CyberKnife Radiosurgery for the Treatment of Intraocular and Periocular Lymphoma

Objectives  To evaluate efficacy, patient tolerance, and adverse effects of CyberKnife radiosurgery for the treatment of intraocular and periocular lymphoma.

Methods  Retrospective case series of 13 patients who underwent CyberKnife radiosurgery was conducted.

Results  Fourteen eyes of 13 patients were included. The tissue location of the lymphoma was orbit (7 eyes), conjunctiva (3 eyes), choroid (2 eyes), and retina (2 eyes). The lymphoma type was classified as extranodal marginal zone B-cell lymphoma in 7 eyes (50%), diffuse large B-cell lymphoma in 3 eyes (21%), follicular lymphoma in 2 eyes (14%), and benign reactive lymphoid hyperplasia in 2 eyes (14%). The mean treatment dose was 1718 centigrays (cGy) (range, 1350-2250 cGy) given over a mean of 5 days (range, 3-5 days) with a mean dose rate of 320 cGy per fraction. Complete tumor resolution without local recurrence over a mean follow-up of 23 months was documented in all cases. Radiation-associated adverse effects included mild dry eye in 2 patients and cataract in 1 patient with conjunctival lymphoma. There was no radiation retinopathy or papillopathy, and visual acuity was preserved or improved in 13 eyes and decreased in 1 eye due to the presence of cataract.

Conclusions  CyberKnife radiosurgery is a well-tolerated technique for the treatment of intraocular and periocular lymphoma, allowing for local resolution of the lesions. An important benefit is that treatment was completed over 5 days.

The treatment of intraocular and periocular lymphoma varies according to specific tumor features such as size, shape, location, and grade of the lesion as well as the systemic status of the patient. Treatment options include surgical excision, chemotherapy, immunotherapy with rituximab for patients with positive CD 20 immunohistochemistry, or radiosurgery.^1-30 For patients without systemic involvement, external beam radiotherapy (EBRT) is often used after histopathologic confirmation. Both primary intraocular lymphoma of the vitreous and retina and uveal lymphoma are treated with EBRT. Soft tissue periocular involvement with lymphoma also responds to EBRT. Local tumor control rates have been greater than 70% for retinovitreal lymphoma and greater than 92% for periocular lymphoma of the conjunctiva and orbit.^21-30 Standard EBRT is delivered over 20 days (4 weeks); requires complete immobilization, often with a frame or mask; and can lead to radiation adverse effects in the periocular region, including loss of vision or the eye.^21-30

The concept of stereotactic radiosurgery was first developed by Swedish neurosurgeon Lars Leksell in 1951 with the goal of providing nonsurgical ablation of tumors deep within the brain.^31-34 Initially considered an investigational device, radiosurgery has evolved to gain wide acceptance, becoming a routine procedure for the treatment of multiple disorders.^35-53 More recently, the advancement of image-guided radiosurgery has allowed for frameless treatment, often given over several sessions and at various extracranial sites.^35-53 In 1994, American neurosurgeon John Adler (who had previously trained with Leksell in Sweden) explored robotic technology and engineered the first robotic linear accelerator capable of precise radiosurgery delivery to targets as small as 5 mm. He called this robotically driven technique CyberKnife.^32-34

CyberKnife is an image-guided radiosurgical technique that uses a linear accelerator mounted on a robotic arm, delivering up to 1300 beams from a multidirectional perspective, limiting surface tissue damage.^32-34 It allows paramount conformal precision, thus limiting damage to surrounding healthy tissue. CyberKnife provides treatment with frameless comfortable delivery over 3 to 5 consecutive days in an outpatient setting.^32-34

In this report, we review the experience of a single institution with CyberKnife radiosurgery for the treatment of intraocular and periocular lymphoma. We explore treatment efficacy, patient tolerance, and adverse effects.

The medical records of all patients with intraocular or periocular lymphoma who were evaluated at the Ocular Oncology Service, Wills Eye Institute, Philadelphia, Pennsylvania, were reviewed. Patients who received CyberKnife radiosurgery were selected for further analysis. Institutional review board approval was obtained.

The data collected included age, race, sex, best-corrected Snellen visual acuity, initial ocular symptom, involved tissue (choroid, retina-vitreous, orbit, eyelid, or conjunctiva), and lymphoma type (extranodal marginal zone B-cell lymphoma [MALT], diffuse large B-cell, follicular, or benign reactive lymphoid hyperplasia). Imaging studies performed (magnetic resonance imaging, computed tomography, or ultrasound) were recorded. Maximum tumor dimensions were recorded in millimeters and measured from ophthalmoscopy and ultrasound for intraocular lesions, magnetic resonance imaging or computed tomography for orbital lesions, slitlamp evaluation for conjunctival lesions, and external examination for eyelid lesions. The location of the tumor was classified according to quadrant involved (superior, inferior, medial, or lateral), coronal location (intraocular, muscle involvement, lacrimal gland, orbital fat, or eyelid), and axial location (anterior, midorbit, posterior, or intraocular). Previous ocular treatments and procedures (pars plana vitrectomy, incisional biopsy, excisional biopsy, or fine-needle aspiration biopsy) were recorded. The systemic status of the patient at the first visit was also recorded. Follow-up visits were scheduled at 2, 6, 12, and 24 months and then on a yearly basis. Follow-up information included visual acuity, funduscopy, Hertel exophthalmometry, Ishihara color plates, Humphrey visual field, response to therapy with local tumor status (completely regressed, partially regressed, stable, or local recurrence), systemic outcome (alive without systemic involvement, alive with systemic involvement, or dead), and complications of therapy (dry eye syndrome [defined as the presence of superficial punctate keratitis on slitlamp examination, tear film break-up test of <10 seconds, or abnormal results of the Schirmer test], radiation retinopathy, radiation papillopathy, radiation cataract, or radiation glaucoma) and their management.

CyberKnife radiosurgery variables that were recorded included tumor volume measurements, total radiation dose in centigrays (cGy), number of fractions, number of beams, isodose, maximum dose, and conformality index. Complications, events during treatment, and patient comfort were also recorded.

Fourteen eyes from 13 patients underwent CyberKnife radiosurgery for treatment of biopsy-proven intraocular or periocular lymphoma between November 1, 2006, and September 30, 2009. The mean age of the patients was 68 years (median, 66; range, 46-91 years). One patient was Hispanic (8%) and the rest were white (92%). There were 5 men (38%) and 8 women (62%). The initial visual acuity and ocular symptoms are listed in Table 1.

Seven patients (54%) had involvement of 1 intraocular or periocular structure (orbit in 5 patients [Figure 1], conjunctiva in 1 patient, and choroid in 1 patient); the remaining 6 (46%) had simultaneous involvement of more than 1 ocular structure (conjunctiva and orbit in 2 patients, retina and vitreous in 2 patients, eyelid and orbit in 1 patient, and conjunctiva and choroid in 1 patient). Six patients (46%) were diagnosed with MALT, 3 (23%) with diffuse large B-cell lymphoma, 2 (15%) with follicular lymphoma, and 2 (15%) with benign reactive lymphoid hyperplasia. Size and location of the lesions are described in detail in Table 1. Patient 3 had bilateral conjunctival involvement for which he underwent simultaneous treatment (Figure 2). All patients underwent both magnetic resonance imaging and computed tomography before the procedure (Figure 3), and 3 had additional B-scan ultrasound performed due to the intraocular location of the lymphoma (Figure 4). All patients had previous confirmation of the diagnosis by either fine-needle aspiration biopsy (patients 6, 9, and 11; Figure 5) or excisional biopsy (the remaining 10 patients). Ten patients (77%) had no systemic involvement at the initial visit. Of the remaining 3 patients (23%), 1 patient had involvement of the gastrointestinal tract (patient 6), 1 had brain lymphoma (patient 9), and 1 had multiorgan disease (patient 7).

After a median follow-up of 23 months (mean, 23; range, 6-37 months), all patients had complete regression of the local lesions with no signs of local tumor recurrence. Systemic status of the patients at the last follow-up visit is shown in Table 1. Of the 2 patients with retina-vitreous lymphoma, 1 showed the presence of a stable systemic disease with brain involvement at the last visit, having received treatment to the brain with EBRT and systemic chemotherapy before the application of CyberKnife (patient 6), and the other (patient 9) developed brain involvement during follow-up and was treated with EBRT, with stable systemic status at date last seen. Patient 2 developed contralateral lacrimal gland lymphoma that was treated with excisional biopsy, and patient 7 died after developing multiorgan involvement that was treated with chemotherapy. One patient with orbital benign reactive lymphoid hyperplasia (patient 12) was found to have cervical lymph node involvement during the follow-up, with the pathology specimen showing MALT.

Visual acuity at the last visit remained stable in 5 eyes (36%), improved in 8 eyes (57%), and decreased in 1 eye (7%). The reason for the decreased vision in this eye with conjunctival involvement was the presence of posterior subcapsular radiation cataract. At the final visit, the visual field and color vision remained stable in 13 eyes and improved in 1 eye (patient 12). The radiation-associated adverse effects recorded at the last visit included dry eye syndrome in 2 patients (patients 3 and 8), which was treated with artificial tears in patient 3 and with punctal plug placement in patient 8, and cataract formation in patient 8, which was managed with cataract extraction. None of the patients developed skin erythema, radiation retinopathy, papillopathy, or glaucoma secondary to treatment during follow-up. Patient 10 reported tearing at the last follow-up visit, and patient 12 reported persistent diplopia.

CyberKnife radiosurgery variables for each patient are described in detail in Table 2. The median initial tumor volume was 14.55 mL, the median dose was 1750 cGy, the median number of treatment fractions was 5, the median number of beams was 133, the median isodose was 75 cGy, the median maximum dose was 2333.33 cGy, and the median conformality index was 1.4. Overall, treatment was well tolerated by all patients. Of 13 patients, 2 (patients 3 and 8) developed itching during treatment and 2 (patients 2 and 13) developed conjunctival erythema.

Lymphoma is a multiorgan condition that can affect intraocular and periocular structures such as the choroid, retina, vitreous, conjunctiva, and orbit.^1-20 Systemic non-Hodgkin lymphoma is estimated to affect 43 000 persons per year in the United States; ocular involvement is uncommon, representing approximately 2% of all extranodal lymphoma cases.^1-20 Lymphoproliferative lesions include a wide range of diseases such as benign reactive lymphoid hyperplasia, atypical lymphoid hyperplasia, and lymphoma, based on varying degrees of histopathologic and immunophenotypic features.^1-30

The approach to treatment of a lymphoid tumor of the eye and its adnexa includes an assessment of the patient's ocular and systemic status.^1-30 If the patient shows no sign of systemic lymphoma after an evaluation by the oncologist, then the ocular tumor is addressed with complete excision plus cryotherapy if the mass is small and accessible (lymphoma of the conjunctiva or anterior orbit) or incisional biopsy and low-dose radiosurgery if the mass is large (posterior orbital tumors). In the case of choroidal or retinovitreal lymphoma, fine-needle aspiration biopsy followed by EBRT or chemotherapy is usually performed. If medical evaluation reveals systemic lymphoma, treatment of the ocular disease is secondary to systemic treatment (usually chemotherapy).^1-30

The effectiveness of EBRT for the treatment of lymphoma of the eye and its adnexa has been widely documented.^21-30 In the ophthalmic literature, Kennerdell and associates²¹ described local tumor control of 95% at a mean follow-up of 7 years in 54 patients with orbital lymphoma treated with low-dose (24 Gy) EBRT. Similarly, Isobe and colleagues²⁴ reported the use of moderate-dose EBRT (median tumor dose of 30.6 Gy) for the treatment of orbital lymphoma in 24 patients, with complete remission in 92% of the patients. The same authors³⁰ described the use of EBRT in 15 patients with primary intraocular lymphoma with a control rate of 74% at 2 years. Local recurrence was found in only 1 of these patients.³⁰ Uno and associates²⁹ reported that 98% of 50 patients with MALT of the ocular adnexa had tumor control at 24 months. Berenbom et al²⁵ described no cases of tumor recurrence in 7 patients with primary intraocular lymphoma treated with EBRT. Liao and associates²⁷ described local disease control in all 20 patients with orbital lymphoma treated with EBRT after a mean follow-up of 4.7 years. In the present series, all 14 eyes showed complete resolution of lymphoma after CyberKnife radiosurgery with stability at a mean follow-up of 23 months. Since these are early results, longer follow-up is needed to assess the status of the lesions.

Adverse effects of conventional EBRT are well known.^21-30 In a series of 54 patients with orbital lymphoma treated with EBRT, Kennerdell and colleagues²¹ found that 33% of the patients treated with EBRT for orbital lymphoma developed chronic dry eye syndrome after a mean follow-up of 7 years. Similarly, Isobe and associates²⁴ reported that after a median follow-up of 37 months, 13% of the patients developed cataract after use of moderate-dose EBRT for the treatment of orbital lymphoma. Berenbom and associates²⁵ described dry eye syndrome and cataract in 57% and radiation retinopathy in 29% of patients after EBRT for the treatment of primary intraocular lymphoma. Liao and et al²⁷ reported that 45% of patients developed dry eye syndrome and 35% had cataract after EBRT for the treatment of orbital lymphoma after a mean follow-up of 4.7 years. In the current series of 14 eyes with lymphoma treated with CyberKnife radiosurgery, the only adverse effects observed were mild dry eye syndrome in 2 eyes (14%) and cataract in 1 eye (7%) after a mean follow-up of 23 months (both patients had conjunctival involvement). To our knowledge, this is the largest series in the ophthalmic literature of lymphomatous lesions of the eye and adnexa treated with CyberKnife radiosurgery, and early results demonstrate relative safety. Longer follow-up is necessary to explore long-term adverse effects.

In the ophthalmic literature, Hirschbein and associates³⁵ described the use of CyberKnife radiosurgery for the treatment of orbital tumors, including 2 patients with conjunctival lymphoma and 3 with orbital lymphoma, resulting in complete resolution of the lesions after a mean follow-up of 8 months. In that small series, there were no cases of intraocular lymphoma (choroid or retina).³⁵

The advantages of CyberKnife include, among others, patient comfort during frameless radiosurgery and the ability to fractionate the treatment dose.^32-34 Dose fractionation can enable the delivery of a higher total dose to the tumor while maintaining safe daily doses to surrounding critical structures. Fractionation of treatment is an important concept in radiation oncology because it takes advantage of the disparate biologies of healthy and neoplastic tissues, allowing preservation of the former and cell damage to the latter.^32-34 Fractionation allows time for healthy cells to recover, while tumor cells are generally less efficient in achieving repair between fractions. CyberKnife is usually fractionated over 5 days. Last, CyberKnife treatment allows for conformal fields that can be round or irregular so that the robotic machine can sculpt an unusually shaped area (nonisocentric planning) for radiosurgery treatment and avoid adjacent structures.^32-34

Conventional radiation therapy has proven to be effective for the treatment of ocular lymphoma. However, it uses a wide field of delivery with the risk of radiation-induced tissue injury to normal structures and even a risk for related secondary cancers.^21-30 This technique is well established and reliable, although some high-dose treatments may be limited by the radiation toxicity of healthy tissues near the target area. CyberKnife has the capacity to minimize irradiation of nearby critical structures through the use of multiple beams coming from different directions, with the intent to reduce collateral damage.^35-53 This capacity could be beneficial in the treatment of ocular lesions, in which radiation tolerance of the globe and optic nerve is considered. In a series of 49 patients with perioptic tumors (within 2 mm of the optic nerve) treated with CyberKnife radiosurgery, Adler and associates³⁹ found that 94% of patients maintained or improved their visual field after a mean follow-up of 49 months. In a series of 16 patients with orbital tumors treated with CyberKnife, Hirschbein and associates³⁵ found that all patients had stable or improved visual field after 8 months of follow-up. Romanelli and associates³⁸ described the use of CyberKnife in 3 patients with optic nerve sheath meningiomas with improvement in the visual field in all patients after the procedure.

In summary, CyberKnife radiosurgery for the treatment of intraocular and periocular lymphoma in 14 eyes allowed for complete resolution without local recurrence. The 5-day frameless procedure was well tolerated by all patients. Radiation-associated adverse effects included mild dry eye syndrome in 2 eyes (14%) and cataract in 1 eye (7%). From these observations, it is evident that CyberKnife radiosurgery is a useful procedure for the treatment of intraocular and periocular lymphoma with minimal adverse effects. Further follow-up is warranted.

Correspondence: Carol L. Shields, MD, Ocular Oncology Service, Wills Eye Institute, 840 Walnut St, Ste 1440, Philadelphia, PA 19107 (carol.shields@shieldsoncology.com).

Submitted for Publication: February 9, 2010; final revision received March 11, 2010; accepted March 29, 2010.

Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Financial Disclosure: None reported.