Phase II clinical trials

In 2015 a phase IIb trial investigated the efficacy of baricitinib at 1 mg, 2 mg, 4 mg, or 8 mg vs. placebo [38]. The study involved 301 patients from 69 institutions in 9 countries who had failed prior treatment with methotrexate. The primary endpoint was the proportion of study volunteers in the 4 mg or 8 mg cohort that received a positive result on the American College of Rheumatology 20% (ACR20) score at 12 weeks. The ACR20 is a tool that uses multiple measures to objectively evaluate improvement in rheumatoid arthritis symptoms. This trial showed a significant response to treatment in the combined 4 mg or 8 mg baricitinib cohort as compared to placebo (76% vs. 41% respectively) at 12 weeks of therapy.

Additionally, baricitinib was well tolerated by most of the study participants, and all baricitinib groups showed an increased response compared to placebo in the other secondary endpoints, including ACR50, ACR70, and remission.

At 12 weeks the baricitinib and placebo groups experienced similar proportions of treatment-emergent adverse events. The mean neutrophil count declined in all baricitinib cohorts, but there was no significant decline in mean lymphocyte count as compared to the placebo. Mean low-density lipoproteins and mean high-density lipoproteins also increased in all the baricitinib cohorts in comparison to placebo. This study shows the effectiveness of baricitinib treatment in patients with active rheumatoid arthritis who were unresponsive to methotrexate therapy.

The inhibition of IL-6 signaling by baricitinib may in part explain how the drug affects lipoprotein metabolism and particle distribution, and was explored in a separate analysis of this trial in 2017 [39]. Changes in lipoprotein particle size and particle number, as well as changes in lipid profile, were assessed at weeks 12 and 24 in association with clinical efficacy.

Interestingly, the study found that following 2 weeks of baricitinib, patients experienced dose-related increases in serum lipid levels. Increases in HDL cholesterol, LDL cholesterol, and triglyceride levels were observed and remained elevated through 24 weeks. The observed increase in LDL cholesterol was associated with a shift in LDL particle size to large LDL particles [39].

Furthermore, treatment with baricitinib resulted in an increase in apolipoprotein A-I and a reduction in the serum amyloid A (SAA) content of HDL particles, which rendered HDL particles more efficient for reverse cholesterol transport.

These increased HDL cholesterol levels correlated with improved clinical outcomes at 12 weeks, as evidenced by improvement in Disease Activity Score 28-joint assessment using the Simplified Disease Activity Index [SDAI] and C-reactive protein level (DAS28-CRP) [39]. This association provides support for a potential relationship between increases in HDL levels and reduction of disease activity scores and inflammation in patients with rheumatoid arthritis. No such relationship was observed in patients treated with placebo.

Further evaluation of this trial using population pharmacokinetic/pharmacodynamics models to determine dose/exposure-response relationships was performed to assess the efficacy and safety of different doses of baricitinib for a potential phase III trial in the future [40].

The primary efficacy endpoint assessed was the ACR20/50/70 and the primary safety endpoint assessed was anemia. The results of the study showed a faster onset of ACR20 improvement in the 4 mg and 8 mg dosed groups along with a slight increase in incidence of anemia for the 8 mg dose group.

The evaluation concluded stating that 4 mg QD was the optimal dosing from a risk-benefit standpoint, and that 2 mg QD may potentially be efficacious, however this should be explored further in a phase III trial. Importantly, this study also found no benefit to BID dosing from a safety and efficacy standpoint.

Another phase IIb randomized control trial in 2016 demonstrated similar effectiveness of baricitinib treatment in Japanese patients with rheumatoid arthritis who were concurrently taking methotrexate [41]. The study enrolled 145 patients. The primary endpoint was the ACR20 response rate of patients taking 4 mg or 8 mg daily of baricitinib compared to placebo at 12 weeks. A significantly greater proportion of patients in the 4 mg or 8 mg group responded to treatment relative to placebo (77% vs. 31%) at 12 weeks.

Improvement of symptoms including remission and physical function was demonstrated as early as 2 weeks in the 4 mg and 8 mg group. The adverse effects of baricitinib treatment were limited. The 1 mg, 2 mg, and 4 mg baricitinib group showed similar rates of adverse events as compared to placebo.

However, there was a slightly increased rate of adverse events and abnormal laboratory values in the 8 mg group. There were 3 serious adverse events (SAEs) reported during the study period: cholecystitis in the placebo group, acute pancreatitis in the 2 mg baricitinib group, and cataract in the 8 mg baricitinib group. It is unclear if these SAEs were related to the experimental drug.

The safety of baricitinib in this Japanese cohort showed similar results to other trials with non-Japanese patients, with no malignancies, serious infection, tuberculosis, pneumocystis pneumonia, herpes zoster cases, or GI perforations reported. This trial justifies the need to further study the benefit to risk ratio of 2 mg and 4 mg baricitinib treatment in patients with rheumatoid arthritis concurrently taking methotrexate.

This 12-week study was then extended to a 64-week study [42]. In this extension, patients that were originally randomized to placebo, 1 mg, or 2 mg doses were re-randomized to either 4 mg or 8 mg doses following the conclusion of the 12-week study. However, after analysis of data from other phase II trials occurring at this time, a decision was made to switch patients from 8 mg to 4 mg, so that all patients were taking 4 mg of baricitinib by the conclusion of the extension study. Of the 142 patients who completed the 12-week study, 109 of these patients continued and completed the extension study. During the extension period, patients who were originally assigned to the placebo group noted a significant improvement in ACR20, while those originally treated with baricitinib successfully maintained or improved on the progress made in the first 12 weeks.

No changes in safety of baricitinib were noted with a longer period of treatment. Notably, although no incidences of herpes zoster were noted in the first 12 weeks of the study, 11 patients (7.8%) developed herpes zoster during the extension and dropped out of the study. This was the most common reason for discontinuation of the study overall.

While this may seem remarkable, reactivation of herpes zoster is a common side effect of disease-modifying antirheumatic drug (DMARD) therapy and it is even recommended by the American College of Rheumatology for patients to get the herpes zoster vaccine prior to starting DMARD therapy. This trial highlights the beneficial effects of baricitinib in patients concurrently taking methotrexate and shows no difference in safety and efficacy between short- and long-term dosing of baricitinib.

A phase III RA-BEACON trial

A phase III RA-BEACON trial in 2016 involving 527 patients compared the use of 2 mg or 4 mg of daily baricitinib with placebo at 24 weeks [43]. The primary endpoint used the ACR20 response to determine clinical improvement at 12 weeks. The study population consisted of patients > 18 years old with moderate-to-severe RA who had discontinued prior treatment with conventional tumor necrosis factor inhibitors (TNFis) or biologic disease-modifying antirheumatic drugs (bDMARDs) due to insufficient response or intolerance after > 3 months [44]. The results showed significantly more patients in the 4 mg baricitinib group had an ACR20 response than the placebo group (55% vs. 27%). Additionally, the 4 mg group as compared to the placebo showed a difference in response to the HAQ-DI and DAS28-CRP score, but not the SDAI score.

The rates of adverse events at 24 weeks were higher among the 2 mg and 4 mg group (71% and 77%) relative to the placebo (64%). The most common cause of adverse events were infections, including respiratory infections, bronchitis, and urinary tract infections. However, the rates of serious adverse events were similar throughout 24 weeks: 4% in the placebo group, 10% in the 2 mg group, and 7% in the 4 mg group. The two baricitinib treatment groups were also associated with a drop in neutrophil count versus placebo at 24 weeks (–560, –630 vs. +130). The serum creatinine and LDL levels were already elevated in the baricitinib groups compared to placebo. These results show evidence of clinical improvement with baricitinib treatment in a study population with disease particularly refractory to multiple biologic therapies.

Furthermore, the efficacy of baricitinib demonstrated in RA-BEACON was reflected by clinically significant changes in patient reported outcomes (PROs). These are core measures established by the American College of Rheumatology that assess disease activity in clinical trials and may improve the patient–physician relationship by addressing patient concerns such as onset of drug action, sustainability or risk of relapse, impact on quality of life, and efficacy plateau.

Patient reported outcomes utilize standardized scales (PROs) measuring fatigue, duration of morning joint stiffness, severity of worst joint pain, physical and mental quality of life, Short Form-36, work productivity and impairment of daily activities, and physical function. RA-BEACON used a minimum clinically important difference (MCID) to assess the clinical relevance of changes to the aforementioned scores. Of note, baseline PROs were similar across treatment groups and represented significant disease burden. PROs were assessed at baseline as well as weeks 1, 2, 4 and every 4 weeks thereafter to week 24 [44]. Over 24 weeks, the majority of PROs significantly improved among patients receiving baricitinib compared with placebo, with patients receiving baricitinib 4 mg demonstrating a faster and greater magnitude of change than the baricitinib 2 mg group. In addition, the PRO improvements were not influenced by the type or number of previous bDMARDs used.

Furthermore, the groups receiving baricitinib reported significant improvement in morning joint stiffness (MJS) duration and fatigue as well as in regular activity (WPAI-RA), EQ-5D scores, and pain when compared to placebo. In contrast, no significant differences were observed for the SF-36 MCS measure between patients treated with baricitinib compared with placebo [44].

A post-hoc analysis was performed following the RA-BEACON trial to examine sub-groups of the patient population and determine if history of prior bDMARD use had an effect on the efficacy and safety of baricitinib [45]. The study examined ACR20s in multiple sub groups of the 527 patients and found no interactions with age, weight, geographic region, disease duration, seropositivity, corticosteroid use, number of prior bDMARDs used, number of prior TNF inhibitors used, or type of TNF inhibitors used, with 2 mg or 4 mg baricitinib. Notably, the only change in safety was an increased risk of infections and serious adverse events in patients who had taken three or more bDMARDs prior to baricitinib treatment. However, the small number of patients in the subgroup with a history of three or more bDMARD use limits the significance of this finding.

A phase III RA-BUILD trial

The phase III RA-BUILD studied the effectiveness of 2 mg or 4 mg of baricitinib therapy compared to placebo in 684 randomized patients who also did not respond to conventional synthetic disease-modifying antirheumatic drugs (csDMARD) [46]. The study population was naïve biologic DMARD treatment. The primary endpoint was the ACR20 at 12 weeks. Most of the study population was receiving simultaneous csDMARD therapy and only 7% of the patients were not taking concomitant csDMARD therapy. A statistically significant increase in the ACR20 response rate for patients taking 4 mg of baricitinib (62%) compared to placebo (39%) was seen at 12 weeks. There was no difference in treatment effect of patients taking concomitant csDMARDs or no csDMARD. Additionally, a reduction in radiographic progression of structural joint damage was observed at 24 weeks in both groups receiving baricitinib therapy relative to placebo. There was no significant difference in the rate of adverse events between both baricitinib groups (67%, 71%) and placebo (71%).

The baricitinib groups showed an increased incidence of low neutrophil count and elevated LDL and HDL values. However, similar rates of serious infections were observed in placebo, 2 mg and 4 mg baricitinib groups (2%, < 1%, and 2%). Both baricitinib groups also exhibited a small increase in the serum creatinine. The results of this trial demonstrated symptomatic relief and a beneficial effect on joint damage without significant side effects in patients who failed to benefit from prior csDMARD therapy.

A phase III RA-BEAM trial

In 2017 the RA-BEAM phase III trial was the first trial to investigate the use of baricitinib versus another biologic, adalimumab, in patients worldwide taking background methotrexate therapy [47]. Adalimumab is an anti-tumor necrosis factor a monoclonal antibody, a biologic that was used as standard of care in combination with methotrexate for patients with moderate to severe RA. The primary endpoint was the response to ACR20 at 12 weeks. A significant increase in the response rate was demonstrated in 4 mg of daily baricitinib use compared to placebo (70% vs. 40%) as well as compared to 40 mg of adalimumab (70% vs. 61%). Structural joint damage was also decreased at 24 weeks in both the baricitinib and adalimumab treatment groups relative to placebo. Adverse events through week 24 occurred at a rate of 60% in the placebo group, 71% in the baricitinib group, and 68% in the adalimumab group. The baricitinib and adalimumab groups both exhibited a decrease in neutrophil count and an elevation in aminotransferase, creatinine, LDL, and HDL levels as compared to placebo.

This groundbreaking trial showed the superiority of baricitinib plus methotrexate to what was considered standard of care treatment at the time, adalimumab plus methotrexate.

A phase III RA-BEGIN trial

The 52-week phase III RA-BEGIN trial was the first to investigate the difference between methotrexate monotherapy, baricitinib monotherapy, and the combination of baricitinib and methotrexate therapy [48]. The primary endpoint assessed the ACR20 response rate at 24 weeks between baricitinib monotherapy and methotrexate monotherapy. At 24 weeks the response rate of baricitinib monotherapy (77%) was higher than methotrexate monotherapy (62%). Symptomatic improvement was observed in the baricitinib group as quickly as 1 week in comparison to methotrexate alone.

Similar results to baricitinib monotherapy was seen in baricitinib plus methotrexate (MTX) combination therapy. The baricitinib plus MTX group showed significant superior radiographic benefit relative to MTX monotherapy, but not baricitinib monotherapy. At 52 weeks the rates of serious adverse events were similar between MTX monotherapy, baricitinib monotherapy, and baricitinib plus MTX combination therapy (10%, 8%, and 8%, respectively). There were more non-serious infections recorded in the baricitinib plus MTX group than either baricitinib or MTX monotherapy.

This trial supported the known benefits of MTX monotherapy in RA patients, while demonstrating superior outcomes in patients taking baricitinib alone or in combination with MTX. Additional benefit in radiographic evidence and measures of inflammation was shown in baricitinib plus MTX compared to baricitinib alone. However, combination therapy increased the risk of adverse events, such as non-serious infection.

Furthermore, RA-BEGIN utilized PROs to assess the safety and efficacy of baricitinib in adults with moderate-to-severe RA. The study found comparable effects on PRO improvements in the baricitinib monotherapy and baricitinib +MTX groups when compared to MTX monotherapy, with both baricitinib regimens appearing consistently more effective than MTX alone. Over 24 weeks, the majority of PROs significantly improved among patients receiving baricitinib compared with placebo, with patients receiving baricitinib 4 mg demonstrating a faster and greater magnitude of change than the baricitinib 2 mg group. In addition, the PRO improvements were not influenced by the type or number of previous bDMARDs used. Interestingly, the majority of PRO measures of joint pain, tiredness, duration of MJS, pain, fatigue, physical function, HRQOL, and PtGA improved to a greater extent in many or all time points measured in patients in both baricitinib groups compared to MTX monotherapy [48].

Furthermore, the groups receiving baricitinib reported significant improvement in MJS duration and fatigue as well as in regular activity (WPAI-RA), EQ-5D scores, and pain when compared to placebo. In contrast, no significant differences were observed for the SF-36 MCS measure between patients treated with baricitinib compared with placebo [44]. These statistically significant improvements in PROs suggest that baricitinib alone or combined with MTX may be an alternative therapy for patients for whom MTX monotherapy is undesirable.

Of note, a post-hoc analysis was performed following phase 3 of the RA-BEGIN study evaluating structural damage progression (as measured via radiographs of the hands and feet at weeks 12, 24 and 52) between each of the treatment groups [49]. The study looked particularly at two markers for RA disease activity, the Disease Activity Score for 28-joint count with serum high-sensitivity C-reactive protein (DAS28-hsCRP) and Simplified Disease Activity Index (SDAI). Typically, acceptable management of RA symptoms is indicated by a DAS28-CRP score < 3.2 and a SDAI score < 11. This study found that if patients were able to stay below those thresholds, the risk of structural damage was lower in those treated with baricitinib alone or combination therapy with MTX. In patients with a DAS28-CRP > 3.2 or a SADI > 11, only combination therapy with MTX and baricitinib was sufficient to decrease the risk of structural damage.

The results of the four phase three trials assessing the efficacy and safety of baricitinib (RA-BEGIN, RA-BEAM, RA-BUILD, and RA-BEACON) showed significant improvement in ACR20 with 4 mg baricitinib when compared to placebo, MTX, and adalimumab standard of care treatment at 12 or 24 weeks. A subgroup analysis was performed in July of 2018 to see if the 394 Japanese patients treated across these trials showed similar benefit to treatment as the general study population [50]. The authors of this subgroup analysis found that across all four trials, equivalent improvements were made in ACR20 in Japanese patients when compared to the overall study population. The study did highlight differences in the Japanese population from other participants in these studies, including a lower body weight and lower average dose of MTX. However, an equal response to baricitinib was noted across all trials despite these differences.

The authors concluded that while this data is promising, long-term effects of baricitinib in this study population have yet to be published, noting that 328 of the 394 Japanese patients are enrolled in the 84 month long RA-BEYOND study that is expected to conclude in 2024.

Trials testing drug-drug interactions

Baricitinib undergoes active renal tubular secretion, and is predominantly eliminated from the body unchanged in urine [51]. Baricitinib’s secretion is dependent on the basolaterally expressed OAT3 transporter and the apically expressed P-GP, BRCP, and MATE2-K transporters [51]. In the presence of probenecid, a strong OAT3 inhibitor, CL_r and CL/F of baricitinib decreased 69% and 51% respectively, and the AUC_(0–∞) of baricitinib doubled in healthy subjects [51].

Additionally, diclofenac and ibuprofen, OAT3 inhibitors with less inhibition potential than probenecid, were evaluated using physiologically based pharmacokinetic (PBPK) modeling and in vitro inhibition data to predict the inhibition potential for the OAT3 mediated secretion of baricitinib. The aforementioned study used the in vitro IC50 value of 4.4 µM to reproduce the renal clearance of baricitinib and the inhibitory effect of probenecid. Using ibuprofen and diclofenac in vitro IC50 values of 4.4 µM and 3.8 µM toward OAT3, 1.2 and 1.0 AUC_(0–∞) ratios of baricitinib were predicted, suggesting that co-administration of diclofenac or ibuprofen is safe as it does not cause clinically relevant drug-drug interactions with baricitinib [51]. These predictions are relevant for furthering our understanding of the safety profile of new RA therapies.

To sum up baricitinib, currently a second line drug for RA, is a novel oral drug that is based on the above mentioned research. It is a once-a-day drug that is orally available and targets the JAK/STAT signaling pathway by acting as a JAK inhibitor. It is approved in Europe, Japan and the US for treatment of RA after failure of treatment with methotrexate (MTX) and TNF inhibitors. Acting upstream in the JAK/STAT pathway, baricitinib inhibits specifically JAK1 and JAK2 and decreases the expression of IL-6 and TNF-α, as well as other pro-inflammatory cytokines such as IL-10, IL-8, IFN-α. While decreasing inflammation, baricitinib interferes with other pathways involved with JAK/STAT signaling and may induce elevated levels of LDL and HDL and decreased hemoglobin production.

Phase I and II trials conducted so far proved the safety and efficacy of baricitinib in RA, including with concurrent use with MTX. The accepted dose currently is that of daily 4 mg which appears to balance the side effects (anemia, hypercholesterolemia, re-activation of latent infection) with the benefit (as measured by ACR20) and improvement in patients’ symptoms at 12-weeks. The RA-BEACON Phase III trial re-iterated these findings, found no safety increased in 2 mg over 4 mg dosing and proved that 4 mg dosing significantly improved symptoms after 12-weeks of treatment, as measured by ACR20. Post-hoc analysis pointed to extensive prior treatment as an increased risk of latent infection reactivation, but otherwise could not point to any other safety or efficacy factors.

The RA-BUILD tested 2 mg and 4 mg dosing and used ACR20 at 12-weeks for end point. The results, while not as impressive as RA-BEACON, similarly demonstrated significant relief of symptoms in the baricitinib over placebo, with or without concurrent DMARD use. Importantly, this study also showed decreased joint damage at the 24-weeks mark. RA-BEAM and RA-BEGIN compared baricitinib to MTX and other biologics and found some evidence to support its used over alternative treatments and not just in addition to current DMARD treatment.