OBJECTIVEIn moderate-to-severe Graves' orbitopathy (GO), rituximab is recommended as second-line therapy in patients non-responsive to intravenous glucocorticoids. We aimed to evaluate rituximab as early second-line therapy, as data is scarce and contradictory.METHODSIn this non-randomized, controlled, interventional study, patients with GO started on intravenous glucocorticoids. After 4 weeks, patients with <2 points improvement in clinical activity score (CAS) were switched to rituximab [Non-Responders Rituximab (NR-RTX) group] and were compared to the remaining patients who continued with intravenous glucocorticoids for 12 weeks [Responders-Glucocorticoid (R-GC) group]. A retrospective group of non-responsive patients who were provided regular care with intravenous glucocorticoids for 12 weeks was used as control [Non-Responders-Regular Care (NR-RC) group]. Background data and CAS were recorded for all groups at 0, 4, 12, 18, and 68 weeks. Quality of life (QoL) and safety data were collected from the NR-RTX and R-GC groups.RESULTSThe NR-RTX group (n=12) was similar to the others at baseline except for a 1-point lower median CAS compared to the NR-RC group (n=12) (p=0.03), and for having twice as many men compared to the R-GC group (n=13) (p=0.03). At 4 weeks, a linear mixed model indicated that the R-GC group had a 1.21-point (95%CI: -2.40 to -0.02) lower value for CAS compared to the NR-RTX group. CAS for all groups converged over time. Similar models for QoL revealed no treatment or time effects.CONCLUSIONSwitch to RTX early in the treatment course did not result in better CAS or QoL, compared to continuous intravenous glucocorticoids.