Original Article
Survival in Granular Cell Astrocytomas Benjamin Voellger1, Jorge Humberto Tapia-Perez1, Rosita Rupa1 Christian Mawrin2 Elmar Kirches2 Thomas Schneider1 1 Department of Neurosurgery, Otto von Guericke-University,
Magdeburg, Germany 2 Department of Neuropathology, Otto von Guericke-University, Magdeburg, Germany Benjamin Voellger and Jorge Humberto Tapia-Perez contributed equally.
Dimitrios Karagiannis1
Address for correspondence Benjamin Voellger, Department of Neurosurgery, Otto von Guericke-University, Leipziger Str. 44, Magdeburg 39120, Germany (e-mail: [email protected]).
J Neurol Surg A 2015;76:30–38.
Abstract
Keywords
► granular cell astrocytoma ► adjuvant therapy ► survival
Background Granular cell astrocytomas (GCAs) are rarely encountered aggressive glial neoplasms. Treatment options comprise surgery, radiotherapy, and chemotherapy. Due to the small number of cases, a standard therapeutic regimen for GCA does not exist. Material and Methods We report on the case of a 64-year-old woman with GCA subjected to tumor biopsy followed by radiochemotherapy with temozolomide. We provide clinical, histopathologic, and magnetic resonance imaging findings as well as a complete follow-up. To assess the relation of age, gender, time of publication, and different treatment options with survival we performed log-rank tests and calculated Cox regression models and hazard ratios in data from all available reports on GCA. Results A significant difference in survival rates in favor of adjuvant therapy (radiotherapy or radiochemotherapy) at 12 months was found. Age > 70 years at the time of diagnosis had a significantly unfavorable impact on survival at 12 months. Although not statistically significant, a tendency toward higher probability of survival at 12 months was found in cases reported after 2002. In surgically treated patients, we could not find a significant impact of extent of resection on survival. A significant impact of gender on survival was not found. Conclusion Adjuvant therapy is significantly related to a higher probability of survival at 12 months and may therefore be recommended for patients with a GCA. Further analysis of these rare neoplasms is warranted.
Introduction Granular cell astrocytomas (GCAs) are rarely encountered glial neoplasms, first described in 1973.1 As of December 1, 2012, 63 cases, including our case, have been reported worldwide.1–27 GCAs do not exhibit unique magnetic resonance imaging (MRI) characteristics.20,25,28 GCAs predominantly arise within the parietal lobe (44%).25 Various authors have described the aggressive behavior of GCAs.20,28 GCAs do not always exhibit astrocytic features.25 In GCA, the astrocytic component of the tumor is supposed to determine the progression of the disease.20 In particular,
received October 11, 2013 accepted after revision March 3, 2014 published online July 29, 2014
high-grade CGAs have a worse prognosis as compared with other high-grade glial tumors.20 Histopathologic features of GCA include periodic acid-Schiff stain positive granular cells, intercellular reticulin, and lymphocytic infiltration.20 GCAs are reported to be positive for CD68, epithelial membrane antigen, S100, ubiquitin, lysosomal membrane glycoproteins (LAMP-1, LAMP-2), and, to some extent, for glial fibrillary acidic protein (GFAP), whereas Kiel67 (Ki-67) antigen and the corresponding molecular immunology Borstel (MIB)-1 antibody index are low and B cell lymphoma protein (BCL)-2 is negative.18–20,25,28 Higher frequencies of losses of heterozygosity as compared with other
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0034-1382781. ISSN 2193-6315.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
30
Survival in Granular Cell Astrocytomas
Voellger et al.
31
Table 1 Cases included in our statistical analysis on survival of patients with granular cell astrocytomas Gender
Markesbery et al
1
Ule et al2 Pialat et al3 Pasquier et al
4
Sakurama et al5 Sakurama et al5 Lechevalier et al
6
Hori et al7 Dickson et al
8
Age, y
Survival, mo
EOR
EOR confirmation NS
Adjuvant treatment
M
58
17
Partial
M
55
6
None
Radiotherapy
M
61
17
Partial
F
64
8
Gross total
NS
Radiotherapy
M
50
12
Partial
NS
Radiotherapy
M
37
24
Partial
NS
Radiotherapy
M
62
11
Partial
NS
Radiotherapy
F
56
0.25
Gross total
CCT
None
NS
Radiotherapy
None NS
None
F
27
12
Partial
Kornfeld
9
M
49
2
None
Kornfeld
9
M
46
8
Partial
NS
Radiotherapy
M
62
9
Partial
NS
Radiochemotherapy
M
41
15
Partial
NS
Radiochemotherapy
M
50
14
Partial
NS
Radiotherapy
F
46
6
Partial
NS
Radiotherapy
M
65
12
Biopsy
F
62
5
Partial
NS NS
Kawano et al10 Nakamura et al Claassen et al
11
12
Gambini et al13 Harris et al14 Albuquerque et al 16
15
None
Radiotherapy None
F
25
96
Partial
Snipes et al16
M
36
18
Biopsy
Radiotherapy
Melaragno et al17
M
60
4
Biopsy
Radiotherapy
Snipes et al
Geddes et al
a
F
57
12
Gross total
Geddes et ala
M
65
4
Biopsy
Radiotherapy
Geddes et ala
Radiotherapy
F
66
9
Biopsy
19
F
78
7
Partial
Chorny et al19
F
83
2
Biopsy
19
Chorny et al
NS
Radiotherapy
NS
None
Radiochemotherapy Radiotherapy
F
66
25
Partial
NS
Radiochemotherapy
Brat et al
b
F
49
51
Gross total
NS
Radiotherapy
Brat et al
b
M
69
3
Gross total
NS
Radiotherapy
Brat et alb
M
59
1
Partial
NS
None
Brat et alb
M
47
11
Partial
NS
Radiotherapy
b
M
29
3
Partial
NS
Radiotherapy
Brat et alb
M
72
10
Partial
NS
Radiotherapy
Brat et al
b
M
45
12
Partial
NS
Radiotherapy
Brat et al
b
M
49
1
Partial
NS
None
Brat et alb
M
60
5
Partial
NS
Radiochemotherapy
Brat et al
b
M
52
12
Partial
NS
Radiotherapy
Brat et al
b
F
55
7
Partial
NS
Radiochemotherapy
Brat et al
b
F
70
9
Partial
NS
Radiotherapy
Brat et alb
F
66
5
Gross total
NS
Radiotherapy
Brat et al
b
F
75
11
Partial
NS
Radiotherapy
Brat et al
b
M
75
1
Partial
NS
Radiotherapy
Brat et alb
M
68
9
Gross total
NS
None
b
M
72
9
Partial
NS
None
Chorny et al
Brat et al
Brat et al
(Continued)
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Study
Survival in Granular Cell Astrocytomas
Voellger et al.
Table 1 (Continued) Study
Gender b
Age, y
Survival, mo
EOR
EOR confirmation NS
M
47
7
Partial
Saad et al21
F
9
0.5
None
Shin et al22
F
28
10
Gross total
MRI NS
Brat et al
Kinjo et al
23
Adjuvant treatment Radiochemotherapy None None
M
68
6
Partial
Lee et al24
M
47
22
Biopsy
Schittenhelm and Psaras25
F
49
5
Gross total
MRI
Radiotherapy
Ishii et al
26
Radiotherapy Radiotherapy
F
56
12
Partial
NS
Radiochemotherapy
Joo et al27
F
55
18c
Gross total
NS
Radiochemotherapy
Our case
F
64
13
Biopsy
Radiochemotherapy
Abbreviations: CCT, cranial computerized tomography; EOR, extent of resection; F, female; M, male; mo, months; MRI, magnetic resonance imaging; NS, not stated; y, years. a Due to missing information in the remaining two cases, only cases 1, 4, and 5 from the cumulative report on five cases of granular astrocytoma (GCA) by Geddes et al18 were included in our analysis. b Due to missing information in the remaining four cases, only cases 1–9, 11, 13–15, 17–19, 21, and 22 from the cumulative report on 22 cases of GCA by Brat et al20 were included in our analysis. c Information on survival time was obtained through personal correspondence with Joo et al.27
astrocytomas of similar grades have been reported.29 Anaplastic features have been reported to occur in 61% of CGAs.25 Glioblastoma multiforme has repeatedly been reported to arise within GCAs.20,25 Differential diagnoses include other tumors of glial origin, lymphoma, demyelinating diseases, and infarction.20,28 Due to the malignant behavior of GCAs, aggressive multimodal therapy has been advocated.28 Therapeutic options include surgical tumor reduction, radiation, and chemotherapy.20,25,28 Considering the small number of only 63 cases, including our case, that have been reported since the first description of this entity in 1973, a prospective study on the impact of treatment modalities on survival in GCA appears unfeasible. We report on the case of a 64-year-old woman with GCA and provide a statistical analysis of the currently available literature on GCA (►Table 1) to assess the potential relations of age, gender, time of publication, and different treatment modalities with survival at 12 months in this rare neoplasm.
“granular cell astrocytoma” (“GCA”) consistently throughout this article. Patients with granular cell tumors of the spine or of the pituitary gland were not included. Age, gender, year of publication, treatment modalities, and survival times were recorded. We considered treatment modalities to be distinct as follows: no surgery, biopsy, partial resection, total resection, no adjuvant therapy, adjuvant radiotherapy, and adjuvant radiochemotherapy. We attempted to obtain missing information through personal correspondence with the author(s) of the respective case reports. Deadline was December 1, 2012. Statistical analysis was conducted using NeoOffice v.3.1.2 patch 9 (Planamesa, Santa Clara, California, United States) and the R software package v.2.10.1 (R Foundation, Vienna, Austria) on a Mac OS X 10.6.8 (Apple, Cupertino, California, United States) We performed log-rank tests to assess probabilities of survival at 12 months after diagnosis. Cox proportional hazard models and hazard ratios (HRs) at 12 months after diagnosis were calculated. The p values < 0.05 were considered statistically significant.
Results Material and Methods
Case Report
Case Report We report on the case of a 64-year-old woman with GCA who underwent tumor biopsy followed by radiochemotherapy with temozolomide (TMZ). We provide clinical, histopathologic, and MRI findings as well as a complete follow-up.
Survival Analysis To assess the relation of survival with age, gender, time of publication, and different treatment modalities, we retrieved the available reports on cases of GCA (►Table 1) through a PubMed database search (http://www.pubmed.org) using the keyword “granular cell brain tumor.” Although astrocytic features have not been reported in all cases included in our statistical analysis, we decided to name this tumor entity Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
A 64 year-old woman presented with a history of 14 days of progressive headache and impaired memory function. Neurologic examination did not reveal focal deficits or signs of increased intracranial pressure. Cranial MRI showed a bilateral periventricular space-occupying lesion, pronounced in the parietal and occipital lobes, more so on the left side (►Fig. 1). The localization and the homogeneous contrast media enhancement of the lesion (►Fig. 1) resembled a lymphoma. A stereotactic biopsy in the left occipital lobe was performed. Histopathologic features of the specimen included granular astrocytic cells (►Fig. 2A), positive staining for GFAP (►Fig. 2B), S100 and lymphocytic infiltrates positive for CD4, and less CD8. Staining for HMB45 was negative. CD68 immunostaining was strongly positive (►Fig. 2C). Analyses
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
32
Voellger et al.
Fig. 1 A 64 year-old woman presented with a history of 14 days of progressive headache and impaired memory function. Cranial magnetic resonance imaging revealed a space-occupying contrast-enhancing periventricular lesion predominantly in the left occipital lobe.
for isocitrate dehydrogenase (IDH)-1 R132H mutation and p53 were negative. A GCA was diagnosed. Early concomitant radiochemotherapy according to the Stupp protocol30 was initiated. The patient received 60 Gy irradiation with 30 fractions of 2 Gy, accompanied by 75 mg/m2 TMZ daily for 6 weeks, followed by three cycles of 150 to 200 mg/m2 TMZ daily for 5 days every 4 weeks. At a scheduled outpatient appointment at 4 months after diagnosis, the patient reported improvement of headache and memory function. Again, we did not detect focal deficits or signs of increased intracranial pressure. Cranial MRI showed a substantial regression of the lesion with a remnant in the left occipital lobe (►Fig. 3). At a scheduled outpatient appointment at 6 months after diagnosis, the patient reported worsening of headache and memory function. Neurologic examination revealed slight disorientation. Cranial MRI showed a new contrast-enhancing lesion in the pons and the floor of the fourth ventricle (►Fig. 4). A single-dose stereotactic irradiation with 15 Gy was conducted. The patient subsequently received dosedense (intensified) TMZ therapy (i.e., 150 mg/m2 TMZ daily every second week). At a scheduled outpatient appointment at 8 months after diagnosis, the patient reported further increase of headaches, accompanied by visual disturbances and nausea. Neurological examination revealed progressive disorientation, ptosis of the left upper eyelid and gait disturbance with a tendency of falling toward the left. Cranial MRI showed a regression of the lesion at the pons and the floor of the fourth ventricle but a new contrast enhancing lesion in the left frontal lobe and an increase in size of the ventricles and, to some extent, of the basal cerebrospinal fluid (CSF) cisterns (►Fig. 5A, B). To exclude symptomatic hydrocephalus, admission on our ward and insertion of an external ventricular drain for measurement of intracranial pressure and clinical testing was advised. We did not detect an increased intracranial pressure.
Fig. 2 Histopathologic specimen from a left occipital stereotactic biopsy of the lesion depicted in Fig. 1. (A) The specimen consisted of large granular astrocytic cells. (B) Staining for glial fibrillary acidic protein was positive. (C) Immunostaining for CD68 was strongly positive. A granular cell astrocytoma was diagnosed.
Disorientation and gait disturbance did not improve after drainage of CSF. A ventriculoperitoneal shunt was not advised. The patient continued to receive dose-dense TMZ until 12 months after diagnosis and died of the GCA at 13 months after diagnosis.
Survival Analysis Patients In our analysis of the literature, data on age, gender, time of publication, surgical procedures, extent of resection (EOR), Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
33
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Survival in Granular Cell Astrocytomas
Survival in Granular Cell Astrocytomas
Voellger et al. confirmation of EOR, adjuvant treatment, and survival were available in 52 of 63 cases including our case (►Table 1). We found a median follow-up of 9 months (range: 1 week to 96 months). At 12 months after diagnosis, 34 of 52 patients (65.4%) had died.
Adjuvant Therapy
Fig. 3 At 4 months, a substantial regression of the periventricular lesion was found after initiation of radiochemotherapy with temozolomide.
Radiotherapy alone has been used in 57.7% (30 of 52 patients). Radiotherapy combined with chemotherapy was used in 19.2% (10 of 52 patients). The reported radiochemotherapeutic regimens comprise nimustine in combination with 5bromo-2-deoxyuridine (BrdU) (one patient); nimustine in combination with vincristine, cisplatin, and bleomycin (one patient); lomustine (two patients); carmustine (two patients); carboplatin (one patient); and TMZ (three patients including our patient). Overall, 23.1% (12 of 52 patients) were reported not to have received adjuvant treatment. Survival rates at 12 months after diagnosis were 40.0% after radiotherapy alone, 40.0% after radiochemotherapy, and 16.7% without adjuvant treatment (►Table 1). Log-rank tests for comparison of the survival rates of patients without adjuvant treatment as opposed to patients who received any kind of adjuvant treatment (i.e., radiotherapy alone or radiochemotherapy) showed a significant difference in the probability of survival in favor of adjuvant treatment at 12 months (p ¼ 0.019; ►Table 2; ►Fig. 6). Further differentiation of adjuvant treatment modalities (radiotherapy alone versus radiochemotherapy) did not demonstrate a significant difference in survival (p ¼ 0.780; ►Table 2). An analysis of the relation of different chemotherapeutic agents with survival was not feasible due to the small numbers of individuals who received the respective drugs.
Age Median age at the time of diagnosis was 57 years (range: 9–83 years) (►Table 1). We found an unfavorable impact of age > 70 years at the time of diagnosis on survival, which was significant at 12 months after diagnosis (p ¼ 0.040; ►Table 2; ►Fig. 7). Fig. 4 At 6 months, a new contrast-enhancing lesion at the pons and the floor of the fourth ventricle was detected,
Fig. 5 (A, B) At 8 months, regression of the lesion at the pons and the floor of the fourth ventricle after stereotactic irradiation with 15 Gy and intensified temozolomide therapy was found, but a new contrast-enhancing lesion in the left frontal lobe was detected. Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
34
Survival in Granular Cell Astrocytomas
Voellger et al.
35
Table 2 The p values of differences in probabilities of survival at 12 months as related to age, gender, time of publication, and treatment modalities in patients with granular cell astrocytoma
Age > 70 years at diagnosis
0.040a
Gender
0.865
EOR, all patients (n ¼ 52)
0.004a
EOR, surgically treated patients (n ¼ 49)
0.468
Adjuvant therapy
0.019a
Radiotherapy
0.019a
Radiochemotherapy
0.439
Radiotherapy versus radiochemotherapy
0.780
Case reported before 1990
0.537
Case reported after 2002
0.179
Abbreviation: EOR, extent of resection. a Statistically significant.
Time of Publication The exact year of diagnosis remains unknown to us in almost all cases. We therefore used the year of publication as a surrogate parameter to assess the potential impact of general improvement of treatment over time on survival (►Table 1). Because the report of Brat et al20 on 22 cases of GCA, which were collected between 1990 and 2002, contributed 18 of 52 cases (34.6%) to our data, we subdivided patients into groups reported on before 1990 and after 2002. We did not find a
Fig. 7 Kaplan-Meier plots of probabilities of survival in patients with granular cell astrocytoma (GCA) who were 70 years of age at the time of diagnosis (black) as opposed to patients with GCA who were > 70 years of age at the time of diagnosis (gray). A significant difference in probabilities of survival in favor of the younger cohort was found at 12 months (p ¼ 0.040; Table 2); n ¼ 52 patients; þ censored; dashed lines representing survival curves as predicted from the Cox regression model in Table 3.
significant relation of the time of publication with probability of survival (p ¼ 0.179 and above; ►Table 2).
Extent of Resection
Fig. 6 Kaplan-Meier plots of probabilities of survival in patients with granular cell astrocytoma (GCA) who received adjuvant therapy (black) as opposed to patients with GCA who did not receive adjuvant therapy (gray). A significant difference in probabilities of survival in favor of adjuvant treatment was found at 12 months (p ¼ 0.019; Table 2; n ¼ 52 patients; þ censored; dashed lines representing survival curves as predicted from the Cox regression model in Table 3.
In 19.2% (10 of 52 patients), total resection was reported. In 59.6% (31 of 52 patients), partial resection was reported. Biopsy was reported in 15.4% (8 of 52 patients including our patient). Purely conservative treatment with the establishment of diagnosis at an autopsy was reported in 5.8% (three cases). In reports claiming total resection of the tumor, cranial computerized tomography for resection control was reported in 10.0% (1 of 10 patients); MRI for resection control was reported in another 20.0% (2 of 10 patients). In reports claiming partial resection of the tumor, use of an imaging modality for resection control was reported in 0 of 31 patients. Survival rates at 12 months after diagnosis were 30.0% after total resection, 32.3% after partial resection, 62.5% after biopsy, and 0% after purely conservative treatment (►Table 1). In a log-rank test including all patients (n ¼ 52), we found a favorable impact of EOR on survival that was significant at 12 months (p ¼ 0.004; ►Table 2). However, in a log-rank test confined to surgically treated patients (n ¼ 49), we did not find a significant impact of the reported EOR on survival at 12 months (p ¼ 0.468; ►Table 2).
Gender A total of 22 of 52 patients (42.3%) were female (►Table 1). We did not find a significant impact of gender on probability of survival (p ¼ 0.865; ►Table 2). Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
p value
Survival in Granular Cell Astrocytomas
Voellger et al.
Table 3 Small Cox regression model Coefficient
ecoefficient (hazard ratio)
Standard error of the coefficient
z value
p value
Age > 70 y at diagnosis
0.862
2.367
0.465
1.854
0.064
Adjuvant therapy
0.831
0.436
0.381
2.179
0.029a
Note: Cox regression model of survival in granular cell astrocytomas containing the variables “age over 70 years at diagnosis” and “adjuvant treatment”. Likelihood ratio test value of the model: 7.53; p value of the model: 0.023; degrees of freedom: 2. a Statistically significant.
Cox Regression Models Because at least two variables (age > 70 years at the time of diagnosis and adjuvant therapy) had a significant impact on survival according to the log-rank tests, two Cox proportional hazard models were fitted to allow for multivariate regression of probability of survival at 12 months, and the respective HRs were calculated (►Tables 3 and 4; ►Figs. 6 and 7). The best fitted Cox regression model included the variables “age over 70 years at the time of diagnosis” and “adjuvant therapy.” In this model, a reduction of 56% chance to die within the first year after diagnosis was found in case an adjuvant therapy (i.e., radiotherapy alone or radiochemotherapy) was applied (HR: 0.436; p ¼ 0.029; ►Table 3; ►Fig. 6), and age > 70 years at the time of diagnosis resulted in a greater than twofold chance to die within the first year after diagnosis (HR: 2.367; p ¼ 0.064; ►Table 3; ►Fig. 7). In a more comprehensive but less significant model (►Table 4), we assessed treatment modalities and other variables in detail, and here we found radiotherapy to reduce significantly the chance to die within the first year after diagnosis (HR: 0.402; p ¼ 0.025).
Discussion Our analysis, aiming to assess the relation of adjuvant treatment with survival in GCA, is based on 63 case reports including our case. In 52 cases, we were able to record data on age, gender, year of publication, treatment modalities, and survival (►Table 1). The rather small number of patients does not allow us to deliberately define subgroups of patients (e.g., to assess the
relation of dosage and timing of irradiation or the relation of distinct chemotherapeutic agents with survival). Attempts to assess statistically more than two subgroups of patients did not always return significant differences (►Table 2). Collecting the data, we had to rely on the written reports of and on personal communication with the respective author(s). Because we were not able to retrieve all missing information, some of the reported cases could not be included. Unreported cases have not been included at all. Data were analyzed retrospectively. As a consequence, potentially important details may have escaped our notice, and it may be questioned whether our study is actually representative. In summary, the conclusiveness of our findings may be limited due to the rarity of the disease and our retrospective secondary approach. Survival rates were not higher after radiochemotherapy as compared with radiotherapy alone, possibly due to the small number of cases. We could, however, demonstrate a significant relation of adjuvant therapy (i.e., radiotherapy alone or radiochemotherapy, as opposed to no adjuvant treatment) with a higher probability of survival at 12 months (p ¼ 0.019; ►Table 2; ►Fig. 6). A reduction of 56% chance to die within the first year after diagnosis was found in patients who had received any kind of adjuvant treatment. Therefore, adjuvant treatment (i.e., radiotherapy or radiochemotherapy) may be recommended in GCA. We found an age > 70 years at the time of diagnosis to be significantly related to a reduced probability of survival at 12 months (p ¼ 0.040; ►Table 2; ►Fig. 7). This finding may primarily reflect the increased probability to die due to risks associated with advanced age itself.
Table 4 Comprehensive Cox regression model Coefficient
ecoefficient (hazard ratio)
Standard error of the coefficient
z value
p value
Age > 70 y at diagnosis
0.752
2.121
0.469
1.602
0.110
Gender
0.183
1.201
0.390
0.469
0.640
Extent of resection
0.221
0.802
0.278
0.793
0.430
Radiotherapy
0.911
0.402
0.407
2.236
0.025a
Radiochemotherapy
0.081
1.084
0.502
0.161
0.870
Reported after 2002
0.750
0.473
0.655
1.145
0.250
Note: Cox regression model of survival in granular cell astrocytomas containing the variables “age over 70 years at diagnosis”, “gender”, “extent of resection”, “radiotherapy”, “radiochemotherapy”, and “reported after 2002”. Likelihood ratio test value of the model: 9.54; p value of the model: 0.146; degrees of freedom: 6. a Statistically significant. Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
36
Survival in Granular Cell Astrocytomas
Conclusion Adjuvant therapy, that is, radiotherapy or radiochemotherapy, is significantly related to a higher probability of survival at 12 months and may therefore be recommended for patients with GCAs. Further analysis of these rare neoplasms is warranted. Acknowledgment We are very grateful to the radiologists Mr. Peters, Helmstedt, Germany, and Dr. med. Ammari, Magdeburg, Germany for conducting and providing the MRI scans.
References 1 Markesbery WR, Duffy PE, Cowen D. Granular cell tumors of the
2
3 4
5
6
7
8
9
10
11
12
13
14
15
16
17
18 19
20
Conflict of Interest The authors have nothing to disclose.
37
central nervous system. J Neuropathol Exp Neurol 1973;32(1): 92–109 Ule G, Tschahargane C, Haag D, Berlet H, Volk B. [Malignant granular-cell tumor of the cerebral white matter: morphological, cytophotometrical and neurochemical studies (author’s transl)]. Acta Neuropathol 1975;32(2):143–155 Pialat J, Carrier H, Pierluca P, et al. Tumeur à cellules granuleuses de l’encéphale. Lyon Med 1978;239:269–274 Pasquier B, Pasquier D, N’Golet A, Panh MH, Couderc P. [Astrocytic origin of a granular cell tumor of the cerebral hemispheres. Preliminary study of a case (author’s transl)]. Sem Hop 1980; 56(17–18):891–892 Sakurama N, Matsukado Y, Marubayashi T, Kodama T. Granular cell tumour of the brain and its cellular identity. Acta Neurochir (Wien) 1981;56(1–2):81–94 Lechevalier B, Mandard JC, Adam Y, Da Silva DC, Bazin C, Courtheoux P. Granular cell tumor of a cerebral hemisphere: value of gliofibrillary protein acid assay. [in French]. Rev Neurol (Paris) 1982;138(8–9):619–629 Hori A, Altmannsberger M, Spoerri O, Beuche W. Granular cell tumour in the third ventricle. Case report with histological, electron-microscopic, immunohistochemical and necropsy findings. Acta Neurochir (Wien) 1985;74(1–2):49–52 Dickson DW, Suzuki KI, Kanner R, Weitz S, Horoupian DS. Cerebral granular cell tumor: immunohistochemical and electron microscopic study. J Neuropathol Exp Neurol 1986;45(3):304–314 Kornfeld M. Granular cell glioblastoma: a malignant granular cell neoplasm of astrocytic origin. J Neuropathol Exp Neurol 1986; 45(4):447–462 Kawano H, Hayashi M, Sato K, Hosotani K, Kubota T, Hirano A. Histological study of malignant cerebral granular cell tumor. [in Japanese]. No To Shinkei 1989;41(10):955–960 Nakamura T, Hirato J, Hotchi M, Kyoshima K, Nakamura Y. Astrocytoma with granular cell tumor-like changes. Report of a case with histochemical and ultrastructural characterization of granular cells. Acta Pathol Jpn 1990;40(3):206–211 Claassen U, Kuntz G, Schmitt HP. Malignant intracerebral granular cell tumor reacts positively with anti-alpha-1-antichymotrypsin and the MB2 antibody: a clue to the histogenesis of the brain granular cell? Clin Neuropathol 1990;9(2):82–88 Gambini C, Ruelle A, Palladino M, Boccardo M. Intracerebral granular cell tumor. Case report. Pathologica 1990;82(1077): 83–88 Harris CP, Townsend JJ, Brockmeyer DL, Heilbrun MP. Cerebral granular cell tumor occurring with glioblastoma multiforme: case report. Surg Neurol 1991;36(3):202–206 Albuquerque L, Pimentel J, Costa A, Cristina L. Cerebral granular cell tumors: report of a case and a note on their nature and expected behavior. Acta Neuropathol 1992;84(6):680–685 Snipes GJ, Horoupian DS, Shuer LM, Silverberg GD. Pleomorphic granular cell astrocytoma of the pineal gland. Cancer 1992;70(8): 2159–2165 Melaragno MJ, Prayson RA, Murphy MA, Hassenbusch SJ, Estes ML. Anaplastic astrocytoma with granular cell differentiation: case report and review of the literature. Hum Pathol 1993;24(7): 805–808 Geddes JF, Thom M, Robinson SF, Révész T. Granular cell change in astrocytic tumors. Am J Surg Pathol 1996;20(1):55–63 Chorny JA, Evans LC, Kleinschmidt-DeMasters BK. Cerebral granular cell astrocytomas: a Mib-1, bcl-2, and telomerase study. Clin Neuropathol 2000;19(4):170–179 Brat DJ, Scheithauer BW, Medina-Flores R, Rosenblum MK, Burger PC. Infiltrative astrocytomas with granular cell features (granular cell astrocytomas): a study of histopathologic features, grading, and outcome. Am J Surg Pathol 2002;26(6):750–757
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Although we saw a tendency toward higher probability of survival in cases reported after 2002 (p ¼ 0.179; HR: 0.473; ►Tables 2 and 4), the probability of survival was not proven to depend significantly on the time of publication. In our analysis, we did not find a significant relation of EOR with survival in surgically treated patients at 12 months (p ¼ 0.468; ►Table 2). This result may currently justify advising no extensive tumor resection but rather a (multimodal) adjuvant treatment as a consequence of histologic confirmation of GCA, as long as the tumor does not qualify for a glioblastoma. However, tumor characteristics indicative of a glioblastoma may be absent in tissue samples from biopsy or partial resection while potentially present in the remaining tissue. Moreover, the rarity of GCA and missing information on resection control in the overwhelming majority of the reported cases may well explain the absence of statistical significance regarding the relation of EOR with survival. There is strong evidence that EOR has a beneficial effect on progression-free survival and overall survival in high-grade gliomas31–33 as well as growing evidence that patients with lowgrade gliomas may also benefit from early and extensive tumor resection.34,35 Considering the rather aggressive behavior of GCA, one may therefore argue that extensive resection should be attempted in these tumors. This, however, is not always feasible due to anatomical or functional boundaries (e.g., due to the extension of tumor mass over the midline, as our case illustrates) (►Fig. 1). Information on methylguanine-o-methyltransferase (MGMT) promotor methylation, IDH-1 mutation status, or p53 expression has not been reported in cases of GCA before 2007 and could therefore be obtained in only 7 of 52 cases (13.5%) including our case.22–27 Hence an analysis of the relation of such tumor characteristics with the clinical course in GCA is currently infeasible. Because histologic features of GCA,20,25,28,29 the MRI morphology of the tumor in our case (►Figs. 1, 5), and the fact that we were unable to prove a statistically significant relation of EOR with survival at 12 months after diagnosis in GCA (►Table 2) indicate that these neoplasms may represent a separate tumor entity, further analysis of cases of GCA is warranted.
Voellger et al.
Survival in Granular Cell Astrocytomas
Voellger et al.
21 Saad A, Mo J, Miles L, Witte D. Granular cell astrocytoma of the
30 Stupp R, Mason WP, van den Bent MJ, et al; European Organisa-
cerebellum: report of the first case. Am J Clin Pathol 2006;126(4): 602–607 Shin E, Ki Chung C, Park SH. Granular cell astrocytoma. Pathol Res Pract 2007;203(1):57–62 Kinjo S, Yokoo H, Hirato J, Nakazato Y. Anaplastic astrocytoma with eosinophilic granular cells. Neuropathology 2007;27(5):457–462 Lee D, Suh YL, Nam H. Cerebral granular cell tumor. Neuropathology 2008;28(4):417–421 Schittenhelm J, Psaras T. Glioblastoma with granular cell astrocytoma features: a case report and literature review. Clin Neuropathol 2010;29(5):323–329 Ishii T, Mizukawa K, Sasayama T, et al. Immunohistochemical and molecular genetics study of a granular cell astrocytoma: a case report of malignant transformation to a glioblastoma. Neuropathology 2013;33(3):299–305 Joo M, Park SH, Chang SH, et al. Cytogenetic and molecular genetic study on granular cell glioblastoma: a case report. Hum Pathol 2013;44(2):282–288 Shi Y, Morgenstern N. Granular cell astrocytoma. Arch Pathol Lab Med 2008;132(12):1946–1950 Castellano-Sanchez AA, Ohgaki H, Yokoo H, et al. Granular cell astrocytomas show a high frequency of allelic loss but are not a genetically defined subset. Brain Pathol 2003;13(2):185–194
tion for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups; National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352(10): 987–996 Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen HJ; ALA-Glioma Study Group. Fluorescence-guided surgery with 5-aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. Lancet Oncol 2006; 7(5):392–401 Stummer W, Reulen HJ, Meinel T, et al; ALA-Glioma Study Group. Extent of resection and survival in glioblastoma multiforme: identification of and adjustment for bias. Neurosurgery 2008; 62(3):564–576; discussion 564–576 Pichlmeier U, Bink A, Schackert G, Stummer W; ALA Glioma Study Group. Resection and survival in glioblastoma multiforme: an RTOG recursive partitioning analysis of ALA study patients. Neuro Oncol 2008;10(6):1025–1034 Jakola AS, Myrmel KS, Kloster R, et al. Comparison of a strategy favoring early surgical resection vs a strategy favoring watchful waiting in low-grade gliomas. JAMA 2012;308(18):1881–1888 Sanai N, Chang S, Berger MS. Low-grade gliomas in adults. J Neurosurg 2011;115(5):948–965
22 23 24 25
26
27
28 29
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
31
32
33
34
35
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
38
Copyright of Journal of Neurological Surgery. Part A. Central European Neurosurgery is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Survival in Granular Cell Astrocytomas Benjamin Voellger1, Jorge Humberto Tapia-Perez1, Rosita Rupa1 Christian Mawrin2 Elmar Kirches2 Thomas Schneider1 1 Department of Neurosurgery, Otto von Guericke-University,
Magdeburg, Germany 2 Department of Neuropathology, Otto von Guericke-University, Magdeburg, Germany Benjamin Voellger and Jorge Humberto Tapia-Perez contributed equally.
Dimitrios Karagiannis1
Address for correspondence Benjamin Voellger, Department of Neurosurgery, Otto von Guericke-University, Leipziger Str. 44, Magdeburg 39120, Germany (e-mail: [email protected]).
J Neurol Surg A 2015;76:30–38.
Abstract
Keywords
► granular cell astrocytoma ► adjuvant therapy ► survival
Background Granular cell astrocytomas (GCAs) are rarely encountered aggressive glial neoplasms. Treatment options comprise surgery, radiotherapy, and chemotherapy. Due to the small number of cases, a standard therapeutic regimen for GCA does not exist. Material and Methods We report on the case of a 64-year-old woman with GCA subjected to tumor biopsy followed by radiochemotherapy with temozolomide. We provide clinical, histopathologic, and magnetic resonance imaging findings as well as a complete follow-up. To assess the relation of age, gender, time of publication, and different treatment options with survival we performed log-rank tests and calculated Cox regression models and hazard ratios in data from all available reports on GCA. Results A significant difference in survival rates in favor of adjuvant therapy (radiotherapy or radiochemotherapy) at 12 months was found. Age > 70 years at the time of diagnosis had a significantly unfavorable impact on survival at 12 months. Although not statistically significant, a tendency toward higher probability of survival at 12 months was found in cases reported after 2002. In surgically treated patients, we could not find a significant impact of extent of resection on survival. A significant impact of gender on survival was not found. Conclusion Adjuvant therapy is significantly related to a higher probability of survival at 12 months and may therefore be recommended for patients with a GCA. Further analysis of these rare neoplasms is warranted.
Introduction Granular cell astrocytomas (GCAs) are rarely encountered glial neoplasms, first described in 1973.1 As of December 1, 2012, 63 cases, including our case, have been reported worldwide.1–27 GCAs do not exhibit unique magnetic resonance imaging (MRI) characteristics.20,25,28 GCAs predominantly arise within the parietal lobe (44%).25 Various authors have described the aggressive behavior of GCAs.20,28 GCAs do not always exhibit astrocytic features.25 In GCA, the astrocytic component of the tumor is supposed to determine the progression of the disease.20 In particular,
received October 11, 2013 accepted after revision March 3, 2014 published online July 29, 2014
high-grade CGAs have a worse prognosis as compared with other high-grade glial tumors.20 Histopathologic features of GCA include periodic acid-Schiff stain positive granular cells, intercellular reticulin, and lymphocytic infiltration.20 GCAs are reported to be positive for CD68, epithelial membrane antigen, S100, ubiquitin, lysosomal membrane glycoproteins (LAMP-1, LAMP-2), and, to some extent, for glial fibrillary acidic protein (GFAP), whereas Kiel67 (Ki-67) antigen and the corresponding molecular immunology Borstel (MIB)-1 antibody index are low and B cell lymphoma protein (BCL)-2 is negative.18–20,25,28 Higher frequencies of losses of heterozygosity as compared with other
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0034-1382781. ISSN 2193-6315.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
30
Survival in Granular Cell Astrocytomas
Voellger et al.
31
Table 1 Cases included in our statistical analysis on survival of patients with granular cell astrocytomas Gender
Markesbery et al
1
Ule et al2 Pialat et al3 Pasquier et al
4
Sakurama et al5 Sakurama et al5 Lechevalier et al
6
Hori et al7 Dickson et al
8
Age, y
Survival, mo
EOR
EOR confirmation NS
Adjuvant treatment
M
58
17
Partial
M
55
6
None
Radiotherapy
M
61
17
Partial
F
64
8
Gross total
NS
Radiotherapy
M
50
12
Partial
NS
Radiotherapy
M
37
24
Partial
NS
Radiotherapy
M
62
11
Partial
NS
Radiotherapy
F
56
0.25
Gross total
CCT
None
NS
Radiotherapy
None NS
None
F
27
12
Partial
Kornfeld
9
M
49
2
None
Kornfeld
9
M
46
8
Partial
NS
Radiotherapy
M
62
9
Partial
NS
Radiochemotherapy
M
41
15
Partial
NS
Radiochemotherapy
M
50
14
Partial
NS
Radiotherapy
F
46
6
Partial
NS
Radiotherapy
M
65
12
Biopsy
F
62
5
Partial
NS NS
Kawano et al10 Nakamura et al Claassen et al
11
12
Gambini et al13 Harris et al14 Albuquerque et al 16
15
None
Radiotherapy None
F
25
96
Partial
Snipes et al16
M
36
18
Biopsy
Radiotherapy
Melaragno et al17
M
60
4
Biopsy
Radiotherapy
Snipes et al
Geddes et al
a
F
57
12
Gross total
Geddes et ala
M
65
4
Biopsy
Radiotherapy
Geddes et ala
Radiotherapy
F
66
9
Biopsy
19
F
78
7
Partial
Chorny et al19
F
83
2
Biopsy
19
Chorny et al
NS
Radiotherapy
NS
None
Radiochemotherapy Radiotherapy
F
66
25
Partial
NS
Radiochemotherapy
Brat et al
b
F
49
51
Gross total
NS
Radiotherapy
Brat et al
b
M
69
3
Gross total
NS
Radiotherapy
Brat et alb
M
59
1
Partial
NS
None
Brat et alb
M
47
11
Partial
NS
Radiotherapy
b
M
29
3
Partial
NS
Radiotherapy
Brat et alb
M
72
10
Partial
NS
Radiotherapy
Brat et al
b
M
45
12
Partial
NS
Radiotherapy
Brat et al
b
M
49
1
Partial
NS
None
Brat et alb
M
60
5
Partial
NS
Radiochemotherapy
Brat et al
b
M
52
12
Partial
NS
Radiotherapy
Brat et al
b
F
55
7
Partial
NS
Radiochemotherapy
Brat et al
b
F
70
9
Partial
NS
Radiotherapy
Brat et alb
F
66
5
Gross total
NS
Radiotherapy
Brat et al
b
F
75
11
Partial
NS
Radiotherapy
Brat et al
b
M
75
1
Partial
NS
Radiotherapy
Brat et alb
M
68
9
Gross total
NS
None
b
M
72
9
Partial
NS
None
Chorny et al
Brat et al
Brat et al
(Continued)
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Study
Survival in Granular Cell Astrocytomas
Voellger et al.
Table 1 (Continued) Study
Gender b
Age, y
Survival, mo
EOR
EOR confirmation NS
M
47
7
Partial
Saad et al21
F
9
0.5
None
Shin et al22
F
28
10
Gross total
MRI NS
Brat et al
Kinjo et al
23
Adjuvant treatment Radiochemotherapy None None
M
68
6
Partial
Lee et al24
M
47
22
Biopsy
Schittenhelm and Psaras25
F
49
5
Gross total
MRI
Radiotherapy
Ishii et al
26
Radiotherapy Radiotherapy
F
56
12
Partial
NS
Radiochemotherapy
Joo et al27
F
55
18c
Gross total
NS
Radiochemotherapy
Our case
F
64
13
Biopsy
Radiochemotherapy
Abbreviations: CCT, cranial computerized tomography; EOR, extent of resection; F, female; M, male; mo, months; MRI, magnetic resonance imaging; NS, not stated; y, years. a Due to missing information in the remaining two cases, only cases 1, 4, and 5 from the cumulative report on five cases of granular astrocytoma (GCA) by Geddes et al18 were included in our analysis. b Due to missing information in the remaining four cases, only cases 1–9, 11, 13–15, 17–19, 21, and 22 from the cumulative report on 22 cases of GCA by Brat et al20 were included in our analysis. c Information on survival time was obtained through personal correspondence with Joo et al.27
astrocytomas of similar grades have been reported.29 Anaplastic features have been reported to occur in 61% of CGAs.25 Glioblastoma multiforme has repeatedly been reported to arise within GCAs.20,25 Differential diagnoses include other tumors of glial origin, lymphoma, demyelinating diseases, and infarction.20,28 Due to the malignant behavior of GCAs, aggressive multimodal therapy has been advocated.28 Therapeutic options include surgical tumor reduction, radiation, and chemotherapy.20,25,28 Considering the small number of only 63 cases, including our case, that have been reported since the first description of this entity in 1973, a prospective study on the impact of treatment modalities on survival in GCA appears unfeasible. We report on the case of a 64-year-old woman with GCA and provide a statistical analysis of the currently available literature on GCA (►Table 1) to assess the potential relations of age, gender, time of publication, and different treatment modalities with survival at 12 months in this rare neoplasm.
“granular cell astrocytoma” (“GCA”) consistently throughout this article. Patients with granular cell tumors of the spine or of the pituitary gland were not included. Age, gender, year of publication, treatment modalities, and survival times were recorded. We considered treatment modalities to be distinct as follows: no surgery, biopsy, partial resection, total resection, no adjuvant therapy, adjuvant radiotherapy, and adjuvant radiochemotherapy. We attempted to obtain missing information through personal correspondence with the author(s) of the respective case reports. Deadline was December 1, 2012. Statistical analysis was conducted using NeoOffice v.3.1.2 patch 9 (Planamesa, Santa Clara, California, United States) and the R software package v.2.10.1 (R Foundation, Vienna, Austria) on a Mac OS X 10.6.8 (Apple, Cupertino, California, United States) We performed log-rank tests to assess probabilities of survival at 12 months after diagnosis. Cox proportional hazard models and hazard ratios (HRs) at 12 months after diagnosis were calculated. The p values < 0.05 were considered statistically significant.
Results Material and Methods
Case Report
Case Report We report on the case of a 64-year-old woman with GCA who underwent tumor biopsy followed by radiochemotherapy with temozolomide (TMZ). We provide clinical, histopathologic, and MRI findings as well as a complete follow-up.
Survival Analysis To assess the relation of survival with age, gender, time of publication, and different treatment modalities, we retrieved the available reports on cases of GCA (►Table 1) through a PubMed database search (http://www.pubmed.org) using the keyword “granular cell brain tumor.” Although astrocytic features have not been reported in all cases included in our statistical analysis, we decided to name this tumor entity Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
A 64 year-old woman presented with a history of 14 days of progressive headache and impaired memory function. Neurologic examination did not reveal focal deficits or signs of increased intracranial pressure. Cranial MRI showed a bilateral periventricular space-occupying lesion, pronounced in the parietal and occipital lobes, more so on the left side (►Fig. 1). The localization and the homogeneous contrast media enhancement of the lesion (►Fig. 1) resembled a lymphoma. A stereotactic biopsy in the left occipital lobe was performed. Histopathologic features of the specimen included granular astrocytic cells (►Fig. 2A), positive staining for GFAP (►Fig. 2B), S100 and lymphocytic infiltrates positive for CD4, and less CD8. Staining for HMB45 was negative. CD68 immunostaining was strongly positive (►Fig. 2C). Analyses
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
32
Voellger et al.
Fig. 1 A 64 year-old woman presented with a history of 14 days of progressive headache and impaired memory function. Cranial magnetic resonance imaging revealed a space-occupying contrast-enhancing periventricular lesion predominantly in the left occipital lobe.
for isocitrate dehydrogenase (IDH)-1 R132H mutation and p53 were negative. A GCA was diagnosed. Early concomitant radiochemotherapy according to the Stupp protocol30 was initiated. The patient received 60 Gy irradiation with 30 fractions of 2 Gy, accompanied by 75 mg/m2 TMZ daily for 6 weeks, followed by three cycles of 150 to 200 mg/m2 TMZ daily for 5 days every 4 weeks. At a scheduled outpatient appointment at 4 months after diagnosis, the patient reported improvement of headache and memory function. Again, we did not detect focal deficits or signs of increased intracranial pressure. Cranial MRI showed a substantial regression of the lesion with a remnant in the left occipital lobe (►Fig. 3). At a scheduled outpatient appointment at 6 months after diagnosis, the patient reported worsening of headache and memory function. Neurologic examination revealed slight disorientation. Cranial MRI showed a new contrast-enhancing lesion in the pons and the floor of the fourth ventricle (►Fig. 4). A single-dose stereotactic irradiation with 15 Gy was conducted. The patient subsequently received dosedense (intensified) TMZ therapy (i.e., 150 mg/m2 TMZ daily every second week). At a scheduled outpatient appointment at 8 months after diagnosis, the patient reported further increase of headaches, accompanied by visual disturbances and nausea. Neurological examination revealed progressive disorientation, ptosis of the left upper eyelid and gait disturbance with a tendency of falling toward the left. Cranial MRI showed a regression of the lesion at the pons and the floor of the fourth ventricle but a new contrast enhancing lesion in the left frontal lobe and an increase in size of the ventricles and, to some extent, of the basal cerebrospinal fluid (CSF) cisterns (►Fig. 5A, B). To exclude symptomatic hydrocephalus, admission on our ward and insertion of an external ventricular drain for measurement of intracranial pressure and clinical testing was advised. We did not detect an increased intracranial pressure.
Fig. 2 Histopathologic specimen from a left occipital stereotactic biopsy of the lesion depicted in Fig. 1. (A) The specimen consisted of large granular astrocytic cells. (B) Staining for glial fibrillary acidic protein was positive. (C) Immunostaining for CD68 was strongly positive. A granular cell astrocytoma was diagnosed.
Disorientation and gait disturbance did not improve after drainage of CSF. A ventriculoperitoneal shunt was not advised. The patient continued to receive dose-dense TMZ until 12 months after diagnosis and died of the GCA at 13 months after diagnosis.
Survival Analysis Patients In our analysis of the literature, data on age, gender, time of publication, surgical procedures, extent of resection (EOR), Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
33
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Survival in Granular Cell Astrocytomas
Survival in Granular Cell Astrocytomas
Voellger et al. confirmation of EOR, adjuvant treatment, and survival were available in 52 of 63 cases including our case (►Table 1). We found a median follow-up of 9 months (range: 1 week to 96 months). At 12 months after diagnosis, 34 of 52 patients (65.4%) had died.
Adjuvant Therapy
Fig. 3 At 4 months, a substantial regression of the periventricular lesion was found after initiation of radiochemotherapy with temozolomide.
Radiotherapy alone has been used in 57.7% (30 of 52 patients). Radiotherapy combined with chemotherapy was used in 19.2% (10 of 52 patients). The reported radiochemotherapeutic regimens comprise nimustine in combination with 5bromo-2-deoxyuridine (BrdU) (one patient); nimustine in combination with vincristine, cisplatin, and bleomycin (one patient); lomustine (two patients); carmustine (two patients); carboplatin (one patient); and TMZ (three patients including our patient). Overall, 23.1% (12 of 52 patients) were reported not to have received adjuvant treatment. Survival rates at 12 months after diagnosis were 40.0% after radiotherapy alone, 40.0% after radiochemotherapy, and 16.7% without adjuvant treatment (►Table 1). Log-rank tests for comparison of the survival rates of patients without adjuvant treatment as opposed to patients who received any kind of adjuvant treatment (i.e., radiotherapy alone or radiochemotherapy) showed a significant difference in the probability of survival in favor of adjuvant treatment at 12 months (p ¼ 0.019; ►Table 2; ►Fig. 6). Further differentiation of adjuvant treatment modalities (radiotherapy alone versus radiochemotherapy) did not demonstrate a significant difference in survival (p ¼ 0.780; ►Table 2). An analysis of the relation of different chemotherapeutic agents with survival was not feasible due to the small numbers of individuals who received the respective drugs.
Age Median age at the time of diagnosis was 57 years (range: 9–83 years) (►Table 1). We found an unfavorable impact of age > 70 years at the time of diagnosis on survival, which was significant at 12 months after diagnosis (p ¼ 0.040; ►Table 2; ►Fig. 7). Fig. 4 At 6 months, a new contrast-enhancing lesion at the pons and the floor of the fourth ventricle was detected,
Fig. 5 (A, B) At 8 months, regression of the lesion at the pons and the floor of the fourth ventricle after stereotactic irradiation with 15 Gy and intensified temozolomide therapy was found, but a new contrast-enhancing lesion in the left frontal lobe was detected. Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
34
Survival in Granular Cell Astrocytomas
Voellger et al.
35
Table 2 The p values of differences in probabilities of survival at 12 months as related to age, gender, time of publication, and treatment modalities in patients with granular cell astrocytoma
Age > 70 years at diagnosis
0.040a
Gender
0.865
EOR, all patients (n ¼ 52)
0.004a
EOR, surgically treated patients (n ¼ 49)
0.468
Adjuvant therapy
0.019a
Radiotherapy
0.019a
Radiochemotherapy
0.439
Radiotherapy versus radiochemotherapy
0.780
Case reported before 1990
0.537
Case reported after 2002
0.179
Abbreviation: EOR, extent of resection. a Statistically significant.
Time of Publication The exact year of diagnosis remains unknown to us in almost all cases. We therefore used the year of publication as a surrogate parameter to assess the potential impact of general improvement of treatment over time on survival (►Table 1). Because the report of Brat et al20 on 22 cases of GCA, which were collected between 1990 and 2002, contributed 18 of 52 cases (34.6%) to our data, we subdivided patients into groups reported on before 1990 and after 2002. We did not find a
Fig. 7 Kaplan-Meier plots of probabilities of survival in patients with granular cell astrocytoma (GCA) who were 70 years of age at the time of diagnosis (black) as opposed to patients with GCA who were > 70 years of age at the time of diagnosis (gray). A significant difference in probabilities of survival in favor of the younger cohort was found at 12 months (p ¼ 0.040; Table 2); n ¼ 52 patients; þ censored; dashed lines representing survival curves as predicted from the Cox regression model in Table 3.
significant relation of the time of publication with probability of survival (p ¼ 0.179 and above; ►Table 2).
Extent of Resection
Fig. 6 Kaplan-Meier plots of probabilities of survival in patients with granular cell astrocytoma (GCA) who received adjuvant therapy (black) as opposed to patients with GCA who did not receive adjuvant therapy (gray). A significant difference in probabilities of survival in favor of adjuvant treatment was found at 12 months (p ¼ 0.019; Table 2; n ¼ 52 patients; þ censored; dashed lines representing survival curves as predicted from the Cox regression model in Table 3.
In 19.2% (10 of 52 patients), total resection was reported. In 59.6% (31 of 52 patients), partial resection was reported. Biopsy was reported in 15.4% (8 of 52 patients including our patient). Purely conservative treatment with the establishment of diagnosis at an autopsy was reported in 5.8% (three cases). In reports claiming total resection of the tumor, cranial computerized tomography for resection control was reported in 10.0% (1 of 10 patients); MRI for resection control was reported in another 20.0% (2 of 10 patients). In reports claiming partial resection of the tumor, use of an imaging modality for resection control was reported in 0 of 31 patients. Survival rates at 12 months after diagnosis were 30.0% after total resection, 32.3% after partial resection, 62.5% after biopsy, and 0% after purely conservative treatment (►Table 1). In a log-rank test including all patients (n ¼ 52), we found a favorable impact of EOR on survival that was significant at 12 months (p ¼ 0.004; ►Table 2). However, in a log-rank test confined to surgically treated patients (n ¼ 49), we did not find a significant impact of the reported EOR on survival at 12 months (p ¼ 0.468; ►Table 2).
Gender A total of 22 of 52 patients (42.3%) were female (►Table 1). We did not find a significant impact of gender on probability of survival (p ¼ 0.865; ►Table 2). Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
p value
Survival in Granular Cell Astrocytomas
Voellger et al.
Table 3 Small Cox regression model Coefficient
ecoefficient (hazard ratio)
Standard error of the coefficient
z value
p value
Age > 70 y at diagnosis
0.862
2.367
0.465
1.854
0.064
Adjuvant therapy
0.831
0.436
0.381
2.179
0.029a
Note: Cox regression model of survival in granular cell astrocytomas containing the variables “age over 70 years at diagnosis” and “adjuvant treatment”. Likelihood ratio test value of the model: 7.53; p value of the model: 0.023; degrees of freedom: 2. a Statistically significant.
Cox Regression Models Because at least two variables (age > 70 years at the time of diagnosis and adjuvant therapy) had a significant impact on survival according to the log-rank tests, two Cox proportional hazard models were fitted to allow for multivariate regression of probability of survival at 12 months, and the respective HRs were calculated (►Tables 3 and 4; ►Figs. 6 and 7). The best fitted Cox regression model included the variables “age over 70 years at the time of diagnosis” and “adjuvant therapy.” In this model, a reduction of 56% chance to die within the first year after diagnosis was found in case an adjuvant therapy (i.e., radiotherapy alone or radiochemotherapy) was applied (HR: 0.436; p ¼ 0.029; ►Table 3; ►Fig. 6), and age > 70 years at the time of diagnosis resulted in a greater than twofold chance to die within the first year after diagnosis (HR: 2.367; p ¼ 0.064; ►Table 3; ►Fig. 7). In a more comprehensive but less significant model (►Table 4), we assessed treatment modalities and other variables in detail, and here we found radiotherapy to reduce significantly the chance to die within the first year after diagnosis (HR: 0.402; p ¼ 0.025).
Discussion Our analysis, aiming to assess the relation of adjuvant treatment with survival in GCA, is based on 63 case reports including our case. In 52 cases, we were able to record data on age, gender, year of publication, treatment modalities, and survival (►Table 1). The rather small number of patients does not allow us to deliberately define subgroups of patients (e.g., to assess the
relation of dosage and timing of irradiation or the relation of distinct chemotherapeutic agents with survival). Attempts to assess statistically more than two subgroups of patients did not always return significant differences (►Table 2). Collecting the data, we had to rely on the written reports of and on personal communication with the respective author(s). Because we were not able to retrieve all missing information, some of the reported cases could not be included. Unreported cases have not been included at all. Data were analyzed retrospectively. As a consequence, potentially important details may have escaped our notice, and it may be questioned whether our study is actually representative. In summary, the conclusiveness of our findings may be limited due to the rarity of the disease and our retrospective secondary approach. Survival rates were not higher after radiochemotherapy as compared with radiotherapy alone, possibly due to the small number of cases. We could, however, demonstrate a significant relation of adjuvant therapy (i.e., radiotherapy alone or radiochemotherapy, as opposed to no adjuvant treatment) with a higher probability of survival at 12 months (p ¼ 0.019; ►Table 2; ►Fig. 6). A reduction of 56% chance to die within the first year after diagnosis was found in patients who had received any kind of adjuvant treatment. Therefore, adjuvant treatment (i.e., radiotherapy or radiochemotherapy) may be recommended in GCA. We found an age > 70 years at the time of diagnosis to be significantly related to a reduced probability of survival at 12 months (p ¼ 0.040; ►Table 2; ►Fig. 7). This finding may primarily reflect the increased probability to die due to risks associated with advanced age itself.
Table 4 Comprehensive Cox regression model Coefficient
ecoefficient (hazard ratio)
Standard error of the coefficient
z value
p value
Age > 70 y at diagnosis
0.752
2.121
0.469
1.602
0.110
Gender
0.183
1.201
0.390
0.469
0.640
Extent of resection
0.221
0.802
0.278
0.793
0.430
Radiotherapy
0.911
0.402
0.407
2.236
0.025a
Radiochemotherapy
0.081
1.084
0.502
0.161
0.870
Reported after 2002
0.750
0.473
0.655
1.145
0.250
Note: Cox regression model of survival in granular cell astrocytomas containing the variables “age over 70 years at diagnosis”, “gender”, “extent of resection”, “radiotherapy”, “radiochemotherapy”, and “reported after 2002”. Likelihood ratio test value of the model: 9.54; p value of the model: 0.146; degrees of freedom: 6. a Statistically significant. Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
36
Survival in Granular Cell Astrocytomas
Conclusion Adjuvant therapy, that is, radiotherapy or radiochemotherapy, is significantly related to a higher probability of survival at 12 months and may therefore be recommended for patients with GCAs. Further analysis of these rare neoplasms is warranted. Acknowledgment We are very grateful to the radiologists Mr. Peters, Helmstedt, Germany, and Dr. med. Ammari, Magdeburg, Germany for conducting and providing the MRI scans.
References 1 Markesbery WR, Duffy PE, Cowen D. Granular cell tumors of the
2
3 4
5
6
7
8
9
10
11
12
13
14
15
16
17
18 19
20
Conflict of Interest The authors have nothing to disclose.
37
central nervous system. J Neuropathol Exp Neurol 1973;32(1): 92–109 Ule G, Tschahargane C, Haag D, Berlet H, Volk B. [Malignant granular-cell tumor of the cerebral white matter: morphological, cytophotometrical and neurochemical studies (author’s transl)]. Acta Neuropathol 1975;32(2):143–155 Pialat J, Carrier H, Pierluca P, et al. Tumeur à cellules granuleuses de l’encéphale. Lyon Med 1978;239:269–274 Pasquier B, Pasquier D, N’Golet A, Panh MH, Couderc P. [Astrocytic origin of a granular cell tumor of the cerebral hemispheres. Preliminary study of a case (author’s transl)]. Sem Hop 1980; 56(17–18):891–892 Sakurama N, Matsukado Y, Marubayashi T, Kodama T. Granular cell tumour of the brain and its cellular identity. Acta Neurochir (Wien) 1981;56(1–2):81–94 Lechevalier B, Mandard JC, Adam Y, Da Silva DC, Bazin C, Courtheoux P. Granular cell tumor of a cerebral hemisphere: value of gliofibrillary protein acid assay. [in French]. Rev Neurol (Paris) 1982;138(8–9):619–629 Hori A, Altmannsberger M, Spoerri O, Beuche W. Granular cell tumour in the third ventricle. Case report with histological, electron-microscopic, immunohistochemical and necropsy findings. Acta Neurochir (Wien) 1985;74(1–2):49–52 Dickson DW, Suzuki KI, Kanner R, Weitz S, Horoupian DS. Cerebral granular cell tumor: immunohistochemical and electron microscopic study. J Neuropathol Exp Neurol 1986;45(3):304–314 Kornfeld M. Granular cell glioblastoma: a malignant granular cell neoplasm of astrocytic origin. J Neuropathol Exp Neurol 1986; 45(4):447–462 Kawano H, Hayashi M, Sato K, Hosotani K, Kubota T, Hirano A. Histological study of malignant cerebral granular cell tumor. [in Japanese]. No To Shinkei 1989;41(10):955–960 Nakamura T, Hirato J, Hotchi M, Kyoshima K, Nakamura Y. Astrocytoma with granular cell tumor-like changes. Report of a case with histochemical and ultrastructural characterization of granular cells. Acta Pathol Jpn 1990;40(3):206–211 Claassen U, Kuntz G, Schmitt HP. Malignant intracerebral granular cell tumor reacts positively with anti-alpha-1-antichymotrypsin and the MB2 antibody: a clue to the histogenesis of the brain granular cell? Clin Neuropathol 1990;9(2):82–88 Gambini C, Ruelle A, Palladino M, Boccardo M. Intracerebral granular cell tumor. Case report. Pathologica 1990;82(1077): 83–88 Harris CP, Townsend JJ, Brockmeyer DL, Heilbrun MP. Cerebral granular cell tumor occurring with glioblastoma multiforme: case report. Surg Neurol 1991;36(3):202–206 Albuquerque L, Pimentel J, Costa A, Cristina L. Cerebral granular cell tumors: report of a case and a note on their nature and expected behavior. Acta Neuropathol 1992;84(6):680–685 Snipes GJ, Horoupian DS, Shuer LM, Silverberg GD. Pleomorphic granular cell astrocytoma of the pineal gland. Cancer 1992;70(8): 2159–2165 Melaragno MJ, Prayson RA, Murphy MA, Hassenbusch SJ, Estes ML. Anaplastic astrocytoma with granular cell differentiation: case report and review of the literature. Hum Pathol 1993;24(7): 805–808 Geddes JF, Thom M, Robinson SF, Révész T. Granular cell change in astrocytic tumors. Am J Surg Pathol 1996;20(1):55–63 Chorny JA, Evans LC, Kleinschmidt-DeMasters BK. Cerebral granular cell astrocytomas: a Mib-1, bcl-2, and telomerase study. Clin Neuropathol 2000;19(4):170–179 Brat DJ, Scheithauer BW, Medina-Flores R, Rosenblum MK, Burger PC. Infiltrative astrocytomas with granular cell features (granular cell astrocytomas): a study of histopathologic features, grading, and outcome. Am J Surg Pathol 2002;26(6):750–757
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Although we saw a tendency toward higher probability of survival in cases reported after 2002 (p ¼ 0.179; HR: 0.473; ►Tables 2 and 4), the probability of survival was not proven to depend significantly on the time of publication. In our analysis, we did not find a significant relation of EOR with survival in surgically treated patients at 12 months (p ¼ 0.468; ►Table 2). This result may currently justify advising no extensive tumor resection but rather a (multimodal) adjuvant treatment as a consequence of histologic confirmation of GCA, as long as the tumor does not qualify for a glioblastoma. However, tumor characteristics indicative of a glioblastoma may be absent in tissue samples from biopsy or partial resection while potentially present in the remaining tissue. Moreover, the rarity of GCA and missing information on resection control in the overwhelming majority of the reported cases may well explain the absence of statistical significance regarding the relation of EOR with survival. There is strong evidence that EOR has a beneficial effect on progression-free survival and overall survival in high-grade gliomas31–33 as well as growing evidence that patients with lowgrade gliomas may also benefit from early and extensive tumor resection.34,35 Considering the rather aggressive behavior of GCA, one may therefore argue that extensive resection should be attempted in these tumors. This, however, is not always feasible due to anatomical or functional boundaries (e.g., due to the extension of tumor mass over the midline, as our case illustrates) (►Fig. 1). Information on methylguanine-o-methyltransferase (MGMT) promotor methylation, IDH-1 mutation status, or p53 expression has not been reported in cases of GCA before 2007 and could therefore be obtained in only 7 of 52 cases (13.5%) including our case.22–27 Hence an analysis of the relation of such tumor characteristics with the clinical course in GCA is currently infeasible. Because histologic features of GCA,20,25,28,29 the MRI morphology of the tumor in our case (►Figs. 1, 5), and the fact that we were unable to prove a statistically significant relation of EOR with survival at 12 months after diagnosis in GCA (►Table 2) indicate that these neoplasms may represent a separate tumor entity, further analysis of cases of GCA is warranted.
Voellger et al.
Survival in Granular Cell Astrocytomas
Voellger et al.
21 Saad A, Mo J, Miles L, Witte D. Granular cell astrocytoma of the
30 Stupp R, Mason WP, van den Bent MJ, et al; European Organisa-
cerebellum: report of the first case. Am J Clin Pathol 2006;126(4): 602–607 Shin E, Ki Chung C, Park SH. Granular cell astrocytoma. Pathol Res Pract 2007;203(1):57–62 Kinjo S, Yokoo H, Hirato J, Nakazato Y. Anaplastic astrocytoma with eosinophilic granular cells. Neuropathology 2007;27(5):457–462 Lee D, Suh YL, Nam H. Cerebral granular cell tumor. Neuropathology 2008;28(4):417–421 Schittenhelm J, Psaras T. Glioblastoma with granular cell astrocytoma features: a case report and literature review. Clin Neuropathol 2010;29(5):323–329 Ishii T, Mizukawa K, Sasayama T, et al. Immunohistochemical and molecular genetics study of a granular cell astrocytoma: a case report of malignant transformation to a glioblastoma. Neuropathology 2013;33(3):299–305 Joo M, Park SH, Chang SH, et al. Cytogenetic and molecular genetic study on granular cell glioblastoma: a case report. Hum Pathol 2013;44(2):282–288 Shi Y, Morgenstern N. Granular cell astrocytoma. Arch Pathol Lab Med 2008;132(12):1946–1950 Castellano-Sanchez AA, Ohgaki H, Yokoo H, et al. Granular cell astrocytomas show a high frequency of allelic loss but are not a genetically defined subset. Brain Pathol 2003;13(2):185–194
tion for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups; National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352(10): 987–996 Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen HJ; ALA-Glioma Study Group. Fluorescence-guided surgery with 5-aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. Lancet Oncol 2006; 7(5):392–401 Stummer W, Reulen HJ, Meinel T, et al; ALA-Glioma Study Group. Extent of resection and survival in glioblastoma multiforme: identification of and adjustment for bias. Neurosurgery 2008; 62(3):564–576; discussion 564–576 Pichlmeier U, Bink A, Schackert G, Stummer W; ALA Glioma Study Group. Resection and survival in glioblastoma multiforme: an RTOG recursive partitioning analysis of ALA study patients. Neuro Oncol 2008;10(6):1025–1034 Jakola AS, Myrmel KS, Kloster R, et al. Comparison of a strategy favoring early surgical resection vs a strategy favoring watchful waiting in low-grade gliomas. JAMA 2012;308(18):1881–1888 Sanai N, Chang S, Berger MS. Low-grade gliomas in adults. J Neurosurg 2011;115(5):948–965
22 23 24 25
26
27
28 29
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
31
32
33
34
35
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
38
Copyright of Journal of Neurological Surgery. Part A. Central European Neurosurgery is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
