According to our analysis of HPV research over the past
decade, the work of Dr. Keerti Shah ranks at #3 by total
citations, based on 55 papers cited a total of 4,752 times.
Four of these papers also appear on our list of the 20
most-cited papers in this topic over the past decade.
Science IndicatorsSM from
Reuters, Dr. Shah's record includes 83 papers cited a
total of 5,050 times between January 1, 1998 and February
29, 2008. The majority of these papers are classified in
the field of Clinical Medicine.
Dr. Shah is a Professor in the Department of Molecular Microbiology and
Immunology at the Johns Hopkins Bloomberg School of Public Health in
Baltimore. He also holds a joint affiliation with their Department of
In the interview below,
he talks with ScienceWatch.com correspondent Gary
Taubes about his highly cited
How did you get started studying human
I’m a virologist. I’ve always been interested in causal
connections between viruses and diseases. I’ve been working on
papillomaviruses since the late 1970s, soon after it was first proposed
that they might be involved with cervical cancer.
How did the link with cervical cancer evolve, and
what prompted your group to study it?
The groups working on papillomavirus used to meet every year. These were
fairly small meetings, maybe a few dozen people in the early years.
We’d discuss what was going on. And there happened to be at these
meetings two epidemiologists who were very interested in this—both of
them from the International Agency for Research on Cancer (IARC) in Lyon,
Dr. Nubia Muñoz and Dr. Xavier Bosch. They had worked with many
human cancers and they’d been studying the epidemiology of cervical
cancer around the world. They would come to these meetings looking to find
virology support for what they were doing.
"...with cancer of the
uterine cervix, HPV is responsible for nearly 100% of
the cancers, and this is true for all parts of the
They were already collecting material from their first studies in Colombia
and Spain, and I got to know them and offered to work with them. It was
good for me, and it turned out to be a long and happy collaboration. And
the methodologies were all very straightforward. They had invasive and
in situ cervical cancer cases and controls consisting of women
without cancer. They also collected material from the sexual partners of
the controls and the cases. It was a very elaborate study, and we did all
the virology for it on blinded specimens. While we were working on this
study—this was now the late 1980s—PCR methodology had just come
into the field, so we also tested specimens using PCR.
When did you publish your first findings?
The first article was published in 1992 where we described the correlation
in the results with the different assays (Guerrero E, et al.,
"Comparison of ViraPap, Southern hybridization and polymerase chain
reaction methods for human papillomavirus identification in an
epidemiological investigation of cervical cancer," J. Clin.
Microbiol. 30: 2951-9, 1992).
In the same year, with Dr. Muñoz as the first author, we proposed a
causal relationship between this virus and cervical cancer (Muñoz N,
et al., "The causal link between HPV and invasive cervical
cancer—a population-based case-control study in Colombia and Spain,"
Int. J. Cancer 52:743-9, 11 November 1992). That sparked a lot
of interest in the subject—many more laboratories followed this up.
In a subsequent IARC study, in which a lot of laboratories participated,
including us, nearly 1,000 cervical cancers from all over world were
What did you consider the most surprising or
remarkable aspect of the research?
Well, with cancer of the uterine cervix, HPV is responsible for nearly 100%
of the cancers, and this is true for all parts of the world. That’s
very surprising, very unusual. It’s the only example of a major human
cancer that has a single etiology. And that means whatever we learned in
the US, for example, could be applied in Zambia, what was learned in India
could be applied in the US, and so on. It was a very straightforward
infectious disease model.
Was this the work published in your highly cited
1999 Journal of Pathology paper (Walboomers JMM, et
al., "Human papillomavirus is a necessary cause of invasive
cervical cancer worldwide," J. Pathol. 189: 12-19,
Yes. That is the study with nearly 1,000 cancers from more than 20
different countries from around the world. The first tests on these cancer
specimens were performed in our lab and we found that 93% of all cervical
cancers had HPV. (Bosch FX, et al., "Prevalence of human
papillomavirus in cervical cancer: A worldwide perspective," J. Natl.
Cancer Inst. 87: 796-802, 1995).
In the ’99 paper, Dr. Walboomers's laboratory in the Netherlands
retested the tissues that were found to be HPV-negative in our lab. He did
many additional tests including type-specific PCRs and re-checked the
histology of the specimens. Dr. Walboomers was able to identify HPVs in
most of the tissues which were HPV-negative in the initial assay. It was
estimated that 99.7% of the cancers were due to HPV. Maybe there were two
samples which remained HPV-negative, but the results were conclusive enough
that we could say that HPV is a necessary cause of cervical cancer. That
was the significance of that paper.
And that’s why that particular paper has
been so highly cited?
Yes. It was more or less the final word. Lots of laboratories were coming
up with similar results around the world. But this paper was from the IARC,
which gave it a lot of credibility. And it was easy to see that every
attempt was made to do this work as rigorously as possible.
Are you surprised by how influential this paper
Not really. The 1992 paper on Colombia and Spain that Dr. Muñoz
wrote is also very highly cited. The first time we discussed these same
1,000 specimens, the 1995 JNCI paper with Dr. Bosch as the first
author (referred to above) was also very highly cited.
I think after a while, people say, "Okay, so all cervical cancers are
caused by HPV and we want to cite one paper," and this particular 1999
paper showed them all to be due to HPV—essentially 100%. It also had
so many participants—IARC investigators, our lab, NIH scientists, and
Dr. Walboomers's lab in the Netherlands. Everyone was on it. So if you
decide you’re going to cite any one paper, this would be the one you
If you were writing that paper today, is there
anything you would do or say differently?
No. That was a very nice paper, and, again, it showed what people have
showed again and again since. That conclusion is quite clear. You can
perhaps never say that 100% of cervical cancers are caused by HPV, but
you’d have a hard time finding one that isn’t. And you must
remember that there are over 100 human papillomaviruses. About 12 account
for 99% of cancers. So if you find an HPV-negative cervical cancer, you
cannot be completely sure that you didn’t just miss an HPV type that
was not tested for, or one that we’ve yet to identify.
What was the most challenging aspect of this
Actually, in many ways the research turned out to be very straightforward.
It has been suspected for over a hundred years that cervical cancer may
have something to do with a sexually transmitted infection because women
with cervical cancer often had more sexual partners than other women. In
the older literature, in the 1960s, it was thought herpes simplex virus,
type 2, might be causing cervical cancer. Various other sexually
transmitted infections were linked to cervical cancer at various times.
"With HPVs everything just
fit together, from the epidemiology to the virology to
the molecular mechanisms."
When we look at HPVs, they have two oncogenes, E6 and E7, which disrupt
cell cycle control; they were just perfect for producing cancers. With HPVs
everything just fit together, from the epidemiology to the virology to the
molecular mechanisms. All lines of enquiry led to the same conclusions. It
turned out to be a very simple story: that all cervical cancers were due to
HPV; the viruses were very stable; the same viruses were everywhere; and
HPV-16 was the predominant oncogenic virus.
Was there any element of serendipity in the
Not serendipity exactly, but I guess you could say that we were lucky it
was such a simple infectious disease model. Those who work with other
cancers certainly haven’t been so lucky. And the luck continued with
the vaccine. It wasn’t hard to make. It works perfectly and it
provides excellent protection. So within a span of 30 years, much was
accomplished. HPV etiology suspected, then proven, and then absolutely
established, then the vaccine prepared and shown to be completely
effective. Investigators from all continents and from many countries worked
on this problem with spectacular results.
How has the field changed since your 1999
Well, you can ask what else these viruses do. We were involved in research
on cancer of the vulva and it turns out, in that case, that unlike cervical
cancers, vulvar cancers are etiologically heterogeneous. One subset is due
to HPV, and these occur in younger women. The other subset occurs in older
women and is not due to HPV. . And the most exciting thing in recent years
is that probably a majority of tonsillar cancers are due to HPV. Annually,
there are about half a million new cases of squamous cell carcinomas of the
head and neck in the world. Most of them are related to alcohol and tobacco
use. But a subset of the cancers in the tonsils and the base of the tongue,
about 25,000-30,000 cases are due to HPV. The evidence for this is as
conclusive as for cervical cancer. So the range of cancers caused by HPV
began with cancer of the cervix, but now includes subsets of cancers of the
vulva, vagina, anal canal, penis, and tonsils/oropharynx.
In addition, genital warts and respiratory papillomas, which are not
malignant tumors and occur in both men and women, are also due to HPVs but
not the same HPV types that produce the cancers, but rather non-oncogenic
HPVs. So HPV disease is not confined only to cancers, but also involves
nonmalignant disease, and is not confined only to women, but is also found
There have also been important questions about the vaccine, particularly,
how to get it to the nations that need it the most. Cervical cancers are
most prevalent in Asia, Latin America, and Africa, and right now the
vaccine is too expensive for use there. A way has to be found to make the
vaccine affordable to all countries.
Which of your professional achievements brings you
the most satisfaction?
Well, the cervical cancer work, of course, and what we found about cancers
at these other sites—we participated in studies that showed the
etiologic heterogeneity of vulvar and head and neck cancers. We also found
that children with respiratory papillomas acquire their HPV infection from
the genital tract of their mothers, at the time of birth.
The one thing that is not related to HPVs is our work on simian virus 40
(SV 40). SV 40 is a virus of rhesus macaques which is oncogenic in
experimental animals. SV 40 contaminated the Salk polio vaccine, long ago,
back in the 1950’s. About 100 million people in the US had taken the
vaccine during the time when part of it was contaminated, so there was a
suspicion that SV 40 in the contaminated vaccine may cause human cancer.
In the 1990s, a number of laboratories reported that SV 40 may have
contributed to human cancers—to mesotheliomas, brain tumors, and some
lymphoid tumors. We participated in many studies to investigate these
claims. We found that it was a false alarm and that the cancers were not
related to SV 40 infection. This may have reduced the anxiety related to
possible adverse effects of the poliovaccines.
Keerti Shah, DrPH
Department of Molecular Microbiology and Immunology
Johns Hopkins Bloomberg School of Public Health
Baltimore, MD, USA
Shah's most-cited paper with
1,468 cites to date:
Walboomers JMM, et al., "Human papillomavirus is a
necessary cause of invasive cervical cancer worldwide,"
J. Pathol. 189(1): 12-19, September 1999. 1,468
Essential Science Indicators from