Dr. Cosette M.
From the Special Topic of
In our Special Topics analysis of HPV research over the
past decade, the work of Dr. Cosette Wheeler ranks at #7 by
total cites, with 51 qualifying papers cited a total of
3,064 times. She also has five papers on both the 10-year
and 2-year top 20 papers lists, in addition to having 10
papers featured in the
Map on HPV.
Science IndicatorsSM from
Reuters, Dr. Wheeler's work ranks in the top 1% in the
field of Clinical Medicine. Her overall citation record in
the database includes 62 papers, mostly classified in the
field of Clinical Medicine, cited a total of 3,799 times
between January 1, 1998 and April 30, 2008.
Dr. Wheeler is a Professor in the Departments of Molecular Genetics and
Microbiology and Obstetrics and Gynecology, Chief of Translational Research
for the Department of Molecular Genetics and Microbiology at the University
of New Mexico Health Sciences Center.
In the interview below,
she talks with
Gary Taubes about her HPV research.
What do you consider your most significant
contributions to HPV research?
My group has had a couple of significant contributions related to both
primary and secondary HPV prevention: one was in contributing to describing
the broad spectrum of papilloma virus types infecting the human genital
tract. There is a whole group of different viruses that infect the genital
tract—actually over 40 of them. We worked with several groups all
over the world trying to dig these out. Then we asked the question, "How
stable are these viruses, and how much variation do they have?" If I see
HPV type 16 in Asia, is that the same as the one in South America?
We described what might be called extended genomic intratypic sequence
variance. So we first found that there were over 40 of these viruses. Then
we went and looked at some of them closely, and realized there are very
fixed variants of each individual HPV type studied. We demonstrated that
co-variation across all regions of HPV genomes was common. This means that
you can often predict the expected sequence changes that will be found in
multiple regions of an HPV genome if you know the changes in just one
region. Overall this information was important to the development of
appropriate broad-spectrum HPV diagnostic tests and to considering possible
differences of variants in risk of disease.
"The vaccine is not a trivial
expenditure of money when you consider that it
does not eradicate risk or make Pap smears
We worked with colleagues at the US National Cancer Institute to
characterize an HPV assay that is now routinely used in cervical screening
programs for the clinical management of women with atypical Pap tests. Our
HPV genotype data generated through the NCI ALTS trial has also contributed
to understanding the natural history of HPV infections and their disease
I think you know that we have also been a significant contributor to the
development of HPV vaccines for more than a decade.
Could you explain what you mean by "variants" and
the "stability" of the viruses?
Well, this is not like HIV, which mutates in our time period and becomes
something else. Even though HPV types vary, they varied a long time ago. In
our era, they’re essentially fixed. Working with my colleague Uli
Bernard, who was then in Singapore and is now at UC Irvine, we contributed
information important to a molecular clock that tells you the rate at which
these viruses have been evolving—how frequently they change.
Because human papillomaviruses, to our knowledge, are not changing rapidly
and thus are apparently somewhat "fixed," we don’t think some new
monster virus is going to arise from the evolutionary pressures of HPV
vaccination. What we showed is that HPVs vary—as compared to saying
they mutate—and they did so over millennia. Nearly all animals
studied have their own species-specific papilloma viruses. So you can
actually correlate speciation events using phylogenetic information for
these animal viruses. A collaboration of our group with Long Fu Xi at the
University of Washington supported the idea that HPVs co-evolved with
individual human racial groups as well.
Does that mean these viruses evolve along with the
Yes. The idea is bigger than just what is seen in HPV infections. Common
persistent viruses, parasites, and bacteria have often been infecting their
host species for so long that they have co-evolved. Because of the
co-evolution, these viruses don’t necessarily trigger much of an
immune response. These are things we all get infected with, that can go
underground or cause virtually no symptoms or noticeable problems and, in
some people, these are the viruses that are later associated with cancer.
Epstein-Barr virus (EBV) is another example. These are super-common
infectious agents—bacteria like Helicobacter are yet another
example—and they do well in populations. They are common infections
that persist without much of an immune response, but in some people, over
time as they persist they can make the cells they are associated with go
So most people carry human papilloma virus, but it
only causes disease in some of them?
Most people who get these infections get rid of them. They might have a
low-grade abnormal Pap smear, but that’s it—the virus almost
always goes away on its own. If you look at the incidence of invasive
cervical cancer in a country that has no Pap screening program, it might be
between 200 and 300 persons per 100,000. But maybe 50,000 out of every
100,000 have been infected. So only 200 to 300 get a bad outcome.
It’s not as if the outcome is bad in everyone, or even most people.
And it’s not as if any of these viruses or bacteria have the
intention of doing something useful for themselves by causing disease.
Actually, they’re very well adapted to not cause disease.
It’s when there’s failure, probably due to complex host genetic
factors, as well as other co-factors that are not well understood that
Was the highly cited 2002 New England Journal
of Medicine vaccine the proof of principle for the HPV vaccine?
(Koutsky A, et al., "A controlled trial of a human
papillomavirus type 16 vaccine," N. Engl. J. Med. 347:
1645-51, 21 November 2002)
Yes. That work told us that this vaccine actually worked, based on this
study in about 2,000 women. Then there were the phase III trials, which, of
course, are always required and the most influential. My group at the
University of New Mexico did the first phase I trial of the Merck vaccine
with investigators at Indiana University. The phase I study was actually
published after the phase II study reported in the New England
Journal. This was because they were conducted so close in time and it
was more important to present the proof of principle results. We have been
one of the leaders bringing HPV vaccines forward beginning at phase I all
the way through phase III.
How has the vaccine itself evolved through this
It started off as a monovalent vaccine—one papilloma virus type in
the vaccine. The first trial we did in New Mexico was HPV type 11. That was
the first phase I of the Merck "product." The one in that 2002 New
England Journal of Medicine paper was the monovalent type 16. Type 11
was a model for various complicated laboratory reasons. Type 16 is the main
virus in cervical cancer—that’s one you want for a vaccine. So
type 16 was the proof of principle. In this study, there were41 cases of
HPV16 infection in the placebo group and none in the vaccine group.
Then came the quad-vaccine that had types 6, 11, 16, and 18. That was the
prototype for the Gardasil vaccine. GlaxoSmithKline was trying to bring
forth their bivalent vaccine— for HPV 16 and 18 during this same time
and we have also worked on these vaccines for the past seven years.
Do you think that all young women should be
vaccinated to prevent cervical cancer?
It depends on what is meant by "all young women." I disagree with the
recommendation to vaccinate all girls and women between the ages of 11 and
Because Pap screening works well in sexually active women. We have reached
a plateau on the impact of Pap screening. We’ve reduced rates of
cervical cancer by 70-80% of what they used to be 40 years ago. We pay
about five to six billion dollars a year to do that in the United States.
We have a residual amount of the disease that we’ve been unable to
conquer. As much as 60% of that is attributable to people who don’t
go to the doctor as recommended for Pap tests. Without getting these women
screened we are potentially not going to reduce cervical cancer rates any
further. That’s where the vaccine comes on the scene. We know these
vaccines are prophylactic and do not seem to help therapeutically. If
you’re already infected, they therefore won’t help you. As the
population ages—we’re talking populations, not
individuals—the average number of lifetime sexual partners increases.
Genital HPVs are sexually transmitted infections. By the time women are 23
or 24 years old, the average number of lifetime sex partners in a
population is about four. And that number, that kind of information, is
probably underestimated because women underreport in a society that
considers multiple lifetime sexual partners to be undesirable. The bottom
line is that the maximum benefit of current HPV vaccines is primarily
realized in presexually active adolescents.
There are other aspects of my perspective. Can you think of a single
example where the US government has said, "Let’s go out there and
spend billions of dollars to vaccinate a population that’s already
largely been exposed to any agent particularly one where most people will
get rid of it on their own?" In effect, we’re saying. "We can’t
tell exactly who has been exposed, but even if they have been exposed, they
may get some benefit if there is one or more of the HPV types in the
vaccine that they haven’t already seen." We’re not talking
about the normal, garden-variety vaccine at $10 a shot. We’re talking
about $120 a shot with three shots needed to achieve protection.
How effective is the vaccine? Won’t it at
least protect women who don’t already have the virus?
"...the ultimate question we want to
answer is what difference will all this work over
the past two decades make?"
These vaccines are highly effective in females who are naïve to the
HPV types in the vaccine. This means that they work best or are most
effective in sexually naïve individuals. The primary mode of genital
HPV transmission is sexual intercourse although there are other modes that
are much less common. These current HPV vaccines however only target a few
types of virus. They don’t allow vaccinated girls or women to not get
Pap smears. You still have to get Pap tests because 30-40% of cervical
cancers are caused by virus types that aren’t in the vaccines. So the
five to six billion we spend in the US on Pap smears doesn’t go away
at least until we really figure out what screening modifications are safe
and have achieved very high vaccine coverage.
The vaccine is not a trivial expenditure of money when you consider that it
does not eradicate risk or make Pap smears irrelevant. On the other hand,
if we had a multivalent vaccine that targeted, say, 8-10 virus types,
constituting maybe 90-95% of the human papilloma viruses that cause
cervical cancer, then that would be a good thing. In theory we could
virtually eliminate screening as we know it. Then I’d say, "OK,
vaccinate women up to age 40," because now there is greater benefit to more
women and better cost-effectiveness in terms of healthcare expenditures.
So what message would you give to women in general
when they’re thinking about getting this vaccine?
I would tell women 18 and under that the vaccine is a good and reasonable
cervical cancer prevention approach. They must still remember to get
regular Pap tests even if they receive the vaccine. For women over 18, they
should discuss their past sexual history with their providers to determine
whether HPV vaccines are likely to be beneficial. All women vaccinated or
not should get Pap tests and continue with cervical screening until we
have, hopefully, a better vaccine and new recommendations. Better vaccines
are certainly on the horizon. Here’s my philosophy: I’m hopeful
that the next generation vaccine will be realized and, in the meantime,
with current first-generation HPV vaccines, we can invest in implementing
the clinical platforms that are necessary to deliver to those who will
The thing is that after kids are five years old and they get their last
vaccine and go off to school, they basically go to the doctor only as
required for sports physicals or acute illnesses. There’s no routine
health care for adolescents, and we’re going to need some kind of
adolescent vaccine platform that really works if we’re going to get
the vaccine to people who are young enough to really benefit.
What are you focusing on in your current
Two things. We are evaluating a large population sample of all the
circulating HPVs in the state of New Mexico by sampling residual liquid Pap
smear materials. Then we have established an information system that is
monitoring all Pap smears, all HPV tests, and all cervical pathology
collected on New Mexico residents. After spending a career hoping to make a
difference in the clinical management and prevention of HPV related
disease, I want to see if the improvements that have been developed will
really make a significant difference.
Don’t we know that it makes a
The truth is that we have no idea what really happens in the US with
screening. That’s not the same thing as knowing that screening works.
The questions are who gets screened, and how often? Never before have there
been US population-based surveillance programs for Pap smears. There are
cancer registries. We know a lot about cancer. We know virtually little
about population-based cervical screening. Then I want to see what HPV
vaccines actually reduce including abnormal Pap tests which should be
reduced long before we ever see any reductions in cervical cancers. To do
that we need a Pap smear registry that can tell you what was happening
before the vaccine coverage was high and then what it looked like
afterward; what’s changed? We will want to ultimately link the Pap
smear information with the New Mexico Statewide Immunization Information
So one component of doing molecular epidemiology is conducting
population-based surveillance. I want to know what diseases are being
diagnosed and what the behavior of the population is. Do women really get
Pap smears every year? Every two years or every three years, etc.?
What’s the average screening interval and how does this differ by
age, race, and ethnicity? Right now, no one really knows, and much of what
we believe happens is based on telephone surveys of a few women and
obviously only women with telephones. This is the first population-based
data that will be able to tell us precisely what is occurring.
And because of our other project, we’ll be able to say what the
prevalence is of the specific HPV types circulating in the population. So
if the vaccine works in the population, not just in a clinical trial, which
I think it will, what will be the impact? Will it be what we expected? And
will there be other problems that we didn’t anticipate? Will women
get Pap tests as recommended who are vaccinated? Maybe HPV types that
aren’t in the vaccine will increase in their prevalence? We
don’t expect they will, but as with many interventions, we
can’t really predict everything when we are in fact dealing with
individual people and Mother Nature.
So that’s what I’m currently doing, evaluating our state
population. My lab now collects about 17,000 or 18,000 specimens per month
and we now have a database of about 1.5 million records. This is one of the
largest projects like this ever done. And the ultimate question we want to
answer is what difference will all this work over the past two decades
make? I also wanted to develop and contribute something that I could pass
on when I retire, so my work wouldn’t come to an end just because I
was no longer actively involved. What we are doing now is for the long-term
good of the public. We will be able to tell people including policymakers
what this vaccine does and does not do. And as we make changes in
screening, we can see if they’re really good or bad. We expect that
over time HPV testing will possibly replace Pap tests in primary screening
and we’ll have established an ability to collect data in real
populations beyond clinical trials. We won’t just be relying on
theory or phone surveys. I’m very excited about this.
When do you expect to begin having meaningful
We just started to see the tip of something meaningful with our data from
2006 through the end of 2007. That’s our first full two-year
interval. We could see how many women got a Pap smear in the population
during that time; how many got one in 2006 and then came back again one
year later. By the end of 2008, we’ll have three full years of data
and that will begin to become really useful. In particular we hope to
improve our abilities to identify disparities in the delivery of primary
and secondary cervical cancer prevention strategies.
Cosette Wheeler, Ph.D.
Departments of Molecular Genetics and Microbiology (MGM) and Obstetrics and
Chief, MGM Translational Research
University of New Mexico Health Sciences Center
Albuquerque, NM, USA
Dr. Cosette M.
Wheeler's most-cited paper with
721 cites to date:
Koutsky LA, et al., "A controlled trial of a human
papillomavirus type 16 vaccine," N. Engl. J. Med.
347(21): 1645-51, November 21, 2002. Source:
Essential Science Indicators from
Keywords: HPV, papilloma virus types, sequence
variance, molecular clock, viral evolution, vaccine, pap screening,
cervical cancer, vaccine cost, population-based screening.