576
Prophylactic Intrathecal Methotrexate and
Hydrocortisone Reduces Central Nervous System
Recurrence and Improves Survival in Aggressive
Non-Hodgkin Lymphoma
Naoto Tomita, M.D.1
Fumio Kodama, M.D.2
Heiwa Kanamori, M.D.1
Shigeki Motomura, M.D.2
Yoshiaki Ishigatsubo, M.D.1
1
First Department of Internal Medicine, Yokohama
City University School of Medicine, Yokohama, Ja-
pan.
2
Department of Chemotherapy, Kanagawa Cancer
Center, Yokohama, Japan.
Presented at the 43rd Annual Meeting of the Amer-
ican Society of Hematology (ASH), Orlando, Florida,
December 8, 2001.
Address for reprints: Naoto Tomita, M.D. First De-
partment of Internal Medicine, Yokohama City Uni-
versity School of Medicine, 3-9 Fukuura, Ka-
nazawa-ku, Yokohama 236-0004, Japan; Phone:
81-45-787-2630; Fax: 011-44-81-45-786-3444;
E-mail: cavalier@ch-yamate.dlenet.com
Received January 2, 2002; revision received
March 1, 2002; accepted March 7, 2002.
© 2002 American Cancer Society
BACKGROUND. Central nervous system (CNS) recurrence is almost invariably fatal
in patients with aggressive non-Hodgkin lymphoma (NHL). Although some proto-
cols are intended to prevent CNS disease, the value of CNS prophylaxis in patients
with aggressive NHL remains to be determined.
METHODS. We retrospectively analyzed a cohort of 68 adults with NHL who had
been treated uniformly with systemic chemotherapy and had attained complete
remission (CR) of disease. Patients ranged in age from 15 to 77 years (median, 56
years). Median follow-up after CR was 40 months. After CR was attained, 29
patients (Group A) received CNS prophylaxis consisting of four doses of intrathecal
methotrexate 10 mg/m2 and hydrocortisone 15 mg/m2 as soon as they could
tolerate it. The other 39 patients (Group B) did not receive CNS prophylaxis.
RESULTS. Although bulky mass (45% vs. 21%, P ⫽ 0.03) was more frequent in Group
A than in Group B, none of the patients in Group A experienced CNS recurrence
(0%), whereas CNS recurrence occurred in six patients in Group B (15%). This
difference was significant (P ⫽ 0.03). Multivariate logistic regression analysis for
CNS recurrence identified no CNS prophylaxis (P ⫽ 0.01) and bone marrow
involvement (P ⫽ 0.02) as independent predictors. Among patients without CNS
disease, systemic recurrence occurred in 5 patients in Group A and in 11 patients
in Group B (P ⫽ 0.12). The 5-year overall survival rate from CR was 80% in Group
A and 58% in Group B (P ⫽ 0.05). The 5-year recurrence-free survival rate from CR
was 85% in Group A and 51% in Group B (P ⫽ 0.01).
CONCLUSIONS. Prophylactic intrathecal methotrexate and hydrocortisone injection
reduces the incidence of CNS recurrence following CR in patients with aggressive
NHL and improves the chance of long-term survival. Cancer 2002;95:576 – 80.
© 2002 American Cancer Society.
DOI 10.1002/cncr.10699
KEYWORDS: aggressive non-Hodgkin lymphoma, central nervous system, prophy-
laxis, methotrexate.
C entral nervous system (CNS) recurrence of non-Hodgkin lym-
phoma (NHL) is almost always fatal with a rapid progression of
disease. Although CNS recurrence is rare in patients with low-grade
NHL, patients with lymphoblastic lymphoma and small noncleaved
cell lymphoma often are threatened with this type of recurrence.1,2
Neoplasia classified according to the working formulation3 (WF) as
aggressive lymphoma of intermediate grade or large cell immunoblas-
tic lymphoma shows an incidence of CNS recurrence of approxi-

mately 5% when no CNS prophylaxis is added to sys-
temic chemotherapy. Several risk factors for CNS
recurrence in patients with complete remission (CR)
of disease have been identified and efforts are under-
way to define patient subgroups for whom CNS pro-
phylaxis is warranted.4 –7 However, the effectiveness of
CNS prophylaxis remains to be clarified. Although re-
cent progress in systemic chemotherapy has raised the
CR rate and reduced systemic recurrences, control of
CNS recurrence remains important for improving the
results of chemotherapy. We retrospectively analyzed
therapeutic outcome in aggressive NHL patients with
or without CNS prophylaxis using intrathecal metho-
trexate (MTX) and hydrocortisone injection.
MATERIALS AND METHODS
The subjects of this study were 68 adult patients with
aggressive NHL diagnosed between 1991 and 2000 at
our institution and affiliated hospitals. Human immu-
nodeficiency virus lymphoma was not included in the
study. Patients’ age ranged from 15 to 77 years (me-
dian, 56). Median follow-up after CR was 40 months.
Histologic specimens were evaluated by expert pa-
thologists at each institution according to the WF clas-
sification. Tumor categories included diffuse small
cleaved cell, diffuse mixed cell, diffuse large cell, and
large cell immunoblastic lymphoma. For convenience,
a few cases of anaplastic large cell lymphomas were
grouped with difffuse large cell lymphoma. Clinical
staging was performed according to the Ann Arbor
system,8 based on a physical examination, computed
tomography (CT) scan of the chest, abdomen, and
pelvis, a gallium scintigram, endoscopic examination
of the upper gastrointestinal tract, and bone marrow
aspiration and biopsy. Patients with initial CNS in-
volvement were excluded from study.
All patients had received two or more courses of
an ACOMP-B induction regimen9 (MACOP-B-modi-
fied regimen) and achieved CR of disease. In general,
two courses of ACOMP-B and sequential consolidative
treatment including nine drugs10 (ML-Y9) were added
after CR of disease was achieved. The ACOMP-B reg-
imen was given every 4 weeks as follows: doxorubicin,
40 mg/m2 on Days 1 and 15; cyclophosphamide, 600
mg/m2 on Days 1 and 15; vincristine, 1.4 mg/m2 (max-
imum, 2.0 mg) on Days 1 and 15; MTX, 400 mg/m2 on
Day 8 followed by leucovorin rescue; prednisolone, 40
mg/m2 on Days 1–7; and bleomycin, 8 mg/m2 on Day
1. All drugs were given by intravenous injection except
for prednisolone, which was given orally. In patients
who initially had a bulky tumor mass, local irradiation
(40 Gy) was added after CR.
As soon as possible after CR was attained, 29 pa-
CNS Prophylaxis in Aggressive Lymphoma/Tomita et al. 577
tients (Group A) received CNS prophylaxis consisting
of four intrathecal doses of MTX (10 mg/m2) and hy-
drocortisone (15 mg/m2), ordinarily given twice a
week for 2 weeks. Systemic chemotherapy was given
simultaneously or sequentially. The other 39 patients
(Group B) did not receive CNS prophylaxis. The deci-
sion for giving CNS prophylactic chemotherapy or no
prophylaxis was not randomized, but depended on
the decision of each attending physician. Specifically,
we began to use CNS prophylaxis in 1992, mainly for
patients who initially had an advanced stage of dis-
ease. Therefore, most patients who received CNS pro-
phylaxis in this study had advanced disease.
Factors examined to compare patient background
between the two groups included gender, histology,
phenotype, age (older than 60 vs. 60 years old and
younger), serum lactate dehydrogenase (LDH) con-
centration (greater than the upper limit of normal vs.
equal to or less than normal), performance status (2– 4
vs. 0 –1), clinical stage according to the Ann Arbor
classification (III or IV vs. I or II), number of sites of
extra lymph node involvement (two or more vs.
fewer), the international prognostic index11 (IPI), pres-
ence versus absence of bone marrow involvement,
and presence or absence of a bulky mass, defined as a
tumor 10 cm or more in diameter. Received dose
intensity12 for each drug was calculated through the
period of ACOMP-B therapy in the two groups.
CNS recurrence was diagnosed when malignant
cells were detected in cytocentrifuged preparations of
cerebrospinal fluid or when an intracranial mass was
detected by CT scan or magnetic resonance imaging.
Epidural spinal cord compression was not considered
to be CNS recurrence in this study. Patients with
symptoms suggesting CNS disease but no other evi-
dence were not regarded as having CNS recurrence.
Statistical Analysis
The Fisher exact test, the chi-square test, and the
two-group Student t tests for unpaired data were used
to determine statistically significant differences be-
tween groups. The Mann–Whitney U test was used
between ordered groups. Multivariate logistic regres-
sion analysis was used to identify risk factors for CNS
recurrence. A survival curve was constructed using
Kaplan–Meier methods and groups were compared by
the Wilcoxon test. P ⬍ 0.05 indicated significance.
RESULTS
There were more female patients (P ⫽ 0.02) and pa-
tients with a bulky mass (P ⫽ 0.03) in Group A than in
Group B (Table 1). Although serum LDH concentra-
tion and clinical stage were higher in Group A, the
578 CANCER August 1, 2002 / Volume 95 / Number 3

TABLE 1 TABLE 2
Patient Characteristics Received Dose Intensity of Each Drug (mg/m2/week)a

CNS prophylaxis CNS prophylaxis

Group A (ⴙ) Group B (ⴚ) P valuea Group A Group B

(ⴙ) (ⴚ) P Valueb
Gender
Male 12 27 ADR 17.0 ⫾ 2.9 17.0 ⫾ 3.1 NS
Female 17 12 0.02 CPA 255.1 ⫾ 42.2 250.3 ⫾ 47.2 NS
Histology VCR 0.54 ⫾ 0.09 0.55 ⫾ 0.12 NS
DSC 3 6 MTX 79.1 ⫾ 16.4 78.6 ⫾ 23.9 NS
DM 3 1 BLM 1.65 ⫾ 0.68 1.55 ⫾ 0.57 NS
DL 19 29
IBL 4 3 NSb CNS: central nervous system; ADR: doxorubicin; CPA: cyclophosphamide; VCR: vincristine; MTX:
Phenotype methotrexate; BLM: bleomycin.
B 14 22 a
Received dose intensity of each drug indicated as mean ⫾ SD.
b
T/NK 6 9 Student t test for unpaired data.
ND 9 8 NS
Age (yrs)
⬎ 60 7 15 TABLE 3
ⱕ 60 22 24 NS CNS and Systemic Recurrence
LDH
Above normal 21 21 CNS prophylaxis
Normal or less than normal 8 18 0.10
Performance status Group A (ⴙ) Group B (ⴚ) P valuea
2–4 12 12
0–1 17 27 NS CNS recurrence (⫹) 0 6b
Stage CNS recurrence (⫺) 29 33 0.03
III, IV 25 26 Systemic recurrence (⫹) 5 11
I, II 4 13 0.06 Systemic recurrence (⫺) 24 22 0.12c
EN
ⱖ2 8 11 CNS: central nervous system.
a
ⱕ1 21 28 NS Fisher exact test.
b
International index Five isolated and one associated with simultaneous systemic recurrence.
c
Low 7 10 Patients with CNS recurrence were excluded from the examination.
Low-intermediate 9 13
High-intermediate 5 12
High 8 4 NSc TABLE 4
BM Covariates Associated with CNS Recurrence in Multivariate Logistic
Positive 6 5 Regression Analysis
Negative 23 34 NS
Bulky mass P value Relative risk (95% confidence interval)
Positive 13 8
Negative 16 31 0.03 CNS prophylaxis (⫹) 0.01 0.0000006 (0–⬁)
BM involvement (⫹) 0.02 60.9 (1.02–3645)
CNS: central nervous system; DSC: diffuse small cleaved cell; DM: diffuse mixed cell; DL: diffuse large
cell; IBL: large cell immunoblastic; NS: not significant; LDH: lactate dehydrogenase; EN: number of
CNS: central nervous system; BM: bone marrow.
extra lymph node sites; BM: bone marrow involvement.
a
Fisher exact test.
b
Chi-square test.
c
ated with simultaneous systemic recurrence; this case
Mann–Whitney U test.
was considered CNS rather than systemic recurrence
in the data analysis. Recurrence in the other five pa-
differences were not statistically significant. Received tients was limited to the CNS. In the six patients with
dose intensity of each drug included in the ACOMP-B CNS recurrence, the phenotype of lymphoma cell was
regimen showed no difference between the two B in three patients, T/NK in one patient, and not
groups (Table 2). determined in two patients. The initial IPI was low in
None of the patients in Group A developed CNS one patient, low–intermediate in two patients, high–
recurrence compared with six patients in Group B intermediate in two patients, and high in one patient.
(Table 3); this difference was significant (P ⫽ 0.03). In Times from CR to CNS recurrence were 1, 1, 2, 5, 13,
one patient in Group B, CNS recurrence was associ- and 56 months. Median survival from CNS recurrence

FIGURE 1. Kaplan–Meier curves for overall survival (A) and recurrence-free
survival (B) in patients with aggressive non-Hodgkin lymphoma who received
or did not receive central nervous system prophylaxis with intrathecal meth-
otrexate and hydrocortisone.
in these six patients was 6.5 months. Multivariate lo-
gistic regression analysis for risk factors of CNS recur-
rence identified absence of CNS prophylaxis (P ⫽ 0.01)
and presence of bone marrow involvement (P ⫽ 0.02)
as independent predictors (Table 4).
In the 62 patients without CNS recurrence, 16
systemic recurrences were seen (5 in Group A and 11
in Group B). The difference between the two groups
did not reach statistical significance (Table 3).
Overall survival from CR in Group A was signifi-
cantly better than in Group B (P ⫽ 0.05; Fig. 1A). The
5-year overall survival rate was 80% in Group A and
58% in Group B. The difference was even greater for
recurrence-free survival (P ⫽ 0.01; Fig. 1B). The 5-year
recurrence-free survival rate was 85% in Group A and
51% in Group B. The end-point was the first recur-
rence, whether CNS or systemic.
DISCUSSION
The role of CNS prophylaxis in patients with aggres-
sive NHL is still uncertain. Although many reports
CNS Prophylaxis in Aggressive Lymphoma/Tomita et al. 579
have described CNS involvement and initial risk fac-
tors in patients with aggressive NHL, subjects gener-
ally included not only those with CNS recurrence dur-
ing systemic CR but also CNS involvement in the
presence of active systemic disease. Analysis of the
efficacy of CNS prophylaxis must be limited to pa-
tients with CR of disease. We found four reports de-
scribing circumstances of CNS recurrence in patients
with CR of aggressive NHL. Van Besien et al.4 identi-
fied an elevated serum LDH concentration and in-
volvement of more than one extra lymph node site as
independent risk factors for CNS recurrence. Zinzani
et al.5 found only advanced stage to be a risk factor for
CNS recurrence. Bos et al.6 could not find a subgroup
for which CNS prophylactic treatment might be ex-
pected to provide substantial benefit. The patients in
these three studies did not receive any CNS prophy-
laxis. Haioun et al.7 reported a large cohort of 974
patients with CR of disease who had been treated
uniformly with the same systemic chemotherapeutic
regimen, including intravenous high-dose MTX and
CNS prophylaxis by intrathecal injection of MTX. Oc-
currence of isolated CNS recurrence according to IPI
was 0.6% in low and low–intermediate patients but
4.1% in high–intermediate and high patients. Haioun
et al. concluded that the prophylaxis notably reduced
the risk of CNS recurrence in higher-risk patients
(high–intermediate and high) compared with the re-
ported rates for patients who did not receive prophy-
laxis.4 – 6
In the current study, we retrospectively analyzed
therapeutic outcome in patients with aggressive NHL
with or without CNS prophylaxis using intrathecal
MTX and hydrocortisone. All patients had been
treated uniformly with systemic chemotherapy using
the ACOMP-B regimen and had attained CR of dis-
ease. CNS prophylactic chemotherapy or no prophy-
laxis depended on the decision of each attending phy-
sician. However, most patients receiving CNS
prophylaxis in this study had initial advanced stage,
because we considered intrathecal prophylactic che-
motherapy particularly important for these patients.
We also considered the dose intensity of each drug
given in ACOMP-B therapy and found it similarly dis-
tributed in the two groups. Although ACOMP-B induc-
tion included 400 mg/m2 of intravenous MTX per cy-
cle, this dose probably was too low for adequate CNS
prophylaxis.13 As for the background variables, female
gender was overrepresented among the 29 patients
with CNS prophylaxis, as was a bulky tumor, which
usually is considered to be a risk factor for CNS in-
volvement. In spite of this, all six CNS recurrences
occurred among patients without CNS prophylaxis,
580 CANCER August 1, 2002 / Volume 95 / Number 3
representing a significant difference. Moreover, mul-
tivariate analysis showed lack of CNS prophylaxis to
be the strongest independent predictor of CNS recur-
rence, followed by bone marrow involvement. This
indicates that CNS prophylaxis is beneficial for pa-
tients with aggressive NHL. The overall survival and
recurrence-free survival curves showed better out-
comes in the patient group with CNS prophylaxis.
When the six patients with CNS recurrence were cen-
sored from the curves at the time of recurrence, the
difference between the two groups was no longer sta-
tistically significant for both overall and recurrence-
free survival rates (data not shown), indicating that the
survival benefits were the result of CNS prophylaxis.
As for timing, CNS prophylaxis should be considered
as soon as possible after patients achieve CR of disease
because four of the six CNS recurrences in our study
occurred within 5 months from CR. We also examined
the possibility that intrathecal MTX injection for CNS
prophylaxis might penetrate the blood– brain barrier
and reduce systemic recurrence, but no significant
prevention of systemic recurrence was evident in the
62 patients without CNS recurrence.
We conclude that prophylactic intrathecal MTX
and hydrocortisone injection reduce the incidence of
CNS recurrence following CR in patients with aggres-
sive NHL and improves survival rates. Further ran-
domized controlled studies are warranted to define
subgroups with particular survival benefit from CNS
prophylaxis.
REFERENCES
1. Ersboll J, Schultz HB, Thomsen BL, Keiding N, Nissen NI.
Meningeal involvement in non-Hodgkin’s lymphoma:
symptoms, incidence, risk factors and treatment. Scand J
Haematol. 1985;35:487– 496.
2. Liang R, Chiu E, Loke SL. Secondary central nervous system
involvement by non-Hodgkin’s lymphoma: the risk factors.
Hematol Oncol. 1990;8:141–145.
3. The Non-Hodgkin’s Lymphoma Pathologic Classification
Project. National Cancer Institute sponsored study of clas-
sifications of non-Hodgkin’s lymphomas: summary and de-
scription of a working formulation for clinical usage. Cancer.
1982;49:2112–2135.
4. van Besien K, Ha CS, Murphy S, et al. Risk factors, treatment,
and outcome of central nervous system recurrence in adults
with intermediate-grade and immunoblastic lymphoma.
Blood. 1998;91:1178 –1184.
5. Zinzani PL, Magagnoli M, Frezza G, et al. Isolated central
nervous system relapse in aggressive non-Hodgkin’s lym-
phoma: the Bologna Experience. Leuk Lymph. 1999;32:571–
576.
6. Bos GM, van Putten WL, van der Holt B, van den Bent M,
Verdonck LF, Hagenbeek A. For which patients with aggres-
sive non-Hodgkin’s lymphoma is prophylaxis for central
nervous system disease mandatory? Ann Oncol. 1998;9:191–
194.
7. Haioun C, Besson C, Lepage E, et al. Incidence and risk
factors of central nervous system relapse in histologically
aggressive non-Hodgkin’s lymphoma uniformly treated
and receiving intrathecal central nervous system prophy-
laxis: A GELA study on 974 patients. Ann Oncol. 2000;11:
685– 690.
8. Carbone PP, Kaplan HS, Musshoff K, Smithers DW, Tubiana
M. Report of the committee on Hodgkin’s disease staging.
Cancer Res. 1971;31:1860 –1861.
9. Tomita N, Kodama F, Sakai R, et al. Predictive factors for
central nervous system involvement in non-Hodgkin’s lym-
phoma: significance of very high serum LDH concentra-
tions. Leuk Lymph. 2000;38:335–343.
10. Tomita N, Kodama F, Kanamori H, et al. Treatment out-
come in elderly patients with aggressive non-Hodgkin’s
lymphoma. Jpn J Clin Hematol. 2000;41:568 –575.
11. The International Non-Hodgkin’s Lymphoma Prognostic
Factors Project. A predictive model for aggressive non-
Hodgkin’s lymphoma. N Engl J Med. 1993;329:987–994.
12. Hryniuk W, Bush H. The importance of dose intensity in
chemotherapy of metastatic breast cancer. J Clin Oncol.
1984;2:1281–1288.
13. Bollen EL, Brouwer RE, Hamers S, et al. Central nervous
system relapse in non-Hodgkin’s lymphoma: a single-center
study of 532 patients. Arch Neurol. 1997;54:854 – 859.