From your second article:

> There is some evidence that the bivalent and quadrivalent vaccines already provide some protection against additional strains of HPV. A recent paper in Clinical and Vaccine Immunology reporting on the clinical trial for Cervarix reported that the vaccine is more than 96 percent effective against disease associated with strains 16 and 18, and more than 50 percent effective against diseases associated with any HPV strain.

There are network effects in disease transmission. In order to catch a disease, not only do you have to be susceptible, but you have to catch it from someone. If you cut the risk of contracting HPV by 90%, then not only has the chance of a vaccinated person contracting the disease gone down by 90%, but the chance that an unvaccinated person will come in contact with a carrier goes down by 90%. Eliminate all the most prevalent strains and provide some protection against the remainder, and you can cut the virus's ability to replicate itself below the threshold needed to sustain itself in the population. Same reason that anti-vaxxers can get away without vaccinating their kids for measles as long as they don't come in contact with too many other unvaccinated people.

I assumed the tetravalent vaccine only worked against the four named strains. If it's semi-effective against all strains, then I see how eradication is possible. Thank you for the clarification.

Edit: Wait, the article does not say the vaccine is semi-effective against all strains. It says the vaccine is semi-effective against cervical cancer caused by all strains. My confusion is not removed.

I would assume that the vaccine couldn't be semi-effective against cervical cancer caused by all strains without it being effective against all strains, since your first link says 95%+ of cervical cancer is caused by HPV and there's no known mechanism for the vaccine to act directly against cervical cancer without first acting against the HPV virus.

I apologize for the ambiguity of my prior comment. To illustrate what I'm thinking: Suppose cervical cancer is caused by two strains, 1 and 2, which are equally present in the population. A vaccine works 100% against strain 1 and 0% against strain 2. That vaccine is 50% effective against all strains causing cancer, despite having 0 power to eradicate strain 2.

Therefore, it's not clear to me that being 'semi-effective against all strains' implies 'semi-effective against each strain.'

I am now looking for sources that say Gardasil is semi-effective against each strain of HPV.

Edit: Reading this source[1] I see that Gardasil protects against strains 31/33/45/51 (grouped) with 33%-51% effectiveness. Still can't find conclusive evidence that Gardasil protects against each strain.

[1] http://cvi.asm.org/content/early/2015/01/29/CVI.00591-14.ful...

I would assume that the context ("already provide some protection against additional strains of HPV") implies that it provides additional protection against additional, individual strains, but will grant that the wording doesn't conclusively say that and alternate interpretations are possible.

Agreed. From the evidence I was able to find, certainly Gardasil seems to prevent at least some closely related strains. I am wondering if it prevents all cancer-causing strains though, which would be needed for eradication. Probably, given the title of the article, but I wish I could find a primary or secondary source.

I'm not clear on the distinction that you're drawing - cervical cancer is almost always caused by HPV. What were you alluding to?

  (Cervarix) is more than 96 percent effective against disease associated with strains 16 and 18

... but those are the two strains that Cervarix is specifically designed to prevent against!

In other words, that's admitting to a 4% failure rate where not only did the HPV infections occur despite those subjects being vaccinated, but they proceeded to cause disease (aside from the HPV presence itself).

Please re-read your first link. "Most HPV infections of the cervix are cleared rapidly by the immune system and do not progress to cervical cancer (see below the Clearance subsection in Virology). Because the process of transforming normal cervical cells into cancerous ones is slow, cancer occurs in people having been infected with HPV for a long time, usually over a decade or more (persistent infection)." 1) Not every person with HPV gets cervical cancer. 2) Many people will get into a monogamous relationship, it will not spread further. 3) It is not necessarily spread in every sexual encounter.

Therefore eradication is definitely possible.

90% is still a ginormous reduction over 0%. And as with most communicable diseases, once their incidence falls below a certain threshold, they fail to spread in effective numbers to keep themselves alive.

You're misunderstanding though, it only works on certain strains of HPV, for the unaffected strains we wouldn't expect a reduction in incidence. You are thinking of Herd Immunity, which is what allows for a disease to be wiped out and protects unvaccinated/immunocompromised individuals in a population.

No, it projects a 50% reduction in disease caused by HPV without giving supporting science. The bivalent vaccine does not prevent infection with other strains (it can't; it doesn't work via recruiting the immune system like most vaccines do).

  90% is still a ginormous reduction over 0%

But those aren't the numbers. The pre-vaccine infection rate was 22.7% (not 96% nor 90%), and the vaccinated infection rate was at least 4% (probably much higher; the 4% represents the cases where the vaccine failed to protect and the subject proceeded to develop HPV-caused disease), using their figures. So, you're looking at a best-case 18-point difference, not 90 points.

Still valuable, but hardly eradication.

This is as much a marketing press release as an interim report on the efficacy of HPV vaccinations. For the points you mention as well as the 70+ lifespan ahead of vaccine recipients.

The spread of infectious disease is a dynamics problem, as is eradication.

It doesn't require 100% effective vaccines nor 100% coverage of vaccines to wipe out an infectious disease, typically it requires far less.

Break it down to the simplest terms. On average in an environment where nobody has immunity to disease an average infectious person will spread the disease to N other people, this is the "basic reproduction number" for that disease. If that number is more than 1 then in a simple model more people will get the disease over time, the disease has the ability to spread. If that number is less than 1 then fewer and fewer people will get the disease over time, leading eventually to eradication.

If you introduce immunity into the situation via vaccines then you change the whole scenario. If, an infectious person has a chance to infect some number of people but now most of those people are immune, the disease won't spread to them, it'll instead only spread to some of the people in the non-immune subset. Meaning that the effective reproduction number is reduced. And if that number is reduced below 1 due to immunization then on average the disease won't spread, it will die out. Infectious diseases rely on luck, and we can hurt their luck substantially with mass immunization programs. All that matters is pushing that effective reproduction numbers across the line from above 1 to below 1, and then ordinary population dynamics will play their part. This is how "herd immunity" works. It doesn't require 100% of people to be immune, it just requires enough immunity to flip the switch between a disease that is commonly circulating in the bulk of the population (measles, polio, HPV, etc.) to one that does not circulate and only a very small number of individuals may have at any time.