ARVO 2026: The natural history of TED

Madhura Tamhankar, MD, a professor of ophthalmology and neurology at the Scheie Eye Institute, University of Pennsylvania, introduces herself and her roles, including running the neuro-ophthalmology service, performing adult strabismus surgery, and serving as co-director of the thyroid eye disease (TED) clinic since 2018. She is presenting work at ARVO 2026 in Denver, Colorado on what she describes as the first large-scale characterization of the natural history of TED.

Historically, the course of TED has been described by the Rundle curve, which proposes an initial active (inflammatory) phase lasting roughly 18–24 months after the diagnosis of Graves’ disease, followed by a chronic or cicatricial phase in which the disease supposedly stabilizes. In that model, patients either improve or are left with residual problems (such as proptosis or diplopia) that may require surgery. Hanker emphasizes that this influential model was, surprisingly, based on only 2 patients (an “n of 2”), making it an extremely limited evidence base.

Fast-forwarding to 2026, her team analyzed more than 30,000 patients with thyroid eye disease using data from the IRIS Registry (the American Academy of Ophthalmology’s large, ophthalmology-specific registry) combined with Komodo Health claims data. Patients were followed for over one year, and disease progression was defined by treatment changes or worsening clinical characteristics, including proptosis, clinical activity score, or double vision.

They found that about one-third of patients experienced worsening Graves’ ophthalmopathy over the follow-up period. This challenges the long-held assumption that TED naturally “burns out” after an initial active phase. Instead, the data suggest that progression can occur over a patient’s lifetime, indicating the Rundle’s curve is not a reliable description of the typical disease course.

Crucially, over 92% of patients were initially detected with mild TED, and only a little over 5% progressed to moderate-to-severe disease, the form most associated with disabling proptosis and diplopia and major quality-of-life impairment. This underscores the importance of early detection and close monitoring.

Hanker stressed that ophthalmologists should maintain a high index of suspicion when seeing patients with recurrent tearing, redness, or subtle eye prominence, and should look carefully for eyelid retraction. They should consider thyroid eye disease as a presenting feature of underlying hyperthyroidism and feel empowered to order thyroid function tests for early diagnosis. Early identification and management—such as restoring euthyroidism, smoking cessation, lifestyle changes, and possibly selenium supplementation—may help prevent progression.

For patients who do progress, disease-modifying therapies like teprotumumab are now available, and a smaller subset with cicatricial sequelae may require surgical intervention. The overarching takeaway of the presentation is that early detection and prevention of progression are central in managing thyroid eye disease, and that the traditional dogma of Rundle curve must be reconsidered in light of robust, large-cohort data.