Efficacy and safety of bronchial thermoplasty in clinical practice: a prospective, longitudinal, cohort study using evidence from the UK Severe Asthma Registry

Objectives Use data from the UK Severe Asthma Registry (UKSAR) to assess the efficacy and safety of bronchial thermoplasty (BT) in routine UK clinical practice and to identify characteristics of ‘responders’. Design Prospective, longitudinal, cohort, multicentre registry study. Setting All (11) UK centres performing BT. Participants and intervention Patients receiving BT in the UK between 01/06/2011 and 30/09/2016 who had consented to data entry into UKSAR (n=133). Efficacy data were available for 86 patients with a BT baseline and at least one follow-up record. Safety data were available for 131 patients with at least one BT procedure record. Primary and secondary outcome measures Efficacy: AQLQ, ACQ, EuroQol, HADS anxiety and HADS depression scores, FEV1 (% predicted), rescue steroid courses, unscheduled healthcare visits (A&E/Asthma clinic/GP), hospital admissions and days lost from work/school. Safety: peri-procedural events, device problems and any other safety-related findings. Responder analysis: differences in baseline characteristics of ‘responders’ (≥0.5 increase in AQLQ at 12 months) and ‘non-responders’. Results Following Bonferroni correction for paired comparisons, mean improvement in AQLQ at 12 months follow-up compared with BT baseline was statistically and clinically significant (0.75, n=28, p=0.0003). Median reduction in hospital admissions/year after 24 months follow-up was also significant (−1.0, n=26, p<0.0001). No deterioration in FEV1 was observed. From 28 patients with AQLQ data at BTBL and 12-month follow-up, there was some evidence that lower age may predict AQLQ improvement. 18.9% (70/370) of procedures and 44.5% (57/128) of patients were affected by an adverse event; only a minority were considered serious. Conclusions Improvement in AQLQ is consistent with similar findings from clinical trials. Other efficacy outcomes demonstrated improving trends without reaching statistical significance. Missing follow-up data impacted this study but multiple imputation confirmed observed AQLQ improvement. The safety review suggested BT is being performed safely in the UK.


GENERAL COMMENTS
The study sets out to describe the clinical experience of bronchial thermoplasty within UK severe asthma centres using the UK severe asthma registry. The manuscript is well written, clearly set out and straight forward to read. The data collected by UKSAR is comprehensive for a registry and the discussion is balanced with mention made of the obvious limitations associated with using registry data. The advantage of this study is that this is real life data and the UKSAR has a track record of collecting robust data over many years. There are obvious limitations which have been identified in the paper such as the lack of comparator group and the fact that improvements in outcomes may relate to other aspects of specialist care the patients receive within a UK severe asthma service. However, there is a sustained improvement in AQLQ and HCU. I would be interested in comparisons with QoL data using patients on biologics which the UKSAR is no doubt also collecting. Current biologic trials in severe asthma have struggled to demonstrate clinically significant improvements in QoL. Although most patients gain benefit in QoL with BT I note some patients have a significant worsening in AQLQ and there is still some uncertainty as to which phenotype of severe asthma is more likely to benefit from this treatment. This study provides no clarity on that issue for clinicians. The paper highlights the current UK experience with bronchial thermoplasty. Given the comprehensive nature of the UK severe asthma registry I would agree that this gives the best current data with which to review BT. There are two aspects to the study. 1. efficacy N=86 2. safety N=128 It is pleasing that health related quality of life has been given prominence as a primary outcome measure as it is often neglected by pharma. I think this fits well with a patient centred approach to asthma management which is gaining prominence particularly in European circles. The safety data is useful and well presented. As with any new novel treatment there is some caution but I think with BT this has been more so given it is a mechanical intervention (one off) rather than a drug that can be stopped at any point. I think readers will find this paper helpful and reassuring.
Major points: 1. The patient demographics does not state how many patients were on a biologic. Were any patients started on a biologic agent post BT as this would obviously potentially affect outcomes? 2.
Were any patients on maintenance oral steroids and if so what was the OCS reduction?
The quality of life measures used (disease specific = AQLQ, generic = Euroqol could be slightly more detailed. Is it the mini AQLQ or the 32 question AQLQ? This has a bearing on MCID accuracy. Is it the EQ5D 5L that has been used (was it just the descriptive or was there VAS data)? When considering the responder analysis it may be useful to perhaps discuss some of the potential limitations. Given that the AQLQ was not designed specifically for severe asthma and the calculation of the MCID was made using the 32 question AQLQ in patients with mild asthma(many of whom were not even on an ICS). The MCID was then extrapolated to the mini-AQLQ. Therefore, it's worth bearing in mind that the MCID may not be completely accurate in this patient population and with the mini-AQLQ (i.e. it's more likely to underestimate responders as the bar may well be too high for what patients perceive as a clinically meaningful response). The other issue with historical MCID calculations is whether the patients' perceptions of a clinically meaningful response change with severity of disease. The issue of transitivityi.e., the assumption that an MCID remains the same over the whole range of the scale. E.g. If a patient already has a good QoL will they value improvement to the same degree as a patient with a poor QoL. This may be more relevant given that the baseline AQLQ in the BT trials ranges from 3.9 to 5.6/7 and 3.6 in UKSAR. I raise it as a point worth mentioning in the paper if one is going to use a HRQoL measure effectively particularly as a primary outcome measure and undertake responder analyses based on the MCID. It may be worth just stating that the AQLQ may not be as sensitive in severe asthma and therefore may underestimate effect.

2.
Is there a possible health economic analysis given the QoL benefit and HCU reduction? The benefits appear to be sustained to 2 years although the lack of complete follow up data is a limitation.
Overall I think this paper provides a good contribution to the literature.

REVIEWER
Rita Amaral Faculdade de Medicina da Universidade do Porto, Portugal REVIEW RETURNED 11-Feb-2019

GENERAL COMMENTS
This study aimed to assess the long-term efficacy for Bronchial Thermoplasty in UK clinical practice, providing a perspective from clinical practice to identify the characteristics of patients who could most benefit from BT.
It is an important topic and the study provide a comprehensive coverage of UK clinical practice and BT procedures. However, there are some methodologic issues that should be clarify and improved. 1. The introduction is well written but last paragraph needs to be a bit shorter. Specifically, the last sentence appears to belong to the methods section rather the paragraph where the study aims/hypotheses are described. 2. Inclusion criteria and data preprocessing are well described. However, in statistical analysis the authors only used parametric tests for continuous variables. Did the authors check the data distribution? If the normality of the distribution is not assumed, then non-parametric test must be applied, and results must be presented accordingly.
3. To facilitate the reader, I suggest the presentation of a graph with the long-term changes of the main continuous variables (e.g. error bar with 95% confidence intervals. 4. Using only measures of central tendency is not the most adequate to describe data. Please add the respective measures of dispersion. 5. Despite the authors described the use of cut-offs, such as the ≥0.5 increase in AQLQ at month follow-up, it is not clear that this is considered minimal clinically important difference. Did the authors considered the concept of the MCID? If yes, please mention in the manuscript and discuss accordingly. 6. I understand that the selection bias was minimized, however, the authors must discuss the generalization of the findings or present a power calculation. This paper is useful, since it supplies some answers which are currently not adequately addressed in the literature: The outcomes from bronchial thermoplasty in patients with SEVERE asthma.

VERSION 1 -AUTHOR RESPONSE
Overall, the description of the methods, results and conclusions are accurate and balanced. The conclusions in particular are well-drawn.
This manuscript will be useful for physicians treating patients with severe asthma, and will help them to explain the current state of knowledge of patient selection and benefits to patients.
In particular, this work helps put the AIR-2 trial into context, since AIR-2 patients were milder than the patients treated in this UK study.
A particular strength is the pan-UK selection.
Dear Dr Gore, Thank you for your review and positive comments. I raise the following points: Was there any data on median steroid dose at BTBL and follow-up?
Measuring steroid dose before and after BT was not one of our prospective outcome measures, however we acknowledge that this may be confounding factor and have therefore added a section for 'Potential confounding factors' into the methods and results. We have included a comparison of the proportions of patients who were taking maintenance oral steroids at BTBL and 12 month follow-up as well as comparing the median steroid dose for those patients who were taking oral steroids at both time points. We have also included the numbers of patients who either remained on oral steroids, off oral steroids or started/stopped oral steroids between BTBL and 12 month follow-up. Dear Dr Cardenas-Garcia, Thank you for your review and positive comments.
-In Table 6, it is mentioned that 4 patients had incomplete BT sessions (370 procedures in 128 patients) , and they should be excluded, thus statistical analysis might need to be performed again as might impact conclusions. If authors disagree, please explain.
The 4 patients who had incomplete BT sessions were included as we followed ITT analysis rather than per protocol for the study. They also fulfilled the inclusion criteria, having received BT treatment, had a valid BT baseline (BTBL) record and at least one follow-up record. ITT analysis is preferable in this study as it avoids overoptimistic estimates of the efficacy of the treatment.
Thank you for pointing this out. We have modified 'sob' to 'shortness of breath' -Any records of how BT was performed? General anesthesia or conscious sedation. This might impact the number of activations (and indirectly the efficacy) ? Please discuss this.
We acknowledge that the use of general anaesthesia is relevant and could potentially impact the number of activations. There were 9 procedures in 8 patients reported to have used general anaesthesia. The median number of activations for all procedures was 34 (min 2max 115). For the procedures carried out under general anaesthesia the median number of activations was 38 (min 12max 50). We have added this information into the 'Safety outcomes' section.
Reviewer: 3 Reviewer Name: Matthew Masoli Institution and Country: Royal Devon and Exeter Hospital, UK Please state any competing interests or state 'None declared': none The study sets out to describe the clinical experience of bronchial thermoplasty within UK severe asthma centres using the UK severe asthma registry. The manuscript is well written, clearly set out and straight forward to read. The data collected by UKSAR is comprehensive for a registry and the discussion is balanced with mention made of the obvious limitations associated with using registry data. The advantage of this study is that this is real life data and the UKSAR has a track record of collecting robust data over many years. There are obvious limitations which have been identified in the paper such as the lack of comparator group and the fact that improvements in outcomes may relate to other aspects of specialist care the patients receive within a UK severe asthma service. However, there is a sustained improvement in AQLQ and HCU. I would be interested in comparisons with QoL data using patients on biologics which the UKSAR is no doubt also collecting. Current biologic trials in severe asthma have struggled to demonstrate clinically significant improvements in QoL.
Although most patients gain benefit in QoL with BT I note some patients have a significant worsening in AQLQ and there is still some uncertainty as to which phenotype of severe asthma is more likely to benefit from this treatment. This study provides no clarity on that issue for clinicians. The paper highlights the current UK experience with bronchial thermoplasty. Given the comprehensive nature of the UK severe asthma registry I would agree that this gives the best current data with which to review BT. There are two aspects to the study. 1. efficacy N=86 2. safety N=128 It is pleasing that health related quality of life has been given prominence as a primary outcome measure as it is often neglected by pharma. I think this fits well with a patient centred approach to asthma management which is gaining prominence particularly in European circles. The safety data is useful and well presented. As with any new novel treatment there is some caution but I think with BT this has been more so given it is a mechanical intervention (one off) rather than a drug that can be stopped at any point. I think readers will find this paper helpful and reassuring.
Dear Dr Masoli, Thank you for your review and positive comments.
Major points: 1. The patient demographics does not state how many patients were on a biologic. Were any patients started on a biologic agent post BT as this would obviously potentially affect outcomes?
We acknowledge that this is a potential confounding factor and have added details into a new 'Potential confounding factors' section in the methods and results. There is a registry (Yes/No) field for anti-IgE medication, so we have included a comparison of the proportions of patients for whom this was ticked at BTBL and 12 month follow-up.

Were any patients on maintenance oral steroids and if so what was the OCS reduction?
Some patients were on maintenance oral steroids, but measuring steroid dose before and after BT was not one of our pre-set objectives. However we acknowledge that this is also a potential confounding factor and have included it in the new 'Potential confoundingfactors' section in the methods and results. We have included a comparison of the proportions of patients who were taking maintenance oral steroids at BTBL and 12 month follow-up as well as comparing the median steroid dose for those patients who were taking oral steroids at both time points. We have also included the numbers of patients who either remained on oral steroids, off oral steroids or started/stopped oral steroids between BTBL and 12 month follow-up.
Minor points: 1. The quality of life measures used (disease specific = AQLQ, generic = Euroqol could be slightly more detailed. Is it the mini AQLQ or the 32 question AQLQ? This has a bearing on MCID accuracy.
The 32 question AQLQ was used. We have added this detail to the 'Methods' section.
Is it the EQ5D 5L that has been used (was it just the descriptive or was there VAS data)?
The EQ-5D-3L was used. This work was commissioned by NICE, and as the EQ-5D VAS is not generally accepted by decision-making bodies including NICE we did not include it in our study. Only the descriptive index score is included in the analysis.
When considering the responder analysis it may be useful to perhaps discuss some of the potential limitations. Given that the AQLQ was not designed specifically for severe asthma and the calculation of the MCID was made using the 32 question AQLQ in patients with mild asthma(many of whom were not even on an ICS). The MCID was then extrapolated to the mini-AQLQ. Therefore, it's worth bearing in mind that the MCID may not be completely accurate in this patient population and with the mini-AQLQ (i.e. it's more likely to underestimate responders as the bar may well be too high for what patients perceive as a clinically meaningful response). The other issue with historical MCID calculations is whether the patients' perceptions of a clinically meaningful response change with severity of disease. The issue of transitivityi.e., the assumption that an MCID remains the same over the whole range of the scale. E.g. If a patient already has a good QoL will they value improvement to the same degree as a patient with a poor QoL. This may be more relevant given that the baseline AQLQ in the BT trials ranges from 3.9 to 5.6/7 and 3.6 in UKSAR. I raise it as a point worth mentioning in the paper if one is going to use a HRQoL measure effectively particularly as a primary outcome measure and undertake responder analyses based on the MCID. It may be worth just stating that the AQLQ may not be as sensitive in severe asthma and therefore may underestimate effect.
We agree that this is a very relevant point and we have added a paragraph into the discussion.
2. Is there a possible health economic analysis given the QoL benefit and HCU reduction? The benefits appear to be sustained to 2 years although the lack of complete follow up data is a limitation.
Yes there is scope for a health economic analysis although that is beyond the scope of this study. We have added a sentence to this effect in the conclusion.
Overall I think this paper provides a good contribution to the literature. This study aimed to assess the long-term efficacy for Bronchial Thermoplasty in UK clinical practice, providing a perspective from clinical practice to identify the characteristics of patients who could most benefit from BT.