The AIDS Vaccine Research Subcommittee (AVRS) met in public session on February 8–9, 2011, at the Bethesda North Marriott Hotel in Bethesda, MD.
AVRS members present: Nancy Haigwood (chair), James Bradac (executive secretary), Jay Berzofsky (ex officio), Deborah Birx, Satya Dandekar, Jeffrey Lifson, Douglas Nixon, Nina Russell, George Shaw, and William Snow.
Other NIH personnel participating:
Nancy Haigwood called the meeting to order at 8:30 a.m. and asked the committee members and other participants to introduce themselves. Jim Bradac recognized the service of two members who are rotating off the subcommittee—Bali Pulendran and Dennis Burton—and Debby Birx, whose appointment has been extended for another 3 years. He also introduced Alan Fix, who is serving as acting director of the Vaccine Research Program in DAIDS, replacing Peggy Johnston, who has recently retired.
James Bradac presented a summary of DAIDS-supported research that included AIDS vaccine grants funded in 2009-2010, funding initiatives planned for 2011, overall plans for 2012 and the AIDS vaccine preclinical pipeline. In the area of basic vaccine discovery research, there were 15 awards made under theHighly Innovative Tactics to Interrupt Transmission (HIT-IT) program; ten 4-year R01 awards in 2009 and five awards in 2010; there were 34 awards made under the Basic Vaccine Discovery (BVD) program; twenty 5-year R01 awards in 2009 and fourteen awards in 2010; and there were three U19, cooperative agreement awards made under the B-Cell Immunology Partnerships program in 2010. In addition, there were six awards made under the Phased Innovation Award (PIA) program and 3 awards made under the HIV Vaccine Research and Development (HIVRAD) program in 2010.
In 2011, the HIT-IT, BVD and B-Cell Immunology Parnership programs will not be active. However, NIAID will launch a new Mechanisms and Prevention of Sexual Transmission of HIV/SIV program and a new Consortia for AIDS Vaccine Research in Nonhuman Primates (NHPs) program in 2011. In 2012, the Phased Innovation Awards (R21) will be replaced by a new R01 programInnovation for HIV Vaccine Discovery (IHVD). There are plans to readvertise the Corsortia for AIDS Vaccine Research in Nonhuman Primates in 2012, but threre is a possibility that this initiative will not be put out should the Division decide to fund two awards from the 2011 initiative. Finally, the new Center for HIV/AIDS Vaccine Immunolgy and Immunuogen Discovery (CHAVI-ID) award will be made, replacing the current CHAVI award that expires in 2012.
Two important research meetings supported by DAIDS are also planned for 2011—a workshop on Defining Innate Immune Responses through Functional Genomics on June 8, and a B-cell Workshop in August.
Alan Fix discussed the renewal of the HIV Vaccine Trials Network. There will be three awards—for Operations, Laboratories, and Data compilation and analysis—and 2011 will see the addition of networks for vaccine research in tuberculosis (TB) and hepatitis C virus (HCV). The network will continue to pursue Phase 1, 2 and 3 trials to evaluate candidate vaccines and investigate immune correlatives. There will be a new emphasis on collaboration with other prevention modalities and on the attraction and mentoring of new investigators. Reviewers have been impressed with a new emphasis on “right-sizing” the network and on collaboration (and budget sharing) with other disciplines and diseases.
In the discussion that followed, AVRS members and other participants pointed out that the TB community, in particular, will be an important source of expertise and collaborators. There is of course a continuing need to coordinate the network’s activities with those of other nations and NGOs. Declining budgets will be a problem for all the networks, but Fix does not anticipate the need to eliminate more than one existing network. There will probably be an increased focus on integrated prevention initiatives, including not only vaccines but also microbicides, PrEP, and other modalities.
José Esparza described the Gates mission as "scientific solutions delivered for human health," with the long-term goal of reducing incidence and extending life expectancy. As a result, their focus is on translational research to support that mission. In terms of vaccines, this translates into a candidate with a high efficiency (e.g., 70 percent) and broad coverage (e.g., 40 percent). Geographically, their focus is on the poor, high-incidence countries of sub-Saharan Africa. Gates would like to see a follow-up to RV144, but they would also like to see trials of new candidates and new modalities that increase efficiency and lower the cost of treatment. The new Concept Pipeline RFP, issued in 2010, aims to diversify the product development and evaluation in four areas:
Application review is scheduled for April 2011, with plans to fund between 20 and 30 projects at up to $1 million per year for up to 4 years.
Michael Pensiero introduced a review of early-phase research projects from the DAIDS-supported vaccine pipeline. He noted that work on additional adenovirus (Ad) vectors would be presented at the next AVRS meeting.
Louis Picker explained that, because vaccines must operate in the T cell “functional landscape,” most of them focus on the early stages of immune response (i.e., central memory) and for this reason are typically slow to replicate (14 days or more post-challenge) and nonpersistent. The real widow of opportunity to block HIV infection is in the first few days after challenge, and for this a strategy based on effector memory is required.
CMV has proven tobe one of the best inducers of effector memory, and NHP trials have demonstrated the speed, specificity, and persistence of this response in controlling early infection. Protection lasts 52 weeks or until necropsy, and can be boosted. This points to the desirability of TEM-mediated response, as well as antibodies stimulated by TCM cells; the question is whether we can develop a CMV-based vaccine that would be safe in pregnant and/or immune-suppressed recipients. Investigators are currently making a variety of replication-deficient CMV to test as vectors.
In response to questions, Picker added that CMV is not the only way to generate a TEM response, and that eradication of infection requires continuous surveillance as well as short-term efficacy. The challenge dose used in these NHP studies has been relatively low; notably, repeated challenge was required to infect all animals. Challenge was homologous, and cost has prohibited repeating the study with a heterologous challenge thus far. However, protection seems to be very broad and would probably withstand heterologous challenge. This approach promises to prevent acquisition; the challenge is to make it work in 100 percent of animals and make it safe for humans.
Wayne Koff suggested that it might take a combination of neutralizing antibodies and cell-mediated immune response to prevent infection. IAVI is conducting preliminary research in a wide range of viral vectors—replicating, attenuated, and replication-deficient—including Ad26 and Ad35, CMV, Sendai virus (SeV), canine distemper virus (CDV), recombinant vesicular stomatitis virus (rVSV), and adeno-associated virus (AAV). A small Phase 1 trial of a CMV vaccine in seropositive males is underway in Oregon, with results expected in 2014; small animal toxicology is underway with other candidates.
In response to questions, Koff explained that investigators are working backward from the desired antibodies to determine whether they need a persistent vector to ensure ongoing immunogen exposure. Maturation of the antibody response is a central concern and is being studied in NHP. In NHP, the use of DNA IL-12 boost does not appear to block replication of the original Ad vector, and it achieves a two- or three-log reduction in viral load that not seen with the vector or boost alone.
Phil Johnson’s team is currently working to develop a vaccine based on vector-mediated gene delivery. The vector is a nonreplicating parvovirus, which disseminated throughout the body but has never been associated with any pathological conditions. It is a safe vector, already tested in over 80 clinical trials involving vaccines for a variety of diseases. Demonstration projects have already been conducted in mice and NHP, and the next step will be Phase 1 safety trials in at-risk recipients, using escalating doses of AAV-1 vector and a PG9 monoclonal antibody gene insert. There could be better vectors or inserts, of course, so the best approach would be to look at multiple combinations. This should eventually yield a vaccine that will elicit both TEM and broadly neutralizing antibodies.
In response to questions, Johnson said that investigators are trying to identify the right monoclonals—the ones that interfere with a broad range of HIV isolates—from the sera of elite controllers. Preexisting immunity to AAV-1 is not expected to be a problem, since it is far less common than AAV-2, but there are many other vectors that could be used if necessary.
Hildegund Ertl reported on the development of recombinant chimpanzee adenoviruses AdC6, 7, and 9 HIV/AIDS vaccines. In Indian rhesus macaques, vaccines using AdC6 and AdC7 vectors and SIV gag and tat inserts produced a relatively broad response to single infectious dose rectal challenge with SIVmac239, and an even better response to the more relevant repeat dose rectal challenge. Two of the 10 control animals developed SIV-like symptoms and had to be euthanized at weeks 8 and 28; none of the 20 experimental animals had developed overt symptoms at week 42. A new NHP experiment has begun using gag and gp160 inserts. The conclusion is that AdC vectors induce a potent, sustained antibody response in NHP. In short, chimp adenoviruses are a versatile, malleable, well tolerated, and potent class of vectors.
John Eldridge described a vaccination strategy that would stimulate both antibodies to block infection and central memory to clear them. pDNA IL-12 has demonstrated its ability to enhance the CMI response to SIV gag in NHP, and delivery with electroporation has been shown to increase the magnitude, breadth, duration, and specificity of the CMI response. Replication-competent recombinant VSV (rVSV) is a safe and effective vector to which there is relatively little preexisting immunity in the human population. Preliminary NHP experiments with HCV showed that pDNA prime followed by rVSV boost increased the magnitude and breadth of the CMI response compared with pDNA alone or pDNA boosted by itself. Clinical trials are already underway to evaluate these components separately—pDNA plus electroporation in ACTG 5281, and rVSV in HVTN 090—and the next step will be a trial (HVTN 087) combining a multi-epitope pDNA prime with rVSV boost. The toxicology study for this trial is already complete, and enrollment is scheduled to begin 4Q2011.
Shan Lu explained that DNA elicits a good T cell response, and it’s also a good B cell prime, but it needs a protein boost to get the broadest possible antibody response This approach, utilizing a polyvalent boost (with Env antigens from several major subtypes), has produced sterilizing protection in NHP and cross-clade neutralizing antibodies in a Phase 1 human trial.
To find the optimal polyvalent formulation, investigators screened antigens from circulating isolates and then tested different combinations, randomly at first and later using rational design techniques. The resulting 5-valent protein boost formulation has been tested in healthy volunteers, showing potency but some safety problems, including skin rash and vascularization. Investigators are looking for a safer adjuvant, possibly MPL rather than QS-21, as well as ways to incorporate the adjuvant into the formulation. Manufacturing studies will follow.
In response to questions, Lu said that boosting with the combination of epitopes produced better results than boosting with each of them separately.
Susan Barnett stated that basic studies have achieved preclinical proof of concept for prime-boost vaccines, and the next step is to work on durability. Her work focuses on the use of virus replicon particles (VRPs) from the Alphavirus, a very small RNA virus. VRPs do not encode for structural protein and thus do not encounter anti-vector immunity, and NHP trials have achieved protection with a VRP prime + Env boost. A three-armed trial in NHPs, testing VRPs + protein versus protein alone, is due to begin in early 2011, with results due late summer, 2011.
Investigators are also working on improved selection, combination, and presentation of critical Env epitopes to include the most highly conserved epitopes, those that elicit the highest antibody titer and breadth.
In response to questions, Barnett said that the protein boost is administered in very small doses, as small as 50 mg for a single protein and even lower for combinations.
Harriet Robinson explained that granulocyte-macrophage colony stimulating factor (GM-CSF) is an immune cytokine that stimulates the production of white blood cells. It also stimulates the myeloid dendritic cells to produce retinoic acid, which induces mucosal responses. GM-CSF is a well studied vaccine adjuvant, best produced in vivo and at controlled levels, so as not to suppress adaptive responses. In NHP trials, the use of GM-CSF as an adjuvant resulted in a significant improvement in avidity against the E660 Env protein, as well as higher titers of neutralizing antibodies. Seven of nine animals have been protected against infection and will be boosted after a full year and rechallenged. To date there is no sign of occult infection. Several mechanisms of protection have been hypothesized. This suggests that GM-CSF could be a valuable component in a combined vaccine regime.
In answer to questions, Robinson added that GeoVax is moving as quickly as possible to advance this adjuvant. Participants raised several questions about virion capture, whether the affinity studies had been conducted with cleaved or whole protein, antibody isotypes, and other technical topics.
Julie McElrath discussed three challenges in evaluating adjuvants for human clinical trials:
Poly-IC/LC is a well studied immunostimulant with over 30 years of clinical data, but because safety is an overarching concern, investigators conducted a number of trials in vitro and in mice, macaques and humans to determine the safety of poly-IC/LC as an adjuvant for HIV vaccines. These trials confirmed its safety and showed that, compared with other adjuvants, poly-IC/LC produces more antibodies, more polyfunctional CD8+ cells, and more durable CD4+ cells and cytokine responses. At week 15 in a mouse studies, the combination of gp140 and poly-IC/LC prime + NYVAC boost looks the best so far. A new NHP study is currently underway, building on the lessons learned in RV144, and additional adjuvant/vaccine studies are planned. One study will look at HIV-1 CN54 Env protein in combination with six different adjuvants.
In response to questions, McElrath said that investigators do not yet understand the range of responses observed in NHP trials, nor do they understand all of the messaging mechanisms that are initiated by poly-IC/LC.
In the general discussion that followed, several AVRS members noted the vital role played by nonhuman primate (NHP) studies in this research. Some of them wanted to know what impact the ongoing reorganization of the National Center for Research Resources (NCRR) would have on the operations of NCRR’s various primate research and resource centers; the answer isn’t yet clear, and NIH staff indicated that the community will have to wait to see if funding will be affected.
The meeting recessed for the evening at 5:00 p.m.
Nancy Haigwood reconvened the meeting at 8:30 a.m. and presented a brief summary of the presentations and discussion from Day One, stressing promising data from early research and the growing collaboration between NIAID and other players in moving toward a deployable vaccine.
Barton Haynes, joining the meeting by telephone, presented an outline of the process that will be used to identify correlates of protection from RV144. Pilot studies are being conducted by 35 investigators from 20 institutions, comparing 35 different assays from week 0 and week 26 samples from 100 uninfected subjects (80 vaccinated and 20 placebo) in order to determine which assays to use in the larger case-control study, which will look at 240 vaccinated and 41 placebo using a broad range of assays and technologies. This approach should be able to identify one or more correlates of risk and may be able to identify surrogate endpoints for HIV infection, but it is unlikely to identify absolute correlates of protection, which would have to be validated in further trials.
Preliminary results have identified three regions of HIV gp120 where neutralizing antibodies are binding, as well as antibodies that promote the killing of virus-infected cells, evidence that the RV144 vaccinees have refocused their immune responses on functional Env regions. IgG responses were also specific. However, the duration of this protection was relatively short (less than 26 weeks), and the responses were weaker than those observed in VAX003. These findings raise additional questions to be answered in the case-control studies.
In response to questions, Haynes added the case-control assays will be conducted between March 15 and April 15, with analysis to begin on May 1. Analysis should take about a month. This approach is limited by the scarcity of cells for analysis, which puts a premium on assays that require only a small number of cells. Any follow-up to RV144 should use as similar a vaccine as possible, in order to facilitate comparison and interpretation. The biologically active gD region has emerged as a promising target, with at least seven antibodies that bind to some configuration of gD. The reasons are unclear.
Nina Russell noted that the vaccine efficiency (VE) in RV144 was 60 percent at month 12 but only 31 percent at month 36, and vaccine had no impact on viral load. To address these shortcomings, Gates is supporting the creation of PPCPP, whose goals are to complete the RV144 correlates analysis, characterize the resulting immune profile, and improve or sustain those results with a new adjuvant and/or a protein boost at 12 months. PPCPP will also move to expand this proof of concept immediately into higher-incidence populations, including MSM in Thailand and high-risk heterosexuals in South Africa. PPCPP will also support the development and evaluation of next-generation products as quickly as possible. PPCPP includes Gates, NIAID, HVTN, MHRP, Sanofi, Novartis, and host-country representatives; partnership governance is currently being negotiated.
Jerome Kim, joining the meeting by telephone, described a global strategy to build on RV144 and accelerate the development of an effective HIV vaccine. RV144 achieved VE of 60 percent, and a 12-month boost and/or new adjuvant should improve durability. A VE of 50 percent, or even 30 percent, would offer significant public health benefits in Thailand and other epidemic regions. In the next phase, RV305 will try to extend the efficacy of the vaccine with a protein boost, and RV306 will attempt to expand its immunogenicity with a new adjuvant. New Phase 2b trials are under development to include MSM in Thailand and boosting and adjuvant, while extending it to high-risk populations in Thailand and (later) South Africa. In addition, future trials will look at vaccines in the context of other prevention modalities, including circumcision, microbicides and PrEP. In all of this, however, investigators will “follow the science”—the correlates of protection that flow from RV144 will inform any follow-up, with the next round of clinical trials to start in 2014.
James Kublin discussed plans for follow-up trials in South Africa, where HVTN proposes an adaptive approach—a series of trials that build on one another to improve the efficacy and durability of the vaccine, with a view to licensure of a new version of the vaccine. The current plans call for a “rolling Phase 2a/2b” trial with about 2000 volunteers in each of three arms—two vaccine and a shared placebo—with a sequential, adaptive design that allows the opportunity to accelerate forward or cycle back, as needed. The overall goal is to “fill in the immunological space,” with the primary endpoint being VE at 18 months and the secondary endpoints durability, correlates, and comparison between the two vaccines. This approach promises to weed out weak vaccines rapidly, and to detect both harm and high VE just as rapidly. To enrich the analysis of correlates of protection, uninfected vaccinees from an eliminated arm would receive “crossover vaccination” into the other experimental arm. Kublin anticipates a rapid succession of trials, perhaps one or two per year over 4 years.
In the discussion that followed, participants asked whether the vaccine pipeline would support this approach. More important, does the research community have the resources and the will to support this intensive, highly collaborative approach? Several speakers praised the current levels of collaboration and suggested increased communication and cooperation will lead to unexpected discoveries. Others praised the opportunity to look at South Africa in greater detail and ask some of the same questions there that were asked by RV144 in Thailand. But some speakers pointed to differences in the South African context, including the growing use of other preventive modalities that might give rise to confounders. Kublin was confident that South Africa had the capability to handle multiple overlapping studies of different modalities, as well as different vaccines. Several speakers urged the research community to wait for the results of the RV144 case-control studies before deciding on next steps.
Elizabeth Adams presented a summary of a statistical workshop that was held on January 11 to evaluate a proposal from Peter Gilbert for just such an adaptive clinical trial design. The overall response to that proposal had been strongly positive, with participants praising the innovative design and retrospective testing through simulations of the results it might have achieved in previous vaccine trials. They agreed with the assertion that this design could weed out weak vaccines rapidly, and they agreed with the use of baseline predictors and crossover vaccination to increase the power of analysis for correlates of protection.
However, participants also asked whether the monitoring was too frequent (monthly vs. 6-month) and whether there was an adequate rationale for the assays that were to be conducted. They urged investigators to limit the trials to three vaccine arms and to pay special attention to confounders and covariates. The proposal to “roll over” a Phase 2b trial into a Phase 3 trial was controversial, largely because FDA and other regulatory authorities are looking for safety and other data for licensure that might not be forthcoming from the proposed design. Preplanning and careful execution would be vital to success in an adaptive trial, which also raises unresolved questions about recruitment, informed consent, and unintentional unblinding.
Jim Bradac announced that the next meeting of AVRS will be on May 24–25 at the Natcher Building.Correlatives data from RV144 will not be available until September, but the May meeting should provide an opportunity to hear updates on chimpanzee adenoviruses and on PPCPP. Members should forward other topics and suggestions.
The meeting adjourned at 12:00 noon.
Last Updated June 27, 2011
Last Reviewed June 21, 2011