Após 25 anos, vacina contra HIV permanece inalcançável

Seqüência de falhas em testes promissores fez Nobel dizer que não há esperança de sucesso; para cientistas, é preciso elucidar biologia do vírus

Métodos que ajudam a evitar contágio, como gel e circuncisão, têm problemas, afirma infectologista brasileiro

EDUARDO GERAQUE
DA REPORTAGEM LOCAL FSP

Indomável é o melhor adjetivo para qualificar o vírus HIV. Após 25 anos da publicação, na revista "Science", do primeiro isolamento desse parasita, a comunidade científica só acumula frustrações. Nenhuma barreira bioquímica desenvolvida até agora conseguiu conter a infecção. O desenvolvimento da tão almejada vacina ainda é apenas um sonho distante.
"A comunidade científica está deprimida, porque nós não vemos esperança de sucesso", declarou em fevereiro o biólogo americano David Baltimore, que ganhou o Nobel por ter descoberto o mecanismo de replicação de vírus como o HIV.
"Penso que o maior erro que cometemos até agora foi colocar muito foco nos testes em larga escala de vacinas e não dar atenção suficiente para a pesquisa básica. Nós precisamos aprender com isso", afirma à Folha Dennis Burton, do Instituto de Pesquisa Scripps, na Califórnia (EUA).
O pesquisador assina um dos artigos do especial de 25 anos da descoberta do HIV, publicado na última sexta-feira pela mesma "Science".
O retrocesso científico do fim do ano passado é exemplar. O teste mais avançado de uma vacina contra a Aids, criada pela farmacêutica Merck, foi suspenso após 82 voluntários (49 vacinados e 33 do grupo de controle) terem sido contaminados pelo vírus.
"O número foi pequeno e não foi a vacina que causou a contaminação, o que seria inaceitável", diz Esper Kallás, infectologista, professor da Unifesp e coordenador dos testes com a vacina no Brasil. No mundo, 3.000 pessoas receberam doses do medicamento.
"A defesa imunológica que a vacina conferiu aos participantes não foi suficiente. O vírus continuou passando [pelas defesas do corpo]", afirma.
Apesar de desde o começo das pesquisas, há 25 anos, o HIV ser qualificado como versátil, o desafio a cada novo estudo é maior. O vírus não pára de surpreender por causa da sua alta velocidade em desenvolver novas formas genéticas.
"O HIV consegue driblar o sistema imunológico com mais habilidade do que nós imaginávamos", diz Kallás, que concorda com a necessidade da volta rápida para a pesquisa básica. "Apesar de o vírus ser o mais estudado do mundo, temos de aprender mais sobre o comportamento dele", diz.
Toda a comunidade científica concorda com a volta para a bancada. E, mesmo depois dos últimos insucessos, em continuar na busca pela vacina. A urgência é grande. Números da OMS (Organização Mundial de Saúde) mostram que a cada dia mais de 6.800 pessoas são contaminadas. E 5.750 morrem.
Como os resultados do fim de 2007 que levaram o projeto da vacina a dar errado ainda não estão prontos, afirma o infectologista da Unifesp, a lista de perguntas que precisam ser respondidas é enorme.
"Nós colocamos pequenos fragmentos do HIV na vacina que foi aplicada. Será que precisava ser mais? Será que foi a quantidade de vacina em cada uma das doses? Será que temos de fazer combinações de vacinas? Ou mudar a forma de apresentar os fragmentos do vírus para o organismo"?
Para John Moore, outro pesquisador que participou do especial da "Science", a questão do desconhecimento sobre como fazer uma vacina é a que tem mais peso no momento.
"Efetivamente, nós não sabemos como fazer uma vacina que seja realmente eficiente", diz o pesquisador da Universidade Cornell (EUA).

Circuncisão e gel
O fracasso de uma vacina não é o fim do mundo, segundo os cientistas ouvidos pela Folha. Pelo menos há uma base de onde é possível continuar.
"Não podemos abandonar a vacina. Ela tem de continuar a ser um sonho", afirma Kallás. "Não podemos dar uma declaração de derrota e voltar as costas para aquilo que pode ser a maior esperança para combater a epidemia. Hoje são 33 milhões de pessoas contaminadas. Daqui a 20 anos, talvez sejam 150 milhões."
Segundo Kallás, existem poucas opções que possam ser aplicadas como políticas de saúde pública, apesar de a circuncisão, por exemplo, já ter se mostrado eficaz em barrar a transmissão e vários géis vaginais estarem em teste hoje com o mesmo fim. "A circuncisão e o uso de géis têm problemas.Temos que partir do princípio que a maioria das pessoas gosta de fazer sexo e o vírus se transmite por via sexual."
Mesmo com pelo menos 20 projetos em curso hoje no mundo para o desenvolvimento de uma vacina contra a Aids, é difícil encontrar um otimismo rasgado entre os atores da comunidade científica.
O pesquisador da Escola Médica de Harvard, Bruce Walker, é um dos que fogem um pouco dessa regra. "Certas pessoas estão infectadas com o HIV há 30 anos e não estão doentes. Estou otimista que poderemos usar esse dado para criar uma vacina efetiva", diz. "Mas isso, como foi visto nos últimos 25 anos, não será fácil."

AIDS/HIV:
A STEP into Darkness or Light?

Science 9 May 2008:
Vol. 320. no. 5877, pp. 753 - 755

John P. Moore^1* P. J. Klasse,¹ Matthew J. Dolan,² Sunil K. Ahuja^3*

¹Weill Medical College of Cornell University, New York, NY 10065, USA

²Infectious Disease Clinical Research Program, San Antonio Military Medical Center, Ft. Sam Houston and Lackland AFB, TX 78234, USA

³Veterans Administration Research Center for AIDS and HIV-1 Infection, South Texas Veterans Health Care System, San Antonio, TX 78229, and University of Texas Health Science Center, San Antonio, TX 78229, USA

The outcome of the efficacy trial of an adenovirus serotype 5 (Ad5) vectorbased HIV-1 vaccine last November (STEP trial) was unexpected. Not only was the vaccine ineffective at lowering plasma viremia postinfection, but it may have increased the risk of acquiring HIV-1 infection. Although firm conclusions cannot be drawn based on the small number of infections that occurred (49 in the vaccinated patient group, 33 in the placebo group), it has been suggested that vaccineinduced generalized immune activation, which can promote HIV-1 replication, might have increased the infection risk (1).

The vector-based vaccine used in the STEP trial was a recombinant Ad5 virus expressing immunogenic HIV-1 proteins. A higher number of HIV-1 infections occurred in the subset of vaccinees with high, preexisting titers of Ad5-specific antibodies, compared with placebo recipients. One possible explanation is that anti-Ad5 antibodies facilitate cellular uptake of the Ad5 vector (perhaps by cells other than the ones normally targeted), inducing an immune response that enhances HIV-1 infection. Although immune responses to viral infections are usually protective, they can also be harmful (as with West Nile, dengue, measles, and respiratory syncytial virus infections). For example, a low-titer antibody response to West Nile virus can enhance viral replication and exacerbate disease (2). Whether similar events occur after vaccination with an Ad5 or similar viral vector is now something to consider.

One way to examine the apparent effect of anti-Ad5 antibodies is to plot the relative risk of HIV-1 infection in the STEP trial groups as a function of antibody titer (see the figure). Unexpectedly, the higher the anti-Ad5 antibody titer in the placebo group, the lower the HIV-1 infection rate. By contrast, infection risk in the vaccinated group appears similar at high and low titers. Given statistical limitations, a conservative explanation is that these patterns arise by chance. However, because of the need to understand all aspects of the STEP trial, we here consider whether the data patterns are meaningful.

It seems implausible that Ad5-specific antibodies directly protect against HIV-1 infection, as no reasonable mechanism is apparent. On the other hand, it may be that an individual with a high-titer Ad5-specific antibody response can better resist HIV-1 infection naturally. The anti-Ad5 antibody titer could thus be a surrogate marker for a host genetic constitution that confers reduced susceptibility to HIV-1. Other factors, perhaps including immune activation, might outweigh this effect in Ad5 vaccine recipients.

A possible precedent arises from two earlier HIV-1 vaccine trials with the gp120 glycoprotein protein of HIV-1 (AIDSVAX). The infection rates were almost the same in vaccine and placebo recipients, but the titers of gp120-binding antibodies and the risk of HIV-1 infection were inversely correlated (3) (see the figure). It is not likely that this correlation arose because binding antibodies are themselves protective. For the overall trial outcome to be neutral (the vaccine conferred no protection compared to the placebo), if above-average titers of gp120- binding antibodies directly protected against infection, then below-average antibody titers would also have to act directly, to place individuals at a greater risk of infection than placebo recipients. This seems improbable. The authors of this study argued that the ability to mount a strong anti-gp120 antibody response inversely measures susceptibility to HIV-1 infection. This scenario might also apply to high anti-Ad5 antibody titers in the STEP trial. Hence, individuals that can mount strong antibody and possibly other relevant immune responses to pathogen antigens may be inherently more resistant to HIV-1 infection--they have "better immune systems."

Possible influence of antibody titers on risk of HIV-1 infection. Data from (1) show the relative risk of HIV-1 infection in the Ad5-vaccinated (triangles) and placebo (squares) groups from the STEP trial, as a function of the anti-Ad5 antibody titer range. An anti-Ad5 antibody titer of <18>3) (circles) show the relative risk of HIV-1 infection in recipients of AIDSVAX (gp120) as a function of the titer range of antibodies that block gp120 from binding to CD4, a receptor used by HIV-1 to gain entry into host T cells (other measures of the antibody response to gp120 yield broadly similar plots). The confidence limits on both data sets (not shown) are broad. Both studies are large (5403 participants in the AIDSVAX trial; analysis of the STEP data is based on a subset of 1836 of those enrolled). The statistical significance of the trend shown for the AIDSVAX data is established (3), whereas significance of the trend among the STEP data is debatable (1). Comparison of the associations is therefore speculative. Not every individual may have been exposed to Ad5 in the STEP trial, whereas in the AIDSVAX trial, every vaccine recipient was given gp120.

There is a modest but significant tendency for individuals with weak antibody responses to one component of a vaccine against measles, mumps, and rubella (MMR) to also respond poorly to the other vaccines (4). Moreover, antibody responses to measles and mumps vaccines are influenced by host genetics, including genes encoding antigen-presenting proteins [human leukocyte antigen (HLA)] that are important for immune function, and singlenucleotide polymorphisms in genes encoding cytokines or their receptors, such as interleukin- 2 (IL-2), IL-10, and IL-12 receptor (5, 6). An extrapolation to HIV-1 vaccines suggests that a broad range of host factors, not just those affecting humoral immunity, might influence protection. For example, IL-10 might affect vaccine responses (and susceptibility to infection) by influencing both cellular and humoral immunity (7, 8); gp120 triggers IL-10 production by specific immune cells (dendritic cells) to an extent that varies greatly between individuals (9), which may be relevant to understanding the wide range of anti-gp120 antibody titers seen in gp120 vaccinees (3).

Host genetic factors may confound HIV vaccine trial evaluation (10-15). For example, genotypes of CCR5, the major HIV-1 co-receptor, and the gene copy number of CCL3L1, the most potent and HIV-1-suppressive CCR5 ligand, together influence cell-mediated immunity in both HIV-1-negative and -positive individuals (15). Genotypes associated with reduced cell-mediated immune responses were similar in the control and HIV-1-infected groups, and predicted an enhanced risk of acquiring HIV-1 and a faster disease course (15). Thus, an individual with a "better immune system" might indeed resist HIV-1 infection or partially control replication. This is consistent with observations that pre-seroconversion immune status predicts the rate of HIV-1 infection and rate of immune cell (CD4+ T cell) depletion postinfection (15-17). A caveat against focusing narrowly upon the antibody response in the STEP trial is supported by the use of cell-mediated immunity-related parameters to define preseroconversion immune status in the aforementioned studies (15-17). Thus, risks for HIV-1 infection may be associated with risks for "a broader spectrum of immunological challenges" that are "reflected in the T cell repertoires of exposed individuals" (18).

More complexity is created when the same vaccine is tested in different geographic areas with genetically diverse populations. The STEP trial was conducted mostly in North America and the Caribbean, but a second, now abandoned, trial (PHAMBILI) was initiated in southern Africa. HIV-1-host interactions relevant to the natural history of the epidemic and vaccine responsiveness may be population specific. Indeed, CCR5 genetic determinants influencing AIDS progression rates differ in European Americans and African Americans (10). HIV-1 acquisition risk is a product of the susceptibility of uninfected persons and the communicability of HIV-1 fromthe infected person (19, 20). Communicability is dictated in part by the transmitter's viral load (19-22), which is influenced by host genetics (14, 15), and also by the genetic makeup of both sexual partners. For example, transmission is more efficient when sexual partners share similar HLA class I alleles (23). Thus, even small differences in the frequencies of disadvantageous genetic variants in different vaccine trial cohorts might have a disproportionately large effect on the likelihood that genetically "atrisk" transmitters will encounter similarly "atrisk" recipients. Clinical trial design should take into account the genetically defined individual differences in both susceptibility and transmissibility to better understand puzzling outcomes. Knowledge of population-specific host factors might also help identify what protective effects are attributable to the vaccine or to the host genotype.

Perhaps the STEP trial outcome signals that a step back is needed to seek more illumination on correlates of protection and susceptibility, rather than initiating trials of broadly similar vaccines (24). Additional studies on HIV-1 vaccine cohorts, coordinated with studies of natural infection, might yield useful information about genetic factors influencing both variable vaccine responses and variable susceptibility to infection. Do titers of antibodies to common viral and nonviral pathogens, including Ad5, correlate with anti-gp120 antibody titers and HIV-1 infection status? What host genetic factors correlate with strong and weak antibody responses? HIV-1 vaccine research must finally step away from its roots in empiricism and embrace new discoveries in immunology and host genetics (24-26).

References

1. J. Cohen, Science 318, 1048 (2007).
2. T. C. Pierson et al., Cell Host Microbe 1, 135 (2007).
3. P. B. Gilbert et al., J. Infect. Dis. 191, 666 (2005).
4. J. L. St. Sauver et al., Vaccine 19, 1363 (2001).
5. I. G. Ovsyannikova et al. Vaccine 25, 3090 (2007).
6. N. Dhiman et al., J. Infect. Dis. 195, 21 (2007).
7. H. D. Shin et al., Proc. Natl. Acad. Sci. U.S.A. 97, 14467 (2000).
8. K. W. Moore et al., Annu. Rev. Immunol. 19, 683 (2001).
9. M. Shan et al., PLoS Pathog. 3, e169 (2001).
10. E. Gonzalez et al., Proc. Natl. Acad. Sci. U.S.A. 96, 12004 (1999).
11. E. Gonzalez et al., Proc. Natl. Acad. Sci. U.S.A. 98, 5199 (2001).
12. S. J. O'Brien, G. W. Nelson, Nat. Genet. 36, 565 (2001).
13. J. Cohen, Science 307, 23 (2005).
14. E. Gonzalez et al., Science 307, 1434 (2005).
15. M. J. Dolan et al., Nat. Immunol. 12, 1324 (2007).
16. F. A. Koning et al., J. Immunol. 175, 6117 (2005).
17. L. Van Asten et al., AIDS 18, 1885 (2004).
18. M. S. Killian et al., AIDS 18, 161 (2004).
19. R. Anderson, M. Hanson, J. Infect. Dis. 191 (suppl. 1), S85 (2005).
20. S. Blower, E. J. Schwartz, J. Mills, AIDS Rev. 5, 113 (2003).
21. R. H. Gray et al., Lancet 357, 1149 (2001).
22. T. C. Quinn et al., N. Engl. J. Med. 342, 921 (2000).
23. S. K. Ahuja, G. Catano, Lancet 363, 2103 (2004).
24. R. C. Desrosiers, Nat. Med. 10, 221 (2004).
25. D. R. Burton et al., Nat. Immunol. 5, 233 (2004).
26. R. C. Gallo, Lancet 366, 1894 (2005).