One of our colleagues has posed the questions from Mitchel and others to one of the developers of the HPV vaccine, Doug Lowy, who is also a coauthor of the paper in J Clin Invest that Phil referred us to. He has responded as follows.
I'm in southern California for a few days with intermittent email access. If you share my comments with others, please disclose that I am an inventor of the vaccine, and the NIH has licensed its HPV vacciine technology to Merck (and to GlaxoSmithKline, the other pharmaceutical company that is developing the vaccine). I have not been involved in the Merck trials.
I have commented on the questions below after each of them. Here I comment on the relationship between HPV infection and cervical cancer.
As you suspect, the evidence that HPV causes cervical cancer goes far beyond an epidemiological association with the cancer. Observations with clinical samples include: cervical HPV infection precedes the development of cervical precancer and cancer, virtually all cervical cancers contain HPV DNA, and the cancers preferentially retain and express the viral genes E6 and E7, which continue to be expressed in cell lines derived from cervical cancer.
There is a wealth of experimental data that E6 and E7 contribute to the development and maintenance of the tumor. Suppression of E6 and E7 expression in cervical cancer cell lines eliminates their tumor forming activity. In the model, expression of E6 and E7 in the genital tract can result in cervical cancer. In cultured cells, E6 and E7 together can immortalize human epithelial cells, and when you add an oncogene such as Ras to the immortalized cells the cells will form a tumors in mice. E6 and E7 induce the inactivation of several cell-encoded tumor suppressor genes, most notably (but not exclusively) p53 and pRb (the retinoblastoma susceptibility protein). In cultured cells and animal models, this inactivation can be shown to be important for the biological activities of E6 and E7.
In addition, a noteworthy molecular feature of human cervical cancer is that both p53 and pRb are wild type in almost all tumors. This feature is in marked contrast most other cancers, where p53 and pRb are frequently inactivated by mutation. The simplest (and most plausible) explanation for this finding is that E6 and E7 inactivate p53 and pRb sufficiently so there is no selective advantage to mutation of p53 or pRb in the tumor. In this regard, I personally find the situation with oral cancer especially instructive. Some oral cancers appear to be caused by HPV infection (predominantly HPV16, the HPV type found most frequently in cervical cancer), while others are not (smoking and/or alcohol are other major causes of oral cancer). In the oral cancers that contain HPV DNA and express E6 and E7, p53 and pRb are almost always wild type. In the other oral cancers, p53 and pRb are frequently mutated. As noted above, these observations are all consistent with the conclusion that E6 and E7 contribute to the development and maintenance of the tumor.
The results of the vaccine clinical trials also support the conclusion that genital HPV infection causes cervical cancer. For me, the GSK results published last year are noteworthy. Without even considering the question of whether or not there was HPV infection, there were 11 cases of high grade cervical dysplasia in the placebo group, versus 3 cases of in the vaccinated group (the vaccine was a bivalent HPV16/HPV18 vaccine).
Finally, since the 1930s, infection by some animal papillomaviruses has been known to cause cancer.
I think that the evidence for "HPV causes cancer" is not "just"
epidemiological. Can you help me address the questions raised below?
From: Science for the People Discussion List
[mailto:[log in to unmask]] On Behalf Of Mitchel Cohen
Sent: Monday, February 26, 2007 9:25 AM
To: [log in to unmask]
Subject: Re: HPV is the major cause of cervical cancer
I am not saying that HPV is not correlated with cervical cancer.
Correlation and "cause" are two different things, though I appreciate
the acceptance of epidemiological findings as legitimate science,
here. (Please remember this for other discussions on HIV/AIDS, etc.)
The main questions on Gardasil are:
1) Is there sufficient evidence in the research studies Merck
performed to conclude, as Merck does, that Gardasil prevents cervical
Comment: Merck is simply repeating the conclusion of the FDA. When they approved the vaccine, the FDA stated, "Gardasil is a vaccine indicated...for the prevention of...cervical cancer." It would be unethical to carry out a phase III placebo-controlled study that used cervical cancer as an end-point, as such a study requires standard of care, and the standard of care requires regular cervical cancer screening and treatment of pre-cancers. The causal relationship between cervical precancer and cervical cancer has been accepted for decades, and the treatment of precancers is almost certainly what has changed the United States from being a high rate cervical cancer country prior the introduction of the Pap smear into a low rate cervical cancer country. More recently, persistent infection with a high-risk HPV type (i.e., an HPV type associated with cervical cancer) has been shown to be associated with a high risk of progression to precancer and cancer. As Gardasil prevents the infection and the precancer, it is valid to conclude that it will prevent cervical cancer.
2) Have there been sufficient (or ANY) studies done on the longterm
effects of the vaccine, particularly when given to young girls?
Comment: For girls under 16, the answer is no, to the best of my knowledge. Merck reported last year on 4 year follow-up data for a monovalent vaccine (HPV16 only). Protection continued to be excellent. GSK also reported last year on 4 year data for their bivalent vaccine (HPV16 and 18). Again, protection continued to be excellent. In these trials, 400-700 women who received the vaccine were analyzed, and a similar number received the placebo.
3) Are there environmental/toxic-exposure causes for HPV? For cervical
Comment: Some (but not all) studies suggest cigarette smoking may be associated with a higher risk of progression to precancer and cancer. Erosions in the genital tract may also be associated with viral persistence and progression. Immunosuppression (e.g., by HIV infection or with drugs in renal transplant recipients) is associated with viral persistence and progression.
4) Can HPV be addressed by other, more natural and less toxic ways,
including removal of any causes due to exposure to pollutants/toxins
and supplementation with L-Lysine?
Comment: First, see comments to #3 above. Second, I am not aware of relevant clinical trials. However, a recent publication looked at the effects of a combination of lysine, proline, ascorbic acid, and green tea extract on the growth of a couple of cervical cancer cell lines in culture (Roomi et al, Int J Gynecol Cancer 16:1241-7, 2006). The clinical relevance of this study is uncertain. Third, it is not clear to me that vaccination is toxic. From my perspective, the vaccine exposes vaccinees to a higher dose of the same protein, in essentially the same form, that they are exposed to during natural infection. The main difference is that, instead of being exposed to the protein through the genital tract, the vaccinees are exposed to the protein through an intramuscular injection, because it is more effective and socially more acceptable, especially for girls who are not sexually active. An adjuvant is used in the vaccine to boost the immune response to the vaccine. The adjuvant is needed primarily because it permits the use of lower levels of the viral protein, and the multivalent nature of the vaccine (i.e., it contains protein from several different HPV types) makes it useful to use less protein for each HPV type.