ast year, organizers of the annual Retrovirus conference—the largest US HIV research meeting—inaugurated a separate AIDS vaccine conference, leading many to expect that vaccines would not be featured at Retrovirus 2002. But the massive event, held 24-28 February in Seattle, found room for several vaccine talks, including the much-anticipated debut of data from human trials of Merck’s new adenovirus-based vaccine candidate.

Merck’s early trial data

In a packed plenary session Emilio Emini, Merck’s head of vaccine research, presented early immunogenicity results (up to 42 weeks post-vaccination) from HIV-negative volunteers. Merck is also testing these vaccines as potential therapeutics in HIV-positive people. The ongoing trials, which will enroll about 600 people, include one study of a gag-expressing DNA vaccine (1 mg or 5 mg) given either in saline or with one of two adjuvants (aluminum phosphate or CRL-1005, an experimental polymer). A second trial is testing different doses of a gag-containing vaccine made from an adenovirus vector (Ad5).

Merck's phase I AIDS vaccine trials

VACCINE	TIME OF SAMPLING
DNA vaccine (saline)	week 8	week 42a
1 mg	N/A	7/34b (20.6%) [105]c
5 mg	N/A	16/38 (42.1%)
Ad5 vector	week 8	week 42
108 vp (viral particles)	6/9 (67%) [149]	6/9 (67%) [239]
109 vp	4/9 (44%) [149]	4/7 (incomplete data) [337]
1010 vp	5/9 (56%) [257]	No data

^a 4 weeks after final immunization
^b number of people showing immune responses
^c number of spots in Elispot assay (geometric mean)
See Retrovirus conference website for webcast presentation.

Emini reviewed partial data on T-cell responses so far. Starting with the DNA vaccine in saline, Emini called the responses "moderate," while stressing that, with adjuvant, they could be sufficient to prime HIV-specific T-cells prior to an Ad5 boost. Immunogenicity of Ad5 was stronger than with DNA alone.

Emini then addressed a key question relating to the Ad5-based vaccine: the effect of pre-existing neutralizing antibody (NAb) to adenovirus on immunogenicity. Since many people are naturally exposed to adenoviruses (which cause severe colds), they already have NAb to Ad5, which could dampen the vaccine’s effect. The prevalence of these antibodies is estimated at about 40% in North America, but is unknown for many other populations.

Because the Elispot data showed that some participants did not respond at all to Ad5, Emini drew a plot to assess whether this non-response was related to a pre-existing anti-adenovirus NAb titer above 1:200 (30% of North Americans exposed to adenovirus have titers above this level). All responders in the 108 and 109 group had titers below 1:200. Of the non-responders, 1/3 from the 108 group had NAb below the cut-off; for the 109 group, the number rose to 4/5, hinting that NAbs do indeed have a dampening effect. This uneven distribution of pre-existing NAb titers may also help account for the fact that there was no obvious dose response seen in Ad5 vaccinees.

Merck then tested whether the effect could be overcome with a higher dose of Ad5 vaccine. Data from the highest dose (1010) group suggested that this might be the case: Non-responders had NAb titers above 200, but so did 4 of the 5 responders.

Cross clade responses

Emini also presented data on cross-clade reactivity of HIV-specific T-cells, a key question for the HIV vaccine field. The study tested whether T-cells from the blood of HIV-infected individuals from Brazil, Thailand, Malawi and the US—which he said represented infection with clades A, B and C—respond to a panel of clade B consensus peptides (covering the viral proteins Gag, Pol, Nef, Rev and Tat) in an Elispot assay. The results showed that the frequency and magnitude of responses to Gag, Pol and Nef was indistinguishable by region. Rev and Tat induced a lower response in all samples, except those from Malawi, where no responses were seen. The researchers then looked at responses to clades A and C Gag peptides in people who responded to B and found them nearly indistinguishable in magnitude, although there were a few individuals who showed poor cross-clade reactivity.

Complementing the cross-clade studies done in HIV-positive individuals, Merck also looked at cross-reactivity in HIV-negative volunteers from the Ad5 trials, using consensus Gag peptides from clade A or C. Encouragingly, the results showed that T-cells from 10 of 13 (77%) responding individuals cross-reacted.

In closing, Emini looked to the future: Immunogenicity data from the highest (1011) Ad5 dose group and the DNA/Ad5 prime-boost vaccinees, as well as safety and immunogenicity data from ongoing trials in HIV-infected individuals, should be available sometime in mid-2002.

A novel adenovirus vector from chimpanzees

While Merck’s vaccine data underscored the challenge posed by pre-existing immunity to adeno-viruses, a late-breaker presentation from Hildegund Ertl and colleagues at the Wistar Institute in Philadelphia suggested at a potential solution: a chimpanzee adenovirus virus called AdC68 that is not neutralized by antibodies to human adenoviruses.

The Wistar team is in the early stages of testing whether AdC68 can be developed as a vector for vaccines, and presented results obtained so far in mice. They began by deleting an essential region (called E1) of AdC68 to create a replication-defective vector expressing a truncated form of HIV-gag and found that it induced strong CD8 T-cell responses in mice, as measured by Elispot. They plan to study AdC68 in non-human primates and, eventually, to conduct clinical trials.

DNA prime/NYVAC boost in Rhesus macaques

Genoveffa Franchini presented results on macaques challenged with the highly pathogenic SIVmac251, after immunization with DNA and NYVAC (attenuated vaccinia virus) vaccines. Pre-challenge immunogencity data from this study was published last year (J. Immunol., 2001; 167:7180-7191). SIVmac251 is considered one of the most stringent challenge strains, and there have been only a few studies showing good vaccine-induced protection against it.

Franchini found that 5 out of 8 animals immunized with NYVAC alone showed some reduction in viral replication, to a set point be-low 100,000 copies after a typical acute phase. In contrast, 5/8 animals given a DNA/NYVAC prime-boost combination had significantly lower viral load peaks and controlled SIVMac251 replication to undetectable levels at set point. These five animals have preserved their CD4 cell counts and maintained control of SIV replication out to a year. Looking at correlates of protection, Franchini found that pre-challenge T-helper responses to p27gag and CD8 T-cell responses to the CM9 Gag epitope correlated with post-challenge virus control.

One caveat concerns the genotype of the animals used: 5 out of 8 monkeys in the prime-boost group were Mamu *A01 positive (an MHC class I type associated with a natural ability to control SIVmac251); these animals accounted for 4/5 of those that contained viral load. But 4/8 animals receiving NYVAC alone (and which controlled viral load only partially) were also Mamu *A01 positive, and the less robust control of viremia in this group indicates that the prime-boost regimen had a significant effect, independent of Mamu *A01.

------------
^* Richard Jefferys is the Basic Science Project Director at the Treatment Action Group, a New York-based organization advocating for HIV research. He was previously director of the Access Project at the AIDS Treatment Data Network, and most recently worked as a writer for the IAVI Report.