Journal of Neuro-Oncology 32: 63–69, 1997.

 1997 Kluwer Academic Publishers. Printed in the Netherlands.

Clinical Study

Glucocorticoid-induced long-term remission in primary cerebral lymphoma:

case report and review of the literature

Benoı̂t Pirotte1, Marc Levivier1, Serge Goldman2, Jean-Marie Brucher3, Jacques Brotchi1 and Jerzy Hilde-
brand4
Departments of 1 Neurosurgery and 4 Neurology, 2 PET/Biomedical Cyclotron Unit, Erasme Hospital, Uni-
versité Libre de Bruxelles; 3 Department of Neuropathology, Saint Luc Hospital, Université Catholique de
Louvain, Brussels, Belgium

Key words: central nervous system, brain neoplasm, non-Hodgkin’s lymphoma, brain lymphoma, glucocorti-
coids

Abstract

We report a 25-year old immunocompetent woman with a high grade primary non-Hodgkin’s lymphoma of
the central nervous system (PNHL-CNS) in whom the administration of dexamethasone alone during three
months produced a complete clinical and radiological response lasting over four years. If complete remission
of PNHL-CNS induced by glucocorticoids are well known, the opportunity to observe glucocorticoid-induced
remission for a long period of time without radio- and chemotherapy is rare. Only nine other cases of PNHL-
CNS with complete remission induced by glucocorticoids lasting from 6 to 60 months, were found in the
literature and are summarized here. Duration of glucocorticoids therapeutic effect in PNHL-CNS is probably
underestimated. Glucocorticoids cannot be recommended as sole initial treatment for PNHL-CNS. However,
we suggest standard therapies to be delayed in those patients responding completely to glucocorticoids where
radio- and chemotherapy should be contraindicated (kidney, liver, bone marrow failure, pregnancy).

Introduction Case report

The combination of chemotherapy (CT) and radi-
ation therapy (RT) has substantially improved life
expectancy in patients with primary intracerebral
malignant non-Hodgkin’s lymphomas (PNHL-
CNS) [1–3]. The neurotoxicity of most of the re-
gimen used is not negligible and the best combina-
tion is yet to be established [2, 4–6]. In one third of
PNHL-CNS, glucocorticoids (GC) have a partial
and transient lymphocytolytic effect and more
rarely induce complete remission [2, 7–14]. How-
ever, the duration of this effect may possibly be un-
derestimated as illustrated by the present case re-
port and the review of the literature [2, 5, 11, 15–
24].
In June 1989, an immunocompetent 25-year-old
woman was admitted for right fronto-parietal head-
ache, vomiting and left hemiparesis evolving over
ten days. These symptoms appeared one month af-
ter an uneventful vaginal delivery of a healthy child.
She had no fever, no adenopathy or hepatospleno-
megaly. Neurological examination showed a left he-
miparesis with increased tendon reflexes and a Ba-
binski sign. Magnetic resonance imaging (MRI) re-
vealed an enhanced supratentorial intraparenchy-
mal mass, which extended from the right cerebral
peduncle to the right thalamo-capsular region (Fig-
ure 1A). Laboratory screening showed no sign of
inflammatory or infectious processes. Plasma pro-
lactin was normal (15 ng/ml). Human immunodef-

Please indicate author’s corrections in blue, setting errors in red

108913 NEON ART.NO 678-95 (779/flop) ORD.NO 234779

64

A B
Figure 1. (A) In June 1989, MRI showed an intraparenchymal mass enhanced after contrast injection which extended from the right
cerebral peduncle to the right thalamocapsular region. (B) In August 1989, MRI showed a more than 95% reduction of the contrast-
enhanced area after GC treatment.

iciency virus (HIV) serology was negative. Chest X-
rays were normal. Because of clinical and radiologi-
cal signs of intracranial hypertension, oral dexa-
methasone 16 mg/day was started at admission. All
the symptoms and signs disappeared within 48
hours. A stereotactic CT-guided biopsy was per-
formed 13 days after admission. Microscopic exam-
ination of tissue samples was performed as de-
scribed [25] and showed chronic inflammation
without sign of malignancy (Figure 3A). Interest-
ingly, CT scan obtained at the time of biopsy re-
vealed a 30% reduction of the contrast-enhanced
lesion (not shown). An extensive workup looking
for a systemic disease including sarcoidosis, infec-
tious process and vasculitis was performed and was
negative. Cerebrospinal fluid (CSF) examination
including immunoelectrophoresis and cytopatholo-
gy was normal. Fundoscopy was normal. A PNHL-
CNS was suspected. During the next two months,
the patient was maintained on oral dexamethasone
12 mg/day and phenytoin 300 mg/day. MRI per-
formed 2 months after the biopsy showed a reduc-
tion of more than 95% of the contrast-enhanced ar-
ea (Figure 1B). GC and antiepileptic medications
were gradually tapered and stopped. The patient
resumed her work three months after admission
and remained neurologically normal until Novem-
ber 1993. Sequential MRI or CT scans revealed no
tumor recurrence or other lesion (Figure 2A).
In November 1993, she was readmitted after a
partial seizure followed by a persistent left hemipa-
resis. MRI showed a contrast-enhanced intraparen-
chymal mass in the right parietothalamic region,
which was adjacent to the initially abnormal area
(Figure 2B). The patient had no fever, adenopathy
or hepatosplenomegaly. No uveal or vitreous de-
posit were observed. CSF contained 200 lympho-
cytes/mm3 (60% T-type and 32% B-type). Cytome-
galovirus, Epstein Barr virus (EBV) and HIV serol-
ogy was negative. Chest and abdomen CT scan were

Please indicate author’s corrections in blue, setting errors in red

108913 NEON ART.NO 678-95 (779/flop) ORD.NO 234779

 65

A B
Figure 2. (A) MRI follow-up during 52 months showed no tumor recurrence or other lesion. (B) In November 1993, MRI revealed a
contrast-enhanced intraparenchymal mass located in the right parieto-thalamic area, adjacent to the site of the initial lesion.

normal. A stereotactic biopsy of the newly-ap-
peared lesion was scheduled. The biopsy was guid-
ed by combined CT and positron emission tomo-
graphy (PET) with 18F-fluorodeoxyglucose (FDG)
as previously described [26, 27].
Microscopic examination revealed a B-type high
grade malignant lymphoma in all biopsy samples
(Figure 3B). Oral dexamethasone 16 mg/day ad-
ministered during ten days did not induce any neur-
ological and radiological response. The patient was
treated by chemotherapy followed by radiation
therapy according to the protocol described by De
Angelis et al. [2]. This treatment regimen produced
a complete clinical and radiological remission, now
lasting for over 18 months.
Review of the literature
Marked and even complete regression of parenchy-
mal brain lesions is seen in a significant number of
patients with PNHL-CNS after GC. For instance,
De Angelis et al. [2] observed a complete radiologi-
cal remission in 3 out of 20 patients in which the ef-
fect of GC could be assessed independently of the
effect of RT and CT. However, long-lasting obser-
vation of the therapeutic effect of GC without other
treatment is rare.
A review of the literature revealed nine patients
with radiologically documented complete remis-
sions (CR) of PNHL-CNS lasting over 6 months af-
ter GC treatment alone (Table 1). The diagnosis of
high grade PNHL-CNS was eventually made in all
the cases, but in five cases (A 3, 4, 7, 8 and 10), biop-
sy at the time of the first manifestation revealed no
tumor. In the five remaining cases, the diagnosis
was made at the time of recurrence (biopsy or au-
topsy). In most papers, the dose and the duration of
GC administration are not available. The longest
specified treatment lasted for 20 weeks, and the
highest daily dose reported was 32 mg of dexameth-
asone. The duration of the first CR ranged from 6 to

Please indicate author’s corrections in blue, setting errors in red

108913 NEON ART.NO 678-95 (779/flop) ORD.NO 234779

66

Figure 3.(A) First stereotactic biopsy (June 1989) showed some hypertrophic reactive astrocytes and sponginess of the brain tissue, which
contained diffuse and perivascular lymphocytic infiltrates, suggesting chronic inflammatory process (paraffin, haematoxylin and eosin,
× 330). (B) Second stereotactic biopsy (November 1993) showed numerous large B-lymphocytes infiltrating the brain tissue, which also
contained perivascular lymphocytic cuffings mainly of the T-cell type (paraffin, immunostaining with monoclonal mouse anti-human B
cell Dako L26, × 330).

60 months. In 8 cases, CR lasted over one year.
Cases A 2 and 4 had a second CR lasting for 6 and 12
months respectively, and case A 5 had a second and
a third CR for 7 and 6 months respectively. How-
ever, these patients had eventually a poor response
to GC at the next relapse. In contrast, cases A 1, 3, 6,
7, 8, 9 and our patient (A 10) did not exhibit the
same rapid and dramatic response to GC at the time
of the first recurrence after the GC-induced CR.
However, in most of these cases, other antineoplas-
tic therapies were initiated soon after recurrence
which is a confounding factor for the evaluation of
GC effect.
Discussion
Our observation emphasizes that prompt adminis-
tration of GC in patients with PNHL-CNS may pro-
duce a rapid lysis of neoplastic cells and may pre-
clude correct neuropathological diagnosis even in
samples obtained by carefully targeted stereotactic
biopsy [28–31]. An interesting aspect of our report
is that both clinical and radiological CR lasted over
4 years after complete discontinuation of GC and in
absence of any other treatment. Indeed, we have
not considered the residual contrast enhancement
on MRI, which persisted unchanged for about 4
years, as evidence of residual tumoral tissue (Figure
2A). Nevertheless, the local recurrence of a tumor,
diagnosed as PNHL-CNS, suggests that the eradica-
tion of the neoplastic cells was not complete.
In PNHL-CNS, partial or complete regression of
brain lesions induced by GC is seen in approximate-
ly one third of the cases [2, 11, 28, 29, 31]. As stated
by De Angelis et al. [2], the clinical and radiological
response is often considered as a diagnostic test and
a specific treatment of the suspected lymphoma is
usually started. Therefore, long-lasting observation
of the therapeutic effect of GC without other treat-
ment is rare. Our observation and an extensive re-
view of the literature indicate that long-lasting CR
induced by GC alone is not uncommon. However,
since the sensitivity of these tumors to CT and RT
has been recognized [1, 2], the opportunity to ob-
serve and to evaluate the effect of GC alone has vir-
tually disappeared. Indeed, in the 10 cases reviewed
here, there was a lack of pathological diagnosis at
first which justified therapeutic abstention. This al-
lowed the observation of a rapid and long-lasting
remission of clinical and radiological signs under
GC alone. In these cases, we, as others, have not
considered the response to GC as a diagnostic test
of PNHL-CNS, and did not started appropriate
therapy without histological diagnosis. Indeed, in
our case, RT combined to CT was started when the
diagnosis of PNHL-CNS was confirmed at the time
of relapse. Also, some of the cases reviewed date
from an era when suspected PNHL-CNS were not
treated as aggressively as it is today.

Please indicate author’s corrections in blue, setting errors in red

108913 NEON ART.NO 678-95 (779/flop) ORD.NO 234779

 67

In all the cases reviewed, the response to GC in
the absence of any other antineoplastic treatment
has been considerably longer than the period of
their administration. Therefore, the duration of the
GC-induced CR is an aspect which may be current-
ly underestimated in therapeutic management [28,
32–34]. The response to GC seen at recurrence is
either much shorter than the initial one, or only par-
tial, or even absent. The mechanism of this resist-
ance has not been clarified. It could correspond to
selection of clones with no or fewer GC-receptors.
Also it is not known whether the development of
GC resistance is accompanied by an increased re-
sistance to other treatment modalities. Therefore,
despite the long lasting remissions to GC observed
in some patients, the use of GC alone cannot be rec-
ommended to treat PNHL-CNS. However, the
question of whether some antineoplastic treatment
can be delayed remains open [1]. At least in the
cases reviewed here, the administration of a poten-
tially highly neurotoxic treatment has been delayed
by the use of GC alone and a high quality of life has
been maintained from months to years. We suggest
that (1) the dose and the schedule of GC adminis-

Table 1. Complete clinical and radiological remission longer than 6 months after glucocorticoids alone

No Age gender Stage of disease Pathology GC therapy Radiological response Reference

Autopsy (A) Type Duration Type Duration

Biopsy (B) Daily dose (Weeks) (Months)
(mg)

1 -/M 1st location ACTH – CR 36 Ruff et al. 1979

Recurrence (A) Histiocytic L – – PR ?
2 34/M 1st location * Dxm 8 Before CT
era 60 Williams et al.
1979
1st recurrence Dxm 4 CR 12
2nd recurrence (A) Malignant L Dxm – PR 6
3 55/M 1st location (B) No tumor Dxm – CR 8
Recurrence (B) Malignant L Dxm – PR ?
4 25/M 1st location (B) No tumor Dxm 16 – CR 12 Singh et al. 1982
1st recurrence (B) Lymphoblastic L Dxm – CR 6
2nd recurrence Dxm – PR ?
5 59/M 1st location – 2 CR 12 Todd et al. 1986
1st recurrence – – CR 7
2nd recurrence – 2 CR 6
3rd recurrence (A) Histiocytic L – – PR ?
6 73/M 1st location Dxm 32 1 CR 18 Pohl et al. 1989
Recurrence (B) Immunoblastic L Dxm – PR ?
7 – 1st location (B) No tumor Dxm 16 – CR 6 De Angelis et al.
1990
Recurrence (B) High grade L Dxm – PR ?
8 – 1st location (B) No tumor Dxm 16 – CR 15
Recurrence (B) High grade L Dxm – PR ?
9 63/M 1st location Dxm 5 20 CR 30 Van den Bent et
al. 1992
Recurrence (B) Centroblastic L Dxm 15 – PR ?
10 25/F 1st location (B) No tumor Dxm 16 8 CR 52 Present case
Recurrence (B) Lymphoblastic L Dxm 16 2 PR 24

A: Autopsy, B: Biopsy, CR: Complete remission, CT: Chemotherapy, Dxm: Dexamethasone, GC: Glucocorticoids, L: Lymphoma, PR:
Partial response, RT: Radiation therapy, *: considered as Multiple Sclerosis, –: Not mentioned, ?: Clinical and radiological partial re-
sponse to glucocorticoids whose duration is not evaluable because complementary treatment including radiation therapy and/or chemo-
therapy was rapidly administered.

Please indicate author’s corrections in blue, setting errors in red

108913 NEON ART.NO 678-95 (779/flop) ORD.NO 234779

68

tration should be carefully standardized in studies 11. Hochberg FH, Miller DC: Primary central nervous system
lymphoma. J Neurosurg 68: 835–853, 1988
comparing the efficiency of different therapeutic
12. Homo-Delarche F: Glucocorticoids receptors and steroid
protocols in PNHL-CNS, and (2) when patients sensitivity in normal and neoplastic human lymphoid tis-
with suspected PNHL-CNS present an initial CR to sues: a review. Cancer Res 44: 431–437, 1984
GC alone, further treatments could be possibly de- 13. Shapiro WR, Posner JB: Corticosteroid hormones. Effects
layed until recurrence especially when CT and RT in an experimental brain tumour. Arch Neurol 30: 217–221,
1974
are contraindicated (kidney, liver or bone marrow
14. Smith T: Tumor lysis syndrome after steroid therapy for ana-
failure, pregnancy). This second point should be phylaxis. South Med J 81: 415–416, 1988
further evaluated. 15. Coca A, Goday A, Font J, Ingelmo M, Balcells A: Letter to
the editor: Primary cerebral lymphoma: the ‘Ghost tumor’. J
Neurosurg 61: 202, 1984
16. Gray RS, Abrahams JJ, Hufnagel TJ, Kim JH, Lesser RL,
References
Spencer DD: Ghost-cell tumor of the optic chiasm. Primary
CNS lymphoma. J Clin Neuro Ophtalmol 9: 98–104, 1989
1. Boiardi A, Silvani A, Valentini S, Salmaggi A, Allegranza A,
17. O’Neill BP, Kelly PJ, Earle JD, Scheithauer B, Banks PM:
Broggi G: Chemotherapy as first treatment for primary ma-
Computer-assisted stereotaxic biopsy for the diagnosis of
lignant non-Hodgkin’s lymphoma of the central nervous
primary central nervous system lymphoma. Neurology 37:
system: preliminary data. J Neurol 241: 96–100, 1993
1160–1164, 1987
2. De Angelis LM, Yahalom J, Heinemann MH, Cirrincione C,
18. Pohl P, Oberhuber G, Dietze O, Vogl G, Pallua AK, Plang-
Thaler HT, Krol G: Primary CNS lymphoma: combined
ger CA, Huber H: Steroid-induced complete remission in a
treatment with chemotherapy and radiotherapy. Neurology
case of primary cerebral non Hodgkin’s lymphoma. Clin
40: 80–86, 1990
Neurol Neurosurg 91: 247–250, 1989
3. Rosenthal MA, Sheridan WP, Green MD, Liew K, Fox RM:
19. Ruff RL, Petito CK, Rawlinson DG: Primary cerebral lym-
Primary cerebral lymphoma: an argument for the use of ad-
phoma mimicking multiple sclerosis. Arch Neurol 36: 598,
junctive systemic chemotherapy. Aust N Z J Surg 63: 30–32,
1979 (letter)
1993
20. Singh A, Strobos RJ, Singh BM, Rothballer AB, Reddy V,
4. Pollack JF, Lunsford LD, Flickinger JC, Dameshek HL:
Pulijc S, Pui Poon T: Steroid-induced remissions in CNS lym-
Prognostic factors in the diagnosis and treatment of primary
phoma. Neurology 32: 1267–1271, 1982
central nervous system lymphoma. Cancer 63: 939–947, 1989
21. Todd II FD, Miller CA, Yates AJ, Mervis LJ: Steroid-in-
5. Buell WF, Hart RG: Spontaneous temporary remission in
duced remission in primary malignant lymphoma of the cen-
primary CNS lymphoma (letter). Can J Neurol Sci 15: 88,
tral nervous system. Surg Neurol 26: 79–84, 1986
1988
22. Van den Bent MJ, Vanneste JAL, Ausink BJJ: Prolonged
6. Ott RJ, Brada M, Flower MA, Babich JW, Cherry SR, Dee-
remission of primary central nervous system lymphoma af-
han BJ: Measurements of blood-brain barrier permeability
ter discontinuation of steroid therapy. J Neurooncol 13: 257–
in patients undergoing radiotherapy and chemotherapy for
259, 1992
primary cerebral lymphoma. Eur J Cancer 27: 1356–1361,
23. Vaquero J, Martinez R, Rossi E, Lopez R: Primary cerebral
1991
lymphoma: the ‘Ghost tumor’. Case report. J Neurosurg 61:
7. Distelhorst CW, Howard KJ: Kinetic pulse-chase labeling
174–176, 1984
study of the glucocorticoid receptor in mouse lymphoma
24. Williams RS, Crowell RM, Fisher CM, Davis K, Lavyne
cells. Effect of glucocorticoid and antiglucocorticoid hor-
MH, Ropper H, Bremer AM: Clinical and radiological re-
mones on intracellular receptor half-life. J Biol Chem 264:
mission in reticulum cell sarcoma of the brain. Arch Neurol
13080–13085, 1989
36: 206–210, 1979
8. Dowd DR, MacDonald PN, Kom M, Haussler MR, Miesfeld
25. Brucher JM: Neuropathological diagnosis with stereotactic
R: Evidence for early induction of calmodulin gene expres-
biopsies. Possibilities, difficulties and requirements. Acta
sion in lymphocytes undergoing glucocorticoid-mediated
Neurochir 124: 37–39, 1993
apoptosis. J Biol Chem 266: 18423–18426, 1991
26. Levivier M, Goldman S, Bidaut LM, Luxen A, Stanus E,
9. Gametchu B, Watson CS, Pasko D: Size and steroid-binding
Prezdborski S, Balériaux D, Hildebrand J, Brotchi J: Pos-
characterization of membrane-associated glucocorticoid re-
itron emission tomography-guided stereotactic brain biop-
ceptor in S-49 lymphoma cells. Steroids 56: 402–410, 1991
sy. Neurosurgery 31: 792–797, 1992
10. Gametchu B, Watson CS, Shih CC, Dashew B: Studies on
27. Levivier M, Goldman S, Pirotte B, Brucher JM, Balériaux D,
the arrangement of glucocorticoid receptors in the plasma
Luxen A, Hildebrand J, Brotchi J: Diagnostic yield of ster-
membrane of the S-49 lymphoma cells. Steroids 56: 411–419,
eotactic brain biopsy guided by positron emission tomogra-
1991
phy with (18F)fluorodeoxyglucose. J Neurosurg 82: 445–452,
1995

Please indicate author’s corrections in blue, setting errors in red

108913 NEON ART.NO 678-95 (779/flop) ORD.NO 234779

 69

28. Geppert M, Ostertag CB, Seitz G, Kiessling M: Glucocorti-
coid therapy obscures the diagnosis of cerebral lymphoma.
Acta Neuropathol 80: 629–634, 1990
29. Kikuchi K, Watanabe K, Miura S, Kowada M: Steroid-in-
duced regression of primary malignant lymphoma of the
brain. Surg Neurol 26: 291–296, 1986
30. Sherman ME, Erozan YS, Mann RB, Kumar AA, McArthur
JC, Royal W, Uematsu S, Nauta HJ: Stereotactic brain biop-
sy in the diagnosis of malignant lymphoma. Am J Clin Path-
ol 95: 878–883, 1991
31. Yamasaki T, Kikuchi H, Yamashita J, Moritake K, Shibamo-
to Y, Paine JT, Shima N, Yamabe H: Intracerebral malignant
lymphoma with fluctuating regression and spatial evolution.
Surg Neurol 34: 235–244, 1990
32. Filla A, De Michele G, D’Armiento FP, Di Salle F, Cirillo S,
Marano E, Chiacchio L, Rapuano A, Buscaino GA: Steroid-
induced disappearance of primary central nervous system
lymphoma. Clinical, neuroradiological and pathological
findings. Acta Neurol (Napoli) 11: 400–407, 1989
33. Ranney HM, Gellhorn A: The effect of massive prednisone
and prednisolone therapy on acute leukemia and malignant
lymphoma. Am J Med 22: 405–413, 1957
34. Takahashi Y, Mikami J, Ito K, Ueda M, Sato H, Matsuoka T,
Takedo S, Ohkawara S, Fujisawa Y: The effect of steroid
treatment on intracranial malignant lymphoma and its serial
CT findings. Prog Comput Tomogr (Tokyo) 6: 75–88, 1984
Address for offprints: B. Pirotte, Department of Neurosurgery,
Erasme Hospital, 808 Route de Lennik, B-1070 Brussels, Belgi-
um

Please indicate author’s corrections in blue, setting errors in red

108913 NEON ART.NO 678-95 (779/flop) ORD.NO 234779