Prostate cancer accounted for an estimated 1,100,000 new cases and 300,000 deaths in 2012, and globally it is the second most common malignancy among men.1 The natural history of prostate cancer is variable, and when metastatic it most commonly involves the regional lymph nodes, bones, liver, or lungs.2,3
Intracranial dissemination is considered a rare event, estimated to occur in 0.63% of prostate cancer patients.4 However, with the introduction of new therapies, an increase in the frequency of prostate cancer brain metastases from 0.8% to 2.8% has been reported.5 This increase was postulated to be related to increasing survival of those with advanced cancer5 and the inability of new therapies like docetaxel to cross the blood-brain barrier.6,7
Historically, patients with intracranial prostate metastases have been treated with whole-brain radiation therapy (WBRT) alone,4,8,9 resulting in a median survival of 4–9 months,4,9 or resection and WBRT, with a mean survival of 9.2 months reported.10 Nevertheless, intracranial dissemination of prostate cancer usually occurs late in the disease, and resection of a brain metastasis in prostate cancer patients may not convey a favorable benefit-to-risk profile for many such patients. Stereotactic radiosurgery (SRS) has been proven to be a safe and effective management option for brain metastases from other more common primary malignancies, including melanoma, lung, and breast. Moreover, SRS has shown promise in extending the overall survival (OS) of patients with intracranial metastases from prostate cancer, with three small studies reporting survivals of 9–13 months.4,11,12 However, there have been no large studies to date investigating the role of SRS for brain metastasis from prostate cancer.4,11,12
In this international multicenter study, we investigated the safety and efficacy of SRS in the management of prostate cancer brain metastases. Prognostic factors and treatment outcomes were also determined following SRS treatment.
Methods
Data were collected for patients from nine institutions treated with SRS using the Gamma Knife (Elekta AB) for cerebral metastases from prostate carcinoma from July 2005 to June 2020. All centers obtained local institutional review board approval to participate in the study, and patient or next of kin consent was obtained when required. The pooled data were then de-identified and logged in a spreadsheet. An International Gamma Knife Research Foundation (IGKRF) coordinator verified data completeness and compliance with patient protection. The data were then sent to the institution of the first and last author for further analysis. Inconsistencies in the data were addressed by the participating centers.
Inclusion criteria for the study included the following: 1) diagnosis of prostate cancer, 2) age > 18 years, 3) no other malignancy, and 4) SRS of brain metastases.
Clinical Assessment
Data collected included patient demographics and primary tumor characteristics, including prostate cancer histology, date of diagnosis, and treatments received. Regarding the intracranial metastatic disease, data were collected on the time from prostate cancer diagnosis to intracranial dissemination diagnosis; the presence of neurological signs and symptoms; the patient Karnofsky Performance Scale (KPS) score and extent of extracranial disease at SRS; the number, volume, and maximum diameter of intracranial metastases treated; and prior or concurrent treatments, including concurrent systemic therapy, and WBRT.
SRS Technique
All patients were treated using either frame-based SRS or a mask-based, frameless stereotactic radiotherapy approach. Radiosurgery was delivered via a Gamma Knife platform (Elekta AB), although the Gamma Knife model depended on the year of treatment and availability at each institution. Prior to treatment, all plans were reviewed and approved by the radiosurgery team at each participating institution.
The SRS treatment plans were reviewed, and radiosurgical treatment parameters were collected, including margin dose, maximum dose, number of fractions, and time interval between fractions.
Radiological and Clinical Follow-Up
Patients were followed clinically and radiographically at approximately 2- to 3-month intervals after SRS. Neuroimaging was evaluated by the treating neurosurgeon or a neuroradiologist. Follow-up data collected included local control, date of local or distant brain metastatic disease progression, salvage therapy, OS, date of last imaging and clinical follow-up, date and cause of death, and toxicity such as radionecrosis. The Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 definitions were used for scoring radionecrosis.13 OS was defined as the time period from the day of prostate cancer diagnosis to last follow-up or death. Intracranial progression-free survival was measured from the day of first SRS treatment to progression of intracranial disease or death or last follow-up. Local tumor control was defined as the period from SRS to neuroimaging-documented progression of the treated lesion, and local SRS failure was defined as a radiographic increase in tumor size > 10% after SRS.
Statistical Analysis
Data for continuous variables are presented as median and range or mean ± SD and for categorical variables as frequencies and percentages. Kaplan-Meier plots of overall and progression-free survival from the date of SRS were constructed. Log-rank tests were used to investigate possible differences between these curves after stratification by various known prognostic factors such as age and KPS at SRS, number of intracranial metastases, active extracranial disease and presence and location of extracranial metastases, and margin dose. In addition, we analyzed for an association between clinical outcomes and potential prognostic factors using univariate and multivariate analysis by Cox regression analysis. Statistical significance was defined as p < 0.05, and statistics were calculated using R programming (R Core Team [2016]; R Foundation for Statistical Computing, https://www.R-project.org).
Results
Patient and Cancer Attributes
A total of 46 patients were treated in 51 SRS procedures for 120 prostate cancer intracranial metastases. All patients had a tissue diagnosis of prostate cancer. The histological type of the primary prostate cancer was available for 35 patients and included 33 adenocarcinomas, 1 tumor of neuroendocrine histology, and 1 small cell carcinoma. The mean patient age at the time of prostate cancer diagnosis was 62.82 ± 8.95 years. The mean patient age at radiosurgery was 68.04 ± 9.05 years, and the mean time interval from prostate cancer diagnosis to SRS was 4.82 ± 4.89 years (Table 1). Data on extracranial disease activity at SRS were available for 42 (91.3%) patients. The primary prostate cancer was reported to be well controlled in 8 (17.4%) patients. At SRS, extracranial dissemination was noted in 34 (73.9%) patients, including 24 with single-organ involvement (18 patients with osseous, 5 with pulmonary, and 1 with hepatic involvement) and 8 patients with involvement of two organ systems (7 patients with osseous and pulmonary metastases and 1 patient with hepatic and osseous metastases). The median patient KPS score at SRS was 80 (range 50–90), and neurological symptoms attributed to intracranial involvement were present prior to 39 (76%) SRS procedures (Table 1).
Patient demographics and primary tumor characteristics
| Value | |
|---|---|
| No. of pts | 46 |
| Pt age at initial prostate cancer diagnosis, yrs | 62.82 ± 8.95 |
| Pt age at SRS, yrs | 68.04 ± 9.05 |
| Time from initial diagnosis to SRS treatment, yrs | 4.82 ± 4.59 |
| KPS score at SRS | 80 (50–90) |
| Neurological symptoms present before each SRS session | 39 (76.47%) |
| Prostate cancer histopathology | |
| Adenocarcinoma | 33 (71.74%) |
| Nonadenocarcinoma | 2 (4.35%) |
| Neuroendocrine carcinoma | 1 (2.17%) |
| Small cell carcinoma | 1 (2.17%) |
| Unknown type | 11 (23.91%) |
Pt = patient.
Values are presented as number (%) of patients, mean ± SD, or median (range) unless otherwise indicated.
Other Therapies for Extracranial and Intracranial Prostate Cancer
Eighteen of 46 patients in our series underwent prostatectomy, 11 of whom received adjuvant radiation therapy to the prostate, while radiation therapy alone was administered to 11 patients. In our series, prior to SRS, 10 patients underwent craniotomy and resection of a metastatic lesion and 1 patient underwent transnasal biopsy of a parasellar metastasis. Three patients received WBRT prior to and 2 patients received a WBRT boost immediately after the SRS procedure (Table 2). Antiandrogen therapies (ADTs) were administered at some point of the disease to 24 patients. At SRS, 11 patients were reported to have castration-resistant prostate cancer, 9 were receiving ADTs, and 4 were receiving systemic chemotherapy.
SRS treatment parameters, intracranial metastatic tumor characteristics, and response of metastases to radiosurgery and other treatments
| Value | |
|---|---|
| Intracranial metastases treated per SRS session | |
| ≤4 | 46 (90.20%) |
| >4 | 5 (9.80%) |
| Metastasis characteristics | |
| Total no. treated | 120 |
| Diameter, mm | 18.39 ± 13.43 |
| Vol, ml | 5.50 ± 7.41 |
| Radiation dose per metastasis treated, Gy | |
| Margin | 18 (6–28) |
| Max | 36 (20–75) |
| Lesion response to SRS at last follow-up | |
| Lesions w/ radiologically observed response | 82/120 (68.3%) |
| Size unchanged (<10% variance) | 12/82 (14.63%) |
| Size decreased | 56/82 (68.29%) |
| Size increased | 14/82 (17.07%) |
| Lesions w/ unknown response | 38/120 (31.66%) |
| Op, no. of pts | |
| Craniotomy & resection of metastatic lesion | 10 |
| Biopsy | 1 |
| WBRT timing relative to SRS, no. of pts | |
| Before | 3 |
| After | 2 |
Values are presented as number (%) of metastases, mean ± SD, or median (range) unless otherwise indicated. Local SRS failure was defined as a radiographic increase in tumor size > 10% after SRS.
SRS Parameters
Single-session SRS was utilized in all but 2 patients. Of these, 1 patient was managed with 5-fraction SRS (margin dose 6 Gy/fraction). The second patient underwent treatment three times, including one single-session SRS procedure, one 3-fraction SRS procedure (margin dose 9 Gy/fraction), and one 5-fraction procedure for 4 metastases (margin dose 6 Gy/fraction). The median diameter of the 120 treated lesions was 17 (range 0.32–60) mm. The median maximum dose was 36 (range 20–75) Gy, and the median margin dose was 18 (range 6–28) Gy (Table 2).
Local Tumor Response
The radiological response to SRS was available for 82 (68.3%) of 120 treated metastatic lesions (Table 2). The mean radiological follow-up duration was 16.12 ± 28.97 months, and the median radiological follow-up was 6 (range 0–156) months. For the 82 lesions with available radiological response at the last imaging follow-up, 56 (68.29%) decreased in size and 12 (14.63%) remained stable (Table 2). Thus, local control defined as tumor decrease or stability was achieved in 82.92% of tumors (Fig. 1). The post-SRS 1-year intracranial progression-free survival probability was 44.1% (95% CI 28.2%–69.2%) (Fig. 2). At the last imaging follow-up, 11 patients were noted to have local progression of 14 (17.07%) lesions. Prior to SRS, neurological symptoms attributed to the metastases were present in 10 (90.91%) of 11 and active extracranial disease was noted in 7 (63.64%) of 11 of these patients. The mean diameter of these lesions was 15.07 ± 11.74 mm, and the mean lesion volume was 3.19 ± 4.47 ml. The mean margin dose to lesions that progressed after SRS was 16.53 ± 4.53 Gy. Local progression of the 14 (17.07%) treated lesions was subsequently managed with repeat SRS (n = 11), WBRT (n = 2), or systemic therapy (n = 1).
Kaplan-Meier curve demonstrating the time period of local control per lesion from SRS treatment.
Kaplan-Meier curve demonstrating patient intracranial progression-free survival from SRS treatment to last follow-up or death.
Clinical Outcomes After SRS
Clinical follow-up data were available for 41 patients. The mean clinical follow-up was 19.6 ± 29.62 months, and the median follow-up was 6 (range 0–156) months. Neurological symptoms attributed to the target lesion(s) were reported prior to 39 SRS procedures. According to Kaplan-Meier calculations, the median OS from time of prostate cancer diagnosis was 75.3 (range 29.9–169.1) months, and the median survival following SRS was 15.18 (range 6.51–43.04) months (Fig. 3).
Kaplan-Meier curve demonstrating OS from SRS treatment.
An improved KPS score at SRS was noted in univariate analysis (p = 0.0009) and multivariate analysis (p = 0.02) to be associated with improved OS and progression-free survival (p = 0.03). In multivariate analysis, age ≥ 65 years at SRS was noted to be associated with decreased OS (p = 0.04). SRS was well tolerated and no grade 3–5 toxicities were noted. One patient required steroid treatment due to grade 2 radiation toxicity manifesting as fluctuating mental status. At the last follow-up, 29 (63%) patients had died. Causes of death included progressive systemic disease in 14 (48.3%), progressive neurological disease in 3 (10.3%), and sepsis in 2 (6.9%) patients. The cause of death for 10 (34.4%) patients was unknown (Table 3).
Patient outcomes
| Value | |
|---|---|
| Pretreatment neurological symptoms post-SRS (n = 39 SRS sessions) | |
| Resolved | 16 (41.03%) |
| Remained | 15 (38.46%) |
| Unknown | 8 (20.51%) |
| Post-SRS follow-up & survival rates | |
| Clinical follow-up, mos | 19.6 ± 29.62 |
| Radiological follow-up, mos | 16.12 ± 28.97 |
| 1-yr post-SRS survival rate (n = 41 pts) | 23 (56.1%) |
| Cause of death (n = 29 pts) | |
| Progressive neurological disease | 3 (10.34%) |
| Progressive systemic disease | 14 (48.29%) |
| Sepsis | 2 (6.9%) |
| Unknown | 10 (34.48%) |
Values are presented as number (%) or mean ± SD.
An improved KPS score at SRS was noted in univariate analysis (p = 0.0009) and multivariate analysis (p = 0.02) to be associated with improved OS and progression-free survival (p = 0.03). In multivariate analysis, age ≥ 65 years at SRS was noted to be associated with decreased OS (p = 0.04). SRS was well tolerated and no grade 3–5 toxicities were noted. One patient required steroid treatment due to grade 2 radiation toxicity manifesting as fluctuating mental status. At the last follow-up, 29 (63%) patients had died. Causes of death included progressive systemic disease in 14 (48.3%), progressive neurological disease in 3 (10.3%), and sepsis in 2 (6.9%) patients. The cause of death for 10 (34.4%) patients was unknown (Table 3).
Discussion
Brain metastases were historically reported to occur in < 1% of prostate cancer patients.4,5 However, since the introduction of docetaxel, which was followed by the chemotherapeutic agent cabazitaxel, the androgen deprivation agent abiraterone,14 the androgen receptor blocker enzalutamide,15 the vaccine Sipuleucel-T,16 and radium-223,17 the incidence of prostate cancer brain metastases is expected to increase as patients have more years at risk with their disease. Indeed, an increase in the frequency of prostate cancer brain metastases from 0.8% in the pre-docetaxel era to 2.8% in the post-docetaxel era was noted in an Italian study.5
Prostate cancer intracranial dissemination usually occurs at the late phase of the disease. It most commonly involves the epidural space as a result of local invasion from skull metastases.10 However, since the dura serves as a barrier to cancer cell dissemination, intradural dissemination is most likely the result of hematogenous spread with two main potential mechanisms suggested. First, intradural metastases occur as a single-step process in which metastatic prostate cancer cells gain access to the intradural compartment via the paravertebral venous plexus. Second, spread to the intradural compartment occurs due to secondary hematogenous spread from a primary osseous or pulmonary metastasis. The latter mechanism is supported by the development of intradural metastases late in the disease and the presence of active systemic disease in most prostate cancer patients at the time of brain metastasis diagnosis.10 Thus, due to poor OS and evidence of active extracranial disease, the majority of prostate cancer patients presenting with brain metastases do not undergo open cranial resection.18 Without treatment, a median survival of 1 month was reported in a series of prostate cancer patients with brain metastasis.4 WBRT is associated with improved outcomes, with one systemic review reporting a median survival ranging from 4 to 9 months.18
SRS is a minimally invasive procedure that has been proven to be safe and effective in the management of brain metastases from cancer types that more commonly metastasize to the brain. However, data concerning the treatment of prostate cancer brain metastasis with SRS are limited to three small studies4,11,12 (Table 4). In an analysis from the Cleveland Clinic, 5 patients with prostate cancer brain metastases treated with SRS were reported.11 Four had a single metastasis, while 1 patient had 4 metastases. Three patients received SRS alone, 1 received both WBRT and SRS, and 1 patient received a combination of surgery, WBRT, and SRS. All patients were initially symptomatic, but all improved neurologically and in KPS score at 3 months post-SRS. In this small case series, local intracranial control was 100%, with no new intracranial lesions reported. The same series also reported a mean OS of 10.0 ± 6.7 months, with 2 patients dying from conditions unrelated to prostate cancer, 2 dying from systemic progression, and 1 patient alive and asymptomatic at last follow-up.11 The second published study included 10 patients with intracranial metastatic prostate cancer treated with SRS.12 In total, 15 brain metastases were treated: 9 dural-based and 6 parenchymal. Six patients had a single metastasis and 4 patients had multiple metastases. Eight patients carried the diagnosis of adenocarcinoma, while a subset of 2 patients had small cell carcinoma. The patients received a combination of SRS, partial brain radiation therapy, WBRT, and surgery. A median margin SRS dose of 16 Gy yielded a local control rate of 85%, which included patients receiving upfront and boost SRS, as well as those undergoing salvage SRS. In 2 patients, disease progressed despite fractionated partial brain radiation therapy. One of 3 patients subsequently developed new distant brain metastases, and 1 patient needed repeat SRS for recurrence. Overall, the median survival in that series was 13 months.12 Tremont-Lukats et al. reported a median survival of 9 months in a small series of 5 patients managed with SRS for brain metastatic disease from prostate cancer.4
Series of patients treated with radiosurgery for prostate cancer brain metastases
| Authors & Year | No. of Pts | Histology | Total Brain Metastases Treated | Mean Tumor Vol | Median Margin Dose | 12-Mo Local Control | Overall Survival (mos) | Severe Toxicity (grade 3–5) |
|---|---|---|---|---|---|---|---|---|
| Tremont-Lukats et al., 20034 | 5 | NR | NR | NR | NR | NR | 9* | None |
| Kim et al., 200811 | 5 | AD | 8 | 3.1 ml | NR | 100% | 10† | None |
| Flannery et al., 201012 | 10 | 8 AD, 2 small cell carcinoma | 15 | 7.7 ml | 16 Gy | 85% | 13* | None |
| Present study | 46 | 33 AD, 2 non-AD | 120 | 5.50 ml | 18 Gy | 83% | 15* | None |
AD = adenocarcinoma; NR = not reported.
Median OS.
Mean OS.
Our study is the largest and only multicenter SRS study for prostate cancer patients with brain metastasis. It confirms the reported effectiveness of SRS in the management of prostate cancer patients with brain metastases. The median patient survival in our series was 15.18 (range 6.51–43.04) months, and at last follow-up, local tumor control was seen in 82.9% (68/82) of the treated lesions. A KPS score > 70 was noted to be associated with improved OS (p = 0.02) and improved local control of the treated lesion (p = 0.03). Age ≥ 65 years at SRS (p = 0.04) and active extracranial disease (p = 0.03) were associated with decreased OS and distant intracranial disease progression, respectively. SRS for brain metastases from prostate cancer appears to be a safe and well-tolerated procedure with a low risk of complications reported in the literature.4,11,12 In our series, there was 1 patient who developed grade 2 radiation toxicity that was managed with corticosteroids.
Study Limitations
Limitations of this multicenter study include its retrospective nature, patient selection bias, and limited statistical power. The lack of imaging follow-up for 38 of 120 SRS-treated cerebral metastases represents another limitation of our study. Nevertheless, clinical and imaging follow-up were available for 41 (89%) of 46 patients and for 82 cerebral metastases, respectively. Thus, the current study represents the largest to date in a relatively rare cancer histology for brain metastasis formation.
Conclusions
As novel systemic therapies are more regularly employed, brain metastases from prostate cancer will likely increase in frequency. SRS appears to be a safe, well-tolerated, and overall effective management option for patients with prostate cancer intracranial metastases. A median survival of 15 months and local intracranial control were achieved in the vast majority of brain metastases treated with SRS. There were no serious grade 3–5 toxicities noted. Further, well-designed studies evaluating the efficacy of SRS in combination with adjuvant treatments are necessary to improve outcomes of patients with intracranial dissemination from prostate cancer.
Acknowledgments
This research study (IRRF-02-04-2020) was supported by a grant from the International Radiosurgery Research Foundation (IRRF).
Disclosures
Dr. Zacharia: speakers bureau, NICO Corp.; consultant, Medtronic. Dr. Lunsford: direct stock ownership, Elekta AB; consultant, Insightec Data Safety Monitoring Board.
Author Contributions
Conception and design: Sheehan. Acquisition of data: Pikis, Bunevicius, Lee, Yang, Zacharia, Liščák, Simonova, Tripathi, Kumar, Mathieu, Perron, Peker, Samanci, Gurewitz, Bernstein, Kondziolka, Niranjan. Analysis and interpretation of data: Sheehan, Pikis, Bunevicius, Mantziaris. Drafting the article: Sheehan, Pikis. Critically revising the article: Sheehan, Pikis, Lee, Yang, Zacharia, Liščák, Simonova, Tripathi, Mathieu, Peker, Kondziolka, Niranjan, Lunsford. Reviewed submitted version of manuscript: Sheehan, Pikis, Bunevicius, Lee, Yang, Zacharia, Liščák, Simonova, Tripathi, Kumar, Mathieu, Peker, Samanci, Gurewitz, Bernstein, Kondziolka, Niranjan, Lunsford. Approved the final version of the manuscript on behalf of all authors: Sheehan. Statistical analysis: Mantziaris. Study supervision: Sheehan.
References
- 3↑
Gandaglia G, Abdollah F, Schiffmann J, et al. Distribution of metastatic sites in patients with prostate cancer: a population-based analysis. Prostate. 2014;74(2):210–216.
- 4↑
Tremont-Lukats IW, Bobustuc G, Lagos GK, et al. Brain metastasis from prostate carcinoma: The M. D. Anderson Cancer Center experience. Cancer. 2003;98(2):363–368.
- 5↑
Caffo O, Veccia A, Fellin G, et al. Frequency of brain metastases from prostate cancer: an 18-year single-institution experience. J Neurooncol. 2013;111(2):163–167.
- 6↑
van der Veldt AA, Hendrikse NH, Smit EF, et al. Biodistribution and radiation dosimetry of 11C-labelled docetaxel in cancer patients. Eur J Nucl Med Mol Imaging. 2010;37(10):1950–1958.
- 7↑
Sanson M, Napolitano M, Yaya R, et al. Second line chemotherapy with docetaxel in patients with recurrent malignant glioma: a phase II study. J Neurooncol. 2000;50(3):245–249.
- 8↑
McCutcheon IE, Eng DY, Logothetis CJ. Brain metastasis from prostate carcinoma: antemortem recognition and outcome after treatment. Cancer. 1999;86(11):2301–2311.
- 9↑
Lawton A, Sudakoff G, Dezelan LC, Davis N. Presentation, treatment, and outcomes of dural metastases in men with metastatic castrate-resistant prostate cancer: a case series. J Palliat Med. 2010;13(9):1125–1129.
- 10↑
Salvati M, Frati A, Russo N, et al. Brain metastasis from prostate cancer. Report of 13 cases and critical analysis of the literature. J Exp Clin Cancer Res. 2005;24(2):203–207.
- 11↑
Kim SH, Chao ST, Toms SA, et al. Stereotactic radiosurgical treatment of parenchymal brain metastases from prostate adenocarcinoma. Surg Neurol. 2008;69(6):641–646.
- 12↑
Flannery T, Kano H, Niranjan A, et al. Stereotactic radiosurgery as a therapeutic strategy for intracranial metastatic prostate carcinoma. J Neurooncol. 2010;96(3):369–374.
- 13↑
National Cancer Institute. Common Terminology Criteria for Adverse Events (CTCAE). Version 5.0. Accessed June 3, 2021. https://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/CTCAE_v5_Quick_Reference_8.5x11.pdf
- 14↑
Fizazi K, Scher HI, Molina A, et al. Abiraterone acetate for treatment of metastatic castration-resistant prostate cancer: final overall survival analysis of the COU-AA-301 randomised, double-blind, placebo-controlled phase 3 study. Lancet Oncol. 2012;13(10):983–992.
- 15↑
Scher HI, Fizazi K, Saad F, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;367(13):1187–1197.
- 16↑
Kantoff PW, Higano CS, Shore ND, et al. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med. 2010;363(5):411–422.
- 17↑
Parker C, Nilsson S, Heinrich D, et al. Alpha emitter radium-223 and survival in metastatic prostate cancer. N Engl J Med. 2013;369(3):213–223.
- 18↑
Sita TL, Petras KG, Wafford QE, et al. Radiotherapy for cranial and brain metastases from prostate cancer: a systematic review. J Neurooncol. 2017;133(3):531–538.
