Primary extranodal non-Hodgkin’s lymphoma (NHL): the impact of alternative definitions tested in the Comprehensive Cancer Centre West population-based NHL registry
Annals of Oncology
Primary extranodal non-Hodgkin's lymphoma (NHL): the impact of alternative definitions tested in the Comprehensive Cancer Centre West population-based NHL registry
A. D. G. Krol
S. le Cessie 1
S. Snijder 0
J. C. Kluin-Nelemans
P. M. Kluin 2
E. M. Noordijk
0 The Comprehensive Cancer Centre West, Leiden; Departments of
1 Medical Statistics, Leiden University Medical Centre , Leiden
2 Pathology, University Hospital Groningen , Groningen , The Netherlands
Background: The definition of primary extranodal non-Hodgkin's lymphoma (NHL) is a controversial issue, especially in patients where both nodal and extranodal sites are involved. Patients and methods: The impact of different definitions of primary extranodal NHL on incidence and prognosis is explored using data from a population-based NHL registry. Results: Using liberal criteria, 389 (34%) cases were classified as primary extranodal NHL. Overall survival (OS) rates of nodal and extranodal NHL patients defined this way were comparable; however, extranodal NHL patients had a better disease-free survival (DFS). When strict criteria were applied, 231 cases (20%) were classified as primary extranodal NHL. OS and DFS rates of extranodal NHL patients defined this way were superior to nodal NHL patients; however, the difference in OS was reversed after correction for differences in International Prognostic Index and malignancy grade. Conclusion: This study illustrates the selection bias that is introduced when a strict definition of primary extranodal NHL, that excludes cases with disseminated disease, is used. Patients with primary extranodal NHL were found to have a superior DFS, irrespective of which definition of primary extranodal NHL was used.
Introduction
Most non-Hodgkins lymphomas (NHL) arise in lymph nodes or
other lymphatic tissues such as the spleen, Waldeyers ring and
thymus. Involvement of so-called extranodal organs is a common
finding after staging investigation, however, and a substantial
part of NHL even arises in these sites. The latter form is often
referred to as primary extranodal NHL.
Although 2540% of NHL patients present with a primary
extranodal lymphoma [14], and numerous papers dealing with
extranodal NHL originating in almost every organ in the body
have been published, the literature on primary extranodal
lymphomas as a group is limited. Moreover, the definition of primary
extranodal lymphoma is a controversial issue, especially in
patients where both nodal and extranodal sites are involved. In
some series on primary extranodal NHL this problem is
circumvented, and only patients with localised disease are included
[2, 58]. As primary extranodal lymphomas have the potential to
disseminate, these series thus present a picture of extranodal
NHL that is incomplete. On the other hand, studies that use more
liberal criteria for extranodal NHL [1, 4, 913], and include
patients with disseminated disease, might erroneously include
patients with disease that originated in lymph nodes or other
nodal sites.
In the Comprehensive Cancer Centre West (CCCW) NHL
registry, we were also confronted with the problems associated
with the definition of primary extranodal NHL as outlined above,
especially in patients with disseminated disease at both nodal and
extranodal sites. For these patients, a third pattern of presentation
was proposed: extensive disease [14, 15]. Data from our registry
were used to define and explore the relevance of this subdivision
of NHL in primary nodal, extranodal and extensive involvement.
Its prognostic significance was tested, together with the two
alternative subdivisions of nodal and extranodal NHL commonly
used in the literature, by relating them to prognostic factors,
response to therapy and survival.
Patients and methods
Inclusion criteria and lymphoma classification
Included in the CCCW NHL registry were all newly diagnosed NHL patients
living in the region covered by the CCCW (determined by postal code).
Registration started on 1 June 1981 and ended on 31 December 1989. Once a
diagnosis of NHL was made by the hospital pathologist, slides and frozen
material were sent for review to a panel of four regional pathologists,
experienced in hemato-pathology. Additional immunological and
enzymehistochemical techniques were performed to enable classification according
to the Kiel classification, with some modifications, and the International
Working Formulation. Only patients with a diagnosis of NHL confirmed
by the pathology panel were included in the database, while patients with
primary cutaneous lymphoma, multiple myeloma, acute lymphoblastic
leukaemia and chronic lymphocytic leukaemia, and all cases diagnosed upon
post-mortem examination were excluded from registration. After the
inclusion of new patients had stopped, the Revised EuropeanAmerican
Lymphoma Classification (REAL) for lymphomas was proposed by the
International Lymphoma Study Group. As this new scheme for classification of
lymphomas wa (...truncated)