• DMARDs (synthetic)
• rheumatoid arthritis
• treatment

Patients with rheumatoid arthritis (RA) have a high prevalence of comorbidities.1 This post-hoc analysis investigated the effect of select comorbidities (depression, osteoporosis, hepatic, cardiovascular or pulmonary disorders) on the efficacy and safety of baricitinib 4 mg once daily in patients with moderate-to-severe active RA and an inadequate response to conventional synthetic disease-modifying antirheumatic drugs (csDMARDs).

Data from the placebo-controlled periods of five baricitinib studies2–6 were pooled for baricitinib 4 mg; data for baricitinib 2 mg were not analysed due to low (n=302) patient numbers. Additional data for all baricitinib-treated patients with a median exposure of 2 years were derived from an ongoing open-label, long-term extension (LTE) study that included patients from phase II and III studies (RA-BEYOND; NCT01885078).⁷ Efficacy outcomes were evaluated at week 12 (vs placebo). Interaction of comorbidity-by-treatment was analysed using logistic regression or analysis of covariance. Safety observations up to week 16 (vs placebo) and during the LTE were summarised by the Medical Dictionary for Regulatory Activities preferred term.

Data from 1684 patients (803, baricitinib 4 mg; 881, placebo) from the placebo-controlled periods were analysed. The mean (SD) age was 52.7 (12.1) years, with only 38 (2.3%) patients aged ≥75 years; most patients (1506, 89.4%) were receiving background methotrexate, alone or in combination with another csDMARD. The numbers of patients receiving placebo and baricitinib 4 mg combined (with or without each comorbidity, respectively) were 133/1551 for depression, 247/1437 for osteoporosis, 424/1260 for hepatic disorders, 731/953 for cardiovascular disorders, and 166/1518 for pulmonary disorders. Demographic and clinical characteristics within each comorbidity subgroup were similar between patients randomised to baricitinib or placebo.

Higher proportions of patients achieved all clinical endpoints with baricitinib 4 mg than with placebo at week 12 across subgroups (table 1). Response rates in patients with or without each comorbidity who received baricitinib were generally close to overall response rates, with the exception of depression, where numerically lower response rates were observed in patients with versus without depression and overall (table 1). Responses to baricitinib versus placebo were similar between patients with or without comorbidity within each comorbidity subgroup, even for those with/without depression (p>0.1).

The safety analysis set from the placebo-controlled periods included 1683 patients (802, baricitinib; 881, placebo) with 235.2 and 246.9 patient-years of exposure (PYE) to baricitinib 4 mg or placebo, respectively. Similar proportions of patients experienced ≥1 treatment-emergent adverse events (TEAEs) between baricitinib and placebo across comorbidity subgroups (table 1). The most common TEAEs for the overall population were nasopharyngitis (5.6%, baricitinib vs 5.2%, placebo), upper respiratory tract infections (4.5% vs 3.5%), urinary tract infections (3.5% vs 2.6%) and bronchitis (2.9% vs 2.6%). Serious adverse events (SAEs) and discontinuations were infrequent, occurring at similar rates across subgroups for baricitinib and placebo. SAE and discontinuation rates for baricitinib-treated patients with/without each comorbidity were generally close to overall rates, although numerical differences were observed between patients with depression (lower rates), osteoporosis (higher rates) or pulmonary disorders (higher rates) and those without these comorbidities.

In the LTE study, safety data were available from 3439 patients with a total of 6633 PYE to baricitinib. Overall, the incidence of TEAEs, SAEs, discontinuations and reported deaths was similar between patients with or without comorbidity within each comorbidity subgroup (table 1). Further, the exposure-adjusted incidence rates for all patients were lower in the LTE than during the placebo-controlled period for all safety outcomes assessed. The numerical differences between patients with/without each comorbidity observed after 16 weeks (see above) were not observed in the LTE study.

This is the first analysis assessing the impact of select comorbidities on the efficacy and safety of baricitinib 4 mg. Limitations of this post-hoc analysis include the analysis of patient data derived from randomised clinical trials rather than real-world patient data, such that patients with certain acute or historical conditions, unstable illness, or certain laboratory abnormalities were excluded from the trials (see online supplementary text); the limited number of patients in each comorbidity subgroup and the limited duration of analyses; the non-inclusion of all known comorbidities in RA; the mix of comorbidities included in each subgroup, which could vary in severity and overall impact on patients; the inability to assess adverse events of specific interest and whether comorbidity severity improved or worsened following baricitinib treatment; and the inability to conclude on the efficacy and safety of baricitinib in patients aged ≥75 years and/or exposed to baricitinib 2 mg due to low patient numbers. Planned analyses of safety data from registries or healthcare databases in the USA and Europe are expected to address many of the current limitations.

Baricitinib 4 mg showed similar efficacy and safety during placebo-controlled and LTE observation periods regardless of the presence or absence of select comorbidities in patients with moderate-to-severe active RA and an inadequate response to csDMARDs. No trends for increased risk of safety events related to comorbidity were observed across comorbidity subgroups for baricitinib during a median exposure of 2 years. Further studies are needed to confirm the data presented, which provide a hypothesis only.

Trial registration numbers: NCT01185353, NCT01469013, NCT00902486, NCT01710358 and NCT01721057.