Article Text
PDF
Abstract
Objectives To update the long-term safety profile of filgotinib, a Janus kinase-1 preferential inhibitor, in patients with moderate-to-severe rheumatoid arthritis.
Methods Data from seven trials were integrated (NCT01888874, NCT01894516, NCT02889796, NCT02873936, NCT02886728, NCT02065700 and NCT03025308). Patients received once-daily filgotinib 100 mg or 200 mg. Exposure-adjusted incidence rates (EAIRs)/100 patient-years of exposure (PYE) were calculated for treatment-emergent adverse events (TEAEs). Post hoc analyses assessed patients aged <65 and ≥65 years.
Results Patients (N=3691) received filgotinib for a median (maximum) of 3.8 (8.3) years (12 541 PYE). Rates of TEAEs of interest: serious infections, malignancies, major adverse cardiovascular events (MACE) and venous thromboembolism were stable over time and comparable between doses. In the overall population, numerically lower EAIR (95% CI)/100 PYE of herpes zoster was observed for filgotinib 100 mg versus 200 mg (1.1 (0.8 to 1.5) vs 1.5 (1.2 to 1.8)). Incidence of serious infections, herpes zoster, MACE, malignancies and all-cause mortality was higher in patients aged ≥65 versus <65 years. In patients aged ≥65 years, EAIRs (95% CI)/100 PYE for non-melanoma skin cancer (NMSC) (0.4 (0.1 to 1.1) vs 1.4 (0.8 to 2.2)), malignancies excluding NMSC (1.0 (0.5 to 1.9) vs 2.0 (1.3 to 2.9)) and all-cause mortality (1.3 (0.7 to 2.2) vs 1.6 (1.0 to 2.5)) were numerically lower for filgotinib 100 mg versus 200 mg.
Conclusions In the overall population, TEAEs of interest were stable over time and similar between filgotinib 100 mg and 200 mg dose groups, except for herpes zoster. A dose-dependent relationship between malignancies and all-cause mortality was suggested in patients ≥65 years old.
- Antirheumatic Agents
- Arthritis, Rheumatoid
- Therapeutics
Data availability statement
Data are available upon reasonable request. Anonymised individual patient data will be shared upon request for research purposes, dependent upon the nature of the request, the merit of the proposed research and the availability of the data and their intended use. The full data sharing policy for Gilead Sciences, Inc., can be found at https://www.gilead.com/about/ethics-and-code-of-conduct/policies. Data sharing policy for Galapagos NV can be found at https://www.clinicaltrials-glpg.com/us/en/data-transparency.html.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Filgotinib is an oral Janus kinase (JAK)-1-preferential inhibitor approved for the treatment of moderate-to-severe rheumatoid arthritis (RA).
In a previous long-term integrated analysis of clinical trial data over a median (maximum) of 1.6 (5.6) years of exposure, the safety and tolerability of filgotinib in the overall RA population were similar between the 100 mg and 200 mg dose groups, with a lower incidence of herpes zoster with the lower dose.
The summaries of product characteristics of JAK inhibitors for inflammatory disease have been recently updated; in patients aged 65 years or older, those with a history of atherosclerotic cardiovascular disease or other cardiovascular risk factors (such as current or past long-time smokers), or with malignancy risk factors (eg, current malignancy or history of malignancy), JAK inhibitors should only be used if no suitable treatment alternatives are available.
WHAT THIS STUDY ADDS
These results provide an update on the long-term integrated safety data from seven clinical trials over a median (maximum) treatment duration of 3.8 (8.3) years.
In the overall population, incidence rates of treatment-emergent adverse events were stable over time with no dose-dependent relationships observed, except for herpes zoster, which was lower in the filgotinib 100 mg versus 200 mg dose group.
Rates of malignancies and all-cause mortality were numerically lower in patients aged ≥65 years receiving filgotinib 100 mg versus 200 mg; rates of major adverse cardiovascular events, thromboembolic events and serious infections were similar for both dose groups in those ≥65 years of age.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Analyses of long-term integrated safety data demonstrate that filgotinib was generally well tolerated, with no new safety findings identified in the overall RA population. A dose-related trend towards increased incidence of malignancies and all-cause mortality in patients aged ≥65 years was observed. Therefore, the recommended filgotinib dose for patients aged ≥65 years with RA is 100 mg once daily. This may be escalated to 200 mg once daily in case of insufficient disease control and for long-term treatment, the lowest effective dose should be used.
Introduction
Rheumatoid arthritis (RA) is a persistent inflammatory condition with no cure; patients rely on long-term treatment strategies.1 In the past two decades, there has been an increase in the number of available treatments for patients with RA, with orally administered targeted synthetic disease-modifying antirheumatic drugs being the most recently approved treatment options.
Filgotinib is a preferential Janus kinase (JAK)-1 inhibitor approved for patients with moderate-to-severe active RA in the European Union, UK and Japan. The efficacy and safety of filgotinib have been evaluated in the phase 2 DARWIN and phase 3 FINCH trials, including their respective long-term extensions (LTEs).
Previous long-term integrated safety data analyses from the DARWIN and FINCH trials over a median (maximum) exposure of 1.6 (5.6) years and 2.2 (6.8) years reported that the safety and tolerability of filgotinib 100 mg and 200 mg doses were broadly similar.2 3 However, small numerical differences in the rates of some treatment-emergent adverse events (TEAEs) were identified. It was shown that infections and serious infections were numerically higher in patients receiving filgotinib 100 mg compared with filgotinib 200 mg. The rate of herpes zoster infection was numerically lower in the filgotinib 100 mg versus 200 mg group and remained stable over time.2 3 Rates of venous thromboembolism (VTE) were low, with malignancies and major adverse cardiovascular events (MACE) demonstrating low and similar rates across the filgotinib doses. The previous integrated safety analysis also reported a numerically lower incidence of mortality in patients receiving the lower filgotinib dose (0.3/100 patient-years of exposure (PYE)) than in those receiving the higher dose (0.5/100 PYE), though CIs overlapped.2
Recently, differentiated safety risks were reported in Oral Rheumatoid Arthritis Trial (ORAL) Surveillance, an events-driven safety study of the JAK inhibitor tofacitinib versus tumour necrosis factor (TNF) inhibitors in patients aged 50 years or older, and with at least one cardiovascular (CV) risk factor.4 ORAL Surveillance identified an increased risk of malignancies (HR 1.48) and MACE (HR 1.33) in patients receiving tofacitinib compared with TNF inhibitors, prompting regulatory agencies to conduct an extensive safety review of all JAK inhibitors. The review by the European Medicines Agency safety committee resulted in updates to the summary of product characteristics (SmPC) for JAK inhibitors, including filgotinib. This states that filgotinib should only be used if no suitable treatment alternatives are available in patients who are 65 years of age or older, have a history of atherosclerotic CV disease or other CV risk factors (such as current or past long-term smokers) or with malignancy risk factors (eg, current malignancy or history of malignancy). In patients with known VTE risk factors, other than CV and malignancy risk factors, JAK inhibitors should be used with caution. For long-term treatment, the lowest effective dose should be used.5
Following the results of the ORAL Surveillance study and subsequent filgotinib SmPC updates, a longer-term analysis of filgotinib safety data is warranted. The current analyses provide an update on the long-term safety profile of filgotinib in patients with RA over a median (maximum) treatment duration of 3.8 (8.3) years. Subgroup analyses using a cut-off age of 65 years are reported to align with the recent SmPC updates.5
Methods
Study design
Patient-level integrated data were reported from the phase 2 DARWIN 1 and 2 trials (NCT01888874, NCT01894516), phase 3 FINCH 1, 2 and 3 trials (NCT02889796, NCT02873936, NCT02886728), and the respective phase 2 and phase 3 LTE studies DARWIN 3 (NCT02065700) and FINCH 4 (NCT03025308). Data from the ongoing LTEs were included through 2 May 2022 (DARWIN 3) and 6 May 2022 (FINCH 4). Data for MACE and VTE only include events positively adjudicated by an independent committee (adjudication cut-off 3 April 2022).
Patient and public involvement
Patients were not involved in the design, conduct or reporting of these studies. Patient recruitment was performed by individual sites, with all patients providing written, informed consent.
Analysis set
The long-term, as-treated analysis set included data from patients enrolled in the DARWIN/FINCH trials who received at least one dose of once-daily filgotinib 200 mg or 100 mg. Patient data were included from both the original treatment received in the parent studies, and after re-randomisation to filgotinib in LTE studies; as such, patients’ exposure time may contribute to more than one treatment group. Post hoc subgroup analyses were performed in patients aged <65 and ≥65 years.
Events were recorded for the treatment and dose received by patients at the time of the event. TEAEs are defined as events with onset during treatment or up to 30 days after treatment. TEAEs of interest included MACE, VTE, arterial systemic thromboembolism (ASTE), non-melanoma skin cancer (NMSC), malignancies excluding NMSC, herpes zoster, serious infections (including COVID-19) and all-cause mortality (all deaths in the long-term, as-treated population over time, including those occurring off study drug). Adverse events were coded according to MedDRA V.25.0. Grade 3 and 4 laboratory abnormalities were reported and graded using the Common Terminology Criteria for Adverse Events (CTCAE) V.4.03.
Statistical analysis
Descriptive statistics were used to summarise baseline demographics and disease characteristics, TEAEs and laboratory safety data. Filgotinib PYE for each of the exposure groups (and age subgroups) were calculated as (last dose date – first dose date + 1)/365.25. Exposure-adjusted incidence rates (EAIRs) per 100 PYE, censored at the time of the first event, were determined for TEAEs of interest and laboratory safety data. Analysis used censored PYE, which is the PYE until the onset of the first event if the patient has the event of interest or the total PYE for patients without the event. For patients without an event, exposure was censored at the data cut-off date for ongoing studies or last patient contact for completed or discontinued patients. Exact 95% CIs of the EAIRs were calculated based on the Poisson distribution.6
Changes in EAIRs over time were assessed at 6-monthly intervals of filgotinib exposure up to 240 weeks. Over 264 weeks, time to the first event was analysed using the Kaplan-Meier method, calculated as (onset date of the first event – first dose date + 1).
Results
Patient population and exposure
In the overall RA population, baseline demographics and disease characteristics were well balanced between the filgotinib 100 mg and 200 mg dose groups. For patients receiving conventional synthetic (cs) disease-modifying antirheumatic drugs (DMARDs), a higher percentage received methotrexate in the filgotinib 100 mg versus 200 mg group (66.8% vs 53.8%) (online supplemental table 1). Additional baseline demographics and disease characteristics have been reported previously.2
Supplemental material
At data cut-off (May 2022), 1647 patients received filgotinib 100 mg for 4532.4 PYE, and 2267 patients received filgotinib 200 mg for 8008.6 PYE. The total as-treated population included 3691 patients with 12541.0 PYE (online supplemental table 2). Median (maximum) filgotinib exposure was 3.3 (7.8) years for patients receiving filgotinib 100 mg, 3.8 (8.3) years for patients receiving filgotinib 200 mg and 3.8 (8.3) years in the pooled filgotinib group (online supplemental table 2).
At baseline, approximately 19% of patients were aged ≥65 years (online supplemental table 1), comprising 326 and 407 patients in the filgotinib 100 mg and 200 mg dose groups, respectively (online supplemental table 3). Baseline demographics were generally well balanced in patients aged <65 and ≥65 years in both dose groups (online supplemental table 3). A higher percentage of patients received non-study csDMARDs in the filgotinib 100 mg versus 200 mg dose group in both age categories. Regionally, patients from North America comprised a greater proportion of patients aged ≥65 years than those aged <65 years in both filgotinib dose groups.
Patients aged ≥65 years had higher proportions of CV risk factors at baseline than those aged <65 years. Approximately 40% of patients aged <65 years had at least one CV risk factor, compared with up to approximately 75% of patients aged ≥65 years.
Summary of TEAEs
As summarised in table 1, EAIRs (95% CI)/100 PYE for any TEAE were 110.5 (104.6 to 116.7) and 96.0 (91.9 to 100.3) for the filgotinib 100 mg and filgotinib 200 mg dose group, respectively. The most common TEAEs were nasopharyngitis, upper respiratory tract infection, urinary tract infection, hypertension and COVID-19 (online supplemental table 4).
Summary of EAIRs of TEAEs in pooled parent and ongoing long-term extension clinical trials in RA
In the long-term, as-treated population, EAIRs of Grade ≥3 TEAEs and serious adverse events (SAEs) were numerically higher with filgotinib 100 mg versus filgotinib 200 mg (table 1). In the overall RA population, EAIRs of TEAEs leading to premature treatment discontinuation were numerically lower in the filgotinib 100 mg versus 200 mg dose group; the TEAEs most commonly leading to discontinuation were Mycobacterium tuberculosis complex test positive (EAIR (95% CI)/100 PYE: 0.45 (0.34 to 0.59)) (no cases of active tuberculosis were reported), COVID-19 (EAIR (95% CI)/100 PYE: 0.27 (0.19 to 0.38)), pneumonia (EAIR (95% CI)/100 PYE: 0.24 (0.16 to 0.34)) and COVID-19 pneumonia (EAIR (95% CI)/100 PYE: 0.19 (0.12 to 0.28)).
EAIRs (95% CI)/100 PYE for any TEAE were generally higher in patients aged ≥65 years (120.4 (111.1 to 130.2)) versus <65 years (97.9 (94.2 to 101.7)). Patients aged ≥65 years also had higher EAIRs (95% CI)/100 PYE of serious TEAEs (11.3 (9.9 to 13.0) vs 5.5 (5.0 to 6.0)) and TEAEs leading to treatment discontinuation (7.3 (6.2 to 8.6) vs 4.1 (3.7 to 4.5)).
TEAEs of interest
Over 240 weeks, EAIRs of TEAEs of interest remained stable when measured across 6-monthly intervals (figure 1).
EAIRs of TEAEs: 6-monthly intervals. EAIR, exposure-adjusted incidence rate; MACE, major adverse cardiovascular events; NMSC, non-melanoma skin cancer; PYE, patient-years of exposure; QD, once daily; TEAE, treatment-emergent adverse event; VTE, venous thromboembolism.
Infections
In the overall RA population, EAIRs (95% CI)/100 PYE of serious infections were comparable between filgotinib dose groups: 2.2 (1.8 to 2.7) for 100 mg and 1.9 (1.6 to 2.2) for 200 mg (table 2). The EAIR of serious infections remained stable over time (figure 1). Other than COVID-19, the most common serious infections were pneumonia, cellulitis and acute pyelonephritis.
Frequencies and EAIRs of TEAEs in pooled parent and ongoing long-term extension clinical trials in RA
EAIRs (95% CI)/100 PYE of herpes zoster were numerically lower in the filgotinib 100 mg versus 200 mg dose group, although 95% CIs overlapped (1.1 (0.8 to 1.5) vs 1.5 (1.2 to 1.8); table 2). Long-term data indicate that EAIRs of herpes zoster infection did not vary over time (figure 1). Rates of herpes zoster infections in the placebo-controlled period have previously been reported for these studies.2 Most cases of herpes zoster were mild or moderate. Eight patients had treatment-emergent SAEs of herpes zoster, with five cases leading to treatment interruption and two cases to permanent treatment discontinuation; one SAE of herpes zoster did not result in a change in the study drug and all events resolved.
For the overall population, in the filgotinib 100 mg and 200 mg dose groups, EAIRs (95% CI)/100 PYE of TEAEs of COVID-19 were 3.3 (2.8 to 3.9) and 3.2 (2.8 to 3.6) (table 2). SAEs of COVID-19 occurred in 72 (2.0%) patients, with incidence rates the same between dose groups (EAIR (95% CI)/100 PYE was 0.6 (0.4 to 0.8) for both filgotinib 100 mg and 200 mg).
In patients aged ≥65 years, EAIRs (95% CI)/100 PYE of serious infections were higher than in patients aged <65 years (2.9 (2.2 to 3.7) vs 1.8 (1.5 to 2.1); table 3). EAIRs (95% CI)/100 PYE of herpes zoster were also higher in the older age subgroup than in younger patients (2.1 (1.5 to 2.8) vs 1.2 (1.0 to 1.4); table 3).
Frequencies and EAIRs of TEAEs in patients aged <65 and ≥65 years
There was no dose-dependent relationship between the filgotinib 100 mg and 200 mg dose for EAIRs of serious infections in the older population. EAIRs (95% CI)/100 PYE of herpes zoster in patients aged ≥65 years were numerically lower in the filgotinib 100 mg versus filgotinib 200 mg dose group (1.8 (1.0 to 2.9) vs 2.3 (1.5 to 3.3)), with overlapping CIs (table 3). For COVID-19, EAIRs were similar in patients aged <65 and ≥65 years in the pooled population (table 3), with no dose-dependent difference between the dose groups in either age subgroup.
CV events—MACE and myocardial infarction
EAIRs of MACE were low across the filgotinib dose groups and did not vary over time (figure 1); EAIRs (95% CI)/100 PYE of MACE were 0.5 (0.3 to 0.7) and 0.3 (0.2 to 0.5) in the filgotinib 100 mg and 200 mg dose group, respectively (table 2). In the age subgroup analyses, the EAIR of MACE was higher in patients aged ≥65 years than in younger patients; EAIRs (95% CI)/100 PYE for MACE were the same in patients aged ≥65 years who received filgotinib 200 mg (1.0 (0.5 to 1.7)) or filgotinib 100 mg (1.0 (0.5 to 1.9)) (table 3). EAIRs of myocardial infarction were low, with overlapping 95% CIs between dose groups (tables 2 and 3).
CV events—thromboembolic events
For TEAEs of VTE and ASTE, EAIRs were low and comparable across filgotinib dose groups (table 2). EAIRs of VTE did not vary over time in long-term analysis (figure 1). There were no differences in EAIRs of VTE between the age subgroups and no dose-dependent differences.
Malignancies and NMSC
In the pooled filgotinib dose group, EAIRs (95% CI)/100 PYE of NMSC and malignancies excluding NMSC were 0.3 (0.2 to 0.4) and 0.7 (0.6 to 0.9), respectively; EAIRs for malignancies were similar for filgotinib 100 mg and 200 mg (table 2) and stable over time (figure 1). EAIRs (95% CI)/100 PYE were higher in patients aged ≥65 years than in younger patients for both NMSC (1.0 (0.6 to 1.5) vs 0.1 (0.1 to 0.2)) and malignancies excluding NMSC (1.6 (1.1 to 2.2) vs 0.5 (0.4 to 0.7)) (table 3). Numerically lower EAIRs (95% CI)/100 PYE were observed with filgotinib 100 mg versus 200 mg for both NMSC (0.4 (0.1 to 1.1) vs 1.4 (0.8 to 2.2)) and malignancies excluding NMSC (1.0 (0.5 to 1.9) vs 2.0 (1.3 to 2.9)) in patients aged ≥65 years (table 3).
All-cause mortality
EAIRs (95% CI)/100 PYE of all-cause mortality (including deaths beyond treatment-emergent deaths) were 0.6 (0.4 to 0.8) and 0.7 (0.5 to 0.9) in the filgotinib 100 mg and 200 mg dose groups, respectively (table 2); all-cause mortality rate slightly increased over time (figure 1). In total, 83 fatal cases were reported with 81 treatment-emergent deaths, and the most common causes of mortality were infections (including COVID-19), cardiac disorders and malignancies (online supplemental table 5).
In subgroup analyses of the pooled dose groups, EAIRs (95% CI)/100 PYE of all-cause mortality were higher in patients aged ≥65 years versus <65 years (1.5 (1.0 to 2.1) vs 0.5 (0.4 to 0.6)). Incidence of all-cause mortality was numerically lower with filgotinib 100 mg (EAIR (95% CI) 1.3 (0.7 to 2.2)/100 PYE) versus filgotinib 200 mg (EAIR (95% CI) 1.6 (1.0 to 2.5)/100 PYE) in patients aged ≥65 years (table 3).
There were 26 COVID-19 events leading to death (online supplemental table 6). In the overall population, age and baseline body mass index were higher and medical history of hypertension was more common in patients who died with a TEAE of COVID-19 than in the rest of the study population (online supplemental table 7). In patients aged ≥65 years, EAIRs of death related to COVID-19 were comparable between dose groups and similar to that in the overall population (online supplemental table 6).
Laboratory abnormalities
The rate for decreases in lymphocytes was lower in the filgotinib 100 mg versus 200 mg group (online supplemental table 8). Between the two dose groups, frequencies and EAIRs of CTCAE Grade 3 or 4 laboratory abnormalities were similar for anaemia, decreased platelets, decreased neutrophils, increased alanine aminotransferase, increased aspartate aminotransferase, increased creatinine, hyperglycaemia, hypophosphataemia, hyperkalaemia and hyponatraemia (online supplemental table 8).
Discussion
This updated analysis of filgotinib integrated safety data from seven clinical trials, including LTEs, evaluated 3691 patients with RA, with a median (maximum) treatment duration of 3.8 (8.3) years. Overall, filgotinib was found to be well tolerated and have an acceptable safety profile. In the long-term, as-treated population, the most common TEAEs reported were nasopharyngitis, upper respiratory tract infection and urinary tract infection. EAIRs of Grade ≥3 TEAEs and SAEs were numerically higher with filgotinib 100 mg versus filgotinib 200 mg, while EAIRs of TEAEs leading to treatment discontinuation were numerically lower with filgotinib 100 mg versus filgotinib 200 mg; this differs from the findings of the previous long-term safety analysis where the CIs of these EAIRs completely overlapped.2 3 In the overall RA population, the incidence of TEAEs of interest were stable over 240 weeks of exposure and a dose-dependent relationship was only observed for herpes zoster.
The EAIRs of serious infections reported here in patients receiving filgotinib 100 mg or 200 mg are 2.2/100 PYE and 1.9/100 PYE, respectively; those reported in studies of other JAK inhibitors are 2.5–6.2/100 PYE.7–10 In contrast to the previous long-term analysis, where EAIRs (95% CI)/100 PYE for serious infections were numerically different between dose groups (100 mg: 3.1 (2.1 to 4.5); 200 mg: 1.6 (1.2 to 2.1)),2 the current longer-term analysis shows the difference between rates of serious infections has reduced for patients receiving filgotinib 100 mg versus 200 mg.
A lower EAIR of herpes zoster was observed in patients receiving filgotinib 100 mg versus 200 mg. Of note, the difference between the EAIRs of herpes zoster between the dose groups reported here (1.1/100 PYE and 1.5/100 PYE for filgotinib 100 mg and 200 mg, respectively) was smaller than that reported in the previous filgotinib long-term analysis (1.1/100 PYE and 1.8/100 PYE, respectively).2 Rates of herpes zoster reported in other JAK inhibitor studies are 2.8–7.0 per 100 PYE.7 8 10 Consistent with the earlier long-term analysis, and data from other JAK inhibitor studies, most cases of herpes zoster were not serious.7 8 10 Further studies are needed to investigate the interaction between JAK inhibitor selectivity and safety.
Rates of MACE were similar between filgotinib doses in the overall population. The EAIR of all-cause mortality was 0.7/100 PYE (filgotinib 100 mg: 0.6/100 PYE; filgotinib 200 mg: 0.7/100 PYE), which is slightly higher than reported in the previous analysis of these studies (100 mg: 0.3/100 PYE; 200 mg: 0.5/100 PYE), perhaps reflecting the impact of the COVID-19 pandemic,2 but consistent with rates observed for other RA treatments.7 11 12 In the previous analysis of these studies, EAIRs of all deaths (including those occurring off study drug) were numerically lower in patients receiving filgotinib 100 mg versus 200 mg.2 The current analysis shows similar rates of all-cause mortality between the dose groups in the total population, with lower rates for filgotinib 100 mg versus 200 mg in patients aged ≥65 years.
The current analysis indicates that infections, cardiac disorders and malignancies were the most common reasons for death. This is consistent with the previous analysis,2 as well as with studies reporting data from patients with RA in general.12–14 In the current analysis, approximately one-third of deaths were reported as related to COVID-19, which might be considered a confounding factor in the interpretation of the data. Of note, most deaths related to COVID-19 occurred prior to the roll-out of COVID-19 vaccinations. In subgroup analyses, treatment-emergent COVID-19 infections occurred in a similar proportion of patients, regardless of age or filgotinib dose.
Filgotinib long-term efficacy in LTE studies has been reported previously.15 16 Efficacy was also maintained for both doses in patients aged ≥65 years.17 As noted in the SmPC, although the recommended filgotinib dose for patients with risk factors for VTE, MACE or malignancies is 100 mg once daily, this may be escalated to 200 mg once daily if disease control is insufficient.5
To align with the recent SmPC changes recommending the use of filgotinib in patients aged 65 years or older only if alternative treatments are not available, a cut-off age of 65 years was applied in a subgroup analysis. As expected, patients aged ≥65 years had higher EAIRs of serious infections, herpes zoster, MACE, malignancies and all-cause mortality than younger patients.
In patients aged ≥65 years, a potential dose-dependent relationship was seen in the occurrence of herpes zoster and malignancies, with a lower incidence of these TEAEs occurring in patients receiving 100 mg versus 200 mg filgotinib. Similarly, the incidence of all-cause mortality was numerically lower with filgotinib 100 mg versus filgotinib 200 mg in the older patient group. However, EAIR CIs were overlapping for herpes zoster, malignancies and all-cause mortality across the dose groups for patients aged ≥65 years and additional long-term safety data may be needed to draw firm conclusions about these events in older patients. A dose-dependent relationship was not observed in the older patient group for the incidence of serious infections, COVID-19, MACE and VTE.
There were differences in regional demographics between patients in the two age subgroups. Notably, patients from North America and Western Europe (and others) accounted for a greater proportion of those aged ≥65 years, whereas the opposite was true for patients from Asia and Southeast Asia and South and Central America. A previous study examining the efficacy and safety of filgotinib reported that safety was generally consistent across geographical regions.18 Small differences in the safety event rates reported between regions may have resulted from low event numbers and/or regional differences in treatment.18 As such, regional differences are unlikely to impact on the risk of TEAEs observed in older and younger patients in the current long-term analysis.
Real-world safety data have also been recently reported for patients treated with JAK inhibitors. In an observational study using real-world data, patients receiving baricitinib for up to 17 months had a higher risk of VTE and a numerically greater risk of MACE and serious infections compared with those receiving TNF inhibitors.19 The results were largely driven by data from the Anti-Rheumatic Therapy in Sweden study (Swedish RA registry) and the Système National des Données de Santé (the French national health data system), which contributed ~70% of the overall data.
Although some real-world data suggest that there is a higher risk of VTE with JAK inhibitors versus TNF inhibitors and other biologic (b) DMARDs,19–21 other early data showed no difference in VTE risk between these treatments.22 Further emerging real-world evidence also suggests no significant difference in risk of MACE following treatment with JAK inhibitors versus TNF inhibitors or other bDMARDs.21–25
To date, real-world data do not show a significant increase in the risk of malignancy among patients treated with JAK inhibitors or TNF inhibitors; however, longer-term data are required before any risk can be either confirmed or excluded.26–28
In a post hoc analysis of ORAL Surveillance, being aged ≥65 years was identified as a risk factor for MACE and malignancies.4 A further recent publication examined the safety of upadacitinib in a post hoc analysis of patients at higher risk of CV events, similar to that reported for ORAL Surveillance.29 Within the higher-risk population, rates of serious infections, MACE and malignancies excluding NMSC were numerically higher in patients aged ≥65 years versus 50 to <65 years for those receiving upadacitinib 15 mg once daily. Rates of VTE did not differ between these age categories.29 The age subgroup analysis for filgotinib reported herein yields similar results, although the younger age bracket in the upadacitinib analysis included all patients below the age of 65 years, and the population specifically included patients at higher risk of CV events. In the absence of head-to-head studies, no conclusions can be drawn regarding clinically meaningful differences between JAK inhibitors for the treatment of RA.
Study limitations
In the previous analysis of long-term integrated safety data, EAIRs/100 PYE of TEAEs were uncensored and therefore calculated using total exposure time for all patients.2 This contrasts with the current analysis, which used a censored approach, in which exposure was calculated as time to the first event. A censored approach avoids the underestimation of EAIRs. As a previous analysis has shown no difference in filgotinib safety outcomes when using either a censored or uncensored approach to calculating EAIRs, the different methodologies used are not expected to affect the reported incidence rates of safety events.30
The median exposure differed between dose groups in the current analysis, with PYE lower in patients receiving filgotinib 100 mg than 200 mg. This is partially mitigated by the use of EAIRs, which account for these differences, though exposure may have an impact on events that become evident in the longer term, such as malignancies.
At later time points, the number of patients and event numbers for some TEAEs were low, including for MACE and VTE, which makes interpretation difficult. Further, a placebo control group was lacking in the LTE studies. During the trials, clinicians were able to alter patients’ background therapy, based on their clinical judgement. This is not captured in the baseline information on concurrent treatments.
Comparison of the current data for filgotinib with those of other JAK inhibitors should be done cautiously. Notably, the COVID-19 pandemic may have impacted data for all-cause mortality.
Conclusions
Over a median of 3.8 and a maximum of 8.3 years of exposure, the safety and tolerability of filgotinib were similar following 100 mg and 200 mg dosing. Rates of TEAEs of interest remained stable over time. In general, small differences in the rates of some TEAEs between filgotinib doses, which were observed in a previous analysis of these studies, were reduced or absent in the current analysis, including for serious infections, herpes zoster and all-cause mortality. In patients aged ≥65 years, the incidences of most TEAEs of interest were similar between the two doses. Potential dose-dependent relationships for herpes zoster, malignancies and all-cause mortality were observed in patients aged ≥65 years, which cannot be excluded based on the current data set. This indicates the potential impact of age on the safety profile of filgotinib, and that some patients aged ≥65 years may benefit from the 100 mg dose option. These analyses demonstrate that long-term filgotinib exposure was well tolerated in patients with moderate-to-severe active RA, with a stable rate of TEAEs over time.
Data availability statement
Data are available upon reasonable request. Anonymised individual patient data will be shared upon request for research purposes, dependent upon the nature of the request, the merit of the proposed research and the availability of the data and their intended use. The full data sharing policy for Gilead Sciences, Inc., can be found at https://www.gilead.com/about/ethics-and-code-of-conduct/policies. Data sharing policy for Galapagos NV can be found at https://www.clinicaltrials-glpg.com/us/en/data-transparency.html.
Ethics statements
Patient consent for publication
Ethics approval
The trials were conducted in accordance with the Declaration of Helsinki and the International Council for Harmonisation guidelines. The protocols were approved by the institutional review board or ethics committee at each site. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
Medical writing support was provided by Iain Haslam, PhD, CMPP (Aspire Scientific Ltd, Bollington, UK), and funded by Galapagos NV (Mechelen, Belgium). Publication coordination was provided by Jo-Ann E West, MSc, a consultant funded by Galapagos NV. Statistical programming support was provided by Pascal Hanssens, Mattijs Beirinckx, Mohsine El Ghazi, Gerard Hermus and Benjamin Pett, Galapagos Statistical Programming Team. Data were previously presented at EULAR 2023.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Handling editor Kimme L Hyrich
Contributors All authors contributed to acquisition, analysis or interpretation of data; had full access to the data; reviewed the manuscript critically for important intellectual content; and approved the final version for publication. KVB is responsible for the overall content as guarantor.
Funding This study was funded by Galapagos NV (Mechelen, Belgium).
Competing interests GRB reports consultancy fees from AbbVie, Amgen, BMS, Galapagos, Lilly, Pfizer and Sanofi; and speakers’ bureau fees from AbbVie, Amgen, BMS, Chugai, Galapagos, Lilly, Pfizer and Sanofi. J-EG reports consultancy fees from AbbVie, BMS, Galapagos, Gilead, Lilly, MSD, Novartis and Pfizer; and grant/research support from BMS and Pfizer. RC reports consultancy fees from AbbVie, Fresenius, Galapagos, Lilly, Pfizer, Novartis and UCB; speakers’ bureau fees from AbbVie, Amgen, BMS, Celltrion, Fresenius, Galapagos, Janssen, Lilly, Novartis, Pfizer and UCB. KLW reports consultancy fees from AbbVie, AstraZeneca, BMS, Eli Lilly & Company, Galapagos, Gilead, GSK, Novartis, Pfizer, Regeneron, Roche, Sanofi and UCB; and grant/research support from BMS and Pfizer. YT has received speaker fees and/or honoraria from AbbVie, Asahi Kasei, AstraZeneca, BMS, Boehringer Ingelheim, Chugai, Eisai, Eli Lilly, Gilead, GSK, Taiho, Taisho and Pfizer; and research grants from Asahi Kasei, Chugai, Eisai, Mitsubishi Tanabe and Taisho. EVEO is a consultant for Galapagos. VR is a former employee of, and shareholder in, Galapagos. PVH is a consultant for AOP Health, Aspen and Galapagos; and a previous employee of Janssen, MSD and Schering-Plough. KVB is an employee of, and shareholder in, Galapagos. TT reports consultancy fees from AbbVie GK, Astellas, Chugai, Eli Lilly Japan, Gilead, Janssen Pharma K.K., Mitsubishi Tanabe and Pfizer Japan; speakers’ bureau fees from AbbVie GK, AYUMI, BMS, Chugai, Daiichi Sankyo, Eisai, Eli Lilly Japan, Gilead, Janssen Pharma K.K., Mitsubishi Tanabe, Pfizer Japan and Sanofi K.K.; and research/grant support from AbbVie GK, Asahi Kasei, Astellas, AYUMI, Chugai, Daiichi Sankyo, Eisai, Eli Lilly Japan, Mitsubishi Tanabe, Nippon Kayaku and UCB Japan. RW acted as an adviser and speaker for Celltrion, Galapagos and Gilead. DA reports consultancy fees, speakers’ bureau fees and grant/research support from AbbVie, Eli Lilly, Galapagos, Gilead, Janssen, Merck, Novartis, Pfizer, Sandoz and Sanofi.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.