J Neurooncol DOI 10.1007/s11060-014-1447-9
TOPIC REVIEW
Brain metastasis from ovarian cancer: a systematic review Shabnam Pakneshan • Damoun Safarpour Fattaneh Tavassoli • Bahman Jabbari
•
Received: 27 January 2014 / Accepted: 13 April 2014 Ó Springer Science+Business Media New York 2014
Abstract To review the existing literature on brain metastasis (BM) from ovarian cancer and to assess the frequency, anatomical, clinical and paraclinical information and factors associated with prognosis. Ovarian cancer is a rare cause of brain metastasis with a recently reported increasing prevalence. Progressive neurologic disability and poor prognosis is common. A comprehensive review on this subject has not been published previously. This systematic literature search used the Pubmed and Yale library. A total of 66 publications were found, 57 of which were used representing 591 patients with BM from ovarian cancer. The median age of the patients was 54.3 years (range 20-81). A majority of patients (57.3 %) had multiple brain lesions. The location of the lesion was cerebellar (30 %), frontal (20 %), parietal (18 %) and occipital (11 %). Extracranial metastasis was present in 49.8 % of cases involving liver (20.7 %), lung (20.4 %), lymph nodes (12.6 %), bones (6.6 %) and pelvic organs (4.3 %). The most common symptoms were weakness (16 %), seizures (11 %), altered mentality (11 %) visual disturbances (9 %) and dizziness (8 %). The interval from diagnosis of breast cancer to BM ranged from 0 to 133 months (median 24 months) and median survival was 8.2 months. Local radiation, surgical resection, stereotactic radiosurgery and medical therapy were used. Factors that significantly increased the survival were younger age at the time of ovarian cancer diagnosis and brain metastasis diagnosis, lower grade of the primary tumor, higher KPS score and multimodality treatment for the brain metastases. Ovarian cancer is a rare cause of brain metastasis. Development of brain metastasis among older patients and lower KPS score correlate with less favorable prognosis. The S. Pakneshan (&) D. Safarpour F. Tavassoli B. Jabbari Department of Neurology, Yale University School of Medicine, 15 York Street, LCI Building, New Haven, CT 06520, USA e-mail: [email protected]
more prolonged survival after using multimodality treatment for brain metastasis is important due to potential impact on management of brain metastasis in future. Keywords Ovarian cancer Brain metastasis Survival Radiation Chemotherapy Multimodality treatment CA-125 Serouse cystadenocarcinoma
Introduction Brain metastasis, although a common and severe complication in lung and breast cancer [1, 2], is considered a rare and late event in ovarian cancer [2–4]. A variable incidence of 0.3–12 % [5–11] has been reported in different studies as well as its association with poor prognosis [12]. The reported incidence of brain metastasis is higher in more recent studies [3, 4] possibly reflecting advances in imaging and longer survival from primary ovarian cancer [13–21]. The rarity and small number of patients affected, have prevented the establishment of a consensus for optimal therapy [3]. The increasing incidence of brain metastasis in ovarian cancer, however, mandates special attention [22] since the treatment strategy clearly affects the prognosis [22–25]. We conducted a comprehensive review of the existing literature on this subject in order to create a database to recognize and compare different patient characteristics, treatment modalities and prognostic factors that may impact the ultimate outcome and survival.
Materials and methods A total of 66 articles published between 1978 and 2013 were reviewed following a Yale web based search of
123
J Neurooncol
Medline, PubMed, Ovid and other medical databases. Nine manuscripts were excluded because either they did not separate ovarian malignancy from other gynecologic cancers or they did not separate brain metastasis from metastasis to other sites. The remaining 57 articles provided a total of 591 patients with brain metastasis from ovarian neoplasm. Of these, five were from a single prospective study [24, 26] and the rest were retrospective observations. Among retrospective cases, 62 were from 28 articles reflecting a single case or small case series, 524 were from 37 manuscripts of larger and mixed case series with little detail of individual cases. Data from all the 57 studies regarding patient characteristics, initial approach to ovarian cancer management, interval between primary diagnosis to brain metastasis, treatment modalities, signs, symptoms and location of brain metastases, different serum markers of the tumor and tumor cell receptors as well as survival outcomes were reviewed and compared. The mean time interval between primary ovarian cancer diagnosis and brain metastasis, and the survival time after the appearance of brain metastasis were compared in different groups using student T test. P value was calculated.
Results The reported incidence of brain metastases among these studies varies from 0.49 to 11.54 % with an average of Table 1 Incidence of brain metastasis in different studies
123
2.55 % (Table 1). The median age of patients at the time of primary ovarian malignancy diagnosis was 52 years (range 13–92) and at the time of brain metastasis was 54.3 years (range 20–81). The median time interval between initial ovarian malignancy diagnosis and development of brain metastasis was 24 months (0–133). In three cases, the brain metastasis was found before the diagnosis of ovarian cancer. A majority of patients with brain metastasis had advanced stage ovarian cancer-using The International Federation of Gynecology and Obstetrics (FIGO) staging system (83 % FIGO stage III/IV vs. 17 % FIGO stage I/II). Serous cyst-adenocarcinoma was the predominant histologic subtype (54.5 %). The histologic grade of the primary ovarian tumor was reported in 26 studies for 436 patients with brain metastasis, the majority of patients who developed brain metastasis had high grade initial tumors (14.2 % G1, 17 % G2, 66.5 % G3, 2.3 % unreported grade),the presence of higher grade primary ovarian tumor significantly increased the likelihood of brain metastasis (P value \0.0001). Presence of regional lymph node involvement at the time of initial diagnosis of ovarian cancer, mentioned in 17 studies for 68 cases was 73.8 %. A majority of patients had multiple brain lesions (57.3 %). Among 582 patients with brain metastasis, extra-cranial metastasis was present in 292 cases (49.8 %), extra cranial lesions were noted in the liver (20.7 %), lung (20.4 %), lymph nodes (12.6 %), bone (6.6 %) and pelvic organs (4.3 %). The most common
Author
Published in
Years included
OVC cases
BM cases
Incidence
Sekine et al.
2013
1983–2007
340
7
2.1
Matsuo et al.
2011
1995–2009
302
5
1.66
Sehouli et al.
2010
1981–2008
4,077
74
1.82
Ratner et al.
2009
1983–2007
2,097
23
1.1
Chen et al.
2009
2000–2007
539
10
1.86
Yang et al.
2008
1986–2007
1,055
7
0.66
KIM et al.
2007
1996–2005
490
13
2.65
Gadducci et al. Kastritis et al.
2007 2006
1995–2005 1995–2004
195 150
12 8
6.15 5.33
Tay et al.
2005
1993–2003
605
4
0.66
Cohen et al.
2004
1975–2001
8,225
72
0.88
Kumar et al.
2003
1991–2001
658
18
2.74
Anupol et al.
2002
1986–2000
1,042
15
1.44
Kolomainen et al.
2002
1980–2000
3,690
18
0.49
Sanderson et al.
2001
1995–2000
1,222
13
1.06
Geisler et al.
1995
1979–1992
479
16
3.34
Bruzzone et al.
1993
1981–1989
413
9
2.18
Rodriguez et al.
1992
1977–1990
795
15
1.89
Hardy et al.
1989
1981–1984
52
6
11.54
Barker et al.
1981
1969–1979
430
4
0.93
J Neurooncol
Fig. 1 Signs and symptoms of brain lesion in patients with brain metastasis from ovarian cancer Miscellaneous symptoms include: Memory loss, Personality changes, Jerky movements of extremities, Neuropathy, Cerebrovascular accidents and etc
Fig. 2 Location of brain metastasis in patients with primary ovarian cancer Miscellaneous locations include: Basal ganglia, cerebral aqueducts and suprasellar regions
symptoms in patients with brain metastasis were: sensory and motor disturbances, ataxia, seizure and alteration of consciousness (Fig. 1). Among anatomic sites, cerebellum was the most common site of intracranial cerebral metastasis (Fig. 2). In three patients, brain metastases were diagnosed prior to the discovery of the ovarian cancer. They were 36, 39 and 56 years of age. Two of the three women had endometrioid carcinoma and the third one had undifferentiated carcinoma. Two of these three patients did not have any pelvic lymph node involvement. The CA-125 measurement at the time of initial diagnosis of ovarian cancer was mentioned in 17 studies for 52 patients; it was positive in 78.8 %. In 82 patients with brain metastasis, CA-125 measurement was available, and was
positive in 58.5 %. The quantitative CA-125 analysis was available in 12 patients with brain metastasis. The median value was 107 and the average was 170.67 (range: 4–4,428, normal values \35). Other markers including CA 72-4, LDH, EMA, Chromogranin, CD-56, pan-CK along with serum levels of HCG, and AFP have been rarely reported. A documented BRCA gene mutation status was only reported in 4 studies, with 5 of 15 patients showing mutation. The Karnowsky performance score was reported in 10 studies for 217 patients, 107 of whom had scores\70 and 111 cases had [70. The reported treatment options for primary ovarian cancer consist of: surgery (51.6 %), chemotherapy (28.8 %) and combination of chemotherapy and surgery (16.6 %). Of 218 patients in whom type of surgery was described, 39.9 % had optimal (\1 cm residual tumor) cytoreduction and 60.1 % had suboptimal cytoreduction ([1 cm residual tumor). Second look surgery status was reported for 193 patients with 22.3 % positive, 39.9 % negative and 37.8 % not performed. Response to treatment of primary malignancy was mentioned in 148 patients (66.9 % complete and 33.1 % incomplete). Treatment options for brain metastasis included whole brain radiotherapy(WBRT: 30 %), WBRT and surgery (15.4 %), WBRT and chemotherapy (13 %), all three modalities (11.1 %), systemic chemotherapy (9.1 %), steroids (6.5 %), palliative care or no treatment (6.5 %), surgery (5.3 %), intrathecal (IT) chemotherapy (0.5 %), surgery and chemotherapy (1.9 %) and WBRT and IT chemotherapy (0.4 %). Patients with primary serous cystadenocarcinoma had significantly longer interval between the primary diagnosis and development of metastasis (25.6 vs. 18.6 months, P value = 0.031) compared to other subtypes. Patients who had extra cranial metastasis at the time of initial diagnosis of ovarian cancer had a shorter time interval between diagnosis of ovarian cancer and brain metastasis (24.6 vs. 61.7 months, P value = 0.040). Different treatment options for primary ovarian cancer did not significantly impact this interval. Patients age at the time of diagnosis of ovarian cancer, positive or negative CA-125 did not influence the time interval between primary ovarian cancer and diagnosis of brain metastasis significantly either. The median survival of patients after confirmation of brain metastasis was 8.2 months (average of 16 months). A total of 26 studies compared the presence of multiple brain lesions with solitary metastasis. The average survival was 9.2 months for multiple lesions compared to 21.4 months for single lesions (P value = 0.17). Patients who were \50 years of age at the time of diagnosis of the primary ovarian cancer, had a better survival than older patients (38.9 months vs. 13 months respectively, P value = 0.002). Likewise, in patients with brain metastasis, survival was
123
J Neurooncol
longer for these younger patients (30.6 vs. 8.1 months, P value = 0.01). A Karnowsky performance scale (KPS) of [70 increased survival rates compared to lower KPS (28 months vs. 11.7 respectively. P value = 0.031). Combination of surgery, WBRT and chemotherapy was associated with longer survival compared to WBRT alone (20.5 months vs. 9.1 months, P value = 0.04); palliative therapy alone, compared to the combination of surgery, chemotherapy and WBRT significantly reduces the survival rate (9.6 months, P value = 0.045). The type of ovarian cancer, presence or absence of extra cranial metastasis, and positive or higher positive levels of CA-125 did not correlate with survival after brain metastasis.
Table 2 Different factors involved in the time interval between primary ovarian cancer and appearance of brain metastasis Factor
OVC to BM time interval
P value
Age at BM diagnosis \50 [50 Primary pathology
49.6
0.139
23.4
SCAC
25.6
Others
18.6
0.031
LN involvement Positive
20.3
Negative
14.3
0.462
Number of brain lesions Multiple lesions
Discussion
Single lesions
23.13
0.352
47
CA 125 status
The reported incidence for brain metastasis from ovarian cancer ranges widely from less than 1–12 % (Table 1). The average figure of 2.55 % found in our literature review is consistent with the most recent post mortem review with the incidence of 2.1 % (7 out of 340 cases) [12]. However, since most reports are from clinical studies and very few postmortem studies have been performed, it is possible that the actual incidence of brain metastasis from ovarian cancer is higher since brain imaging is not a routine component of follow up for patients with ovarian cancer to detect asymptomatic lesions. From an anatomical point of view, this review found cerebellum to be the most common site of intracranial metastasis for ovarian cancer (Fig. 1). This may be due to the rich blood supply of the cerebellum. Rostami et al. [27] reported a similar observation for brain metastasis from breast cancer. From a histological standpoint, high grade primary ovarian cancer (G3) as well as advanced stage at presentation of the primary malignancy at the time of diagnosis significantly increases the likelihood of development of brain metastasis (P value \0.0001). The finding that serous cystadenocarcinoma (SCAC) is associated with longer interval (Table 1) is surprising since many CACs reflect a high grade, aggressive form of ovarian carcinoma. Possibly most cases of the studied cohorts were of low grade SCAC variant. It is also worth mentioning that none of the three patients in whom brain metastases preceded detection of the ovarian cancer, had serous cystadenocarcinoma. Factors that correlated with better survival after development of cerebral metastasis were young age both at the time of initial diagnosis and at the time of brain metastasis detection (30.6 months for age \50 vs. 8.1 for [50, P value = 0.002), and a high KPS score. Correlation with high KPS scores is in agreement with previous observations
123
Positive
6.8
Negative
23.3
0.594
CA 125 level \100 [101 Extracranial metastasis
15
0.997
28.6
Present
61.7
Absent
24.5
0.04
BM breast metastasis, SCAC serous cystadenocarcinoma, LN lymph node Statistically significant difference considered as P value \ 0.05
(6, 22, 25, and 26). The type of ovarian pathology, grade and stage of tumor, interval between diagnosis of ovarian cancer and diagnosis of brain metastasis, high CA-125 titers (initially or later) and concurrent presence of motor or sensory symptoms did not correlate with prognosis. Presence of extracranial metastasis, although significantly influencing the interval between ovarian cancer and brain metastasis, is not a predictor of survival after central nervous system involvement. The presence of multiple brain lesions showed a considerably reduced survival (9.2 months for multiple lesions vs. 21.4 months for single lesions), however, the statistical analysis failed to reveal a significant difference, possibly due to the small sample size. The impact of initial pelvic lymph node involvement on interval to brain metastasis and overall survival was not evaluated due to the few number of studies in which, the detailed information about lymph node involvement and survival of the patients were included (Table 3). Previous studies have shown inconsistent results regarding the impact of extra cranial metastasis on outcome; association with lower survival (after detection of brain metastasis) was suggested by Cormio et al. [23] and Anupol et al. [25], while Sehouli et al. [26], like us, did not find such
J Neurooncol Table 3 Different Prognostic factors involved in survival for patients with brain metastasis due to ovarian cancer Factor
Survival (months)
P value
\50
38.9
0.001
[50
13.0
Age at BM diagnosis
Age when ovarian cancer was diagnosed Age \50
30.6
Age [50
8.1
Primary pathology SCAC Other
16.7
0.002
0.715
14.3
Number of brain lesions Multiple lesions Single lesions
9.2 21.4
0.17
Positive
23.2
0.118
Negative
28.9
CA 125 status
CA 125 level \100
6.7
[101
1.7
0.674
In conclusion, brain metastasis is an uncommon complication of ovarian cancer but its true incidence is not well established. Cerebellum was the most common site of cerebral metastasis from ovarian cancer. Absence of distant metastasis at the time of diagnosis of ovarian cancer correlate with delayed manifestation of brain metastasis. Survival following the detection of brain metastasis is influenced by patients’ age, absence of multiple metastasis and performance status (KPS score), but presence of extra cranial metastasis, and increased serum CA-125 titers do not have prognostic significance. Since the therapeutic options have significant impact on survival after appearance of brain metastasis, it is reasonable to use multimodality therapy when feasible, depending on patients’ condition. At present, neither a standard screening method to diagnose brain metastasis in ovarian cancer patients, nor a standardized therapeutic guideline is available. The rarity of brain metastasis precludes performance of large scale studies; therefore, establishment of multicenter collaborative studies to configure standard management approaches is warranted.
KPS score \70
11.7
[70
28
Treatment option No treatment WBRT ? Chemo ? SX
10.8
0.031
Conclusion 0.045
20.5
BM breast metastasis, SCAC serous cystadenocarcinoma, LN lymph node Statistically significant considered as P value \ 0.05
association. This may be due to the smaller size of the population studied in the former studies. Cormio et al. [28] also reported longer interval between diagnosis of brain metastasis and death in patients with longer interval between the diagnosis of the primary ovarian cancer and the subsequent brain metastasis (Table 2). Our review which encompasses the literature to date and a similar study by Pietzner et al. [2] found no such correlation. Brain metastasis from ovarian cancer was treated differently by different groups using a range of palliative and no treatment to multi modal treatment strategy using surgery, radiation and chemotherapy [15–18, 23, 25]. Among different therapeutic choices, combination of surgery, radiation and chemotherapy produced a more effective response and was associated with better survival (Table 3). The relationship between KPS score, and treatment options and survival is of practical value and can be used effectively in tailoring the management plan for patients with brain metastasis from ovarian cancer based on their performance status.
Brain metastasis is a rare and fatal outcome of ovarian cancer that is accompanied by a very poor survival. However, based on the data presented in this review, there are factors that are associated with more favorable outcomes. These factors- patient age, histologic subtype of the cancer, primary KPS score indicating the performance status, existing extra cranial metastasis—constitute important elements that should be considered in selection of optimal therapy for each patient. The association between multimodal therapy and better survival is important information that suggests more personalized decision making in terms of therapeutic choice based on patients’ health status and chemotherapy side effects versus the overall benefit of this approach. The inconclusive data regarding the role of other factors in the overall disease progression may be due to the paucity of available data and therefore, the potential role of these factors (i.e.; multiplicity of metastatic lesions, CA 125 level, primary location of tumor, etc.) on overall disease progression should be investigated in larger, multicenter studies in the future.
Conflict of interests disclose.
The authors have no conflicts of interest to
123
J Neurooncol
References 1. Wong J, Hird A, Kirou-Mauro A, Napolskikh J, Chow E (2008) Quality of life in brain metastases radiation trials: a literature review. Curr Oncol 15:25–45 2. Pietzner K, Oskay-Oezcelik G, El Khalfaoui K, Boehmer D, Lichtenegger W, Sehouli J (2009) Brain metastases from epithelial ovarian cancer: overview and optimal management. Anticancer Res 29:2793–2798 3. Hu XQ, Imitola J, Kim Y, Mahta A, Kesari S (2012) Brain metastasis from ovarian cancer: case report and review of the literature. Med Oncol 29:1250–1252 4. Chen YL, Cheng WF, Hsieh CY, Chen CA (2011) Brain metastasis as a late manifestation of ovarian carcinoma. Eur J Cancer Care (Engl) 20:44–49 5. Larson DM, Copeland LJ, Moser RP, Malone JM Jr, Gershenson DM, Wharton JT (1986) Central nervous system metastases in epithelial ovarian carcinoma. Obstet Gynecol 68:746–750 6. Cohen ZR, Suki D, Weinberg JS, Marmor E, Lang FF, Gershenson DM et al (2004) Brain metastases in patients with ovarian carcinoma: prognostic factors and outcome. J Neurooncol 66:313–325 7. D’Andrea G, Roperto R, Dinia L, Caroli E, Salvati M, Ferrante L (2005) Solitary cerebral metastases from ovarian epithelial carcinoma: 11 cases. Neurosurg Rev 28:120–123 8. Kaminsky-Forrett MC, Weber B, Conroy T, Spaeth D (2000) Brain metastases from epithelial ovarian carcinoma. Int J Gynecol Cancer 10:366–371 9. Pectasides D, Aravantinos G, Fountzilas G, Kalofonos C, Efstathiou E, Karina M et al (2005) Brain metastases from epithelial ovarian cancer. The Hellenic Cooperative Oncology Group (HeCOG) experience and review of the literature. Anticancer Res 25:3553–3558 10. Bruzzone M, Campora E, Chiara S, Giudici S, Merlini L, Simoni C et al (1993) Cerebral metastases secondary to ovarian cancer: still an unusual event. Gynecol Oncol 49:37–40 11. Ricke J, Baum K, Hosten N (1996) Calcified brain metastases from ovarian carcinoma. Neuroradiology 38:460–461 12. Sekine M, Yoshihara K, Komata D, Haino K, Nishino K, Tanaka K (2013) Increased incidence of brain metastases in BRCA1related ovarian cancers. J Obstet Gynaecol Res 39:292–296 13. Patchell RA (2003) The management of brain metastases. Cancer Treat Rev 29:533–540 14. Ewend MG, Carey LA, Morris DE, Harvey RD, Hensing TA (2001) Brain metastases. Curr Treat Options Oncol 2:537–547 15. Growdon WB, Lopez-Varela E, Littell R, Oliva E, Seiden M, Krasner C et al (2008) Extent of extracranial disease is a powerful
123
16.
17.
18.
19.
20. 21.
22.
23.
24. 25.
26.
27.
28.
predictor of survival in patients with brain metastases from gynecological cancer. Int J Gynecol Cancer 18:262–268 Choo BA, Walji N, Spooner D, Barber P, Fernando IN (2010) Prolonged relapse-free survival in two patients with an isolated brain metastasis from epithelial ovarian carcinoma. J Clin Oncol 28:271–272 Melichar B, Urminska´ H, Kohlova´ T, Nova´ M, Cesa´k T (2004) Brain metastases of epithelial ovarian carcinoma responding to cisplatin and gemcitabine combination chemotherapy: a case report and review of the literature. Gynecol Oncol 94:267–276 McMeekin DS, Kamelle SA, Vasilev SA, Tillmanns TD, Gould NS, Scribner DR et al (2001) Ovarian cancer metastatic to the brain: what is the optimal management? J Surg Oncol 78: 194–201 Root K, Armaghany T (2012) Solitary brain metastasis in a patient with ovarian cancer with BRCA2 mutation. J Clin Oncol 30(25):e239–e240 Piura E, Piura B (2011) Brain metastases from ovarian carcinoma. ISRN Oncol 2011:527453 Lee YK, Park NH, Kim JW, Song YS, Kang SB, Lee HP (2008) Gamma-knife radiosurgery as an optimal treatment modality for brain metastases from epithelial ovarian cancer. Gynecol Oncol 108:505–509 Matsuo K, Eno ML, Ahn EH, Shahzad MM, Im DD, Rosenshein NB et al (2011) Multi-drug resistance gene (MDR-1) and risk of brain metastasis in epithelial ovarian, fallopian tube, and peritoneal cancer. Am J Clin Oncol 34:488–493 Cormio G, Loizzi V, Falagario M, Lissoni AA, Resta L, Selvaggi LE (2011) Changes in the management and outcome of central nervous system involvement from ovarian cancer since 1994. Int J Gynaecol Obstet 114:133–136 Clark TG, Stewart ME, Altman DG, Gabra H, Smyth JF (2001) A prognostic model for ovarian cancer. Br J Cancer 85:944–952 Anupol N, Ghamande S, Odunsi K, Driscoll D, Lele S (2002) Evaluation of prognostic factors and treatment modalities in ovarian cancer patients with brain metastases. Gynecol Oncol 85:487–492 Sehouli J, Pietzner K, Harter P, Munstedt K, Mahner S, Hasenburg A et al (2010) Prognostic role of platinum sensitivity in patients with brain metastases from ovarian cancer: results of a German multicenter study. Ann Oncol 21:2201–2205 Rostami R, Safarpour D, Tavassoli FA, Jabbari B (2013) Intramedullary metastasis in breast cancer—a comprehensive literature review. J Neurol Sci 332:16–20 Cormio G, Rossi C, Cazzolla A, Resta L, Loverro G, Greco P et al (2003) Distant metastases in ovarian carcinoma. Int J Gynecol Cancer 13:125–129
TOPIC REVIEW
Brain metastasis from ovarian cancer: a systematic review Shabnam Pakneshan • Damoun Safarpour Fattaneh Tavassoli • Bahman Jabbari
•
Received: 27 January 2014 / Accepted: 13 April 2014 Ó Springer Science+Business Media New York 2014
Abstract To review the existing literature on brain metastasis (BM) from ovarian cancer and to assess the frequency, anatomical, clinical and paraclinical information and factors associated with prognosis. Ovarian cancer is a rare cause of brain metastasis with a recently reported increasing prevalence. Progressive neurologic disability and poor prognosis is common. A comprehensive review on this subject has not been published previously. This systematic literature search used the Pubmed and Yale library. A total of 66 publications were found, 57 of which were used representing 591 patients with BM from ovarian cancer. The median age of the patients was 54.3 years (range 20-81). A majority of patients (57.3 %) had multiple brain lesions. The location of the lesion was cerebellar (30 %), frontal (20 %), parietal (18 %) and occipital (11 %). Extracranial metastasis was present in 49.8 % of cases involving liver (20.7 %), lung (20.4 %), lymph nodes (12.6 %), bones (6.6 %) and pelvic organs (4.3 %). The most common symptoms were weakness (16 %), seizures (11 %), altered mentality (11 %) visual disturbances (9 %) and dizziness (8 %). The interval from diagnosis of breast cancer to BM ranged from 0 to 133 months (median 24 months) and median survival was 8.2 months. Local radiation, surgical resection, stereotactic radiosurgery and medical therapy were used. Factors that significantly increased the survival were younger age at the time of ovarian cancer diagnosis and brain metastasis diagnosis, lower grade of the primary tumor, higher KPS score and multimodality treatment for the brain metastases. Ovarian cancer is a rare cause of brain metastasis. Development of brain metastasis among older patients and lower KPS score correlate with less favorable prognosis. The S. Pakneshan (&) D. Safarpour F. Tavassoli B. Jabbari Department of Neurology, Yale University School of Medicine, 15 York Street, LCI Building, New Haven, CT 06520, USA e-mail: [email protected]
more prolonged survival after using multimodality treatment for brain metastasis is important due to potential impact on management of brain metastasis in future. Keywords Ovarian cancer Brain metastasis Survival Radiation Chemotherapy Multimodality treatment CA-125 Serouse cystadenocarcinoma
Introduction Brain metastasis, although a common and severe complication in lung and breast cancer [1, 2], is considered a rare and late event in ovarian cancer [2–4]. A variable incidence of 0.3–12 % [5–11] has been reported in different studies as well as its association with poor prognosis [12]. The reported incidence of brain metastasis is higher in more recent studies [3, 4] possibly reflecting advances in imaging and longer survival from primary ovarian cancer [13–21]. The rarity and small number of patients affected, have prevented the establishment of a consensus for optimal therapy [3]. The increasing incidence of brain metastasis in ovarian cancer, however, mandates special attention [22] since the treatment strategy clearly affects the prognosis [22–25]. We conducted a comprehensive review of the existing literature on this subject in order to create a database to recognize and compare different patient characteristics, treatment modalities and prognostic factors that may impact the ultimate outcome and survival.
Materials and methods A total of 66 articles published between 1978 and 2013 were reviewed following a Yale web based search of
123
J Neurooncol
Medline, PubMed, Ovid and other medical databases. Nine manuscripts were excluded because either they did not separate ovarian malignancy from other gynecologic cancers or they did not separate brain metastasis from metastasis to other sites. The remaining 57 articles provided a total of 591 patients with brain metastasis from ovarian neoplasm. Of these, five were from a single prospective study [24, 26] and the rest were retrospective observations. Among retrospective cases, 62 were from 28 articles reflecting a single case or small case series, 524 were from 37 manuscripts of larger and mixed case series with little detail of individual cases. Data from all the 57 studies regarding patient characteristics, initial approach to ovarian cancer management, interval between primary diagnosis to brain metastasis, treatment modalities, signs, symptoms and location of brain metastases, different serum markers of the tumor and tumor cell receptors as well as survival outcomes were reviewed and compared. The mean time interval between primary ovarian cancer diagnosis and brain metastasis, and the survival time after the appearance of brain metastasis were compared in different groups using student T test. P value was calculated.
Results The reported incidence of brain metastases among these studies varies from 0.49 to 11.54 % with an average of Table 1 Incidence of brain metastasis in different studies
123
2.55 % (Table 1). The median age of patients at the time of primary ovarian malignancy diagnosis was 52 years (range 13–92) and at the time of brain metastasis was 54.3 years (range 20–81). The median time interval between initial ovarian malignancy diagnosis and development of brain metastasis was 24 months (0–133). In three cases, the brain metastasis was found before the diagnosis of ovarian cancer. A majority of patients with brain metastasis had advanced stage ovarian cancer-using The International Federation of Gynecology and Obstetrics (FIGO) staging system (83 % FIGO stage III/IV vs. 17 % FIGO stage I/II). Serous cyst-adenocarcinoma was the predominant histologic subtype (54.5 %). The histologic grade of the primary ovarian tumor was reported in 26 studies for 436 patients with brain metastasis, the majority of patients who developed brain metastasis had high grade initial tumors (14.2 % G1, 17 % G2, 66.5 % G3, 2.3 % unreported grade),the presence of higher grade primary ovarian tumor significantly increased the likelihood of brain metastasis (P value \0.0001). Presence of regional lymph node involvement at the time of initial diagnosis of ovarian cancer, mentioned in 17 studies for 68 cases was 73.8 %. A majority of patients had multiple brain lesions (57.3 %). Among 582 patients with brain metastasis, extra-cranial metastasis was present in 292 cases (49.8 %), extra cranial lesions were noted in the liver (20.7 %), lung (20.4 %), lymph nodes (12.6 %), bone (6.6 %) and pelvic organs (4.3 %). The most common
Author
Published in
Years included
OVC cases
BM cases
Incidence
Sekine et al.
2013
1983–2007
340
7
2.1
Matsuo et al.
2011
1995–2009
302
5
1.66
Sehouli et al.
2010
1981–2008
4,077
74
1.82
Ratner et al.
2009
1983–2007
2,097
23
1.1
Chen et al.
2009
2000–2007
539
10
1.86
Yang et al.
2008
1986–2007
1,055
7
0.66
KIM et al.
2007
1996–2005
490
13
2.65
Gadducci et al. Kastritis et al.
2007 2006
1995–2005 1995–2004
195 150
12 8
6.15 5.33
Tay et al.
2005
1993–2003
605
4
0.66
Cohen et al.
2004
1975–2001
8,225
72
0.88
Kumar et al.
2003
1991–2001
658
18
2.74
Anupol et al.
2002
1986–2000
1,042
15
1.44
Kolomainen et al.
2002
1980–2000
3,690
18
0.49
Sanderson et al.
2001
1995–2000
1,222
13
1.06
Geisler et al.
1995
1979–1992
479
16
3.34
Bruzzone et al.
1993
1981–1989
413
9
2.18
Rodriguez et al.
1992
1977–1990
795
15
1.89
Hardy et al.
1989
1981–1984
52
6
11.54
Barker et al.
1981
1969–1979
430
4
0.93
J Neurooncol
Fig. 1 Signs and symptoms of brain lesion in patients with brain metastasis from ovarian cancer Miscellaneous symptoms include: Memory loss, Personality changes, Jerky movements of extremities, Neuropathy, Cerebrovascular accidents and etc
Fig. 2 Location of brain metastasis in patients with primary ovarian cancer Miscellaneous locations include: Basal ganglia, cerebral aqueducts and suprasellar regions
symptoms in patients with brain metastasis were: sensory and motor disturbances, ataxia, seizure and alteration of consciousness (Fig. 1). Among anatomic sites, cerebellum was the most common site of intracranial cerebral metastasis (Fig. 2). In three patients, brain metastases were diagnosed prior to the discovery of the ovarian cancer. They were 36, 39 and 56 years of age. Two of the three women had endometrioid carcinoma and the third one had undifferentiated carcinoma. Two of these three patients did not have any pelvic lymph node involvement. The CA-125 measurement at the time of initial diagnosis of ovarian cancer was mentioned in 17 studies for 52 patients; it was positive in 78.8 %. In 82 patients with brain metastasis, CA-125 measurement was available, and was
positive in 58.5 %. The quantitative CA-125 analysis was available in 12 patients with brain metastasis. The median value was 107 and the average was 170.67 (range: 4–4,428, normal values \35). Other markers including CA 72-4, LDH, EMA, Chromogranin, CD-56, pan-CK along with serum levels of HCG, and AFP have been rarely reported. A documented BRCA gene mutation status was only reported in 4 studies, with 5 of 15 patients showing mutation. The Karnowsky performance score was reported in 10 studies for 217 patients, 107 of whom had scores\70 and 111 cases had [70. The reported treatment options for primary ovarian cancer consist of: surgery (51.6 %), chemotherapy (28.8 %) and combination of chemotherapy and surgery (16.6 %). Of 218 patients in whom type of surgery was described, 39.9 % had optimal (\1 cm residual tumor) cytoreduction and 60.1 % had suboptimal cytoreduction ([1 cm residual tumor). Second look surgery status was reported for 193 patients with 22.3 % positive, 39.9 % negative and 37.8 % not performed. Response to treatment of primary malignancy was mentioned in 148 patients (66.9 % complete and 33.1 % incomplete). Treatment options for brain metastasis included whole brain radiotherapy(WBRT: 30 %), WBRT and surgery (15.4 %), WBRT and chemotherapy (13 %), all three modalities (11.1 %), systemic chemotherapy (9.1 %), steroids (6.5 %), palliative care or no treatment (6.5 %), surgery (5.3 %), intrathecal (IT) chemotherapy (0.5 %), surgery and chemotherapy (1.9 %) and WBRT and IT chemotherapy (0.4 %). Patients with primary serous cystadenocarcinoma had significantly longer interval between the primary diagnosis and development of metastasis (25.6 vs. 18.6 months, P value = 0.031) compared to other subtypes. Patients who had extra cranial metastasis at the time of initial diagnosis of ovarian cancer had a shorter time interval between diagnosis of ovarian cancer and brain metastasis (24.6 vs. 61.7 months, P value = 0.040). Different treatment options for primary ovarian cancer did not significantly impact this interval. Patients age at the time of diagnosis of ovarian cancer, positive or negative CA-125 did not influence the time interval between primary ovarian cancer and diagnosis of brain metastasis significantly either. The median survival of patients after confirmation of brain metastasis was 8.2 months (average of 16 months). A total of 26 studies compared the presence of multiple brain lesions with solitary metastasis. The average survival was 9.2 months for multiple lesions compared to 21.4 months for single lesions (P value = 0.17). Patients who were \50 years of age at the time of diagnosis of the primary ovarian cancer, had a better survival than older patients (38.9 months vs. 13 months respectively, P value = 0.002). Likewise, in patients with brain metastasis, survival was
123
J Neurooncol
longer for these younger patients (30.6 vs. 8.1 months, P value = 0.01). A Karnowsky performance scale (KPS) of [70 increased survival rates compared to lower KPS (28 months vs. 11.7 respectively. P value = 0.031). Combination of surgery, WBRT and chemotherapy was associated with longer survival compared to WBRT alone (20.5 months vs. 9.1 months, P value = 0.04); palliative therapy alone, compared to the combination of surgery, chemotherapy and WBRT significantly reduces the survival rate (9.6 months, P value = 0.045). The type of ovarian cancer, presence or absence of extra cranial metastasis, and positive or higher positive levels of CA-125 did not correlate with survival after brain metastasis.
Table 2 Different factors involved in the time interval between primary ovarian cancer and appearance of brain metastasis Factor
OVC to BM time interval
P value
Age at BM diagnosis \50 [50 Primary pathology
49.6
0.139
23.4
SCAC
25.6
Others
18.6
0.031
LN involvement Positive
20.3
Negative
14.3
0.462
Number of brain lesions Multiple lesions
Discussion
Single lesions
23.13
0.352
47
CA 125 status
The reported incidence for brain metastasis from ovarian cancer ranges widely from less than 1–12 % (Table 1). The average figure of 2.55 % found in our literature review is consistent with the most recent post mortem review with the incidence of 2.1 % (7 out of 340 cases) [12]. However, since most reports are from clinical studies and very few postmortem studies have been performed, it is possible that the actual incidence of brain metastasis from ovarian cancer is higher since brain imaging is not a routine component of follow up for patients with ovarian cancer to detect asymptomatic lesions. From an anatomical point of view, this review found cerebellum to be the most common site of intracranial metastasis for ovarian cancer (Fig. 1). This may be due to the rich blood supply of the cerebellum. Rostami et al. [27] reported a similar observation for brain metastasis from breast cancer. From a histological standpoint, high grade primary ovarian cancer (G3) as well as advanced stage at presentation of the primary malignancy at the time of diagnosis significantly increases the likelihood of development of brain metastasis (P value \0.0001). The finding that serous cystadenocarcinoma (SCAC) is associated with longer interval (Table 1) is surprising since many CACs reflect a high grade, aggressive form of ovarian carcinoma. Possibly most cases of the studied cohorts were of low grade SCAC variant. It is also worth mentioning that none of the three patients in whom brain metastases preceded detection of the ovarian cancer, had serous cystadenocarcinoma. Factors that correlated with better survival after development of cerebral metastasis were young age both at the time of initial diagnosis and at the time of brain metastasis detection (30.6 months for age \50 vs. 8.1 for [50, P value = 0.002), and a high KPS score. Correlation with high KPS scores is in agreement with previous observations
123
Positive
6.8
Negative
23.3
0.594
CA 125 level \100 [101 Extracranial metastasis
15
0.997
28.6
Present
61.7
Absent
24.5
0.04
BM breast metastasis, SCAC serous cystadenocarcinoma, LN lymph node Statistically significant difference considered as P value \ 0.05
(6, 22, 25, and 26). The type of ovarian pathology, grade and stage of tumor, interval between diagnosis of ovarian cancer and diagnosis of brain metastasis, high CA-125 titers (initially or later) and concurrent presence of motor or sensory symptoms did not correlate with prognosis. Presence of extracranial metastasis, although significantly influencing the interval between ovarian cancer and brain metastasis, is not a predictor of survival after central nervous system involvement. The presence of multiple brain lesions showed a considerably reduced survival (9.2 months for multiple lesions vs. 21.4 months for single lesions), however, the statistical analysis failed to reveal a significant difference, possibly due to the small sample size. The impact of initial pelvic lymph node involvement on interval to brain metastasis and overall survival was not evaluated due to the few number of studies in which, the detailed information about lymph node involvement and survival of the patients were included (Table 3). Previous studies have shown inconsistent results regarding the impact of extra cranial metastasis on outcome; association with lower survival (after detection of brain metastasis) was suggested by Cormio et al. [23] and Anupol et al. [25], while Sehouli et al. [26], like us, did not find such
J Neurooncol Table 3 Different Prognostic factors involved in survival for patients with brain metastasis due to ovarian cancer Factor
Survival (months)
P value
\50
38.9
0.001
[50
13.0
Age at BM diagnosis
Age when ovarian cancer was diagnosed Age \50
30.6
Age [50
8.1
Primary pathology SCAC Other
16.7
0.002
0.715
14.3
Number of brain lesions Multiple lesions Single lesions
9.2 21.4
0.17
Positive
23.2
0.118
Negative
28.9
CA 125 status
CA 125 level \100
6.7
[101
1.7
0.674
In conclusion, brain metastasis is an uncommon complication of ovarian cancer but its true incidence is not well established. Cerebellum was the most common site of cerebral metastasis from ovarian cancer. Absence of distant metastasis at the time of diagnosis of ovarian cancer correlate with delayed manifestation of brain metastasis. Survival following the detection of brain metastasis is influenced by patients’ age, absence of multiple metastasis and performance status (KPS score), but presence of extra cranial metastasis, and increased serum CA-125 titers do not have prognostic significance. Since the therapeutic options have significant impact on survival after appearance of brain metastasis, it is reasonable to use multimodality therapy when feasible, depending on patients’ condition. At present, neither a standard screening method to diagnose brain metastasis in ovarian cancer patients, nor a standardized therapeutic guideline is available. The rarity of brain metastasis precludes performance of large scale studies; therefore, establishment of multicenter collaborative studies to configure standard management approaches is warranted.
KPS score \70
11.7
[70
28
Treatment option No treatment WBRT ? Chemo ? SX
10.8
0.031
Conclusion 0.045
20.5
BM breast metastasis, SCAC serous cystadenocarcinoma, LN lymph node Statistically significant considered as P value \ 0.05
association. This may be due to the smaller size of the population studied in the former studies. Cormio et al. [28] also reported longer interval between diagnosis of brain metastasis and death in patients with longer interval between the diagnosis of the primary ovarian cancer and the subsequent brain metastasis (Table 2). Our review which encompasses the literature to date and a similar study by Pietzner et al. [2] found no such correlation. Brain metastasis from ovarian cancer was treated differently by different groups using a range of palliative and no treatment to multi modal treatment strategy using surgery, radiation and chemotherapy [15–18, 23, 25]. Among different therapeutic choices, combination of surgery, radiation and chemotherapy produced a more effective response and was associated with better survival (Table 3). The relationship between KPS score, and treatment options and survival is of practical value and can be used effectively in tailoring the management plan for patients with brain metastasis from ovarian cancer based on their performance status.
Brain metastasis is a rare and fatal outcome of ovarian cancer that is accompanied by a very poor survival. However, based on the data presented in this review, there are factors that are associated with more favorable outcomes. These factors- patient age, histologic subtype of the cancer, primary KPS score indicating the performance status, existing extra cranial metastasis—constitute important elements that should be considered in selection of optimal therapy for each patient. The association between multimodal therapy and better survival is important information that suggests more personalized decision making in terms of therapeutic choice based on patients’ health status and chemotherapy side effects versus the overall benefit of this approach. The inconclusive data regarding the role of other factors in the overall disease progression may be due to the paucity of available data and therefore, the potential role of these factors (i.e.; multiplicity of metastatic lesions, CA 125 level, primary location of tumor, etc.) on overall disease progression should be investigated in larger, multicenter studies in the future.
Conflict of interests disclose.
The authors have no conflicts of interest to
123
J Neurooncol
References 1. Wong J, Hird A, Kirou-Mauro A, Napolskikh J, Chow E (2008) Quality of life in brain metastases radiation trials: a literature review. Curr Oncol 15:25–45 2. Pietzner K, Oskay-Oezcelik G, El Khalfaoui K, Boehmer D, Lichtenegger W, Sehouli J (2009) Brain metastases from epithelial ovarian cancer: overview and optimal management. Anticancer Res 29:2793–2798 3. Hu XQ, Imitola J, Kim Y, Mahta A, Kesari S (2012) Brain metastasis from ovarian cancer: case report and review of the literature. Med Oncol 29:1250–1252 4. Chen YL, Cheng WF, Hsieh CY, Chen CA (2011) Brain metastasis as a late manifestation of ovarian carcinoma. Eur J Cancer Care (Engl) 20:44–49 5. Larson DM, Copeland LJ, Moser RP, Malone JM Jr, Gershenson DM, Wharton JT (1986) Central nervous system metastases in epithelial ovarian carcinoma. Obstet Gynecol 68:746–750 6. Cohen ZR, Suki D, Weinberg JS, Marmor E, Lang FF, Gershenson DM et al (2004) Brain metastases in patients with ovarian carcinoma: prognostic factors and outcome. J Neurooncol 66:313–325 7. D’Andrea G, Roperto R, Dinia L, Caroli E, Salvati M, Ferrante L (2005) Solitary cerebral metastases from ovarian epithelial carcinoma: 11 cases. Neurosurg Rev 28:120–123 8. Kaminsky-Forrett MC, Weber B, Conroy T, Spaeth D (2000) Brain metastases from epithelial ovarian carcinoma. Int J Gynecol Cancer 10:366–371 9. Pectasides D, Aravantinos G, Fountzilas G, Kalofonos C, Efstathiou E, Karina M et al (2005) Brain metastases from epithelial ovarian cancer. The Hellenic Cooperative Oncology Group (HeCOG) experience and review of the literature. Anticancer Res 25:3553–3558 10. Bruzzone M, Campora E, Chiara S, Giudici S, Merlini L, Simoni C et al (1993) Cerebral metastases secondary to ovarian cancer: still an unusual event. Gynecol Oncol 49:37–40 11. Ricke J, Baum K, Hosten N (1996) Calcified brain metastases from ovarian carcinoma. Neuroradiology 38:460–461 12. Sekine M, Yoshihara K, Komata D, Haino K, Nishino K, Tanaka K (2013) Increased incidence of brain metastases in BRCA1related ovarian cancers. J Obstet Gynaecol Res 39:292–296 13. Patchell RA (2003) The management of brain metastases. Cancer Treat Rev 29:533–540 14. Ewend MG, Carey LA, Morris DE, Harvey RD, Hensing TA (2001) Brain metastases. Curr Treat Options Oncol 2:537–547 15. Growdon WB, Lopez-Varela E, Littell R, Oliva E, Seiden M, Krasner C et al (2008) Extent of extracranial disease is a powerful
123
16.
17.
18.
19.
20. 21.
22.
23.
24. 25.
26.
27.
28.
predictor of survival in patients with brain metastases from gynecological cancer. Int J Gynecol Cancer 18:262–268 Choo BA, Walji N, Spooner D, Barber P, Fernando IN (2010) Prolonged relapse-free survival in two patients with an isolated brain metastasis from epithelial ovarian carcinoma. J Clin Oncol 28:271–272 Melichar B, Urminska´ H, Kohlova´ T, Nova´ M, Cesa´k T (2004) Brain metastases of epithelial ovarian carcinoma responding to cisplatin and gemcitabine combination chemotherapy: a case report and review of the literature. Gynecol Oncol 94:267–276 McMeekin DS, Kamelle SA, Vasilev SA, Tillmanns TD, Gould NS, Scribner DR et al (2001) Ovarian cancer metastatic to the brain: what is the optimal management? J Surg Oncol 78: 194–201 Root K, Armaghany T (2012) Solitary brain metastasis in a patient with ovarian cancer with BRCA2 mutation. J Clin Oncol 30(25):e239–e240 Piura E, Piura B (2011) Brain metastases from ovarian carcinoma. ISRN Oncol 2011:527453 Lee YK, Park NH, Kim JW, Song YS, Kang SB, Lee HP (2008) Gamma-knife radiosurgery as an optimal treatment modality for brain metastases from epithelial ovarian cancer. Gynecol Oncol 108:505–509 Matsuo K, Eno ML, Ahn EH, Shahzad MM, Im DD, Rosenshein NB et al (2011) Multi-drug resistance gene (MDR-1) and risk of brain metastasis in epithelial ovarian, fallopian tube, and peritoneal cancer. Am J Clin Oncol 34:488–493 Cormio G, Loizzi V, Falagario M, Lissoni AA, Resta L, Selvaggi LE (2011) Changes in the management and outcome of central nervous system involvement from ovarian cancer since 1994. Int J Gynaecol Obstet 114:133–136 Clark TG, Stewart ME, Altman DG, Gabra H, Smyth JF (2001) A prognostic model for ovarian cancer. Br J Cancer 85:944–952 Anupol N, Ghamande S, Odunsi K, Driscoll D, Lele S (2002) Evaluation of prognostic factors and treatment modalities in ovarian cancer patients with brain metastases. Gynecol Oncol 85:487–492 Sehouli J, Pietzner K, Harter P, Munstedt K, Mahner S, Hasenburg A et al (2010) Prognostic role of platinum sensitivity in patients with brain metastases from ovarian cancer: results of a German multicenter study. Ann Oncol 21:2201–2205 Rostami R, Safarpour D, Tavassoli FA, Jabbari B (2013) Intramedullary metastasis in breast cancer—a comprehensive literature review. J Neurol Sci 332:16–20 Cormio G, Rossi C, Cazzolla A, Resta L, Loverro G, Greco P et al (2003) Distant metastases in ovarian carcinoma. Int J Gynecol Cancer 13:125–129
