Are Babies in a Blind Spot?
Pediatric Workshop Highlights Barriers to Neonatal HIV Vaccine Development
By Emily Bass
This October, the Elizabeth Glaser Pediatric AIDS Foundation (EGPAF) co-sponsored its second meeting on HIV vaccines and immunoprophylaxis in infants.* Held in Seattle, the 2002 gathering shared a location and many participants with the biannual meeting of the HIV Vaccine Trials Network (HVTN). But in spite of this neighborliness, the Glaser meeting underscored the fact that pediatric studies are still marginal to AIDS vaccine research. The last pediatric AIDS vaccine trial started in 1997 (see Table 1), and it is still the only one out of an estimated 24 ongoing studies of preventive vaccines which includes infants. There has never been a pediatric vaccine trial in the developing world, where there is an acute need for strategies to prevent breastmilk transmission of HIV.
|
Table 1: Pediatric Vaccine Trials—A Brief History* |
|
|
Planned Trials Vaccine |
Vaccine |
| HPTN 027 Phase I preventive vaccine study in 40 Ugandan infants of HIV-infected women Protocol to be re-submitted to Ugandan regulatory authorities Q1 2003 |
ALVAC vCP1521 (clade E) |
| PACTG 1033 Phase I prime-boost therapeutic vaccine study in 16-20 US HIV-infected children on HAART |
MVA (gag, pol, env, nef, rev, tat) + Fowlpox vaccine (gag, pol, env, nef, rev, tat) Both vaccine based on pediatric clade B isolate and manufactured by Therion Biologicals (Clade C version planned ) |
| Protocol #TBA Phase I study of topical, therapeutic DNA vaccine in 18 HIV-infected, HAART-treated children 3 years or older |
DermaVir DNA vaccine (clade B) with all HIV genes except integrase (see IAVI Report, Jul-Sep 2002) |
|
Ongoing Trials |
Vaccine |
|
PACTG 326 |
Group 1: ALVAC vCP205 (2 doses) Group 2: ALVAC cCP1452 +/- gp120 B/B |
|
Completed Trials |
Vaccine |
|
PACTG 230 |
Group 1: SF-2 (Chiron) Group 2: MN gp120 (Genentech) |
|
PACTG 218 |
Group1: SF-2 (Chiron) Group 2: gp120 (MN) protein subunits (Genentech) Group 3: gp160 (MicroGenSys) |
|
Not included in this table: AVEG 104, a trial of rgp120 in HIV-infected pregnant women; PACTG 185, an early passive immunization study |
|
Nor are there signs of imminent change. The only trials in the cards are the oft-postponed HIV Prevention Trials Network protocol (HPTN 027) for Uganda, which originated in 1999, and two protocol concepts that have been approved by the Pediatric AIDS Clinical Trials Group (PACTG). But none of these trials has an official start date, making it unlikely that they will begin before 2004 at the earliest.
Seen from a research perspective, the big news from the Seattle meeting was that there is almost no news at all. Attended by nearly 70 people, the gathering provided an unusual degree of cross-talk between pediatric researchers, pharmaceutical companies and the FDA. Unlike the 2001 meeting, which had no industry representation, this year's event was attended by most of the major vaccine companies, including Aventis Pasteur, Chiron, Glaxo- SmithKline, Merck and Wyeth. However, with the exception of Aventis, which will collaborate on the 027 protocol and has worked on a similar US trial, none of the companies has plans for new pediatric studies of HIV vaccines.
Why Test HIV Vaccines in Babies?
The inaction belies an urgent need for a neonatal vaccine. Recent figures from UNICEF and UNAIDS indicate that an estimated 1,700 infants are born with HIV every day, and another 300 acquire the virus through breastmilk transmission. While there are several cheap antiretroviral regimens to help prevent in utero and intrapartum infection, there is still no completely effective way to minimize the risk of transmission via breastfeeding, which remains a common choice among women in developing countries for a variety of reasons.
Treating HIV-positive women is one important strategy, since antiretrovirals (ARVs) would improve maternal health and reduce the amount of virus in breastmilk. This approach is being rolled out, albeit slowly, through MTCT Plus programs that offer medications to women and their families. But in practice, it will be many years before all women can access these drugs. Should a neonatal HIV vaccine become available, it could be added to the prevention package and without ARVs, an immunization approach is even more important.
Despite the urgency of the problem, the current paradigm for HIV vaccine development is to push forward with adult trials and hold off on pediatric studies until a product is much farther down the testing pipeline in adults. Jeff Safrit, senior programs officer at the EGPAF and a co-organizer of the Seattle meeting, says he's been told by the HVTN and the Division of AIDS (DAIDS) that they will move into new pediatric trials once there's a vaccine in hand and ready for Phase III trials in adults. This is due, in part, to the explicit mandate of some networks. The hand that's been dealt [the HVTN] is adult trials, HVTN head Larry Corey told the gathering. Our aim is to get safety and immunogenicity data to the pediatric vaccine community as soon as possible. IAVI, which currently conducts its trials exclusively in adults, is also considering pediatric trials once the adult program is further along. At the moment, VaxGen, which is poised to release data from its Phase III trials in 2003, has no plans for pediatric bridging studies if AIDSVAX shows some efficacy in adults.
Statistics aside, there are good scientific reasons for pursuing an HIV vaccine for babies. One is that it may be easier to develop a vaccine which protects infants than adults, since protection is only needed for the duration of breastfeeding. There are also examples of currently licensed vaccines—including chickenpox and flu—which are more effective in children than in adults. Put another way, if the world waits for adult efficacy before launching pediatric trials, it could inadvertently discard a worthwhile vaccine. On a practical level, global vaccine delivery systems are set up to serve babies, not adults—so an effective HIV vaccine could probably be rolled into existing childhood immunization programs. In contrast, in many countries, adult vaccination programs will require delivery systems that have yet to be developed.
All these rationales have been forward for many years, to little or no avail. It’s a difficult state of affairs—particularly to veterans of the field. Pat Fast, IAVI’s Medical Affairs Director, is a pediatrician who watched the launch of the first pediatric vaccine trials in 1993 as a member of the HIV vaccine program at the National Institutes of Health (NIH). She recalls that the early studies were seen as a first step to testing candidates in developing countries, following the logic that it would encourage pediatric vaccine trials in Africa, for example, if similar studies had already been done in the US. But so far, the world has yet to take that step.
Presenters at the Seattle meeting catalogued the sources of delay. Frances Mmiro (Mulago Hospital, Kampala) and Laura Guay (Johns Hopkins University, Baltimore) each gave updates on HPTN 027, which has been delayed by indecision over whether to use a well-studied but non-clade matched vaccine—even though Uganda tested an unmatched vaccine in 2000-2001, and detected cross-reactive immune responses (XIVth International AIDS Conference, Abs. #TuOrA1226, ThPeA7081). The trial was further delayed as investigators waited for additional data from PACTG 326. The current plan is to re-submit the HPTN 027 protocol with a clade E canarypox-based vaccine (ALVAC vCP1521, another unmatched construct, which is slated for use in the planned US Army-Thai Phase III trial). Guay says that the decision to use the unmatched construct was based on vaccine availability—a clade A version of vCP1521 is being developed, but will not be ready until 2004.
Mmiro and Guay described a domino effect, where delays in US trials can topple pediatric studies in other parts of the world. It’s a situation that is likely to get worse, warned University of Rochester’s Colleeen Cunningham, since it is becoming increasingly difficult to enroll either HIV-positive or -negative infants in pediatric vaccine trials in the US and Europe. ARVs and the relative ease of formula feeding in these regions have slashed rates of MTCT, so very few HIV-positive babies are born there. Against this background, many parents are reluctant to enroll either HIV-negative or -positive babies in research studies. “Any expectation that a trial has to be done in babies here [in the US] before it goes elsewhere is going to be a problem,” Cunningham said.
FDA regulatory requirements can also add to the delays, says Katherine Luzuriaga (University of Massachusetts, Worcester), a meeting co-organizer and principal investigator of a proposed therapeutic trial in HIV-infected, HAART-treated babies (PACTG 1033) that would test MVA- and fowlpox-based HIV vaccines. The standard FDA practice is to require “safety and immunogenicity data in relevant adult populations” before approving pediatric clinical trials. While this guideline reflects an understandable degree of caution, it is also somewhat vague about how much, and what type, of adult data is sufficient. In the case of PACTG 1033, for instance, Luzuriaga was surprised by an FDA request for Phase I trial data in both HIV-positive and HIV-negative populations. Since the protocol proposed testing the vaccine in HIV-positive children, the investigators only anticipated a request for data in infected adults.
The FDA requirement for immunogenicity data was also questioned at the meeting. “If we all agree that there are no correlates of protection, why do we have to show immunogenicity in adults before going to infants?” asked Colleen Cunningham. Other participants agreed, pointing to examples of age-related differences in vaccine effects as they argued that it may be unreasonable to require adult immunogenicity data before moving to pediatric populations.
Another regulatory issue: the FDA perspective that each new genetic insert—for example, from a different clade—constitutes a new vaccine, even if the vector has received prior approval. In the case of MVA, there have been several Phase I trials of MVA-based HIV vaccines, both in uninfected and infected adults, while an MVA-malaria vaccine (now in Phase III adult studies in The Gambia) has also been tested in a Phase I pediatric study (see IAVI Report, Oct/Nov 2002, Malaria Vaccine Trials Underway in Africa). Despite this prior experience, the investigators of PACTG 1033 must submit their own portfolio of data, specific to their vaccine. And if they carry through with plans to develop a clade C version of the vaccine, they may well be required to repeat a full set of adult trials with this candidate before starting pediatric trials in the developing world. With fowlpox, the situation is more ambiguous: there is one ongoing adult trial of a fowlpox-based HIV vaccine in HIV-infected people in Australia, but there is less data so far than for MVA. Overall, the situation comes back to the need voiced by many participants: an appropriate, clearly-defined standard for safety data from other trials, which could support applications for new studies in lieu of launching completely new adult trials.
During the meeting, FDA representative Joseph Toerner was peppered with questions about data requirements under different scenarios, and with requests for a special FDA advisory committee to review the specific issues related to HIV vaccine research in children. In his talk, Toerner re-stated the FDA’s standard—laid out by the International Council on Harmonization and the American Academy of Pediatrics—that adult safety and immunogenicity data with the identical candidate be collected before a pediatric trial moves forward. In a follow-up interview with the IAVI Report, Toerner reported that he brought feedback from Seattle to his colleagues, and that they had decided to maintain the current guidelines but to consider departures on a case-by-case basis.
Scientific Support for Infant Trials
The meeting’s smattering of new data bolstered the scientific rationale for infant vaccines—and pointed at some of the outstanding questions about current animal models. Colleen Cunningham presented unpublished data from PACTG 230, a study of ALVAC vCP205 in 20 HIV-negative infants. The findings: cellular immune responses (lymphoproliferation and CTL) were similar to those seen in adults. Further insight into optimal dosage and intervals should come from final analysis of this trial and of the ongoing PACTG 326, a study comparing two slightly different canarypox-based vaccines (ALVAC vCP205 and ALVAC vCP1452), with or without a gp120 boost.
Primate researcher Marta Marthas (University of California, Davis) then presented her latest data on vaccine-induced protection of immunized infant monkeys who receive multiple, low-dose oral challenges of SIVmac251, a model that is currently the best approximation of breastfeeding exposure. Marthas has now conducted two trials of ALVAC and MVA-based vaccines, and found that both appear to protect some immunized animals against SIV infection (see Table 2). The first of these studies, presented at last year’s meeting, used different immunization schedules for ALVAC and MVA; the newer study standardized the vaccination regimen, and again found evidence of protection. (The different infection rates in the 2001 and 2002 experiments are not statistically significant, according to Marthas, due to the small sample sizes; protection of ALVAC vaccinees versus controls was significant at p<0.005.) Interestingly, the ALVAC vector alone appeared to provide some protection. This may be a sign that general immune stimulation could improve the newborn immune system’s ability to fight off disease—a theory also discussed at the last meeting.
|
Table 2: Low-Dose Oral Challenge Following Vaccination of |
||
|
Vaccine |
Animals infected** / Total | |
|
2002 |
2001 |
|
| None | 7/8 | 7/8 |
| MVA-SIVgpe weeks (0,3 in 2001; 0, 2, 3 in 2002) |
4/8 | 7/9 |
| alvac-sivGPE (0,2,3 weeks) | 4/8 | 2/8 |
| ALVAC vector alone | 2/4 | N/A |
| * data from Marta Mathas, University of California, Davis ** infection defined as persistent viremia (SIV in the blood) 6 months after oral challenge |
||
Pushing Ahead with Passive Immunization
Ruth Ruprecht (Harvard Medical School, Boston) speaking on behalf of Hoosen Coovadia (University of Natal, South Africa), presented a proposed South African protocol to test whether a cocktail of monoclonal antibodies (MAbs) to different HIV epitopes can prevent breastmilk transmission. The first phase of the study would enroll HIV-infected infants (diagnosed at birth using PCR), so that the safety profiles can be directly compared with data from a completed trial in infected adults. The babies would not receive antiretrovirals, which are not widely available in South Africa. Once safety data from the HIV-positive babies are in hand, the investigators propose a Phase Ib study in HIV-negative infants. The rationale for this approach is that even an effective HIV vaccine given to newborns would leave them vulnerable to breastmilk transmission for the first few weeks of life, before protective immunity was established—and that passive immunization might provide “cover” during this high-risk period. Proof-of-concept studies by Ruprecht and others have shown that high doses of MAbs can protect infant monkeys against high-dose oral challenge. At the meeting, Ruprecht presented un-published data showing that MAbs also protect when delivered up to one hour after challenge.
Laura Guay reported on a Phase I/II study of a related, relatively low-tech approach using “hyperimmune serum” made from purified antibodies from HIV-infected individuals. In a safety and dose-escalation study of one such product, called HIVIGLOB, 29 mother-infant pairs received infusions; infants were monitored for adverse events and levels of HIVIGLOB in the blood for 30 months following the infusion. This study found that HIVIGLOB was safe and well-tolerated, and had a half-life of 30 days. The study was not designed to measure efficacy, and gave the mothers and infants a single infusion of antibodies. (An earlier trial, PACTG 185, gave multiple doses to the mother, but was discontinued after the advent of AZT regimens to reduce MTCT.) The next step is to compare HIVIGLOB and short-course nevirapine (NVP) given to mother and baby with short-course NVP to both plus six weeks of NVP prophylaxis for the babies during breastfeeding. The control arm will be short-course NVP alone.
At the meeting, enthusiasm for passive immunization was mixed with skepticism about its feasibility, compared with the alternative approach of short-term, prophylactic antiretroviral therapy. MAbs might have fewer toxicities and side effects than antiretrovirals. But they have several drawbacks, including difficulty of administration compared with ARVs (they are given via intravenous infusion or intramuscular injection), the large amounts of antibody needed for protection (which translates into a high cost), and the relatively short duration of protection, which wanes as antibodies are cleared from the body. (Data from HIV-infected adults showed a half-life ranging from 7.94 days to 16.48 days for different MAbs; there is no comparable data for infants.) Another disadvantage, pointed out by Dorothy Mbori-Ngacha (University of Nairobi): many women deliver their babies outside a hospital setting, which would make it difficult to administer the infant MAb dose within 24 hours of birth. In contrast, mothers could deliver ARVs to the babies at home. Even in a passive immunization context, Mbori-Ngacha suggested that women should be given ARV prophylaxis for their infants.
A Field with No Home Base
Under the current system, advancing candidate vaccines into pediatric trials depends largely on investigator passion and tenacity. It’s a selection process which does not guarantee that the best products will be tested—or that trials will go forward if the investigators move on. But there are alternatives, says Cathy Wilfert, who points out that the HPTN and PACTG could develop and advance a pediatric research agenda. A key obstacle, however, is the lack of funds: Wilfert, who chairs the HPTN perinatal working group, says that the entire US budget for the PACTG is only $30 million per year.
Despite the discouraging state of the field, the Seattle gathering nonetheless seemed to catalyze interest in these issues, with many of the industry and trial network representatives expressing enthusiasm for future collaborations—although time will tell whether this translates into concrete studies, and on what timescale. In lively discussions, participants shared diverse opinions about the best and speediest way to proceed in the times ahead. Some argued that passive immunization trials are ready to launch, and should therefore move forward. Others said that this strategy was unlikely to be practical, especially if multiple infusions are required, and that the push should be for vaccine trials. Still others proposed a trial that would compare MAbs alone, vaccine alone, MAbs plus vaccine, and ARV prophylaxis during breastfeeding.
But, while there was still no consensus at the end of the meeting, there was considerable energy for action. As Bonnie Mathieson, of the Office of AIDS Research, summed up, “We need to get one pediatric trial started as soon as possible, and start to prepare for two or three more. The time to act is now.