Brain metastases

Abstract

Introduction

Metastases to the central nervous system may occur with tumours of any primary origin. Brain (cerebral) metastases may be either single or multiple, with or without disseminated disease elsewhere. Brain metastases may present with focal or generalised symptoms, although up to a third of patients may be asymptomatic.

Methods and outcomes

We conducted a systematic review and aimed to answer the following clinical question: What are the effects of interventions for managing brain metastases in adults? We searched: Medline, Embase, The Cochrane Library, and other important databases up to June 2007 (Clinical Evidence reviews are updated periodically, please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare Regulatory Agency (MHRA).

Results

We identified 18 systematic reviews, RCTs, or observational studies that met our inclusion criteria. We have performed a GRADE evaluation of the quality of evidence for interventions included in this review.

Conclusions

In this systematic review we present information relating to the effectiveness and safety of the following interventions: corticosteroids; cytotoxic chemotherapy (systemic); radiation sensitisers plus whole-brain radiotherapy (external beam); surgery; radiosurgery; surgery plus radiosurgery; surgery plus radiosurgery plus whole-brain radiotherapy (external beam); surgery plus whole-brain radiotherapy (external beam); whole-brain radiotherapy (external beam); and whole-brain radiotherapy plus radiosurgery.

Key Points

Brain (cerebral) metastases may be either solitary or multiple, with or without disseminated disease elsewhere.

They may present with focal or generalised symptoms, although up to a third of people may be asymptomatic.

• Headache is the most common presenting symptom. Focal weakness, mental change, and seizures are also common.

The incidence of brain metastases is between 8-11/100,000 people a year.

• The lung is the most common primary site for brain metastases.

This review only includes adults with brain metastases (cerebral hemispheres and posterior fossa structures) confirmed with a biopsy or by computed tomography or magnetic resonance imaging.

We found no direct evidence comparing corticosteroids versus no corticosteroids. Such RCTs are unlikely to be undertaken.

• Although we found no direct evidence, there is consensus that corticosteroids are effective for the relief of symptoms.

Whole brain radiotherapy (external beam) (WBRT) may be effective in some selected people with brain metastases.

• However, there are adverse effects associated with the use of WBRT, which need to be weighed against any potential benefits on an individual basis.

We don't know whether WBRT plus radiosurgery is more effective than WBRT alone at improving survival in people with between one and four brain metastases.

• However, subgroup analysis in one large RCT found that, in people with a single unresectable brain metastasis, WBRT plus radiosurgery may increase median survival compared with WBRT alone.

We found insufficient evidence on the effects of systemic cytotoxic chemotherapy, surgery, surgery plus WBRT, and radiosurgery.

We don't know whether surgery plus radiosurgery or surgery plus radiosurgery plus WBRT are effective as we found no evidence of their effects.

Current evidence suggests that adding radiation sensitisers to WBRT is unlikely to produce any additional benefit compared with giving WBRT alone.

About this condition

Definition

Metastases to the central nervous system may occur with tumours of any primary origin. Brain (cerebral) metastases may be either solitary or multiple, with or without disseminated disease elsewhere. Brain metastases may present with focal or generalised symptoms, although up to a third of patients may be asymptomatic. A high index of suspicion is required when managing patients with cancer. Headache is the most common presenting symptom (50% of people). Focal weakness, mental change, and seizures are also common. Although clinical signs can be helpful to localise the lesion(s), the initial diagnostic evaluation is commonly performed with pre- and post-contrast computed tomography (CT) scan. Although CT is commonly done, magnetic resonance imaging (MRI) is considered imaging modality of choice. MRI with gadolinium contrast is performed following the detection of a solitary lesion on CT, or if clinical suspicion remains high. MRI may detect lesions as small as 1.9 mm, and is superior to CT for detection of posterior fossa lesions. More than 10% of solitary lesions will not be metastatic and therefore warrant biopsy. In the case of solitary or multiple metastases in the absence of known malignancy, further investigations should be directed towards the identification of a primary lesion, most commonly from the chest. In this review, we have included only adults with brain metastases (cerebral hemispheres and posterior fossa structures) from any primary source that have been confirmed with biopsy or by CT or MRI, and excluded metastasis to the leptomeninges and peripheral nervous system, where management may be more case specific.

Incidence/ Prevalence

The incidence of brain metastasis is 8-11/100,000 people a year. The proportion of people with primary cancers developing brain metastasis varies widely, between 9.6-50.0% depending on the series selected. The lung is the most common primary site, with 9.7-64.0% of people developing brain metastases, while melanoma (6.9-7.4%), renal (6.5-9.8%), breast (5.0-5.1%), and colorectal (1.2-1.9 %) account for most of the remaining cases. Cancer of unknown primary origin represents 15% of cases of brain metastasis.

Aetiology/ Risk factors

Brain metastases are most common in the advanced stages of disseminated disease, but can occur in isolation. Tumour seeding of the brain parenchyma involves a number of steps, including intravasation (reaching the brain vasculature), breaching of the blood-brain barrier, and proliferation and neoangiogenesis (the formation of new blood vessels/vasculature) within the brain. These steps are dependant on the expression of specific regulatory molecules such as matrix metalloproteinases and growth factors.

Prognosis

People with untreated brain metastases have a median survival of about 4 weeks from diagnosis. The addition of corticosteroids may extend this by another 4 weeks. Whole-brain radiotherapy further extends median survival to 3-6 months. The additional benefit of surgery, radiotherapy, chemotherapy, and biological agents alone or in combination depends on tumour type. Prognostic factors predicting a better outcome are solitary lesions, surgical resection, and the use of combined chemotherapy and radiotherapy.

Aims of intervention

To cure, to maintain or improve quality of life and symptoms, to increase overall survival, progression-free survival, with minimal adverse effects of treatment.

Outcomes

Overall survival, progression-free survival, pain and other symptoms, neurological function, objective response rates (complete response, partial response), improvement in performance status according to validated scales of daily functioning/activity, quality of life, adverse effects.

Methods

Clinical Evidence search and appraisal June 2007. The following databases were used to identify studies for this systematic review: Medline and Embase 1986 to June 2007, and The Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Clinical Trials 2007, Issue 2. Additional searches were carried out using these websites: NHS Centre for Reviews and Dissemination (CRD) — for Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA), Turning Research into Practice (TRIP), and NICE. We also searched for retractions of studies included in the review. Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the author for additional assessment, using pre-determined criteria to identify relevant studies. Study design criteria for evaluation in this review were: published systematic reviews and RCTs in any language. RCTs could be open or blinded, and had to contain 20 or more individuals, of whom 80% or more were followed up. There was no minimum length of follow-up required to include studies. In addition, we use a regular surveillance protocol to capture harms alerts from organisations such as the FDA and the UK Medicines and Healthcare products Regulatory Agency (MHRA), which are added to the reviews as required. We have included only adults with brain metastases that have been confirmed with biopsy or by CT or MRI. We have compared each included intervention versus usual (supportive) care and versus any other included intervention and reported any studies we found. Studies generally included people with one or more brain metastasis and reported results for trial participants as a whole, and we have reported these overall results. However, where studies additionally presented a separate analysis for people with single and multiple brain metastases, we have also reported these results. We have tabulated baseline population data for all RCTs included in this review, including Karnofski score and Recursive partitioning analysis (RPA) status where this has been reported (see table 1 ). To aid readability of the numerical data in our reviews, we round many percentages to the nearest whole number. Readers should be aware of this when relating percentages to summary statistics such as RRs and ORs. We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table ).

Table 1.

Baseline population data for included RCTs.

Reference number	Participants	Comparison	Single/multiple brain metastases	Other baseline characteristics
	48 people, ages not reported, people with focal signs due to metabolic encephalopathy excluded	Corticosteroids v corticosteroids plus WBRT	Not stated (before CT scanning available, published 1971) — people had evidence of parenchymal metastasis (clinical symptoms and signs or abnormal tests such as EEGs, radioisotope brain scans, etc)	Prognostic/severity data not reported. Primary site: lung (30 people); breast (7 people); melanoma (4 people); kidney (3 people); other (4 people)
	42 people, median age 60–63 years, all had non-small cell lung cancer, all people had inoperable CNS disease or had refused surgery	Carboplatin plus WBRT v WBRT alone	Single brain metastasis 11/42 (26%) people; multiple brain metastases 31/42 (74%) people	WHO performance status: score 0 = 6 people; score 1 = 21 people; score 2 = 15 people. Neurological function status: score 1 = 21 people; score 2 = 15 people; score 3 = 6 people. Metastasis at sites other than brain not reported
	82 people, mean age 57.8–58.3 years, all people not suitable for surgery or radiosurgery	Temozolomide plus WBRT v WBRT alone	Median number of brain metastases: 3 with WBRT v 2 with WBRT plus temozolomide	Karnofski performance status (KPS): score 70 or greater = 65/82 (79%) people; score less than 70 = 17/82 (21%) people. Extracranial metastasis: yes = 56/82 (68%) people; no = 26/82 (32%) people. Recursive partitioning analysis (RPA) class: I = 5/82 (6%); II = 60/82 (73%); III = 17/82 (21%). Primary site: lung (42 people); breast (13 people); other (27 people). Previous chemotherapy: yes = 62/82 (76%) people; no = 20/82 (24%) people
	52 people (data only presented for 48 people), median age 61–62 years	Temozolomide plus WBRT v WBRT alone	Single brain metastasis 13/48 (27%) people; multiple brain metastases 35/48 (73%) people	Life expectancy of at least 3 months. Eastern Cooperative Oncology Group (ECOG) performance status: 0 = 14/48 (29%) people; 1 = 28/48 (58%) people; 2 = 6/48 (13%) people. Neurological function evaluation: level I = 12 people; level II = 25 people; level III = 11 people. Other organ metastasis = 12/48 (25%) people. Primary site: lung (40 people); breast (5 people); unknown (3 people)
	95 people, median age 58–60 years, people with single metastasis to the brain treated with complete surgical resection (verified by postoperative MRI)	Surgery plus WBRT versus surgery alone	All people had single brain metastasis	Karnofski score: median, 90 (range 70–100) in both groups. Extent of disease other than brain metastasis: none = 34 people; primary tumour only = 37 people; disseminated = 24 people. Primary site: lung (57 people); breast (9 people); other (29 people)
	84 people, mean age 59 years, all people with a single brain metastasis at an operable site	WBRT alone v WBRT plus surgery	All people had single brain metastasis	Karnofski performance status: 50% = 6 people; 60% = 12 people; 70% = 15 people; 80% = 15 people; 90% = 26 people; 100% = 6 people. Extent of primary disease: no evidence of primary disease = 18 people; local primary and intracranial metastasis = 28 people; extracranial metastasis = 38 people. Primary site: lung (45 people); breast (10 people); colon or rectum (13 people); skin (4 people); renal (3 people); head and neck (1 person); other (4 people); unknown primary (4 people)
	63 people, mean age 59 years, all with single brain metastasis at an operable site	WBRT alone v WBRT plus surgery	All people had single brain metastasis	WHO status: score 0 = 7 people; score 1 = 39 people; score 2 = 17 people. Neurological functional scale: I = 19 people; II = 32 people; III = 12 people. Status of disease: “stable” = 43 people; “progressive” = 20 people. Primary site: lung (33 people); breast (12 people); kidney (4 people); melanoma (6 people); others (8 people)
	48 people, median age 59–60 years, all people with a single brain metastasis at an operable site	WBRT alone v WBRT plus surgery	All people had single brain metastasis	Karnofski score for both groups: median, 90% (range 70–100%). Extent of disease other than brain metastasis: none = 10 people; primary tumour only = 20 people; disseminated = 18 people. Primary site: lung (37 people); breast (3 people); gastro-intestinal (3 people); genito-urinary (2 people); melanoma (3 people)
Two systematic reviews included (5 RCTs)	58 people, histologically confirmed cancer and brain metastasis, with no prior WBRT	WBRT plus radiation sensitisers v WBRT alone	Multiple brain metastases (further details not reported by reviews)	Further details not reported by reviews
	116 people, adults, neurological symptoms, no prior cranial irradiation or prior treatment with systemic chemotherapy agents that cross the blood–brain barrier	WBRT plus radiation sensitisers v WBRT alone	Multiple brain metastases (further details not reported by reviews)	Expected survival more than 4 weeks. Further details not reported by reviews
	859 people, aged 18–75 years	WBRT plus radiation sensitisers v WBRT alone	Multiple brain metastases (further details not reported by reviews)	Karnofski performance status of 40 or above; neurological function class (NFC) 1, 2, or 3
	401 people, required WBRT, no prior cranial irradiation, small cell lung cancer and lymphoma and germ-cell tumours excluded, no chemotherapy planned during WBRT	WBRT plus radiation sensitisers v WBRT alone	Multiple brain metastases (further details not reported by reviews)	Karnofski performance status of at least 70
	72 people, age above 18 years, no previous brain radiotherapy, no concurrent chemotherapy	WBRT plus radiation sensitisers v WBRT alone	Multiple brain metastases (further details not reported by reviews)	Karnofski performance status of at least 70; neurological function class (NFC) 1, or 2.
	515 people, age 18–65 years = 145 people, age 65 years or above = 55 people, no prior treatment for brain metastasis other than resection and with a measurable lesion remaining	WBRT plus radiation sensitisers v WBRT alone	Single brain metastasis, 20% in control group and 17% in efaproxiral group; 2–3 brain metastasis, 32% in control group and 30% in efaproxiral group; 3 or more brain metastasis, 47% in control group and 52% in efaproxiral group	Baseline Karnofski performance score: 100 = 16% in control group and 13% in efaproxiral group; 90 = 37% in control group and 46% in efaproxiral group; 80 = 31% in control group and 23% in efaproxiral group; 70 = 16% in control group and 17% in efaproxiral group. Recursive partitioning analysis (RPA): class I = 10% in control group and 11% in efaproxiral group; class II = 90% in control group and 89% in efaproxiral group. Primary cancer: controlled, approximately 25%; uncontrolled, approximately 75%. Primary site: non-small cell lung cancer (55–58% of groups); breast (20–22% of groups); other (22–23% of groups). Extracranial metastasis sites: 0 = 31–36% of groups; 1 to 2 = 4–48% of groups; 3 or above = 18–22% of groups
First RCT reported in three systematic reviews	27 people, lesion 25 mm or less in size	WBRT plus radiosurgery v WBRT alone	2–4 metastatic brain tumours (further details not reported by reviews)	Karnofski performance status 70 or above
Second RCT reported in three systematic reviews	331 people, mean age 58.8–59.9 years, with 1–3 newly diagnosed brain metastasis	WBRT plus radiosurgery v WBRT alone	Single brain metastasis 186/331 (56%) people; 2 brain metastases 85/331 (26%) people; 3 brain metastases 60/331 (18%) people	Karnofski performance status 90–100 = 198/313 (60%) people; 70–80 = 133/331 (40% people). Recursive partitioning analysis (RPA): class 1 = 91/331 (27%) people; class II = 240/331 (73%) people. Metastasis: brain alone = 31%, brain and one or more extracranial site = 69%. Neurological function: no symptoms = 121 people; minor symptoms = 153 people; moderate symptoms = 55 people; information missing = 2 people. Primary site: lung (211 people); breast (34 people); skin/melanoma (16 people); other (40 people); kidney (4 people); bladder (3 people); colon (6 people); ovarian (3 people); unknown primary (7 people)
	132 people, mean age 62 years, with 1–4 brain metastases, maximum diameter 3 cm by MRI scan	Radiosurgery alone v WBRT plus radiosurgery	Single brain metastasis 64/132 (48%) people; 2–4 brain metastases 68/132 (52%) people	Karnofski performance status 90–100 = 78/132 (59%) people; 70–80 = 54/132 (41%) people. Recursive partitioning analysis (RPA): class 1 (aged less than 65 years, no active extracranial disease) = 19/132 (14%) people; class II (aged 65 or over, active extracranial disease) = 113/132 (86%) people. Extracranial metastasis: stable = 79 people; active = 53 people. Primary tumour status: stable = 63 people; active = 69 people. Neurological function: no symptoms, 85 people; minor symptoms, 25 people; moderate symptoms (may or may not require assistance), 22 people; severe symptoms, 0 people. Primary site: lung (88 people); breast (9 people); colorectal (11 people); kidney (10 people); other (14 people)

Table.

GRADE evaluation of interventions for brain metastases in adults

Important outcomes	Mortality, treatment success, adverse effects
Number of studies (participants)	Outcome	Comparison	Type of evidence	Quality	Consistency	Directness	Effect size	GRADE	Comment
What are the effects of interventions for managing brain metastases in adults?
1 (48)	Mortality	Corticosteroids v corticosteroids plus WBRT	4	–1	0	–2	0	Very low	Quality points deducted for sparse data. Directness points deducted for no statistical comparison between groups and lack of baseline data
1 (48)	Treatment success	Corticosteroids v corticosteroids plus WBRT	4	–2	0	–2	0	Very low	Quality points deducted for sparse data and incomplete reporting of results. Directness points deducted for no statistical analysis between groups and lack of baseline data
3 (176)	Mortality	Cytotoxic chemotherapy (with carboplatin or temozolomide) plus WBRT v WBRT alone	4	–2	0	–1	0	Very low	Quality points deducted for sparse data and incomplete reporting of results. Directness point deducted for early termination of 2 RCTs
2 (134)	Treatment successs	Cytotoxic chemotherapy (with carboplatin or temozolomide) plus WBRT v WBRT alone	4	–2	–1	–1	0	Very low	Quality points deducted for sparse data and incomplete reporting of results. Consistency point deducted for conflicting results (inconsistent results for radiological outcomes). Directness point deducted for early termination of 1 RCT
1 (95)	Mortality	Surgery plus WBRT v surgery alone	4	–1	0	–1	0	Low	Quality point deducted for sparse data. Directness point deducted for uncertainty of generalisability of radiotherapy regimen
1 (95)	Treatment success	Surgery plus WBRT v surgery alone	4	–1	0	–1	0	Low	Quality point deducted for sparse data. Directness point deducted for uncertainty of generalisability of radiotherapy regimen
3 (195)	Mortality	Surgery plus WBRT v WBRT alone	4	–1	–1	–1	0	Very low	Quality point deducted for sparse data. Consistency point deducted for heterogeneity among RCTs. Directness point deducted for uncertainty about generalisability of results (highly selected population)
3 (195)	Treatment success	Surgery plus WBRT v WBRT alone	4	–2	0	–1	0	Very low	Quality point deducted for sparse data and incomplete reporting of results. Directness point deducted for uncertainty about generalisability of results (highly selected population)
6 (1522)	Mortality	WBRT plus radiation sensitisers v WBRT	4	0	0	–1	0	Moderate	Directness point deducted for subgroup analysis
At least 3 (at least 410)	Treatment success	WBRT plus radiation sensitisers v WBRT	4	–1	0	0	0	Moderate	Quality point deducted for incomplete reporting of results
2 (at least 331)	Mortality	WBRT plus radiosurgery v WBRT	4	–1	0	–2	0	Very low	Quality point deducted for incomplete reporting of results. Directness points deducted for premature termination of 1 RCT and subgroup analysis
2 (358)	Treatment success	WBRT plus radiosurgery v WBRT	4	0	–1	–1	0	Low	Consistency point deducted for heterogeneity among RCTs. Directness point deducted for high withdrawal rate in performance and mental-status analysis
1 (132)	Mortality	Radiosurgery alone v WBRT plus radiosurgery	4	–1	0	0	0	Moderate	Quality point deducted for sparse data
1 (132)	Treatment success	Radiosurgery alone v WBRT plus radiosurgery	4	–1	0	0	0	Moderate	Quality point deducted for sparse data

Type of evidence: 4 = RCT; Consistency: similarity of results across studies; Directness: generaliseability of population or outcomes; Effect size: based on relative risk or odds ratio; WBRT, whole-brain radiotherapy

Glossary

Karnofsky score

Is a measure of performance status based on physical ability (scale 0–100). 100: normal, no complaints or evidence of disease; 90: able to perform normal activity, minor signs and symptoms of disease; 80: able to perform normal activity with effort, some signs and symptoms of disease; 70: cares for self, unable to perform normal activity or to do active work; 60: requires occasional assistance but is able to care for most of own needs; 50: requires considerable assistance and frequent medical care; 40: requires special care and assistance, disabled; 30: hospitalisation indicated, although death not imminent, severely disabled; 20: hospitalisation necessary, active supportive treatment required, very sick; 10: fatal processes progressing rapidly, moribund; 0: death.

Low-quality evidence

Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.

Moderate-quality evidence

Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

Recursive partitioning analysis

The Radiation Therapy Oncology Group (RTOG) recursive partitioning analysis (RPA) groups patients with brain metastases into three classes based on clinical criteria with differing outcomes. Patients with a controlled primary tumour, no other metastatic sites, age under 65 years, and a Karnofsky performance status above 70 (RPA class I) have the best outcomes. RPA class II includes those with KPS at least 70, uncontrolled primary disease, age greater or equal to 65 years, or other metastatic sites than brain. RPA class III represents patients with KPS less than 70 and is associated with the poorest outcomes.

Very low-quality evidence

Any estimate of effect is very uncertain.

Disclaimer

The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.