Neurochirurgie 60 (2014) 12–16
Disponible en ligne sur
ScienceDirect www.sciencedirect.com
Original article
Brain metastasis from renal cell carcinoma Métastases cérébrales de cancer du rein O. Bennani a,∗,b , S. Derrey a , O. Langlois a , H. Castel d , A. Laquerriere c , P. Freger a , F. Proust a a
Service de neurochirurgie, CHU de Rouen, 1, rue de Germont, 76031 Rouen cedex, France Service de neurologie, CHU de Rouen, 76031 Rouen, France Service d’anatomopathologie, CHU de Rouen, 76031 Rouen, France d U982 Inserm, Astrocyte and Vascular Niche, DC2N, université de Rouen, 76821 Mont-Saint-Aignan, France b c
a r t i c l e
i n f o
Article history: Received 27 November 2012 Received in revised form 19 November 2013 Accepted 8 December 2013 Available online 14 March 2014 Keywords: Brain metastasis Renal cell carcinoma Prognostic factor
a b s t r a c t Background. – Patients with brain metastasis (BM) from renal cell carcinoma (RCC) have a poorly known prognosis due to the rarity of this disease. The aim of our study was to assess the outcome of patients with a BM due to RCC, and to determine the predictive factors for survival. Methods. – Consecutive patients who underwent treatment between 1997 and 2012 were identified retrospectively from a database (n = 28, median age of 57.8 years, sex ratio M/F: 3.7). Main criteria collected concerned survival time. Other data collected were relative to initial histology, clinical findings at the time of BM diagnosis (diagnosis circumstances, KPS), radiological findings and BM characteristics (number, size and localization), treatment of BM (including surgery, stereotactic radiosurgery [SRS], systemic treatments, whole brain radiotherapy [WBRT]) and the outcome of surgery if performed. Statistical analysis of survival was performed using the Kaplan-Meier method. Results. – Median survival was 13.3 months, 1-year survival was 60.2%, 2-year survival was 16.4%. Univariate analysis showed the existence of intracranial hypertension (P = 0.01), other systemic metastasis (P = 0.049), the absence of deep metastasis (P = 0.03) which are all linked to shorter survival. Age, KPS, initial histology of RCC, number, size, localization, and hemorrhage in BM were not correlated to survival. The median survival in the surgical resection group was 25.3 months versus 8.6 months (P = 0.02). The main criteria for the selection of the surgical group were a single BM (P = 0.04), and superficial metastasis (P = 0.02). Conclusions. – Three predictive factors for longer survival in BMRCC were the absence of intracranial hypertension, the absence of acute metastasis and the absence of extracranial metastasis. Surgical removal, when possible, seems to benefit patient survival. © 2014 Elsevier Masson SAS. All rights reserved.
r é s u m é Mots clés : Métastase cérébrale Cancer du rein Facteur pronostique
Introduction. – Les métastases cérébrales (MC) de cancer du rein (CR) sont des atteintes rares dont le pronostic est mal connu. L’objectif est d’analyser la survie chez les patients porteurs de MC de CR et de déterminer les facteurs prédictifs de survie. Patients et méthodes. – La population est composée d’une série consécutive de 28 patients (âge : 57,8 ; sex-ratio H/F : 3,7) traités pour une MC de CR entre 1997 et 2012. Le recueil rétrospectif de données (histologie du CR, état clinique initial, [circonstances de découverte, score de Karnofsky {KPS}], données radiologiques, traitements [résection chirurgicale, radiochirurgie, thérapies ciblées systémiques et radiothérapie] et survie) a été réalisé. L’analyse statistique de la survie est réalisée par la méthode de Kaplan-Meier. Résultats. – La survie médiane des patients avec MC est de 13,3 mois, la survie à un an est de 60,2 %, et à 2 ans de 16,4 %. Les variables associées à une survie plus longue sont l’absence d’hypertension intracrânienne (p = 0,01), l’absence de métastases systémiques (p = 0,049) et l’absence de métastases profondes (p = 0,03). L’âge, le KPS, l’histologie des CR, le nombre, la taille, le caractère hémorragique des MC ne sont pas
∗ Corresponding author. E-mail address: [email protected] (O. Bennani). 0028-3770/$ – see front matter © 2014 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.neuchi.2013.12.001
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
13
significatifs. La survie dans le groupe traité chirurgicalement est de 25,3 mois contre 8,7 mois dans le groupe non traité (p = 0,02). Les principaux critères de sélection des patients opérés sont l’existence d’une métastase cérébrale unique (p = 0,04) et l’existence d’une métastase superficielle (p = 0,02). Conclusions. – Trois facteurs prédictifs de survie prolongée des MC de CR ont pu être identifiés : l’absence d’hypertension intracrânienne, l’absence de métastases systémiques et l’absence de métastases profondes. Quand la résection chirurgicale est possible, elle semble apporter un gain de survie. © 2014 Elsevier Masson SAS. Tous droits réservés.
1. Introduction Brain metastasis (BM) affects 3.5% [1] to 17% [2] of patients with renal cell carcinoma (RCC). In these patients, median survival varied between 4.1 months [3] and 13.4 months [4] respectively. The armamentarium against these metastases is limited, primarily due to resistance to classical modalities of treatment i.e., chemotherapy [5] and radiotherapy [6]. For these reasons, radiosurgery [7] and surgery have remained an effective conventional treatment proposition [8], despite the resection difficulty related to the vascular nature of the disease [9]. Due to the rarity of the disease, no prospective study has yet been published on this subject. Various prognostic factors have been proposed, including the number of BM [3,10], initial clinical status [10], or the existence of extracerebral metastasis [4]. The most significant prognostic factors isolated in the Radiotherapy and Oncology Group (RTOG) stratification analysis have been used in the form of composite prognostic factor to predict survival and response to radiotherapy and systemic therapy [11]. However, these prognostic factors have been proposed in studies with selective recruitment, based on the access to a surgical, medical or radiotherapy treatment, leading to contradictory results. Thus, much is still needed to enhance our knowledge about the natural history of the BMRCC, the prognostic factors, and the surgical criteria for the resection of BM. The aim of this study was to attempt to determine the outcome of a series of consecutive patients with BM from RCC, as well as the prognostic factors of survival, while analyzing the factors that led to surgical treatment for these patients.
irregularity and nucleolar prominence in the malignant cells and describing the severity of the cytologic malignancy [13]. The prognostic index used to stratify our population was the Disease Specific Graded Prognostic Assessment for RCC (DS-GPA [11]), taking into account 2 prognostic factors: KPS and the number of intracranial metastasis. The score ranges from 0 to 4, 0 being the worst prognostic score and 4 the best prognostic score. Treatments included surgical resection, whole brain radiotherapy (WBRT), systemic targeted treatments (sunitinib, sorafenib, temsirolimus or interferon), stereotactic radio surgery (SRS), a combination of treatments, or a conservative treatment. Quantitative variables (age, number and size of BM, median survival and median delay to BM) were considered as continuous variables with normal distribution, and were expressed as median ± standard deviation (S.D.). Qualitative variables (sex ratio, KPS ≥70, location of BM, DS-GPA, initial T grade ≥ 3, Fuhrman grade ≥3, extracranial metastasis, type of treatments) were expressed in proportion to a confidence interval at 95%). Survival from BM onset was calculated using the Kaplan-Meier method and groups were compared with the log-rank test. The surviving patients were censored at the end of the follow-up period. To find patterns in the surgical decision to treat a BM, we compared patients who did not undergo surgery with patients who underwent surgery. Comparison was performed on all the variables described previously, using Fisher’s exact test. All statistical tests were performed using SPSS 20 (SPSS, Inc., Chicago, IL). Tests were considered statistically significant if P < 0.05. 3. Results
2. Patients and methods 3.1. Population This retrospective longitudinal and observational study included all the consecutive patients who presented with a brain metastasis due to renal cell carcinoma between 1997 and 2012, collected from a computerized database. Exclusion criteria were the incomplete availability of the histological, clinical or radiological data at the time of inclusion. Our primary objective was to attempt to determine the median survival of BMRCC patients. Secondary objectives were the analysis of the prognostic factors affecting survival and the decision concerning surgical treatment. The primary endpoint was survival time, defined as the delay between clinical diagnosis and death. Secondary endpoints were organized into 5 categories: diagnostic circumstances, BM characteristics, RCC histology, prognostic score, and treatments. The diagnostic circumstances included the initial symptomatology: epilepsy, intracranial hypertension (ICH) and focal deficit, and the Karnofsky Performance Status (KPS [12]). The BM characteristics were the number, size (maximum diameter on T1 gadolinium weighed sequence MRI), location (supratentorial, infratentorial, superficial or deep), hemorrhage, and delay of RCC diagnosis (synchronous or metachronous). The RCC histology included the T grade ranked from 1 to 4 according to the locoregional and general extension of the RCC, and the Fuhrman grade ranked 1 to 4 according to nuclear size,
Patient characteristics are outlined in Table 1. Our population included 28 patients with 56 BM (median age: 57.8 ± 8.86 years; range: 41–76 years; sex ratio M/F: 3.7) treated for BMRCC. At the time of the diagnosis, 26 patients (92.8 ± 9.3%) exhibited symptoms associated to their BM and 2 patients were asymptomatic (7.2 ± 9.3%), the discovery of the BM was fortuitous on a brain-imaging follow-up to their cancer. ICH was the only symptom in 6 patients (out of 18 patients with ICH), and was related to a hydrocephalus in 3 cases, an acute hemorrhage in 2 patients, and size, number and edema associated to the BM in the 13 other patients. A neurological deficit was the only sign in 6 patients (out of 18 with a neurological deficit). The deficits presented were motor deficits in 9 patients, a sensory deficit in 2 patients, aphasia in 3 patients, visual impairment in 2 patients, cerebellar ataxia in 3 patients, and frontal lobe disorder in 1 patient. One patient also had a neuroendocrine disorder due to the location of the metastasis in the hypothalamus. Only one patient had an isolated epilepsy that led to the diagnosis of BM. All the patients had an MRI. Brain imaging showed a single metastasis in 13 patients (47.4 ± 18.5%), 2 BM in 6 patients (21.4 ± 15.5%), 3 BM in 7 patients (25 ± 16.04%) and 5 BM in 2 patients (10.7 ± 9.3%) totaling 56 BM, an average of 2 ± 1.16 brain metastasis per patient. The mean size of the BM was 16 ± 9.9 mm.
14
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
Table 1 Baseline characteristics of 28 patients with brain metastasis (BM) from renal cell carcinoma (RCC) with survival and univariate analysis of prognostic factors. Caractéristiques de 28 patients porteurs de metastases cérébrales de cancer du rein (BMRCC), survie globale et analyse univariée des facteurs pronostiques. Variable
Number (percent)
Number of patients
28
Sex ratio M/F
3.7
Mean age of discovery of BM
57.8 years
Diagnosis circumstances Symptoms Epilepsy Intracranial hypertension Neurological deficit KPS ≥70 Brain metastasis characteristics Multiple intracranial metastasis Mean number Mean size Location Supratentorial Infratentorial Multifocal Deep Intratumoral hemorrhage Synchronous metastasis
5 (17.9%) 18 (64.3%) 18 (64.3%) 20 (71.4%)
Median survival from BMRCC discovery in months (S.D.)
P-value (log-rank)
0.97
14.8 (16.2) 7.8 (7.7) 12.8 (11.5) 14.1 (17.9)
0.98 0.01 0.72 0.86
9.1 (7.7)
0.12
18 (64.3%) 4 (14.3%) 6 (21.4%) 6 (21.4%) 9 (32.1%) 4 (14.3%)
13.1 (30.8) 10.9 (13.2) 5.6 (7.9) 4.5 (3.9) 13.6 (13.5) 20.2 (13.7)
0.22
15 (53.6%) 2 ± 1.16 16 ± 9.9 mm
0.03 0.86 0.42
Renal cell carcinoma histology Initial T grade ≥3 Fuhrman grade ≥3 Other systemic metastasis
16 (57.1%)
15 (14)
0.69
19 (67.9%) 20 (71.4%)
14.1 (13.1) 10 (10.6)
0.92 0.049
Disease Specific Graded Prognostic Assessment 0 to 1 2 3 4
6 (21.4%) 11 (39.3%) 9 (32.1%) 2 (7.2%)
7.9 (5.8) 11.2 (12.9) 11.2 (13.2) 50.5 (24.5)
0.24
Treatment Surgical resection of brain metastasis WBRT Targeted treatmenta Combined modalities of treatments
9 (32.1%) 17 (60.7%) 15 (53.6%) 15 (53.6%)
25.3 (22.7) 17.5 (18.3) 13.4 (10.8) 17.6 (19.2)
0.02 0.36 0.94 0.39
KPS: Karnofsky Performance Status; Fuhrman grade: nuclear grade ranked 1 to 4 according to nuclear size, irregularity and nucleolar prominence in the malignant cells; T grade: ranked from 1 to 4 according to the locoregional and general extension of the RCC; WBRT: whole brain radiotherapy. a Systemic treatment is either sunitinib, sorafenib or interferon therapy.
Forty-nine BM were supratentorial (87.5 ± 8.7%), and 7 were infratentorial (12.5 ± 8.7%). BM hemorrhage was found in 9 patients (32.1 ± 17.2%), and was fortuitous in 2 patients, while 4 patients were synchronously diagnosed for BM and RCC (14.3 ± 12.9%). Median time of occurrence for BM was 15 months for metasynchronous metastases. Cytopathology confirmed a clear cell renal carcinoma in all 28 patients. 3.2. Treatment Surgery of the BM was performed in 9 cases (32.1 ± 17.2%). Complications occurred in 3 patients: 1 developed a systemic nosocomial infection, 1 had a new neurological deficit (hemiplegia), 1 had a subdural hematoma and brain edema leading to an uncal herniation causing death 10 days after surgery. One patient died in the post-operative period (11.1%). Two patients had a recurrence of brain metastasis, 10 months and 40 months respectively after brain surgery. WBRT was performed in 17 cases (60.7 ± 18.1%), while SRS was performed in 1 case. Patients received a systemic targeted treatment in 13 cases (46.4 ± 18.4%). The treatment varied with the period of treatment, from interferon in the mid-1990s to sunitinib, sorafenib and temsirolimus more recently. A combination of 2 of these modalities was used in 15 cases (53.6 ± 18.4%).
The comparison between patients treated surgically and the other patients showed 2 significant variables influencing the surgical treatment decision: multiple cerebral metastasis (P = 0.04) and deep metastasis (P = 0.02). Other variables (age, gender, symptoms, KPS, other systemic metastasis, size of the brain metastasis) were non-significant in this instance. Out of the 13 patients who had a single BM, 7 were surgically treated. The other 6 did not receive surgical treatment due to a deep metastasis (2 patients), or a disseminated RCC (4 patients). The arguments used to justify surgically treating the 2 patients presenting multiple BM were ICH resistant to corticosteroids (1 patient) and to determine the diagnosis (i.e. RCC was discovered during brain surgery). The 3 patients who presented with a hydrocephalus were treated surgically: 2 had shunt surgery (without BM removal), 1 patient had an endoscopic ventriculostomy preceding the BM removal. 3.3. Prognosis Median survival from BM onset was 13.3 ± 17months. One-year survival rate was 60.2 ± 9.3% (16 patients), and two-year survival rate was 16.4 ± 7.4% (4 patients) (Fig. 1). As detailed in Table 1, statistically significant variables for enhanced survival shown in the univariate analysis were the
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
Fig. 1. Kaplan-Meier curve showing a median survival of 25.3 months for patients receiving surgical resection of brain metastasis (BM), compared to 8.6 months for patients not receiving surgery. Log-rank P-value was 0.021. Courbe de Kaplan-Meier montrant une survie médiane de 25,3 mois pour les patients recevant un traitement chirurgical, contre 8,6 mois pour les patients ne recevant pas de traitement chirurgical ; p (Log rank) = 0,021.
absence of ICH (P = 0.01), the absence of other systemic metastases (P = 0.049), the absence of deep metastasis (P = 0.03) and surgical resection (P = 0.02, Fig. 1).
15
surgery (SRS) could be of interest for these patients. Extracranial metastasis are frequently found in the context of BMRCC. Vogl et al. [4] demonstrated their prognostic significance for survival in BMRCC in an SRS treated population. Thus, the existence of extracranial metastasis at the time of BM diagnosis should be regarded as a significant negative prognostic factor for BMRCC survival. Analysis of the DS-GPA did not correlate with survival in our study, and the median survivals in our patient population exceed those reported by Sperduto et al. [15] in the RTOG analysis. The discrepancies between this study and the series published in the literature may be explained by the lower number of patients and the surgical setting. In conclusion, our study showed that BMRCC patients had 13.3 months survival since the discovery of their brain disease. Some limitations could be due to the use of a small patient population and the retrospective design of the study. In spite of these limitations, we found three significant factors associated with longer patient survival: i.e. absence of ICH, superficial metastasis, and the absence of other systemic metastasis. Surgery represents an important therapeutic option in patients with a single superficial metastasis, and seems to result in a 13-month gain compared to the rest of the BMRCC population. A larger database is required for multivariate analysis of prognostic factors in surgical patients with BMRCC. The results of our study showed three predictive factors for longer survival in BMRCC: the absence of intracranial hypertension, the absence of acute metastasis and the absence of extracranial metastasis. Also surgical removal, when possible, seems to benefit patient survival.
4. Discussion Disclosure of interest In our study reporting 28 cases of BMRCC at Rouen University Hospital, median survival was 13.3 months from the discovery of the BM and median age of discovery of the BM was 57.8 years. The major variables affecting survival from the discovery of the BM were the existence of ICH (P = 0.01), other systemic metastasis (P = 0.049), the absence of acute metastasis (P = 0.03) and the surgical resection (P = 0.02). The importance of 3 criteria that have influenced the surgical selection of patients should be underlined, thus making the surgical resection a dependent variable: a single brain metastasis, an accessible location and a systematic controlled disease. Considering that the criteria involved in the surgical decision are multiple including intracranial hypertension, number and localization of BM, age, it is subsequently questioned whether BM surgery in patients with RCC may prolong patient survival. The patients who underwent surgery exhibited a marked longer survival (25.7 months) whereas non-operated patients had a shorter survival (8.5 months). Patient selection should take into consideration the number of BM, accessibility, systemic metastasis and other treatments. Our study spanned 15 years from 1997 to 2012, thus the analysis of the targeted treatments do not reflect recent improvements in survival for metastatic RCC. Targeted therapies offer a prolonged survival in BMRCC patients managed with local treatment [14]. Thus, brain surgery should be considered relevant to patients benefiting from targeted therapies. ICH has been reported as a negative prognostic factor in BMRCC. In our patients, ICH was due to a combination of factors: the size of the BM (exceeding 2 cm diameter in 14 patients), the number of BM (2 or more in 10 cases), and edema. Surgery was found to be an effective treatment option in patients with a hydrocephalus and ICH resistant to corticosteroids. Age had little correlation with survival, as suggested by the RTOG database analysis [11,15]. However, it still remains as a limiting factor for surgery, but alternative treatments equivalent to
The authors declare that they have no conflicts of interest concerning this article. References [1] Ljungberg B, Alamdari FI, Rasmuson T, Roos G. Follow-up guidelines for nonmetastatic renal cell carcinoma based on the occurrence of metastases after radical nephrectomy. BJU Int 1999;84:405–11. [2] Levy DA, Slaton JW, Swanson DA, Dinney CP. Stage specific guidelines for surveillance after radical nephrectomy for local renal cell carcinoma. J Urol 1998;159:1163–7. [3] Nieder C, Spanne O, Nordøy T, Dalhaug A. Treatment of brain metastases from renal cell cancer. Urol Oncol Semin 2011;29:405–10. [4] Vogl UM, Bojic M, Lamm W, Frischer JM, Pichelmayer O, Kramer G, et al. Extracerebral metastases determine the outcome of patients with brain metastases from renal cell carcinoma. BMC Cancer 2010;10:480. [5] Yu DS, Ma CP, Chang SY. Establishment and characterization of renal cell carcinoma cell lines with multidrug resistance. Urol Res 2010;28:86–92. [6] Kirova YM, Servois V, Chargari C, Amessis M, Zerbib M, Beuzeboc P. Further developments for improving response and tolerance to irradiation for advanced renal cancer: concurrent (mTOR) inhibitor RAD001 and helical tomotherapy. Invest New Drugs 2012;30:1241–3. [7] Amendola BE, Wolf AL, Coy SR, Amendola M, Bloch L. Brain metastases in renal cell carcinoma: management with gamma knife radiosurgery. Cancer J 2000;6(6):372–6. [8] Wronski M, Arbit E, Russo P, Galicich JH. Surgical resection of brain metastases from renal cell carcinoma in 50 patients. Urology 1996;47:187–93. [9] Bitoh S, Hasegawa H, Ohtsuki H, Obashi J, Fujiwara M, Sakurai M. Cerebral neoplasms initially presenting with massive intracerebral hemorrhage. Surg Neurol 1984;22:57–62. [10] Sperduto PW, Berkey B, Gaspar LE, Mehta M, Curran PW. A new prognostic index and comparison to three other indices for patients with brain metastases: an analysis of 1,960 patients in the RTOG database. Int J Radiat Oncol 2008;70:510–4. [11] Sperduto PW, Chao ST, Sneed PK, Luo PX, Suh PJ, Roberge D, et al. Diagnosisspecific prognostic factors, indexes, and treatment outcomes for patients with newly diagnosed brain metastases: a multi-institutional analysis of 4,259 patients. Int J Radiat Oncol 2010;77:655–61. [12] Karnofsky DA, Burchenal JH. The clinical evaluation of chemotherapeutic agents in cancer. In: MacLeod CM, editor. Evaluation of chemotherapeutic agents. New York (USA): Columbia University Press; 1949. p. 196.
16
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
[13] Fuhrman SA, Lasky LC, Limas C. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 1982;6:55–63. [14] Cochran DC, Chan MD, Aklilu M, Lovato JF, Alphonse NK, Bourland JD, Cochran DC, et al. The effect of targeted agents on outcomes in patients with brain metastases from renal cell carcinoma treated with Gamma Knife surgery. J Neurosurg 2012;116:978–83.
[15] Sperduto PW, Kased N, Roberge D, Xu Z, Shanley R, Luo X, et al. Summary report on the graded prognostic assessment: an accurate and facile diagnosisspecific tool to estimate survival for patients with brain metastases. J Clin Oncol 2012;30:419–25.
Disponible en ligne sur
ScienceDirect www.sciencedirect.com
Original article
Brain metastasis from renal cell carcinoma Métastases cérébrales de cancer du rein O. Bennani a,∗,b , S. Derrey a , O. Langlois a , H. Castel d , A. Laquerriere c , P. Freger a , F. Proust a a
Service de neurochirurgie, CHU de Rouen, 1, rue de Germont, 76031 Rouen cedex, France Service de neurologie, CHU de Rouen, 76031 Rouen, France Service d’anatomopathologie, CHU de Rouen, 76031 Rouen, France d U982 Inserm, Astrocyte and Vascular Niche, DC2N, université de Rouen, 76821 Mont-Saint-Aignan, France b c
a r t i c l e
i n f o
Article history: Received 27 November 2012 Received in revised form 19 November 2013 Accepted 8 December 2013 Available online 14 March 2014 Keywords: Brain metastasis Renal cell carcinoma Prognostic factor
a b s t r a c t Background. – Patients with brain metastasis (BM) from renal cell carcinoma (RCC) have a poorly known prognosis due to the rarity of this disease. The aim of our study was to assess the outcome of patients with a BM due to RCC, and to determine the predictive factors for survival. Methods. – Consecutive patients who underwent treatment between 1997 and 2012 were identified retrospectively from a database (n = 28, median age of 57.8 years, sex ratio M/F: 3.7). Main criteria collected concerned survival time. Other data collected were relative to initial histology, clinical findings at the time of BM diagnosis (diagnosis circumstances, KPS), radiological findings and BM characteristics (number, size and localization), treatment of BM (including surgery, stereotactic radiosurgery [SRS], systemic treatments, whole brain radiotherapy [WBRT]) and the outcome of surgery if performed. Statistical analysis of survival was performed using the Kaplan-Meier method. Results. – Median survival was 13.3 months, 1-year survival was 60.2%, 2-year survival was 16.4%. Univariate analysis showed the existence of intracranial hypertension (P = 0.01), other systemic metastasis (P = 0.049), the absence of deep metastasis (P = 0.03) which are all linked to shorter survival. Age, KPS, initial histology of RCC, number, size, localization, and hemorrhage in BM were not correlated to survival. The median survival in the surgical resection group was 25.3 months versus 8.6 months (P = 0.02). The main criteria for the selection of the surgical group were a single BM (P = 0.04), and superficial metastasis (P = 0.02). Conclusions. – Three predictive factors for longer survival in BMRCC were the absence of intracranial hypertension, the absence of acute metastasis and the absence of extracranial metastasis. Surgical removal, when possible, seems to benefit patient survival. © 2014 Elsevier Masson SAS. All rights reserved.
r é s u m é Mots clés : Métastase cérébrale Cancer du rein Facteur pronostique
Introduction. – Les métastases cérébrales (MC) de cancer du rein (CR) sont des atteintes rares dont le pronostic est mal connu. L’objectif est d’analyser la survie chez les patients porteurs de MC de CR et de déterminer les facteurs prédictifs de survie. Patients et méthodes. – La population est composée d’une série consécutive de 28 patients (âge : 57,8 ; sex-ratio H/F : 3,7) traités pour une MC de CR entre 1997 et 2012. Le recueil rétrospectif de données (histologie du CR, état clinique initial, [circonstances de découverte, score de Karnofsky {KPS}], données radiologiques, traitements [résection chirurgicale, radiochirurgie, thérapies ciblées systémiques et radiothérapie] et survie) a été réalisé. L’analyse statistique de la survie est réalisée par la méthode de Kaplan-Meier. Résultats. – La survie médiane des patients avec MC est de 13,3 mois, la survie à un an est de 60,2 %, et à 2 ans de 16,4 %. Les variables associées à une survie plus longue sont l’absence d’hypertension intracrânienne (p = 0,01), l’absence de métastases systémiques (p = 0,049) et l’absence de métastases profondes (p = 0,03). L’âge, le KPS, l’histologie des CR, le nombre, la taille, le caractère hémorragique des MC ne sont pas
∗ Corresponding author. E-mail address: [email protected] (O. Bennani). 0028-3770/$ – see front matter © 2014 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.neuchi.2013.12.001
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
13
significatifs. La survie dans le groupe traité chirurgicalement est de 25,3 mois contre 8,7 mois dans le groupe non traité (p = 0,02). Les principaux critères de sélection des patients opérés sont l’existence d’une métastase cérébrale unique (p = 0,04) et l’existence d’une métastase superficielle (p = 0,02). Conclusions. – Trois facteurs prédictifs de survie prolongée des MC de CR ont pu être identifiés : l’absence d’hypertension intracrânienne, l’absence de métastases systémiques et l’absence de métastases profondes. Quand la résection chirurgicale est possible, elle semble apporter un gain de survie. © 2014 Elsevier Masson SAS. Tous droits réservés.
1. Introduction Brain metastasis (BM) affects 3.5% [1] to 17% [2] of patients with renal cell carcinoma (RCC). In these patients, median survival varied between 4.1 months [3] and 13.4 months [4] respectively. The armamentarium against these metastases is limited, primarily due to resistance to classical modalities of treatment i.e., chemotherapy [5] and radiotherapy [6]. For these reasons, radiosurgery [7] and surgery have remained an effective conventional treatment proposition [8], despite the resection difficulty related to the vascular nature of the disease [9]. Due to the rarity of the disease, no prospective study has yet been published on this subject. Various prognostic factors have been proposed, including the number of BM [3,10], initial clinical status [10], or the existence of extracerebral metastasis [4]. The most significant prognostic factors isolated in the Radiotherapy and Oncology Group (RTOG) stratification analysis have been used in the form of composite prognostic factor to predict survival and response to radiotherapy and systemic therapy [11]. However, these prognostic factors have been proposed in studies with selective recruitment, based on the access to a surgical, medical or radiotherapy treatment, leading to contradictory results. Thus, much is still needed to enhance our knowledge about the natural history of the BMRCC, the prognostic factors, and the surgical criteria for the resection of BM. The aim of this study was to attempt to determine the outcome of a series of consecutive patients with BM from RCC, as well as the prognostic factors of survival, while analyzing the factors that led to surgical treatment for these patients.
irregularity and nucleolar prominence in the malignant cells and describing the severity of the cytologic malignancy [13]. The prognostic index used to stratify our population was the Disease Specific Graded Prognostic Assessment for RCC (DS-GPA [11]), taking into account 2 prognostic factors: KPS and the number of intracranial metastasis. The score ranges from 0 to 4, 0 being the worst prognostic score and 4 the best prognostic score. Treatments included surgical resection, whole brain radiotherapy (WBRT), systemic targeted treatments (sunitinib, sorafenib, temsirolimus or interferon), stereotactic radio surgery (SRS), a combination of treatments, or a conservative treatment. Quantitative variables (age, number and size of BM, median survival and median delay to BM) were considered as continuous variables with normal distribution, and were expressed as median ± standard deviation (S.D.). Qualitative variables (sex ratio, KPS ≥70, location of BM, DS-GPA, initial T grade ≥ 3, Fuhrman grade ≥3, extracranial metastasis, type of treatments) were expressed in proportion to a confidence interval at 95%). Survival from BM onset was calculated using the Kaplan-Meier method and groups were compared with the log-rank test. The surviving patients were censored at the end of the follow-up period. To find patterns in the surgical decision to treat a BM, we compared patients who did not undergo surgery with patients who underwent surgery. Comparison was performed on all the variables described previously, using Fisher’s exact test. All statistical tests were performed using SPSS 20 (SPSS, Inc., Chicago, IL). Tests were considered statistically significant if P < 0.05. 3. Results
2. Patients and methods 3.1. Population This retrospective longitudinal and observational study included all the consecutive patients who presented with a brain metastasis due to renal cell carcinoma between 1997 and 2012, collected from a computerized database. Exclusion criteria were the incomplete availability of the histological, clinical or radiological data at the time of inclusion. Our primary objective was to attempt to determine the median survival of BMRCC patients. Secondary objectives were the analysis of the prognostic factors affecting survival and the decision concerning surgical treatment. The primary endpoint was survival time, defined as the delay between clinical diagnosis and death. Secondary endpoints were organized into 5 categories: diagnostic circumstances, BM characteristics, RCC histology, prognostic score, and treatments. The diagnostic circumstances included the initial symptomatology: epilepsy, intracranial hypertension (ICH) and focal deficit, and the Karnofsky Performance Status (KPS [12]). The BM characteristics were the number, size (maximum diameter on T1 gadolinium weighed sequence MRI), location (supratentorial, infratentorial, superficial or deep), hemorrhage, and delay of RCC diagnosis (synchronous or metachronous). The RCC histology included the T grade ranked from 1 to 4 according to the locoregional and general extension of the RCC, and the Fuhrman grade ranked 1 to 4 according to nuclear size,
Patient characteristics are outlined in Table 1. Our population included 28 patients with 56 BM (median age: 57.8 ± 8.86 years; range: 41–76 years; sex ratio M/F: 3.7) treated for BMRCC. At the time of the diagnosis, 26 patients (92.8 ± 9.3%) exhibited symptoms associated to their BM and 2 patients were asymptomatic (7.2 ± 9.3%), the discovery of the BM was fortuitous on a brain-imaging follow-up to their cancer. ICH was the only symptom in 6 patients (out of 18 patients with ICH), and was related to a hydrocephalus in 3 cases, an acute hemorrhage in 2 patients, and size, number and edema associated to the BM in the 13 other patients. A neurological deficit was the only sign in 6 patients (out of 18 with a neurological deficit). The deficits presented were motor deficits in 9 patients, a sensory deficit in 2 patients, aphasia in 3 patients, visual impairment in 2 patients, cerebellar ataxia in 3 patients, and frontal lobe disorder in 1 patient. One patient also had a neuroendocrine disorder due to the location of the metastasis in the hypothalamus. Only one patient had an isolated epilepsy that led to the diagnosis of BM. All the patients had an MRI. Brain imaging showed a single metastasis in 13 patients (47.4 ± 18.5%), 2 BM in 6 patients (21.4 ± 15.5%), 3 BM in 7 patients (25 ± 16.04%) and 5 BM in 2 patients (10.7 ± 9.3%) totaling 56 BM, an average of 2 ± 1.16 brain metastasis per patient. The mean size of the BM was 16 ± 9.9 mm.
14
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
Table 1 Baseline characteristics of 28 patients with brain metastasis (BM) from renal cell carcinoma (RCC) with survival and univariate analysis of prognostic factors. Caractéristiques de 28 patients porteurs de metastases cérébrales de cancer du rein (BMRCC), survie globale et analyse univariée des facteurs pronostiques. Variable
Number (percent)
Number of patients
28
Sex ratio M/F
3.7
Mean age of discovery of BM
57.8 years
Diagnosis circumstances Symptoms Epilepsy Intracranial hypertension Neurological deficit KPS ≥70 Brain metastasis characteristics Multiple intracranial metastasis Mean number Mean size Location Supratentorial Infratentorial Multifocal Deep Intratumoral hemorrhage Synchronous metastasis
5 (17.9%) 18 (64.3%) 18 (64.3%) 20 (71.4%)
Median survival from BMRCC discovery in months (S.D.)
P-value (log-rank)
0.97
14.8 (16.2) 7.8 (7.7) 12.8 (11.5) 14.1 (17.9)
0.98 0.01 0.72 0.86
9.1 (7.7)
0.12
18 (64.3%) 4 (14.3%) 6 (21.4%) 6 (21.4%) 9 (32.1%) 4 (14.3%)
13.1 (30.8) 10.9 (13.2) 5.6 (7.9) 4.5 (3.9) 13.6 (13.5) 20.2 (13.7)
0.22
15 (53.6%) 2 ± 1.16 16 ± 9.9 mm
0.03 0.86 0.42
Renal cell carcinoma histology Initial T grade ≥3 Fuhrman grade ≥3 Other systemic metastasis
16 (57.1%)
15 (14)
0.69
19 (67.9%) 20 (71.4%)
14.1 (13.1) 10 (10.6)
0.92 0.049
Disease Specific Graded Prognostic Assessment 0 to 1 2 3 4
6 (21.4%) 11 (39.3%) 9 (32.1%) 2 (7.2%)
7.9 (5.8) 11.2 (12.9) 11.2 (13.2) 50.5 (24.5)
0.24
Treatment Surgical resection of brain metastasis WBRT Targeted treatmenta Combined modalities of treatments
9 (32.1%) 17 (60.7%) 15 (53.6%) 15 (53.6%)
25.3 (22.7) 17.5 (18.3) 13.4 (10.8) 17.6 (19.2)
0.02 0.36 0.94 0.39
KPS: Karnofsky Performance Status; Fuhrman grade: nuclear grade ranked 1 to 4 according to nuclear size, irregularity and nucleolar prominence in the malignant cells; T grade: ranked from 1 to 4 according to the locoregional and general extension of the RCC; WBRT: whole brain radiotherapy. a Systemic treatment is either sunitinib, sorafenib or interferon therapy.
Forty-nine BM were supratentorial (87.5 ± 8.7%), and 7 were infratentorial (12.5 ± 8.7%). BM hemorrhage was found in 9 patients (32.1 ± 17.2%), and was fortuitous in 2 patients, while 4 patients were synchronously diagnosed for BM and RCC (14.3 ± 12.9%). Median time of occurrence for BM was 15 months for metasynchronous metastases. Cytopathology confirmed a clear cell renal carcinoma in all 28 patients. 3.2. Treatment Surgery of the BM was performed in 9 cases (32.1 ± 17.2%). Complications occurred in 3 patients: 1 developed a systemic nosocomial infection, 1 had a new neurological deficit (hemiplegia), 1 had a subdural hematoma and brain edema leading to an uncal herniation causing death 10 days after surgery. One patient died in the post-operative period (11.1%). Two patients had a recurrence of brain metastasis, 10 months and 40 months respectively after brain surgery. WBRT was performed in 17 cases (60.7 ± 18.1%), while SRS was performed in 1 case. Patients received a systemic targeted treatment in 13 cases (46.4 ± 18.4%). The treatment varied with the period of treatment, from interferon in the mid-1990s to sunitinib, sorafenib and temsirolimus more recently. A combination of 2 of these modalities was used in 15 cases (53.6 ± 18.4%).
The comparison between patients treated surgically and the other patients showed 2 significant variables influencing the surgical treatment decision: multiple cerebral metastasis (P = 0.04) and deep metastasis (P = 0.02). Other variables (age, gender, symptoms, KPS, other systemic metastasis, size of the brain metastasis) were non-significant in this instance. Out of the 13 patients who had a single BM, 7 were surgically treated. The other 6 did not receive surgical treatment due to a deep metastasis (2 patients), or a disseminated RCC (4 patients). The arguments used to justify surgically treating the 2 patients presenting multiple BM were ICH resistant to corticosteroids (1 patient) and to determine the diagnosis (i.e. RCC was discovered during brain surgery). The 3 patients who presented with a hydrocephalus were treated surgically: 2 had shunt surgery (without BM removal), 1 patient had an endoscopic ventriculostomy preceding the BM removal. 3.3. Prognosis Median survival from BM onset was 13.3 ± 17months. One-year survival rate was 60.2 ± 9.3% (16 patients), and two-year survival rate was 16.4 ± 7.4% (4 patients) (Fig. 1). As detailed in Table 1, statistically significant variables for enhanced survival shown in the univariate analysis were the
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
Fig. 1. Kaplan-Meier curve showing a median survival of 25.3 months for patients receiving surgical resection of brain metastasis (BM), compared to 8.6 months for patients not receiving surgery. Log-rank P-value was 0.021. Courbe de Kaplan-Meier montrant une survie médiane de 25,3 mois pour les patients recevant un traitement chirurgical, contre 8,6 mois pour les patients ne recevant pas de traitement chirurgical ; p (Log rank) = 0,021.
absence of ICH (P = 0.01), the absence of other systemic metastases (P = 0.049), the absence of deep metastasis (P = 0.03) and surgical resection (P = 0.02, Fig. 1).
15
surgery (SRS) could be of interest for these patients. Extracranial metastasis are frequently found in the context of BMRCC. Vogl et al. [4] demonstrated their prognostic significance for survival in BMRCC in an SRS treated population. Thus, the existence of extracranial metastasis at the time of BM diagnosis should be regarded as a significant negative prognostic factor for BMRCC survival. Analysis of the DS-GPA did not correlate with survival in our study, and the median survivals in our patient population exceed those reported by Sperduto et al. [15] in the RTOG analysis. The discrepancies between this study and the series published in the literature may be explained by the lower number of patients and the surgical setting. In conclusion, our study showed that BMRCC patients had 13.3 months survival since the discovery of their brain disease. Some limitations could be due to the use of a small patient population and the retrospective design of the study. In spite of these limitations, we found three significant factors associated with longer patient survival: i.e. absence of ICH, superficial metastasis, and the absence of other systemic metastasis. Surgery represents an important therapeutic option in patients with a single superficial metastasis, and seems to result in a 13-month gain compared to the rest of the BMRCC population. A larger database is required for multivariate analysis of prognostic factors in surgical patients with BMRCC. The results of our study showed three predictive factors for longer survival in BMRCC: the absence of intracranial hypertension, the absence of acute metastasis and the absence of extracranial metastasis. Also surgical removal, when possible, seems to benefit patient survival.
4. Discussion Disclosure of interest In our study reporting 28 cases of BMRCC at Rouen University Hospital, median survival was 13.3 months from the discovery of the BM and median age of discovery of the BM was 57.8 years. The major variables affecting survival from the discovery of the BM were the existence of ICH (P = 0.01), other systemic metastasis (P = 0.049), the absence of acute metastasis (P = 0.03) and the surgical resection (P = 0.02). The importance of 3 criteria that have influenced the surgical selection of patients should be underlined, thus making the surgical resection a dependent variable: a single brain metastasis, an accessible location and a systematic controlled disease. Considering that the criteria involved in the surgical decision are multiple including intracranial hypertension, number and localization of BM, age, it is subsequently questioned whether BM surgery in patients with RCC may prolong patient survival. The patients who underwent surgery exhibited a marked longer survival (25.7 months) whereas non-operated patients had a shorter survival (8.5 months). Patient selection should take into consideration the number of BM, accessibility, systemic metastasis and other treatments. Our study spanned 15 years from 1997 to 2012, thus the analysis of the targeted treatments do not reflect recent improvements in survival for metastatic RCC. Targeted therapies offer a prolonged survival in BMRCC patients managed with local treatment [14]. Thus, brain surgery should be considered relevant to patients benefiting from targeted therapies. ICH has been reported as a negative prognostic factor in BMRCC. In our patients, ICH was due to a combination of factors: the size of the BM (exceeding 2 cm diameter in 14 patients), the number of BM (2 or more in 10 cases), and edema. Surgery was found to be an effective treatment option in patients with a hydrocephalus and ICH resistant to corticosteroids. Age had little correlation with survival, as suggested by the RTOG database analysis [11,15]. However, it still remains as a limiting factor for surgery, but alternative treatments equivalent to
The authors declare that they have no conflicts of interest concerning this article. References [1] Ljungberg B, Alamdari FI, Rasmuson T, Roos G. Follow-up guidelines for nonmetastatic renal cell carcinoma based on the occurrence of metastases after radical nephrectomy. BJU Int 1999;84:405–11. [2] Levy DA, Slaton JW, Swanson DA, Dinney CP. Stage specific guidelines for surveillance after radical nephrectomy for local renal cell carcinoma. J Urol 1998;159:1163–7. [3] Nieder C, Spanne O, Nordøy T, Dalhaug A. Treatment of brain metastases from renal cell cancer. Urol Oncol Semin 2011;29:405–10. [4] Vogl UM, Bojic M, Lamm W, Frischer JM, Pichelmayer O, Kramer G, et al. Extracerebral metastases determine the outcome of patients with brain metastases from renal cell carcinoma. BMC Cancer 2010;10:480. [5] Yu DS, Ma CP, Chang SY. Establishment and characterization of renal cell carcinoma cell lines with multidrug resistance. Urol Res 2010;28:86–92. [6] Kirova YM, Servois V, Chargari C, Amessis M, Zerbib M, Beuzeboc P. Further developments for improving response and tolerance to irradiation for advanced renal cancer: concurrent (mTOR) inhibitor RAD001 and helical tomotherapy. Invest New Drugs 2012;30:1241–3. [7] Amendola BE, Wolf AL, Coy SR, Amendola M, Bloch L. Brain metastases in renal cell carcinoma: management with gamma knife radiosurgery. Cancer J 2000;6(6):372–6. [8] Wronski M, Arbit E, Russo P, Galicich JH. Surgical resection of brain metastases from renal cell carcinoma in 50 patients. Urology 1996;47:187–93. [9] Bitoh S, Hasegawa H, Ohtsuki H, Obashi J, Fujiwara M, Sakurai M. Cerebral neoplasms initially presenting with massive intracerebral hemorrhage. Surg Neurol 1984;22:57–62. [10] Sperduto PW, Berkey B, Gaspar LE, Mehta M, Curran PW. A new prognostic index and comparison to three other indices for patients with brain metastases: an analysis of 1,960 patients in the RTOG database. Int J Radiat Oncol 2008;70:510–4. [11] Sperduto PW, Chao ST, Sneed PK, Luo PX, Suh PJ, Roberge D, et al. Diagnosisspecific prognostic factors, indexes, and treatment outcomes for patients with newly diagnosed brain metastases: a multi-institutional analysis of 4,259 patients. Int J Radiat Oncol 2010;77:655–61. [12] Karnofsky DA, Burchenal JH. The clinical evaluation of chemotherapeutic agents in cancer. In: MacLeod CM, editor. Evaluation of chemotherapeutic agents. New York (USA): Columbia University Press; 1949. p. 196.
16
O. Bennani et al. / Neurochirurgie 60 (2014) 12–16
[13] Fuhrman SA, Lasky LC, Limas C. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 1982;6:55–63. [14] Cochran DC, Chan MD, Aklilu M, Lovato JF, Alphonse NK, Bourland JD, Cochran DC, et al. The effect of targeted agents on outcomes in patients with brain metastases from renal cell carcinoma treated with Gamma Knife surgery. J Neurosurg 2012;116:978–83.
[15] Sperduto PW, Kased N, Roberge D, Xu Z, Shanley R, Luo X, et al. Summary report on the graded prognostic assessment: an accurate and facile diagnosisspecific tool to estimate survival for patients with brain metastases. J Clin Oncol 2012;30:419–25.
