|
|  |
|
in-cites,
December 2004
http://www.in-cites.com/papers/NancyLeeHarris_ILSG.html
|
|
|
|
An
interview with:
Dr. Nancy Lee Harris and the ILSG |
|
|
|
 n
the interview below, Dr. Nancy Lee Harris and her colleagues
discuss their highly cited paper, "A revised
European-American classification of lymphoid neoplasms—a
proposal from the International Lymphoma Study Group," (Blood
84[5]: 1361-92, 1 September 1994). According to the ISI
Essential Science Indicators
Web product, their REAL classification paper has been cited
3,523 times to date, making it one of the top five papers in
the field of Clinical Medicine over the past decade.
|
Why do you think your paper is highly cited?
The late 1990s saw an explosion in both clinical and basic
research in lymphomas, with a large number of publications. In this
decade, as often in the past, advances in understanding the biology
of lymphomas preceded and paved the way for advances in other tumor
types. The REAL classification, with its well-defined entities and
multimodality approach to diagnosis, facilitated the application of
emerging techniques to study the pathogenesis and biology of
specific entities and to define new targets for treatment. For this
reason, the REAL classification became the standard lymphoma
classification cited in basic and clinical research papers during
the 1990s.
What are the circumstances that led you to your work?
Lymphoma classifications have always been controversial. Over the
years, much of this controversy arose from the assumption that there
had to be a single guiding principle—a "gold standard"—for
classification. Many early classifications were based purely on
morphology; others utilized primarily clinical features; and others
were based primarily on cell lineage and differentiation, in the
belief that each neoplasm corresponded to a recognizable normal cell
or differentiation stage. In addition, most classifications were the
work of a single author or a small group under a strong leader—thus
there was no theoretical limit to the number of competing
classifications.
|
“The adoption of the REAL classification by increasing numbers of centers has resulted in greater precision of clinical studies and greater ease of interpretation of studies carried out in different centers.”
|
|
From the 1970s through the early 1990s, multiple lymphoma
classifications were used in different parts of the world—most
recently, the Kiel Classification in Europe, and the Working
Formulation, and, to a lesser extent, the Lukes-Collins
classification in the U.S. This lack of consensus on lymphoma
classification and terminology caused problems for both pathologists
and clinicians, and created difficulty in interpreting published
studies. In addition, in the 1980s and 1990s, many new disease
entities were described, which were not included in either
classification, leading to confusion among both pathologists and
oncologists about which were "real" diseases that they
should be recognizing in daily practice. Finally, the introduction
of immunophenotyping and molecular genetic analysis led to confusion
about what, if anything, should be the modern "gold
standard" for defining disease entities.
In 1990, Peter Isaacson in London and Harald Stein in Berlin
founded the International Lymphoma Study Group (ILSG), believing
that new knowledge about normal and neoplastic lymphocytes as a
result of advances in immunophenotyping and molecular biology were
likely to permit resolution of some former controversies. The ILSG,
an informal, invited group consisting of members from the U.S.,
Europe, and Asia, began meeting annually in 1991, to discuss both
current research and problem areas in lymphoma diagnosis and
classification. It rapidly became apparent that there was
substantial agreement among the members on diseases that they were
recognizing in daily practice and which appeared to be distinct
clinical entities. However, a consensus was needed on disease
definitions, criteria for diagnosis, and nomenclature. After the
first two meetings, the group prepared manuscripts describing our
consensus on two controversial lymphoma entities.
At the second meeting of the ILSG, Harald Stein, Elaine Jaffe,
and Nancy Lee Harris proposed to the group that we collaborate on a
new lymphoma classification. Because of the complexity of arriving
at an overarching principle for a classification, we felt that the
best approach would be to attempt to reach a consensus on a list of
diseases that we could all agree were "real" entities—what
pathologists were actually doing in practice, not a hypothetical
scheme. In this classification, all available information—morphology,
immunophenotype, genetic features, and clinical features—is used
to define a disease entity. The relative importance of each of these
features varies among diseases, and there is no one "gold
standard." Furthermore, in contrast to previous
classifications, we did not attempt to sort these disparate diseases
into broad categories such as "low grade" and "high
grade" based on either morphologic or clinical features. Both
clinicians and pathologists need to recognize each disease as a
distinct entity, and understand its spectrum of morphology and
clinical behavior.
In 1993, Harald Stein hosted the ILSG meeting in Berlin, where 19
of us met and worked out a consensus on the list of diseases over
three days. Over the next year we drafted the manuscript and
presented it to a meeting of international clinical experts; it was
published in the fall of 1994. Because it was based on our
experiences with the previous European (Kiel) and American (Working
formulation) classifications, we called it the "revised
European-American Lymphoma Classification," which gave rise to
the convenient—if provocative—acronym "REAL,"
emphasizing our commitment to defining real disease entities.
Would you describe the significance of this work for your
field?
The availability of a list of well-defined disease entities,
developed by a consensus of leading hematopathologists, has allowed
both clinical and basic research in the field to proceed at a
greater rate and in a more focused manner. A major development of
the 1990s and early 2000s has been the application of molecular
techniques, most recently microarray analysis, to the diagnosis and
classification of lymphoid neoplasms. New knowledge is easily
incorporated into this classification, since all available data are
used to define and diagnose diseases. The adoption of the REAL
classification by increasing numbers of centers has resulted in
greater precision of clinical studies and greater ease of
interpretation of studies carried out in different centers.
Where has this research gone since the publication of your
paper? Where do you see it going 10 years from now?
When the REAL classification was first published, it was very
controversial. The same issue of the journal contained a highly
critical editorial by one of the most prominent clinicians working
in the field. Some prominent clinicians and pathologists felt
"ownership" of classifications and resented this upstart
group, whose very existence some regarded as subversive. However,
several oncologists, including Vincent DeVita at Yale and James
Armitage at Nebraska, saw the REAL classification as a paradigm
shift. Armitage proposed an international study of the
classification, to test its clinical validity in defining distinct
entities, and to see if pathologists could use it reproducibly. This
study was carried out over the year after publication of the
classification, with the participation of Hans-Konrad Muller-Hermelink
from the ILSG and four "neutral" pathologists. It showed
that pathologists could use the classification with better
reproducibility than other classifications and that it provided
better discrimination among diseases with distinct clinical features
and prognosis than either the Working formulation or the Kiel
Classification. Presentation of these results at meetings and later
in journals resulted in rapid adoption of the classification by
hemato-oncologists.
Shortly after publication of the REAL classification, the World
Health Organization decided to produce a new "blue book"
on hematopoietic and lymphoid malignancies. Under the leadership of
Elaine Jaffe and together with members of the ILSG, this became an
international effort, involving over 50 pathologists under the
auspices of the two major hematopathology societies—the Society
for Hematopathology in the U.S. and the European Association of
Hematopathologists—in order to broaden the consensus on lymphoid
neoplasms and extend the concepts of the REAL classification to the
myeloid and histiocytic neoplasms. This effort included a Clinical
Advisory Committee of over 50 international hematologists and
oncologists, who provided input into issues of clinical relevance
and helped to ensure that the outcome would be useful and used in
clinical practice. All the authors of prior classifications
eventually agreed to use the new WHO classification after its
publication in 2001, and it represents the first true international
consensus on classification of hematopoietic and lymphoid neoplasms.
The process of discussions among experts—both pathologists and
clinicians—that occurred during the development of the REAL and
WHO classifications was truly remarkable. All the members of the
group made important contributions, and there was no one who did not
learn something from the process. It has resulted in an
unprecedented level of communication among pathologists and between
pathologists and clinicians in this once very contentious area.
What lessons would you draw from your work to share with the
next generation of researchers?
It is clear from the many citations of the REAL classification
that research progress is facilitated by having a rational platform
on which new research projects can be developed. The REAL/WHO
classification is such a platform, and illustrates the importance of
continuous efforts to combine research results and daily experience
in practice into new concepts and understanding of diseases.
On a practical level, if pathologists are to have an impact on
patient care, they have to develop consensus on disease
classifications and diagnostic criteria. Multiple competing
classifications result in loss of credibility with clinical
colleagues, and complicate patient care and research. Reaching a
consensus requires compromise, but the only thing worse than an
imperfect classification is multiple classifications. The most
important factor in reaching a consensus is a prior agreement that a
consensus will result. If anyone enters the process thinking he or
she can exit if unhappy with the result, the process will fail. We
even resorted to voting when it appeared that consensus was
impossible, and often found that there was a substantial majority
and that only one or two individuals were holding up the discussion.
Another important lesson for pathologists is the importance of
recognizing the respective roles of pathologists and clinicians in
developing pathologic classifications of disease. Pathologists may
become overly enthusiastic about details that have little clinical
or biological relevance, and clinicians may become overly focused on
survival (which nowadays usually means response to whatever
treatments are available). Pathologists should take responsibility
for developing classifications, but should involve clinical
colleagues during the process, so that the result is both familiar
and useful to clinicians.
A final lesson is that in order to retain "power" you
have to share it. Authors of several preceding classifications
insisted that only they could revise or update them; this refusal to
include a wider group of contributors ultimately led to the
classifications being discarded. Involving a large group in
consensus development results in a better product and one that is
more likely to be used by others. Many of us in the ILSG who worked
on the REAL classification were reluctant to have it superceded so
quickly by the WHO—but again, the only sure road to long-term
success was to relinquish individual credit and try to achieve the
largest possible consensus. In the future, the WHO classification
will need to be modified by adding new diseases and by improving the
definitions of existing diseases. We are now working on a process
whereby the classification can be updated through committees of the
Hematopathology societies, to ensure that the "patho-Babel"
of the 20th century does not recur.
Dr. Nancy L. Harris and the International Lymphoma Study Group
Dr. Harris is Professor of
Pathology at Massachusetts General Hospital and Harvard Medical
School, and the Editor of Case Records of Massachusetts General
Hospital for the New England Journal of Medicine
Boston, MA, USA
|
|
|
in-cites, December 2004
http://www.in-cites.com/papers/NancyLeeHarris_ILSG.html
|
|
cites
