Keywords
brain tumor - HIV - lymphoma - primary CNS lymphoma
Introduction
Primary central nervous system lymphoma (PCNSL) is a rare and aggressive extra nodal
non-Hodgkin's lymphoma (NHL).[1]
[2] PCNSL most commonly presents as an intracerebral mass; however, it may involve the
brain, spinal cord, leptomeninges, or eyes without evidence of systemic involvement
at the time of diagnosis.[2]
[3] With an overall incidence rate of 0.44 per 100,000 populations, it accounts for
2% of all primary CNS tumors.[4] The optimal treatment for PCNSL is still unclear.[5] New therapeutic approaches have markedly improved the survival; however, the optimal
management of this disease still poses a challenge in neuro-oncology.[6]
To date, an increasing number of cases have been reported from Western countries.
However, we did not find any report about PCNSL from middle east countries in our
review of the literature. In this study, we present our experience with management
of PCNSL in a tertiary care center in Iran and compare the clinical and paraclinical
features of our patients with those of the others.
Methods
In this retrospective study, patients with tissue diagnosis of PCNSL between January
2011 and December 2018 were studied. All patients were examined both by the neurosurgeons
and oncologist of our center. Patients underwent a complete evaluation for evidence
of extra-CNS disease including ophthalmological evaluation, serum and urine protein
electrophoresis, lumbar puncture, positron emission tomography (PET) scan, testicular
ultrasound in male patients, bone marrow biopsy, as well as thoraco-abdomino-pelvis
computed tomography (CT) imaging. Inclusion criteria were a diagnosis of PCNSL, confirmed
by histopathological evaluation and consistent longitudinal follow-up at our institution
for a minimum of 1 year.
All patients were treated with chemotherapy including intravenous high-dose methotrexate,
rituximab, and temozolomide and radiotherapy by the same oncologist (ME).
Medical records included age, sex, Karnofsky performance score (KPS), sign and symptoms,
ocular involvement, leptomeningeal involvement (on imaging and/or CSF cytology), laboratory
data (HIV tests, CSF cytology), findings of whole CNS CT scan and Magnetic resonance
imaging (MRI), and follow-up data (progression and survival). This study was conducted
in accordance with the Declaration of Helsinki and the ethics committee of our institution
approved the study.
Statistical analysis was performed using SPPS (IBM SPSS Statistics for Windows, Version
24.0. Armonk, NY: IBM Corp; 2016). Descriptive statistics were mean and standard deviation
(SD) for the quantitative variables and frequencies for the qualitative variables.
Comparisons between the categorical variables were performed by Pearson's chi-square
test and Fisher exact test. Mann–Whitney was applied for inter-group comparison of
continuous variables between the two groups.
Results
Fifty-eight patients with mean age of 48 years (range of 24 to 78 years) were enrolled
including 32 females (55.1%) and 26 males (44.9%). The mean fallow-up period was 40.2
months (ranging from 12 to 66 months).
The most common locations were frontal lobe (16, 27.6%) and periventricular area (15,
25.8%). The histologic diagnosis of lymphoma was confirmed in all cases via stereotactic
biopsy (55, 94.8%) or open surgery (3 [5.2%]: two with the impression of glioma and
one with uncal herniation needing emergent decompression). The most common histological
finding was diffuse large B-cell lymphoma (DLBCL), which was found in 56 patients
(96.5%), the two others had Burkitt's and T cell lymphoma. Contrast-enhanced brain
MRI was done in all patients that revealed multiple lesions in 7 patients (12%) and
one lesion in the remaining 51 (88%). Systemic evaluations did not reveal any additional
abnormality except positive HIV antibody detected in three patients (5.1%). All patients
underwent ophthalmologic examination for chorio-retinal involvement, which was observed
in three patients (5.1%). Six (10.3%) patients had positive CSF cytology for lymphoma
cells, 3 of them had also evidence of leptomeningeal enhancement over cerebral hemispheres.
Two of the 6 patients (33.3%) with positive CSF cytology had HIV infection and 4 cases
(66.6%) had multiple lesions in brain. Eight patients (13.7%) needed at least another
admission for disease progression: three of them died from disease progression; two
patients with Burkitt's (after 13 months) and T cell lymphoma (after 18 months) and
one with brain stem involvement (after 12 months), in the other five patients, the
disease progression was controlled after treatment. In comparison to the other patients,
the 8 patients who experienced progression had no significant difference in the rate
of positive CSF cytology (1 patient, 12.5%, vs. 5, 10%, p = 0.72) but had higher rate of HIV infection (2, 25%, vs. 1, 2%, p = 0.000) and multiple lesions on their initial MRI (3, 37.5% vs. 4, 8%, p = 0.001). The mean age of patients who had a progression was 60.8 ± 5.3 years as
compared to the others (45.9 ± 14.1 years) (p = 0.003).
The mean overall survival in this study was 37.4 ± 13.6 months and the mean progression-free
survival (PFS) was 35.1 ± 9.8 months. The mean time to progression was 15.2 ± 8.79
months (range: 8–26 months) among 8 patients who experienced progression in this series.
There was no mortality related to the surgical procedure (biopsy or craniotomy) in
this series. [Table 1] summarizes the clinical features and the patient outcomes in this study.
Table 1
Characteristic features of the patients
|
Variables
|
Number (%)
|
|
Age
|
< 50
|
38
|
|
50–70
|
17
|
|
> 70
|
3
|
|
Sex
|
Male
|
26 (44.9)
|
|
female
|
32 (55.1)
|
|
Karnofsky performance score (KPS)
|
< 70
|
3 (5.1)
|
|
> 70
|
55 (94.9)
|
|
Signs and symptoms
|
FND*
|
10 (17.2)
|
|
Seizure
|
19 (32.7)
|
|
RICP*
|
41 (70.7)
|
|
Others
|
23 (39.6)
|
|
Location of tumor
|
Lobar
|
28 (48.2)
|
|
Bi-hemispheric
|
7 (12.0)
|
|
Peri-ventricular
|
15 (25.8)
|
|
Brain stem
|
1 (1.7)
|
|
Multiple lesion
|
7 (12.0)
|
|
Multifocal
|
Yes
|
14 (24.1)
|
|
No
|
44 (75.9)
|
|
Ocular involvement
|
Yes
|
3 (5.1)
|
|
No
|
55 (94.9)
|
|
Leptomeningeal involvement
|
Yes
|
6 (10.3)
|
|
No
|
52 (89.7)
|
|
HIV positive
|
Yes
|
3 (5.1)
|
|
No
|
55 (94.9)
|
*FND: focal neurological deficit, RICP: raised intracranial pressure.
Discussion
The overall prognosis for PCNSL remains poor for the majority of cases despite remarkable
medical progression and less than 30% of patients can be cured successfully.[1]
[7] This is the first report of patients with PCNSL in the middle east according to
our review of literature.
The majorities (∼ 95%) of PCNSLs are classified as DLBCL, and the remaining cases
are low-grade B cell lymphoma, or rarely, T cell or Burkitt's lymphoma.[1] In a study on 167 patients with PCNSLs, Yuan et al reported the histological type
as DLBCL in 150 cases (89.8%).[1] In the study of Yadav et al, all 32 patients with PCNSL were reported to have DLBCL.[8] In our study, the most common histological type was DLBCL which was found in 56
patients (96.5%).
Although most patients are immunocompetent, human immunodeficiency virus (HIV) infection
and organ transplantation are considered as the predisposing factors for PCNSL. Epstein–Barr
virus (EBV) infection is also suggested to play a role in the oncogenesis among the
immunocompromised patients.[1]
[2] Dandachi et al in a retrospective study on 144 patients with PCNSL in 2019, reported
the HIV infection to be positive in 19% of the cases. Patients with HIV were more
likely to have multiple lesions, hemorrhage, and peripheral rim enhancement on imaging;
to receive palliative care or whole brain radiation; and less likely to receive chemotherapy.[5] In in our study, HIV infection was detected in three patients (5.1%) and all the
three survived until the end of the study and experienced no progression. In our study,
two of the three patients with HIV infection experienced disease progression, but
none of them died from disease progression. It seems that HIV infection is associated
with higher rate of progression/recurrence but does not affect the overall survival.
In our study, 10.3% of patients had positive CSF cytology for lymphoma cells, three
of them had also evidence of leptomeningeal enhancement over cerebral hemispheres.
Kiewe et al in 2010 reported 92 patients with PCNSL; median event-free survival (EFS)
of patients with meningeal dissemination (MD) was 26 months, of those without MD 34.1
months; median OS of these two patients' groups was 45.5 and 42.5 months, respectively.[9] In a study by Ghesquières et al, CSF cytology was not identified as a prognostic
factor for OS and they suggest that the evaluation of CSF provides no additional information
for survival although this test should be necessary to diagnose leptomeningeal involvement.[10] In our study, finding of a positive CSF cytology was more probable among HIV-infected
patients (2 out of 3) and those with multiple involvement site on brain MRI (4 out
of 7). Nevertheless, based on our results, only one of the six patients with positive
CSF cytology had a progressive disease, who had the features of HIV infection and
multiple brain lesions as well.
According to the review of literature, high-dose methotrexate (MTX)-based polychemotherapy
regimen prolongs the survival and is currently considered the most effective treatment
and the standard of care for patients with PCNSL.[5]
[11] The role of intrathecal chemotherapy and whole brain radiation therapy is controversy.[5] A population-based cohort of 189 patients with PCNSL in 2019, variable patterns
of first-line treatments with increasing use of rituximab and high-dose MTX-based
poly-chemotherapy after 2007, were shown to be associated with an increase in median
OS restricted to patients aged below 70 years.[12] Dalia et al reported that patients treated initially with high-dose MTX-based therapy
had improved overall and progression-free survival, with no improvement shown with
added radiation or intrathecal therapy.[3] In our study, all patients were treated with rituximab and high-dose MTX and the
mean overall survival and PFS in this study were 37.4 ± 13.6 and 35.1 ± 9.8 months,
respectively. The mean time to progression was 15.2 ± 8.79 months among eight patients
who experienced progression in this series. Yuan et al showed the median OS and PFS
were 37 months and 17 months, respectively, with a median follow-up of 25 months.[1] In a study on 89 Chinese patients with PCNSL, Liu et al reported a median OS of
45.3 months and median progression-free survival of 30.0 months.[13]
In a study by Ney et al, a retrospective review of 174 PCNSL patients aged older than
65 years was done; age and performance status were identified as the most important
predictors of survival.[14] Lin et al investigated the risk factors of early mortality (death within 60 days
after diagnosis) in 133 patients with PCNSL: approximately 9.8% had early mortality;
and in multivariate analysis, age 80 years or more and involvement of the basal ganglia
were identified as independent risk factors of early mortality, and MTX-based chemotherapy
served as an independent protective factor.[15] Yuan et al reported that residual tumor after operation, chemotherapy without HD-MTX,
and palliative treatment were independent prognostic markers and Eastern Cooperative
Oncology Group (ECOG) Performance Status more than 3, multifocal lesions, and palliative
treatment were revealed as unfavorable independent prognostic markers for PFS.[1] Ghesquière et al evaluated 132 patients with PCNSL with a median follow-up of 102
months and reported that the time-dependent effect of age, performance status and
tumor site lost their prognostic value after 6 months from diagnosis.[10] In our study, patients who experienced disease progression had statistically significant
older ages in comparison to the other patients (60.8 vs. 45.9 years).
Our study is limited by being retrospective and the number of the patients. In particular,
the number of the patients with HIV infection, positive CSF cytology and multifocal
lesions was not large enough and thus the comparisons made in this article need to
be proved in larger studies from similar populations.
Conclusion
Our study shows that HIV infection is less common among PCNSL patients in comparison
to the series coming from the western countries. Finding of a positive CSF cytology
was not linked to disease progression, while HIV infection and multifocal involvement
at initial presentation were strongly linked to a lower progression-free survival.
The single most important factor affecting the OS was the histopathologic type of
the PCNSL; two of the three patients who died from their disease in this series had
non-B cell PCNSL, whereas only one patient with DLBCL died because of brainstem involvement.
