Neurosurgical Interventions for Cerebral Metastases of Solid Tumors

Abstract

Background

Metastases are the most common malignant tumors affecting the central nervous system and occur in 20–40 percent of patients with solid systemic tumors. The aim of this review is to discuss the role of neurosurgical procedures in a modern, multidisciplinary treatment approach.

Methods

An expert panel of neurosurgeons, neurologists, and radio-oncologists conducted a selective literature review on neurosurgical interventions for the diagnosis and treatment of cerebral metastases. Original articles, meta-analyses, and systematic reviews were included.

Results

There is a lack of prospective randomized studies. Based on retrospective case series, international guidelines recommend the harvesting (if required, stereotactically guided) of tissue for histological and molecular diagnosis in cases of unknown or possibly competing underlying systemic malignant diseases, in cases of suspected tumor recurrence, and with regard to the evaluation of targeted therapies taking into account molecular heterogeneity of primary and secondary tumors. Surgical resection is particularly valuable for the treatment of up to three space-occupying cerebral metastases, especially to achieve clinical stabilization to allow further non-surgical treatment. For cystic metastasis, a combination of stereotactic puncture and radiotherapy may be useful. Meningeal carcinomatosis can be treated with intrathecal medication via an intraventricular catheter system. Ventriculo-peritoneal shunts represents an effective treatment option for patients with tumor-associated hydrocephalus.

Conclusion

Neurosurgical procedures are of central importance in the multimodal treatment of cerebral metastases. The indications for neurosurgical interventions will be refined in the light of more effective radiation techniques and systemic treatments with new targeted therapeutic approaches and immunotherapies on the horizon.

cme plus

This article has been certified by the North Rhine Academy for Continuing Medical Education. Participation in the CME certification program is possible only over the internet: cme.aerzteblatt.de. The deadline for submission is 9 March 2024.

Metastases of solid tumors are the most common tumors in the brain, with an incidence ten times that of primary brain tumors (1, 2, e1– 3). The most common entities are lung cancer (40–60%), breast cancer (10–41.5%), and melanoma (10–30%) (e3). Brain metastases are already found at the time of initial diagnosis in 23–28% of cases (1, 2, e1). The rising incidence of brain metastases can be explained by the aging of the population, better treatment of primary tumors (particularly with new pharmacotherapeutic approaches), and better diagnostic imaging. The median survival of patients with brain metastases has improved over the past 30 years from five to 8–16 months, depending on the entity (3).

The main therapeutic modalities are open microneurosurgery, radiosurgery, fractionated stereotactic radiotherapy, and antitumor pharmacotherapy. These are increasingly used in combination (4, 5). Radiation techniques have been progressively refined to mitigate their side effects and complications, and new types of drug treatment are available, including immune modulation. The continuing evolution of the available therapies necessitates a critical risk-benefit analysis of the role of neurosurgery within a multimodal approach to treatment (6, e4). Surgery plays a well-established role in the alleviation of acute manifestations due to solitary space-occupying brain metastases in patients whose extracerebral tumor status is stable (either without current treatment, or under ongoing treatment) (4, e5). Yet surgery can also be useful under some circumstances in patients with advanced cancer, and may even be a prerequisite to further treatment as part of a multimodal treatment plan (3– 5). The histopathological and molecular genetic characterization of metastases, and especially of recurrent metastases, is becoming increasingly important in the context of new options for targeted therapy (7, 8, e6). Interdisciplinary coordination of treatment in tumor boards and a profound understanding of the available treatment options are needed in order to determine the optimal treatment for each individual patient. In this review, we discuss the role of neurosurgical procedures in the context of the multimodal management of brain metastases.

Methods

A selective literature search in the PubMed and Cochrane databases was carried out by three independent experts with the search terms “brain,” “CNS,” “central nervous system,” “metastasis,” “secondary,” “tumor,” “cancer,” “lung cancer,” “melanoma,” “surgery,” “resection,” “operation,” “radiotherapy,” “radiosurgery,” “outcome,“ and “survival.” The final reference list was compiled by consensus between the authors. Only articles in German or English were included in the literature search.

Results

The histological and molecular characterization of brain metastases

In a number clinical situations, neurosurgical intervention is primarily useful for diagnostic purposes:

• if the clinical and imaging findings do not yield a clear diagnosis;

• if the primary tumor is unknown;

• if the patient has more than one kind of cancer;

• and if the patient has a type of cancer that only rarely metastasizes to the brain (4, 5).

Obtaining tissue for histopathological diagnosis may also be indicated when the imaging studies do not alone suffice to distinguish treatment sequelae, such as radionecrosis or pseudoprogression after immunotherapy, from tumor progression (9, 10, e7). For deep-seated tumors in areas of the brain where resective surgery would entail high risk, tissue can be obtained in minimally invasive fashion with stereotactically guided serial biopsy, under either general or local anesthesia. The probability of establishing a diagnosis from 1 mm³ tissue samples is over 95%, and the risk of complications is acceptably low even for lesions located in the brainstem (2.6% transient morbidity). (11, e8). Seeding of tumor cells along the biopsy trajectory has only been reported only in very rare cases (e9, e10).

Comprehensive molecular genetic tissue analysis now increasingly provides the basis for treatment with new types of targeted drug that cross the blood-brain barrier to enter the central nervous system (CNS) (table 1) (7, 8, 12). Approximately 53% of brain metastases manifest clinically relevant mutations that are not found in the primary tumor (13, e6). The mutational status of a tumor can change because of treatment-induced selection pressure or tumor progression. In non-small cell lung cancer with an anaplastic lymphoma kinase (ALK) translocation or an EGFR mutation, there are discordance rates of up to 13% and 33%, respectively, between the primary tumor and the brain metastasis, in the untreated setting (e11). A similar discordance rate of 14% for HER-2/new-status in brain metastases of breast cancer has been reported in two independent studies (e11, e12). The molecular signature of some other tumor entities appears to be more stable: for example, a BRAF mutation in an extracranial melanoma is almost certainly predictive of a BRAF mutation in the associated brain metastasis (e11, e13). The detection of predictive biomarkers in metastatic tumor tissue is also becoming increasingly important for immune therapies, just as it already has an established role to play in the immune therapy of primary tumors (e.g., PD-L1 expression).

Table 1. Molecular markers for prediction and for drug therapy of brain metastases*.

Primary tumor	Molecular markers
NSCLC	ALK, BRAF, EGFR, KRAS, MET, NTRK, PD-1/PD-L1, RET, ROS1
Breast cancer	BRCA1/2, CDK4/6, ER/PR, HER2, PD-L1, PIK3CA
Melanoma	BRAF, KIT, NF1, NRAS, PD-L1
Colon cancer	BRAF, KRAS, NRAS, MSI, PD-L1
Upper gastrointestinal tract cancer	HER2, MET

* Exemplary listing of the clinically most relevant molecular markers that serve as therapeutic guideposts for targeted therapy, or as prognostic markers.

These markers can be expressed both in the primary tumor and in the corresponding brain metastasis.

Expression in the primary tumor is not perfectly correlated with expression in the brain metastasis.

NSCLC, non-small-cell lung cancer

The surgical resection of brain metastases

Patient selection

Aside from the need to obtain tissue for diagnostic purposes, the following indications for the surgical resection of brain metastases are internationally accepted:

• a solitary brain metastasis, even in an eloquent area of the brain, especially if it is too large to be treated with radiosurgery;

• a metastasis causing acute neurological manifestations by mass effect, especially in the posterior cranial fossa;

• a dominant brain metastasis that can be resected in a patient with other brain metastases for which there are other therapeutic options;

• and in cases of suspected local tumor recurrence in a surgically accessible location after previous surgery, radiosurgery, or radiotherapy (4, 5).

No reliable figures are available on the percentage of patients with brain metastases who undergo surgery, but retrospective studies indicate that the rate of resection of metastases has declined over time (from 30.8% in the 1990s to 19.5% in the past decade) in favor of single-session irradiation, particularly with radiosurgery, with small brain metastases being diagnosed at an earlier stage than before (3).

In general, the surgical resection of brain metastases is an option only for patients who are in adequate physical condition (Karnofsky Performance Score ≥ 70%) or whose condition is expected to improve after resection (e.g., after the relief of intracranial hypertension) so that they can undergo further oncological treatment (e5). Clinical nomograms are available as aids to prognostication on the basis of the type of primary tumor and the intra- and extracranial tumor burden (14). Old age and the presence of neurological manifestations are unfavorable prognostic factors whatever the tumor entity (15, 16). In cases of acute neurologic deterioration because of mass effect, resection can improve and stabilize the patient’s clinical condition, enabling further therapy (17). Frailty must be considered in the assessment of operability: In a retrospective study of 3500 patients with brain metastases, a Modified Frailty Index of ≥ 2 points was associated with higher mortality after hospital discharge (odds ratio 2.79; 95% confidence interval: [1.6; 4.9]) (18). The histological type of the primary tumor is also relevant to the indication for surgery: for example, chemo- and radiosensitive tumors that are prone to diffuse seeding (e.g., small-cell lung cancer) should not be primarily resected if the patient is clinically stable. This illustrates the importance of a critical assessment of the indications for surgery and of proper patient selection. The available evidence regarding the value of surgery for brain metastases is of low quality (level IV). There are no prospective trials evaluating the effect of the extent of resection on clinical outcome, neither for primary (intrinsic) brain tumors nor for brain metastases, as such trials cannot be performed for ethical reasons.

The timing of surgery

Acute neurological deterioration (e.g., due to intracranial hypertension) can be an indication for urgent or emergency surgery. This is especially true of metastases in the posterior fossa. For surgical procedures that can be planned electively, the timing should accord with the overall treatment plan. Proper timing may require weighing the risks of operating earlier or later: poor wound healing, perioperative bleeding, and infections are more likely if surgery is preceded by longer-term steroid treatment, or if it is closely preceded by radio- or chemotherapy (nadir of the blood counts), or even by anti-angiogenic drug treatment (e.g., the VEGF inhibitor bevacizumab); on the other hand, if surgery is delayed for these or other reasons, the metastasis can grow larger during the interval (e14, e15). The optimal timing of radiotherapy in relation to the surgical resection of metastases is being studied in prospective clinical trials (e.g., NCT04474925). A trial that is now recruiting patients in Germany and elsewhere is also investigating neo-adjuvant radiotherapy (NCT03368625).

Technical aspects of surgery for brain metastases

As metastases in the brain grow, they usually remain well circumscribed, displacing rather than invading the surrounding brain tissue. However, autopsy studies have shown that brain metastases can infiltrate a few millimeters into the adjacent brain tissue, regardless of the histology of the primary tumor (19). If the metastasis is located at the surface of the brain, tumor cells can spread along the sulci (local meningeosis). Brain metastases are of highly variable consistency: some are hard, some are soft or even deliquescent, while others are primarily cystic. Brain metastases can be resected “supramaximally,” i.e., together with adjacent brain tissue that may contain infiltrating tumor cells, when this is judged to be acceptable in terms of brain function (20, e16, e17). The surgeon should avoid the accidental transfer of tumor cells via the CSF to decrease the risk of further distant metastases or leptomeningeal seeding. Brain metastases should thus be resected en bloc whenever possible, if they are of suitable consistency and at an accessible location (21– 24). A magnetic resonance (MR) scan of the brain should be performed early, i.e., within 72 hours, to assess the extent of resection (4, 25). It is not known at present whether supramarginal resection beyond the visible boundary of the tumor might lessen the need for postoperative radiotherapy (see below) (10).

The preservation or restoration of neurologic function has priority; clinical worsening may affect the timing and extent of further treatment with chemotherapy and/or radiotherapy 15, e18). Brain metastases can now routinely be treated safely and effectively in major neurosurgical centers, with a refined approach to surgical indications together with modern technical aids to microsurgery, including the routine use of the high-power surgical microscope, image-guided neuronavigation, and pre- and intraoperative electrophysiological functional testing (26). Depending on the site of the metastasis, a transient morbidity of 10–20% has been described, particularly when the tumor is a recurrence or is located in an eloquent brain region; most complications are mild, mainly consisting of wound-healing disturbances, wound infections, or concomitant systemic illnesses, such as infections (17, 26). In general, an appropriate assessment of the benefits and risks of resection is a key element of the overall oncological approach.

The role of resection for solitary brain metastases

Surgery plays a well-established role in the treatment of solitary brain metastases as well as oligometastatic disease (i.e., two or three brain metastases), as long as the patient’s systemic disease is stable, either with or without chemotherapy (Table 2). For smaller metastases that are no larger than 3 cm in diameter, radiosurgery results in local tumor control at a rate comparable to resection (89–93% at 12 months), as was found in a prospective, randomized trial (27). The larger the tumor, however, the higher the risk that radiosurgery will cause symptomatic brain edema or medically intractable radionecrosis (6–17%) (e19, e20). For patients with a larger volume of metastatic tumor, hypofractionated stereotactic radiotherapy or radiosurgery is better tolerated and also yields considerable local tumor control (ca. 86% at 12 months) (e21).

Table 2a. Clinical trials concerning the neurosurgical resection of solitary brain metastases* (prospective trials only).

Trial design,reference number	Intervention (number of patients)	Primary tumor	Inclusion criteria	Outcome with respect to tumor control and survival (resection vs. other treatment)
RCT (e30)	resection→WBRT (n = 10) vs. SRS→WBRT (n = 11)	lung Ca: n = 10 (48%) colon Ca: n = 4 (19%) other: n = 7 (33%)	diameter ≤ 4 cm; ≥ 18 years old; KPS ≥ 70% (except tumor-related)	PFS (incl. local PFS): 3.1 vs. 1.7 months (HR: 0.55; p = 0.2) OS: 2.8 vs. 6.2 months (HR 0.53; p = 0.2)
RCT (27)	resection→WBRT (n = 33) vs. GKRS (n = 31)	lung Ca: n = 22 (34%) breast Ca: n = 11 (17%) urogenital Ca: n = 10 (16%) melanoma: n = 9 (14%) CUP: n = 6 (9%) breast Ca: n = 4 (6%) liver Ca: n = 2 (3%)	diameter ≤ 3 cm; age 18–80 years; KPS ≥ 70%; stable systemic disease with life expectancy ≥ 4 months	1-year local recurrence: 82% vs. 96.8% (HR: n.a.; p = 0.06) survival without distant recurrence: 3% vs. 25.8% (HR: n.a.; p < 0.05) os: 9.5 vs. 10.3 months (hr: n.a.; p = 0.8)
RCT (e31)	resection→WBRT (n = 41) vs. biopsy→WBRT (n = 43)	lung Ca: n = 45 (54%) gastrointest. Ca: n = 13 (16%) breast Ca: n = 10 (12%) melanoma: n = 4 (5%) CUP: n = 4 (5%) Renal Ca: n = 3 (4%) head and neck Ca: n = 1 (1%) other: n = 4 (5%)	age ≤ 80 years; KPS > 50 %	OS: 5.6 vs. 6.3 months (HR: n.a.; p = 0.24)
RCT (e32)	resection→WBRT (n = 32) vs. biopsy→WBRT (n = 31)	lung Ca: n = 33 (52%) breast Ca: n = 12 (19%) melanoma: n = 6 (10%) renal Ca: n = 4 (6%) other: n = 8 (13%)	age ≥ 18 years; ECOG ≤ 2; life expectancy > 6 months	survival with ECOG ≤ 1: 7.5 vs. 3.5 (HR: n.a.; p = 0.06) OS: 10 vs. 6 months (HR: n.a.; p = 0.04)
RCT (e33)	resection→WBRT (n = 25) vs. biopsy→WBRT (n = 23)	lung Ca: 37 (77%) breast Ca: 3 (6%) gastrointestinal Ca: 3 (6%) urogenital Ca: 2 (4%) melanoma: 3 (6%)	age ≥ 18 years; KPS ≥ 70 %	local recurrence up to data cutoff: 20 vs. 52% (RR: n.a.; p < 0.02) local pfs: > 59 vs. 21 weeks (RR: 7.1; p < 0.001) os: 40 vs. 15 weeks (rr 2.2; p < 0.01)

* Available prospective studies on resective surgery for solitary brain metastases compared to alternative treatments.

Ca, cancer; CUP, cancer of unknown primary; ECOG, Eastern Cooperative Oncology Group; GKRS, Gamma Knife radiosurgery; HR, hazard ratio; KPS, Karnofsky Performance Score; n.a., not available for review; OS, overall survival; PFS, progression-free survival; RCT, randomized controlled trial; RR, relative risk; SRS, stereotactic radiosurgery; vs., versus; WBRT, whole-brain radiotherapy

The relative utility and risk profiles of radiation and resection for larger metastases (> 3 cm) have not been studied in any prospective, randomized trial to date. The international guidelines therefore favor surgery for (symptomatic) lesions of diameter greater than 3 cm with mass effect, particularly if located in the posterior fossa (figure) (4, 5). A randomized phase III trial has shown that postoperative stereotactic irradiation of the tumor bed improves local tumor control (hazard ratio 0.46, [0.2; 0.9]) (28). Cystic metastases in functionally relevant areas, can be treated with minimally invasive stereotactic cyst puncture followed by tumor irradiation (e22, e23).

Figure.

Microsurgical resection of a symptomatic solitary brain metastasis in a 64-year-old woman with breast cancer who presented with progressive dizziness, visual disturbances, and diplopia.

a) There is a large right temporo-occipital mass involving the optic radiation and compressing the midbrain. b) Intraoperative neuronavigation (left) and photographic documentation (right) of tumor tissue (arrows) and of the trochlear nerve (asterisk) after midbrain decompression. c) Visual field examinations: a homonymous hemianopsia that was present before surgery (above) has fully resolved by the time of the follow-up examination at six months (below). d) Early postoperative MRI (within 72 hours of surgery) shows no residual tumor tissue.

The role of resection for multiple brain metastases

30–50% of patients with brain metastases have more than one metastasis (2, e1). There have not been any randomized, prospective trials of surgical resection in this situation (Table 2) (29). Resecting a clinically dominant metastasis can stabilize the patient’s condition and lessen the need for steroids, which may be relevant in the context of multimodal treatment, e.g., in combination with single-session radiosurgery, whole-brain radiation (now only rarely considered to be indicated), or newer types of pharmacotherapy, including immune checkpoint inhibitors (4, 5, 29, 30, 31) (efigure 2). In the absence of data from prospective trials, decisions regarding combined treatment for patients with multiple brain metastases are taken on a case-by-case basis and should be critically evaluated by an interdisciplinary tumor board (32). With a variety of targeted therapies for brain metastases now being developed, there is a need for prospective, tumor- and target-specific trials so that the diagnostic and therapeutic value of neurosurgical intervention in these patients can be more precisely defined.

eFigure 2.

Combined surgery and radiotherapy in a 72-year-old man with melanoma who presented with two brain metastases: a) a right parietal metastasis with marked perifocal edema and left hemiparesis (motor strength 4 out of 5) and a right temporal metastasis measuring 4 mm in diameter (inset). b) Neuronavigation (left) and intraoperative photo documentation both before and after (top/bottom right) complete resection of the parietal metastasis (arrows). c) MRI follow-up six weeks after stereotactic irradiation of the resection cavity according to the VMAT scheme (total dose 25 Gy, in 5 fractions of 5 Gy each) and single-session radiosurgery of the temporal metastasis (1 x 20 Gy at the 80% isodose), followed by immunotherapy as directed by a dermato-oncologist. VMAT, volumetric intensity-modulated arc therapy.

The role of resection for recurrent metastases and radionecrosis

Local recurrence of a metastasis does not have the same implications for treatment as a new metastasis arising at a different site. The decision whether to operate on a local recurrence must be made individually in the light of all prior treatments and the patient’s systemic disease status. No prospective, randomized trials have been performed on the resection of local recurrences of metastases. Retrospective studies have shown that favorable prognostic factors include good clinical status, a longer time since previous treatment at the same site, and breast cancer as the primary tumor (33, 34, e24).

The detection of a recurrent a brain metastasis poses a diagnostic challenge: oncologic follow-up is based on MRI imaging, which cannot definitively distinguish a local tumor recurrence from post-therapeutic changes (so-called pseudoprogression) or from symptomatic radiation necrosis. In fact, the risk of radionecrosis after single-session stereotactic irradiation (depending on the radiation dose and the treatment volume) is 12–20%. Moreover, radionecrosis can occur at any time from 10 to 36 months after treatment (9). Positron emission tomography (PET) with amino acid tracers, when it is available, enables greater differential-diagnostic specificity (35, e25), but generally does not obviate the need for histopathological examination, preferably by biopsy (29). Histopathology is also needed if changes in the molecular profile of the tumor, compared to the primary tumor from which it is derived, might affect the therapeutic strategy (4). In the future, new approaches such as liquid biopsy from cerebrospinal fluid or peripheral blood may yield important differential diagnostic information (e26).

Surgical treatments for leptomeningeal tumor seeding

About 5% of all patients with solid tumors develop leptomeningeal tumor seeding (4). In this situation, the median overall survival across entities is only approximately 4 months (36). Therapeutic options include systemic and/or intrathecal drug therapy as well as cranial irradiation down to the C2 level. The implantation of a ventricular catheter connected to a subcutaneous Ommaya or Rickham reservoir seems to enable a more homogeneous distribution of drugs in the CSF space compared to repeated lumbar punctures and is associated with a median survival time of 5.2 months in retrospective studies; the risks accompanying repeated lumbar punctures are thereby avoided as well (37). When needed, CSF can be obtained repeatedly by puncture of the reservoir in order to monitor the response to treatment over time.

Ventriculoperitoneal shunting: symptom control in patients with hydrocephalus

Two-thirds of patients with leptomeningeal tumor seeding develop a CSF circulation disturbance leading to hydrocephalus (e27, e28). The implantation of a ventriculoperitoneal shunt yields adequate symptom control in more than 90% of cases, with an associated improvement in the quality of life, enabling further treatment if indicated (38, e29). The periprocedural morbidity of shunt implantation is slightly higher (4–10%) in this situation than in patients without malignant tumors (39). In two retrospective series of 37 and 59 patients, shunt malfunctions necessitating revision arose in 8% an 13% of cases; peritoneal seeding, though occasionally feared, was not demonstrable in any case (38, 39). In summary, neurosurgical intervention can be helpful as a palliative measure in patient with leptomeningeal tumor seeding.

The role of neurosurgery in clinical window-of-opportunity studies

Valuable pharmacokinetic information can be obtained during neurosurgical procedures for the resection of brain metastases, e.g., information on the achievable levels of systemically administered drugs within tumor tissue (so-called window-of-opportunity studies) (40). Aside from the measurement of drug concentrations, the surgical specimen can be subjected to further study in the laboratory, e.g., concerning drug binding to molecular target structures and the biological effects of drugs on tumor tissue and normal tissue (pharmacodynamics).

Conclusion

Brain metastases are an increasingly common problem and neurosurgery remains a key component of their multidisciplinary treatment, all the more so after the recent introduction of immuno-oncological and targeted therapies. The spectrum of neurosurgical procedures also includes minimally invasive ones, such as the placement of catheter systems for the palliative treatment of patients with leptomeningeal tumor cell seeding. The scientific evidence for these statements is of poor quality (level IV) in the absence of controlled, prospective trials. In the future, multicenter databases with long-term follow-up may yield more detailed information on the utility of neurosurgical intervention. Moreover, window-of-opportunity studies using neurosurgical tissue specimens may yield important information on pharmacokinetics and pharmacodynamics that can serve as a basis for the implementation of new therapeutic options.

Table 2b. Studies on the neurosurgical resection of multiple brain metastases*.

Study design, reference no.	Size of patient cohort	Primary tumor	Inclusion criteria	Outcome with respect to quality of life, tumor control, and survival
Retrospective (e34)	n = 704 (n = 372 with solitary metastasis, n = 122 with two metastases, n = 142 with three metastases, n = 68 with > 3 metastases)	breast Ca, lung CA, other (precise number not stated)	histopathologic diagnosis of brain metastasis; surgical resection of ≥ 1 brain metastases; pre- and postoperative MRI	OS: 6 months (postoperative tumor volume with HR 1.02, p = 0.004)
Retrospective (e35)	n = 216 (n = 129 with solitary metastasis, n = 64 with 2–3 metastases, n = 23 with ≥ 4 metastases)	non-small-cell lung Ca: n = 216 (100%)	> 18 years; surgical resection of brain metastases of non-small cell lung cancer	KPS: significant improvement postoperatively (raw data n.a.; p < 0.001) local pfs: 38% of patients with radiological local progression after median radiological follow-up of 8 months (the metastasis count had no significant effect on hr) os (entire cohort): 12.7 months (the metastasis count had no significant effect on hr)
Retrospective (17)	n = 750 (n = 462 with solitary metastasis, n = 185 with 2–3 metastases, n = 103 with ≥ 4 metastases)	lung Ca: n = 318 (42%) melanoma: n = 114 (15%) breast Ca: n = 116 (16%) gastrointest. Ca: n = 72 (10%) renal Ca: n = 24 (3%) CUP: n = 33 (4%) other: n = 73 (10%)	> 18 years; surgical resection of brain metastases	KPS: 80% preoperative vs. 90% postoperative (p < 0.001) rtog groups (pre- vs. postoperative): 1: 18.5 vs. 19.3%; 2: 62.9 vs. 70.1%; 3: 18.5% vs. 10.5% (p < 0.001) os (entire cohort): 10.9 months (hr 0.63 for 1–3 metastases vs. ≥ 4 metastases, p < 0.001)
Retrospective (e24)	n = 127 (n = 81 with 2–3 metastases, n = 24 with 4–5 metastases, n = 22 with ≥6 metastases)	lung Ca: n = 44 (35%) melanoma: n = 22 (17%) gastrointest. Ca: n = 16 (13%) breast Ca: n = 15 (12%) renal Ca: n = 15 (12%) CUP: n = 8 (6%) other: n = 7 (6%)	tumor volume ≥ 27 cm3 ± posterior fossa mass effect ± progression despite prior radiotherapy ± CUP syndrome	KPS: improved in 23% of patients postoperatively, worsened in 16% (p-value not stated) RTOG groups: improved in 12% of patients postoperatively (p-value not stated) OS (entire cohort): 6.5 months (HR 1.81 for ≥ 5 metastases vs. ≤ 4 metastases, p = 0.019)
Retrospective (e36)	n = 208 (n = 132 with solitary metastasis, n = 76 with ≥ 2 metastases)	lung Ca: n = 105 (51%) breast Ca: n = 28 (14%) melanoma: n = 18 (9%) colon Ca: n = 12 (6%) renal Ca: n = 13 (6%) other: n = 25 (12%) CUP: n = 15 (7%) (note: in the paper, the numbers of patients do not add up to the stated total)	surgical resection of brain metastases	KPS: improved postoperatively in 30% of patients with solitary metastasis vs. in 37% of patients with ≥ 2 metastases (HR: not significant) OS (entire cohort): 8 months (8 months in patients with solitary metastasis vs. 11 months in patients with ≥ 2 metastases; p = 0.77)
Retrospective (e37)	n = 56 (all with ≥ 2 metastases)	melanoma: n = 25 (45%) breast Ca: n = 11 (20%) lung Ca: n = 7 (13%) sarcoma: n = 5 (9%) colon Ca: n = 3 (5%) renal Ca: n = 2 (4%) ovarian Ca: n = 1 (2%) CUP: n = 2 (4 %)	presence of multiple brain metastases; surgical resection	local PFS: local recurrence in 4% (1/26) of patients with resection of all brain metastases OS: 10 months (6 months with some brain metastases still present postoperatively vs. 14 months with resection of all brain metastases, p = 0.003)

* No prospective studies are available for multiple brain metastases, so retrospective studies that included more than 50 patients with multiple brain metastases who underwent surgical resection are listed here.

Ca, cancer; CUP, cancer of unknown primary; HR, hazard ratio; KPS, Karnofsky Performance Score; MRI, magnetic resonance imaging, n.a., not available for review; OS, overall survival; PFS, progression-free survival; RCT, randomized controlled trial; RR, relative risk; RTOG, Radiation Therapy Oncology Group; vs., versus

eFigure 1.

The microsurgical resection of multiple symptomatic brain metastases

A 48-year-old man with colon cancer presented with progressive visual disturbances.

a) The preoperative MRI shows multiple brain metastases (arrows) in the occipital lobes and elsewhere. b) The postoperative follow-up MRI, obtained within 72 hours of surgery, confirms total resection of the metastases; the patient’s vision was markedly improved.

Questions on the article in issue 10/2023:

Neurosurgical Interventions for Cerebral Metastases of Solid Tumors

The submission deadline is 9 March 2024. Only one answer is possible per question. Please select the answer that is most appropriate.

Question 1

What is the most common primary tumor of brain metastases?

1. breast cancer

2. lung cancer

3. melanoma

4. ovarian cancer

5. cervical cancer

Question 2

What Karnofsky performance score is generally considered necessary to qualify a patient for the resection of brain metastases?

1. < 30%

2. < 10%

3. < 60%

4. ≥ 50%

5. ≥ 70%

Question 3

What examination should be performed to assess the extent of resection of a brain metastasis?

1. lumbar puncture after at least 12 hours

2. MRI within 72 hours

3. CT at 48 hours

4. ultrasound at 24 hours

5. EEG at 36 hours

Question 4

Approximately what percentage of patients with solid tumors develop leptomeningeal tumor seeding?

1. 1%

2. 5%

3. 10%

4. 15%

5. 25%

Question 5

What problem (also described in the text) arises in two out of three patients with leptomeningeal tumor seeding?

1. epilepsy

2. meningitis

3. hydrocephalus

4. encephalitis

5. strabismus

Question 6

According to international guidelines, at what size (diameter) is surgery preferable to radiosurgery for the treatment of a symptomatic, space-occupying brain metastasis, especially if located in the posterior cranial fossa?

1. > 0.3 cm

2. > 2 cm

3. > 3 cm

4. > 5 cm

5. > 7 cm

Question 7

As described in the text, what is a potentially effective treatment for a cystic metastasis in a functionally relevant area of the brain?

1. (a) chemotherapy followed by irradiation

2. (b) surgical cyst removal followed by irradiation

3. (c) radiotherapy followed by chemotherapy

4. (d) minimally invasive stereotactic puncture followed by radiotherapy

5. (e) surgical cyst removal followed by chemotherapy

Question 8

What is the risk of radionecrosis after single-session stereotactic irradiation?

1. 0.2–0.8%

2. 1–2%

3. 5–8%

4. 12–20%

5. 30–40%

Question 9

Which of the following mutations, if present in an extracranial primary tumor, is almost certainly present in the associated brain metastasis as well?

1. BRAF in melanoma

2. HER-2 in breast cancer

3. EGFR in non-small-cell lung cancer

4. BRCA1 in breast cancer

5. MYD88 in lymphoma

Question 10

Approximately what percentage of brain metastases have clinically relevant mutations that are not found in the primary tumor?

1. 5%

2. 12%

3. 23%

4. 32%

5. 53%