A Closer Look
Despite Early Skepticism, HPV Vaccines Prove Effective
Cervical cancer is the second most deadly cancer in women worldwide. Although incidence has fallen dramatically in the United States and other developed countries, where routine screening measures—most notably the Pap test—are widely available, the disease remains a major public health concern in the developing world. Fortunately, the group of viruses responsible for virtually all cases of cervical cancer has characteristics that make the viruses particularly amenable to vaccine development.
The discovery that human papillomaviruses (HPV) are responsible for cervical cancer initiation led long-time collaborators Dr. Douglas Lowy, chief of NCI’s Laboratory of Cellular Oncology (LCO), and Dr. John Schiller, head of the LCO’s Neoplastic Disease Section, to begin studies in the early 1980s to understand how these viruses infect cells. The goal of designing a vaccine to prevent HPV infection and, thus, cancer wasn’t adopted until the early 1990s.
“When we started this work,” said Dr. Schiller, “there was no greater optimism for an HPV vaccine than there was for an HIV vaccine. In fact, there was skepticism that it could work at all.”
The researchers’ persistence paid off when, in 2006, Gardasil became the first prophylactic vaccine against HPV to be approved for cervical cancer prevention in females in the United States. Approval for a second vaccine, Cervarix, followed in 2009. Women and girls who complete all three doses of either vaccine before becoming sexually active receive nearly 100 percent protection from infection by select HPV types. The duration of immunity is not known, but it has been shown to last for at least 5 years. Both vaccines protect against infection with HPV-16 and HPV-18, two high-risk, or carcinogenic, HPV types that cause approximately 70 percent of all cervical cancers. Gardasil also protects against infection with HPV-6 and HPV-11, which cause 90 percent of genital warts.
“There was a poor track record for developing vaccines against local genital infection,” commented Dr. Lowy, citing the long history of unsuccessful attempts to develop a herpes simplex virus vaccine as an example. “I think the surprise was really how well the HPV vaccine works,” Dr. Lowy continued, “and we now think several different factors account for this very high level of protection.”
A Combination of Serendipity and Good Science
Drs. Schiller and Lowy and their colleagues began their studies by examining the proteins on the surface of the HPV virus, which presumably were responsible for binding to target epithelial cells to initiate an infection.
“The best way to make antibodies that will neutralize a virus is to give the body something that looks like the real virus,” explained Dr. Schiller. “We couldn’t give killed or live attenuated HPV because it has oncogenes, so we decided to try to make something that looks like the outer shell of the virus.” The key discovery was that the major HPV surface protein, L1, alone or in combination with the minor surface protein, L2, spontaneously forms noninfectious particles that closely resemble the original virus structure and can induce the body to produce high levels of antibodies that prevent infection of cells in culture.
To begin studying the HPV infection process, the researchers incubated the L1 or L1 plus L2 virus-like particles (VLPs) with cells in culture. Both types of VLP were able to attach to the surface of the cells, suggesting that L1 was responsible for this attachment. This realization led the researchers to begin vaccine development with the L1 VLPs. Remarkably, the highly repetitive structure and the spatial arrangement of the L1 VLPs are ideal for activating the immune receptor that controls the production of antibodies, explaining the robust levels of antibodies observed following HPV vaccination.
The antibodies produced by intramuscular injection of the HPV vaccine circulate in the blood. Because HPV infections occur on the surface of the cervix, however, it was unclear how these circulating antibodies would come into contact with the virus to prevent infection. Further research into the virus’ life cycle shed light on this apparent conundrum.
Using a mouse model, Drs. Schiller and Lowy and their colleagues found that the virus must attach to the basement membrane of the cervix, a layer beneath the epithelial cells, to initiate infection. The basement membrane is exposed only after the cells above it are physically or chemically damaged. This disruption (or “microtrauma”) stimulates the body to release antibodies at the sites of trauma, where the antibodies can then come into contact with the virus.
The animal studies also illuminated the mechanism by which HPV infects cervical epithelial cells. Binding of the viral L1 protein to the basement membrane causes the proteins on the viral surface to reorganize, revealing the L2 protein. An enzyme on the basement membrane then clips off a piece of L2, exposing a previously hidden portion of the L1 protein. This L1 domain binds to a receptor on the epithelial cell surface, allowing the virus to enter the cell.
“For most viruses, the first step in the viral life cycle is the binding to the cell-surface receptor, whereas papillomaviruses apparently spend several hours on the basement membrane before they bind to the cell surface. This gives the antibodies more opportunities to bind to the virus particle and prevent infection,” Dr. Lowy explained.
One other characteristic of HPV that has lent itself well to vaccine development, Dr. Schiller explained, is its double-stranded DNA genome, which makes it difficult for the DNA to accumulate mutations. This prevents the HPV virus from evading the immune system as effectively as viruses that have rapidly evolving single-stranded RNA genomes, like HIV.
More Work to Be Done
Although the current HPV vaccines are important advances in the prevention of HPV infection and cervical cancer, there is still room for improvement.
Gardasil and Cervarix target only a few types of HPV because they were developed using the unique L1 VLPs of those types. To expand the number of HPV types targeted by Gardasil, Merck is adding L1 VLPs of the five types known to cause the most cervical cancers after HPV-16 and HPV-18 to its next-generation vaccine. Other companies are taking a new approach by developing vaccines that use the L2 protein. A portion of L2 appears to trigger the generation of antibodies that will block infection by most known HPV types, not just the type from which the L2 was derived.
There are also more practical obstacles to overcome. The need for multiple injections and the vaccines’ high cost pose significant barriers to their widespread use, especially for people in the developing world. “Right now, most of the girls and women getting vaccinated will get screened [for cervical cancer], so they are really pretty low risk; we’re mostly preventing premalignant lesions,” said Dr. Schiller. “The real impact will be if you can get this vaccine to the group of women who aren’t going to get adequately screened. That will be the real payoff for this vaccine.”