Stereotactic Radiosurgery of Mobile Extraocular Muscle Metastasis: A Noninvasive Approach Using MRI Simulation

Clinical Correspondence Section Editors: Robert Avery, DO Karl C. Golnik, MD Caroline Froment, MD, PhD An-Guor Wang, MD Stereotactic Radiosurgery of Mobile Extraocular Muscle Metastasis: A Noninvasive Approach Using MRI Simulation Benjamin J. Rich, MD, Joshua Pasol, MD, Michael E. Ivan, MD, Michael A. Schaffer, MD, John C. Ford, PhD, Eric A. Mellon, MD, PhD A 63-year-old woman with invasive ductal carcinoma (estrogen and progesterone receptor positive and human epidermal growth factor receptor 2 negative) presented to her ophthalmologist with intermittent, vertical diplopia and impaired ocular motility. She had no previous history of central nervous system or orbital involvement. On examination, she had a positive forced duction to upgaze on the left eye with an elevation deficit, and her diplopia was corrected with prism glasses. MRI of the orbits demonstrated a 4-mm enhancing lesion in the distal left inferior rectus muscle. During the workup to treatment period of about 2 months, the vertical diplopia continued to worsen with a final size at treatment of 6 mm (Fig. 1). The patient was referred to radiation oncology for management. Motion of the inferior rectus nodule during treatment is expected because the extraocular muscles and optic nerve move significantly with changes in gaze (1), which could be unpredictable with partial gaze palsy. The margin required for stereotactic radiotherapy would put the patient at unacceptable risk of toxicity, limiting treatment to conventional fractionation (e.g., 30 Gy in 10 fractions) with still significant risks of cataract and conjunctivitis. To better define the target, a separate MRI simulation was performed the day before treatment using a MRIdian hybrid MRI-linear accelerator system (ViewRay, Inc, Oakwood Village, OH) with the patient immobilized in an MRI-compatible thermoplastic mask (Civco MRSeries, Coralville, IA). The MRIdian uses a wide bore 0.35 T field strength MRI, and the default imaging uses balanced steady-state free precession with heavy T2 (2). MRIs of the orbit were acquired on the MRIdian; each while the Department of Radiation Oncology (BJR, JCF, EAM), Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, Miami, Florida; Department of Ophthalmology (JP), Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida; Department of Neurosurgery (MEI), University of Miami Miller School of Medicine, Miami, Florida; and Delray Eye Associates P.A. (MAS), Delray Beach, Florida The authors report no conflicts of interest. Address correspondence to Eric A. Mellon, MD, PhD, Department of Radiation Oncology, Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, 1475 NW 12th Avenue, D-31, Miami, FL 33136; E-mail: [e-mail] Rich et al: J Neuro-Ophthalmol 2022; 42: e551-e553 patient was instructed to gaze neutral, up, down, left, or right. The patient had difficulty with upward gaze, and the patient was instructed not to look up during stereotactic radiosurgery (SRS) treatment. Comparing the images in all eye positions, the total anterior–posterior excursion of the tumor nodule and superior–inferior excursion of the optic nerve was 1.0 cm. The excursion was asymmetric, with essentially no right–left excursion of the tumor nodule. The gross tumor volume (GTV) was contoured in MIM (MIM Software Inc, Cleveland, OH) in neutral, down, left, and rightward gaze, and an internal target volume (ITV) of 0.434 cc was created by combining the GTVs (Fig. 2). A planning risk volume (PRV) for the left optic nerve was created combining the optic nerve position in neutral, up, down, left, and rightward gaze plus 1 mm for safety margin. An avoidance structure termed “front of the left eye” included the lens in all gaze positions, anterior chamber, and conjunctiva. We decided to perform SRS because it offers a high probability of local control while minimizing the radiation dose to surrounding structures. Given that our center has Gamma Knife (GK) (Perfexion; Elekta Instruments, Stockholm, Sweden), we chose GK for frame-based delivery without planning target volume (PTV) expansion and excellent dosimetric characteristics for this small target next to the retina. GammaPlan software (Elekta Instruments) was used to calculate radiation doses. The rectus muscle tumor ITV was treated with a dose of 20 Gy to the 50% isodose line by 12 shots of mixed 4- and 8-mm collimation over 15 minutes. Maximum doses to nearby organs at risk were 5.2 Gy left optic nerve PRV, 6.0 Gy front of the left eye, 0.5 Gy brainstem, 0.3 Gy optic chiasm, and 0.6 Gy right optic nerve. Two months after treatment, an MRI of the brain and orbits demonstrated decreased volume of the left inferior rectus muscle metastasis. At the last follow-up 12 months from treatment, the patient’s diplopia persisted, although stable from treatment, and MRI showed the lesion to have decreased to 4 mm. The patient’s vision was corrected with prism glasses. The patient had no acute or late toxicity from radiation treatment. Although most radiation-related toxicities would be evident at 1 year, some late toxicities of radiotherapy such as optic neuropathy and retinopathy may occur e551 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical Correspondence FIG. 1. Stereotactic MRI of the brain postcontrast demonstrating left inferior rectus tumor. Postcontrast T1 images before treatment with axial (A) and coronal (B) views. The left inferior rectus tumor is demonstrated on both images with a white arrow. years after treatment, although optic neuropathy is unlikely at a maximum 5.2 Gy optic nerve dose and retinopathy should be confined to the most inferior retina given tight dose fall off and lack of PTV in our SRS delivery (3). Although radiosurgery is an uncommon treatment modality for palliating orbital metastasis, it has been shown to be successful at resolving diplopia secondary to orbital metastasis (4). Treating a tumor in the mobile extraocular muscles poses a challenge for radiosurgery. A stereotactic frame immobilizes the skull, but does not fix the eyes in place. Treatment with GK radiosurgery can take hours, and ocular movements must be accounted for when targeting mobile orbital structures. Techniques to immobilize the eye include tethering sutures, suction fixation devices, or a local nerve block (4). Unique in this case was the use of a target consisting of an ITV generated with a dynamic MRI simulation. This provides a noninvasive solution for the treatment of mobile extraocular muscle metastases for radiation oncology departments capable of MRI simulation. This may provide more accuracy than a uniform PRV, particularly in a patient with ophthalmoplegia. Although movements at the anterior end of the optic nerve at extremes of gaze average greater than 1 cm, another report has suggested that PRV and PTV sizes of up to 5 mm are sufficient in the treatment of optic nerve sheath meningiomas (5). Therefore, as demonstrated here, patient-specific measurements may be helpful. In the future, an MRI-linear accelerator system could also provide treatment of the target; however, at the present time, we have found the GK to provide more conformal radiation plans than the MRIdian for smaller targets. Realtime tracking of the eye position during radiation by MRIdian would also be very appealing with 4D cine; however, current cine imaging during radiotherapy is only used in the inferior–superior axis to account for body respiratory motion. In conclusion, we report a case of progressing diplopia secondary to an extraocular muscle metastasis treated with radiosurgery using a novel noninvasive MRI technique for FIG. 2. Gross target volumes used to create internal target volume. Gross target volume (GTV) of the inferior rectus tumor was contoured in MIM using 0.35 T MRI. Depicted is the GTV contour overlying an MRI in the sagittal plane with the left eye in the (A) neutral, (B) down, (C) left, (D) right, and (E) up position. The GTVs are contoured on each image in violet. Organs at risk included the front of the eye (contoured in blue), the lens (contoured in brown), and the optic nerve (contoured in red). F. The GTVs from the eye position (A–D) are used to create the stereotactic radiosurgery-treated internal target volume (contoured in orange) shown on the MRI with the eye in neutral. Similarly, the lens and left optic nerve in each eye position are shown. A left optic nerve planning risk volume to account for all possible positions of the optic nerve for avoidance is depsicted in light green. e552 Rich et al: J Neuro-Ophthalmol 2022; 42: e551-e553 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical Correspondence treatment planning to account for all eye and target positions. This technique may be used for other types of mobile orbital malignancies, increasing accuracy and reducing toxicity in this critical anatomical region. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: B. J. Rich and E. A. Mellon; b. Acquisition of data: B. J. Rich, M. A. Schaffer, and E. A. Mellon; c. Analysis and interpretation of data: B. J. Rich, J. Pasol, Mi. E. Ivan, M. A. Schaffer, J. C. Ford, and E. A. Mellon. 2. Category 2: a. Drafting the manuscript: B. J. Rich, J. C. Ford, and E. A. Mellon; b. Revising it for intellectual content: B. J. Rich, J. Pasol, J. C. Ford, and E. A. Mellon. 3. Category 3: a. Final approval of the completed manuscript: B. J. Rich, J. Pasol, M. E. Ivan, M. A. Schaffer, J. C. Ford, and E. A. Mellon. Rich et al: J Neuro-Ophthalmol 2022; 42: e551-e553 REFERENCES 1. Abràmoff MD, Van Gils AP, Jansen GH, Mourits MP. MRI dynamic color mapping: a new quantitative technique for imaging soft tissue motion in the orbit. Invest Ophthalmol Vis Sci. 2000;41:3256–3260. 2. Mehta S, Gajjar SR, Padgett KR, Asher D, Stoyanova R, Ford JC, Mellon EA. Daily tracking of glioblastoma resection cavity, cerebral edema, and tumor volume with MRI-guided radiation therapy. Cureus. 2018;10:e2346. 3. Durkin SR, Roos D, Higgs B, Casson RJ, Selva D. Ophthalmic and adnexal complications of radiotherapy. Acta Ophthalmol Scand. 2007;85:240–250. 4. Ares WJ, Flickinger JC, Lunsford LD. Leksell radiosurgery for orbital, uveal, and choroidal tumors. Prog Neurol Surg. 2019;34:298–305. 5. Geneser S, Chang J, Chen J, Yom S, Garsa A. Expansions for planning target volume (PTV) and planning risk volume (PRV) to account for optic nerve and lens motion. Int J Radiat Oncol Biol Phys. 2015;93:E622–E623. e553 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.