Article Text

Download PDFPDF

Original research
Increased risk of cardiovascular events under the treatments with Janus kinase inhibitors versus biological disease-modifying antirheumatic drugs in patients with rheumatoid arthritis: a retrospective longitudinal population-based study using the Japanese health insurance database
  1. Ryoko Sakai1,2,
  2. Eiichi Tanaka2,3,
  3. Eisuke Inoue4 and
  4. Masayoshi Harigai2,3
  1. 1Department of Publich Health and Epidemiology, Meiji Pharmaceutical University, Kiyose, Tokyo, Japan
  2. 2Division of Rheumatology, Department of Internal Medicine, Tokyo Women's Medical University School of Medicine, Shinjuku-ku, Tokyo, Japan
  3. 3Institute of Rheumatology, Tokyo Women’s Medical University Hospital, Shinjuku-ku, Tokyo, Japan
  4. 4Showa University Research Administration Center, Showa University, Shinagawa-ku, Tokyo, Japan
  1. Correspondence to Dr Masayoshi Harigai; harigai.masayoshi{at}


Objectives To compare the risk of cardiovascular events among Janus kinase inhibitors (JAKIs), biological disease-modifying antirheumatic drugs (bDMARDs) (tumour necrosis factor inhibitors (TNFIs) and non-TNFIs) and methotrexate (MTX) in Japanese patients with rheumatoid arthritis (RA).

Methods Using Japanese claims data, patients with RA were enrolled in this study if they had at least one ICD-10 code (M05 or M06), were new users of JAKIs, bDMARDs or MTX between July 2013 and July 2020 and being 18 years old or older. The incidence rate (IR), IR ratio and adjusted hazard ratio (aHR (95% CI)) of cardiovascular events including venous thromboembolism, arterial thrombosis, acute myocardial infarction and stroke were calculated. A time-dependent Cox regression model adjusted for patient characteristics at baseline was used to calculate aHR.

Results In 53 448 cases, IRs/1000 patient-years of the overall cardiovascular events were 10.1, 6.8, 5.4, 9.1 and 11.3 under the treatments with JAKIs, bDMARDs, TNFIs, non-TNFIs and MTX, respectively. The adjusted HRs of JAKIs for overall cardiovascular events were 1.7 (1.1 to 2.5) versus TNFIs without MTX and 1.7 (1.1 to 2.7) versus TNFIs with MTX.

Conclusions Among patients with RA, individuals using JAKIs had a significantly higher risk of overall cardiovascular events than TNFIs users, which was attributed to the difference in the risk between JAKIs and TNFIs versus MTX. These data should be interpreted with caution because of the limitations associated with the claims database.

  • arthritis, rheumatoid
  • epidemiology
  • biological therapy
  • cardiovascular diseases

Data availability statement

No data are available. No data are available due to the regulation of Medical Data Vision (MDV) Co., Ltd.

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:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • In patients with rheumatoid arthritis (RA), reports of the association between Janus kinase inhibitors (JAKIs) and the risk of cardiovascular events has not been consistent.


  • This study showed a significantly higher risk of overall cardiovascular events in JAKIs users than in tumour necrosis factor inhibitors (TNFIs) users irrespective of concomitant MTX, and a significantly lower risk in TNFIs users than in MTX users using a large Japanese claims database.

  • JAKIs users without MTX also showed a significantly higher cardiovascular risk than non-TNFIs users without MTX.


  • The results of this study suggest the need for appropriate risk management of cardiovascular events among patients with RA, especially, those treated with JAKIs.


Many medications for rheumatoid arthritis (RA) have become available over the past 20 years and have enabled better control of disease activity in large number of patients.1 Of these, Janus kinase inhibitors (JAKIs) are new treatment options,2 3 and five agents have been approved for RA in Japan (tofacitinib was approved in 2013, baricitinib in 2017, peficitinib in 2019 and filgotinib and upadacitinib in 2020). Clinical trials and observational studies have shown clinical efficacy and effectiveness in patients with RA, whereas, concerns about safety has increased.4 Current clinical practice guidelines of the Japan College of Rheumatology5 and recommendations from the European Alliance for Associations for Rheumatology6 call for a careful approach to the use of JAKIs.

Among safety issues under the treatment of JAKIs, the risk of cardiovascular events such as venous thromboembolism (VTE), arterial thromboembolism (ATE), acute myocardial infarction (AMI) and stroke is clinically important. It has been reported that patients with RA have a 60%–140% increased risk of VTE compared with the general population,7–9 and the incidence rate (IR) of VTE in patients treated with biological disease-modifying antirheumatic drugs (bDMARDs) or conventional synthetic DMARDs is estimated to be 0.4–0.8/100 patient-years (PYs).10 11 Additionally, patients with RA have a higher risk of ATE, AMI and stroke than that of the general population.12 13 A meta-analysis showed that use of TNFI reduced the risk of myocardial infarction (pooled adjusted relative risk 0.81 (0.68 to 0.96)) and cerebrovascular accident (RR: 0.69 (0.53 to 0.89)).14 German cohort study revealed that better physical function decreased the risk of stroke significantly (adjusted HR: 0.9 (0.8 to 0.96),15 indicating better disease control by biologics could lead to lower risk of stroke. To optimise the treatment of RA, it is necessary to determine the differences in the risk of these cardiovascular events in patients with RA receiving different categories of DMARDs.

The association between JAKIs and cardiovascular events requires further investigation. A case report and review of the literature of JAKIs showed a numerically higher IR of VTE than placebo16; however, a meta-analyses of randomised clinical trials did not show significantly elevated risks of VTE, pulmonary embolism (PE) and deep vein thrombosis (DVT) in patients with autoimmune diseases who were treated with JAKI.17 A review of observational studies have also shown similar risks of major adverse cardiovascular events (MACEs) and VTE between bDMARDs and JAKIs,18 except for one study on VTE.19 However, tofacitinib did not show non-inferiority to tumour necrosis factor inhibitors (TNFIs) in MACE and malignancy in the ORAL Surveillance, a controlled, prospective, clinical trial,20and safety warnings from the US Food and Drug Administration,21 and the European Medicines Agency22 also were announced. The Pharmaceuticals and Medical Devices Agency in Japan updated the package inserts of all JAKIs, despite the small proportion of Asian patients enrolled in the ORAL Surveillance. Ethnic differences in the risk of cardiovascular events have been reported, with Japanese patients with RA having a lower risk of cardiovascular events than those in Western countries.23 Therefore, it is essential to clarify the risk of cardiovascular events in Japanese patients undergoing treatment with JAKI in the clinical setting.

Investigating rare cardiovascular events during treatment of RA with a particular drug or a class of drugs requires a large number of patients, which can be implemented by a large randomised controlled trial or by using a large registry of the disease or a claims database. A claims database usually contains a larger number of patients than the others, but major drawbacks are it has no or does not have enough information about previously reported risk factors such as body mass index, smoking status, disease activity of RA and history and family history of cardiovascular disease.

This study aimed to estimate and compare the risk of new cardiovascular events among patients with RA treated with JAKIs, bDMARDs (ie, TNFIs or non-TNFIs) and methotrexate (MTX) using a large Japanese health insurance database.


Data source

This retrospective, longitudinal, population-based study was conducted using nationwide hospital-based claims data provided by Medical Data Vision (MDV) Co (Tokyo, Japan). Previous studies have used the MDV database.24 Briefly, the MDV database covers 37.4 million patients from more than 450 hospitals that participate in the Diagnostic Procedure Combination/per diem payment system (DPC/PDPS) in Japan (as of July 2021), which corresponds to 26% of the hospitals that participate in the DPC/PDPS. The data in the MDV database included patient age, sex, diagnoses, drug prescriptions, medical procedures and reimbursement costs for inpatients and outpatients. No personal identifiable information such as patient name, residence, employment or economic status was included in the MDV database. In addition, linkage of databases with other databases or patient medical records has not been allowed in Japan to date.

Study population

This study included new users of JAKIs, bDMARDs and MTX users. We defined the patients who met all the following criteria as new users: (1) having at least one ICD-10 code for RA (M05 or M06); (2) having at least one prescription of JAKI (tofacitinib, baricitinib, peficitinib, upadacitinib or filgotinib), bDMARDs (listed in online supplemental table 1) or oral MTX between July 2013 and July 2020 (subcutaneous (sc) MTX was not yet approved in Japan during these calendar years); (3) being 18 years old or over; and (4) having prescription of neither JAKIs, bDMARDs or MTX in 6 months prior to the index month for MTX new users; having no prescription of bDMARDs irrespective of MTX or JAKIs in 6 months prior to the index month for bDMARDs new users; having no prescription of JAKIs irrespective of MTX or bDMARDs for JAKIs new users. The index month was defined as the first month in which patients met all the above criteria: (1), (2) and (3). To identify new cardiovascular events, patients were excluded from the study if they had a diagnosis of cardiovascular events 6 months prior to the index month. The definitions of cardiovascular events are described in the ‘Definition of cardiovascular events’ section. If MTX+JAKI or MTX+bDMARD was prescribed in the index month, the patient was classified as a new JAKI or bDMARD user.


Patients were followed-up from the index month until the last exposure to JAKIs, bDMARDs, MTX, date of loss to follow-up or the end of follow-up (July 2021), whichever came first. The last day of exposure to MTX, JAKI or bDMARDs for sc injection was defined as the last day of prescription of these drugs plus supply days and 30 days as a grace period. The last day of exposure to bDMARDs for intravenous injection was defined as the last day of bDMARD prescription plus the interval in days of each agent.25 The observation was continued even after changing the drugs to JAKIs or bDMARDs. We implemented an on-drug analysis of the events, considering that the patients were censored at the end of exposure to MTX, bDMARDs or JAKIs, with or without a 30-day grace period, as described above (online supplemental figure 1).

Definition of cardiovascular events

Cardiovascular events, including VTE, DVT, PE, arterial thrombosis, AMI and stroke, were defined as instances where at least one ICD-10 code and medication or medical practice for each disease were recorded during hospitalisation (online supplemental table 2), considering the clinical setting in Japan. Cardiovascular death, typically included in MACE, was not defined as an outcome in this study because it was impossible to define it accurately, due to a lack of data regarding the certification of death and clinical information. In this study, we performed on-drug analysis (online supplemental figure 1). If a cardiovascular event developed during MTX, bDMARD or JAKI monotherapy, it was attributed to the respective drug. If a cardiovascular event occurred during MTX+bDMARD or MTX+JAKI treatment, it was attributed to bDMARD or JAKI. None of the patients received JAKI or bDMARD simultaneously.

Statistical analysis

We described the patients’ comorbidities reported 6 months before the index month (baseline period). Diabetes mellitus (DM), hypertension (HT), dyslipidaemia (DL), antiphospholipid syndrome (APS) and atrial fibrillation (AF) were defined by having at least one ICD-10 code and at least one prescription for each disease (online supplemental table 3). Malignancy was defined using ICD-10 codes (C.x, D.x except for D10.x-D36.x (benign tumour)) and medical procedures (chemotherapy, surgery and radiation) for malignancy or cancer management. Hip fracture was defined as the presence of ICD-10 code S72.x and surgery for hip fracture with hospitalisation. Chronic kidney disease (CKD), congestive heart failure and liver disease were defined using ICD-10 codes according to a previous report.26 A history of hormone replacement treatment was defined as at least one prescription of oestrogen, progestogen or these compounds during the baseline period. Age, sex and medications for RA were defined as index months. Body mass index, smoking status, disease activity of RA, and family history of cardiovascular disease were not included in the database.

Descriptive patient characteristics at baseline, such as age, sex, comorbidities and RA medications, were analysed for the new users of each medication. The IR per 1000 patient-years (PY) and crude IR ratio (IRR) (JAKIs vs bDMARDs, JAKIs vs TNFIs, JAKIs vs non-TNFIs (interleukin-6 (IL-6) inhibitors and abatacept (ABT)) and JAKIs vs MTX) of cardiovascular events were calculated. Time-dependent Cox regression models were employed to calculate the HRs of JAKIs versus bDMARDs, JAKIs versus MTX, bDMARDs versus MTX after adjusting for age, sex, comorbidities including HT, DM, DL, APS, liver disease, CKD, AF and CHF as baseline covariates, and JAKI use, bDMARD use, MTX use, history of malignancy, prescription of hormone replacement treatment, NSAIDs use and categorical doses of oral glucocorticoids (GCs, prednisolone-equivalent dose) (0 mg/day, 0<GC≤5 mg/day, 5<GC≤10 mg/day, GC>10 mg/day) as time-dependent covariates. We calculated adjusted HRs of JAKIs (vs TNFIs, non-TNFIs and MTX), TNFIs (vs MTX), non-TNFIs (vs MTX) with or without MTX separately. We also calculated adjusted HRs of venous events (VTE) and arterial events (ATEA, AMI and stroke) separately. Analysis were performed using R V.4.1.


Study population

The number of the study population was 53 448 and that of JAKIs, bDMARDs and MTX new users were 4416, 20 964 and 37 926, respectively (figure 1). Table 1 presents the baseline patient characteristics. The median (interquartile) and mean±SD observation period were 26 (12–43), 31.7±24.9 months in JAKIs new users, 31 (15–57), 39.1±30.4 months for bDMARD new users and 24 (6–51), 33.4±31.0 months for MTX new users. The median age was late 60s and the most frequent comorbidity in all three groups was HT, followed by DL. Among JAKI new users, most patients started baricitinib (45.9%) or tofacitinib (43.1%). Among the bDMARDs new users, 58.8% started a TNF inhibitor and the remaining patients started a non-TNF inhibitor. Among the patients who started a non-TNF inhibitor, approximately half started tocilizumab treatment.

Figure 1

Patients selection flowchart. Figure shows the selection process of the study population. *Prescription of either JAKIs, bDMARDs nor MTX 6 months prior to the index month for MTX new users; no prescription of bDMARDs irrespective of MTX or JAKIs 6 months prior to the index month for bDMARDs new users; no prescription of JAKIs irrespective of MTX or bDMARDs for new users of JAKIs. bDMARDs, biological disease-modifying antirheumatic drugs; JAKIs, Janus kinase inhibitors; MTX, methotrexate; RA, rheumatoid arthritis.

Table 1

Patients’ characteristics at baseline

Occurrence of cardiovascular events

The IRs (per 1000 PY) with a 95% CI of the total and each cardiovascular event by medication are shown in table 2. The IR of total cardiovascular events was 10.1 (7.9 to 12.9) for JAKIs, 6.8 (6.2 to 7.5) for bDMARDs and 11.3 (10.6 to 12.0) for MTX. When bDMARDs were divided into TNFIs and non-TNFIs, the IR of total cardiovascular events was 5.4 (4.7 to 6.1) for TNFIs and 9.1 (8.0 to 10.3) for non-TNFIs. VTE accounted for the majority of cardiovascular events, except in MTX users. The IRR (95% CI) for total cardiovascular events comparing to JAKIs with TNFIs was elevated (1.9 (1.4 to 2.5)), whereas the IRRs comparing JAKIs with non-TNFIs or MTX were not (1.1 (0.9 to 1.5), 0.9 (0.7 to 1.2)) (table 3).

Table 2

IR of cardiovascular events per 1000 patient-years

Table 3

Crude incidence rate ratio of cardiovascular events

Comparison of the risk of total cardiovascular events among therapies

In a time-dependent Cox regression model, JAKIs with and without MTX showed a significantly higher risk for overall cardiovascular events compared with TNFIs with (1.7 (1.1 to 2.7)) and without MTX (1.7 (1.1 to 2.5)), respectively. HR for JAKIs without MTX versus non-TNFIs without MTX was also significantly elevated (1.7 (1.1 to 2.5)), but HR for JAKIs with MTX versus non-TNFIs with MTX was not (1.0 (0.6 to 1.5)) (table 4, online supplemental figure 2). When venous events (VTE) and arterial events (ATE, AMI and stroke) were separately analysed, JAKIs with and without MTX had a significantly higher risk of venous events than TNFIs with and without MTX, respectively, but did not for that of arterial events. JAKIs without MTX showed a significantly higher risk of venous events than non-TNFIs without MTX as well, but not that of arterial events (online supplemental table 4 and 5).

Table 4

Adjusted HRs of total cardiovascular events


Using a large claims database, this study demonstrated incidence rates of cardiovascular events in patients with RA using MTX, JAKIs, bDMARDs, TNFIs, non-TNFIs in the real world, and that (1) JAKIs+/−MTX were associated with an increased risk of cardiovascular events versus TNFIs+/−MTX, which derived from numerically increased risk of JAKIs+/−MTX versus MTX and significantly decreased risk of TNFIs+/−MTX versus MTX; and (2) similar relationships were observed when JAKIs without MTX, non-TNFIs without MTX, and MTX were compared.

In this study, the IR of DVT (0.6/100PY), PE (0.1/100PY) or ATE (0.1/100PY) in patients treated with JAKIs was similar to the results of previous studies,27 28 although this study population was relatively older than previous studies from western countries.19 29 In patients with RA treated with 5 mg or 10 mg of tofacitinib, the IRs (95% CI) of DVT and PE were 0.2/100 PYs (0.1 to 0.3) and 0.1/100 PYs (0.1 to 0.2) for both doses, and those of ATE were 0.3/100 PYs (0.2 to 0.5) for 5 mg twice a day and 0.4/100 PYs (0.3 to 0.5) for 10 mg twice a day.28 In the placebo-controlled period of a clinical trial of baricitinib, five DVT/PE cases, including two serious cases, were reported in the 4 mg baricitinib group and none in the placebo group.30 The incidence rates of overall and serious DVT/PE were 0.5 and 0.3 per 100 PYs in the combined dataset of the clinical trials and long-term extension study of baricitinib, respectively.31 In addition, in phase II and III clinical trials of baricitinib in Japanese patients with RA, the incidence of DVT/PE was 0.5 per 100 PY.27 An observational study of tofacitinib in Canada reported an IR for VTE of 0.3/100 PY.32 In this study, more than half of events were VTE (table 2). This can be explained by recent increase of VTE risk in Asian countries including Japan. Although the incidence rates of VTE are supposed to be lower in Asian population than others,33 VTE is becoming more common in Asian countries including Japan.34 In addition, recently direct oral anticoagulants have been used commonly. These results and recent management strategies for DVT indicate that treatments for DVT to prevent occurrence of PE during the hospitalisation may be implemented more frequently than before.

A meta-analysis comparing baricitinib and TNFIs using several claims and registry databases showed an IRR of 1.51 (95% CI: 1.10 to 2.08) for VTE and 1.54 (0.93 to 2.54) for MACE,35 which is similar to the results of the present study. Swedish population-based study19 also showed the significantly higher risk of VTE in JAKI users compared with TNFI users. However, several real-world studies have reported a similar risk of VTE and MACE between JAKIs and TNFIs in patients with RA. Analysis of French national health data system revealed that risk of MACEs and VTE for the JAKIs (tofacitinib or baricitinib)-exposed versus adalimumab-exposed group was not significantly different: weighted HR was 1.0 (95% CI: 0.7 to 1.5) (p=0.99) for MACE and 1.1 (0.7 to 1.6) (p=0.63) for VTE.36 The United States Corrona RA registry also reported unincreased risk of MACE in tofacitinib users compared with bDMARD users. The pooled weighted HR (95% CI) comparing tofacitinib with TNFIs was 1.01 (0.83 to 1.23) in a cohort created from claims databases and 1.24 (0.90 to 1.69) in ORAL-surveillance-duplicate cohort.37 In addition, Swedish and German observational cohort studies also reported that there were not significant differences in the risk of MACE between JAKIs and TNFIs.29 38 These data indicate that the risk of cardiovascular events may be increased in patients with RA treated with JAKIs. However, it should be noted that the absolute risk differences in these studies were quite small, and the number needed to harm was large.

In this study, JAKIs had a 1.5-fold increased risk of cardiovascular events compared with bDMARDs. Previous meta-analyses have shown that bDMARDs are associated with a significantly lower risk of MACE and stroke than csDMARDs.39 40 Another observational study conducted in the USA also reported that TNFIs decreased the risk of cardiovascular diseases compared with csDMARDs.40 The IR of cardiovascular events in this study was also decreased for bDMARDs compared with MTX. The same reduction was observed for TNFIs without MTX, TNFIs with MTX and non-TNFIs without MTX. This decreased risk could be explained by the reduction in vascular inflammation by bDMARDs.41 bDMARDs have been reported to be associated with protective calcification of non-calcified lesions in coronary atherosclerosis.41 Additionally, TNFIs improve endothelial function in patients with inflammatory arthritis.42 However, the detailed mechanism has not been elucidated to date. Due to the nature of observational studies, it is not possible to determine whether bDMARDs decreased the risk of cardiovascular events, JAKI increased it, or both.43

To the best of our knowledge, the possible differences in cardiovascular disease risk due to JAK selectivity have not yet been proven. Integrated safety analyses of clinical trials showed that risk of MACEs and VTE for upadacitinib was 0.6–1.0/100 PY and 0.3–0.6/100 PY, respectively,44 risk of MACEs for filgotinib was 0.4–0.6/100 PY,45 and risk of VTE in Asian patients treated with peficitinib was 0.1/100 PY.46 As approximately 90% of the patients with JAKIs were administered tofacitinib or baricitinib, a similar analysis enrolling more patients using other JAKIs is necessary to determine whether the increased risk of vascular events is a class effect of JAKI.

This study has multiple and significant limitations. First, selection bias may have influenced the results because the MDV database included data from only 26% of the hospitals that participated in the DPC/PDPS system. Second, we could not analyse the clinical and laboratory data owing to a lack of data, especially for previously reported risk factors of cardiovascular events including body mass index, smoking status, disease activity of RA47 and history and family history of cardiovascular disease. In addition, there are possibilities that patients treated with JAKIs had longer RA disease duration than others, but we could not include the data in this analysis because it was impossible to identify the disease duration precisely in a claims data. However, we adjusted for other classical risk factors such as DM, HT and HL, and excluded patients with a diagnosis of cardiovascular events 6 months prior to the index month to identify new cardiovascular events as described in the ‘Methods’ section. Third, the definition of cardiovascular events has not yet been validated in the Japanese claims data. Fourth, cardiovascular death was not included among the cardiovascular events in this study, which might have decreased the IRs of total cardiovascular events. Fifth, we could not compare the risk of vascular events among agents because the numbers of patients using each drug and numbers of cases with events became too small to adjust for covariates. It is needed to analyse it in the larger number of patients using the latest database. Lastly, although the definition of patients with RA has been already validated in Japan,48 there were possibilities to include patients with other rheumatic diseases such as psoriatic arthritis.

In conclusion, the total number of the vascular events were lower in bDMARD users, especially in TNFI users, and users of JAKIs versus TNFIs irrespective of concomitant MTX and users of JAKIs without MTX versus non-TNFI without MTX had a significantly higher cardiovascular risk. The increased risks of JAKIs versus TNFIs, and non-TNFIs were attributed to the differences in the risk between JAKIs (vs MTX) and these bDMARDs (vs MTX). The increased cardiovascular risk in JAKIs users was mainly derived from venous events. These data should be interpreted with caution because of the limitations associated with claims databases in general.

Data availability statement

No data are available. No data are available due to the regulation of Medical Data Vision (MDV) Co., Ltd.

Ethics statements

Patient consent for publication

Ethics approval

This study was approved by the ethics committee of Meiji Pharmaceutical University (approval number 202123) and Tokyo Women’s Medical University (approval number 4910).


We would like to thank Editage ( for English language editing. The abstract of this work has been presented at the ACR Congress 2022.


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.


  • Contributors All authors contributed to the study conception and/or design, data analysis and interpretation of data; critically revised the manuscript; and agreed to its submission. MH takes responsibility for the overall contents as guarantor.

  • Funding This work was supported by Pfizer general research grant (Grant number #60447919) and JSPS KAKENHI Grant number 20K10480 and the Ministry of Health, Labour, and Welfare (Grant number 22FE1002).

  • Competing interests Ryoko Sakai has nothing to declare. Eiichi Tanaka has received research funding from Pfizer Inc., UCB Japan Co. Ltd, and has received lecture fees or consulting fees from AbbVie, Asahi Kasei Pharma Co., Astellas Pharmaceutical, Ayumi Pharmaceutical, Boehringer Ingelheim Japan, Inc., Bristol Myers Squibb Co., Ltd., Chugai Pharmaceutical, Daiichi-Sankyo, Inc., Eisai Pharmaceutical, Eli Lilly Japan K.K., Gilead Sciences, Inc., GlaxoSmithKline K.K., Kyowa Pharma Chemical Co., Ltd., Janssen Pharmaceutical K.K., Mitsubishi Tanabe Pharma Co., Mochida Pharmaceutical Co., Ltd., Nippon Kayaku Co., Ltd., Pfizer Japan Inc., Takeda Pharmaceutical, and Teijin Pharma Ltd., and Viatris Japan. Eisuke Inoue has received lecture fees or consulting fees from Nippn tect systems Co., Ltd., Cyberdine Inc., Bristol-Myers Squibb, Eisai Co., Ltd., and Chugai Pharmaceutical Co., Ltd. Masayoshi Harigai has received grants or contacts from AbbVie Japan GK, Asahi Kasei Corp., Boehringer Ingelheim Japan, Inc., Eli Lilly Japan K.K., Kaken Pharmaceutical Co., Mitsubishi 16 Tanabe Pharma Co., Mochida Pharmaceutical Co., Nippon Shinyaku Co., Ltd., Taisho Pharmaceutical Co., Ltd., Teijin Pharma Ltd., UCB Japan Co., Ltd., and Viatris Japan, and has received lecture fees or consulting fees from Bristol Myers Squibb Co., Ltd., Ono Pharmaceutical Co., Ltd., Eli Lilly Japan K.K., AbbVie Japan GK, AbbVie Japan GK, Asahi Kasei Corp., Ayumi Pharmaceutical Co., Bristol Myers Squibb Co., Ltd., Chugai Pharmaceutical Co., Ltd., Eisai Co., Ltd., Gilead Sciences Inc., Janssen Pharmaceutical K.K., Mitsubishi Tanabe Pharma Co., Taisho Pharmaceutical Co., Ltd., Teijin Pharma Ltd and UCB Japan.

  • 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.