Golimumab for the Treatment of Rheumatoid Arthritis: A Narrative Review of Pivotal Randomized Trials and Real-World Studies Authors: Uta Kiltz, Yi-Ming Chen, Andrea Rubbert-Roth, Sergio Fonseca, José Antunes, Arun Singh, Richard Milek, Jordan Wu, Jeffrey A. Sparks Published: 2026 ABSTRACT Rheumatoid arthritis (RA) is a chronic, inflammatory autoimmune disease characterized by progressive joint involvement. The disease is associated with reduced health-related quality of life, functional ability, work productivity, and daily activity, in addition to pain, fatigue, and inflammation-induced structural damage. RA emerges through a complex interplay of multiple factors, including the release of inflammatory cytokines. One such cytokine is tumor necrosis factor (TNF), which is inhibited by golimumab, an approved treatment for the disease. Golimumab is produced using a recombinant cell line that originated from genetically modified mice immunized with human TNF. This narrative review presents the efficacy and safety outcomes from the golimumab pivotal randomized controlled trials (RCTs) that evaluated the subcutaneous and intravenous formulations. Across the RCTs, treatment was associated with an American College of Rheumatology ≥ 20% improvement. A majority of participants reported at least one AE and the most frequent were upper respiratory infection, nasopharyngitis, and bronchitis. The review also presents data from post hoc analyses and real-world studies, which demonstrated benefits that included improvements in health-related quality of life and daily productivity, prolonged drug survival, and reduced fatigue, pain, disease activity, and structural damage. Lastly, data from meta-analyses, including RCTs and real-world studies, demonstrated that golimumab had an adverse event pattern similar to other TNF inhibitors. Collectively, these studies demonstrated that golimumab is an effective and safe treatment for RA when used according to the approved indication. Key Summary Points Through structured searches of the literature from 2020 to 2025, real-world and safety data specific to golimumab were identified and are presented in this narrative review, along with the results from the golimumab pivotal trials.Based upon the results from the pivotal trials, golimumab treatment was found to be an efficacious and safe treatment for RA when used according to the approved indication. Specifically, it was associated with a reduction in tender and swollen joints, disease activity, pain, physical disability, acute phase reactants, and structural damage. Further, a low percentage of participants developed antibodies against golimumab, and the most frequently reported adverse events were upper respiratory infection, nasopharyngitis, and bronchitis.Based upon the results from the real-world studies, golimumab treatment was associated with improvements in health-related-quality of life, work and daily productivity, fatigue, and physical function, in addition to prolonged drug survival.Based upon the results from the meta-analyses, golimumab treatment was associated with an increased risk of infections and serious infections, similar to the other tumor necrosis factor inhibitors. Through structured searches of the literature from 2020 to 2025, real-world and safety data specific to golimumab were identified and are presented in this narrative review, along with the results from the golimumab pivotal trials.Based upon the results from the pivotal trials, golimumab treatment was found to be an efficacious and safe treatment for RA when used according to the approved indication. Specifically, it was associated with a reduction in tender and swollen joints, disease activity, pain, physical disability, acute phase reactants, and structural damage. Further, a low percentage of participants developed antibodies against golimumab, and the most frequently reported adverse events were upper respiratory infection, nasopharyngitis, and bronchitis.Based upon the results from the real-world studies, golimumab treatment was associated with improvements in health-related-quality of life, work and daily productivity, fatigue, and physical function, in addition to prolonged drug survival.Based upon the results from the meta-analyses, golimumab treatment was associated with an increased risk of infections and serious infections, similar to the other tumor necrosis factor inhibitors. Introduction Rheumatoid arthritis (RA) is a chronic, inflammatory autoimmune disease characterized by progressive joint involvement [1, 2]. It affects approximately 1% of adults and prevalence varies due to geographic and ethnic factors [3]. Further, RA demonstrates a sex imbalance, with women being diagnosed 2–3 times more frequently [2, 3]. Initially, small joints are affected, but as the disease progresses, larger joints are involved, and eventually extra-articular manifestations such as interstitial lung disease, inflammatory eye conditions, serositis, and glomerulonephritis occur [3]. RA profoundly diminishes health-related quality of life (HR-QoL) through the progressive deterioration of joints, coupled with the widespread systemic inflammatory complications [2]. Although RA can develop at any age, it typically emerges in people who are 40 years or older [3]. Diagnosing RA is challenging due to the absence of definitive diagnostic standards [3]. Rheumatologists frequently use the American College of Rheumatology (ACR)/European Alliance of Associations for Rheumatology (EULAR) classification criteria to diagnose RA. These criteria include the number and distribution of joints with synovitis, the presence and levels of RA-related autoantibodies (e.g., anti-citrullinated protein antibodies, rheumatoid factor), acute phase reactants (e.g., C-reactive protein [CRP], erythrocyte sedimentation rate [ESR]), and the duration of symptoms. The etiology of RA currently remains unknown [4]. The disease emerges through a complex interplay of multiple factors, including genetic predisposition, immune system dysregulation, environmental influences, diverse biomarkers, and inflammatory cytokine activity [2, 3]. RA pathogenesis is characterized by dysregulation involving both innate and adaptive immune responses, leading to the secretion of cytokines, which promote persistent synovial inflammation, ultimately resulting in cartilage and bone erosion [3]. Enhanced knowledge of the disease’s pathogenesis has been instrumental in developing targeted therapies. Treatments for RA comprise the biologic disease-modifying anti-rheumatic drugs (bDMARDs), which include tumor necrosis factor (TNF) inhibitors (TNFi), including golimumab, a T-cell costimulatory inhibitor, interleukin-6 (IL-6) receptor inhibitors, an IL-1 inhibitor, and an anti-CD20 antibody [5]. The targeted synthetic DMARDs (tsDMARD), Janus kinase inhibitors (JAKi), are also approved for treatment. Collectively, these drug classes represent a significant advancement in disease management. The objective of this narrative review is to present efficacy/effectiveness and safety data associated with golimumab, a TNFi, for the treatment of RA. The review discusses the primary endpoints and safety data from the pivotal randomized controlled trials (RCTs). Additionally, data from the RCT post hoc analyses and real-world studies that evaluated disease activity, radiographic progression, pain, fatigue, physical function, HR-QoL, work productivity, and drug survival are presented. Safety data, including risk of infection, cardiovascular disease (CVD), immunogenicity, and cancer, are presented, as is use during pregnancy and lactation. The real-world and safety data were identified through structured searches of MEDLINE and Embase (2020–2025) and searches of PubMed for safety data specific for golimumab. This narrative review is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors; formal ethics approval was not required. Golimumab Golimumab (Janssen Biologics B.V., the Netherlands) is a fully human TNFi produced by a recombinant cell line that originated from genetically modified mice [6]. These mice were immunized with human TNF, leading to the development of antibodies that feature human variable and constant regions. Golimumab forms high-affinity, stable complexes with both the soluble and transmembrane bioactive forms of human TNF, which prevents the binding of TNF to its receptors, TNF receptor (TNFR) 1 and TNFR2 [6, 7]. While TNF-TNFR1 signaling primarily drives inflammation and apoptosis, TNF-TNFR2 signaling is mostly involved in T-cell migration and tissue repair [7]. A recent study sought to identify the molecular basis underlying the inhibitory mechanism of golimumab using X-ray crystallography. It was determined that golimumab binds to a unique epitope on TNF that is separate from the TNFR2 binding site. Golimumab inhibits TNF activity through steric hindrance, physically blocking both TNFR1 and TNFR2 from binding to TNF. Further, golimumab does not alter the conformation of TNF in ways that would impact receptor binding. This mechanism of action distinguishes golimumab from the four other TNFi approved for RA. By contrast, golimumab exhibits effects similar to other TNFi that contain constant region components (such as adalimumab, etanercept, and infliximab) [6]. These effects can include complement-dependent cytotoxicity (CDC), antibody-dependent cellular cytotoxicity (ADCC), reverse signaling, and cytokine suppression. Although etanercept binds to transmembrane TNF, it only induces ADCC, without CDC activities. By contrast, certolizumab induces a direct cytotoxic effect through a different mechanism of action related to the lack of constant region components [8]. In conclusion, although across the TNFi there are differences in TNF receptor binding and mechanisms related to cytotoxicity, i.e., differences in structure and function, the clinical significance of these is currently not known. Golimumab has subcutaneous (SC) (recommended to be administered every month) and intravenous (IV) (recommended to be administered at weeks 0 and 4, and then every 8 weeks) formulations and the availability of the two formulations varies across countries. Golimumab was initially approved in 2009 in Canada, the European Union, and the United States. Golimumab Pivotal Randomized Controlled Trials The efficacy and safety of golimumab were assessed in four randomized, double-blind, placebo-controlled trials. Three evaluated the SC formulation (GO-BEFORE [9], GO-FORWARD [10], and GO-AFTER [11]) in different patient populations with active RA. In GO-BEFORE, participants were methotrexate (MTX)-naïve. In GO-FORWARD and GO-AFTER, participants were previously treated with MTX or ≥ 1 dose of a TNFi (infliximab, etanercept, or adalimumab), respectively. In the fourth, GO-FURTHER, the IV formulation was evaluated in participants with active RA despite MTX treatment [12]. To assess the long-term effects of golimumab, extension trials (GO-BEFORE [13], GO-FORWARD [14–16], GO-AFTER [17, 18], and GO-FURTHER [19, 20]) were conducted. The treatment arms, primary endpoints (which varied across the RCTs), and key safety data are presented in Table 1. Additional study outcomes associated with each RCT are summarized below. Table 1Pivotal randomized controlled trials on efficacy and safety of golimumab in active rheumatoid arthritisRCT (clinical trial identifier number)Trial populationTreatment armsPrimary endpoint(s) and outcomesSafety outcomesEvaluation of subcutaneous GLM administered every 4 weeksGO-BEFORE 24-week trial (NCT00264537) [9] Adults (n = 637 randomized) with RA (ACR criteria: ≥ 3 months) who were MTX-naïve PBO + MTX (n = 160) 100 mg + PBO (n = 157) 50 mg + MTX (n = 158) 100 mg + MTX (n = 159)ACR50 at week 24: Difference in proportion meeting ACR50 was not significant for combined GLM (50 mg and 100 mg) + MTX (38.4%) vs. PBO + MTX (29.4%) (p = 0.053) Pairwise comparison was significant for 50 mg + MTX (40.5%) vs. PBO + MTX (29.4%) (p = 0.038), but not 100 mg + MTX (36.5%) (p = 0.177)Week 24: AEs/SAEs was similar: 72.5%/6.9% in PBO + MTX, 68.2%/3.2% in 100 mg + PBO, 81.6%/6.3% in 50 mg + MTX, and 76.1%/6.3% in 100 mg + MTX Antibodies to GLM developed in 13.5%, 3.7%, and 1.9% in the 100 mg + PBO, 50 mg + MTX, and 100 mg + MTX groups, respectively 2 deaths (1 with 50 mg + MTX, 1 with 100 mg + MTX)GO-BEFORE 5-year long-term extension (NCT00264537) [13]See aboveWeek 28: inadequate responders entered early escape from PBO + MTX → 50 mg + MTX; 100 mg + PBO → 100 mg + MTX; 50 mg + MTX → 100 mg + MTXWeek 52: remaining participants in PBO + MTX group crossed over to 50 mg + MTX groupACR20/50/70 at week 256: 74.8%, 58.2%, and 38.4% of participants treated with GLM had an ACR20/50/70 response, respectivelyWeek 268: 94.5% had ≥ 1 AE, 33.1% had ≥ 1 SAE, 9.6% developed antibodies to GLM 12 deaths occurred across the entire trial (6 with 50 mg + MTX, 6 with 100 mg + MTX or + PBO)GO-FORWARD 24-week trial (NCT00264550) [10]Adults with RA (n = 444 randomized) (ACR criteria: ≥ 3 months) on a stable MTX dose (15–25 mg/week) during the 4-week period immediately before screening PBO + MTX (n = 133) 100 mg + PBO (n = 133) 50 mg + MTX (n = 89) 100 mg + MTX (n = 89)Week 16: inadequate responders entered early escape from PBO + MTX → 50 mg + MTX; 100 mg + PBO → 100 mg + MTX; 50 mg + MTX → 100 mg + MTXACR20 at week 14 and HAQ-DI at week 24: Compared with PBO + MTX (33.1%), significantly more in the 50 mg + MTX (55.1%, p = 0.001) and 100 mg + MTX (56.2%, p < 0.001) achieved the ACR20, whereas 100 mg + PBO (44.4%, p = 0.059) did not Compared with PBO + MTX (– 0.13), median improvements from baseline HAQ-DI score were significantly greater for 50 mg + MTX (– 0.38, p < 0.001) and 100 mg + MTX (– 0.50, p < 0.001), but not 100 mg + PBO (– 0.13 p = 0.240)Week 16: ≥ 1 AE was 68.5% for 50 mg + MTX, 69.7% for 100 mg + MTX, 63.2% for 100 mg + PBO, and 60.9% for PBO + MTX ≥ 1 SAE was 5.6% for 50 mg + MTX, 9.0% for 100 mg + MTX, 3.8% for 100 mg + PBO, and 2.3% for PBO + MTXWeek 24: 2.1% developed antibodies to GLM and all were in the 100 mg + PBO group 1 death with 100 mg + PBOGO-FORWARD 5-year trial (NCT00264550) [16]See above Week 24: remaining patients in PBO + MTX group crossed over to 50 mg + MTX groupACR20/50/70 at week 256: 63.1%, 40.8%, and 24.1% of participants treated with GLM had an ACR20/50/70 response, respectivelyWeek 268: 93.1% had ≥ 1 AE, 39.6% had ≥ 1 SAE, and 7.5% developed antibodies to GLM 8 deaths occurred across the entire trial (2 with 50 mg or 100 mg + PBO or MTX, 6 with 100 mg + MTX)GO-AFTER 24-week trial (NCT00299546) [11]Adults (n = 461 randomized) with RA (ACR criteria: ≥ 3 months) who previously received ≥ 1 dose of an TNFi PBO (n = 155) 50 mg (n = 152) 100 mg (n = 152)Co-administered with stable doses of MTX, sulfasalazine, hydroxychloroquine, oral corticosteroids, or NSAIDs in most participantsACR20 at week 14: Compared with PBO (18%), significantly more participants in combined (37%, p < 0.0001) and each dose group (50 mg: 35%, p = 0.0006; 100 mg: 38%, p = 0.0001) achieved ACR20Week 16: ≥ 1 AE reported by 70% of PBO, 61% of 50 mg, and 73% of 100 mg ≥ 1 SAE reported by 7% of PBO, 5% of 50 mg, and 3% of 100 mgWeek 24: 3% developed antibodies to GLM 1 death with PBOGO-AFTER 5-years trial (NCT00299546) [18]n = 304 coadministered MTX, see aboveWeek 16: inadequate responders entered early escape from PBO → 50 mg and from 50 mg → 100 mg Week 24: remaining participants in PBO group crossed over to 50 mg groupACR20/50/70 at week 256: Among participants coadministered MTX, the proportions that achieved ACR responses were 37.6% to 47.0% for ACR20, 21.4–35.0% for ACR50, and 7.8–17.0% for ACR70Week 268: 94.7% had ≥ 1 AE, 35.0% had ≥ 1 SAE, and 8% developed antibodies to GLM 11 deaths occurred across the entire trial (1 with PBO, 10 with GLM)Evaluation of intravenous GLM administered every 8 weeksGO-FURTHER 24-week trial (NCT00973479) [12]Adults (n = 592 randomized) with RA for ≥ 3 months (≥ 6 swollen and tender joints, RF + and/or anticyclic citrullinated protein-positive, CRP ≥ 1.0 mg/dL) despite MTX for ≥ 4 weeks 2 mg/kg + MTX (n = 395) PBO + MTX (n = 197)Week 16: inadequate responders in the PBO + MTX group entered early escape to 2 mg/kg + MTXACR20 at week 14: Compared with PBO + MTX (24.9%), significantly more participants in the GLM + MTX group (58.5%, p < 0.001) achieved an ACR20 responseWeek 16: ≥ 1 AE reported by 43.7% of PBO + MTX and 47.3% of GLM + PBOWeek 24: 3% developed antibodies to GLM 1 death with PBO + MTXGO-FURTHER 112-week trial (NCT00973479) [20]See aboveWeek 24: remaining participants in the PBO + MTX crossed over to 2 mg/kg + MTXACR20/50/70 at week 100: 68.1%, 43.8%, and 23.5% of all participants treated with GLM had an ACR20/50/70 response, respectivelyWeek 100: 6.7% developed antibodies to GLMWeek 112: 79.1% had ≥ 1 AE and 18.2% had ≥ 1 SAE 6 deaths (1 death with PBO + MTX, 5 with 2 mg/kg + MTX)ACR American College of Rheumatology, ACR20/50/70 American College of Rheumatology 20%/50%/70% improvement, AE adverse event, CRP C-reactive protein, GLM golimumab, HAQ-DI Health Assessment Questionnaire-Disability Index, MTX methotrexate, NSAID non-steroidal anti-inflammatory drug, PBO placebo, RA rheumatoid arthritis, RCT randomized controlled trial, RF rheumatoid factor, SAE serious adverse event, TNFi tumor necrosis factor inhibitor, vs versus Table 1Pivotal randomized controlled trials on efficacy and safety of golimumab in active rheumatoid arthritisRCT (clinical trial identifier number)Trial populationTreatment armsPrimary endpoint(s) and outcomesSafety outcomesEvaluation of subcutaneous GLM administered every 4 weeksGO-BEFORE 24-week trial (NCT00264537) [9] Adults (n = 637 randomized) with RA (ACR criteria: ≥ 3 months) who were MTX-naïve PBO + MTX (n = 160) 100 mg + PBO (n = 157) 50 mg + MTX (n = 158) 100 mg + MTX (n = 159)ACR50 at week 24: Difference in proportion meeting ACR50 was not significant for combined GLM (50 mg and 100 mg) + MTX (38.4%) vs. PBO + MTX (29.4%) (p = 0.053) Pairwise comparison was significant for 50 mg + MTX (40.5%) vs. PBO + MTX (29.4%) (p = 0.038), but not 100 mg + MTX (36.5%) (p = 0.177)Week 24: AEs/SAEs was similar: 72.5%/6.9% in PBO + MTX, 68.2%/3.2% in 100 mg + PBO, 81.6%/6.3% in 50 mg + MTX, and 76.1%/6.3% in 100 mg + MTX Antibodies to GLM developed in 13.5%, 3.7%, and 1.9% in the 100 mg + PBO, 50 mg + MTX, and 100 mg + MTX groups, respectively 2 deaths (1 with 50 mg + MTX, 1 with 100 mg + MTX)GO-BEFORE 5-year long-term extension (NCT00264537) [13]See aboveWeek 28: inadequate responders entered early escape from PBO + MTX → 50 mg + MTX; 100 mg + PBO → 100 mg + MTX; 50 mg + MTX → 100 mg + MTXWeek 52: remaining participants in PBO + MTX group crossed over to 50 mg + MTX groupACR20/50/70 at week 256: 74.8%, 58.2%, and 38.4% of participants treated with GLM had an ACR20/50/70 response, respectivelyWeek 268: 94.5% had ≥ 1 AE, 33.1% had ≥ 1 SAE, 9.6% developed antibodies to GLM 12 deaths occurred across the entire trial (6 with 50 mg + MTX, 6 with 100 mg + MTX or + PBO)GO-FORWARD 24-week trial (NCT00264550) [10]Adults with RA (n = 444 randomized) (ACR criteria: ≥ 3 months) on a stable MTX dose (15–25 mg/week) during the 4-week period immediately before screening PBO + MTX (n = 133) 100 mg + PBO (n = 133) 50 mg + MTX (n = 89) 100 mg + MTX (n = 89)Week 16: inadequate responders entered early escape from PBO + MTX → 50 mg + MTX; 100 mg + PBO → 100 mg + MTX; 50 mg + MTX → 100 mg + MTXACR20 at week 14 and HAQ-DI at week 24: Compared with PBO + MTX (33.1%), significantly more in the 50 mg + MTX (55.1%, p = 0.001) and 100 mg + MTX (56.2%, p < 0.001) achieved the ACR20, whereas 100 mg + PBO (44.4%, p = 0.059) did not Compared with PBO + MTX (– 0.13), median improvements from baseline HAQ-DI score were significantly greater for 50 mg + MTX (– 0.38, p < 0.001) and 100 mg + MTX (– 0.50, p < 0.001), but not 100 mg + PBO (– 0.13 p = 0.240)Week 16: ≥ 1 AE was 68.5% for 50 mg + MTX, 69.7% for 100 mg + MTX, 63.2% for 100 mg + PBO, and 60.9% for PBO + MTX ≥ 1 SAE was 5.6% for 50 mg + MTX, 9.0% for 100 mg + MTX, 3.8% for 100 mg + PBO, and 2.3% for PBO + MTXWeek 24: 2.1% developed antibodies to GLM and all were in the 100 mg + PBO group 1 death with 100 mg + PBOGO-FORWARD 5-year trial (NCT00264550) [16]See above Week 24: remaining patients in PBO + MTX group crossed over to 50 mg + MTX groupACR20/50/70 at week 256: 63.1%, 40.8%, and 24.1% of participants treated with GLM had an ACR20/50/70 response, respectivelyWeek 268: 93.1% had ≥ 1 AE, 39.6% had ≥ 1 SAE, and 7.5% developed antibodies to GLM 8 deaths occurred across the entire trial (2 with 50 mg or 100 mg + PBO or MTX, 6 with 100 mg + MTX)GO-AFTER 24-week trial (NCT00299546) [11]Adults (n = 461 randomized) with RA (ACR criteria: ≥ 3 months) who previously received ≥ 1 dose of an TNFi PBO (n = 155) 50 mg (n = 152) 100 mg (n = 152)Co-administered with stable doses of MTX, sulfasalazine, hydroxychloroquine, oral corticosteroids, or NSAIDs in most participantsACR20 at week 14: Compared with PBO (18%), significantly more participants in combined (37%, p < 0.0001) and each dose group (50 mg: 35%, p = 0.0006; 100 mg: 38%, p = 0.0001) achieved ACR20Week 16: ≥ 1 AE reported by 70% of PBO, 61% of 50 mg, and 73% of 100 mg ≥ 1 SAE reported by 7% of PBO, 5% of 50 mg, and 3% of 100 mgWeek 24: 3% developed antibodies to GLM 1 death with PBOGO-AFTER 5-years trial (NCT00299546) [18]n = 304 coadministered MTX, see aboveWeek 16: inadequate responders entered early escape from PBO → 50 mg and from 50 mg → 100 mg Week 24: remaining participants in PBO group crossed over to 50 mg groupACR20/50/70 at week 256: Among participants coadministered MTX, the proportions that achieved ACR responses were 37.6% to 47.0% for ACR20, 21.4–35.0% for ACR50, and 7.8–17.0% for ACR70Week 268: 94.7% had ≥ 1 AE, 35.0% had ≥ 1 SAE, and 8% developed antibodies to GLM 11 deaths occurred across the entire trial (1 with PBO, 10 with GLM)Evaluation of intravenous GLM administered every 8 weeksGO-FURTHER 24-week trial (NCT00973479) [12]Adults (n = 592 randomized) with RA for ≥ 3 months (≥ 6 swollen and tender joints, RF + and/or anticyclic citrullinated protein-positive, CRP ≥ 1.0 mg/dL) despite MTX for ≥ 4 weeks 2 mg/kg + MTX (n = 395) PBO + MTX (n = 197)Week 16: inadequate responders in the PBO + MTX group entered early escape to 2 mg/kg + MTXACR20 at week 14: Compared with PBO + MTX (24.9%), significantly more participants in the GLM + MTX group (58.5%, p < 0.001) achieved an ACR20 responseWeek 16: ≥ 1 AE reported by 43.7% of PBO + MTX and 47.3% of GLM + PBOWeek 24: 3% developed antibodies to GLM 1 death with PBO + MTXGO-FURTHER 112-week trial (NCT00973479) [20]See aboveWeek 24: remaining participants in the PBO + MTX crossed over to 2 mg/kg + MTXACR20/50/70 at week 100: 68.1%, 43.8%, and 23.5% of all participants treated with GLM had an ACR20/50/70 response, respectivelyWeek 100: 6.7% developed antibodies to GLMWeek 112: 79.1% had ≥ 1 AE and 18.2% had ≥ 1 SAE 6 deaths (1 death with PBO + MTX, 5 with 2 mg/kg + MTX)ACR American College of Rheumatology, ACR20/50/70 American College of Rheumatology 20%/50%/70% improvement, AE adverse event, CRP C-reactive protein, GLM golimumab, HAQ-DI Health Assessment Questionnaire-Disability Index, MTX methotrexate, NSAID non-steroidal anti-inflammatory drug, PBO placebo, RA rheumatoid arthritis, RCT randomized controlled trial, RF rheumatoid factor, SAE serious adverse event, TNFi tumor necrosis factor inhibitor, vs versus Pivotal randomized controlled trials on efficacy and safety of golimumab in active rheumatoid arthritis PBO + MTX (n = 160) 100 mg + PBO (n = 157) 50 mg + MTX (n = 158) 100 mg + MTX (n = 159) ACR50 at week 24: Difference in proportion meeting ACR50 was not significant for combined GLM (50 mg and 100 mg) + MTX (38.4%) vs. PBO + MTX (29.4%) (p = 0.053) Pairwise comparison was significant for 50 mg + MTX (40.5%) vs. PBO + MTX (29.4%) (p = 0.038), but not 100 mg + MTX (36.5%) (p = 0.177) Week 24: AEs/SAEs was similar: 72.5%/6.9% in PBO + MTX, 68.2%/3.2% in 100 mg + PBO, 81.6%/6.3% in 50 mg + MTX, and 76.1%/6.3% in 100 mg + MTX Antibodies to GLM developed in 13.5%, 3.7%, and 1.9% in the 100 mg + PBO, 50 mg + MTX, and 100 mg + MTX groups, respectively 2 deaths (1 with 50 mg + MTX, 1 with 100 mg + MTX) Week 28: inadequate responders entered early escape from PBO + MTX → 50 mg + MTX; 100 mg + PBO → 100 mg + MTX; 50 mg + MTX → 100 mg + MTX Week 52: remaining participants in PBO + MTX group crossed over to 50 mg + MTX group ACR20/50/70 at week 256: 74.8%, 58.2%, and 38.4% of participants treated with GLM had an ACR20/50/70 response, respectively Week 268: 94.5% had ≥ 1 AE, 33.1% had ≥ 1 SAE, 9.6% developed antibodies to GLM 12 deaths occurred across the entire trial (6 with 50 mg + MTX, 6 with 100 mg + MTX or + PBO) PBO + MTX (n = 133) 100 mg + PBO (n = 133) 50 mg + MTX (n = 89) 100 mg + MTX (n = 89) Week 16: inadequate responders entered early escape from PBO + MTX → 50 mg + MTX; 100 mg + PBO → 100 mg + MTX; 50 mg + MTX → 100 mg + MTX ACR20 at week 14 and HAQ-DI at week 24: Compared with PBO + MTX (33.1%), significantly more in the 50 mg + MTX (55.1%, p = 0.001) and 100 mg + MTX (56.2%, p < 0.001) achieved the ACR20, whereas 100 mg + PBO (44.4%, p = 0.059) did not Compared with PBO + MTX (– 0.13), median improvements from baseline HAQ-DI score were significantly greater for 50 mg + MTX (– 0.38, p < 0.001) and 100 mg + MTX (– 0.50, p < 0.001), but not 100 mg + PBO (– 0.13 p = 0.240) Week 16: ≥ 1 AE was 68.5% for 50 mg + MTX, 69.7% for 100 mg + MTX, 63.2% for 100 mg + PBO, and 60.9% for PBO + MTX ≥ 1 SAE was 5.6% for 50 mg + MTX, 9.0% for 100 mg + MTX, 3.8% for 100 mg + PBO, and 2.3% for PBO + MTX Week 24: 2.1% developed antibodies to GLM and all were in the 100 mg + PBO group 1 death with 100 mg + PBO ACR20/50/70 at week 256: 63.1%, 40.8%, and 24.1% of participants treated with GLM had an ACR20/50/70 response, respectively Week 268: 93.1% had ≥ 1 AE, 39.6% had ≥ 1 SAE, and 7.5% developed antibodies to GLM 8 deaths occurred across the entire trial (2 with 50 mg or 100 mg + PBO or MTX, 6 with 100 mg + MTX) PBO (n = 155) 50 mg (n = 152) 100 mg (n = 152) Co-administered with stable doses of MTX, sulfasalazine, hydroxychloroquine, oral corticosteroids, or NSAIDs in most participants ACR20 at week 14: Compared with PBO (18%), significantly more participants in combined (37%, p < 0.0001) and each dose group (50 mg: 35%, p = 0.0006; 100 mg: 38%, p = 0.0001) achieved ACR20 Week 16: ≥ 1 AE reported by 70% of PBO, 61% of 50 mg, and 73% of 100 mg ≥ 1 SAE reported by 7% of PBO, 5% of 50 mg, and 3% of 100 mg Week 24: 3% developed antibodies to GLM 1 death with PBO Week 16: inadequate responders entered early escape from PBO → 50 mg and from 50 mg → 100 mg Week 24: remaining participants in PBO group crossed over to 50 mg group ACR20/50/70 at week 256: Among participants coadministered MTX, the proportions that achieved ACR responses were 37.6% to 47.0% for ACR20, 21.4–35.0% for ACR50, and 7.8–17.0% for ACR70 Week 268: 94.7% had ≥ 1 AE, 35.0% had ≥ 1 SAE, and 8% developed antibodies to GLM 11 deaths occurred across the entire trial (1 with PBO, 10 with GLM) 2 mg/kg + MTX (n = 395) PBO + MTX (n = 197) Week 16: inadequate responders in the PBO + MTX group entered early escape to 2 mg/kg + MTX ACR20 at week 14: Compared with PBO + MTX (24.9%), significantly more participants in the GLM + MTX group (58.5%, p < 0.001) achieved an ACR20 response Week 16: ≥ 1 AE reported by 43.7% of PBO + MTX and 47.3% of GLM + PBO Week 24: 3% developed antibodies to GLM 1 death with PBO + MTX ACR20/50/70 at week 100: 68.1%, 43.8%, and 23.5% of all participants treated with GLM had an ACR20/50/70 response, respectively Week 100: 6.7% developed antibodies to GLM Week 112: 79.1% had ≥ 1 AE and 18.2% had ≥ 1 SAE 6 deaths (1 death with PBO + MTX, 5 with 2 mg/kg + MTX) ACR American College of Rheumatology, ACR20/50/70 American College of Rheumatology 20%/50%/70% improvement, AE adverse event, CRP C-reactive protein, GLM golimumab, HAQ-DI Health Assessment Questionnaire-Disability Index, MTX methotrexate, NSAID non-steroidal anti-inflammatory drug, PBO placebo, RA rheumatoid arthritis, RCT randomized controlled trial, RF rheumatoid factor, SAE serious adverse event, TNFi tumor necrosis factor inhibitor, vs versus In GO-BEFORE, all efficacy outcomes were compared across SC 50 mg + MTX, 100 mg + MTX, 100 mg + placebo, and MTX + placebo groups at week 24 (Table 1) [9]. The primary endpoint, the proportion that achieved an ACR ≥ 50% improvement (ACR50), required a significant difference between a combined group (50 mg and 100 mg + MTX) versus placebo + MTX and the individual dose groups (50 mg + MTX and 100 mg + MTX) versus placebo + MTX. The first comparison between the combined group versus MTX + placebo was not significant (p = 0.053). The pairwise comparison between MTX + placebo versus 50 mg + MTX was significant (p = 0.042), but it was not for 100 mg + MTX (p = 0.177). Secondary endpoint analyses found that significantly more participants in the 50 mg + MTX and 100 mg + MTX groups (both 61.6%) achieved an ACR20 response versus placebo + MTX (49.4%) (both: p = 0.028). A significant improvement was also observed in the combined group versus placebo + MTX for the ACR20 (p = 0.011), ACR90 (p = 0.027), and Disease Activity Score in 28 joints (DAS28) remission using CRP (p = 0.031) or ESR (p = 0.031). At week 24, participants who discontinued treatment included ten from placebo + MTX, nine from 100 mg + placebo, eight from 50 mg + MTX, and ten from 100 mg + MTX. Adverse events (AE) were reported in 72.5%, 68.2%, 81.6%, and 76.1% of the placebo + MTX, 100 mg + placebo, 50 mg + MTX, and 100 mg + MTX groups, respectively. Serious AEs (SAE) were reported in 6.9%, 3.2%, 6.3%, and 6.3%, respectively. Across the golimumab groups, the most frequently reported AEs were nausea, upper respiratory infection (URI), increased aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels, dyspepsia, and headache. Nausea, URI, ALT increased, and headache were the most frequently reported AEs by the placebo + MTX group. The percentage of each group that developed antibodies to golimumab were 13.5%, 3.7%, and 1.9% in the 100 mg + placebo, 50 mg + MTX, and 100 mg + MTX groups, respectively. In conclusion, although the primary endpoint was not significantly met (see Table 1), secondary endpoints demonstrated that golimumab + MTX was more efficacious than MTX or golimumab monotherapy. An extension trial evaluated efficacy and safety through 5 years [13]. At week 256, 74.8%, 58.2%, and 38.4% of participants treated with golimumab (n = 318) had an ACR20/50/70 response, respectively, and 84.9% had a good/moderate DAS28-CRP response, demonstrating that efficacy was maintained. Regarding safety outcomes, 94.5% and 33.1% of the golimumab-treated participants reported ≥ 1 AE and ≥ 1 SAE, respectively, and there were no unexpected safety concerns. The most frequently reported AEs across all groups were URI, nausea, bronchitis, and increased ALT. Antibodies to golimumab developed in 9.6% of participants. Lastly, at completion of the GO-BEFORE extension trial, at week 252, drug survival across the golimumab treatment groups remained robust (50 mg + MTX: 69.0%, 100 mg + MTX: 62.3%, and placebo → golimumab + MTX [50 or 100 mg]: 68.8%). In GO-FORWARD [10], participants were randomized to the same four groups as in GO-BEFORE (Table 1) [9]. The co-primary endpoints were week 14 ACR20 response and change from baseline to week 24 in the Health Assessment Questionnaire-Disability Index (HAQ-DI) score. Compared with placebo + MTX (33.1%), significantly more participants in the 50 mg + MTX (55.1%, p = 0.001) and 100 mg + MTX (56.2%, p < 0.001) groups achieved an ACR20 response. Median improvement in HAQ-DI score was significantly greater in the 50 mg + MTX (− 0.38, p < 0.001) and 100 mg + MTX (− 0.50, p < 0.001) groups versus placebo + MTX (− 0.13). Secondary endpoint analyses showed that at weeks 14 and 24, significantly more participants in the golimumab + MTX groups experienced remission per the DAS28-ESR versus placebo + MTX (week 14: 50 mg + MTX: 15.7% p < 0.001, 100 mg + MTX: 18.0% p < 0.001, placebo + MTX: 1.5%; week 24: 50 mg + MTX: 20.2% p = 0.001, 100 mg + MTX: 22.5% p < 0.001; placebo + MTX: 6.0%). At week 14, participants who discontinued treatment included six from placebo + MTX, three from 100 mg + placebo, two from 50 mg + MTX, and four from 100 mg + MTX. Regarding safety outcomes, at week 16, the percentage of participants who reported ≥ 1 AE was 68.5%, 69.7%, 63.2%, and 60.9% for the 50 mg + MTX, 100 mg + MTX, 100 mg + placebo, and placebo + MTX, respectively. The percentage that reported ≥ 1 SAE was 5.6%, 9.0%, 3.8%, and 2.3%, respectively. At week 24, 2.1% of participants developed antibodies to golimumab and all were in the 100 mg + placebo group. Long-term efficacy and safety were evaluated in three GO-FORWARD extension trials [14–16]. In the 5-year extension trial, 63.1%, 40.8%, and 24.1% of participants (n = 444) treated with golimumab had an ACR20/50/70 response, respectively, and 78.2% had a DAS28-CRP response, demonstrating that efficacy was maintained [16]. Regarding safety outcomes, 93.1% and 39.6% of the golimumab-treated participants reported ≥ 1 AE and ≥ 1 SAE, respectively, and antibodies to golimumab developed in 7.5%. The most frequently reported AEs were URI, bronchitis, nasopharyngitis, cough, and worsening of RA. Lastly, at completion of the extension trial, at week 252, drug survival across the golimumab treatment groups remained robust (50 mg + MTX: 80.9%, 100 mg + MTX: 66.3%, and placebo to golimumab [50 or 100 mg] + MTX: 67.7%). In GO-AFTER, participants who discontinued previous TNFi treatment were randomized to receive SC 50 mg, 100 mg, or placebo (Table 1) [11]. In addition, concomitant therapy with stable doses of MTX, sulfasalazine, hydroxychloroquine, oral corticosteroids, or non-steroidal anti-inflammatory drugs (NSAIDs) were allowed. At week 14, significantly more participants in the combined golimumab group (37%, p < 0.0001) and in each dose group (50 mg: 35%, p = 0.0006; 100 mg: 38%, p = 0.0001) achieved an ACR20 (the primary endpoint) versus placebo (18%). Secondary endpoint analyses determined that significantly more participants in the combined and dose groups experienced remission per DAS28 (combined: 10%, p = 0.0001; 50 mg: 8%, p = 0.0009; 100 mg: 12%, p < 0.0001) versus placebo (1%). At week 24, participants who discontinued treatment included 28 from placebo, three from placebo → 50 mg, nine from 50 mg, three from 50 mg → 100 mg, and 14 from 100 mg. Regarding safety outcomes, at week 16, the percentage of participants who reported ≥ 1 AE was similar (placebo: 70%, 50 mg: 61%, and 100 mg: 73%). The percentage of participants who reported ≥ 1 SAE was also similar (placebo: 7%, 50 mg: 5%, 100 mg: 3%). Antibodies to golimumab developed in 3% of participants through week 24. The most frequently reported AEs in both the placebo and golimumab groups were URI, nasopharyngitis, and diarrhea. Long-term efficacy and safety were evaluated in two GO-AFTER extension trials [17, 18]. At week 256, a total of 304 participants received golimumab + MTX and among these, the proportions that achieved ACR responses were 37.6–47.0% for ACR20, 21.4–35.0% for ACR50, and 7.8–17.0% for ACR70, demonstrating that efficacy was maintained [18]. Regarding safety outcomes, through week 268, 94.7% and 35.0% of the golimumab-treated participants reported ≥ 1 AE and ≥ 1 SAE, respectively and antibodies to golimumab developed in 8.0%. The most frequently reported AEs were URI, worsening of RA, nasopharyngitis, sinusitis, back pain, hypertension, arthralgia, bronchitis, diarrhea, urinary tract infection (UTI), nausea, headache, and cough. Regarding drug survival, among participants treated at baseline with MTX and golimumab (n = 311), 41.8% continued treatment through week 252. Among participants who received golimumab monotherapy (n = 148), 35.8% continued treatment through week 252. In GO-FURTHER, participants were randomized to receive IV 2 mg/kg + MTX or placebo + MTX and all were TNFi-naïve [12]. The primary endpoint was the proportion achieving an ACR20 at week 14. Compared with placebo + MTX (25%), significantly more in the 2 mg/kg + MTX group (58.5%, p < 0.001) achieved an ACR20 response. Secondary endpoint analyses determined that there was a significant difference in ACR20 at week 2 in the 2 mg/kg + MTX (33.2%, p < 0.001) group versus placebo + MTX (11.7%). Additionally, significantly more of the 2 mg/kg + MTX group (81.3%, p < 0.001) achieved a moderate/good DAS28-CRP score versus placebo + MTX (40.1%) at week 14 and also at the earlier timepoint of week 2 (2 mg/kg + MTX: 65.1%, placebo + MTX: 19.3%; p < 0.001). Lastly, the 2 mg/kg + MTX group demonstrated a significantly greater mean improvement in the HAQ-DI score at week 14 (0.50, p < 0.001) versus placebo + MTX (0.19). At week 24, participants who discontinued treatment included six from placebo + MTX, 0 from placebo + MTX → 2 mg/kg + MTX, and 16 from 2 mg/kg + MTX. Regarding safety outcomes, at week 16, similar proportions of the placebo + MTX and 2 mg/kg + MTX groups reported AEs (43.7% and 47.3%, respectively). Antibodies to golimumab developed in 3.0% of participants through week 24. The most frequently reported AEs were infections and infestations, including URI, UTI, and nasopharyngitis. In the extension trial, at week 100, 68.1%, 43.8%, and 23.5% of all participants treated with golimumab (n = 486) had an ACR20/50/70 response, respectively, demonstrating that efficacy was maintained [20]. Regarding safety outcomes at week 112, among all participants treated with golimumab, 79.1% had ≥ 1 AE and 18.2% had ≥ 1 SAE. Antibodies developed in 6.7% of golimumab-treated participants through week 100. The most frequently reported AEs were URI, bronchitis, worsening of RA, hypertension, and nasopharyngitis. Lastly, at completion of the GO-FURTHER extension trial, at week 100, drug survival across the golimumab treatment groups remained robust (2 mg/kg + MTX: 82.5% and placebo → golimumab + MTX: 81.2%). Collectively, the four pivotal RCTs demonstrated that golimumab, in participants with active RA, significantly reduced the signs and symptoms of the disease and the extension trials revealed that the beneficial clinical responses were maintained. The safety data showed that a majority of participants reported at least one AE and the most frequent AEs were URI, nasopharyngitis, and bronchitis. Within the pivotal and extension trials, a low percentage of participants developed antibodies against golimumab (range 1.9–13.5%, both from GO-BEFORE). At the completion of all the extension trials (i.e., > 50% of participants completed treatment), except for GO-AFTER, drug survival was found to be robust. A post hoc analysis of pooled data evaluated drug survival from three of the extension trials [21]. First-line treatment with golimumab from GO-BEFORE and GO-FORWARD (n = 1050) was associated with higher drug survival (69.0%, 95% confidence interval [CI] 66.1–71.8%) at year 5. By contrast, drug survival associated with second-line treatment from GO-AFTER (n = 431) was significantly lower (41.6%, 95% CI 36.8–46.3%) (p < 0.0001) at year 5 versus the pooled first-line data. The Effect of Golimumab on Radiographic Progression and MRI Inflammation Effectively treating RA requires not only alleviating signs and symptoms but also preventing the progression of joint damage. A post hoc analysis of GO-BEFORE and GO-FORWARD data evaluated the effect of golimumab (50 mg or 100 mg) + MTX on radiographic progression versus placebo + MTX [22]. At baseline, in GO-BEFORE, the mean (standard deviation [SD]) modified Sharp/van der Heijde Score (SHS) across treatment arms was placebo + MTX: 19.7 (35.4), 100 mg + placebo: 20.4 (30.9), 50 mg + MTX: 18.7 (32.4), and 100 mg + MTX: 18.2 (35.5). At week 52, golimumab was associated with a significant change from baseline in the mean (SD) modified SHS in the 50 mg + MTX: 0.74 (5.23) (p = 0.015) and 100 mg + MTX: 0.07 (1.83) (p = 0.025) groups versus placebo + MTX: 1.37 (4.56). At baseline, in GO-FORWARD, the mean (SD) modified SHS across treatment arms was placebo + MTX: 36.7 (52.1), 100 mg + placebo: 37.4 (52.5), 50 mg + MTX: 29.7 (39.3), and 100 mg + MTX: 39.6 (56.1). At week 24, golimumab treatment was not associated with a significant change from baseline versus placebo + MTX (50 mg + MTX: 0.60 [2.74], p = 0.953; 100 mg + MTX: 0.23 [1.34], p = 0.293; placebo + MTX: 0.55 [2.35]). Three extension trials evaluated the long-term effect of golimumab on radiographic progression using the SHS. In GO-BEFORE, at week 256, golimumab was associated with minimal radiographic progression, as the overall mean (SD) change from baseline in SHS was 1.36 (5.34) [13]. Furthermore, approximately 60% of all participants had a change from baseline in SHS ≤ 0 and 73% had a change of ≤ 0.5. Similarly in GO-FORWARD, at week 256, the mean annual rate of progression per the SHS was 0.65 for all participants and 50.9% had a change from baseline in total SHS of ≤ 0 and 60.1% had a change of ≤ 0.5 [16]. In GO-FURTHER, the mean change from baseline to week 100 in total SHS was significantly reduced in the 2 mg/kg + MTX group (0.74) versus placebo + MTX (2.10, p = 0.005) [20]. At week 100, 59.5% of all participants had a change from baseline in total SHS ≤ 0. Collectively, extension trial data, up to 5 years, demonstrated that treatment with golimumab was associated with minimal radiographic progression. A post hoc analysis using GO-FORWARD data evaluated golimumab treatment on MRI-detected inflammation and structural damage in participants with active RA despite MTX [23]. MRI was performed on the dominant wrist and metacarpophalangeal (MCP) joints. All scans were evaluated using the Rheumatoid Arthritis MRI Scoring (RAMRIS) system. For synovitis, two scoring approaches were employed: wrist-only assessment (scale 0–9) and combined wrist + MCP assessment (scale 0–21). Additionally, osteitis was scored on a scale of 0–69, and bone erosions were evaluated on a scale of 0–230. There were significant improvements in synovitis and osteitis in the combined (50 mg + 100 mg) + MTX group versus placebo + MTX at week 12 (wrist + MCP: − 1.77 versus − 0.15, p < 0.001; osteitis: − 2.00 versus 0.19, p = 0.003) and week 24 (wrist + MCP: − 1.91 versus − 0.38, p < 0.001; osteitis: − 1.74 versus 0.71, p = 0.004). Assessment of the effect of golimumab on bone erosions was limited as < 10% of each group exhibited substantial erosive progression, with most participants showing no progression. In conclusion, at both weeks 12 and 24, golimumab + MTX significantly improved synovitis and osteitis versus placebo + MTX. Of note, these inflammatory markers are known predictors of future structural joint damage [23]. The Effect of Golimumab on HR-QoL, Fatigue, Productivity, and Relationship with Disease Activity Two post hoc analyses of data from GO-FORWARD and GO-FURTHER evaluated the effect of golimumab on HR-QoL, productivity, and fatigue, and the relationship with disease activity. The analysis of GO-FORWARD data evaluated physical function, general health, and fatigue in participants with active RA despite MTX therapy [24]. Through week 52, physical function (HAQ-DI) and health status (Short Form-36 [SF-36] physical and mental component summary scores [PCS, MCS]) were evaluated. Fatigue was assessed through week 24 using the Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F). At week 14, the mean (SD) improvement in HAQ-DI was significantly greater in the 50 mg + MTX group (0.42 [0.50]) and 100 mg + MTX group (0.41 [0.46]) versus placebo + MTX group (0.16 [0.49]; both p < 0.001) and this was maintained at weeks 24 and 52. Regarding current health state, there were significant improvements from baseline in the mean (SD) SF-36 PCS score in both dose groups versus placebo at week 14 (50 mg + MTX: 8.02 [7.17], 100 mg + MTX: 7.41 [8.04], placebo + MTX: 2.39 [7.80]; both p < 0.001) and this was maintained at weeks 24 and 52. Regarding the SF-36 MCS score, at weeks 14 and 24, only the 100 mg + MTX group was significantly different from placebo + MTX (4.62 [10.17] versus 1.63 [9.81], p = 0.039; 4.33 [10.66] versus 0.75 [9.68], p = 0.014). Regarding fatigue, at week 14 there were significantly greater mean (SD) improvements from baseline in FACIT-F scores in the 50 mg + MTX (7.58 [8.93]; p < 0.001) and 100 mg + MTX (6.40 [9.62]; p = 0.002) versus placebo + MTX (2.28 [9.23]), which were maintained at week 24. Similarly, in GO-FURTHER, the effect of golimumab on the same measures of physical function, general health, and fatigue was evaluated in the same patient population [25]. This analysis also analyzed the effect of golimumab on daily productivity at work, school, or home (visual analog scale [VAS]), patient function and HR-QoL (EuroQol-5 dimensions [EQ-5D] assessment of current health state), and disease activity (DAS28-CRP). At weeks 12, 14, 16, and 24, there were significant improvements in HAQ-DI scores from baseline in the 2 mg/kg + MTX group versus placebo + MTX (all p < 0.001). Regarding HR-QoL, at weeks 12, 16, and 24, the 2 mg/kg + MTX group showed significant improvements across all 8 SF-36 subscores and in both the PCS and MCS scores (all p < 0.001). The 2 mg/kg + MTX group exhibited improved EQ-5D, EQ-5D VAS, and FACIT-F scores at weeks 12, 16, and 24 (all p < 0.001), with a higher proportion achieving clinically meaningful improvements in these measures. Regarding productivity, at weeks 12, 16, and 24, the 2 mg/kg + MTX group experienced significant improvement in the VAS versus placebo + MTX (all p < 0.001). Regarding disease activity, a significant proportion of the 2 mg/kg + MTX group achieved a DAS28-CRP < 2.6 versus placebo + MTX at week 12 (9.4% versus 3.0%, p = 0.005), week 16 (12.9% versus 4.6%, p = 0.002), and week 24 (17.7% versus 5.1%; p < 0.001). Lastly, improvements in the DAS28-CRP significantly correlated with improvements in the SF-36 PCS and MCS, the EQ-5D, the FACIT-F, and productivity VAS scores at weeks 12, 16, and 24 (all p < 0.0001). These two post hoc analyses demonstrated that in participants with active RA, treatment with golimumab resulted in a significant improvement in physical function, HR-QoL, productivity, and fatigue and further, these improvements correlated with a decrease in disease activity per the DAS28-CRP. Golimumab Real-World Studies Effectiveness and Safety Real-world studies have demonstrated the effectiveness and safety of golimumab for RA in routine clinical care. A prospective observational study using Canadian registry (2002–2017) data examined clinical outcomes with SC and IV golimumab in patients with RA who were bDMARD-naïve or had ≤ 1 bDMARD exposure [26]. Patients treated with golimumab (n = 687) had significantly reduced pain and disease activity (DAS28) and improved function (HAQ-DI) over time (both p < 0.001), and the proportion of patients in remission (Simple Disease Activity Index ≤ 3.3) at 36 months reached 52.7% with SC golimumab and 61.9% with IV golimumab. Physician and patient global assessment scores were also significantly improved (both p < 0.01 for combined formulations). The most frequently reported AE was drug ineffective (11.9% SC golimumab and 14.6% IV golimumab). Across system organ classes, infections and infestations were most often reported (32.6% SC golimumab, 22.9% IV golimumab), with the most common AEs being URI (8.5%) with SC golimumab and sinusitis and UTI (3.8% each) with IV golimumab. Serious AEs occurred in 15.5% and 3.8% treated with SC and IV golimumab, respectively. The most common serious infection was pneumonia (0.9% SC golimumab, 0.6% IV golimumab) and no patients experienced an opportunistic infection. The incidence of malignancy was 2.41 per 100 patient-years (PY). The types of AEs and SAEs reported in this study were similar to those reported in the pivotal RCTs. Additional real-world studies, which found decreases in disease activity and improvements in physical function and fatigue with up to 2 years of golimumab treatment in bDMARD-naïve and -experienced participants, are presented in Table 2. Table 2Real-world studies on the effectiveness and safety of golimumab in rheumatoid arthritisReal-world study/study design/countryPatient population/duration of follow-upEffectiveness resultsSafety resultsDisease activityPhysical function and fatigueProspective studiesGO-BEYOND/pooled analysis of prospective, observational studies/Europe [27]n = 325 with previous TNFi, mean age (SD): 57.9 (12.6) years/12-mos follow-upDAS28-CRP: low-disease activity at 3, 6, and 12 mos: 43.1%, 58.3%, 63.6%; remission at 3, 6, and 12 mos: 27.7%, 40.8%, 47.5%N/A30.5% reported an AEProspective open-label, post-marketing surveillance study/India [28]n = 120 with failed first-line DMARDs; mean age (SD): 46.3 (13.07) years/6-mos follow-upDAS28-ESR mean (SD) change from baseline (n = 62): − 2.0 (1.25). DAS28-ESR remission: 14.5%; low-disease activity: 4.8%; moderate disease activity: 67.7%; high disease activity: 12.9%HAQ-DI mean (SD) total scores change from baseline: − 0.09 (0.65)FACIT-F mean (SD) score change from baseline: 14.8 (10.20)≥ 1 AE: 51.7%; SAEs: 3.3%GO-Q/Observational, prospective cohort study/Greece [29]n = 145; mean (SD) age: 54.6 (12.2) years/12-mos follow-upDAS28-ESR mean (SD) scores at baseline and 3, 6, and 12 mos: 4.1 (1.2), 3.7 (1.2), 3.2 (1.1), and 5.4 (0.9) (all p < 0.0001)DAS28-ESR remission at 3, 6, and 12 mos: 13.3%,18.9%, and 27.3%, low-disease activity: 24.4%, 35.6%, and 54.6%; treatment response (DAS28-ESR reduction ≥ 1.2): 49.6%, 57.6%, and 76.4%HAQ-DI mean (SD) scores: 1.45 (0.61) at baseline, 0.88 (0.67) at 3 mos, 0.71 (0.57) at 6 mos, and 0.48 (0.46) at 12 mosMean (SD) changes at 3, 6, and 12 mos: − 0.58 (0.53), − 0.75 (0.62), and − 1.04 (0.73) (all p < 0.0001)N/AAWARE/prospective, observational study/United States [30]n = 685; mean age (SD) years: 60.9 (13.4); bDMARD-naïve: 35.3%/52-week follow-upbDMARD-naïve: CDAI mean change from baseline at 6 mos was − 10.7; at 12 mos: − 11.0bDMARD-experienced: CDAI mean (SD) change baseline at 6 mos was − 10.3 (18.8); at 12 mos: − 11.0 (18.4)N/A≥ 1 AE: 51%; ≥ 1 SAE: 10%; most common AE: infection. Cancer: n = 13, including melanoma, NMSC, and leukemiaOpen-label, observational, post-marketing surveillance study/Japan [31, 32]n = 5154 started on GLM (50 mg or 100 mg) with or without MTX (56% bDMARD-experienced); mean (SD) age: 62.7 (13.2) years/24-week follow-upDAS28-ESR mean (SD) scores decreased from 4.72 (1.37) to 3.48 (1.43)DAS28-ESR low/remitted disease increased from 13.69% to 46.21%Rates of remission/low-disease activity were higher with GLM + MTX (50 mg + MTX: 49.74%, 100 mg + MTX: 46.28%) vs. GLM monotherapy (50 mg: 34.31%, 100 mg: 34.04%) and with no prior bDMARD (56.23% bDMARD-naïve vs. 37.93% bDMARD-exposed)In bDMARD-exposed, remission rate declined with increasing number of previous agent (1, 2, or ≥ 3): DAS28-CRP: 23.3%, 21.5%, 14.0%, (p = 0.0097 for 1 vs. ≥ 3 prior bDMARD) and DAS28-ESR: 13.4%, 8.2%, 4.5% (p = 0.0032 for 1 vs. ≥ 3 prior bDMARD)N/AAEs: 20.95%; SAEs: 7.36%. Most common AEs: infections and infestations. TB and HBV-related AEs in 2 patients eachCancer in 14 (0.27%) patientsGO-NICE/Non-interventional, prospective studies/Germany [33–35]n = 474; 64.7% bDMARD-naïve, mean (SD) age range: 54.9 (13.4) years/24-mos follow-upDAS28 mean scores were 5.0 at baseline, 3.7 at 3 mos, and 2.9 at 24 mos (p < 0.0001)DAS28 remission at 3 and 24 mos: 33.7% and 44.6%DAS28 remission with GLM as first, second, or ≥ third-line agent: 27.5%, 19.5%, 14.5% at 3 mos and 45.3%, 50.0%, and 33.3% at 24 mosFFbH scores were 68.7 at baseline, 74.0 at 3 mos (p < 0.001), and 76.1 at 24 mos (p < 0.003)FACIT-F scores were 32.4 at baseline, 35.8 at 3 mos, and 38.3 at 24 mos (p < 0.0001)AEs: 61.1%; SAEs: 14.3%Retrospective studiesRetrospective, longitudinal study from RISE registry/United States [36]n = 163 switched from IFX to IV GLM; mean age (SD) range: 65.3 (11.4) to 67.1 (11.0) years/6–9 mos and 9–12 mos after switch follow-upMean (SD) CDAI at baseline and follow-up: 6-mos cohort: 21.3 (13.1) and 14.1 (9.8) (p < 0.0001) and 9-mos cohort: 22.2 (13.9) and 15.2 (14.1) (p = 0.0002)6 mos cohort with moderate/high disease activity decreased from 73 to 56% and those with low/remitted increased from 27 and 44% (p < 0.001)9 mos cohort with moderate/high disease activity decreased from 78 to 47% and those with low/remitted increased from 22 and 53% (p < 0.001)N/AN/ARetrospective study/Taiwan [37]n = 108 who failed csDMARDs; mean (SD) age at bDMARD initiation: 59.2 (13.3) years/24-mos follow-upDAS28-ESR mean (SD) scores at baseline and 24 mos: 6.7 (0.8) and 3.1 (0.7), respectively (p < 0.001)DAS28-ESR low/remitted disease at 6, 12, 18, and 24 mos: 23.8%, 45.6%, 55%, and 58.9%, respectivelyN/AAEs: 20.4%; Serious infections: 9.3%. No cases of reactivated TB or HBV. Cancer: n = 2 (1.9%)Retrospective cohort study of data from GISEA national registry/Italy [38]n = 370 bDMARD-naïve or experienced, mean age: 56 years/12-mos follow-upBaseline DAS28-ESR mean (SD) scores similar across groups: 4.9 (1.3) in naïve and 4.9 (1.2) with 1 or ≥ 2 prior bDMARDs 6 mos scores were 3.1 (1.3) in naïve, 3.7 (1.1) with 1 prior bDMARD, and 3.7 (1.6) with ≥ 2 prior bDMARDs (both p = 0.002) 12 mos scores were 3.0 (1.2), 3.1 (1.2), and 3.7 (1.4), respectively (p = 0.001 naïve vs. ≥ 2 bDMARDs, p = 0.007 naïve vs. ≥ 2 bDMARDs)CDAI low-disease activity at 6 and 12 mos in naïve: 60% and 71% (p = 0.02 vs. 1 and ≥ 2 bDMARDs); in 1 prior bDMARD: 40% and 74%; in ≥ 2 prior bDMARDs: 40% and 57%CDAI remission at 6 and 12 mos in naïve: 16% and 21%; in 1 prior bDMARD: 7.0% and 20%; in ≥ 2 prior bDMARDs: 11% and 17%HAQ-DI mean (SD) baseline scores were similar across groups: 1.4 (0.7) in naïve, 1.5 (0.7) in 1 prior bDMARD, and 1.5 (0.6) in ≥ 2 prior bDMARDs6 mos scores were 0.9 (0.7), 1.08 (0.6), and 1.25 (0.7), respectively (p = 0.003 naïve vs. ≥ 2 bDMARDs)12 mos scores were 0.8 (0.6), 0.9 (0.6), and 1.1 (1.7); (p = 0.007 naïve vs. ≥ 2 bDMARDs)n = 79 discontinued due to lack of efficacyAE adverse event, bDMARD biologic disease-modifying antirheumatic drug, CDAI Clinical Disease Activity Index, CRP C-reactive protein, csDMARD conventional synthetic disease-modifying antirheumatic drug, DAS28 28-joint Disease Activity Score, DMARD disease-modifying antirheumatic drug, ESR erythrocyte sedimentation rate, FACIT-F Functional Assessment of Chronic Illness Therapy-Fatigue, FFbH Funktionsfragebogen Hannover (functional ability), GLM golimumab, HAQ-DI Health Assessment Questionnaire-Disability Index, HBV hepatitis B virus, IFX infliximab, IV intravenous, mos months, MTX methotrexate, N/A not applicable/not available, NMSC nonmelanoma skin cancer, RISE Rheumatology Informatics System for Effectiveness, SAE serious adverse event, SD standard deviation, TB tuberculosis, TNFi tumor necrosis factor inhibitor, vs. versus Table 2Real-world studies on the effectiveness and safety of golimumab in rheumatoid arthritisReal-world study/study design/countryPatient population/duration of follow-upEffectiveness resultsSafety resultsDisease activityPhysical function and fatigueProspective studiesGO-BEYOND/pooled analysis of prospective, observational studies/Europe [27]n = 325 with previous TNFi, mean age (SD): 57.9 (12.6) years/12-mos follow-upDAS28-CRP: low-disease activity at 3, 6, and 12 mos: 43.1%, 58.3%, 63.6%; remission at 3, 6, and 12 mos: 27.7%, 40.8%, 47.5%N/A30.5% reported an AEProspective open-label, post-marketing surveillance study/India [28]n = 120 with failed first-line DMARDs; mean age (SD): 46.3 (13.07) years/6-mos follow-upDAS28-ESR mean (SD) change from baseline (n = 62): − 2.0 (1.25). DAS28-ESR remission: 14.5%; low-disease activity: 4.8%; moderate disease activity: 67.7%; high disease activity: 12.9%HAQ-DI mean (SD) total scores change from baseline: − 0.09 (0.65)FACIT-F mean (SD) score change from baseline: 14.8 (10.20)≥ 1 AE: 51.7%; SAEs: 3.3%GO-Q/Observational, prospective cohort study/Greece [29]n = 145; mean (SD) age: 54.6 (12.2) years/12-mos follow-upDAS28-ESR mean (SD) scores at baseline and 3, 6, and 12 mos: 4.1 (1.2), 3.7 (1.2), 3.2 (1.1), and 5.4 (0.9) (all p < 0.0001)DAS28-ESR remission at 3, 6, and 12 mos: 13.3%,18.9%, and 27.3%, low-disease activity: 24.4%, 35.6%, and 54.6%; treatment response (DAS28-ESR reduction ≥ 1.2): 49.6%, 57.6%, and 76.4%HAQ-DI mean (SD) scores: 1.45 (0.61) at baseline, 0.88 (0.67) at 3 mos, 0.71 (0.57) at 6 mos, and 0.48 (0.46) at 12 mosMean (SD) changes at 3, 6, and 12 mos: − 0.58 (0.53), − 0.75 (0.62), and − 1.04 (0.73) (all p < 0.0001)N/AAWARE/prospective, observational study/United States [30]n = 685; mean age (SD) years: 60.9 (13.4); bDMARD-naïve: 35.3%/52-week follow-upbDMARD-naïve: CDAI mean change from baseline at 6 mos was − 10.7; at 12 mos: − 11.0bDMARD-experienced: CDAI mean (SD) change baseline at 6 mos was − 10.3 (18.8); at 12 mos: − 11.0 (18.4)N/A≥ 1 AE: 51%; ≥ 1 SAE: 10%; most common AE: infection. Cancer: n = 13, including melanoma, NMSC, and leukemiaOpen-label, observational, post-marketing surveillance study/Japan [31, 32]n = 5154 started on GLM (50 mg or 100 mg) with or without MTX (56% bDMARD-experienced); mean (SD) age: 62.7 (13.2) years/24-week follow-upDAS28-ESR mean (SD) scores decreased from 4.72 (1.37) to 3.48 (1.43)DAS28-ESR low/remitted disease increased from 13.69% to 46.21%Rates of remission/low-disease activity were higher with GLM + MTX (50 mg + MTX: 49.74%, 100 mg + MTX: 46.28%) vs. GLM monotherapy (50 mg: 34.31%, 100 mg: 34.04%) and with no prior bDMARD (56.23% bDMARD-naïve vs. 37.93% bDMARD-exposed)In bDMARD-exposed, remission rate declined with increasing number of previous agent (1, 2, or ≥ 3): DAS28-CRP: 23.3%, 21.5%, 14.0%, (p = 0.0097 for 1 vs. ≥ 3 prior bDMARD) and DAS28-ESR: 13.4%, 8.2%, 4.5% (p = 0.0032 for 1 vs. ≥ 3 prior bDMARD)N/AAEs: 20.95%; SAEs: 7.36%. Most common AEs: infections and infestations. TB and HBV-related AEs in 2 patients eachCancer in 14 (0.27%) patientsGO-NICE/Non-interventional, prospective studies/Germany [33–35]n = 474; 64.7% bDMARD-naïve, mean (SD) age range: 54.9 (13.4) years/24-mos follow-upDAS28 mean scores were 5.0 at baseline, 3.7 at 3 mos, and 2.9 at 24 mos (p < 0.0001)DAS28 remission at 3 and 24 mos: 33.7% and 44.6%DAS28 remission with GLM as first, second, or ≥ third-line agent: 27.5%, 19.5%, 14.5% at 3 mos and 45.3%, 50.0%, and 33.3% at 24 mosFFbH scores were 68.7 at baseline, 74.0 at 3 mos (p < 0.001), and 76.1 at 24 mos (p < 0.003)FACIT-F scores were 32.4 at baseline, 35.8 at 3 mos, and 38.3 at 24 mos (p < 0.0001)AEs: 61.1%; SAEs: 14.3%Retrospective studiesRetrospective, longitudinal study from RISE registry/United States [36]n = 163 switched from IFX to IV GLM; mean age (SD) range: 65.3 (11.4) to 67.1 (11.0) years/6–9 mos and 9–12 mos after switch follow-upMean (SD) CDAI at baseline and follow-up: 6-mos cohort: 21.3 (13.1) and 14.1 (9.8) (p < 0.0001) and 9-mos cohort: 22.2 (13.9) and 15.2 (14.1) (p = 0.0002)6 mos cohort with moderate/high disease activity decreased from 73 to 56% and those with low/remitted increased from 27 and 44% (p < 0.001)9 mos cohort with moderate/high disease activity decreased from 78 to 47% and those with low/remitted increased from 22 and 53% (p < 0.001)N/AN/ARetrospective study/Taiwan [37]n = 108 who failed csDMARDs; mean (SD) age at bDMARD initiation: 59.2 (13.3) years/24-mos follow-upDAS28-ESR mean (SD) scores at baseline and 24 mos: 6.7 (0.8) and 3.1 (0.7), respectively (p < 0.001)DAS28-ESR low/remitted disease at 6, 12, 18, and 24 mos: 23.8%, 45.6%, 55%, and 58.9%, respectivelyN/AAEs: 20.4%; Serious infections: 9.3%. No cases of reactivated TB or HBV. Cancer: n = 2 (1.9%)Retrospective cohort study of data from GISEA national registry/Italy [38]n = 370 bDMARD-naïve or experienced, mean age: 56 years/12-mos follow-upBaseline DAS28-ESR mean (SD) scores similar across groups: 4.9 (1.3) in naïve and 4.9 (1.2) with 1 or ≥ 2 prior bDMARDs 6 mos scores were 3.1 (1.3) in naïve, 3.7 (1.1) with 1 prior bDMARD, and 3.7 (1.6) with ≥ 2 prior bDMARDs (both p = 0.002) 12 mos scores were 3.0 (1.2), 3.1 (1.2), and 3.7 (1.4), respectively (p = 0.001 naïve vs. ≥ 2 bDMARDs, p = 0.007 naïve vs. ≥ 2 bDMARDs)CDAI low-disease activity at 6 and 12 mos in naïve: 60% and 71% (p = 0.02 vs. 1 and ≥ 2 bDMARDs); in 1 prior bDMARD: 40% and 74%; in ≥ 2 prior bDMARDs: 40% and 57%CDAI remission at 6 and 12 mos in naïve: 16% and 21%; in 1 prior bDMARD: 7.0% and 20%; in ≥ 2 prior bDMARDs: 11% and 17%HAQ-DI mean (SD) baseline scores were similar across groups: 1.4 (0.7) in naïve, 1.5 (0.7) in 1 prior bDMARD, and 1.5 (0.6) in ≥ 2 prior bDMARDs6 mos scores were 0.9 (0.7), 1.08 (0.6), and 1.25 (0.7), respectively (p = 0.003 naïve vs. ≥ 2 bDMARDs)12 mos scores were 0.8 (0.6), 0.9 (0.6), and 1.1 (1.7); (p = 0.007 naïve vs. ≥ 2 bDMARDs)n = 79 discontinued due to lack of efficacyAE adverse event, bDMARD biologic disease-modifying antirheumatic drug, CDAI Clinical Disease Activity Index, CRP C-reactive protein, csDMARD conventional synthetic disease-modifying antirheumatic drug, DAS28 28-joint Disease Activity Score, DMARD disease-modifying antirheumatic drug, ESR erythrocyte sedimentation rate, FACIT-F Functional Assessment of Chronic Illness Therapy-Fatigue, FFbH Funktionsfragebogen Hannover (functional ability), GLM golimumab, HAQ-DI Health Assessment Questionnaire-Disability Index, HBV hepatitis B virus, IFX infliximab, IV intravenous, mos months, MTX methotrexate, N/A not applicable/not available, NMSC nonmelanoma skin cancer, RISE Rheumatology Informatics System for Effectiveness, SAE serious adverse event, SD standard deviation, TB tuberculosis, TNFi tumor necrosis factor inhibitor, vs. versus Real-world studies on the effectiveness and safety of golimumab in rheumatoid arthritis HAQ-DI mean (SD) total scores change from baseline: − 0.09 (0.65) FACIT-F mean (SD) score change from baseline: 14.8 (10.20) DAS28-ESR mean (SD) scores at baseline and 3, 6, and 12 mos: 4.1 (1.2), 3.7 (1.2), 3.2 (1.1), and 5.4 (0.9) (all p < 0.0001) DAS28-ESR remission at 3, 6, and 12 mos: 13.3%,18.9%, and 27.3%, low-disease activity: 24.4%, 35.6%, and 54.6%; treatment response (DAS28-ESR reduction ≥ 1.2): 49.6%, 57.6%, and 76.4% HAQ-DI mean (SD) scores: 1.45 (0.61) at baseline, 0.88 (0.67) at 3 mos, 0.71 (0.57) at 6 mos, and 0.48 (0.46) at 12 mos Mean (SD) changes at 3, 6, and 12 mos: − 0.58 (0.53), − 0.75 (0.62), and − 1.04 (0.73) (all p < 0.0001) bDMARD-naïve: CDAI mean change from baseline at 6 mos was − 10.7; at 12 mos: − 11.0 bDMARD-experienced: CDAI mean (SD) change baseline at 6 mos was − 10.3 (18.8); at 12 mos: − 11.0 (18.4) DAS28-ESR mean (SD) scores decreased from 4.72 (1.37) to 3.48 (1.43) DAS28-ESR low/remitted disease increased from 13.69% to 46.21% Rates of remission/low-disease activity were higher with GLM + MTX (50 mg + MTX: 49.74%, 100 mg + MTX: 46.28%) vs. GLM monotherapy (50 mg: 34.31%, 100 mg: 34.04%) and with no prior bDMARD (56.23% bDMARD-naïve vs. 37.93% bDMARD-exposed) In bDMARD-exposed, remission rate declined with increasing number of previous agent (1, 2, or ≥ 3): DAS28-CRP: 23.3%, 21.5%, 14.0%, (p = 0.0097 for 1 vs. ≥ 3 prior bDMARD) and DAS28-ESR: 13.4%, 8.2%, 4.5% (p = 0.0032 for 1 vs. ≥ 3 prior bDMARD) AEs: 20.95%; SAEs: 7.36%. Most common AEs: infections and infestations. TB and HBV-related AEs in 2 patients each Cancer in 14 (0.27%) patients GO-NICE/ Non-interventional, prospective studies/Germany [33–35] DAS28 mean scores were 5.0 at baseline, 3.7 at 3 mos, and 2.9 at 24 mos (p < 0.0001) DAS28 remission at 3 and 24 mos: 33.7% and 44.6% DAS28 remission with GLM as first, second, or ≥ third-line agent: 27.5%, 19.5%, 14.5% at 3 mos and 45.3%, 50.0%, and 33.3% at 24 mos FFbH scores were 68.7 at baseline, 74.0 at 3 mos (p < 0.001), and 76.1 at 24 mos (p < 0.003) FACIT-F scores were 32.4 at baseline, 35.8 at 3 mos, and 38.3 at 24 mos (p < 0.0001) Mean (SD) CDAI at baseline and follow-up: 6-mos cohort: 21.3 (13.1) and 14.1 (9.8) (p < 0.0001) and 9-mos cohort: 22.2 (13.9) and 15.2 (14.1) (p = 0.0002) 6 mos cohort with moderate/high disease activity decreased from 73 to 56% and those with low/remitted increased from 27 and 44% (p < 0.001) 9 mos cohort with moderate/high disease activity decreased from 78 to 47% and those with low/remitted increased from 22 and 53% (p < 0.001) DAS28-ESR mean (SD) scores at baseline and 24 mos: 6.7 (0.8) and 3.1 (0.7), respectively (p < 0.001) DAS28-ESR low/remitted disease at 6, 12, 18, and 24 mos: 23.8%, 45.6%, 55%, and 58.9%, respectively Baseline DAS28-ESR mean (SD) scores similar across groups: 4.9 (1.3) in naïve and 4.9 (1.2) with 1 or ≥ 2 prior bDMARDs 6 mos scores were 3.1 (1.3) in naïve, 3.7 (1.1) with 1 prior bDMARD, and 3.7 (1.6) with ≥ 2 prior bDMARDs (both p = 0.002) 12 mos scores were 3.0 (1.2), 3.1 (1.2), and 3.7 (1.4), respectively (p = 0.001 naïve vs. ≥ 2 bDMARDs, p = 0.007 naïve vs. ≥ 2 bDMARDs) CDAI low-disease activity at 6 and 12 mos in naïve: 60% and 71% (p = 0.02 vs. 1 and ≥ 2 bDMARDs); in 1 prior bDMARD: 40% and 74%; in ≥ 2 prior bDMARDs: 40% and 57% CDAI remission at 6 and 12 mos in naïve: 16% and 21%; in 1 prior bDMARD: 7.0% and 20%; in ≥ 2 prior bDMARDs: 11% and 17% HAQ-DI mean (SD) baseline scores were similar across groups: 1.4 (0.7) in naïve, 1.5 (0.7) in 1 prior bDMARD, and 1.5 (0.6) in ≥ 2 prior bDMARDs 6 mos scores were 0.9 (0.7), 1.08 (0.6), and 1.25 (0.7), respectively (p = 0.003 naïve vs. ≥ 2 bDMARDs) 12 mos scores were 0.8 (0.6), 0.9 (0.6), and 1.1 (1.7); (p = 0.007 naïve vs. ≥ 2 bDMARDs) AE adverse event, bDMARD biologic disease-modifying antirheumatic drug, CDAI Clinical Disease Activity Index, CRP C-reactive protein, csDMARD conventional synthetic disease-modifying antirheumatic drug, DAS28 28-joint Disease Activity Score, DMARD disease-modifying antirheumatic drug, ESR erythrocyte sedimentation rate, FACIT-F Functional Assessment of Chronic Illness Therapy-Fatigue, FFbH Funktionsfragebogen Hannover (functional ability), GLM golimumab, HAQ-DI Health Assessment Questionnaire-Disability Index, HBV hepatitis B virus, IFX infliximab, IV intravenous, mos months, MTX methotrexate, N/A not applicable/not available, NMSC nonmelanoma skin cancer, RISE Rheumatology Informatics System for Effectiveness, SAE serious adverse event, SD standard deviation, TB tuberculosis, TNFi tumor necrosis factor inhibitor, vs. versus Health-Related Quality of Life Real-world studies have evaluated the effectiveness of golimumab for improving HR-QoL, activity, and work productivity (Table 3). Collectively, these studies have shown a benefit with treatment for improvements in general-QoL and HR-QoL and decreases in activity impairment and work productivity loss. In addition, the GO-ACTIVE observational study in patients with RA (n = 95) evaluated golimumab for up to 2 years and found associations between disease activity, work productivity, and daily activity [39]. The Total Work Productivity Impairment (TWPI) and Activity Impairment subscores of the Work Productivity and Activity Impairment (WPAI) questionnaire were significantly decreased at 3, 6, 12, 18, and 24 months (all p < 0.0001), and HR-QoL was significantly improved at each timepoint (all p < 0.0001). At baseline, the WPAI scores (TWPI, Absenteeism, Presenteeism, and Activity Impairment) significantly correlated with the Clinical Disease Activity Index (CDAI) (all p < 0.0001). Through 12 months, each point increase in baseline CDAI equated to a 1.1% increase in the TWPI subscore. Similarly, baseline HR-QoL correlated with TWPI at 3 and 12 months and with Activity Impairment at 3, 12, and 24 months (all p < 0.0001). Table 3Real-world prospective studies on productivity and HR-QoL with golimumab in patients with rheumatoid arthritisReal-world study/study design/countryPatient population/duration of follow-upProductivity and HR-QoL outcomesGO-BEYOND/pooled analysis of prospective, observational studies/Europe [27]n = 325 with previous TNFi treatment, mean age (SD): 57.9 (12.6) years/12-mos follow-upEQ-5D-3L VAS scores increased from a median of 50 (Q1, Q3: 30–60) at baseline to 70 (Q1, Q3: 55–80) at 6 mos, and 75 (Q1, Q3: 60–89) at 12 mosEQ-5D-3L dimensions: proportion of patients reporting no problems at baseline, 6, and 12 mos in mobility: 25%, 49.1%, 56.4%; self-care: 29.6%, 60.9%, 68.7%; pain/discomfort: 1.9%, 31.8%, 34.0%; usual activities: 11.2%, 37.9%, 49.1%; and anxiety/depression: 26.5%, 46.7%, 43.6%AWARE/prospective, non-interventional, study/United States [40]n = 685; mean age (SD) years: 60.9 (13.4); bDMARD-naïve: 35.3%/52-week follow-upPROMIS-29 and PROMIS-short form T-scores: Meaningful improvements (≥ 3-point improvement) in all 7 domains (depression, anxiety, physical function, pain interference, fatigue, sleep disturbance, and participation in social roles and activities) and respective short forms (fatigue and pain interference) that were maintained through end of study Proportions of patients with ≥ 3, ≥ 5, or ≥ 10-point improvements in T-scores increased from the second infusion through the last infusion at end of studyGO-Q/observational, prospective cohort study/Greece [29]n = 145; mean (SD) age: 54.6 (12.2) years/12-mos follow-upEQ-VAS scores increased from a mean (SD) of 47.6 (16.8) at baseline to 62.8 (21.2) at 3 mos, 66.8 (21.6) at 6 mos, and 77.4 (16.8) at 12 mosEQ-5D-3L dimensions: proportion reporting no problems at baseline, 3, 6, and 12 mos in mobility: 18.8%, 51.1%, 67.9%, 80.9%; self-care: 25.7%, 60.0%, 71.8%, 81.8%; usual activities: 6.3%, 45.2%, 58.0%, 71.8%; pain/discomfort: 2.8%, 33.3%, 44.3%, 66.4%; and anxiety/depression: 19.4%, 40.7%, 51.9%, 61.8%EQ-5D-3L utility index: mean (SD) changes from baseline to 3, 6, and 12 mos were 0.21 (0.24), 0.28 (0.25), and 0.38 (0.30) (all p < 0.0001)WPAI RA domain scores: mean changes from baseline to 3, 6, and 12 mos in absenteeism: − 8.7, − 10.8, − 11.2; presenteeism: − 20.1, − 24.5, − 28.5; work productivity impairment: − 14.2, − 17.8, − 21.4; activity impairment: − 22.1, − 26.3, − 32.3 (all p < 0.0001)GO-NICE/non-interventional, prospective study/Germany [34]n = 474; 60.4% bDMARD-naïve, mean (SD) age range: 54.9 (13.4) years/24-mos follow-upEQ-VAS score increased from a mean 51.0 at baseline to 60.2 at 6 mos, 62.4 at 12 mos, 64.4 at 18 mos, and 63.4 at 24 mos (all p < 0.0001)EQ-5D-3L dimensions: proportion reporting some or extreme problems at baseline, 6, 12, 18, and 24 mos in pain/discomfort: 94.7%, 81.3%, 77.2%, 78.2%, 74.8%; usual activities: 70.0%, 57.1%, 52.3%, 48.8%, 50.3%; mobility: 56.0%, 45.6%, 36.6%, 36.54%, 40.1%; anxiety/depression: 45.0%, 35.9%, 30.1%, 25.6%, 25.8%; self-care: 32.2%, 24.9%, 24.8%, 17.5%, 18.4% (all p ≤ 0.05)bDMARD biologic disease-modifying antirheumatic drug, EQ-5D-3L EuroQol 5-Dimensions 3-Levels, EQ-VAS EuroQoL visual analog score, HR-QoL health-related quality of life, mos months, PROMIS Patient-Reported Outcomes Measurement Information System, RA rheumatoid arthritis, SD standard deviation, TNFi tumor necrosis factor inhibitor, VAS visual analog scale, WPAI Work Productivity and Daily Activity Impairment Table 3Real-world prospective studies on productivity and HR-QoL with golimumab in patients with rheumatoid arthritisReal-world study/study design/countryPatient population/duration of follow-upProductivity and HR-QoL outcomesGO-BEYOND/pooled analysis of prospective, observational studies/Europe [27]n = 325 with previous TNFi treatment, mean age (SD): 57.9 (12.6) years/12-mos follow-upEQ-5D-3L VAS scores increased from a median of 50 (Q1, Q3: 30–60) at baseline to 70 (Q1, Q3: 55–80) at 6 mos, and 75 (Q1, Q3: 60–89) at 12 mosEQ-5D-3L dimensions: proportion of patients reporting no problems at baseline, 6, and 12 mos in mobility: 25%, 49.1%, 56.4%; self-care: 29.6%, 60.9%, 68.7%; pain/discomfort: 1.9%, 31.8%, 34.0%; usual activities: 11.2%, 37.9%, 49.1%; and anxiety/depression: 26.5%, 46.7%, 43.6%AWARE/prospective, non-interventional, study/United States [40]n = 685; mean age (SD) years: 60.9 (13.4); bDMARD-naïve: 35.3%/52-week follow-upPROMIS-29 and PROMIS-short form T-scores: Meaningful improvements (≥ 3-point improvement) in all 7 domains (depression, anxiety, physical function, pain interference, fatigue, sleep disturbance, and participation in social roles and activities) and respective short forms (fatigue and pain interference) that were maintained through end of study Proportions of patients with ≥ 3, ≥ 5, or ≥ 10-point improvements in T-scores increased from the second infusion through the last infusion at end of studyGO-Q/observational, prospective cohort study/Greece [29]n = 145; mean (SD) age: 54.6 (12.2) years/12-mos follow-upEQ-VAS scores increased from a mean (SD) of 47.6 (16.8) at baseline to 62.8 (21.2) at 3 mos, 66.8 (21.6) at 6 mos, and 77.4 (16.8) at 12 mosEQ-5D-3L dimensions: proportion reporting no problems at baseline, 3, 6, and 12 mos in mobility: 18.8%, 51.1%, 67.9%, 80.9%; self-care: 25.7%, 60.0%, 71.8%, 81.8%; usual activities: 6.3%, 45.2%, 58.0%, 71.8%; pain/discomfort: 2.8%, 33.3%, 44.3%, 66.4%; and anxiety/depression: 19.4%, 40.7%, 51.9%, 61.8%EQ-5D-3L utility index: mean (SD) changes from baseline to 3, 6, and 12 mos were 0.21 (0.24), 0.28 (0.25), and 0.38 (0.30) (all p < 0.0001)WPAI RA domain scores: mean changes from baseline to 3, 6, and 12 mos in absenteeism: − 8.7, − 10.8, − 11.2; presenteeism: − 20.1, − 24.5, − 28.5; work productivity impairment: − 14.2, − 17.8, − 21.4; activity impairment: − 22.1, − 26.3, − 32.3 (all p < 0.0001)GO-NICE/non-interventional, prospective study/Germany [34]n = 474; 60.4% bDMARD-naïve, mean (SD) age range: 54.9 (13.4) years/24-mos follow-upEQ-VAS score increased from a mean 51.0 at baseline to 60.2 at 6 mos, 62.4 at 12 mos, 64.4 at 18 mos, and 63.4 at 24 mos (all p < 0.0001)EQ-5D-3L dimensions: proportion reporting some or extreme problems at baseline, 6, 12, 18, and 24 mos in pain/discomfort: 94.7%, 81.3%, 77.2%, 78.2%, 74.8%; usual activities: 70.0%, 57.1%, 52.3%, 48.8%, 50.3%; mobility: 56.0%, 45.6%, 36.6%, 36.54%, 40.1%; anxiety/depression: 45.0%, 35.9%, 30.1%, 25.6%, 25.8%; self-care: 32.2%, 24.9%, 24.8%, 17.5%, 18.4% (all p ≤ 0.05)bDMARD biologic disease-modifying antirheumatic drug, EQ-5D-3L EuroQol 5-Dimensions 3-Levels, EQ-VAS EuroQoL visual analog score, HR-QoL health-related quality of life, mos months, PROMIS Patient-Reported Outcomes Measurement Information System, RA rheumatoid arthritis, SD standard deviation, TNFi tumor necrosis factor inhibitor, VAS visual analog scale, WPAI Work Productivity and Daily Activity Impairment Real-world prospective studies on productivity and HR-QoL with golimumab in patients with rheumatoid arthritis EQ-5D-3L VAS scores increased from a median of 50 (Q1, Q3: 30–60) at baseline to 70 (Q1, Q3: 55–80) at 6 mos, and 75 (Q1, Q3: 60–89) at 12 mos EQ-5D-3L dimensions: proportion of patients reporting no problems at baseline, 6, and 12 mos in mobility: 25%, 49.1%, 56.4%; self-care: 29.6%, 60.9%, 68.7%; pain/discomfort: 1.9%, 31.8%, 34.0%; usual activities: 11.2%, 37.9%, 49.1%; and anxiety/depression: 26.5%, 46.7%, 43.6% PROMIS-29 and PROMIS-short form T-scores: Meaningful improvements (≥ 3-point improvement) in all 7 domains (depression, anxiety, physical function, pain interference, fatigue, sleep disturbance, and participation in social roles and activities) and respective short forms (fatigue and pain interference) that were maintained through end of study Proportions of patients with ≥ 3, ≥ 5, or ≥ 10-point improvements in T-scores increased from the second infusion through the last infusion at end of study EQ-VAS scores increased from a mean (SD) of 47.6 (16.8) at baseline to 62.8 (21.2) at 3 mos, 66.8 (21.6) at 6 mos, and 77.4 (16.8) at 12 mos EQ-5D-3L dimensions: proportion reporting no problems at baseline, 3, 6, and 12 mos in mobility: 18.8%, 51.1%, 67.9%, 80.9%; self-care: 25.7%, 60.0%, 71.8%, 81.8%; usual activities: 6.3%, 45.2%, 58.0%, 71.8%; pain/discomfort: 2.8%, 33.3%, 44.3%, 66.4%; and anxiety/depression: 19.4%, 40.7%, 51.9%, 61.8% EQ-5D-3L utility index: mean (SD) changes from baseline to 3, 6, and 12 mos were 0.21 (0.24), 0.28 (0.25), and 0.38 (0.30) (all p < 0.0001) WPAI RA domain scores: mean changes from baseline to 3, 6, and 12 mos in absenteeism: − 8.7, − 10.8, − 11.2; presenteeism: − 20.1, − 24.5, − 28.5; work productivity impairment: − 14.2, − 17.8, − 21.4; activity impairment: − 22.1, − 26.3, − 32.3 (all p < 0.0001) EQ-VAS score increased from a mean 51.0 at baseline to 60.2 at 6 mos, 62.4 at 12 mos, 64.4 at 18 mos, and 63.4 at 24 mos (all p < 0.0001) EQ-5D-3L dimensions: proportion reporting some or extreme problems at baseline, 6, 12, 18, and 24 mos in pain/discomfort: 94.7%, 81.3%, 77.2%, 78.2%, 74.8%; usual activities: 70.0%, 57.1%, 52.3%, 48.8%, 50.3%; mobility: 56.0%, 45.6%, 36.6%, 36.54%, 40.1%; anxiety/depression: 45.0%, 35.9%, 30.1%, 25.6%, 25.8%; self-care: 32.2%, 24.9%, 24.8%, 17.5%, 18.4% (all p ≤ 0.05) bDMARD biologic disease-modifying antirheumatic drug, EQ-5D-3L EuroQol 5-Dimensions 3-Levels, EQ-VAS EuroQoL visual analog score, HR-QoL health-related quality of life, mos months, PROMIS Patient-Reported Outcomes Measurement Information System, RA rheumatoid arthritis, SD standard deviation, TNFi tumor necrosis factor inhibitor, VAS visual analog scale, WPAI Work Productivity and Daily Activity Impairment Drug Survival Drug survival measures how long patients remain on therapy uninterrupted in the real-world, typically using pharmacy claims databases, an indirect measure suggesting effectiveness, safety, tolerability, and patient/provider satisfaction. This is a critical metric for evaluating treatment success in chronic conditions such as RA. Long-term use of golimumab, predictive factors for drug survival, and the impact of prior treatment with bDMARDs or JAKi were evaluated using data from a Japanese hospital insurance claims database [41]. A total of 7524 patients with RA received golimumab (bDMARD-naïve: 2283; one previous bDMARD/JAKi: 942; ≥ 2 previous bDMARD/JAKi: 322). The persistence rates in bDMARD-naïve were 58.8%, 32.1%, 21.4%, and 11.4% at 1, 3, 5, and 7 years, respectively, which were significantly higher than the two bDMARD/JAKi-experienced groups (both p < 0.001). Several factors were associated with a high persistence rate in the naïve group, including age (being 61–75 years, p = 0.008), being female (p = 0.019), and coadministration of either low or high dose MTX (both p < 0.001, respectively). Two smaller studies evaluated golimumab drug survival and associated predictive factors. In the first study, 81 participants were bDMARD-naïve and 27 were bDMARD-experienced [37]. The mean duration of treatment was 38.3 months, with 87 participants continuing treatment. The retention rates were 92.6%, 84.8%, and 76.8% at 12, 24, and 70 months, respectively. Factors associated with poor retention included high baseline ESR, high DAS28-ESR, glucocorticoid use, and being bDMARD-experienced (all p < 0.005). In the second study, participants were stratified after treatment with golimumab into retention (n = 28) and discontinued due to lack of efficacy (n = 24) groups [42]. The overall drug survival was 66.3%, 48.3%, and 24.5% at 12, 36, and 72 months, respectively. In addition, the retention rates after excluding participants who discontinued treatment because of lack of efficacy were: 67.0%, 50.0%, and 44.2% at 12, 36, and 72 months, respectively. In the retention group, significant predictive factors were concomitant MTX (p = 0.005) and being bDMARD-naïve (p < 0.001), similar to the study above. Lastly, a large population-based study evaluated drug survival using the National Database of Health Insurance Claims and Specific Health Checkups of Japan, which covers 98.3% of the population [43]. In addition, the study compared retention rates across bDMARDs. The study identified 15,165 bDMARD-naïve (median estimated age interquartile range [IQR]: 65–69) and 5446 bDMARD-experienced (median estimated age IQR: 65–69) who were treated with golimumab. An analysis of differences in retention rate using a log-rank test determined that the golimumab retention rate for the bDMARD-naïve group was significantly higher than that of adalimumab, certolizumab pegol, etanercept, and infliximab (all p < 0.0001). By contrast, golimumab had a significantly lower rate than abatacept and tocilizumab (both p < 0.0001). In the bDMARD-experienced group, the rate with golimumab was significantly higher versus adalimumab and certolizumab pegol (both p < 0.0001) and significantly lower against abatacept, tocilizumab, and tofacitinib (p < 0.0001). Collectively, these studies showed that drug survival with golimumab was robust, and factors that predicted retention included being bDMARD-naïve, having less severe disease, coadministration with MTX, and being female. Table 4 presents data from additional studies that evaluated drug survival in participants treated with golimumab. Consistent with the studies above, the results showed that drug survival was robust and consistently higher in participants who were bDMARD-naïve. Table 4Real-world studies on drug survival with golimumab in patients with rheumatoid arthritisReal-world study/study design/countryPatient population/duration of follow-upDrug survival outcomesProspective studyGO-PRACTICE/prospective observational study of public or private practices/France [44]n = 170 (mean age [SD]: 54.3 [12.3] years); 64.7% bDMARD-naïve/2-year follow-upMonth 12 persistence: 65.4%; persistence was higher in bDMARD-naïve (69.6%) than bDMARD-experienced patients (56.9%)Month 24 persistence: 56.5%; persistence was higher in bDMARD-naïve (60.6%) than bDMARD-experienced patients (48.2%)Retrospective studiesRetrospective analysis of BIOBADASER/Spain [45]n = 267 (mean [SD] age: 57.4 [12.2] years)/8-year follow-upRetention rate at year 1 and 8 was 58.6% (52.6–64.1) and 24.6% (15.4–35.0), respectivelyRetention rate was higher in bDMARD-naïve (log‐rank p < 0.001)Nationwide, population-based real-world study of insurance claim datasets/Taiwan [46]n = 1369 (mean [SD] age: 46.23 [14.52] years)/5-year follow-upGLM had a lower cumulative probability of discontinuation across years 1, 2, 3, 4, and 5 versus etanercept and adalimumab: 17.72 (15.58–19.86), 30.86 (28.12–33.60), 38.50 (35.42–41.58), 46.86 (43.33–50.39), and 51.81 (47.71–55.91), respectivelyRetrospective long-term retention study/Spain [47]n = 61 (mean age: 55 [14.1] years). Approximately 50% bDMARD-experienced; main reasons for switch to GLM were lack of efficacy (57.1%) and AEs (32.1%)/5-year follow-upMedian time to discontinuation was 50.2 monthsRetention rate was 71% at year 1, 57%, at year 2, 51% at year 3, and 51% at year 4, and 39% at year 5GO-BEYOND/multicenter, retrospective study/Turkey [48]n = 60 (median age: 52, range: 39–60 years). bDMARD-naïve: n = 53, bDMARD-experienced: n = 7/24-mos follow-upMonths 3, 6, 12, and 24 retention rates: 95.0%, 86.4%, 74.5%, and 65.5%, respectively, in combined bDMARD-naïve and -experiencedMonth 24 retention rate for bDMARD-naïve: 67.2% and bDMARD-experienced: 57.1%AE adverse event, bDMARD biologic disease-modifying antirheumatic drug, GLM golimumab, RA rheumatoid arthritis, SD standard deviation Table 4Real-world studies on drug survival with golimumab in patients with rheumatoid arthritisReal-world study/study design/countryPatient population/duration of follow-upDrug survival outcomesProspective studyGO-PRACTICE/prospective observational study of public or private practices/France [44]n = 170 (mean age [SD]: 54.3 [12.3] years); 64.7% bDMARD-naïve/2-year follow-upMonth 12 persistence: 65.4%; persistence was higher in bDMARD-naïve (69.6%) than bDMARD-experienced patients (56.9%)Month 24 persistence: 56.5%; persistence was higher in bDMARD-naïve (60.6%) than bDMARD-experienced patients (48.2%)Retrospective studiesRetrospective analysis of BIOBADASER/Spain [45]n = 267 (mean [SD] age: 57.4 [12.2] years)/8-year follow-upRetention rate at year 1 and 8 was 58.6% (52.6–64.1) and 24.6% (15.4–35.0), respectivelyRetention rate was higher in bDMARD-naïve (log‐rank p < 0.001)Nationwide, population-based real-world study of insurance claim datasets/Taiwan [46]n = 1369 (mean [SD] age: 46.23 [14.52] years)/5-year follow-upGLM had a lower cumulative probability of discontinuation across years 1, 2, 3, 4, and 5 versus etanercept and adalimumab: 17.72 (15.58–19.86), 30.86 (28.12–33.60), 38.50 (35.42–41.58), 46.86 (43.33–50.39), and 51.81 (47.71–55.91), respectivelyRetrospective long-term retention study/Spain [47]n = 61 (mean age: 55 [14.1] years). Approximately 50% bDMARD-experienced; main reasons for switch to GLM were lack of efficacy (57.1%) and AEs (32.1%)/5-year follow-upMedian time to discontinuation was 50.2 monthsRetention rate was 71% at year 1, 57%, at year 2, 51% at year 3, and 51% at year 4, and 39% at year 5GO-BEYOND/multicenter, retrospective study/Turkey [48]n = 60 (median age: 52, range: 39–60 years). bDMARD-naïve: n = 53, bDMARD-experienced: n = 7/24-mos follow-upMonths 3, 6, 12, and 24 retention rates: 95.0%, 86.4%, 74.5%, and 65.5%, respectively, in combined bDMARD-naïve and -experiencedMonth 24 retention rate for bDMARD-naïve: 67.2% and bDMARD-experienced: 57.1%AE adverse event, bDMARD biologic disease-modifying antirheumatic drug, GLM golimumab, RA rheumatoid arthritis, SD standard deviation Real-world studies on drug survival with golimumab in patients with rheumatoid arthritis Month 12 persistence: 65.4%; persistence was higher in bDMARD-naïve (69.6%) than bDMARD-experienced patients (56.9%) Month 24 persistence: 56.5%; persistence was higher in bDMARD-naïve (60.6%) than bDMARD-experienced patients (48.2%) Retention rate at year 1 and 8 was 58.6% (52.6–64.1) and 24.6% (15.4–35.0), respectively Retention rate was higher in bDMARD-naïve (log‐rank p < 0.001) Median time to discontinuation was 50.2 months Retention rate was 71% at year 1, 57%, at year 2, 51% at year 3, and 51% at year 4, and 39% at year 5 Months 3, 6, 12, and 24 retention rates: 95.0%, 86.4%, 74.5%, and 65.5%, respectively, in combined bDMARD-naïve and -experienced Month 24 retention rate for bDMARD-naïve: 67.2% and bDMARD-experienced: 57.1% AE adverse event, bDMARD biologic disease-modifying antirheumatic drug, GLM golimumab, RA rheumatoid arthritis, SD standard deviation Collectively, the results from the pivotal RCTs and extension trials, along with real-world studies, demonstrated that golimumab significantly reduced pain, physical disability, fatigue, radiographic progression, disease activity, swollen and tender joints, and acute phase reactants (Fig. 1).Fig. 1Effects of golimumab demonstrated in pivotal trials and real-world studies Effects of golimumab demonstrated in pivotal trials and real-world studies Meta-analyses Indirectly Comparing Golimumab Efficacy, Safety, and Drug Survival Outcomes to Other TNFi For the treatment of RA, there is a lack of head-to-head studies, RCT or real-world studies that directly compare the efficacy, safety, and persistence between golimumab and the other TNFi; however, four recent meta-analyses were conducted that indirectly compared the TNFi [49–52]. The first analyzed efficacy and safety data from RCTs in which adults with RA were treated with a TNFi in combination with MTX after MTX failure [49]. Regarding the efficacy outcomes of remission (CDAI ≤ 2.8) and low disease activity (CDAI ≤ 10) and the risk of SAEs, the comparisons of the risk ratios (RRs) between golimumab + MTX versus adalimumab + MTX, certolizumab pegol + MTX, etanercept + MTX, and infliximab + MTX were not significant (see Fig. 2). Regarding the risk of infections, golimumab + MTX was associated with a 32% lower risk compared with certolizumab pegol + MTX, but there were no significant differences in the risk of infections between golimumab + MTX and the other TNFi + MTX. Regarding serious infections, golimumab + MTX was associated with a higher risk relative to infliximab + MTX as was certolizumab pegol + MTX. Regarding the variability in the CIs within the meta-analysis, the authors noted that these tended to be wider for outcomes with low event numbers and more narrow for outcomes with high event numbers. Across the safety and efficacy outcomes, the authors concluded that, based on the limited number of statistically significant TNFi comparisons and variable CIs, there were only minor differences across the drug class.Fig. 2Forest plot of remission and disease activity and risk of SAEs, infections, and serious infections (data from [49]). The data presented in the figure represent the RRs and 95% CIs on a log scale. ADA adalimumab, CDAI Clinical Disease Activity Index, CI confidence interval, CTZ certolizumab pegol, ETN etanercept, GOL golimumab, IFX infliximab, MTX methotrexate, RR risk ratio, SAE serious adverse event Forest plot of remission and disease activity and risk of SAEs, infections, and serious infections (data from [49]). The data presented in the figure represent the RRs and 95% CIs on a log scale. ADA adalimumab, CDAI Clinical Disease Activity Index, CI confidence interval, CTZ certolizumab pegol, ETN etanercept, GOL golimumab, IFX infliximab, MTX methotrexate, RR risk ratio, SAE serious adverse event The second meta-analysis evaluated safety data from RCTs of adults with RA treated with a TNFi [50]. Regarding the risk of AEs, the highest surface under the cumulative ranking curve (SUCRA; larger values indicate greater safety risk) was with certolizumab pegol (90.6) and placebo (6.6) had the lowest risk; the SUCRAs for golimumab + DMARD and golimumab were 59.1 and 53.5, respectively. Regarding the risk of SAEs, golimumab + DMARD had the highest SUCRA (94.0) and the lowest was for infliximab (12.3); the SUCRA for golimumab was 16.7. Regarding the risk of infections, infliximab + DMARD had the highest SUCRA (91.0) and the lowest was for a DMARD (21.0); the SUCRAs for golimumab and golimumab + DMARD were 40.9 and 27.4, respectively. Regarding the risk of serious infections, certolizumab pegol had the highest SUCRA (81.7), and etanercept + DMARD had the lowest (28.5); the SUCRAs for golimumab + DMARD and golimumab were 57.5 and 31.8, respectively. Lastly, regarding the risk of malignant tumors, golimumab had the highest SUCRA (77.8), and golimumab + DMARD had the lowest (23.7). The third meta-analysis evaluated the effectiveness of TNFi treatment for patients with RA who were identified in administrative health databases [51]. Overall, golimumab, adalimumab, and etanercept were found to be 19% more effective versus infliximab (RR: 1.19, 95% CI 1.05–1.36, p < 0.01). Comparisons between individual drugs found golimumab treatment was associated with significantly higher effectiveness than infliximab (RR: 1.57, 1.19–2.08, p < 0.01). The fourth meta-analysis evaluated long-term (5 years) drug survival using data from observational studies (registries and cohort studies) and long-term extensions of RCTs of adults with RA who had failed a conventional synthetic DMARD [52]. Among the TNFi, golimumab was found to have the highest persistence (62%, 95% CI 47–75%) followed by certolizumab pegol (52%, 45–59%), etanercept (42%, 24–62%), adalimumab (36%, 23–53%), and infliximab (35%, 30–41%). Of note, the authors conveyed that most of the studies included in the meta-analysis evaluated adalimumab, etanercept, and infliximab. Collectively, the results were equivocal across the three meta-analyses that evaluated efficacy and/or safety outcomes following TNFi treatment (i.e., golimumab had a higher risk of infections and serious infections in one analysis but not the other) and did not favor one TNFi over another. Risk of Infections Patients with RA are at an increased risk of developing infections [53]. Further, TNFi treatment may potentially compromise the immune system, leaving patients with RA even more vulnerable to infections. To assess this, two studies evaluated the effect of golimumab on the risk of developing infections. The first study analyzed pooled safety data from 3 RCTs that evaluated IV golimumab for the treatment of RA, psoriatic arthritis (PsA), or axial spondyloarthritis (axSpA) [54]. Across the three patient populations, during the placebo-controlled phase, infections were the most common AE and occurred in 23.8% of the golimumab group and 17.3% of placebo. Serious infections occurred in 0.8% of the golimumab group and in 0.4% of placebo. In total, 49 (3.9%) of all golimumab-treated participants, including those who previously received placebo, had a serious infection; the incidence/100 PY was 3.4 (95% CI 2.5–4.4). Serious infections reported included pneumonia (11 cases), UTI (five cases), sepsis (four cases), appendicitis (two cases), empyema (two cases), and erysipelas (two cases). Serious infections occurring as single instances included infected dermal cyst, acute pyelonephritis, periodontitis, and acute hepatitis of mixed etiology. Among the golimumab group, there were six cases of active TB (RA trial: n = 3, PsA trial: n = 2, axSpA trial: n = 1). Of note, all participants screened negative for TB at baseline and lived in countries endemic for TB. Opportunistic infections occurred in the golimumab group (n = 4) and included non-serious esophageal candidiasis infection (n = 2) and the serious infections cryptococcal pneumonia (n = 1) and localized vertebral candidiasis (n = 1). The second study evaluated real-world incidence and determinants of infection in patients with RA (n = 530) treated with golimumab who were identified in the Biologic Treatment Registry Across Canada [55]. During a mean follow-up period of 27 months, the total infection incidence density rate (IDR) was 35.1 events per 100 PY, with most infections occurring within the first 6 months. The IDRs were 32.9 events per 100 PY for non-serious infections and 2.2 events per 100 PY for serious infections. No predictive factors for non-serious infection were identified within the first 6 months. Patients with pulmonary comorbidities had significantly higher odds of developing both total and non-serious infections over time. Advanced age and high corticosteroid dosage (> 5 mg/day) were predictors for increased serious infection risk. The occurrence of serious infections, but not non-serious infections, was significantly associated with a higher likelihood of golimumab discontinuation. Two studies compared the risk of developing infection across TNFi in participants from RCTs. In the first study, a network meta-analysis, a total of 61 RCTs involving 20,458 participants with RA, axSpA, or PsA were included [56]. The pairwise meta-analysis that compared placebo to TNFi, determined that certolizumab pegol was associated with the highest significant risk of serious infection (odds ratio [OR]: 2.28, 95% CI 1.13–4.62). By contrast, no significant differences were detected between golimumab (OR: 0.69, 95% CI 0.36–1.35), infliximab (OR: 1.39, 95% CI 0.72–2.68), adalimumab (OR: 1.27, 95% CI 0.85–1.88), or etanercept (OR: 0.79, 95% CI 0.46–1.34) versus placebo. Regarding any infection, both adalimumab and certolizumab pegol were associated with a significantly higher risk versus placebo (OR: 1.18, 95% CI 1.06–1.30, OR: 1.40, 95% CI 1.11–1.76, respectively). Regarding opportunistic infections and herpes zoster (HZ), there was no increased risk with TNFi versus placebo; however, there was an increased risk for TB (OR: 2.21, 95% CI 1.05–4.66). Similarly, the network meta-analysis found that certolizumab pegol was associated with a significantly increased risk of serious infection versus placebo, golimumab, and etanercept (OR: 1.85, 95% CI 1.09–3.14; OR: 2.67, 95% CI 1.14–6.26 and OR: 2.41, 95% CI 1.14–5.08, respectively). The cumulative probability for the risk of serious infection was certolizumab pegol (90.7%), infliximab (73.1%), adalimumab (69.5%), etanercept (17.0%) and golimumab (12.8%). Regarding risk of any infection, adalimumab was associated with a significantly higher risk versus placebo (OR: 1.18, 95% CI 1.06–1.30) and the cumulative ranking demonstrated that certolizumab pegol had the highest risk (87.7%), followed by golimumab (56.5%), infliximab (53.3%), adalimumab (53.6%) and etanercept (38.1%). Overall, these results suggest that golimumab and etanercept were associated with a lower risk of infection. In the second study, a meta-analysis, the risk of TB in participants with RA treated with a TNFi was evaluated [57]. Among the 50 identified RCTs, eight evaluated golimumab, nine evaluated etanercept, 18 evaluated adalimumab, nine evaluated infliximab, and six evaluated certolizumab pegol. Thirteen registry and longitudinal cohort studies were also identified and all included ≥ 2 TNFi cohorts. Regarding the RCTs, no cases of TB were identified in the etanercept studies. The RRs for golimumab, infliximab, adalimumab, and certolizumab pegol were 1.18, 1.65, 1.01, and 1.02, respectively, and there was no significant difference in TB risk observed between each TNFi and placebo. Regarding the TB incidence in patients with RA versus the general population, among the 13 non-randomized controlled studies, 11 reported the local TB incidence rate. Patients with RA treated with non-bDMARDs demonstrated a 3.17-fold increased risk of TB (95% CI 2.12–4.73) versus the general population. By contrast, TNFi treatment was associated with a substantially higher TB risk, with incidence rates 17.07 times greater (95% CI 13.85–21.04) than the general population. The analysis examined approximately 323,709 PY to compare TB risk between patients with RA who received TNFi and those who did not. The results showed that TNFi treatment was associated with a significant 4.03-fold increase in TB incidence rate (95% CI 2.36–6.88). The risk of reactivation of hepatitis B virus (HBV) [58] and HZ [59] following treatment with golimumab and other bDMARDs was evaluated. The incidence and risk factors for HBV reactivation were assessed in participants with RA treated with golimumab (n = 26), abatacept (n = 29), etanercept (n = 25), tocilizumab (n = 25), adalimumab (n = 19), infliximab (n = 17) and certolizumab pegol (n = 11) [58]. At the beginning of treatment, a total of 133 (88%) participants had antibodies against HBV surface antigens (anti‐HBs). Throughout the observation period of 15 months (IQR: 4–34 months), HBV reactivation occurred in seven participants (4.6% of the study population). Five experienced spontaneous resolution with HBV-DNA levels becoming undetectable or remaining below the detection threshold without treatment. The remaining two participants exhibited HBV-DNA levels > 20 IU/ml but achieved viral clearance following treatment. The analysis showed that participants lacking hepatitis B surface antibodies demonstrated a significantly higher risk of HBV reactivation (p = 0.013). Among participants treated with golimumab (n = 26), one participant experienced HBV reactivation. A second study evaluated the incidence and recurrence of HZ in patients (n = 11,720) with seropositive RA who were treated with first-line bDMARDs or tsDMARDs and identified in the Korean Health Insurance Review & Assessment Service database [59]. Throughout the study period, which encompassed 34,702 PY of follow-up, a total of 1686 HZ cases, representing 14.4% of the study population, were identified. These cases included 1372 incident cases (11.7%) and 314 recurrent cases (2.7%). The overall risk of HZ, per first-line treatment, was significantly higher with tofacitinib (p < 0.001), infliximab (p = 0.017), and adalimumab (p = 0.032) versus abatacept. The same analysis with golimumab was not significant (p = 0.692). Independent factors associated with overall risk for HZ included age (p < 0.001), being female (p = 0.032), Charlson comorbidity index (p < 0.001), year of the index date, glucocorticoid use per 1 mg/day increase (p = 0.033), and history of HZ (p < 0.001). Regarding HZ reoccurrence, the risk was only significantly higher for tofacitinib (p < 0.001) compared with abatacept (golimumab: p = 0.268). Collectively, the studies presented above suggest that golimumab treatment was associated with a risk of infection; however, it is low. Risk of Cardiovascular Outcomes CVD is a significant comorbidity and a leading cause of morbidity and mortality [53]. Risk of developing CVD arises from traditional cardiovascular (CV) risk factors, inflammation processes associated with RA, and potential side effects from RA treatments, including corticosteroids and NSAIDs. Several studies evaluated the association between TNFi treatment, including golimumab, and risk of CV outcomes [54, 60, 61]. The effect of golimumab on major adverse cardiac events (MACE) was analyzed using data from 3 RCTs (1 each for RA, PsA, and axSpA) [54]. Golimumab was administered until week 100 for participants with RA (golimumab: n = 395) and the incidence (95% CI) of MACE was 0.8/100 PY (0.4–1.6). By comparison, the incidence (95% CI) of MACE among the combined placebo (RA, PsA, and axSpA: n = 539) group was 1.4/100 PY (0.3–4.0). The results from two studies suggest that TNFi, including golimumab, may be associated with positive CV effects [60, 61]. A meta-analysis of RCTs and observational studies with follow-up ≥ 1 year evaluated the association between CV events (CVEs) and antirheumatic drugs in participants with RA (28 studies, n = 236,525) or PsA/psoriasis (six studies, n = 220,209) [60]. The CV outcomes included all CVEs, myocardial infarction (MI), congestive heart failure (CHF), stroke and/or MACE. TNFi treatment (specific drugs not reported) was associated with significantly reduced risk for all CVEs (p = 0.005), MI (p = 0.04), stroke (p = 0.02), and MACE (p = 0.0003). Treatment with MTX was associated with significantly reduced risk of all CVEs (p = 0.007) and MI (p = 0.01). By contrast, NSAIDs significantly increased the risk of all CVEs (p = 0.04) and stroke (p = 0.01), which may have been specifically related to the effects of rofecoxib. Corticosteroids significantly increased the risk of all CVEs (p < 0.00001), MI (p < 0.00001), CHF (p = 0.006), stroke (p < 0.03), and MACE (p = 0.0010). The effect of golimumab on left ventricle (LV) myocardial dysfunction in participants with RA was evaluated using ultrasound carotid assessment and comprehensive echocardiographic evaluation (conventional and speckle-tracking calculation) [61]. Participants (n = 30) were assessed at baseline and following 18 months of treatment and compared with healthy controls (n = 30). Disease activity was assessed per DAS28. At baseline, 90% of the participants with RA were treated with MTX. None of the participants with RA showed any signs or symptoms of CVD, pulmonary involvement, or any other complication. Further, the mean LV ejection fraction and E/A ratio were not significantly different from controls (p = 0.08 and 0.233, respectively). By contrast, the speckle-tracking analysis revealed a global longitudinal strain deformation in the apical four-chamber view that was significantly lower in participants with RA (p < 0.05). Also, compared with controls, the median right and left common carotid intima-media thickness (cIMT) were significantly greater (both p < 0.05). Following 18 months of golimumab treatment, the median end-systolic LV longitudinal strain was significantly improved (p < 0.01), with a minimal nonsignificant reduction in arterial stiffness and cIMT parameters. Disease activity was also significantly reduced (p < 0.01). In conclusion, LV myocardial longitudinal strain was impaired in participants with RA at baseline and this was reversed following golimumab treatment. The results from the studies above suggest that treatment with golimumab was not associated with poorer CV outcomes and it may be cardioprotective. Risk of Malignancies In RA, prolonged immune dysregulation and the resulting inflammatory response may also lead to an increased risk of cancer [62]. RA has been linked to a higher incidence of non-Hodgkin’s lymphoma, and emerging research suggests an increased risk of lung cancer. RA has also been linked to an increased risk of nonmelanoma skin cancer versus the general population [53]. Two studies evaluated the effect of golimumab on the risk of developing a malignancy [54, 63]. An analysis of pooled safety data from three RCTs evaluated the effect of IV golimumab on malignancies in participants with RA, PsA, or axSpA [54]. Malignancies occurred in three participants in the placebo group (non-small cell lung cancer, esophageal neoplasm, and non-treatment-emergent lung adenocarcinoma). Among the golimumab group, there were eight malignancies that occurred in seven participants. The malignancies included basal cell carcinoma, breast cancer, cervical carcinoma in situ, chronic lymphocytic leukemia, gastric cancer, and colon cancer and one participant with both Bowen’s disease and basal cell carcinoma. In participants with RA who were treated with golimumab, the incidence of malignancies/100 PY was 0.5 (95% CI 0.2–1.1). There were no cases of lymphoma. A retrospective nationwide study compared participants with RA or psoriasis treated with a TNFi including golimumab (n = 1219), adalimumab (3705), infliximab (1766), and etanercept (2195) versus non-users [63]. Pooled data across the two patient populations and four TNFi showed that the risk of site-specific cancer per adjusted hazard ratios (aHR) (95% CIs) were not significantly increased or decreased for colorectal, pancreatic, liver, lung, kidney, breast, and thyroid cancer: 0.92 (0.61–1.38), 1.64 (0.92–2.92), 0.90 (0.53–1.53), 1.00 (0.73–1.37), 1.20 (0.62–2.34), 0.91 (0.62–1.34), and 0.91 (0.66–1.26), respectively. The data were similar for golimumab, collapsed across the patient populations, for the following cancers: colorectal, lung, kidney, breast, and thyroid cancer: 1.02 (0.33–3.17), 0.92 (0.34–2.45), 0.82 (0.26–2.55), and 1.00 (0.42–2.41), respectively. Collectively, the results from these two studies suggest that treatment with TNFi, including golimumab, is associated with a risk of cancer; however, it is low. Risk of Immunogenicity Using data from the Norwegian DMARD study, a longitudinal observational study of adult participants with inflammatory joint disease (RA: n = 20, axSpA: n = 41, PsA: n = 30) who initiated a bDMARD, the frequency of antibodies to golimumab was determined [64]. After 3 months of treatment, six participants (6%) developed antibodies to golimumab (two each across the RA, axSpA, and PsA groups) and showed significantly lower golimumab serum concentrations versus those who tested negative (p = 0.001). Among the six participants who developed antibodies, only one responded to treatment at 3 and 6 months. By contrast, approximately 50% of antibody-negative patients were responders after 3 and 6 months. Risk During Pregnancy and Lactation Studies that evaluated the effect of golimumab during pregnancy and lactation are limited [65]. A recent observational population-based cohort study evaluated pregnancy and infant outcomes following prenatal golimumab exposure [66]. Registries in Finland (2006 to 2018) and Denmark and Sweden (2019) were used to identify patients who received golimumab for rheumatic diseases or ulcerative colitis. Based on drug-exposure, infants were stratified into golimumab, other TNFi, other bDMARDs, and non-bDMARD cohorts. Prevalence of adverse pregnancy outcomes, mortality, major congenital anomalies (MCA), and inpatient infections in the infant’s first year of life was assessed. No stillbirths or deaths occurred among 134 infants with golimumab exposure. MCAs were observed in 4.5% of golimumab-exposed infants, which was comparable to: 6.8% for other TNFi (n = 2273), 10.9% for other bDMARDs (n = 64), 5.5% for non-bDMARD therapies (n = 10,823), and 4.6% for the general population (n = 3,067,204). Hospitalization for infections occurred in 11% of golimumab-exposed infants, similar to the 9% to 11% observed across the other cohorts. A second analysis, using data from a company’s global safety database, found that rates of adverse pregnancy outcomes and MCAs in cases with golimumab exposure were similar to the general population [67]. Cases were medically confirmed exposures during pregnancy or within 3 months prior to conception with a reported pregnancy outcome. A total of 261 reported pregnancies were identified and among these there were 214 (82.0%) live births, 31 (11.9%) spontaneous abortions, 13 (5.0%) induced/elective abortions, two (0.8%) reported intrauterine death/still birth, and one (0.4%) fetal AE in an ongoing pregnancy. Most pregnancies had exposure in the first trimester. Seven congenital anomalies (7/261; 2.7%) were reported, and five were considered major. In conclusion, although the number of golimumab exposures were low across the two studies, the data did not find an increased risk for infants or pregnant women. Consistent with these results, EULAR recommends that all TNFi can be used throughout pregnancy and that treatment before, during, and after pregnancy should aim at remission or low-disease activity [68]. The guideline conveys that current evidence indicates that the TNFi are not associated with an increased risk of congenital malformations, miscarriage, or any other adverse pregnancy outcomes. However, the potential risk of drug therapy for the fetus or child should be weighed against the risk of untreated maternal disease. Lastly, the choice of treatment should be a shared decision between the treating providers and the patient. Regarding lactation, the EULAR guideline conveys that continuation of TNFi treatment is compatible with breastfeeding [68]. The recommendation is based on current evidence that indicates there is minimal transfer of bDMARDs into breast milk and limited systemic absorption by the breastfed child based on the drugs’ physicochemical and pharmacokinetic properties. Conclusions The data from the golimumab pivotal RCTs in participants with active RA demonstrated that treatment effectively reduced the signs and symptoms of the disease, MRI inflammation, and radiographic progression, and improved physical function, HR-QoL, and work productivity. The RCT data also showed that these benefits correlated with a reduction in disease activity. The long-term extension trials demonstrated that the clinical benefits were maintained, and treatment was associated with robust drug survival. The safety data revealed that most participants reported at least one AE, and the most frequent AEs were URI, nasopharyngitis, and bronchitis. The data from the golimumab real-world studies from the past 5 years confirmed the efficacy results from the pivotal RCTs, extension trials, and post hoc analyses. Data from several recent meta-analyses of RCTs and real-world studies indirectly compared the efficacy/effectiveness and safety outcomes following treatment with golimumab or other TNFi. The results did not favor one drug over another as they were equivocal. Regarding comorbidities associated with RA or TNFi treatment, golimumab was not associated with an increased risk of CVD; however, it was associated with an increased risk of infections and cancer, and reactivation of TB. Lastly, there is limited data in pregnancy; however, the rates of adverse outcomes and MCAs were similar to those of the general population.