Does Gender Matter for HIV Vaccines?

New Approaches to an Open Question

By Emily Bass

Do vaccines work differently in men and women? Over the past few years, this question has been transformed from a far-flung supposition to a serious query for HIV vaccine researchers, even cropping up in the recent analyses of VaxGen data (see Keystone article).

This type of discussion marks a fairly radical shift. Globally, millions of men, women, boys and girls receive immunizations each year, and there is little evidence of gender-specific effects in any of these products.

The first hint of the new paradigm came in September 2000, with a brief but startling announcement from GlaxoSmithKline (GSK): Data from the company’s two Phase III trials appeared to show that a candidate vaccine against herpes simplex virus-2 (HSV-2) was about 74% efficacious in preventing HSV-2 disease in women who did not have HSV-1, a related virus that causes cold sores and confers some natural protection from HSV-2. In contrast, the vaccine showed no significant protection in men. However, the trials were not statistically powered to measure efficacy separately in men and women, so the results—while striking—were not definitive evidence of a gender gap (IAVI Report Jul-Sep 2001).

Since that announcement, the vaccine research field has moved from skepticism about the influence of gender on vaccines to more active investigation, particularly in the context of vaccines against sexually transmitted infections (STIs).

In November 2002, GSK and the US National Institutes of Health (NIH) launched a new 16-center Phase III study of the HSV-2 vaccine, which aims to enroll 7,550 female volunteers—enough to determine efficacy in women only. And if the earlier findings are confirmed, then the world will have the first gender-specific vaccine on its hands, along with a host of questions about the potential for similar effects in other vaccines.

“The herpes trial is at the back of everyone’s minds” in HIV vaccine development, says IAVI’s scientific director, Wayne Koff. “If it turns out that [the findings] are real, it’s going to open a whole new field.”

In fact, the field already exists. November 2002 saw the launch of a second women-only Phase III trial of an STI vaccine, in this case against four sexually-transmitted strains of human papillomavirus (HPV). HPV is a family of more than 100 viral strains, a few of which are linked to genital warts, cervical and anal cancer. Since cervical cancer is the most common malignancy associated with HPV infection, Merck, which is sponsoring these studies, has so far conducted its trials almost exclusively in women. (A small number of men were included in early Phase I safety trials.)

To date, Merck has conducted two proof-of-principle studies of separate HPV vaccines in women—one against the two main strains that cause genital warts, and another against two strains linked to cervical cancer. Both candidates appeared to show strong protective efficacy in women, as measured by the absence of HPV DNA matching the vaccine strain in cervical specimens, or of cervical intraepithelial neoplasia (N Engl J Med347:1645;2002; see IAVI Report, Jul-Sep 2002). The ongoing women-only Phase III trial, which is taking place at sites in North and South America, Southeast Asia, Africa, Europe and the Middle East, tests a combination vaccine against all four strains. This study will follow subjects for at least two years--longer than previous studies--to allow researchers to monitor the key endpoint: Whether the vaccines reduce the risk of cervical cancer, which may develop many years after HPV infection.

Will the same candidate vaccine also protect men from infection with the HPV strains that cause genital warts and anal cancer? To answer this question, Merck may conduct separate trials in men, says Eliav Barr, a lead investigator on the Merck team. “We will not rely on efficacy data in women to make statements [about efficacy] in men.”

Barr’s dogma-defying statement is characteristic of the new—though by no means universal—perspective on STI vaccine research. It’s an approach that is informed by an ever-expanding knowledge of mucosal immunology and hormonal influences on health and disease, fields which provide potential explanations for observed gender differences in susceptibility and prognosis for many STIs. In the AIDS field, researchers are paying attention to differences in rates of transmission from women to men and men to women, and gender gaps in viral set point and viral load following infection. “People are starting to say, ÔFollow the women,’” says IAVI’s Koff.

Pieces of the Gender Puzzle

Beyond the HSV-2 vaccine studies, the only other data bearing on adult gender differences and vaccines address side effects and levels of immune responses, not protection itself. A primary source of these data is the US Army. At a November 2002 meeting* on gender and HIV/AIDS, Philip Pittman (US Army Medical Research Institute of Infectious Diseases, Fort Detrick, Maryland), reviewed these studies, starting with the most recent findings on anthrax vaccines. During Operation Desert Storm, there were numerous reports of severe local reactions in women military personnel. At first, Pittman said, it was assumed that women were “just complaining” more than men.

However, a post-9/11 prospective study of different immunization routes and dose schedules of anthrax vaccine confirmed the initial reports: In women, the traditional, subcutaneous regimen led to more frequent, severe and long-lasting local reactions (including lesions and subcutaneous nodules) than in men. Pittman also reported that women in this group had significantly higher antibody titers than their male counterparts. Moving back in time, Pittman surveyed studies from the mid-90s which showed gender-specific differences in antibody responses to vaccines against Venezuelan Equine Encephalitis virus, yellow fever and botulin toxin. In several of these instances, women had lower antibody titers than men, in contrast to the data from the anthrax study.

VaxGen has added another piece of data to the puzzle: Compared to men, women had higher titers of antibodies against gp120 (in ELISA tests and in neutralization of the HIV-B(MN) strain, one of two strains upon which the vaccine was based). This finding is the latest reports of differences in antibody titers in men and women (or boys and girls). Similar results have been reported in response to other vaccines, including measles and hepatitis B. And mouse and human studies have shown that women have higher levels of serum immunoglobulins than men when exposed to the same pathogen, suggesting an increased propensity for antibody production in women—which could translate into higher vaccine-induced antibody titers. There are also well-documented gender differences in the risk for certain autoimmune diseases. However, none of these gaps has ever been directly linked to a male-female split in vaccine-induced protection, making the data little more than an interesting footnote in vaccine research.

Which Matters Most, Bugs or Bodies?

The GSK findings have moved gender from footnote to foreground, and raised the question of where the differences come from. At this point, answers are purely speculative. David Bernstein (Cincinnati Children’s Hospital, Ohio), a herpes researcher and investigator on the GSK trials, wryly sums up the two prevailing hypotheses. “One is that women are the stronger sex, and make more T-cells or more antibodies, or more of whatever is protective,” he says. “The second is that men and women respond to vaccines equally well, but that the infection begins in different ways.”

Bernstein and Moncef Slaoui, a senior vice-president at GSK and head of the HSV-2 vaccine development effort, favor the latter explanation. In this model, men and women have essentially the same amount and type of immune responses in their blood, but these defenses do not translate into equal protection at the local site of infection. This could be due to gender differences in how well vaccine-induced responses penetrate the genital tract, or to enhancement of vaccine effects by naturally-occurring defenses. Women’s genital tracts are bathed in antibodies, including IgG and IgA, mucins and other innate host factors such as defensins. In contrast, men’s genital tracts are protected by a keratinized outer layer of skin with far fewer immune defenses at the surface. A viral infection that starts at a break in the penile epithelium is therefore less likely to encounter protective defenses than one occurring at the surface of an intact vaginal membrane.

The basic immunologic data that GSK collected in its first two studies have so far not revealed any gender differences. As reported in the New England Journal of Medicine (N Engl J Med 347:1652;2002), there were no statistically significant differences between immunized men and women in blood levels of binding and neutralization antibodies to HSV-2 and glycoprotein-D (a component of the adjuvant). GSK also looked at cell-mediated immune responses in a small subset of vaccinees. Slaoui says that these data, which were not included in the NEJM article, showed no evidence of gender differences.

To get a better picture of what immune responses are protective in women, the HSV-2 vaccine trial team is planning substudies, scheduled to start in late 2003, that will take place at Phase III sites. These studies will look at possible correlates of protection, including mucosal CD4 and CD8 T-cell responses and HSV-2 antibodies in the blood and genital tracts. They will also measure HSV-2 shedding in the genital tract.

Even if these substudies yield insight into how and why women are protected, the broader implications of HSV-2 protection in women only—assuming that this result is confirmed in the Phase III trial—still remain to be seen. “It is hard to really know whether the data are generalizable or whether they are specific to the antigen and disease in question,” says HVTN head Larry Corey (University of Washington, Seattle), who heads the US HIV Vaccine Trials Network and has also studied herpes vaccines.

The Glaxo team agrees that the finding might be related to unique characteristics of HSV-2. Except for episodic outbreaks in the genital mucosa, HSV-2 is contained in the neural tissue, a relatively immune-free zone. This means that the burden of protection and immune control falls almost entirely on mucosal immune defenses, where there are significant differences between men and women. Following this line of argument, vaccines against STIs that remain confined primarily to the genital tract or sequestered compartments—i.e., HSV-2 and HPV—might be more likely to show gender-specific differences than vaccines targeting STIs such as HIV or Hepatitis B, which spread to the blood, where immune defenses in men and women are more similar. Indeed, the hepatitis B vaccine—the only licensed vaccine against a sexually transmitted disease—appears to protect men and women equally well.

But neither HBV nor HSV-2 exactly mirror HIV infection, where there is already evidence of gender-specificity from the point of infection onwards. Julie Overbaugh (University of Washington, Seattle) has found that women are initially infected with a greater number of HIV variants than men (see article, page 7), and that this diversity is linked to more rapid disease progression. Other studies have shown that HIV-infected women also have lower viral loads and higher T-cell counts than matched male counterparts, a finding which is already affecting thinking about HIV vaccine trial design. “Gender is an important issue in evaluating T-cell-based vaccines,” says Corey. “[HIV-positive] women have lower viral loads than men, so using viral load as a surrogate [for vaccine efficacy] requires stratification” by gender in the analysis of results.

With its women-only efficacy trial now underway, Merck is also starting to lay the groundwork for baseline studies of HPV infection in men. Before the company decides whether or not to test its vaccine in men, it will analyze the types of cells and tissues the virus infects, the natural history of disease, dynamics of viral clearance, and outbreaks of warts in heterosexual men, and men who have sex with men. Besides their usefulness for vaccine studies, these data (which already exist for women) will also help guide recommendations for use, marketing and public health messages around HPV immunization. Even if HPV turns out not to be a health problem for heterosexual men, says Barr, “there is the notion of immunizing them to protect the women who will be their partners.”

Practical Implications for the HIV Field

Several general lessons for HIV vaccine trials can be drawn from these studies. One is the importance of gender-related enrollment targets. While it will rarely be feasible to conduct trials with sufficient power to measure efficacy separately in each gender, it should be possible to enroll enough volunteers of each gender to detect trends towards gender-specific effects, which can then be further investigated in single-sex studies like the current HSV-2 trial. The recent Phase III VaxGen study enrolled very few women, and so was unable to either assess protection in women, or detect a trend towards protection in one gender (see VaxGen results article)

Another critical lesson is that it is feasible to enroll young women in trials of vaccines against sexually-transmitted diseases. The lower limit of age in Merck’s Phase III trial is 16, and both Merck and GSK have already begun additional bridging studies to show safety and immunogenicity in younger adolescents and children; both companies also report that many parents are willing to enroll their daughters in these trials. In the case of the HPV vaccine, the fact that the vaccine is designed to prevent cancer—which has a very different stigma from STIs—may have also played a role in parents’ openness to adolescent enrollment.

These studies may also lead to new regulatory precedents. If its candidate proves efficacious, GSK will seek a female-only indication for its HSV-2 vaccine. If efficacy is shown in both genders, then Merck plans to license its vaccine for men and women. However, there will be a lag between trials in women and men. This is also true for microbicides, which are being tested largely in women, although they could also be used by men who have sex with men.

Ultimately, AIDS vaccines will have to advance to Phase III efficacy trials to determine whether and how gender influences vaccine-induced protection. GSK’s Slaoui speculates that the gender differences seen in the HSV-2 vaccine trials will emerge again in the confirmatory Phase III study, and in the context of other sexually-transmitted diseases. “My gut feeling is that it’s going to be a general effect with significant implications for vaccine development for diseases such as HPV—but more importantly, for HIV.”

Gender and Vaccines: A Bibliography


Monath, T. et al. Comparative safety and immunogenicty of two yellow fever 17D vaccines (ARILVAX and YF-VAX) in a phase III, multicenter, double-blind clinical trial. Am J Trop Med Hyg HYPERLINK "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=12201587&dopt=Abstract" 66:533;2002

Vaccines: live attenuated yellow fever vaccines from two different manufacturers
Key findings: Male gender, Caucasian race and smoking associated with higher antibody titers

Pittman, Phillip. Aluminum-containing vaccine associated adverse events: role of route of administration and gender. Vaccine HYPERLINK "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=12184365&dopt=Abstract" 20:S48;2002

Vaccine: Anthrax vaccine adsorbed (containing aluminum hydroxide) Key Findings:The frequency and severity of local reactions was statistically higher in women than in men following first subcutaneous immunization. Women receiving SQ immunization also had higher antibody titers than men receiving the same course (unpublished data presented at Sex and Gender Issues in HIV, Washington DC, November 2002, sponsored by the Forum for Collaborative Research. Available at:www.hivforum.org/publications/PhillipPittman.pdf

Pittman, Phillip "Effect of Gender on Antibody Response and Adverse Events Associated with Immunization," Sex and Gender Issues in HIV, Washington DC, November 2002, sponsored by the Forum for Collaborative HIV Research. Available at: www.hivforum.org/publications/PhillipPittman.pdf

Vaccine: live attenuated yellow fever vaccine
Key findings: Women statistically less likely to respond than men (15% versus 9% nonresponders).

Pittman, et al. Long-term duration of detectable neutralizing antibodies after administration of VEE vaccine and following booster with inactivated VEE vaccineVaccine
HYPERLINK "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=8744562&dopt=Abstract" 14:337;1996

Vaccines: live attenuated Venezuelan Equine Encephalitis (VEE) vaccine with whole-killed boost (formalin-inactivated virus, in adjuvant C-84).
Key findings: Age and gender influence level of antibody responses. Males were 2.2 times more likely to respond to live-attenuated VEE (defined as antibody titers above a specific threshold) than matched female counterparts. 18-39 year olds were 2.1 times more likely to respond than volunteers over age 40.


Holt, EA et al. Differential mortality by measles vaccine titer and sex J Infect Dis
HYPERLINK "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=8228340&dopt=Abstract" 168:1087;1993

Vaccines: live attenuated measles vaccines at10-fold (medium titer) and 100-fold greater titers than standard measles vaccine
Key findings: Increased mortality was associated with high-titer vaccine for girls but not for boys. Similar mortality patterns have been noted in two other populations.

Stanberry, L et alGlycoprotein-D-adjuvant vaccine to prevent genital herpes
N Engl J Med HYPERLINK "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=12444179&dopt=Abstract" 347:1652;2002

Vaccine: herpes simplex virus type 2 (HSV-2) glycoprotein-D-subunit vaccine with alum and 3-O-deacylated-monophosphoryl lipid A
Key findings: Vaccine showed 73-74% efficacy in preventing disease (38-42% in completely preventing infection); 0% efficacy in men