Behind the headlines about vaccine research

11 September 2002. Related: Conference reports, Vaccines and microbicides, World AIDS 14 Barcelona 2002.

Gareth Hardy, HIV i-Base

The basic science content of the XIV International AIDS Conference was considerably less than that of previous years.

While much of the focus of this conference seems to have shifted since Durban 2000 to a more political and humanitarian theme, the majority of the science data currently being reported appears to be presented at the Conference on Retroviruses and Opportunistic Infections, the Keystone Symposia and other smaller meetings.

While a diversity of relevant and interesting oral and poster presentations provided updates on developing research topics, new and novel findings were somewhat thin on the ground. The over riding feeling prevalent at this Conference, even for those following the track A basic science, was one of “business as usual”. Suffice to say there was no major ground breaking data presented.

Of course one of the greatest implications of bringing together 14,000 delegates from just about every country in the world is the huge amount of press attention generated. This may serve to greatly enhance the dissemination of literature, statements, statistics and data, much of which reaches the press’s attention because of the glossiness of the brochures, pamphlets and press releases produced by its protagonists.

A clear example was the thunderous headlines inspired by the Thai AIDS Vaccine Evaluation Group and the US Military HIV Research Programme’s announcement that they are to initiate the world’s biggest phase III preventative HIV-1 vaccine trial at the end of this year: 16,000 volunteers between the ages of 20 and 30 years are to be recruited over one year and followed up for a further three years in what is expected to be a five year trial, ending in 2007. Half the volunteers will receive Aventis Pasteur’s canary pox vector expressing clade E gp120 and clade B gp41, together with gag proteins and protease. Clade E virus is the predominant viral subtype in Thailand. (ALVAC vCP1521) which will be administered at months nought, one, three and six with VaxGen’s whole gp120 (clade B and E) preparation (AIDSVAX B/E) as a booster at months three and six in. The other half will receive placebo at each of the four time points. “Thailand has a long commitment to develop ways to stem this epidemic, and a viable vaccine approach is key to that end” says Dr Vallop, director general of the Department of Communicable Disease Control, Thailand Ministry of Public Health. “The phase III trial is a milestone in the long journey toward an AIDS vaccine.” Says Dr Deborah Birx, director of the US Military HIV Research Programme.

Harriet Robinson of the Yerkes Regional Primate Research Center, Atlanta, USA, presented data on a successful DNA prime, MVA boost system in macaques. Prior studies had determined that this strategy was superior to DNA alone or DNA followed by a whole protein vaccine. Robinson’s team compared the ability of a gag-pol DNA/MVA SHIV vaccination strategy to protect macaques from disease following challenge, to a gag-pol-env DNA/MVA strategy. Animals were challenged at seven months after the boosts with two viral isolates to which cross neutralising antibodies do not occur. The gag-pol construct strategy failed to protect the animals from disease progression as no viral control was seen. However the gag-pol-env construct protected from disease and gave rise to stable control of viraemia and maintenance of normal CD4 T cell numbers in vaccinated animals. Thus the benefit of this construct showed that the env component was important in eliciting immune control following infection. Because of this a whole gp120 preparation was added, modifying the gag-pol-env DNA/MVA regimen. Surprisingly this resulted in loss of control of viraemia and loss of CD4 T cell numbers in these animals. Thus the addition of a whole gp120 preparation abrogated the protection induced by the DNA/MVA vaccine. Further investigations led to the discovery that this gp120 preparation induced the production of non-neutralising antibodies which enhanced SHIV infectivity/pathogenicity which the authors suggested may have contributed to the loss of control of virus in these animals. (WeOrA213)

This effect of a soluble envelope protein preparation may have implications for other similar vaccine strategies, including the phase III prophylactic vaccine studies proposed in Thailand. Indeed one poster presentation by Chien et al showed that in 14 matched pairs of HIV-1 infected rapid and slow progressors antibodies to the CD4 binding domain of gp120 were significantly higher in the rapid progressors compared to the slow progressors, again suggesting a role for such antibodies in enhancing pathogenesis (MoPpA2001). This confirms work previously published by other groups, where higher envelope specific humoral responses are associated with increased risk of disease progression. Taken together these data suggest that while neutralising antibodies to the viral envelope proteins may yet yield an effective means of controlling HIV-1 infection, induction of non-neutralising envelope specific antibodies may have precisely the opposite effect. This is likely to be a very serious consideration for vaccine design and development. However, Thongcharoen presented reults from a phase II, double blind, randomised placebo controlled trial of the Thai vaccine candidates (ALVAC cp1521 and AIDSVAX B/E) where intramuscular injections were given at 0, 4, 12 and 24 weeks (vaccine n=46, placebo n=15). Neutralising antibodies specific to the envelope of clade B lab strain SF2 were found in 27% of vaccinees and to clade E strains in 96% of vaccinees. CD4 lymphocyte proliferative responses to clade B envelope were also detected in 76% of vaccinees and to clade E envelope in 89% of vaccinees. Cytotoxic T cell responses were not yet evaluated (TuOrA1225).

Further evidence leaned toward the belief that mucosal HIV-1 envelope specific IgA antibodies may be protective of infection. In a study by Kebba et al. vaginal secretions were collected and assayed for HIV-1-gp160-specific IgA and IgG antibodies from a cohort of eight Ugandan exposed seronegative (ESN) females with a history of frequent unprotected sexual intercourse with an HIV-1 infected spouse in addition to seven seropositive females (SP). Kebba found that while gp160-specific IgG was not apparent in the vaginal secretions of ESN females, it was readily detectable in six of the seven SP females. Although all ESN females were negative in two HIV serological tests for plasma IgG and negative for HIV-1 proviral DNA gp160-specific IgA was detectable in vaginal secretions and furthermore tended to increase concurrently with increasing plasma viral load in their infected spouse. Whereas IgA in SP females was inversely correlated with autologous plasma viral load. (MoOrA1053)

Jay Levy presented interesting results building on previous work by his group where interferon-a/b producing plasmacytoid cells (IPCs) were found to inversely correlate with viral load and AIDS defining disease. These CD11C+ plasmacytoid dendritic cells are the body’s main interferon-a producing cell. Because they express CD4 and both CCR5 and CXCR4, purified IPCs (>999.5%) were exposed to SI or NSI isolates and stimulated with soluble CD40L to assess their susceptibility to infection. While IPCs could be infected with HIV-1, viral replication in these cultures was found to be minimal, suggesting that they are not major producers of virus. Furthermore infected IPCs did not loose viability, suggesting that they are not killed by HIV-1 and may thus represent an important viral reservoir. In contrast HSV was found to cause loss of viability of IPCs in culture. Following the addition of activated CD4 T cells to IPC cultures, HIV-1 virus was readily recovered, up to seven days post exposure. This transfer of virus to CD4 T cells was found to be independent of DC-SIGN. In addition exposure to free virus did not induce IFN-a production by IPC, whereas exposure to HIV-1 infected CD4 T cells was a potent inducer of IFN-a production.

Levy showed that long-term non-progressors have higher numbers of IPCs than normal progressors and that in normal progression IPCs correlate with CD4 T cell numbers. However a very understudied population of HIV-1 infected individuals who remain disease free for extended periods of time without antiretroviral therapy despite very low CD4 T cell counts appear to have normal numbers of IPCs. Levy suggested that IPCs may thus have a role in protecting from disease if they are sufficient in number in advanced HIV-1 infection. (TuOrA1136)

Nobile et al described analysis of CD4 and CD8 T cell dysfunction during the natural course of HIV-1 infection, using premade cDNA expression arrays spotted with 111 genes associated with cell cycle regulation. Gene expression patterns were compared in CD4 and CD8 T cells between eight uninfected donors, eight HIV-1 infected patients and in cultured T cells stimulated with anti-CD3 and anti-CD28 over 60 hours. While increased expression of cell activation markers such as tyrosine kinases and cell cycle markers such as cyclins and CDKs (cyclin dependent kinases) were observed in uninfected donor CD4 and CD8 T cells after in vitro stimulation, the same pattern was observed in HIV-1 infected patients CD8 T cells only. CD4 T cells of HIV-1 infected patients demonstrated increased kinase expression, suggesting increased cell activation, but specifically lacked expression of the CDK2 gene, crucial to cell cycle progression from G1 phase to S phase. Furthermore, expression of the pro-apoptotic mdm2 gene was elevated in CD8 T cells of HIV-1 infected patients, but not CD4 T cells. This data suggests that while both CD4 and CD8 T cells are hyperactivated in HIV-1 infection, CD8 T cells are actively cycling, where as CD4 T cells are anergic with a cell cycle blockade. (WeOrA1344)

Lending further support to this notion was a presentation by Yassine Diab et al who used newly available HLA DR tetramer/HIV-1peptide complexes with CFSE to identify functional antigen specific CD4 T cell clones in 20 HAART treated patients with primary HIV-1 infection and over three subsequent years of infection. In five out of seven of these patients who were treated early, tetramer stained HIV-1 specific CD4 T cells were detectable and cycled normally as measured by CFSE staining. 80% of these early treated patients maintained these responsive CD4 T cell clones over the study period. However in 11 late treated patients, although tetramer stained HIV-1 specific CD4 T cells were evident, these did not stain for CSFE, demonstrating cell cycle blockade and clone specific T cell anergy. (WeOrA1346)

The clinical implications of this were perhaps illustrated in a presentation by Lange et al in which 29 patients receiving HAART with HIV-1 RNA <400/ml for at least nine months and a CD4 count > 450 were assessed to see if their CD4 nadir before receiving HAART effected their ability to mount antibody and recall lymphocyte proliferative responses following immunisation with tetanus, diptheria typhoid or KLH. Responses were measured at the time of immunisation and four weeks later. An immune response score (IRS) was calculated from the antibody concentration, lymphocyte proliferative response and delayed hypersensitivity reactions induced by each vaccination. Patients were stratified according to whether they had experienced a pre-HAART CD4 count nadir of <250 cells or not. Although there were no differences in the CD4 counts at the time of vaccination (744 cells in the low nadir group and 724 cells in the high nadir group) there was a marked impairment of the IRS in the low nadir group (p=0.005). There was a linear relationship between the IRS and CD4 cell nadir at levels between 150 and 450 cells (r=0.6, p=<0.01), but no relationship was apparent with CD4 count at the time of vaccination. This suggested that even when good immune reconstitution occurs in terms of CD4 counts and control of viral load is maintained, the CD4 nadir, and thus level of progression, prior to HAART predicts the response to vaccination while current CD4 count does not. (LbOr09)

Gotch et al presented a study where 36 patients treated with HAART for 17 weeks were randomised to receive either four injections of the whole killed HIV-1 gp120 depleted immunogen (Remune) three monthly, three cycles of IL-2 for five days four weekly, a combination of the two or just continued HAART. The mean CD4 T cell count of these patients prior to HAART was 303. Again CD4 lymphocyte proliferative responses to the vaccinating agent, or to other HIV-1 antigens, were not induced, even with the addition of IL-2. However it was noted that transient viral blips associated with each cycle of IL-2 in this study, were much less apparent in those patients receiving Remune with IL-2. (ThOrA1483)

An analysis of peripheral and lymph node CTL responses before and during a structured treatment interruption study by Van Lunzun et al may shed some light on why CTLs induced by STI in chronic infection fail to suppress viral load. In this study 15 individuals with <25 HIV-1 RNA/ml plasma and normal CD4/CD8 ratios after a median of 18 months of HAART had a single STI. CTL responses were measured by IFN-g ELIspot using HLA restricted peptides. The viral load set point upon TI remained unchanged and was acquired by a median of four weeks following the stop (median two – six weeks).

This viral load increase correlated with CD38 and HLA-DR activation markers on CD4 and CD8 T cells and inversion of the CD4/CD8 ratio. The numbers of CTL in the periphery strongly correlated with the numbers in the lymph nodes (R=0.92, P=<0.0001). Lunzen showed that HIV-1 specific CTL in the lymph nodes are stronger and broader than in the peripheral blood. In conclusion it was suggested that the HIV-1 specific CTL that emerge in the periphery following STI are expanded from those that persist in the lymph nodes during HAART, which would thus not effect viral set point. (TuOrA1180)

All references to the XIV International AIDS Conference, Barcelona, 7-12 July.