanism, which most commonly introduces a five base pair insertion with two stop codons, resulting in permanent disruption of the CCR5 gene. Lalezari called the ZFN a “designer restriction enzyme.”

At CROI, Lalezari reported results from a Phase I safety study that involved six HIV-infected volunteers on highly active antiretroviral therapy (HAART) with undetectable viral loads (below 50 copies of viral RNA per milliliter of blood), but sub-optimal CD4+ T-cell levels of between 200 and 500. Lalezari said the men were infected with HIV for between 20 and 30 years.

The ZFN in this study was delivered via Sangamo’s adenoviral SB-728 vector to T cells harvested from the HIV-infected volunteers at an outpatient leukapheresis center. The cells were then expanded up to 10-30 billion cells using a proprietary technology, frozen, and shipped back to the centers, where they were re-injected into patients along with Tylenol. “The infusions were safe and well tolerated,” reported Lalezari, who said most of the adverse events were mild flu-like symptoms.

Following the transfusion, researchers found the modified cells engrafted in all six volunteers. “We’re seeing engraftment of the cells beyond what would be expected,” said Lalezari. One patient, who had the highest level of anti-adenovirus (Ad) antibodies before transfusion, had the lowest level of engraftment of the ZFN-modified cells.

There was also evidence that the cells expanded in vivo in some of the volunteers—two weeks after the infusion, there was a median three-fold increase in the number of ZFN-modified cells in PBMCs. However, Lalezari warned that the apparent expansion of the CCR5-modified cells in peripheral blood mononuclear cells (PBMCs) may be an artifact of the activation of the cells during the modification process, rather than actual expansion of these cells in vivo.

The engrafted cells also seemed to persist. Five of six patients had persistent CD4+ T-cell count increases of more than 100 cells. Five of the six volunteers also experienced a normalization of the CD4+/CD8+ T-cell ratio, which is often skewed in HIV-infected individuals. Three months after transfusion, 6%-7% of the CD4+ T cells in peripheral blood remained modified. Lalezari also reported that these cells trafficked to the gut. Sequential rectal biopsies of volunteers showed a median of 1.7% of CD4+ T cells modified there after 90 days.

Although these results are preliminary, Lalezari said ZFNs “offer hope of providing a protected reservoir of CD4+ T cells that are resistant to HIV,” and that this gene therapy approach may be “a feasible approach to HIV therapy.”

Lalezari’s next study will involve treatment-naive, viremic individuals. He said that in these individuals, who have actively replicating virus, the modified T cells should have a selective growth advantage, and therefore there should be even better engraftment of these cells. The big question is whether this will result in a reduction in viral load in these individuals. As for whether this approach could someday be a viable cure, Lalezari warned that it would be an overreach of the data. “We have to be careful when we throw around the C word,” he said.

Another ongoing study of Sangamo’s ZFN is taking place at the University of Pennsylvania (UPenn). So far, six patients have been enrolled who have CD4+ T-cell counts above 450, and of them, four have been infused with the ZFN-modified cells. Another four patients have been enrolled with CD4 counts below 500, and of them, two have received the ZFN treatment.

Similar to Lalezari’s results, most of the adverse events reported so far in this trial stem from infusion-related side effects, according to Carl June, professor of pathology and laboratory medicine at UPenn and an investigator on the trial, who presented the interim analysis. June reported that there has been a persistent increase in CD4+ T-cell levels in the volunteers who have been infused so far, and that in most patients, the CD4+/CD8+ T-cell ratio has normalized. “These are persistent changes through the follow up we have to date,” he said.

Researchers used deep sequencing technology to characterize the insertions and deletions in the CCR5 gene in the volunteers treated so far and found that the most abundant insertion in the CCR5 gene after ZFN treatment is a pentamer insertion sequence that is not found anywhere else in the human genome.

June said that data analyzed from nine patients from both his study and Lalezari’s suggest that 40% of the cells in peripheral blood are modified soon after transfusion. In eight of these nine patients, again combined from both studies, there is also evidence that the CCR5-disrupted cells have expanded.

In June’s trial, the two patients with CD4+ T-cell counts above 450 that received the ZFN treatment also opted to interrupt ARV treatment for a 12-week period. And, according to June, it took 10 weeks after cessation of therapy for the virus to come back to pre-ZFN treatment levels in these individuals. He said these results are the first example of genetic editing used to introduce a disease-resistance gene and that this work will likely have implications beyond HIV therapy.

Humanized mice experiments

Craig Wilen, a member of Robert Doms’s laboratory at UPenn, reported on another ZFN developed by Sangamo that is undergoing pre-clinical testing. This ZFN is designed specifically to bind to and cleave the CXCR4 gene, which controls expression of the CXCR4 receptor, an alternate receptor on CD4+ T cells that HIV can use. In experiments in humanized mice, Wilen showed that delivery of the ZFN using an Ad5 vector resulted in disruption of the CXCR4 gene and subsequent protection from an X4-tropic HIV challenge seven days after modification of the T cells. However, in this model, Wilen did observe a preponderance of dual-tropic virus emerge in the challenged mice. He said the focus of future research would be on disrupting expression of both the CCR5 and CXCR4 genes, blocking both routes of HIV entry. “Our long-term goal is to recapitulate the success of the Berlin patient,” said Wilen.

Rather than modifying T cells, another approach to shutting down CCR5 expression involves modifying HSCs, much as was done in the Berlin patient, but using gene therapy instead of a bone marrow transplant. “Bone marrow transplantation is not something that’s going to be widely available,” said Paula Cannon, an associate professor at the Keck School of Medicine at the University of Southern California. She is studying whether CCR5-modified HSCs engrafted into humanized mice are able to protect the mice against HIV challenge.

In her experiment, one-day-old mice are engrafted with HSCs that have been modified ex vivo using a ZFN delivered in an Ad vector. When the mice are challenged with HIV several months later, they are able to control infection. “The ZFN-treated mice look the same as uninfected controls,” says Cannon. In the control mice, 25% of cells in the spleen and gut are CCR5+, while in the ZFN-treated mice, none of the cells in these compartments express CCR5.

Cannon said her focus now is on getting this approach into clinical trials. To do this, she plans to piggyback on trials that are using HSC therapy to treat HIV-infected individuals with lymphoma. Before beginning such a trial, though, she said more mouse studies will ne necessary to ensure that the Ad vector containing the ZFN doesn’t harm human HSCs. —Kristen Jill Kresge

1. N. Engl. J. Med. 360, 692, 2009
2. Blood 117, 2791, 2011