Lymphoma of the orbit is a rare presentation of non-Hodgkin’s lymphoma (NHL), representing 8% of all extranodal NHL  and only 1% of all NHL. Localized orbital lymphomas are usually managed with radiotherapy alone with excellent local disease control. Because of the rarity of this disease and the effort to preserve the integrity of the orbit without compromising local control, radiation treatment is often technically challenging.

Many of the published studies group lymphomas of various histologies together, making the data difficult to interpret. Orbital lymphomas are predominantly of mucosa-associated lymphoid tissue (MALT) histology, although they include also other histologic subtypes, such as follicular lymphomas, diffuse large B-cell lymphomas (DLCL), and mantle cell lymphomas. Although many series have reported excellent local control and long-term survival for orbital lymphomas of MALT subtype using modest doses of radiation alone, others have observed a higher risk of distant relapse associated with MALT lymphomas of the orbit than with those arising in other extranodal sites

In this study, we summarize our experience in a cohort of patients with orbital lymphoma treated at our institution between 1987 and 2003. The clinical presentation, treatment, local control, long-term clinical outcome, and toxicities are reported, with emphasis on the two most common histologic subtypes, MALT and follicular lymphomas. Various radiation techniques employed at our institution are described.

Forty-six patients (and 62 eyes) with orbital lymphoma treated with radiotherapy between 1987 and 2003 were included. The majority had mucosa-associated lymphoid tissue (48%) or follicular (30%) lymphoma. Seventeen patients had prior lymphoma at other sites, and 29 had primary orbital lymphoma. Median follow-up was 46 months.

Radiation therapy: Dose and techniques

Radiation treatment was individualized based on the location and extent of orbital involvement, as well as the patient’s overall clinical condition (radical vs. palliative). , en face electrons (6 or 9 MeV) were used for 7 patients (9 eyes) in 2–2.5 Gy fractions. Photons (6 MV) were used for the remaining 39 patients (53 eyes) in 1.5–2 Gy fractions. For superficial lesions that were confined to the conjunctiva or eyelid, electron beam therapy with a contact lens block was used when the lesion was small and could be adequately covered in the presence of the contact lens block; otherwise, photons were employed with a central hanging eye block added for cases in which sufficient coverage could be ensured in the presence of the block. For patients with any involvement of intraorbital tissues, including those with superficial lesions that extended to involve deeper retrobulbar tissues, photons were used without any eye blocks. The radiation target volume included the entire orbit for patients with any intraorbital involvement. For superficial lesions without involvement of the orbit, the target volume was defined as tumor plus an adequate margin. The most common techniques for photon treatment of a single orbit were an anterior (AP) field or an anterior wedged pair designed to spare the opposite orbit. For patients with bilateral orbital involvement, opposed lateral fields, bilateral AP fields, or 3 fields (AP and opposed lateral fields) were used based on individual cases. One patient received stereotactic radiotherapy at the discretion of the treating radiation oncologist given the focal nature of the lesion. One patient received intensity-modulated radiation therapy because of prior total body irradiation, to limit the cumulative dose to the surrounding normal structures.

Contact lens shields were used with all electrons, and central hanging eye blocks were used in 8 of 53 eyes treated with photons. Bolus was often used to treat superficial disease in eyelid or conjunctiva when isodose curve could not adequately cover the disease without. Twenty of the 46 patients had CT simulation as part of their treatment planning. CT simulation was routinely used for these lesions after it became available in our institutions.

Results: The median dose was 30.6 Gy; one-third received <30 Gy. Electrons were used in 9 eyes with disease confined to the conjunctiva or eyelid, and photons in 53 eyes with involvement of intraorbital tissues to cover entire orbit. Local control rate was 98% for all patients and 100% for those with indolent lymphoma. Three of the 26 patients with localized primary lymphoma failed distantly, resulting in a 5-year freedom-from-distant-relapse rate of 89%. The 5-year disease-specific and overall survival rates were 95% and 88%, respectively. Late toxicity was mainly cataract formation in patients who received radiation without lens block.

Toxicity of treatment

Orbital radiation therapy was generally well tolerated and required no treatment breaks. Mild acute side effects included transient conjunctivitis, periorbital erythema and swelling, dryness, or excessive tearing.

The major late complication of orbital radiation therapy was cataract formation. Nine eyes had documented cataracts that required surgical correction, with a median time to surgery of 37 months (range, 19–62 months). All 9 eyes had been treated with photons (range, 30.6–40.6 Gy) in the absence of a lens block, resulting in a 20% crude risk of cataract development requiring surgery in those receiving photons without a lens block (Table 4). None of the 17 eyes treated with a lens block (8 photons and 9 electrons) had developed cataract requiring surgery.

Corneal erosion occurred in 1 patient who received electron beam therapy for a conjunctival lymphoma. This patient’s treatment was discontinued after 18 Gy (of planned 30 Gy) at 2 Gy fractions, because of the development of periorbital cellulitis that was difficult to control with antibiotics. She subsequently underwent a keratoplasty for corneal perforation 8 months after radiation therapy. Persistent, mild dry eye requiring up to once-daily eye drops was reported in 30% of patients. No vision deterioration or retinal damage as a result of radiation was documented.

Discussion

This study reports on the clinical presentation, treatment, outcome, and toxicity of 46 patients with orbital lymphoma who received radiation therapy. The results showed that the local control of orbital lymphoma with a modest dose of radiation (median dose, 30.6 Gy) was excellent, and treatment morbidity was minimal.

In this series of patients, MALT lymphoma was the most common histologic subtype in patients with primary disease, accounting for 59% of the cases. Follicular lymphoma was the second most common subtype, accounting for 24% of cases. In comparison, among patients who developed recurrence in the orbit after prior lymphoma elsewhere, follicular histology was the most common (41%), followed by MALT lymphoma (29%).

In the current study, a 98% local control rate was achieved with radiation therapy to a median dose of 30.6 Gy; one-third of patients received less than 30 Gy. The only patient who failed locally had transformed DLCL. All 41 patients with indolent lymphoma achieved local control. Our finding that a modest dose of radiation therapy is sufficient for low-grade orbital lymphoma is consistent with many other studies. For example, Uno et al. showed that a dose of >30 Gy had no impact on local control in their series of 50 patients with orbital MALT lymphoma. A multi-institutional experience with 90 patients with mostly low-grade disease showed a 97% local control rate with a median dose of 34.2 Gy. Le et al. showed a 100% local control rate in 31 cases of orbital lymphoma and observed no difference in outcome between patients treated to 34 Gy and those treated to higher doses; however, 2 patients suffered significant retinal damage after receiving >34 Gy

Furthermore, because one-third of our patients with indolent orbital lymphoma received <30 Gy and there were no local recurrences among these patients, doses of 25–30 Gy may be sufficient for indolent orbital lymphoma. Other investigators have recommended a dose lower than 30 Gy for low-grade orbital lymphoma. Princess Margaret Hospital reported 2 local failures in 30 patients with orbital MALT lymphoma after receiving 25 Gy. However, additional data with more patients are needed to determine the optimal dose that would ensure local control while minimizing morbidity.

An adequate target volume for treating orbital lymphoma is essential. A recent study comparing partial-orbit radiation therapy with whole-orbit radiation therapy showed that 4 of 12 patients (33%) treated with a partial-orbit volume (the lesion seen on CT plus a 1-cm margin) suffered from intraorbital recurrence outside of the initial target volume, as opposed to 100% local control in 11 patients treated with whole-orbit radiation therapy. This may be explained by the existence of microscopic disease in seemingly unaffected intraorbital areas. Therefore, in our practice, we use photons to treat the entire orbit for lesions with any intraorbital involvement. A central hanging eye block is added only in cases when the disease is confined to the conjunctiva or eyelid and can be adequately covered in the presence of a lens shield.

This report demonstrates good systemic control, with a 5-year actuarial freedom-from-distant-relapse rate of 89% for patients with localized primary lymphoma. Moreover, with a median follow-up of nearly 4 years, only 1 of the 16 patients with localized MALT lymphoma failed distantly. However, some studies have shown a higher risk of distant relapse for MALT lymphoma of the orbit when compared to MALT of other sites.   With a median follow-up of 5 years, Tsang et al. showed that 6 of 30 patients with orbital MALT lymphoma recurred in either the contralateral orbit or distant MALT locations, whereas none of 13 gastric MALT lymphomas relapsed. They further proposed the use of early systemic treatment for localized orbital MALT lymphoma. However, these findings have not been confirmed by other studies  or the present study. Certainly, longer follow-up time and larger numbers of patients are needed to determine whether a high risk of distant relapse is associated with MALT lymphoma of the orbit. Nevertheless, it is reassuring that these relapses can generally be well controlled by salvage local therapy, and patients enjoy long disease-free intervals

Treatment of orbital lymphoma with radiation is technically challenging because of the presence of the radiosensitive lens, lacrimal gland, and retina, which are located near or within the target volume. The current study shows that with a median dose of 30.6 Gy, radiation was well tolerated. Thirty percent of patients in this study developed mild dry eye, and none had retinal damage. Cataracts, seen only in patients treated with photons in the absence of a lens block, were the major late complication. Given the minimal morbidity of cataract surgery and the evidence that partial-orbit irradiation is associated with an unacceptably high risk of intraorbital recurrence, we recommend the use of a lens shield only in cases where adequate tumor coverage can be assured. In addition, orbital MRI before treatment may help assess the extent of disease.

The present study was limited by all the inherent weaknesses of any retrospective study. Nevertheless, based on the previously available data and the results of this study, we conclude that a dose of 30 Gy can achieve excellent local control for indolent orbital lymphoma. The distant relapse rate in patients with localized orbital lymphoma was lower than that reported for low-grade lymphoma presenting in other sites, and survival was good. The predominant histology was MALT, and the clinical behavior and treatment outcome of these MALT lymphomas were similar to those of the other subtypes. Cataracts were the major late complication. Electrons (2–2.5 Gy per fraction) with a contact lens block can be used to treat disease confined to conjunctiva or eyelid. Photons (1.5–2 Gy per fraction) should be used for any intraorbital involvement to treat the entire orbit. A central hanging lens block can be added to photons only when adequate tumor coverage is not compromised with the presence of the block.
Conclusions A dose of 30 Gy is sufficient for indolent orbital lymphoma. Distant relapse rate in patients with localized orbital lymphoma was lower than that reported for low-grade lymphoma presenting in other sites. Orbital radiotherapy can be used for salvage of recurrent indolent lymphoma.