Ann Rheum Dis 2020; doi:10.1136/annrheumdis-2020-21841220Both ACR and EULAR recommend adding a biologic or targeted synthetic DMARD in patients who do not achieve treatment goals at follow-up. Findings indicated that an immediate switch in mechanism of action (from JAKi to TNFi and vice versa) following treat-to-target principles is feasible with minimal risk of flare regardless of whether patients are switched due to non-response or incomplete-response.SELECT-COMPARE followed treat-to-target principles to examine the efficacy of switching in two patient subgroups. Patients were randomised to receive background methotrexate plus upadacitinib 15 mg once daily, placebo, or adalimumab 40 mg every other week. Following an insufficient response, patients were blindly switched from upadacitinib to adalimumab and vice versa. Switch was undertaken prior to Week 26 for patients failing to achieve a 20% improvement from baseline in both TJC and SJC (non-responders), and at Week 26 for patients without low disease activity, defined as CDAI <10 (incomplete responders).In total, 39% and 49% of patients originally randomised to upadacitinib and adalimumab were rescued to the alternate therapy, respectively. As expected – given that incomplete responders by definition had at least 20% improvement in TJC and SJC at the rescue visits – most achieved ACR20 at switch. Overall, switching to a drug with a different mechanism of action had a beneficial effect on clinical responses in both groups, although some patients in both groups did experience disease worsening. Following immediate switch without washout, the proportion experiencing treatment-emergent adverse events was similar regardless of whether patients switched to adalimumab or upadacitinib, and no differences were observed when the same patient groups were evaluated prior to and following rescue. These observations support a treat-to-target strategy, in which patients who fail to respond are switched to a therapy with a different mechanism of action.