Journal of Neuro-Oncology 43: 237–239, 1999.

© 1999 Kluwer Academic Publishers. Printed in the Netherlands.

Glucocorticoid treatment of primary CNS lymphoma

Michael Weller
Laboratory of Molecular Neuro-Oncology, Department of Neurology, University of Tübingen, Tübingen, Germany

Key words: primary CNS lymphoma, glucocorticoids, cerebral edema, apoptosis, bcl-2

Summary

Glucocorticoid therapy may result in the rapid resolution of cerebral mass lesions in patients with primary CNS
lymphoma. Since glucocorticoids will obscure the histological diagnosis of primary CNS lymphoma upon biopsy,
steroids should be withheld if primary CNS lymphoma is a likely diagnosis by neuroradiological criteria. The
lympholytic effect of glucocorticoids is mediated by cytoplasmic steroid receptors which are translocated to the
nucleus and signal apoptosis. Glucocorticoid-induced apoptosis of lymphoid cells does not require wild-type p53
activity, seems not to depend on caspase activation, but is attenuated by the bcl-2 protooncogene product. Long-
term glucocorticoid therapy of primary CNS lymphoma is not recommended because relapse is probably inevitable
and because of the prominent side effects of long-term glucocorticoid treatment. Further, long-term glucocorticoid
treatment is contraindicated in immunocompromised patients with primary CNS lymphoma.

Introduction latter domain result in constitutive receptor activation,

The role of glucocorticoids in the management of
patients with primary CNS lymphoma is an impor-
tant and controversial issue for both diagnostic and
therapeutic reasons. First, prior exposure to steroid
medications of patients suspected of harboring a pri-
mary CNS lymphoma is likely to obscure the diagno-
sis upon biopsy; second, the rapid lympholytic effects
of steroids in such patients make these medications an
important adjunct to the medical management of pri-
mary CNS lymphoma, notwithstanding the often tran-
sient nature of tumor remission and the severe side
effects of long-term glucocorticoid treatment.
Biochemical effects of glucocorticoids
The biochemical effects of natural or synthetic gluco-
corticoids are mediated by cytoplasmic receptors which
are translocated to the nucleus when bound by lig-
and and then bind DNA in a sequence-specific manner
to regulate target gene expression [1]. The receptors
form a family of proteins which share two zinc finger
domains for interactions with DNA and a carboxyter-
minal ligand binding domain. Since deletions of the
negative control appears to be the function of the car-
boxyterminus, and ligand binding derepresses receptor
activity. Most transcriptional regulation mediated by
the glucocorticoid receptor is positive but negative tran-
scriptional regulation has been observed as well. Glu-
cocorticoid responsive elements (GRE) usually contain
the sequence GGTACAnnnTGTTCT.
Glucocorticoids are potent inducers of
apoptosis in lymphoid cells
Glucocorticoid-induced apoptosis in lymphoid cells
was among the first paradigms of apoptotic cell death
examined in more detail and was the first to be
associated with nucleosomal size DNA fragmenta-
tion [2]. Glucocorticoid-induced apoptosis does not
involve some of the most common intracellular path-
ways of apoptosis and is thus a rather specific pro-
cess. Thus, apoptosis of lymphoid cells induced by
glucocorticoids does not require wild-type p53 activity
[3] and may not involve caspase activation [4,5].
In this regard, glucocorticoid-induced apoptosis dif-
fers from drug-induced apoptosis which commonly
238
involves p53, and from immune-mediated lymphoid
apoptosis which is probably commonly mediated
by caspases. Interestingly, while glucocorticoids are
thought to mediate most of their effects via receptor-
mediated transcriptional activation of various target
genes, glucocorticoid-induced apoptosis of lymphoma
cells may involve transcriptional repression rather than
activation [6]. Specifically, loss of nuclear factor of
activated T cells (NFAT) transcriptional activity has
been implicated in dexamethasone-induced apopto-
sis of thymocytes [7]. Enhanced expression of the
antiapoptotic BCL-2 protein inhibits glucocorticoid-
induced apoptosis of lymphoid cells [8,9].
Glucocorticoid effects in primary CNS lymphoma
Patients with primary CNS lymphoma commonly show
a dramatic clinical improvement after glucocorticoids
are administered. In contrast to glial neoplasms and
metastases from solid extracerebral tumors, the benefi-
cial effects of glucocorticoids in patients with primary
CNS lymphoma are not solely mediated by a reduc-
tion of cerebral edema but appear to involve cytotoxic
activity. In fact, cerebral mass lesions may resolve com-
pletely or incompletely within a few days of glucocor-
ticoid treatment [10]. This has two important clinical
implications. First, patients suspected of harboring a
primary CNS lymphoma should not receive glucocorti-
coids prior to biopsy, but rather receive osmotic agents,
should increased intracranial pressure necessitate ther-
apy. Second, if rapid clinical improvement in a patient
with one or more intracranial lesions is followed by
dramatic reduction or even disappearance of lesions
on CT or MRI scans, primary CNS lymphoma is the
most probable diagnosis. Inevitably, primary CNS lym-
phoma recurs despite glucocorticoid treatment, indi-
cating that complete tumor cell eradication is never
achieved.
Lymphoma cell resistance to glucocorticoids
The mechanisms underlying the changing pattern of
responsiveness to glucocorticoids from sensitive to
resistant in primary CNS lymphoma cells have not been
elucidated. It is possible that the treatment has selected
for resistant cells with altered glucocorticoid receptors
but this hypothesis has never been supported by exper-
imental or clinical data. Further, resistant cells may
have acquired a more global resistance to proapop-
totic stimuli, e.g., by virtue of enhanced expression
of antiapoptotic oncogenes such as bcl-2. Although
bcl-2 gene transfer enhances B lymphoma cell resis-
tance to glucocorticoid-induced apoptosis [9], there
is so far no evidence that primary CNS lymphomas
recurring after transient glucocorticoid-induced remis-
sion express more bcl-2 than prior to glucocorticoid
exposure [11]. Such changes would be associated
with enhanced resistance to chemotherapy and radio-
therapy as well. Yet, it is not known whether pro-
longed exposure to glucocorticoids modulates the
subsequent response to radiochemotherapy in primary
CNS lymphoma patients.
Long-term remissions of primary CNS
lymphoma with glucocorticoid treatment alone
A significant number of case observations suggest that
glucocorticoid treatment alone results in prolonged
local disease control as assessed by CT or MRI scans
or clinical monitoring [12,13]. Initially, the rates for
complete or partial CT-assessed remissions of primary
CNS lymphoma in response to dexamethasone are 15%
and 25%, respectively [14]. Such patients had often
received glucocorticoids prior to biopsy or tumor resec-
tion, the diagnosis of primary CNS lymphoma was
missed, and the patients were found to suffer from a
recurrent steroid-sensitive disease that sooner or later
turned out to be primary CNS lymphoma. Importantly,
such case histories can commonly not exclude that
the initial lesions were preneoplastic or inflamma-
tory rather than true primary CNS lymphoma. Thus,
since the diagnosis of primary CNS lymphoma usually
results in the initiation of radiochemotherapy, there is
not sufticient data on the course of proven primary CNS
lymphoma with glucocorticoid monotherapy. Recur-
rence of clinical signs in the initially steroid-responsive
patients developed most often, but not always, after glu-
cocorticoids had been withdrawn (reviewed in [12,13]),
suggesting that primary CNS lymphomas may acquire
resistance to glucocorticoids during chronic treatment
as well as during glucocorticoid withdrawal. The clini-
cal experiences summarized above raise the question as
to whether glucocorticoid monotherapy might be a use-
ful option for the management of subgroups of patients
with primary CNS lymphoma. Clearly, because of pre-
existing immunosuppression, long-term glucocorticoid
treatment would not seem to be an option for AIDS

patients or other immunosuppressed patients with
primary CNS lymphoma.
However, even in immunocompetent patients, the
possible benefit of glucocorticoid monotherapy com-
pared with radiochemotherapy must be weighed
against the almost inevitable severe side effects of
prolonged steroid therapy. Major adverse effects of
long-term glucocorticoid treatment include prolonged
immunosuppression, electrolyte disturbances, adrenal
insufficiency, steroid diabetes, osteoporosis, Cushing
syndrome, cataract formation, myopathy, and cognitive
dysfunction.
Conclusions
The lympholytic effects of glucocorticoids may
obscure the diagnosis of primary CNS lymphoma.
Therefore, these agents should be withheld until the
diagnosis can be established.
Once the diagnosis is established, the rapid onset
of action makes steroids a valuable addition to the
treatment options for patients with primary CNS lym-
phoma, notably in the presence of mass effect or in
clinically deteriorating patients.
Despite the casuistic long-term control of putative
primary CNS lymphoma with glucocorticoids alone,
this strategy must not be used in the significant pro-
portion of immunocompromised patients with primary
CNS lymphoma and is unlikely to be superior to cur-
rent chemoradiotherapy either in regard to long-term
side effects or in regard to therapeutic efficacy.
References
1. Bamberger CM, Schulte HM, Chrousos GP: Molecular
determinants of glucocorticoid receptor function and tis-
sue sensitivity to glucocorticoids. Endocr Rev 17: 245–261,
1996
2. Wyllie AH: Glucocorticoid-induced thymocyte apoptosis
is associated with endogenous endonuclease activation.
Nature 284: 555–556, 1980
3. Lowe SW, Schmitt EM, Smith SW, Osborne BA, Jacks T:
p53 is required for radiation-induced apoptosis in mouse
thymocytes. Nature 362: 847–849, 1993
239
4. Smith KGC, Strasser A, Vaux DL: CrmaA expression in
T lymphocytes of transgenic mice inhibits CD95 (Fas/
APO-1)-transduced apoptosis but does not cause lym-
phadenopathy or autoimmune disease. EMBO J 15: 5167–
5176, 1996
5. Geley S, Hartman BL, Kapelari K, Egle A, Villunger A,
Heidacher D, Greil R, Auer B, Kofler R: The interleukin-1β-
converting enzyme inhibitor CrmA prevents APO-1/Fas- but
not glucocorticoid-induced poly(ADP-ribose) polymerase
cleavage and apoptosis in lymphoblastic leukemia cells.
FEBS Lett 402: 36–40, 1997
6. Helmberg A, Auphan N, Caelles C, Karin M:
Glucocorticoid-induced apoptosis of human leukemic cells
is caused by the repressive function of the glucocorticoid
receptor. EMBO J 14: 452–460, 1995
7. Wisniewska M, Stanczyk M, Grzelakowska-Sztabert B,
Kamiska B: Nuclear factor of activated T cells (NFAT) is a
possible target for dexamethasone in thymocyte apoptosis.
Cell Biol Int 21: 127–132, 1997
8. Memon SA, Moreno MB, Petrak D, Zacharchuk CM:
Bcl-2 blocks glucocorticoid- but not Fas- or activation-
induced apoptosis in a T cell hybridoma. J Immunol 155:
4644–4652, 1995
9. Miyashita T, Reed JC: Bcl-2 gene transfer increases rela-
tive resistance of S49.1 and WEHI7.2 lymphoid cells to cell
death and DNA fragmentation induced by glucocorticoids
and multiple chemotherapeutic drugs. Cancer Res 52: 5407–
5411, 1992
10. Geppert M, Ostertag CB, Seitz G, Kiessling M: Glucocorti-
coid therapy obscures the diagnosis of cerebral lymphoma.
Acta Neuropathol 80: 629–634, 1990
11. Jellinger KA, Paulus W: Primary central nervous system
lymphomas – new pathological developments. J Neuro-
Oncol 24: 33–36, 1995
12. Herrlinger U, Schabet M, Eichhorn M, Petersen D, Grote
EH, Meyermann R, Dichgans J: Prolonged corticosteroid-
induced remission in primary central nervous system lym-
phoma: report of a case and review of the literature. Eur
Neurol 36: 241–243, 1996
13. Pirotte B, Levivier M, Goldman S, Brucher JM, Brotchi J,
Hildebrand J: Glucocorticoid-induced long-term remission
in primary cerebral lymphoma: case report and review of the
literature. J Neuro-Oncol 32: 63–69, 1997
14. DeAngelis LM, Yahalom J, Heinemann MH, Cirrincione
C, Thaler HT, Krol G: Primary CNS lymphoma: combined
treatment with chemotherapy and radiotherapy. Neurology
40: 80–86, 1990
Address for offprints: Michael Weller, Laboratory of Molecular
Neuro-Oncology, Department of Neurology, University
of Tübingen, Medical School, Hoppe-Seyler-Strasse 3,
72076 Tübingen, Germany; Tel.: (49)7071 2982141; Fax:
(49)7071 295260; E-mail: michael.weller@uni-tuebingen.de