Dr. Xavier Bosch
From the Special Topic of
Human papillomavirus (HPV) is a virus that infects
the deepest layer of the skin and mucosa, causing a spectrum of
conditions from benign warts in the skin, the genital tract, or
the oral cavity, to invasive cancer in the same locations.
There are about 100 types of HPV, of which around 40 infect the
genital area. Of these, at least 13 are known to cause cervical
cancer, and two of these (types 16 and 18) cause over 70% of
all cases. The realization that a virus is primarily the cause
of cervical cancer has transformed not only the understanding
of the disease but also its management.
A Special Topics analysis of papers on HPV spotlighted the
research achievements of Xavier Bosch, whose specialty is improving the
understanding of the role of HPV in the etiology of cervical cancer. In
this analysis, Dr. Bosch's work ranks at #4 by total cites, and includes 41
qualifying papers cited a total of 4,155 times. In
Clarivate Analytics, Dr.
Bosch's record includes 108 papers, the majority of which are classified in
the field of Clinical Medicine, cited a total of 6,290 times between
January 1, 1998 and April 30, 2008. Dr Bosch is the
Head of the Cancer Epidemiology Research Program and Chief of
International Affairs at the Catalan Institute of Oncology (ICO)
in Barcelona, Spain.
In the interview below,
ScienceWatch.com correspondent Dr. Simon Mitton
talks with Dr. Bosch about his highly cited HPV
How did you become an
I went to medical school first, and my specialization was in internal
medicine and clinical oncology. At the time, oncology was a very difficult
discipline in Spain; the number of jobs was rather limited. In fact, the
very concept of oncology was just beginning. In the hospital, cancers were
treated by a number of different specialists and only came to the oncology
ward when the cancer was at such an advanced stage we could do very little
by way of treatment.
Being interested in research, I wanted to learn about public health and
epidemiology. To that end I went to the US for formal training in cancer
epidemiology at UCLA, where I had the opportunity to view oncology from the
preventive point of view as well as acquiring some skills in the
methodology of epidemiology investigations.
What developments influenced your interest in HPV
as a risk factor in cervical cancer?
"Any preventative strategy that uses
HPV as the guiding point (such as HPV screening
or vaccination) should target all cases of
I received an offer to join the International Agency for Research on Cancer
(IARC), where we launched the cervical cancer program. We were a rather
small group—just two people—Dr. Nubia Muñoz, who was the
unit chief at the time, and myself. A firmly established research structure
and an impressive network of collaborators worldwide aided us. The network
remains active in conducting research on HPV and cancers other than those
of the cervix.
We put together a study to investigate the sexual behavior of males and
females as a critical risk factor for cervical cancer. Just as we got
started, a research group in Heidelberg developed new technology for
isolating and characterizing HPV DNA. Professors Zur Hausen and Lutz
Gissmann noted that HPV DNA might be involved in the etiology of these
We then merged our epidemiological studies with a protocol for taking
specimens and for trying novel biomarkers for HPV in cervical scrapes. This
was one of the earliest molecular epidemiology studies done at IARC. The
working hypothesis stated that there was some association between HPV DNA
and cervical cancer. Subsequently, after examining and testing several tens
of thousands of individuals worldwide we were able to prove that the
association was indeed strong, and that it was universal. We found it in
all countries surveyed.
How prevalent is cervical cancer in developing
Invasive cervical cancer in the majority of countries in the world is still
the #1 or #2 cancer in women. Where there is screening activity—the
Pap test—the fatalities have dropped significantly, and in these
populations it’s typically the sixth or seventh most common cancer in
The number of cases worldwide is close to half a million, of which more
than 80% occur in developing countries, where the lack of access to medical
facilities inevitably leads to high mortality in a relatively young age
group. The HPV infections that are at the origin of these cancers occur
early after the onset of sexual activity, and the first serious cases are
already seen in young women in their 20s and 30s.
So, should screening be introduced in developing
That has been repeatedly shown to be very difficult and highly improbable,
I’m afraid to say. Cervical cancer prevention has been largely a
failure in developing countries, because hundreds of millions of women must
be regularly screened at vast expense. The conventional method for
screening, the Pap smear, is relatively insensitive and has to be repeated
rather often. Screening requires adequate diagnosis, treatment, and follow
up of women, which is only a reality in well-developed societies or in
minority social groups in developing countries. That scenario clearly opens
the door for the introduction of new vaccines. HPV vaccines are expected in
the long run to make a significant breakthrough in cancer prevention in
Next, I’d like to discuss the impact of your
most-cited paper for the time span 1998-2008,
"Human papillomavirus is a necessary cause of
invasive cervical cancer worldwide," (Walboomers JMM, et al.,
Journal of Pathology 189:12-19, 1999). It really
establishes the relationship between HPV and cervical cancer
The precedent for my #1 paper in the analysis is, however, our 1995
publication in the Journal of the National Cancer Institute, in
which I was the lead author, "Prevalence of human papillomavirus in
cervical-cancer—a worldwide perspective," (87: 796-802, 7 June
1995). That paper is one of the most quoted in the field. It is a
prevalence survey describing the HPV DNA profile of over 1,000 specimens
from 33 countries. The breakthrough in that paper was to set the prevalence
of HPV DNA in cervical cancer at the 93% level and to show the consistency
"The number of [HPV] cases worldwide
is close to half a million, of which more than
80% occur in developing countries, where the lack
of access to medical facilities inevitably leads
to high mortality in a relatively young age
However, from the results we were still left with 7% of cases that were
apparently unrelated to HPV. On a second step, we identified those 7%,
retrieved the specimens, and sent them to the late Jan Walboomers, who at
the time was the head of the HPV research laboratories at the Free
University in Amsterdam. He and his colleagues had developed a slightly
different technology for detecting and typing HPV DNA. In the 7% apparently
HPV-negative samples where he could retrieve cancer cells amenable for
proper HPV testing, he found HPV DNA in virtually all of them.
So Paper #1, in which Jan is the leading author, is highly cited because it
expanded and completed the 1995 results to show that virtually 100% of
cervical cancer had HPV DNA—the necessary cause. This paper is
important because it was the first time in cancer epidemiology in which a
necessary cause was declared.
What are the implications of this universal
In the absence of exposure to HPV there won’t be any disease. Any
preventative strategy that uses HPV as the guiding point (such as HPV
screening or vaccination) should target all cases of cervical cancer. That
paper contributed to set the agenda for novel concepts and options of
cancer prevention. The impressive follow up was that technologies were
rapidly developed and tested. In less than a decade, we had firm evidence
that HPV-based screening is indeed significantly better that the
conventional Pap smear.
The second paper I’d like to know more about
is from 2003, on the epidemiologic classification of HPV types in
terms of their risk to induce cervical cancer (Muñoz N, et
al. "Epidemiologic classification of human
papillomavirus types associated with cervical cancer," NEJM
348: 518-27, 2003).
This is the final contribution of the series of case control studies
organized from the IARC by the group led by Dr. Muñoz. In 1992 the
causal link was announced for the first time based on two case-control
studies conducted in Spain and Colombia. By the end of the project we had
11 papers on case-control studies, several of which feature in the Special
Topics rankings. The studies were of large enough samples to allow for a
specific estimate of the risk by HPV type, which is something that few of
the other studies were able to do. The type-specific risk estimates allowed
for the empirical epidemiological classification of the HPV types, which
was a critical contribution.
Paper #3 is a review paper on the causal relation
of HPV to cancer (Bosch FX, et al.,
"The causal relation between human
papillomavirus and cervical cancer," Journal of Clinical
Pathology 55: 244-65, 2002). This one addresses
public health officials and regulatory bodies.
This paper was written because the first time the proposal to use HPV
tests was submitted to regulatory bodies in the US it was turned down
because reviewers claimed that the causal association had not been proven.
So the rationale for HPV screening was in doubt.
Then I was invited to prepare a fairly academic paper discussing the points
at issue. The exercise blew away my 2001 summer vacation! The review put
together a significant fraction of the epidemiological literature at the
time. It went back to the criteria of causality that historically had been
used for human carcinogens, and then reviewed one by one to ascertain which
were the best pieces of evidence that would support the hypothesis one way
or the other.
This paper is written in a language that is relatively easy and accessible.
We reached the people who had to make the decisions on HPV screening. The
paper is a proper review backed up by the best literature, and that was
highly desirable at the time. The writing of this paper included an
extensive group of reviewers who were most of the best in the field. I
integrated their feedback and the effort was duly acknowledged. The names
backing up the paper were unchallengeable, and that’s probably one of
the reasons why so much notice has been taken of the paper. At the next
session of the regulatory hearings, the manuscript was used as sufficient
to prove causality. Paper #3 is also being used extensively in teaching the
way epidemiology tackles cancer causality issues.
Dr. F. Xavier Bosch BSc, M.D., Ph.D.
Cancer Epidemiology Research Program
Catalan Institute of Oncology
Bosch'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. Source:
Essential Science Indicators from
Keywords: human papillomavirus, HPV, epidemiology, cervical
cancer, etiology, developing countries, screening, vaccine, public
health, regulatory bodies.