Radiotherapy in orbital lymphoma: optimizing the backbone to build upon

Abstract

Treatment for primary orbital lymphoma (POL) has not changed signifi cantly for several decades. Radiotherapy is highly eff ective, and the role of chemotherapy or antibiotics (for marginal zone lymphoma [MZL] subtype) is less well defi ned. Th e best practice is still unclear due to the low incidence of POL and lack of prospective clinical trials (PCTs), and this is compounded by the unreliable histological classifi cation and distinction between indolent non-Hodgkin lymphoma (NHL) versus benign lymphoid hyperplasia (BLH) in many older series of POL prior to the era of immunohistochemistry [1]. Th e traditional use of radiotherapy in POL is based on retrospective studies that showed a high rate of local control (95 – 100%), particularly in indolent NHL histological subtypes, using moderate-dose radiotherapy from 25 to 40 Gy [2 – 4]. Radiotherapy-related ocular complications, in particular serious late retinopathy and optic neuropathy, appeared to be increased with doses above 35 Gy [2,5,6], whereas local recurrences, when present, were usually after doses of 25 – 30 Gy [2,3]. In recent years there has been a trend toward employing reduced radiation dose, without apparent compromise in local control; although the optimal radiation dose and fractionation remains unclear, 24 – 30 Gy is widely accepted as the lowest eff ective dose [3,4,7]. In the MZL subtype in particular, a lower dose of radiation compared to that recommended for non-gastric non-orbital sites is preferred [4]. In this issue of Leukemia and Lymphoma , Tran and colleagues [8] report the effi cacy of low-dose radiotherapy and fractionation (total 24 – 25 Gy in 1.5 – 2 Gy fractions) in a retrospective analysis of 24 patients with orbital MZL, 22 of whom were stage IEA. Only one local failure occurred at the electron beam penumbra, attributed to marginal miss. Two further relapses, one in the contralateral orbit and one systemic relapse, occurred in patients with initial stage IEA disease. Five-year progression-free survival and overall survival were 81% and 100%, respectively. Th ese results are comparable, if not better than expected, for non-orbital localized low-grade lymphomas treated with radiotherapy with curative intent, where distal relapse occurs in up to 40 – 50% of patients [9]. MZL, however, is particularly radiation-sensitive, and the effi cacy seen here with 25 Gy, which confi rms results from one other series [4], may not apply to the other 15 – 20% of POL of other indolent and aggressive NHL histological subtypes. Toxicity-wise, a dose of 30 Gy in 1.8 – 2 Gy fractions is not usually associated with signifi cant long-term ocular toxicities [5]. With both lower total dose and fractionations, Tran and colleagues report no severe acute toxicities and, more importantly, no severe late toxicities such as radiation retinopathy, visual loss, corneal ulceration, optic nerve damage or glaucoma, apart from cataracts that are amenable to surgery. Indeed this provides a strong impetus to further investigate a lower-dose schedule; however, until results are confi rmed in prospective clinical trials, this proposal cannot be heralded as standard of care as it is based on retrospective analyses of small numbers of patients. So taking this into consideration, how can we build on the information provided by Tran and colleagues to improve outcome for patients with POL? Clearly radiotherapy is the backbone on which to add, and hence defi ning the minimum eff ective dose that ensures local control yet respects the vulnerability of surrounding ocular structures is important and challenging. In this respect, prospective clinical trials to assess various radiotherapy doses for diff erent histological subtypes and subsites of POL are required. We eagerly await results from the Trans Tasman Radiation Oncology Group and Australasian Lymphoma and Leukaemia Group (TROG/ALLG) phase II trial that tests 24 Gy in 1.5 – 2 Gy fractions for orbital MZL. Questions regarding radiotherapy technique, including the required target volume (entire vs. partial orbital irradiation) for diff erent subsites of disease, degree of protection from smaller fractions and lens shielding, and the use of bolus to improve effi cacy, also need to be evaluated prospectively, as there are few data or little consensus available for POL.