Discordance between proptosis response and other clinical improvements at week 12 during teprotumumab N01 therapy for thyroid eye disease: a multicenter prospective registry study

An early proptosis response threshold may not capture concurrent changes in other manifestations of thyroid eye disease (TED). We examined the coexistence of conventional Hertel nonresponse with improvement in inflammatory and patient-reported outcomes at Week 12 among patients receiving teprotumumab N01 (IBI311; SYCUME) in China. This multicenter prospective observational registry included patients treated with teprotumumab N01, a ready-to-use liquid formulation, at four Chinese centers. The safety set included all treated participants with safety documentation; the Week-12 full analysis set (FAS) comprised participants who had reached the early assessment window. Hertel exophthalmometry and the 7-item Clinical Activity Score (CAS) were assessed in the same predefined study eye, selected by greater baseline proptosis and fixed during follow-up. We examined discordance between study-eye Hertel nonresponse (proptosis reduction < 2 mm) and improvement in CAS and/or Graves Ophthalmopathy Quality of Life (GO-QOL), using outcome-specific available-case denominators. The safety set included 98 participants and the Week-12 FAS included 64. Paired Week-12 data were available for Hertel exophthalmometry ( n = 58), CAS ( n = 58), diplopia ( n = 56), and GO-QOL ( n = 44). Hertel response occurred in 23/58 (39.7%), CAS response in 42/58 (72.4%), diplopia improvement in 23/56 (41.1%), and GO-QOL improvement in 32/44 (72.7%). Hertel nonresponse coexisted with CAS and/or GO-QOL improvement in 29/54 (53.7%) participants classifiable for the combined analysis. Among 33 Hertel nonresponders with at least one evaluable CAS or GO-QOL result, 29 (87.9%) had improvement in one or both domains; this was a conditional composite estimate, not an overall response rate. In separate paired analyses, Hertel nonresponse with CAS improvement occurred in 24/52 (46.2%), and Hertel nonresponse with GO-QOL improvement in 16/41 (39.0%). Any adverse event was recorded in 64/98 (65.3%) participants. In this prospective Chinese registry of teprotumumab N01 treatment, failure to achieve the conventional 2-mm Hertel threshold at Week 12 frequently coexisted with improvement in CAS and/or GO-QOL. The findings support interpreting early anatomical classifications alongside other measured clinical domains; the uncontrolled design does not establish that all observed changes were treatment attributable. This was a prospective observational registry without investigator-assigned treatment allocation; therefore, no interventional trial registration was required.