Radiation therapy for Graves’ ophthalmopathy: trick or treat?

Abstract

Listen

Translate article

In this issue of Ophthalmology, Gorman et al present long-awaited results of a double-masked, randomized, prospective study evaluating external beam radiation for the treatment of Graves’ ophthalmopathy. The ability to generalize their results depends primarily on how effectively the study addresses four major issues: (1) timing of therapy, (2) variable and multiple clinical manifestations, (3) confounding therapies, and (4) scientific validity of the experimental design. Timing is a critical issue regarding efficacy of therapy. Graves’ ophthalmopathy typically goes through three phases—progression, followed by stabilization and, perhaps, some improvement. This cycle typically has a duration between 6 months and 2 years, the so-called Rundle’s curve. If a patient has residual signs of ophthalmopathy after the disease has reached a plateau, then these signs are likely to be the result of fibrosis and other permanent tissue changes rather than persistent inflammation amenable to radiation therapy. Many physicians may be reluctant to treat clinically stable disease with radiation therapy. In Gorman et al’s study, entry varied between 0.2 and 16 years after the onset of eye symptoms (median, 1.3 years). Thus, many of the patients included in the study would not have progressive eye symptoms or signs indicative of an ongoing orbital process. This conclusion is supported by the lack of change in clinical and computed tomography measures in the untreated eye after 6 months. The authors use a positive serum thyroid-stimulating immunoglobulin as a surrogate of active eye disease. However, they follow parameters indicative of clinical progression as the measures of outcome. As a consequence, the authors confuse immunologic activity with clinical progression. In fact, thyroid-stimulating immunoglobulin levels did decrease after treatment! The clinical manifestations of Graves’ ophthalmopathy tend to be highly variable. Only a small percentage of patients experience severe consequences such as persistent diplopia or optic neuropathy. A larger number experience varying and fluctuating degrees of ocular irritation, lid retraction, exophthalmos, and periorbital swelling, alone or in combination. This variability in clinical presentation complicates the process of patient selection and the assessment of therapeutic response. For instance, Gorman et al use computed tomography measurement of extraocular muscle and fat volume, proptosis, range of extraocular motion, area of diplopia fields, and lid fissure width as main outcome measures. But these separate measures may show divergence, with some measures improving and others worsening, making an overall assessment of a patient’s response to therapy ambiguous. Global measures of ophthalmopathy, such as the Mourits’ activity score, also used in this article, have been developed in an attempt to homogenize the clinical presentation into a singular value. The advantage of having a pseudoquantitative lumped value is offset by the danger of equating severity of different clinical signs, or as the saying goes, “mixing apples and oranges.” For example, is severe periorbital edema equivalent to moderate muscle restriction? Is an improvement in periorbital edema meaningful if the patient is reporting diplopia? In the clinical setting, patients who receive orbital radiation often have been treated sequentially or concurrently with corticosteroids, surgery, or both. Nineteen of 42 patients in the study by Gorman et al had been treated previously with corticosteroids. No difference was found between the corticosteroid-treated subgroup and the noncorticosteroid-treated subgroup. However, there is no statement as to the temporal relationship between the radiation and the corticosteroid therapy, except that only two had received steroid therapy within 30 days of starting the study. Patients treated with corticosteroids in the remote past may have had a lasting response to therapy, making radiation therapy redundant and, thereby, less likely to have an effect. However, failure to respond to prior corticosteroid treatment may predict a poor response to radiation. In concluding that radiation therapy has no benefit in the management of Graves’ ophthalmopathy, the study by Gorman et al fully addresses only one of several important issues. Their use of a prospective double-masked, placebo-controlled, randomized study design was excellent and should serve as a model for future research into this important topic. Unfortunately, the usefulness of their results is markedly diminished by the broad patient inclusion criteria that lack rigor in controlling the issues of timing of therapy, clinical variability in presentation, and multiple treatment methods for individual patients. This study does demonstrate that many patients with Graves’ ophthalmopathy will not benefit from radiation therapy. However, considerable caution should be exercised in generalizing the conclusions reached by the authors regarding management of patients with clinically progressive symptoms and signs of Graves’ ophthalmopathy. Although I personally remain skeptical regarding the possible benefits of external beam orbital radiation, a larger, multicenter study using carefully defined clinical progression as a criterion for eligibility must be performed.