Toxicity from anti-TNF therapy

The University of Manchester and the British Society of Rheumatology require data from NHS England in order to enhance the safety data already captured in the British Society for Rheumatology Biologics Register for Rheumatoid Arthritis (BSRBR-RA). This is a long-term observational study to monitor the safety of new biologic and targeted therapies prescribed for rheumatoid arthritis in routine healthcare, specifically to understand if these new drugs increase the risks of developing cancer or premature death above the expected risks in a population with similar disease characteristics not receiving these therapies. Detailed and fully adjusted analyses of the full dataset will take place in the University of Manchester Data Safe Haven where BSRBR-RA data will be combined with the NHS England data on cancer and mortality to see if there is any increased risk of cancer and premature death due to these drugs. The study already captures extensive healthcare data on consented study participants via the NHS rheumatology hospital team. This includes the capture of special category data (namely health data and ethnicity). Data from NHS England on the occurrence of key outcomes of interest (specifically cancer and death) will ensure that the study have robust and complete data on these important outcomes.

As a joint data controller, the British Society for Rheumatology (BSR) has determined the data can be processed for these purposes in support of its legitimate interests. The BSR is an independent health membership charity working to transform lives. The BSR’s members represent the entire rheumatology profession and form a powerful voice for innovation, progression and collaboration at every stage of the patient care pathway. The BSR wants children, young people and adults living with rheumatic and musculoskeletal disease to receive the right health and social care, at the right time, to achieve the best outcome throughout their lifetime. The BSR’s patient registers contribute to increasing knowledge about new and existing treatments, and promote the integration of clinical care, data and evidence-based research outcomes that the BSR hopes will change and improve clinical practice both now and in the future.

The University of Manchester’s justification for processing is for the public interest in the area of medical research to further scientific understanding of inflammatory diseases and therapeutic pathways.

Data will be held securely and the level of personal data processed will be minimised to ensure safeguarding of the data. Published results from the full analysis will be in the form of aggregated datasets and individual personal data will not be shared outside of the study team and only where appropriate legal agreements are in place. No NHS England data will be shared with any third parties including the participating pharmaceutical companies.

The primary objective of the study is to compare the risk of key safety outcomes (including cancer and death) between UK patients with rheumatoid arthritis (RA) starting any new biologic, biosimilar or other new targeted therapy with appropriate comparator cohorts already established in the BSRBR-RA. The data requested will allow the study team to link the hospital and treatment data already captured under the study consent with national cancer and death data on study participants to ensure the register has no missing data on these two major outcomes. This will then allow the study to understand whether there is any increased risk of cancer and death in patients receiving these drugs.

The BSRBR-RA study started in 2001 as a joint venture between the BSR, the pharmaceutical companies who manufacture the drugs, and epidemiologists at the University of Manchester with the aim to monitor the long-term safety of newer drugs prescribed in the NHS to treat patients with RA. The BSR currently receives restricted income from UK pharmaceutical companies, including Abbvie, Celltrion, Eli Lilly, Pfizer, Roche, Samsung Bioepis, Sandoz, Sanofi, UCB, and in the past has received income from Hospira, MSD and SOBI. This income finances a wholly separate contract between the BSR and the University of Manchester. The principal investigator and the BSRBR-RA team at the University of Manchester have full academic freedom and are able to work independently of pharmaceutical industry influence. All decisions concerning analyses, interpretation and publication are made autonomously of any industrial contribution. Members of the BSRBR-RA University of Manchester team, BSR trustees, committee members and staff complete an annual declaration in relation to conflicts of interest. With over 18 years of follow-up in some patients, the study continues to monitor for long-term risks of the original biologic therapies, whilst at the same time, monitoring the safety of newer drugs (including biosimilars and Janus kinase inhibitors (JAKs) that have been later approved for use by the National Institute for Health and Care Excellence (NICE).

The data will only be used for the BSRBR-RA study and is not part of a wider project.

The data subjects for this study consist of 3 types of cohorts. These are as follows:

[1] Exposed Cohort

For each new biologic, biosimilar or other new advanced therapy drug, the exposed cohort are patients under the care of a consultant rheumatologist with rheumatoid arthritis newly starting therapy with that biologic, biosimilar or other new advanced therapy.

Recruitment is co-ordinated at a national level. The study is based in England, Wales, Scotland and Northern Ireland. Historically, patients without a diagnosis of RA (such as psoriatic arthritis, ankylosing spondylitis and other rheumatic conditions) were recruited to the study, but this has since been limited to RA only. Follow-up of these historic non-RA patients continues as per the process outlined in the study protocol. Due to the nature of the study, new cohorts open as new treatments are launched to market. A full up-to-date list of eligible treatments can be found here: http://bsrbr.org/hospitals/eligibility/

[2] Comparator Cohort 1

The first comparator cohort will be patients recruited to the BSRBR-RA with RA who have been registered within 6 months of first exposure to an established tumour necrosis factor inhibitor (TNFi) drug (e.g. Humira, Enbrel or Remicade).

[3] Comparator Cohort 2

The second comparator cohort is a historical RA cohort of patients treated with non-biologic DMARDs recruited to the BSRBR-RA from a select number of sites within the UK (recruited between 2002 and 2008). Patients who subsequently progressed to a biologic, biosimilar or other new targeted therapy will leave this cohort and be eligible to enter an ‘exposed’ cohort but only if the subject consents to being re-registered as an “exposed patient”.

The purpose of the study is to obtain comprehensive long safety data on patients receiving these new therapies, and in particular with reference to NHS England data, to observe if participants receiving these new drugs are at a higher risk of cancer or premature death compared to patients with similar disease not receiving these drugs. For this reason, the study requires Civil Registration Mortality data, Demographics and cancer data for the consented participants taking part in the study. The additional data from NHS England will allow the study to have full outcome data on cancer and death and will enhance the BSRBR-RA dataset.

Recruitment commenced in 2001 and continues to recruit. As of April 2024, ~ 24,000 individuals had been registered to participate in the study.

Notifications for mortality, demographics and cancer data is required from when the study began with the first participant in 2001. Under previous iterations of this Data Sharing Agreement, the study has received data from NHS England on these outcomes since the study started in 2001. The study currently has ethical approval to continue the long-term follow of study participants until 2028 and the data over a long period will play an important part in order to understand the long-term safety of the drugs used.

Although the study will capture most cancer and death outcome data for the study via the NHS hospitals, there are occasions where the study will not be told that one of these events has occurred (i.e. the rheumatology team were not aware and/or the patient may have been treated in a different hospital) and therefore flagging with NHS England will allow complete data on these important outcomes.

The University of Manchester and the British Society for Rheumatology are joint data controllers for the study. The University of Manchester is the only organisation which processes the data. The pharmaceutical companies who manufacture the drugs under study also process the BSRBR-RA study data but not the data provided by NHS England. The pharmaceutical companies have no influence over the dissemination of the results of this study.

The purpose of the data processing activities performed on these data is to produce statistical outputs including reports, submissions to peer reviewed journals and presentations and posters at relevant conferences. BSRBR-RA study data will be combined with NHS England data to maximise and enhance data available and used in the outputs.

The outputs will not contain NHS England Data and will only contain aggregated information with small numbers suppressed as appropriate in line with the relevant disclosure rules for the dataset(s) from which the information was derived.

Dissemination and communication of results to stakeholders includes regular review by BSRBR project boards, including the BSR Registers Steering Committee and BSRBR Data Monitoring and Ethics Committees. The study also has active social media channels (Twitter/Facebook) and a comprehensive study website (www.bsrbr.org) which has areas aimed at hospitals/sites participating, study participants and researchers.

In terms of exploitation of the results/outputs, the British Society for Rheumatology also has links back to the BSRBR-RA study site, in addition to details and the process of applying to access study data for research purposes (https://www.rheumatology.org.uk/practice-quality/registers). To date, over 60 original papers have been published on the BSRBR-RA data since 2001 and the research outputs from the BSRBR-RA study have refined the way TNFi therapies are prescribed in RA with a goal of maximising patient benefit and minimising patient risk. As an example, the research into the influence of TNFi therapy on incidence/risk of solid cancer is cited in the 2019 BSR DMARD safety guidelines in inflammatory arthritis in terms of reassuring patients that their risk of cancer is not increased by taking the medication. The guideline states that, ͞patients should be advised that there is no conclusive evidence for an increased risk of solid tumours or lymphoproliferative disease linked with biologic therapy, but that ongoing vigilance is required. Other European observational studies using national registry data have also been reassuring.

The BSRBR-RA team releases study newsletters which can be accessed via the study website which describes the progress of the study and latest study findings in language targeted at study participants or individuals from a hospital site.

Planned future analysis/output will initially focus on (1) covid-related deaths in this population of people with rheumatoid arthritis. The study team will continue to explore two equally important longer-term research questions as data allows: (2) What is the long-term risk of exposure to established biologic therapies? (3) What is the absolute and relative effectiveness and risk of newer biologic therapies?

(1) With the NHS England death data linked with the study data, this will allow the BSRBR-RA researchers at The University of Manchester to identify all deaths occurring in study patients due to COVID-19 (using International Classification of Diseases (ICD)-10 codes). The rates of COVID-related deaths can be generated in all BSRBR-RA patients, stratified whether they are on or off biologic therapy, and by which biologic therapy. To compare these rates with those in the general population, the Office of National Statistics (ONS) weekly COVID death figures (stratified by age group and gender), and the ONS population estimates (stratified by age group and gender) will be used to generate COVID-related death rates in the general population. From these, standardised mortality ratios (SMRs) and hazard ratios (HRs) can be estimated for the whole cohort, on and off biologic therapy, and for each biologic therapy, to estimate whether the number of COVID-related deaths occurring in the BSRBR-RA is greater than would have been expected given the age and gender of the cohort. Additionally, the researchers will also be able to study those factors associated with death from COVID-19.

(2) What is the long-term risk of exposure to biologic therapies? The first patient was enrolled into the BSRBR-RA in 2001. Therefore, over the next 5 years the study team will start to observe patients who have been receiving TNFi therapies for >20 years. This is an unexplored area due to the very nature of when these drugs were introduced. Hence, the BSRBR-RA can and will offer a unique insight into the very long-term use of TNFi drugs, including treatment persistence and long-term safety, such as the late occurrence of malignancies. The presence of equally long follow-up in an untreated comparison cohort will add to these analyses. Most of the University of Manchester’s long-term analysis has focussed on TNFi but the study now have over 10 years of follow-up data on patients receiving rituximab (RTX)(MabThera)). Within the dataset, RTX is the most common second line biologic drug in RA treatment, with over 5,800 exposures. This drug is different from other biologic therapies not only because of its mechanism of action but also because of its long half-life, making it challenging to study. Therefore, its use in daily practice is less well understood than other RA therapies. Outcomes of interest include long-term effectiveness and safety, including the safety of repeated dosing of MabThera. The University of Manchester plan to continue to study the risk of serious infection and will also move to consider other adverse events such as malignancy and cardiovascular disease in patients treated with MabThera. The University of Manchester will also analyse the data held within the register on work to provide further insight into the effect of response to biologics and related therapies on productivity at work (presenteeism) and the ability for patients to remain in work.

(3) What is the effectiveness and risk of newer biologic therapies? Interleukin (IL)6 inhibitors are one of the newer classes of biologic drugs on the market and therefore, their safety is less well understood. In the UK, it is primarily used as a second (or higher) line biologic therapy following TNFi. The BSRBR-RA has now accrued data from over 2,700 tocilizumab (TCZ) treated patients in the register. The risk of infection over the short term is already being studied as part of collaboration with King’s College London, but specifically a comparative study of infection risk between different choices of second line treatments, including a second TNFi and RTX, will be undertaken. There are concerns over the risk of cardiovascular disease due to aberrations in lipid levels with TCZ and therefore, the risk of MI and stroke following IL6 inhibition will also be investigated. The most challenging analysis will be to study the risks of biosimilar therapies. The switch from originator Remicade and Enbrel to biosimilar products has been swift and extensive in the UK, despite a lack of supporting evidence about the safety of switching. The study team’s analyses in this area will include a comparison of first line biosimilar TNFi use with originator, using recent historical originator data as a comparison, as well as the safety of a switch programme. An analysis of outcomes after switching needs careful consideration due to inherent selection bias (patients who switch between originator and biosimilar have usually experienced a response to the originator and, by definition, have not experienced a treatment limiting adverse event). This selection bias will have to be considered when changes in disease activity and occurrence of adverse events following a switch are analysed and the rich historical data within the BSRBR-RA should allow for this.

The University of Manchester and BSR have already amended the study’s data collection forms to ensure it captures as much disease activity data at the point of switching as possible. JAK inhibitors form a new oral therapeutic option in the treatment of RA. These drugs are now available within the NHS and over the next 5 years, the study team’s efforts will focus on recruitment of patients starting these therapies. Early analyses will include a descriptive study of how JAK inhibitors are being used in the UK and their early effectiveness in routine clinical practice. With the introduction of an oral targeted therapy for RA, the study team will use the opportunity to introduce simple patient questionnaires to gather further data about treatment adherence across modes of therapy.

UPDATES TO OUTPUTS FEBRUARY 2024

Analysis of the BSRBR-RA data continues. To date there have been 100+ scientific publications and our more recent work is summarised in accessible summaries which can be found on our website here: https://www.bsrbr.org/research/lay-summaries/.

The BSRBR-RA study data has been combined with NHS England data to maximise and enhance the rheumatology NHS data available and helped towards not only publications, but also to the National Institute for Health and Clinical Excellence (NICE) guidance, which gives evidence-based good practice recommendations to provide the best care to NHS patients in the UK. The study team hopes to continue to do this with updated NHS England linked data as it becomes available. The data has also contributed to the British Society for Rheumatology (BSR) clinical guidelines which supports evidence-based clinical practice for healthcare providers prescribing rheumatoid arthritis therapies.

Communication of latest results is very important to the study team and the research has been presented at local, national and international rheumatology meetings to help share findings and knowledge amongst the rheumatology community. Also, the data are regularly reviewed by the project committees and researchers from across the UK can apply for datasets using the BSRBR-RA data to answer important research questions.

The study team collaborate closely with the National Rheumatoid Arthritis Society (NRAS), a patient-led charity, who are now a co-investigator on the study. NRAS helps to provide insight into the questions patients have about biologic, biosimilar and other targeted therapies. In 2019, the BSRBR-RA Chief Investigator presented at the NRAS Facebook Live Event Webinar: Understanding the safety of biologic treatments for rheumatoid arthritis NRAS aimed at people with arthritis and there was an interesting Question and Answer (Q&A) session to follow.

In 2021, NRAS and the study team carried out a focus group activity to try to understand what were the most important questions for people living with rheumatoid arthritis. This was followed up with a survey to the wider population of people living with rheumatoid arthritis in January 2023. Amazingly, the survey had to be closed after two hours as over the study limit of 1000 respondents was met so quickly. The results have been published in the NRAS magazine to reach as many people as possible and the team will plan the study's next steps based on these results. In the future, the study team will continue similar work with NRAS and people with rheumatoid arthritis to help guide future research direction.

The study team also produce regular newsletters for both study participants and clinical rheumatology teams to share news and results which are available on our study website: https://www.bsrbr.org/about/newsletters/. A BSRBR-RA blog has also recently been launched to share further the most recent work from the study: https://blogs.manchester.ac.uk/bsrbr-blog/.

As is often the case with research impact, there is a time delay between research publication and clinical impact. Therefore, much of the impact on clinical practice between 2011 and 2021 was from research published in the preceding years. However, the evidence-impact gap for the BSRBR-RA still remains relatively small. The BSRBR-RA has had a significant influence on clinical practice in the UK and more widely. Evidence emerging from the BSRBR-RA has fed directly into National Institute for Health and Care Excellence (NICE) technology appraisals (TA), UK and other clinical practice guidelines, and patient information sheets. These influences have resulted in more consistent prescribing across the country. The data held in the BSRBR-RA have also been used to study the impact of changes in clinical guidelines and safety warnings.

Evidence from the BSRBR-RA has contributed to several NICE Technology Assessments (TAs). For example, two University of Manchester analyses from 2006 (1;2) demonstrated the benefits of combining TNFi treatments with continued background MTX (unless contraindicated); the study contributed directly to NICE TA195 (published 2010; updated TA375, published 2016). Initial guidelines did not specify that MTX treatment should be continued (TA36). Subsequent analysis using the BSRBR-RA data demonstrated that, year on year from 2001-8, the proportion of patients continuing MTX with TNFi increased in the UK, with corresponding improvements in treatment response (3).

Output from the BSRBR-RA has also contributed to BSR Guidelines outlining eligibility criteria for TNFi. The analyses indicated that biologic drugs should no longer be reserved for patients with high disease activity, but will also benefit patients with moderate disease activity despite csDMARDs (4). Unfortunately, the same eligibility criteria were not adopted by NICE in 2016 meaning it remains difficult to prescribe biologic drugs in this setting.

Study research into the risk of infections such as listeria and salmonella (5) has led to new information being included into Arthritis Research UK drug information leaflets (provided to every patient in the UK considering TNFi therapy). It has also prompted the FDA to update product labelling. Specifically, the information now warns of the risk of consuming undercooked or unpasteurised foods, similar to the advice provided to pregnant women. Analysis of the BSRBR-RA dataset in 2013 showed that since updating the Arthritis Research UK drug information leaflets in 2006 there has been a 73% decrease in the rate of new cases of intracellular infection in our patients exposed to TNFi in the UK from 2007 to 12 (6).

Study publications on outcomes among women exposed to TNFi therapy during pregnancy (7; 8) have contributed to a significant change in national pregnancy guidelines. It is now stated that women can continue TNFi therapy in pregnancy if clearly needed (9), whereas previous labelling stated that treatment should be discontinued in the months leading to conception (which often resulted in a disease flare). This is a major advance within rheumatology, as other non-biologic DMARDs, many with the risk of teratogenicity, are contraindicated in pregnancy. The research contributes to the weight of evidence for safer options for disease control in the months leading up to conception and in early pregnancy.

One of the most common questions, by both healthcare professionals and patients, is whether biologic therapy increases the risk of cancer. The initial size of the original Enbrel, Remicade and Humira cohorts were chosen to ensure enough power two detect a doubling in the risk of lymphoma compared to nbDMARD therapy alone. Reassuringly, all findings to date have shown that biologics do not appear to increase the risk of cancer in patients with no prior history of malignancy (10; 11; 12). These findings have been incorporated into national guidelines to help provide reassurance to both patients and their healthcare professionals.

A major challenge for all research studies is the dissemination and implementation of results. The close collaboration with BSR representatives has ensured that study data are disseminated as widely as possible, including to policymakers such as NICE. The study team also worked with rheumatologists to generate ideas for new analysis based on clinically relevant questions, data of which has now been published (examples include moderate disease activity, refractory disease and interstitial lung disease). In recent years, the study team has also used a lunchtime session at the annual BSR Conference to promote the register including its scientific output. The study team have also convened a scientific session at the last four annual conferences to provide a more comprehensive review of the scientific output from the study. The study team have also established an ongoing mutually beneficial collaboration with the National Rheumatoid Arthritis Society (NRAS) as one root of dissemination to patients. On average, 1-2 lay articles per year are written for the NRAS magazine.

The University of Manchester BSRBR-RA team also receive regular queries (average 3 per month) from consultant rheumatologists, nurses, pharmacists as well as via the NRAS helpline, asking about evidence to help support treatment decisions. Often these queries are in areas for which little evidence exists, such as malignancy or pregnancy, but the collective UK experience from the BSRBR-RA can be helpful and reassuring to support these decisions.

The University of Manchester and the British Society of Rheumatology require data from NHS Digital in order to enhance the safety data already captured in the British Society for Rheumatology Biologics Register for Rheumatoid Arthritis (BSRBR-RA). This is a long-term observational study to monitor the safety of new biologic and targeted therapies prescribed for rheumatoid arthritis in routine healthcare, specifically to understand if these new drugs increase the risks of developing cancer or premature death above the expected risks in a population with similar disease characteristics not receiving these therapies. Detailed and fully adjusted analyses of the full dataset will take place in the University of Manchester Data Safe Haven where BSRBR-RA data will be combined with the NHS Digital data on cancer and mortality to see if there is any increased risk of cancer and premature death due to these drugs. The study already captures extensive healthcare data on consented study participants via the NHS rheumatology hospital team. This includes the capture of special category data (namely health data and ethnicity). Data from NHS Digital on the occurrence of key outcomes of interest (specifically cancer and death) will ensure that the study have robust and complete data on these important outcomes.

As a joint data controller, the British Society for Rheumatology (BSR) has determined the data can be processed for these purposes in support of its legitimate interests. The BSR is an independent health membership charity working to transform lives. The BSR’s members represent the entire rheumatology profession and form a powerful voice for innovation, progression and collaboration at every stage of the patient care pathway. The BSR wants children, young people and adults living with rheumatic and musculoskeletal disease to receive the right health and social care, at the right time, to achieve the best outcome throughout their lifetime. The BSR’s patient registers contribute to increasing knowledge about new and existing treatments, and promote the integration of clinical care, data and evidence-based research outcomes that the BSR hopes will change and improve clinical practice both now and in the future.

The University of Manchester’s justification for processing is for the public interest in the area of medical research to further scientific understanding of inflammatory diseases and therapeutic pathways.

Data will be held securely and the level of personal data processed will be minimised to ensure safeguarding of the data. Published results from the full analysis will be in the form of aggregated datasets and individual personal data will not be shared outside of the study team and only where appropriate legal agreements are in place. No NHS Digital data will be shared with any third parties including the participating pharmaceutical companies.

The primary objective of the study is to compare the risk of key safety outcomes (including cancer and death) between UK patients with rheumatoid arthritis (RA) starting any new biologic, biosimilar or other new targeted therapy with appropriate comparator cohorts already established in the BSRBR-RA. The data requested will allow the study team to link the hospital and treatment data already captured under the study consent with national cancer and death data on study participants to ensure the register has no missing data on these two major outcomes. This will then allow the study to understand whether there is any increased risk of cancer and death in patients receiving these drugs.

The BSRBR-RA study started in 2001 as a joint venture between the BSR, the pharmaceutical companies who manufacture the drugs, and epidemiologists at the University of Manchester with the aim to monitor the long-term safety of newer drugs prescribed in the NHS to treat patients with RA. The BSR currently receives restricted income from UK pharmaceutical companies, including Abbvie, Celltrion, Eli Lilly, Pfizer, Roche, Samsung Bioepis, Sandoz, Sanofi, UCB, and in the past has received income from Hospira, MSD and SOBI. This income finances a wholly separate contract between the BSR and the University of Manchester. The principal investigator and the BSRBR-RA team at the University of Manchester have full academic freedom and are able to work independently of pharmaceutical industry influence. All decisions concerning analyses, interpretation and publication are made autonomously of any industrial contribution. Members of the BSRBR-RA University of Manchester team, BSR trustees, committee members and staff complete an annual declaration in relation to conflicts of interest. With over 18 years of follow-up in some patients, the study continues to monitor for long-term risks of the original biologic therapies, whilst at the same time, monitoring the safety of newer drugs (including biosimilars and Janus kinase inhibitors (JAKs)) that have been later approved for use by the National Institute for Health and Care Excellence (NICE).

The data will only be used for the BSRBR-RA study and is not part of a wider project.

The data subjects for this study consist of 3 types of cohorts. These are as follows:

[1] Exposed Cohort

For each new biologic, biosimilar or other new advanced therapy drug, the exposed cohort are patients under the care of a consultant rheumatologist with rheumatoid arthritis newly starting therapy with that biologic, biosimilar or other new advanced therapy.

Recruitment is co-ordinated at a national level. The study is based in England, Wales, Scotland and Northern Ireland. Historically, patients without a diagnosis of RA (such as psoriatic arthritis, ankylosing spondylitis and other rheumatic conditions) were recruited to the study, but this has since been limited to RA only. Follow-up of these historic non-RA patients continues as per the process outlined in the study protocol. Due to the nature of the study, new cohorts open as new treatments are launched to market. A full up-to-date list of eligible treatments can be found here: http://bsrbr.org/hospitals/eligibility/

[2] Comparator Cohort 1

The first comparator cohort will be patients recruited to the BSRBR-RA with RA who have been registered within 6 months of first exposure to an established tumour necrosis factor inhibitor (TNFi) drug (e.g. Humira, Enbrel or Remicade).

[3] Comparator Cohort 2

The second comparator cohort is a historical RA cohort of patients treated with non-biologic DMARDs recruited to the BSRBR-RA from a select number of sites within the UK (recruited between 2002 and 2008). Patients who subsequently progressed to a biologic, biosimilar or other new targeted therapy will leave this cohort and be eligible to enter an ‘exposed’ cohort but only if the subject consents to being re-registered as an “exposed patient”.

The purpose of the study is to obtain comprehensive long safety data on patients receiving these new therapies, and in particular with reference to NHS Digital data, to observe if participants receiving these new drugs are at a higher risk of cancer or premature death compared to patients with similar disease not receiving these drugs. For this reason, the study requires Civil Registration Mortality data and cancer data for the consented participants taking part in the study. The additional data from NHS Digital will allow the study to have full outcome data on cancer and death and will enhance the BSRBR-RA dataset.

Recruitment commenced in 2001 and continues to recruit. As of July 2022, 29,101 individuals had been registered to participate in the study.

Notifications for mortality and cancer data is required from when the study began with the first participant in 2001. Under previous iterations of this Data Sharing Agreement, the study has received data from NHS Digital on these outcomes since the study started in 2001. The study currently has ethical approval to continue the long-term follow of study participants until 2028 and the data over a long period will play an important part in order to understand the long-term safety of the drugs used.

Although the study will capture most cancer and death outcome data for the study via the NHS hospitals, there are occasions where the study will not be told that one of these events has occurred (i.e. the rheumatology team were not aware and/or the patient may have been treated in a different hospital) and therefore flagging with NHS Digital will allow complete data on these important outcomes.

The University of Manchester and the British Society for Rheumatology are joint data controllers for the study. The University of Manchester is the only organisation which processes the data. The pharmaceutical companies who manufacture the drugs under study also process the BSRBR-RA study data but not the data provided by NHS Digital. The pharmaceutical companies have no influence over the dissemination of the results of this study.

The purpose of the data processing activities performed on these data is to produce statistical outputs including reports, submissions to peer reviewed journals and presentations and posters at relevant conferences. BSRBR-RA study data will be combined with NHS Digital data to maximise and enhance data available and used in the outputs. Only aggregated data will be included in any study outputs and data will be safeguarded by ensuring that results which include small numbers which could identify study participants will be excluded. Any numbers smaller than 5 will be round to ≤5 in line with NHS Digital’s HES Analysis Guide.

Dissemination and communication of results to stakeholders includes regular review by BSRBR project boards, including the BSR Registers Steering Committee and BSRBR Data Monitoring and Ethics Committees. The study also has active social media channels (Twitter/Facebook) and a comprehensive study website (www.bsrbr.org) which has areas aimed at hospitals/sites participating, study participants and researchers.

In terms of exploitation of the results/outputs, the British Society for Rheumatology also has links back to the BSRBR-RA study site, in addition to details and the process of applying to access study data for research purposes (https://www.rheumatology.org.uk/practice-quality/registers). To date, over 60 original papers have been published on the BSRBR-RA data since 2001 and the research outputs from the BSRBR-RA study have refined the way TNFi therapies are prescribed in RA with a goal of maximising patient benefit and minimising patient risk. As an example, the research into the influence of TNFi therapy on incidence/risk of solid cancer is cited in the 2019 BSR DMARD safety guidelines in inflammatory arthritis in terms of reassuring patients that their risk of cancer is not increased by taking the medication. The guideline states that, ͞patients should be advised that there is no conclusive evidence for an increased risk of solid tumours or lymphoproliferative disease linked with biologic therapy, but that ongoing vigilance is required. Other European observational studies using national registry data have also been reassuring.

The BSRBR-RA team releases study newsletters which can be accessed via the study website which describes the progress of the study and latest study findings in language targeted at study participants or individuals from a hospital site.

Planned future analysis/output will initially focus on (1) covid-related deaths in this population of people with rheumatoid arthritis. The study team will continue to explore two equally important longer-term research questions as data allows: (2) What is the long-term risk of exposure to established biologic therapies? (3) What is the absolute and relative effectiveness and risk of newer biologic therapies?

(1) With the NHS Digital death data linked with the study data, this will allow the BSRBR-RA researchers at The University of Manchester to identify all deaths occurring in study patients due to COVID-19 (using International Classification of Diseases (ICD)-10 codes). The rates of COVID-related deaths can be generated in all BSRBR-RA patients, stratified whether they are on or off biologic therapy, and by which biologic therapy. To compare these rates with those in the general population, the Office of National Statistics (ONS) weekly COVID death figures (stratified by age group and gender), and the ONS population estimates (stratified by age group and gender) will be used to generate COVID-related death rates in the general population. From these, standardised mortality ratios (SMRs) and hazard ratios (HRs) can be estimated for the whole cohort, on and off biologic therapy, and for each biologic therapy, to estimate whether the number of COVID-related deaths occurring in the BSRBR-RA is greater than would have been expected given the age and gender of the cohort. Additionally, the researchers will also be able to study those factors associated with death from COVID-19.

(2) What is the long-term risk of exposure to biologic therapies? The first patient was enrolled into the BSRBR-RA in 2001. Therefore, over the next 5 years the study team will start to observe patients who have been receiving TNFi therapies for >20 years. This is an unexplored area due to the very nature of when these drugs were introduced. Hence, the BSRBR-RA can and will offer a unique insight into the very long-term use of TNFi drugs, including treatment persistence and long-term safety, such as the late occurrence of malignancies. The presence of equally long follow-up in an untreated comparison cohort will add to these analyses. Most of the University of Manchester’s long-term analysis has focussed on TNFi but the study now have over 10 years of follow-up data on patients receiving rituximab (RTX)(MabThera)). Within the dataset, RTX is the most common second line biologic drug in RA treatment, with over 5,800 exposures. This drug is different from other biologic therapies not only because of its mechanism of action but also because of its long half-life, making it challenging to study. Therefore, its use in daily practice is less well understood than other RA therapies. Outcomes of interest include long-term effectiveness and safety, including the safety of repeated dosing of MabThera. The University of Manchester plan to continue to study the risk of serious infection and will also move to consider other adverse events such as malignancy and cardiovascular disease in patients treated with MabThera. The University of Manchester will also analyse the data held within the register on work to provide further insight into the effect of response to biologics and related therapies on productivity at work (presenteeism) and the ability for patients to remain in work.

(3) What is the effectiveness and risk of newer biologic therapies? Interleukin (IL)6 inhibitors are one of the newer classes of biologic drugs on the market and therefore, their safety is less well understood. In the UK, it is primarily used as a second (or higher) line biologic therapy following TNFi. The BSRBR-RA has now accrued data from over 2,700 tocilizumab (TCZ) treated patients in the register. The risk of infection over the short term is already being studied as part of collaboration with King’s College London, but specifically a comparative study of infection risk between different choices of second line treatments, including a second TNFi and RTX, will be undertaken. There are concerns over the risk of cardiovascular disease due to aberrations in lipid levels with TCZ and therefore, the risk of MI and stroke following IL6 inhibition will also be investigated. The most challenging analysis will be to study the risks of biosimilar therapies. The switch from originator Remicade and Enbrel to biosimilar products has been swift and extensive in the UK, despite a lack of supporting evidence about the safety of switching. The study team’s analyses in this area will include a comparison of first line biosimilar TNFi use with originator, using recent historical originator data as a comparison, as well as the safety of a switch programme. An analysis of outcomes after switching needs careful consideration due to inherent selection bias (patients who switch between originator and biosimilar have usually experienced a response to the originator and, by definition, have not experienced a treatment limiting adverse event). This selection bias will have to be considered when changes in disease activity and occurrence of adverse events following a switch are analysed and the rich historical data within the BSRBR-RA should allow for this.

The University of Manchester and BSR have already amended the study’s data collection forms to ensure it captures as much disease activity data at the point of switching as possible. JAK inhibitors form a new oral therapeutic option in the treatment of RA. These drugs are now available within the NHS and over the next 5 years, the study team’s efforts will focus on recruitment of patients starting these therapies. Early analyses will include a descriptive study of how JAK inhibitors are being used in the UK and their early effectiveness in routine clinical practice. With the introduction of an oral targeted therapy for RA, the study team will use the opportunity to introduce simple patient questionnaires to gather further data about treatment adherence across modes of therapy.

As is often the case with research impact, there is a time delay between research publication and clinical impact. Therefore, much of the impact on clinical practice between 2011 and 2021 was from research published in the preceding years. However, the evidence-impact gap for the BSRBR-RA still remains relatively small. The BSRBR-RA has had a significant influence on clinical practice in the UK and more widely. Evidence emerging from the BSRBR-RA has fed directly into National Institute for Health and Care Excellence (NICE) technology appraisals (TA), UK and other clinical practice guidelines, and patient information sheets. These influences have resulted in more consistent prescribing across the country. The data held in the BSRBR-RA have also been used to study the impact of changes in clinical guidelines and safety warnings.

Evidence from the BSRBR-RA has contributed to several NICE Technology Assessments (TAs). For example, two University of Manchester analyses from 2006 (1;2) demonstrated the benefits of combining TNFi treatments with continued background MTX (unless contraindicated); the study contributed directly to NICE TA195 (published 2010; updated TA375, published 2016). Initial guidelines did not specify that MTX treatment should be continued (TA36). Subsequent analysis using the BSRBR-RA data demonstrated that, year on year from 2001-8, the proportion of patients continuing MTX with TNFi increased in the UK, with corresponding improvements in treatment response (3).

Output from the BSRBR-RA has also contributed to BSR Guidelines outlining eligibility criteria for TNFi. The analyses indicated that biologic drugs should no longer be reserved for patients with high disease activity, but will also benefit patients with moderate disease activity despite csDMARDs (4). Unfortunately, the same eligibility criteria were not adopted by NICE in 2016 meaning it remains difficult to prescribe biologic drugs in this setting.

Study research into the risk of infections such as listeria and salmonella (5) has led to new information being included into Arthritis Research UK drug information leaflets (provided to every patient in the UK considering TNFi therapy). It has also prompted the FDA to update product labelling. Specifically, the information now warns of the risk of consuming undercooked or unpasteurised foods, similar to the advice provided to pregnant women. Analysis of the BSRBR-RA dataset in 2013 showed that since updating the Arthritis Research UK drug information leaflets in 2006 there has been a 73% decrease in the rate of new cases of intracellular infection in our patients exposed to TNFi in the UK from 2007 to 12 (6).

Study publications on outcomes among women exposed to TNFi therapy during pregnancy (7; 8) have contributed to a significant change in national pregnancy guidelines. It is now stated that women can continue TNFi therapy in pregnancy if clearly needed (9), whereas previous labelling stated that treatment should be discontinued in the months leading to conception (which often resulted in a disease flare). This is a major advance within rheumatology, as other non-biologic DMARDs, many with the risk of teratogenicity, are contraindicated in pregnancy. The research contributes to the weight of evidence for safer options for disease control in the months leading up to conception and in early pregnancy.

One of the most common questions, by both healthcare professionals and patients, is whether biologic therapy increases the risk of cancer. The initial size of the original Enbrel, Remicade and Humira cohorts were chosen to ensure enough power two detect a doubling in the risk of lymphoma compared to nbDMARD therapy alone. Reassuringly, all findings to date have shown that biologics do not appear to increase the risk of cancer in patients with no prior history of malignancy (10; 11; 12). These findings have been incorporated into national guidelines to help provide reassurance to both patients and their healthcare professionals.

A major challenge for all research studies is the dissemination and implementation of results. The close collaboration with BSR representatives has ensured that study data are disseminated as widely as possible, including to policymakers such as NICE. The study team also worked with rheumatologists to generate ideas for new analysis based on clinically relevant questions, data of which has now been published (examples include moderate disease activity, refractory disease and interstitial lung disease). In recent years, the study team has also used a lunchtime session at the annual BSR Conference to promote the register including its scientific output. The study team have also convened a scientific session at the last four annual conferences to provide a more comprehensive review of the scientific output from the study. The study team have also established an ongoing mutually beneficial collaboration with the National Rheumatoid Arthritis Society (NRAS) as one root of dissemination to patients. On average, 1-2 lay articles per year are written for the NRAS magazine.

The University of Manchester BSRBR-RA team also receive regular queries (average 3 per month) from consultant rheumatologists, nurses, pharmacists as well as via the NRAS helpline, asking about evidence to help support treatment decisions. Often these queries are in areas for which little evidence exists, such as malignancy or pregnancy, but the collective UK experience from the BSRBR-RA can be helpful and reassuring to support these decisions.

The University of Manchester and the British Society of Rheumatology require data from NHS Digital in order to enhance the safety data already captured in the British Society for Rheumatology Biologics Register for Rheumatoid Arthritis (BSRBR-RA). This is a long-term observational study to monitor the safety of new biologic and targeted therapies prescribed for rheumatoid arthritis in routine healthcare, specifically to understand if these new drugs increase the risks of developing cancer or premature death above the expected risks in a population with similar disease characteristics not receiving these therapies. Detailed and fully adjusted analyses of the full dataset will take place in the University of Manchester Data Safe Haven where BSRBR-RA data will be combined with the NHS Digital data on cancer and mortality to see if there is any increased risk of cancer and premature death due to these drugs. The study already captures extensive healthcare data on consented study participants via the NHS rheumatology hospital team. This includes the capture of special category data (namely health data and ethnicity). Data from NHS Digital on the occurrence of key outcomes of interest (specifically cancer and death) will ensure that the study have robust and complete data on these important outcomes.

As a joint data controller, the British Society for Rheumatology (BSR) has determined the data can be processed for these purposes in support of its legitimate interests. The BSR is an independent health membership charity working to transform lives. The BSR’s members represent the entire rheumatology profession and form a powerful voice for innovation, progression and collaboration at every stage of the patient care pathway. The BSR wants children, young people and adults living with rheumatic and musculoskeletal disease to receive the right health and social care, at the right time, to achieve the best outcome throughout their lifetime. The BSR’s patient registers contribute to increasing knowledge about new and existing treatments, and promote the integration of clinical care, data and evidence-based research outcomes that the BSR hopes will change and improve clinical practice both now and in the future.

The University of Manchester’s justification for processing is for the public interest in the area of medical research to further scientific understanding of inflammatory diseases and therapeutic pathways.

Data will be held securely and the level of personal data processed will be minimised to ensure safeguarding of the data. Published results from the full analysis will be in the form of aggregated datasets and individual personal data will not be shared outside of the study team and only where appropriate legal agreements are in place. No NHS Digital data will be shared with any third parties including the participating pharmaceutical companies.

The primary objective of the study is to compare the risk of key safety outcomes (including cancer and death) between UK patients with rheumatoid arthritis (RA) starting any new biologic, biosimilar or other new targeted therapy with appropriate comparator cohorts already established in the BSRBR-RA. The data requested will allow the study team to link the hospital and treatment data already captured under the study consent with national cancer and death data on study participants to ensure the register has no missing data on these two major outcomes. This will then allow the study to understand whether there is any increased risk of cancer and death in patients receiving these drugs.

The BSRBR-RA study started in 2001 as a joint venture between the BSR, the pharmaceutical companies who manufacture the drugs, and epidemiologists at the University of Manchester with the aim to monitor the long-term safety of newer drugs prescribed in the NHS to treat patients with RA. The BSR currently receives restricted income from UK pharmaceutical companies, including Abbvie, Celltrion, Eli Lilly, Pfizer, Roche, Samsung Bioepis, Sandoz, Sanofi, UCB, and in the past has received income from Hospira, MSD and SOBI. This income finances a wholly separate contract between the BSR and the University of Manchester. The principal investigator and the BSRBR-RA team at the University of Manchester have full academic freedom and are able to work independently of pharmaceutical industry influence. All decisions concerning analyses, interpretation and publication are made autonomously of any industrial contribution. Members of the BSRBR-RA University of Manchester team, BSR trustees, committee members and staff complete an annual declaration in relation to conflicts of interest. With over 18 years of follow-up in some patients, the study continues to monitor for long-term risks of the original biologic therapies, whilst at the same time, monitoring the safety of newer drugs (including biosimilars and Janus kinase inhibitors (JAKs)) that have been later approved for use by the National Institute for Health and Care Excellence (NICE).

The data will only be used for the BSRBR-RA study and is not part of a wider project.

The data subjects for this study consist of 3 types of cohorts. These are as follows:

[1] Exposed Cohort

For each new biologic, biosimilar or other new advanced therapy drug, the exposed cohort are patients under the care of a consultant rheumatologist with rheumatoid arthritis newly starting therapy with that biologic, biosimilar or other new advanced therapy.

Recruitment is co-ordinated at a national level. The study is based in England, Wales, Scotland and Northern Ireland. Historically, patients without a diagnosis of RA (such as psoriatic arthritis, ankylosing spondylitis and other rheumatic conditions) were recruited to the study, but this has since been limited to RA only. Follow-up of these historic non-RA patients continues as per the process outlined in the study protocol. Due to the nature of the study, new cohorts open as new treatments are launched to market. A full up-to-date list of eligible treatments can be found here: http://bsrbr.org/hospitals/eligibility/

[2] Comparator Cohort 1

The first comparator cohort will be patients recruited to the BSRBR-RA with RA who have been registered within 6 months of first exposure to an established tumour necrosis factor inhibitor (TNFi) drug (e.g. Humira, Enbrel or Remicade).

[3] Comparator Cohort 2

The second comparator cohort is a historical RA cohort of patients treated with non-biologic DMARDs recruited to the BSRBR-RA from a select number of sites within the UK (recruited between 2002 and 2008). Patients who subsequently progressed to a biologic, biosimilar or other new targeted therapy will leave this cohort and be eligible to enter an ‘exposed’ cohort but only if the subject consents to being re-registered as an “exposed patient”.

The purpose of the study is to obtain comprehensive long safety data on patients receiving these new therapies, and in particular with reference to NHS Digital data, to observe if participants receiving these new drugs are at a higher risk of cancer or premature death compared to patients with similar disease not receiving these drugs. For this reason, the study requires Civil Registration Mortality data and cancer data for the consented participants taking part in the study. The additional data from NHS Digital will allow the study to have full outcome data on cancer and death and will enhance the BSRBR-RA dataset.

Recruitment commenced in 2001 and continues to recruit. As of June 2019, 24,000 individuals had consented to particpate in this study.

Notifications for mortality and cancer data is required from when the study began with the first participant in 2001. Under previous iterations of this Data Sharing Agreement, the study has received data from NHS Digital on these outcomes since the study started in 2001. The study currently has ethical approval to continue the long-term follow of study participants until 2028 and the data over a long period will play an important part in order to understand the long-term safety of the drugs used.

Although the study will capture most cancer and death outcome data for the study via the NHS hospitals, there are occasions where the study will not be told that one of these events has occurred (i.e. the rheumatology team were not aware and/or the patient may have been treated in a different hospital) and therefore flagging with NHS Digital will allow complete data on these important outcomes.

The University of Manchester and the British Society for Rheumatology are joint data controllers for the study. The University of Manchester is the only organisation which processes the data. The pharmaceutical companies who manufacture the drugs under study also process the BSRBR-RA study data but not the data provided by NHS Digital. The pharmaceutical companies have no influence over the dissemination of the results of this study.

The outputs for the BSRBR-RA study will include reports, submissions to peer reviewed journals and presentations and posters at relevant conferences. BSRBR-RA Study data will be combined with NHS Digital data to maximise data available and used in the outputs. Only aggregated data will be included in any study outputs and data will be safeguarded by ensuring that results which include small numbers which could identify study participants will be excluded.

Dissemination and communication of results to stakeholders includes regular review by BSRBR project boards, including the BSRBR Steering Committee and BSRBR Data Monitoring and Ethics Committees. The study also has active social media channels (Twitter/Facebook) and a comprehensive study website (www.bsrbr.org) which has areas aimed at Hospitals/Sites participating, study participants and researchers. In terms of exploitation of the results/outputs, the British Society for Rheumatology also has links back to the BSRBR study site, in addition to details and the process of applying to access study data for research purposes (https://www.rheumatology.org.uk/practice-quality/registers).

To date, over 60 original papers have been published on the BSRBR-RA data since 2001 and the research outputs from the BSRBR-RA study have refined the way anti-TNF therapies are prescribed in RA with a goal of maximising patient benefit and minimising patient risk. As an example, the research into the influence of anti-TNF therapy on incidence/risk of solid cancer is cited in the 2019 BSR DMARD safety guidelines in inflammatory arthritis in terms of reassuring patients that their risk of cancer is not increased by taking the medication. The guideline states that, “patients should be advised that there is no conclusive evidence for an increased risk of solid tumours or lymphoproliferative disease linked with biologic therapy, but that ongoing vigilance is required (grade 1A, SOA 99%)… Most observational studies using national registry data have also been reassuring; recent studies from BSRBR-RA … registries have all failed to show a significant association between anti-TNF use in RA and overall malignancy, with up to 52 549 patient-years exposure.”

The BSRBR-RA study releases occasional newsletters which can be accessed via the study website which describes the progress of the study and latest study findings in language targeted at study participants or individuals from a hospital site.

Planned future analysis/output will focus on two related and equally important research questions: (1) What is the long term risk of exposure to established biologic therapies? (2) What is the absolute and relative effectiveness and risk of newer biologic therapies?

(1) What is the long term risk of exposure to biologic therapies?

The first patient was enrolled into the BSRBR-RA in 2001. Therefore, over the next 5 years the study team will start to observe patients who have been receiving TNFi therapies for >20 years. This is an unexplored area due to the very nature of when these drugs were introduced. Hence, the BSRBR-RA can and will offer a unique insight into the very long term use of TNFi drugs, including treatment persistence and long-term safety, such as the late occurrence of malignancies. The presence of equally long follow-up in an untreated comparison cohort will add to these analyses.

Most of the University of Manchester’s long-term analysis has focussed on TNFi but the study now have over 10 years of follow-up data on patients receiving rituximab (MabThera). Within the dataset, RTX is the most common second line biologic drug in RA treatment, with over 5,800 exposures. This drug is different from other biologic therapies not only because of its mechanism of action but also because of its long half-life, making it challenging to study. Therefore, its use in daily practice is less well understood than other RA therapies. Outcomes of interest include long-term effectiveness and safety, including the safety of repeated dosing of MabThera. The University of Manchester plan to continue to study the risk of serious infection, and will also move to consider other adverse events such as malignancy and cardiovascular disease in patients treated with MabThera.

The University of Manchester will also analyse the data held within the register on work to provide further insight into the effect of response to biologics and related therapies on productivity at work (presenteeism) and the ability for patients to remain in work.

(2) What is the effectiveness and risk of newer biologic therapies?

IL6 inhibitors are one of the newer classes of biologic drugs on the market and therefore, their safety is less well understood. In the UK, it is primarily used as a second (or higher) line biologic therapy following TNFi. The BSRBR-RA has now accrued data from over 2,700 tocilizumab (TCZ) treated patients in the register. The risk of infection over the short term is already being studied as part of collaboration with King’s College London, but specifically a comparative study of infection risk between different choices of second line treatments, including a second TNFi and RTX, will be undertaken. There are concerns over the risk of cardiovascular disease due to aberrations in lipid levels with TCZ and therefore, the risk of MI and stroke following IL6 inhibition will also be investigated.

The most challenging analysis will be to study the risks of biosimilar therapies. The switch from originator Remicade and Enbrel to biosimilar products has been swift and extensive in the UK, despite a lack of supporting evidence about the safety of switching. The study team’s analyses in this area will include a comparison of first line biosimilar TNFi use with originator, using recent historical originator data as a comparison, as well as the safety of a switch programme. An analysis of outcomes after switching needs careful consideration due to inherent selection bias (patients who switch between originator and biosimilar have usually experienced a response to the originator and by definition, have not experienced a treatment limiting adverse event). This selection bias will have to be considered when changes in disease activity and occurrence of adverse events following a switch are analysed and the rich historical data within the BSRBR-RA should allow for this. The University of Manchester and BSR have already amended the study’s data collection forms to ensure it captures as much disease activity data at the point of switching as possible.

JAK inhibitors form a new oral therapeutic option in the treatment of RA. These drugs are now available within the NHS and over the next 5 years, the study team’s efforts will focus on recruitment of patients starting these therapies. Early analyses will include a descriptive study of how JAK inhibitors are being used in the UK and their early effectiveness in routine clinical practice. With the introduction of an oral targeted therapy for RA, the study team will use the opportunity to introduce simple patient questionnaires to gather further data about treatment adherence across modes of therapy.

As is often the case with research impact, there is a time delay between research publication and clinical impact. Therefore, much of the impact on clinical practice between 2011 and 2018 was from research published in the preceding years. However, the evidence-impact gap for the BSRBR-RA remains relatively small.

The BSRBR-RA has had a significant influence on clinical practice in the UK and more widely. Evidence emerging from the BSRBR-RA has fed directly into National Institute for Health and Care Excellence (NICE) technology appraisals (TA), UK and other clinical practice guidelines, and patient information sheets. These influences have resulted in more consistent prescribing across the country. The data held in the BSRBR-RA have also been used to study the impact of changes in clinical guidelines and safety warnings.

Evidence from the BSRBR-RA has contributed to several NICE TAs. For example, two UoM analyses from 2006 (1;2) demonstrated the benefits of combining TNFi treatments with continued background MTX (unless contraindicated); the study contributed directly to NICE TA195 (published 2010; updated with TA375, published 2016). Initial guidelines did not specify that MTX treatment should be continued (TA36). Subsequent analysis using the register demonstrated that, year on year from 2001-8, the proportion of patients continuing MTX with TNFi increased in the UK, with corresponding improvements in treatment responses (3).

Output from the BSRBR-RA has also contributed to BSR guidelines outlining eligibility criteria for TNFi. The analyses indicated that biologic drugs should no longer be reserved for patients with high disease activity, but will also benefit patients with moderate disease activity despite csDMARDs (4). Unfortunately the same eligibility criteria were not adopted by NICE in 2016, meaning it remains difficult to prescribe biologics in this setting.

Study research into the risk of intracellular infections such as listeria and salmonella (5) has led to new information being incorporated into Arthritis Research UK Drug Information Leaflets (provided to every patient in the UK considering TNFi therapies). It also prompted the FDA to update product labelling. Specifically, the information now warns of the risk of consuming undercooked or unpasteurised foods, similar to the advice provided to pregnant women. Analysis of the BSRBR-RA dataset in 2013 showed that since updating the Arthritis Research UK Drug Information Leaflets in 2006 there has been a 73% decrease in the rate of new cases of intracellular infection in RA patients exposed to TNFi in the UK from 2007-12 (6).

Study publications on outcomes among women exposed to anti-TNF therapy during pregnancy (7;8) have contributed to a significant change in national pregnancy guidelines. It is now stated that women can continue TNFi therapies into pregnancy if clearly needed (9), whereas previous labelling stated that treatment should be discontinued in the months leading up to conception (which often resulted in a disease flare). This is a major advance within rheumatology, as other non-biologic DMARDs, many with the risk of teratogenicity, are contra-indicated in pregnancy. This research contributes to the weight of evidence for safer options for disease control in the months leading up to conception and in early pregnancy.

One of the most common questions which both healthcare professionals and patients have is around whether biologic therapy increases the risk of cancer. The initial size of the original Enbrel, Remicade and Humira cohorts were chosen to ensure enough power to detect a doubling in the risk of lymphoma compared to csDMARD therapy alone. Reassuringly, all of the findings to date have shown that biologics do not appear to increase the risk of cancer in patients with no prior history of malignancy (10;11;12). These findings have been incorporated into national guidelines to help provide reassurance to both patients and their healthcare professionals.

A major challenge for all research studies is the dissemination and implementation of results. The close collaboration with BSR representatives has ensured that study data are disseminated as widely as possible, including to policy makers such as NICE. The study team also work with rheumatologists to generate ideas for new analyses based on clinically relevant questions (examples include moderate disease activity), refractory disease and interstitial lung disease). In recent years the study team have also used a lunchtime session at the annual BSR conference to promote the register including its scientific output. The study team have also convened a scientific session at the last 2 annual conferences to provide a more comprehensive review of the scientific output from the study.

The study have also established an ongoing and mutually beneficial collaboration with the National Rheumatoid Arthritis Society (NRAS) (a patient-led organisation) as one route of dissemination to patients. On average, 1-2 lay articles per year for the NRAS Magazine are written.

The UoM BSRBR-RA team also receive regular queries (average 3 per month) from doctors, nurses, pharmacists as well as via the NRAS helpline asking about evidence to help support treatment decisions. Often these queries are in areas for which little evidence exists, such as malignancy or pregnancy, but the collective UK experience can be helpful and reassuring to support these decisions.

The University of Manchester and the British Society of Rheumatology require data from NHS Digital in order to enhance the safety data already captured in the British Society for Rheumatology Biologics Register for Rheumatoid Arthritis (BSRBR-RA). This is a long-term observational study to monitor the safety of new biologic and targeted therapies prescribed for rheumatoid arthritis in routine healthcare, specifically to understand if these new drugs increase the risks of developing cancer or premature death above the expected risks in a population with similar disease characteristics not receiving these therapies. Detailed and fully adjusted analyses of the full dataset will take place in the University of Manchester Data Safe Haven where BSRBR-RA data will be combined with the NHS Digital data on cancer and mortality to see if there is any increased risk of cancer and premature death due to these drugs. The study already captures extensive healthcare data on consented study participants via the NHS rheumatology hospital team. This includes the capture of special category data (namely health data and ethnicity). Data from NHS Digital on the occurrence of key outcomes of interest (specifically cancer and death) will ensure that the study have robust and complete data on these important outcomes.

As a joint data controller, the British Society for Rheumatology (BSR) has determined the data can be processed for these purposes in support of its legitimate interests. The BSR is an independent health membership charity working to transform lives. The BSR’s members represent the entire rheumatology profession and form a powerful voice for innovation, progression and collaboration at every stage of the patient care pathway. The BSR wants children, young people and adults living with rheumatic and musculoskeletal disease to receive the right health and social care, at the right time, to achieve the best outcome throughout their lifetime. The BSR’s patient registers contribute to increasing knowledge about new and existing treatments, and promote the integration of clinical care, data and evidence-based research outcomes that the BSR hopes will change and improve clinical practice both now and in the future.

The University of Manchester’s justification for processing is for the public interest in the area of medical research to further scientific understanding of inflammatory diseases and therapeutic pathways.

Data will be held securely and the level of personal data processed will be minimised to ensure safeguarding of the data. Published results from the full analysis will be in the form of aggregated datasets and individual personal data will not be shared outside of the study team and only where appropriate legal agreements are in place. No NHS Digital data will be shared with any third parties including the participating pharmaceutical companies.

The primary objective of the study is to compare the risk of key safety outcomes (including cancer and death) between UK patients with rheumatoid arthritis (RA) starting any new biologic, biosimilar or other new targeted therapy with appropriate comparator cohorts already established in the BSRBR-RA. The data requested will allow the study team to link the hospital and treatment data already captured under the study consent with national cancer and death data on study participants to ensure the register has no missing data on these two major outcomes. This will then allow the study to understand whether there is any increased risk of cancer and death in patients receiving these drugs.

The BSRBR-RA study started in 2001 as a joint venture between the BSR, the pharmaceutical companies who manufacture the drugs, and epidemiologists at the University of Manchester with the aim to monitor the long-term safety of newer drugs prescribed in the NHS to treat patients with RA. The BSR currently receives restricted income from UK pharmaceutical companies, including Abbvie, Celltrion, Eli Lilly, Pfizer, Roche, Samsung Bioepis, Sandoz, Sanofi, UCB, and in the past has received income from Hospira, MSD and SOBI. This income finances a wholly separate contract between the BSR and the University of Manchester. The principal investigator and the BSRBR-RA team at the University of Manchester have full academic freedom and are able to work independently of pharmaceutical industry influence. All decisions concerning analyses, interpretation and publication are made autonomously of any industrial contribution. Members of the BSRBR-RA University of Manchester team, BSR trustees, committee members and staff complete an annual declaration in relation to conflicts of interest. With over 18 years of follow-up in some patients, the study continues to monitor for long-term risks of the original biologic therapies, whilst at the same time, monitoring the safety of newer drugs (including biosimilars and Janus kinase inhibitors (JAKs)) that have been later approved for use by the National Institute for Health and Care Excellence (NICE).

The data will only be used for the BSRBR-RA study and is not part of a wider project.

The data subjects for this study consist of 3 types of cohorts. These are as follows:

[1] Exposed Cohort

For each new biologic, biosimilar or other new advanced therapy drug, the exposed cohort are patients under the care of a consultant rheumatologist with rheumatoid arthritis newly starting therapy with that biologic, biosimilar or other new advanced therapy.

Recruitment is co-ordinated at a national level. The study is based in England, Wales, Scotland and Northern Ireland. Historically, patients without a diagnosis of RA (such as psoriatic arthritis, ankylosing spondylitis and other rheumatic conditions) were recruited to the study, but this has since been limited to RA only. Follow-up of these historic non-RA patients continues as per the process outlined in the study protocol. Due to the nature of the study, new cohorts open as new treatments are launched to market. A full up-to-date list of eligible treatments can be found here: http://bsrbr.org/hospitals/eligibility/

[2] Comparator Cohort 1

The first comparator cohort will be patients recruited to the BSRBR-RA with RA who have been registered within 6 months of first exposure to an established tumour necrosis factor inhibitor (TNFi) drug (e.g. Humira, Enbrel or Remicade).

[3] Comparator Cohort 2

The second comparator cohort is a historical RA cohort of patients treated with non-biologic DMARDs recruited to the BSRBR-RA from a select number of sites within the UK (recruited between 2002 and 2008). Patients who subsequently progressed to a biologic, biosimilar or other new targeted therapy will leave this cohort and be eligible to enter an ‘exposed’ cohort but only if the subject consents to being re-registered as an “exposed patient”.

The purpose of the study is to obtain comprehensive long safety data on patients receiving these new therapies, and in particular with reference to NHS Digital data, to observe if participants receiving these new drugs are at a higher risk of cancer or premature death compared to patients with similar disease not receiving these drugs. For this reason, the study requires Civil Registration Mortality data and cancer data for the consented participants taking part in the study. The additional data from NHS Digital will allow the study to have full outcome data on cancer and death and will enhance the BSRBR-RA dataset.

Recruitment commenced in 2001 and continues to recruit. As of June 2019, 24,000 individuals had consented to particpate in this study.

Notifications for mortality and cancer data is required from when the study began with the first participant in 2001. Under previous iterations of this Data Sharing Agreement, the study has received data from NHS Digital on these outcomes since the study started in 2001. The study currently has ethical approval to continue the long-term follow of study participants until 2028 and the data over a long period will play an important part in order to understand the long-term safety of the drugs used.

Although the study will capture most cancer and death outcome data for the study via the NHS hospitals, there are occasions where the study will not be told that one of these events has occurred (i.e. the rheumatology team were not aware and/or the patient may have been treated in a different hospital) and therefore flagging with NHS Digital will allow complete data on these important outcomes.

The University of Manchester and the British Society for Rheumatology are joint data controllers for the study. The University of Manchester is the only organisation which processes the data. The pharmaceutical companies who manufacture the drugs under study also process the BSRBR-RA study data but not the data provided by NHS Digital. The pharmaceutical companies have no influence over the dissemination of the results of this study.

The outputs for the BSRBR-RA study will include reports, submissions to peer reviewed journals and presentations and posters at relevant conferences. BSRBR-RA Study data will be combined with NHS Digital data to maximise data available and used in the outputs. Only aggregated data will be included in any study outputs and data will be safeguarded by ensuring that results which include small numbers which could identify study participants will be excluded.

Dissemination and communication of results to stakeholders includes regular review by BSRBR project boards, including the BSRBR Steering Committee and BSRBR Data Monitoring and Ethics Committees. The study also has active social media channels (Twitter/Facebook) and a comprehensive study website (www.bsrbr.org) which has areas aimed at Hospitals/Sites participating, study participants and researchers. In terms of exploitation of the results/outputs, the British Society for Rheumatology also has links back to the BSRBR study site, in addition to details and the process of applying to access study data for research purposes (https://www.rheumatology.org.uk/practice-quality/registers).

To date, over 60 original papers have been published on the BSRBR-RA data since 2001 and the research outputs from the BSRBR-RA study have refined the way anti-TNF therapies are prescribed in RA with a goal of maximising patient benefit and minimising patient risk. As an example, the research into the influence of anti-TNF therapy on incidence/risk of solid cancer is cited in the 2019 BSR DMARD safety guidelines in inflammatory arthritis in terms of reassuring patients that their risk of cancer is not increased by taking the medication. The guideline states that, “patients should be advised that there is no conclusive evidence for an increased risk of solid tumours or lymphoproliferative disease linked with biologic therapy, but that ongoing vigilance is required (grade 1A, SOA 99%)… Most observational studies using national registry data have also been reassuring; recent studies from BSRBR-RA … registries have all failed to show a significant association between anti-TNF use in RA and overall malignancy, with up to 52 549 patient-years exposure.”

The BSRBR-RA study releases occasional newsletters which can be accessed via the study website which describes the progress of the study and latest study findings in language targeted at study participants or individuals from a hospital site.

Planned future analysis/output will focus on two related and equally important research questions: (1) What is the long term risk of exposure to established biologic therapies? (2) What is the absolute and relative effectiveness and risk of newer biologic therapies?

(1) What is the long term risk of exposure to biologic therapies?

The first patient was enrolled into the BSRBR-RA in 2001. Therefore, over the next 5 years the study team will start to observe patients who have been receiving TNFi therapies for >20 years. This is an unexplored area due to the very nature of when these drugs were introduced. Hence, the BSRBR-RA can and will offer a unique insight into the very long term use of TNFi drugs, including treatment persistence and long-term safety, such as the late occurrence of malignancies. The presence of equally long follow-up in an untreated comparison cohort will add to these analyses.

Most of the University of Manchester’s long-term analysis has focussed on TNFi but the study now have over 10 years of follow-up data on patients receiving rituximab (MabThera). Within the dataset, RTX is the most common second line biologic drug in RA treatment, with over 5,800 exposures. This drug is different from other biologic therapies not only because of its mechanism of action but also because of its long half-life, making it challenging to study. Therefore, its use in daily practice is less well understood than other RA therapies. Outcomes of interest include long-term effectiveness and safety, including the safety of repeated dosing of MabThera. The University of Manchester plan to continue to study the risk of serious infection, and will also move to consider other adverse events such as malignancy and cardiovascular disease in patients treated with MabThera.

The University of Manchester will also analyse the data held within the register on work to provide further insight into the effect of response to biologics and related therapies on productivity at work (presenteeism) and the ability for patients to remain in work.

(2) What is the effectiveness and risk of newer biologic therapies?

IL6 inhibitors are one of the newer classes of biologic drugs on the market and therefore, their safety is less well understood. In the UK, it is primarily used as a second (or higher) line biologic therapy following TNFi. The BSRBR-RA has now accrued data from over 2,700 tocilizumab (TCZ) treated patients in the register. The risk of infection over the short term is already being studied as part of collaboration with King’s College London, but specifically a comparative study of infection risk between different choices of second line treatments, including a second TNFi and RTX, will be undertaken. There are concerns over the risk of cardiovascular disease due to aberrations in lipid levels with TCZ and therefore, the risk of MI and stroke following IL6 inhibition will also be investigated.

The most challenging analysis will be to study the risks of biosimilar therapies. The switch from originator Remicade and Enbrel to biosimilar products has been swift and extensive in the UK, despite a lack of supporting evidence about the safety of switching. The study team’s analyses in this area will include a comparison of first line biosimilar TNFi use with originator, using recent historical originator data as a comparison, as well as the safety of a switch programme. An analysis of outcomes after switching needs careful consideration due to inherent selection bias (patients who switch between originator and biosimilar have usually experienced a response to the originator and by definition, have not experienced a treatment limiting adverse event). This selection bias will have to be considered when changes in disease activity and occurrence of adverse events following a switch are analysed and the rich historical data within the BSRBR-RA should allow for this. The University of Manchester and BSR have already amended the study’s data collection forms to ensure it captures as much disease activity data at the point of switching as possible.

JAK inhibitors form a new oral therapeutic option in the treatment of RA. These drugs are now available within the NHS and over the next 5 years, the study team’s efforts will focus on recruitment of patients starting these therapies. Early analyses will include a descriptive study of how JAK inhibitors are being used in the UK and their early effectiveness in routine clinical practice. With the introduction of an oral targeted therapy for RA, the study team will use the opportunity to introduce simple patient questionnaires to gather further data about treatment adherence across modes of therapy.

As is often the case with research impact, there is a time delay between research publication and clinical impact. Therefore, much of the impact on clinical practice between 2011 and 2018 was from research published in the preceding years. However, the evidence-impact gap for the BSRBR-RA remains relatively small.

The BSRBR-RA has had a significant influence on clinical practice in the UK and more widely. Evidence emerging from the BSRBR-RA has fed directly into National Institute for Health and Care Excellence (NICE) technology appraisals (TA), UK and other clinical practice guidelines, and patient information sheets. These influences have resulted in more consistent prescribing across the country. The data held in the BSRBR-RA have also been used to study the impact of changes in clinical guidelines and safety warnings.

Evidence from the BSRBR-RA has contributed to several NICE TAs. For example, two UoM analyses from 2006 (1;2) demonstrated the benefits of combining TNFi treatments with continued background MTX (unless contraindicated); the study contributed directly to NICE TA195 (published 2010; updated with TA375, published 2016). Initial guidelines did not specify that MTX treatment should be continued (TA36). Subsequent analysis using the register demonstrated that, year on year from 2001-8, the proportion of patients continuing MTX with TNFi increased in the UK, with corresponding improvements in treatment responses (3).

Output from the BSRBR-RA has also contributed to BSR guidelines outlining eligibility criteria for TNFi. The analyses indicated that biologic drugs should no longer be reserved for patients with high disease activity, but will also benefit patients with moderate disease activity despite csDMARDs (4). Unfortunately the same eligibility criteria were not adopted by NICE in 2016, meaning it remains difficult to prescribe biologics in this setting.

Study research into the risk of intracellular infections such as listeria and salmonella (5) has led to new information being incorporated into Arthritis Research UK Drug Information Leaflets (provided to every patient in the UK considering TNFi therapies). It also prompted the FDA to update product labelling. Specifically, the information now warns of the risk of consuming undercooked or unpasteurised foods, similar to the advice provided to pregnant women. Analysis of the BSRBR-RA dataset in 2013 showed that since updating the Arthritis Research UK Drug Information Leaflets in 2006 there has been a 73% decrease in the rate of new cases of intracellular infection in RA patients exposed to TNFi in the UK from 2007-12 (6).

Study publications on outcomes among women exposed to anti-TNF therapy during pregnancy (7;8) have contributed to a significant change in national pregnancy guidelines. It is now stated that women can continue TNFi therapies into pregnancy if clearly needed (9), whereas previous labelling stated that treatment should be discontinued in the months leading up to conception (which often resulted in a disease flare). This is a major advance within rheumatology, as other non-biologic DMARDs, many with the risk of teratogenicity, are contra-indicated in pregnancy. This research contributes to the weight of evidence for safer options for disease control in the months leading up to conception and in early pregnancy.

One of the most common questions which both healthcare professionals and patients have is around whether biologic therapy increases the risk of cancer. The initial size of the original Enbrel, Remicade and Humira cohorts were chosen to ensure enough power to detect a doubling in the risk of lymphoma compared to csDMARD therapy alone. Reassuringly, all of the findings to date have shown that biologics do not appear to increase the risk of cancer in patients with no prior history of malignancy (10;11;12). These findings have been incorporated into national guidelines to help provide reassurance to both patients and their healthcare professionals.

A major challenge for all research studies is the dissemination and implementation of results. The close collaboration with BSR representatives has ensured that study data are disseminated as widely as possible, including to policy makers such as NICE. The study team also work with rheumatologists to generate ideas for new analyses based on clinically relevant questions (examples include moderate disease activity), refractory disease and interstitial lung disease). In recent years the study team have also used a lunchtime session at the annual BSR conference to promote the register including its scientific output. The study team have also convened a scientific session at the last 2 annual conferences to provide a more comprehensive review of the scientific output from the study.

The study have also established an ongoing and mutually beneficial collaboration with the National Rheumatoid Arthritis Society (NRAS) (a patient-led organisation) as one route of dissemination to patients. On average, 1-2 lay articles per year for the NRAS Magazine are written.

The UoM BSRBR-RA team also receive regular queries (average 3 per month) from doctors, nurses, pharmacists as well as via the NRAS helpline asking about evidence to help support treatment decisions. Often these queries are in areas for which little evidence exists, such as malignancy or pregnancy, but the collective UK experience can be helpful and reassuring to support these decisions.