20 Stereotactic Radiosurgery for Tumors of the Brainstem, Thalamus, and Pineal Region

10.1055/b-0039-173911

20 Stereotactic Radiosurgery for Tumors of the Brainstem, Thalamus, and Pineal Region

Amparo Wolf and Douglas Kondziolka

Abstract

Stereotactic radiosurgery (SRS) for tumors of the brainstem, thalamus, and pineal region allows patients to receive focused, conformal, and ionizing radiation to a target that is delineated by high-resolution imaging. This chapter reviews the basic technique of Gamma Knife radiosurgery and the evidence of its effectiveness in the most commonly treated benign and malignant tumors of the brainstem, thalamus, and pineal region, encompassing both intrinsic and extrinsic tumors. Because SRS is highly selective and delivers radiation to narrowly targeted areas, nearby tissue toxicity is reduced, thereby increasing the safety of SRS. This precision is particularly vital when treating tumors located within critical functional tissue. The role of SRS in focal brainstem gliomas, thalamic gliomas, and pineal region tumors is still evolving; SRS may be used as the primary modality treatment or as adjuvant treatment to surgical resection. Most clinicians advocate histological diagnosis of tumors when feasible. However, when a tissue diagnosis is not possible and when growth of a focal tumor within the brainstem, thalamus, or pineal region is documented, SRS may be an appropriate management strategy to maximize tumor control while maintaining neurological function and quality of life.

Introduction

Since the advent of stereotactic radiosurgery (SRS) more than 30 years ago, the management of benign and malignant tumors of the brainstem, thalamus, and pineal region has evolved. SRS delivers focused, conformal, and ionizing radiation to a target that is delineated by high-resolution imaging. SRS is highly selective due to the steep falloff of radiation into the brain structures surrounding the target. This selectivity results in reduced tissue toxicity, thereby increasing the safety of SRS. This precision of SRS is particularly vital for tumors located within critical functional tissue. The biologic effects of radio-surgery are time and dose dependent and include inhibition of tumor cell division, thrombosis of neoplastic blood vessels, induction of apoptosis or necrosis, and alterations of the regional brain blood flow. ¹ ^, ² ^, ³

Both intrinsic and extrinsic tumors of the brainstem can be managed with radiosurgery. This chapter reviews the basic technique of Gamma Knife radiosurgery and the evidence of its effectiveness in the most commonly treated tumors of the brainstem, thalamus, and pineal region. Several other less prevalent tumor types of the brainstem and posterior fossa that may be treated effectively with SRS but are not discussed in this chapter include other schwannomas, glomus tumors, chordomas, hemangioblastomas, hemangiopericytomas, and other tumors.

Gamma Knife Radiosurgery Procedural Basics

Prior to undergoing radiosurgery, patients meet with their clinicians to discuss management options, goals, and potential risks of SRS. On the day of radiosurgery, patients are given a light sedative. The Leksell frame is placed with the patient under local anesthesia. High-resolution magnetic resonance imaging (MRI) or computed tomography (CT) images, including contrast-enhanced images, are acquired with the frame in place. Target volumes are delineated on high-resolution images using software. Radiosurgical dose planning is performed by the neurosurgeon in collaboration with the radiation oncologist and medical physicist. For tumors of irregular shape, multiple isocenters of different sizes (4 mm, 8 mm, or 16 mm) and weighting may be used to conform the target volume to the selected treatment isodose. The margin dose is chosen on the basis of a balance of achieving high local control rates while minimizing risks related to tumor volume and location. Radiation is then delivered. At the end of the procedure, the stereotactic head frame is removed and a local dressing is applied. Most patients are discharged home the same day. Patients with malignant disease commonly have follow-up high-resolution imaging every 2 to 3 months, and those with benign conditions are followed-up every 6 to 12 months during the first few years after radiosurgery. Subsequent follow-up imaging is conducted at less frequent intervals.

With the introduction of the Gamma Knife Icon (Elekta, Stockholm, Sweden) in 2015, it became possible to perform SRS using mask fixation. The unit integrates a cone beam CT, and CT images are coregistered with the preprocedural MRI. The unit has the capacity to detect and measure patient position changes, allowing adaptation in dose planning when necessary. The Icon increases treatment flexibility by allowing the physician to choose either single-session or fractionated delivery.

Stereotactic Radiosurgery for Tumors of the Brainstem

Brainstem Metastases

Brainstem metastases account for 3 to 5% of intracranial metastases. ⁴ The prognosis for patients with brainstem metastases remains poor, with a median overall survival of 4 to 6 months. ⁵ Patients with tumors within the brainstem can present with progressive weakness, diplopia, unsteady gait, dysphagia, dysarthria, and headache, among other symptoms. Local progression of disease within the brainstem may result in an acute neurologic decline. Brainstem metastases are rarely accessible by open surgery. The response of brain metastases to systemic chemotherapies is unclear and unreliable. Compared with SRS, conventional whole-brain radiation therapy may result in lower rates of tumor control and increased global neurocognitive decline. ⁶ ^, ⁷ ^, ⁸ ^, ⁹

Several retrospective studies have evaluated the role of SRS in brainstem metastases and demonstrated its effectiveness. The UCSF group studied 42 patients with 44 tumors (7 midbrain, 31 pons, 6 medulla) and reported local control rates of 90% at 6 months and 77% at 1 year. ¹⁰ Four patients (9.5%) had brainstem adverse radiation effects (AREs). Poor outcomes for patients with brainstem SRS were associated with melanoma histology, renal cell carcinoma histology, and volumes greater than 1 cm³. Another retrospective study looked at 53 patients (8 midbrain, 42 pons, 3 medulla) with 37 patients having followup imaging. ¹¹ Thirty-two of the 37 tumors remained stable or showed partial response, with local progression in 5 tumors at a mean follow up of 9.8 months. The median overall survival was 11 months. The authors concluded that SRS prolongs survival compared with observation alone. Several other retrospective studies have been performed ( Table 20.1 ). ¹⁰ ^, ¹¹ ^, ¹² ^, ¹³ ^, ¹⁴ ^, ¹⁵ ^, ¹⁶ ^, ¹⁷ ^, ¹⁸ ^, ¹⁹ ^, ²⁰ ^, ²¹ The International Gamma Knife Research Foundation recently published the largest series of patients with brainstem metastases, including 547 patients with 596 tumors. ²¹ Actuarial local control at 1 year after SRS was approximately 82%. Higher local control rates were seen with increasing margin dose and maximum dose. Overall survival at 1 year was 33% and depended on age, sex, the number of metastases, tumor histology, and performance score. Severe toxicity (grade ≥ 3 based on the Common Terminology Criteria for Adverse Events) occurred in 7.4% of patients. Predictors of severe toxicity included larger tumor-volume margin dose and a history of whole-brain irradiation. Margin doses of 20 Gy and higher were associated with improved local control but higher toxicity. ²¹ Fig. 20.1 depicts a patient with a large brainstem metastasis who underwent SRS at a margin dose of 14 Gy (maximum dose of 28 Gy) with significant regression on MRI 2 months after SRS.

**Fig. 20.1** An 88-year-old man with gastrointestinal carcinoma metastases to the pons. **(a)** Axial, **(b)** sagittal, and **(c)** coronal T1-weighted gadolinium-enhanced magnetic resonance images (MRIs) with stereotactic radiosurgery (SRS) plan. The marginal dose was 14 Gy at 50% isodose, and the target volume was 2.08 cm³. The 12 Gy isodose volume is delineated in green. Axial **(d)** T1-weighted MRI with gadolinium and **(e)** fluid-attenuated inversion recovery MRI obtained 2 months after SRS show significant regression of the tumor.

**Table 20.1** Summary of key studies of stereotactic radiosurgery in patients with brainstem metastases
Study	N patients (N tumors)	Median dose (Gy), 50% isodose	Median tumor volume (cm³)	Local control	Symptomatic AREs
Huang et al 1999 ¹²	26 (27)	16	1.1	95%, median F/U of 9.5 mo	None
Fuentes et al 2006 ¹³	28 (28)	19.6	2.1	92%, median F/U of 11 mo	12.5% transient worsening 3 d after Gamma Knife
Yen et al 2006 ¹¹	53 (53)	17.6	2.8	86%, mean F/U of 9.5 mo	None
Kased et al 2008 ¹⁰	42 (44)	16	0.26	77% at 1 y	9.5%
Koyfman et al 2010 ¹⁴	43 (43)	15	0.37	85% at 1 y	9% (grade 1 and 2); no grade 3 or 4
Hatiboglu et al 2011 ¹⁵	60	15 (LINAC)	1.0	35% at 1 y	20% early and delayed (hemiparesis, cranial nerve deficit, hemorrhage, nausea/vomiting, headache)
Kawabe et al 2012 ¹⁶	200 (222)	18	0.2	81.8% at 2 y	0.5%
Sengoz et al 2013 ¹⁷	44 (46)	16	0.6	96%	None
Kilburn et al 2014 ¹⁹	44 (52)	18	0.13	74% at 1 y	9.1%
Peterson et al 2014 ¹⁸	41	17^a	0.66 ^a	91%	2.4%
Voong et al 2015 ²⁰	74 (77)	16	0.13	94% at median F/U of 5.5 mo	8%
Trifiletti et al 2016 ²¹	547 (596)	16	0.8	81.8% at 1 y	7.4% (grade ≥ 3)
Abbreviations: ARE, adverse radiation effect; F/U, follow-up; LINAC, linear accelerator. ^aMean value.

The conclusion from most of these retrospective studies is that SRS provides excellent local control of brainstem metastases with low morbidity. The severity of systemic disease remains the predominant factor determining the prognosis of patients with tumor metastases to the brainstem. As systemic therapies improve, including targeted therapies and immunotherapies, their combination with SRS may result in longer overall survival.

Vestibular Schwannomas

The paradigm for the management of vestibular schwannomas (VS) has shifted over the last 20 years. Radiosurgery is an effective alternative to surgical removal and, at certain institutions, the favored approach for small-sized to moderate-sized VS. Patients often prefer radiosurgery, particularly when they are minimally symptomatic, to avoid the risks associated with microsurgery, which include higher rates of facial weakness, hearing loss, wound infection, cerebrospinal fluid leak, meningitis, and hemorrhage. The goals of radiosurgery are to prevent further tumor growth while preserving facial nerve (cranial nerve [CN] VII) and trigeminal nerve (CN V) function. For patients with serviceable hearing, attempts are made to maintain hearing as long as possible.

Multiple matched cohort studies in VS have shown comparable tumor control rates between microsurgery and radiosurgery of 95 to 98% over 10 years. ²² ^, ²³ ^, ²⁴ SRS has the benefit of resulting in facial weakness in less than 1% of cases. ²⁴ ^, ²⁵ Trigeminal dysfunction occurs in 1 to 3% of patients. ²⁶ Symptoms of tinnitus and imbalance are similar between SRS and microsurgery.

With the evolution of high-resolution imaging, patients with VS are now presenting earlier, with tumors that are smaller in size and with hearing that is still preserved. Patients with useful hearing before radiosurgery continue to report an approximate 60 to 85% overall rate for maintenance of serviceable hearing in the first years after the procedure. ²⁷ For patients with intracanalicular tumors, the rate of hearing preservation is greater than 80% and may be superior to rates of hearing preservation when patients are managed conservatively. ²⁸ ^, ²⁹ Predictors of hearing preservation after SRS include age, pre-SRS hearing status, and the mean dose to the cochlea. ²⁵ Maintaining a mean dose to the cochlea of less than 4 Gy appears to result in higher rates of useful hearing preservation. ²⁵ More recent retrospective studies have shown that SRS performed within 2 years of VS diagnosis in normal hearing patients, and prior to subjective hearing loss, results in greater retention of hearing compared with delayed SRS. ³⁰ ^, ³¹

The average dose prescribed to the tumor margin is 12.5 Gy (12–13 Gy) at the 50% isodose line. Lower or higher doses can be used depending on hearing status, tumor volume, and clinical history. SRS has been shown to be effective in properly selected larger-sized VS measuring 3 to 4 cm, with few patients requiring a subsequent procedure. ³² ^, ³³ ^, ³⁴ Although a limited number of reports have suggested that SRS is less effective in cystic VS, ³⁵ this may not be the case, and evidence is lacking ( Fig. 20.2 ). In the early follow-up period after SRS, the VS will commonly show reduced central contrast uptake, and the patient may have transient expansion of the tumor capsule. ³⁵ Overall, radiosurgery has been established as a minimally invasive alternative to microsurgery for VS, with studies showing the long-term efficacy, safety, and cost-effectiveness of SRS.

**Fig. 20.2** An 87-year-old man with reduced hearing in the left ear over several years and new onset of left facial numbness due to a large left cystic vestibular schwannoma. **(a)** Axial, **(b)** sagittal, and **(c)** coronal T1-weighted gadolinium-enhanced magnetic resonance images (MRIs) with stereotactic radiosurgery (SRS) plan to treat the cystic vestibular schwannoma. The marginal dose was 11.5 Gy at 50% isodose, and the target volume was 6.16 cm³. Axial **(d)** T1-weighted gadolinium-enhanced MRI and **(e)** T2-weighted constructive interference steady-state MRI at 6 months and then again **(f,g)** at 14 months after SRS show collapse of the cyst and partial regression of the tumor.

Meningiomas

Meningiomas commonly occur at the skull base, including at the foramen magnum, clivus, and petrous bone. Tumors within these locations can compress the brainstem. The evolution of SRS has impacted the management algorithm of cranial base tumors including meningiomas. In addition to offering surgical resection, observation, or fractionated radiation therapy, the multidisciplinary team can offer radiosurgery as a primary or adjuvant treatment.

Meningiomas are excellent candidates for radiosurgery because their borders are clearly demarcated and they rarely invade the brain. The overall tumor control rate of SRS after prior resection of WHO grade I meningiomas is 93%. ³⁶ Similarly, tumor control is above 90% for meningiomas without a history of prior resection. ³⁶ Furthermore, rather than observing a patient after subtotal resection, such as residual tumor within the cavernous sinus or superior sagittal sinus, we advocate postoperative SRS to reduce the risk of complications from delayed progression. ³⁷ Because multiple studies have shown that untreated meningiomas grow over time, observation no longer seems to be the best choice, particularly for symptomatic meningiomas compressing the brainstem. Higher WHO grade, larger target volume, SRS after failed surgery, and convexity tumors are predictors of lower tumor control. ³⁸ Control rates for WHO grade II and III tumors have been reported as 50% and 17%, respectively. ³⁶ Better control rates are obtained with tumor margin doses above 15 Gy, ³⁹ compared with margin doses of 12 to 14 Gy commonly used for WHO grade I tumors. Overall morbidity rates for all intracranial meningiomas have been reported at approximately 7%. ⁴⁰ In a study of 246 patients with benign skull base tumors causing brainstem compression, including 44 meningiomas, the control rate after SRS was 100% in meningiomas (52.3% regressed, 47.7% stable). ³² Brainstem compression improved in 50% of meningioma patients and was associated with symptom improvement in 43.2% of patients. Only one meningioma patient (2.3%) experienced an ARE of persistent facial weakness. ³²

Overall, radiosurgery is a minimally invasive option for patients with benign skull base tumors that compress or distort the brainstem ( Fig. 20.3 ). SRS results in high rates of tumor control and preservation of neurological function, while avoiding the risks associated with open surgery of skull base lesions.