Primary non-Hodgkin's lymphomas of the liver with nodular and diffuse infiltration patterns have different prognoses.

Abstract

Primary liver non-Hodgkin's lymphomas have peculiar clinical and biological patterns. This study correlates these patterns with pathology and outcome.
Clinical records and histology of patients with primary liver non-Hodgkin's lymphoma, treated at our institution over a 20-year period, were reviewed. Lymphoproliferations occurring after liver transplantation were excluded. Survival analyses were performed with patients from the other published series (62 patients).
Our series included eight patients. Three patients had a nodular liver infiltration, corresponding to a large B-cell lymphoma. Five patients had a diffuse liver infiltration, of whom three had a T-cell lymphoma with predominant sinusoid infiltration, and two had a large B-cell lymphoma. Patients with diffuse liver infiltration presented with hepatomegaly, and two of these also had acute liver failure. Diffuse infiltration had a worse prognosis than nodular infiltration (P = 0.0033). Among these latter patients, those treated with an anthracycline-based chemotherapy had a better outcome (P < 0.0001).
Patients with primary liver lymphomas can be classified in two groups, depending on the type of infiltration. Those with nodular infiltration may benefit from anthracycline-based chemotherapy. Diffuse infiltration has a bad prognosis, and should be suspected in patients presenting with altered liver functions and hepatomegaly.

Full text

Page 1

Annals of Oncology 12: 1005-1010, 2001.
© 2001 Kluwer Academic Publishers. Primed in the Nellie/lands.
Original article
Primary non-Hodgkin's lymphomas of the liver with nodular and diffuse
infiltration patterns have different prognoses
J.-F. Emile,1 D. Azoulay,2 J.-M. Gornet,3 G. Lopes,3 V. Delvart,2 D. Samuel,2 M. Reynes,1
H. Bismuth2 & F. Goldwasser3
1 Service d'anatomie pathologique, 2Centre hepatobdiaire. }Federation des services des maladies sanguines, immumtaireset tumorales,
Hopital Paul Brousse. UPRES 1596 and 1FR89. Universile Paris Sud. Paris. France
Summary
Background: Primary liver non-Hodgkin's lymphomas have
peculiar clinical and biological patterns. This study correlates
these patterns with pathology and outcome.
Patients and methods: Clinical records and histology of
patients with primary liver non-Hodgkin's lymphoma, treated
at our institution over a 20-year period, were reviewed. Lym-
phoproliferations occurring after liver transplantation were
excluded. Survival analyses were performed with patients from
the other published series (62 patients).
Results: Our series included eight patients. Three patients
had a nodular liver infiltration, corresponding to a large B-cell
lymphoma. Five patients had a diffuse liver infiltration, of
whom three had aT-cell lymphoma with predominant sinusoid
infiltration, and two had a large B-cell lymphoma. Patients
with diffuse liver infiltration presented with hepatomegaly, and
two of these also had acute liver failure. Diffuse infiltration
had a worse prognosis than nodular infiltration (P = 0.0033).
Among these latter patients, those treated with an anthracyclin-
based chemotherapy had a better outcome (P < 0.0001).
Conclusions: Patients with primary liver lymphomas can be
classified in two groups, depending on the type of infiltration.
Those with nodular infiltration may benefit from anthracyclin-
based chemotherapy. Diffuse infiltration has a bad prognosis,
and should be suspected in patients presenting with altered
liver functions and hepatomegaly.
Key words: anthracyclin, hepatomegaly, liver, lymphoma
Introduction
Liver is one of the most frequently involved organs in
disseminated non-Hodgkin's lymphomas (NHLs), and
may be detected in up to 40% of patients at the time of
initial staging [1]. By contrast, primary liver NHLs are
rare and account for less than 1% of NHLs. Less than
100 cases of primary liver NHLs were reported in Med-
line indexed literature in 1998 [2], and only seven series
include more than three patients [3-9].
The diagnosis of primary liver NHL is often unsus-
pected before biopsy or surgical resection. Indeed, clin-
ical presentation is variable [10]. A liver mass may be
disclosed by echography in a patient with abdominal
pain, while other patients may be referred for liver trans-
plantation for acute liver failure. Patients have been
treated by surgical resection, radiotherapy, chemotherapy,
or by a combination of these treatments. Because of the
low number of reported cases, the variability of the
clinical presentation and of the treatment, the prognosis
and the optimal treatment of patients with primary liver
NHL are still debated.
We present herein a series of eight patients with
primary liver NHL treated in our institution over a 20-
year period. Our data, and the analysis of the other
published series, suggest that patients with primary liver
NHL should be classified in two distinct clinical entities,
nodular and diffuse liver infiltration. Most lymphomas
with nodular infiltration of the liver are of large B-cell
type, and have a good prognosis, and may benefit from
an anthracyclin-based chemotherapy. Patients with dif-
fuse liver infiltration have a poor prognosis.
Patients and methods
Patients
Patients with histologically proven liver lymphomatous infiltration
were retrieved from the files of the pathology department of Paul
Brousse hospital. Lymphoproliferative disorders (28 cases), occurring
after liver transplantation and Hodgkin's lymphomas (23 cases), were
excluded from the study. Over a 20-year period, from January 1979
to December 1999, slides and medical records from 51 patients were
available for the study. Histological samples and clinical records of
these patients were reviewed. Inclusion criteria were as previously
published (2)- I) symptoms mainly caused by liver involvement at time
of diagnosis: 2) absence of distant lymphadcnopathy, palpable clin-
ically at time of presentation or detected during staging radiological
studies, 3) absence of leukemic blood picture in peripheral blood film.
Lymphomatous infiltration of lymph node of the hepatic hilum. of the
spleen and/or of the bone marrow at time of diagnosis were not
exclusive. Altogether, 8 out of the 51 patient (16%) fit these criteria.
The 43 other patients had an extra-hepatic dissemination.
at University of Portland on May 24, 2011
annonc.oxfordjournals.org
Downloaded from

Page 2

1006
Table I. Clinical and biological presentation of patients with primary liver lymphoma.
Case #
2
3
5
1
6
7
4
8
Age/sexe
58/F
46/ F
49/M
25/F
54/F
51/F
34/M
73/F
Revealing symptoms
Abdominal pain
Sepsis. Digestive bleeding
Jaundice
Abdominal mass
Altered general status
Altered general status
Jaundice. Ascite
Altered general status
HM/SM
+ /-
+ / +
+ / -
+ / +
+ / +
+ / -
+ / +
+ /-
Liver in-
filtration
Nodular
Nodular
Nodular
Diffuse
Diffuse
Diffuse
Diffuse
Diffuse
ALAT
N
2N+
2N+
N
N
2N+
2N+
2N+
GGT
N
2N+
2N+
2N+
N
2N+
2N+
2N+
Biliru-
bine
N
2N +
2N+
N
2N +
2N+
2N+
N
ALP
N
2N+
2N+
NA
N+
2N+
2N+
2N+
PT
N
<N
<N
N
N
<N
<N
N
LDH
NA
NA
2N+
NA
N+
N+
2N+
2N+
Abbreviations: HM - hepatomegaly; SM - splenomegaly; ALAT - alanine aminotransferase; GGT - g-glutamyl-transpeptidase; ALP - alkaline
phosphatase; PT - prothrombin time; LDH - Lactic dehydrogenase; M - male; F - female; N - within normal range; N+ - less than twice the
normal value; 2N+ - more than twice the normal value: NA - not available.
Pathology
Diagnosis of NHL was established, according to R.E.A.L [II] and
WHO [12] classifications on paraffin wax embedded tissue from either
needle biopsies (n — 7) or surgical liver specimens (n = 1). Histological
analysis was performed on hematoxylin and eosin (H&E) and Giemsa-
stained slides.
Immunohistochemistry was performed on four urn thick sections
of paraffin-embedded liver samples with a three step avidine-biotin
technique according to manufacturer's instruction (LSAB2, Dako,
Copenhagen, Denmark). The primary antibodies were anti-CD5 (4C7,
Novocastra Laboratories Ldt. Newcastel uponTyne, UK), CD20 (L26,
Dako), Bcl-2 (Dako), Mib-I (Dako), p53 DO-7 (Dako),TIA-l (Coulter,
Hialeah, FL) mouse monoclonal antibodies, and anti-CD3 (Dako)
rabbit polyclonal antibodies. Immunostaining was also performed on
frozen sections of T-cell lymphomas with deltaTCRI (T-cell Sciences,
Cambridge, Massachusetts), betaFl (T-cell Sciences).
Bibliography and statistics
The data analyzed in this report was accrued through a review of the
English language medical indexed literature from January 1976 to
December 1999. As many individual case reports were lacking impor-
tant biological and/or clinical data, we only used data from series
including more than two cases of primary liver NHL [3-9] for stat-
istical analysis. Altogether, data from 62 patients published in eight
series (including ours), were analyzed. All but two [6, 8] of these eight
series were from a single institution.
Cumulative survival for patients with primary liver NHLs were
estimated by the Kaplan-Meier method based upon the duration of
survival specified for each patient in each report included within our
review. The survival functions were compared with the log rank test.
Statistical analysis was performed with StatView F-4.5 (SAS Institute,
Cary, North Carolina).
Results
The main clinical features of the patients at presentation
are summarized in Table 1. Six patients were female and
two were male, with an age ranging from 25 to 73 years
(median 50 years). All the patients had a liver enlarge-
ment at physical examination, four also had a spleno-
megaly. Six patients had negative serologies for B- and
C-hepatitis virus, one was virus B positive (#3) and one
was not tested (#1).
Clinical presentation
For three patients, liver involvement consisted in nodules
(Table 1). Two had no relevant clinical history, and one
had a virus B-related cirrhosis (#3). Physical examina-
tion revealed hepatomegaly in all cases, jaundice in one
and severe upper digestive bleeding and sepsis in another.
This latter patient was admitted in an intensive care unit.
Liver lymphomatous mass were hyperechogenic in all
cases. The liver nodule was unique in all cases. Liver
enzymes were elevated in two cases.
The five other patients had an hepatomegaly without
any detectable nodule with either echography or tomog-
raphy (Table 1). One of them (#1) had been treated
since the age of 1.5 for a primary immune B cell defi-
ciency by injections of human gammaglobulines. In
1979, at the age of 25 years the patient was admitted for
fever and weight loss without evidence of infection. At
this time, she had a hepatosplenomegaly. A splenectomy
(spleen weight 1.5 kg) and a liver biopsy were per-
formed. Patient # 7 received a kidney transplantation
six years before admission. The three other patients had
no relevant clinical history. The five patients presented
with B symptoms (fever, night sweats and/or weight
loss) and hepatomegaly. Liver enzymes were elevated in
all cases, and two patients (#4 and #7) had altered
liver function, and were initially admitted in an intensive
care unit.
Pathology
The three patients with liver nodules had a diffuse large
cell lymphoma of B-cell phenotype. The tumors had a
destructive growth pattern, without detectable residual
portal tracts inside. Tumor cells had a centroblastic
morphology (Figure la), and were positive for CD20
and negative for CD3 and CDS. Only a few (less than
5%) tumor cells were Bcl-2 positive. The tumor cells
were p53 positive in one case (#2) and negative in the
two others.
Among patients with diffuse liver infiltration, three
had a T-cell lymphoma (#1, # 6 and #7). In these
at University of Portland on May 24, 2011
annonc.oxfordjournals.org
Downloaded from

Page 3

1007
A*
Figure 1. Histology of primary liver lymphomas; (a) nodular infiltra-
tion of the liver corresponding to a large B-cell lymphoma: liver
parenchyma is destroid, and replaced by the tumour proliferation; the
tumour was monomorphic, and consisted in an area of large lympho-
ma cells with centroblastic morphology (H&S, original magnification
x400); (b) diffuse infiltration of the liver corresponding to a yb T-cell
lymphoma: liver architecture was preserved with liver-cell plates and
sinuses; Imphoma cells had a small size and a dense nucleus, they were
mainly localized within hepatic sinuses (arrowheads) (H&S, original
magnification x400).
cases, infiltration and dilatation of hepatic sinusoids
were predominant (Figure lb). Tumor cells were small
to medium sized. Tumor cells were CD3 positive, CD20
negative and contained numerous cytotoxic granules
(TIA-1 positive). A frozen liver biopsy was available for
patient #6, and tumor cells were positive for 5TCR1
and negative for (3F1, thus confirming the diagnostic
of y6 T-cell lymphoma. For the two other patients, no
frozen samples were available, and although the pre-
dominant sinusoid infiltration was suggestive of a y5
origin of these T-cell lymphomas, we failed to demon-
strate it. Necropsy was performed in patient # 1 , one
year after the initial liver biopsy. It revealed an enlarged
liver (5.3 kg) without nodules, and confirmed the pre-
dominant sinusoid infiltration by T-lymphoma cells.
The two other patients with diffuse liver infiltration
had a large cell lymphoma of B-cell phenotype. Tumor
growth pattern was not destructive in both cases. For
patient # 4 tumor infiltration was limited to portal
tracts, while tumor cells were present in portal tract as
well as in sinusoids in patient #8. In both cases, lym-
phoma cells were positive for CD20 and negative for
CD3, CD5 and p53. Bcl-2 was expressed by more than
80% of tumor cells in case #4, and not expressed by
tumor cells in case #8.
Treatments and outcome
Over a 20 years period, treatments of the patients of our
series was highly variable (Table 2). One of the patients
with nodular liver lymphoma (#3) died a few days after
liver biopsy. The two other (#2 and #5) were treated
with anthracyclin-containing regimens. For one chemo-
therapy was given after surgical resection of the tumor.
Both patients achieved a long-lasting complete remission.
Two out of the three patients with diffuse liver infil-
tration by T-cell lymphoma did not receive specific treat-
ment. Indeed, for the older case (#1), the diagnosis of
lymphoma was performed retrospectively. The patient
was treated with anti-infectious agents and corticoids,
and she died 12 months later with multi-organ failure.
Patient # 7 died four days post diagnosis from multi-
organ failure. Patient # 6 was treated with one cycle of
an anthracyclin-containing regimen and 8 cycles of
oxaliplatin and cytarabine. An autologous bone marrow
transplantation performed after BEAM conditioning
regimen. This patient was alive without evidence of
disease 17 months after diagnosis.
One of the two patients with diffuse liver infiltration
by a B-cell lymphoma died the day of the liver biopsy
from acute liver failure. The other patient achieved
partial remission after an anthracyclin-containing che-
motherapy, and finally died of tumor progression de-
spite salvage therapy.
Survival analysis
Although our series was one of the largest published, it
was not possible to perform a survival analysis. We thus
analyzed the data available from the other published
series with our own patients. Altogether 62 patients
were included. Nodular infiltration was diagnosed in 41
cases. Fourteen of them (34%) were dead 0 to 1.6 years
after diagnosis, while 27 were alive 0.1 to 12 years after
diagnosis. The one-year and three-year survivals were
70% and 57%, respectively (Figure 2a). Diffuse infiltra-
tion was diagnosed in 21 patients of the largest pub-
lished series. Most of them (81%) died, and death
occurred during the first year following diagnosis in 3/4
of them. Prognosis of patients with diffuse liver infiltra-
tion was worse than that of patients with nodular liver
infiltration (P = 0.0033).
As anthracyclin-containing chemotherapies are used
to treat the majority of patients with aggressive NHLs,
we tried to determine whether this type of treatment was
at University of Portland on May 24, 2011
annonc.oxfordjournals.org
Downloaded from

Page 4

1008
Table 2.
Case
#
2
3
5
Prognostic factors.
Year of
diagnosis
1983
1988
1995
Age
58
46
49
treatment and outcome of patients
B
symp-
toms
0
+
0
Extra-
hepatic
extension
0
0
0
IPI
1
2
2
with primary liver lymphoma.
Phenotype
B (Bcl-2+)
B
B
Specific treatment
Surgery and 8 cycle anthra
None
8 cycle anthra
Outcome
follow-up
CR 144 months"
DOD < 1 month
CR 52 months"
1
6
7
4
8
1980
1998
1998
1994
1999
25
54
51
34
73
+
+
+
4
+
Spl
Spl. BM
0
0
BM
2
2
2
3
3
T
T
T
B
B
None
6 cycle anthra, ABMT
None
1 cycle chemotherapy without anthra
6 cycle anthra puis 1 AraC oxaly
DOD 12 months
CR 15 months"
DOD < 1 month
DOD < 1 month
DOD 6 month
Abbreviations. B symptoms - fever, night sweas and/or weight loss; IPI - international prognostic index; Spl - sleen; BM - bone marrow,
Anthra - anthracycline containing chemotherapy; ABMT - autologous bone marrow transplantation; AraC - aracytin; Oxaly - oxalyplatin; CR
- complete remission: DOD - dead of disease.
" Patients # 2 , 5 and 6 were in CR without treatment for 135. 44 and 6 months, respectively, after last follow-up.
(a)
1 -
6 -
4
2 -
0 -
KV
T—1
Nodiiar lymphoma
. Difuse lymphoma
P-O.OO33
12 14
Time (years)
(b)
Adnamycin
No admniycin
P<0.0001
0 2 4 6 8 10 12 11
Time (years)
Figure 2. Kaplan-Meier survival curves of patients with primary non-
Hodgkin's lymphoma reported in the largest series (including ours): (a)
survival of patients with nodular liver infiltration (upper curve) and
with diffuse liver infiltration (lower curve); (b) survival of patients with
nodular liver infiltration and treated by adriamycin-containing chemo-
therapy (upper curve) or other treatments (lower curve).
useful for patients with nodular primary liver NHLs.
Fourteen out of the forty-one patients with nodular liver
infiltration received an anthracyclin-containing chemo-
therapy. Survival of these patients (Figure 2b) was sig-
nificantly better than that of the patients receiving other
treatments (P < 0.0001). The number of patients with
diffuse liver infiltration and clinical information con-
cerning the treatment was not sufficient to test the value
of anthracyclin in this subset of patients.
Discussion
We present herein a series of eight patients with primary
liver non-Hodgkin's lymphoma, treated in our institu-
tion over a 20-year period. According to the type of liver
infiltration at presentation, we were able to classify the
patients in two groups. Nodular liver infiltration corre-
sponded to high grade B-cell lymphomas, and diffuse
liver infiltration corresponded to either T- or B-cell
lymphomas. Survival analysis of the published cases of
primary liver lymphomas confirmed the clinical interest
of separating these patients in two groups. Indeed, those
with nodular infiltration of the liver had a significantly
better prognosis than those with diffuse infiltration.
Nodular infiltration accounts for 66% of the 62 cases
of primary NHL of the liver reported in the series ([3-9],
and our series). The mean age of the 41 patients with
nodular liver NHL infiltration was 53 years (median 51,
range 21-87). The male/female sex ratio was 26/15. As
in our series, patients with nodular liver infiltration
usually presented with abdominal pain and/or hepato-
megaly. Fever, night sweats or weight loss (i.e., B symp-
toms) were present in 41% of the cases. The liver mass
was solitary in most cases (71%). All patients had nor-
mal a fetoprotein, and 86% had elevated serum levels of
lactic dehydrogenase. Several pathological classifica-
tions have been used since the first reported cases in
1969 [13]. However, as in our series, most cases (76%)
corresponded to diffuse large cell lymphomas, and phe-
notype was B cell in the recent series.
Diffuse liver lymphomatous infiltration was reported
in 21 (34%) out of the 62 patients of the published series.
The mean age of these patients was 57 years (median 54,
range 25-85). The male/female sex ratio was 12/9.
Most patients (19 of 21) had B signs at initial presenta-
tion. All had an hepatomegaly, and half (6 of 13) of the
patients from single center series had a splenomegaly.
at University of Portland on May 24, 2011
annonc.oxfordjournals.org
Downloaded from

Page 5

1009
Bone marrow was involved in half the cases (6 of 11). All
the patients had normal tx fetoproteins, and most of
them had elevated serum levels of lactic dehydrogenase.
Gaulard and al. described a subtype of lymphoma, with
predominant hepatic and splenic infiltration: y5 T-cell
lymphoma [14-16]. These NHL mainly involve middle-
aged women, some of whom may have received immu-
nosuppressive drugs (organ transplantation, Hodgkin's
lymphoma,...). Liver infiltration predominates in the
sinuses and tumor cells express the yS T-cell receptor.
Three patients in ours series had the clinicopathological
features of y§ T-cell lymphomas, two of which were
immuno-deficient (primary hypogammaglobulinemia
and renal transplantation). Other histological types of
lymphoma may be responsible for a primary diffuse liver
infiltration, some of which may be of B-cell phenotype
[17] as were the two other cases of our series.
Prognosis of primary liver lymphoma is still deatable,
mainly because of the low number of published cases,
the recent improvement of diagnostic tools (i.e., im-
munohistochemistry and molecular biology), the high
variability of treatments and the absence of prospective
studies. Our statistical analysis of the data provided by
published series (including ours), shows that prognosis
of patients with either nodular or diffuse infiltration of
the liver is different (P = 0.0033). The one-year and
three-year survivals were respectively 70% and 57% for
patients with nodular infiltration, and 38% and 18% for
patients with diffuse infiltration. Furthermore, patients
with nodular primary liver NHL had a better prognosis
when treated with an anthracyclin-containing chemo-
therapy (P < 0.0001). This important survival difference
may be partly due to the fact that the patients with the
worse initial performance status did not receive an
anthracyclin-containing chemotherapy. However, half
of the patients without anthracyclin treatment died
more than three months after diagnosis. The 62 patients
from these published series were treated over a period of
20 years. During this period of time, supporting cares
have improved. However, the survival was not related to
the date of diagnosis in our series, nor with the date of
publication in other series. We thus conclude that pa-
tients with nodular infiltration of the liver should be
treated with an anthracyclin-containing chemotherapy.
Lymphomatous diffuse liver infiltration may be asso-
ciated with an alteration of liver functions, and may
eventually be responsible for acute liver failure [18, 19].
Indeed, two patients from our series and one from
another group [9] presented with acute liver failure and
rapidly died. In a recently published series of 18 patients
with acute liver failure caused by tumor infiltration, 9
were related to a NHL [20], most of which were already
diagnosed. The diagnosis of primary liver infiltration by
lymphoma is often unsuspected, and a patient may be
referred for liver transplantation [21], or diagnosis may
be performed postmortem [18, 22]. Therefore, patients
presenting with acute liver failure and hepatomegaly
should undergo a liver biopsy.
In conclusion, primary liver lymphomas are rare tu-
mors. A nodular liver lymphomatous infiltration should
be suspected in either cirrhotic or non-cirrhotic patients
with hypovascularized liver nodule(s), normal alpha-
fetoprotein and elevated serum lactic dehydrogenase. It
usually corresponds to a large B-cell lymphoma. These
patients may benefit from anthracyclin-based chemo-
therapy. Diffuse liver lymphomatous infiltration should
be suspected in patients with hepatic or hepatosplenic
enlargement, B symptoms (i.e., weight loss, night sweats
or fever) and elevated liver enzymes and/or altered liver
functions. The bad prognosis of these patients may be in
part related to the diagnostic delay. An early liver biopsy
of these patients might allow initiation of a specific
treatment, before liver failure and/or lymphoma dis-
semination.
Acknowledgements
This work was supported by grants from the Groupe de
Recherche en Immuno-Pathologie (GRIP).
The authors want to thank medical and nurse teams
of the Centre Hepato-Biliaire and the Federation des
Services des Maladies Sanguines lmmunitaires et
Tumorales for caring for the patients, Pr N. Brousse
and C. Degott for sending slides from patient # 1 , and
M. Ortin-Serrano and F. Clusel for technical assistance.
J. F. Emile wishes to thank Ph. Gaulard for continuous
encouragement and fruitful discussions.
References
1. Chabner BA. Johnson RE. Young RC cl al Sequential non
surgical and surgical staging of non-Hodgkin's lymphoma. Ann
Intern Med 1976. 85 149-54.
2. Lei K1K. Primary non-Hodgkin's lymphoma of the liver. Leuk
Lymph 1998: 29: 293-9
3. Osborne BM. Butler JJ. Guarda LA. Primary lymphoma of the
liver. Ten cases and a review of the literature. Cancer 1985: 56:
2902-10.
4. DeMent SH. Mann RB. Staal SPet al. Primary lymphomas of the
liver. Am J Clin Pathol 1987. 88: 255-63.
5. Ryan J, Straus DJ. Lunge C et al. Primary lymphoma of the liver.
Cancer 1988:61- 370-5.
6. Anthony PP. Sarsfield P. Clarke T. Primary lymphoma of the
liver: Clinical and pathological features of 10 patients. J Clin
Pathol 1990; 43: 1007-13.
7 Scoazec JY. Degott C. Brousse N et al. Non-Hodgkin's lymphoma
preenting as a primary tumor of the liver: Presentation, diagnosis
and outcome in eight patients. Hepatology 1991: 13: 870-5.
8 Ohsawa M. Aozasa K. Horiuchi K et al. Malignant lymphoma of
the liver. Report of five cases and review of the literature. Dig Dis
Sci 1992:37: 1105-9
9. Lei KIK. Chow JHS. Johnson PJ. Aggressive primary hepatic
lymphoma in Chinese patients, presentation, pathologic features
and outcome Cancer 1995: 76' 1336-43.
10. Zafrani ES, Gaulard P. Primary lymphoma of the liver. Liver
1993: 13- 57-61
11. Harris NL. Jaffe ES. Stein H el al. A revised European-American
classification of lymphoid neoplasm. A proposal from the inter-
national lymphoma study group. Blood 1995; S4: 1361-92.
12. Harris NL, JafTe ES. Diebold J et al World Health Organization
classification of neoplastic diseases of the hematopoietic and
at University of Portland on May 24, 2011
annonc.oxfordjournals.org
Downloaded from

Page 6

1010
lymphoid tissue Report of the clinical advisory comittee meeting
- Airhe House. Virginia, November 1997. J Clin Oncol 1999; 17
3835-49.
13. Torres A. Primary lymphocytic follicular lymphoma of liver.
Cancer 1969; 27: 1489-92.
14. Gaulard P. Zafrani ES, Mavier P et al. Peripheral T-cell lym-
phoma presenting as predominant liver disease: A report of three
cases. Hepatology 1986; 6: 864-8.
15. Caulard P. Bourquelot P. Kanavaros Pet al. Expression of the
alpha/beta and gamma/delta T-cell receptors in 57 cases of
peripheral T-cell lymphomas. Am J Pathol 1990; 137: 617-28.
16. Farcet JP, Gaulard P. Marolleau JP et al. Hepatosplenic T-cell
lymphoma. Sinusal/sinusoidal localization of malignant cells ex-
pressing the T-cell receptor y5. Blood 1990; 75: 2213-9.
17. Trudel M, Aramendi T. Caplan S. Large-cell lymphoma presenting
with hepatic sinusoidal infiltration. Arch Pathol Lab Med 1991:
115:821-4
18. Colby TV. LaBrecque DR. Lymphoreticular malignancy pre-
senting as fulminant hepatic disease. Gastroenterology 1982; 82:
339-45
19. Sadahira Y. Ohmoto K. Yamamoto S et al. Expression of cytotoxic
molecule TIA-1 in malignant lymphomas mimicking fulminant
hepatitis. Pathol Int 1998; 48: 695-704.
20. Rowbotham D, Wendon J, Williams R. Acute liver failure secon-
dary to hepatic infiltration: A single centre experience of 18 cases.
Gut 1998; 42: 576-80.
21. Woolf GM, Petrovi LM, Rojter SE et al. Acute liver failure due to
lymphoma. A diagnostic concern when considering liver trans-
plantation. Dig Dis Sci 1994; 39: 1351-8.
22. Harris AC. Kornstein MJ. Malignant lymphoma imitating hepati-
tis. Cancer 1993; 71: 2639-46.
Received II December 2000; accepted 22 February 2001.
Correspondence to •
J.-F. Emile. MD, PhD
Service d'anatomopathologie
Hopital Paul Brousse BP200
94804 VillejuifCedex
France
E-mail: jean-francois.emile@pbr.ap-hop-paris.fr
at University of Portland on May 24, 2011
annonc.oxfordjournals.org
Downloaded from