Key points are not available for this paper at this time.
Airway hyperresponsiveness (AHR) is a key tenet in uncontrolled severe asthma and can be used to characterise the potential sensitivity of the airway smooth muscle to external triggers 1, assessed through either direct or indirect challenge. Indirect challenge, which includes mannitol, is more physiological, acting similarly to exogenous stimuli. Mannitol acts to increase osmolarity in the periciliary liquid, which in turn results in both mast cell degranulation and de novo synthesis of bronchoconstricting mediators. This leads to the release of inflammatory mediators such as histamine or leukotriene D4 alongside the recruitment of eosinophils, which indirectly induce bronchoconstriction 1. This process is demonstrated by the result of sputum eosinophilia to mannitol challenge and therefore reflects the close links between type 2 inflammation and indirect challenge 2. Bronchoconstriction is measured as the provocation dose (PD) of mannitol required to reduce the FEV1 by either 10% (PD10) or 15% (PD15) and is termed challenge sensitivity, as opposed to reactivity, which is the maximum fall in FEV1 divided by the final cumulative dose of mannitol. The introduction of biologics has raised the question as to whether complete asthma remission can be achieved in terms of abolition of AHR. Here, we have focused on AHR response to biologic therapy as assessed by indirect mannitol bronchial challenge. Four studies were identified as outlined in Table 1. Anti-TSLP agent tezepelumab was assessed in UPSTREAM 3 as a primary outcome and CASCADE 4 as an exploratory outcome, both using PD15 and including a placebo control arm. AHR response to anti-interleukin (IL)-5Ra benralizumab and anti-IL-4Ra dupilumab were assessed via BISA (benralizumab in severe asthma) 5 and DISA (dupilumab in severe asthma) 6, respectively, as primary outcomes as PD10; however, these were not placebo-controlled. Only BISA and DISA were performed exclusively in severe uncontrolled asthma patients. All four studies found a significant improvement in their treatment arm in the doubling difference (DD) with the lower 95% confidence interval exceeding the MCID in UPSTREAM, BISA, and DISA but not CASCADE (Figure 1). When compared to the placebo arms, CASCADE was found to be significant but failed to exceed the MCID, and UPSTREAM was not significant (p = 0.06). Mannitol non-responders who did not respond at the maximal cumulative mannitol dose of 635 mg comprised 45% of patients in UPSTREAM and CASCADE, 24% in BISA, and 61% of DISA at the end of treatment, thus censoring the true DD in all four studies. In this regard, challenge reactivity can provide further insight as the dose response ratio (RDR) as it does not censor the patients who reached 635 mg. UPSTREAM, BISA and DISA report the RDR DD as 2.18 (95% CI 1.36, 2.94); 2.32 (95% CI 1.00, 3.32) and 3.48 (95% CI 2.65, 4.32) respectively. As shown in other areas of asthma management, such as assessing response to inhaled corticosteroid 7, measuring the RDR allows for a clinically meaningful description of the slope of the dose-response curve when airway sensitivity is negative. Thus, the next logical steps of applying the RDR is to assess the effects of biologic therapies. Chan et al. indirectly matched a subcohort of 12 participants from each BISA and DISA (from a total of 21 and 23 respectively) based on their baseline PD10 with an in-between mean difference of 3 mg and found that the DD changes were 0.86 (0.12, 1.59) to benralizumab and 1.91 (95% CI 1.20, 2.63) to dupilumab with a DD difference of 1.06 (95% CI 0.09, 2.02) 8. Notably, this is in contrast when compared to the unmatched DD change in the two studies, where both the respective baselines exceeded the MCID of 1 DD with 2.1 (95% CI 1.0, 3.3) to benralizumab and 1.78 (95% CI 1.23–2.33) to dupilumab with a DD difference of 0.32 (95% CI −0.95, 1.59) (Figure 1). Participants receiving dupilumab exhibited a greater proportion of mannitol non-responders at 67% compared to 25% with benralizumab similar to their original studies. Furthermore, the DD in RDR was significantly greater with dupilumab at 2.61 (95% CI 1.57, 3.64) than with benralizumab 0.82 (95% CI 0.00, 1.66) with a DD difference between biologics of 1.78 (0.54, 3.03). This highlights the potential confounder when comparing studies with different baseline AHR as shown by the geometric mean PD values of 67 mg in BISA, 125 mg in DISA, 135 mg in UPSTREAM and 241 mg in CASCADE. A further factor to consider is the inhaled corticosteroid dose. A key potential factor in DISA was that participants also reduced their inhaled corticosteroid dose with a mean reduction in beclomethasone/formoterol 100/6 μg of 1.7 puffs/day (95% CI 0.7, 2.7), which in turn could potentiate AHR 1. One of the key limitations of BISA and DISA was the lack of a placebo arm. To allow an approximate comparison, we pooled the placebo groups of UPSTREAM and CASCADE, which experienced a DD change of 0.75 (95% CI 0.33, 1.18). This coincidence interval does not overlap with the 95% CI reported for either DISA or BISA. The lower 95% CI of both UPSTREAM and DISA exceeded the upper 95% CI of the pooled placebo group despite UPSTREAM not achieving this against its own placebo group. Given the markedly different results seen when participants are matched on the baseline PD10, one would expect this to also be applicable when comparing between treatment arms or to placebo. Given AHR challenge studies tend to involve small numbers, randomisation may not be sufficient to achieve two equally similar groups to facilitate a true comparison 9 which is then further blunted when participants become challenge non-responsive. In conclusion, benralizumab, dupilumab and tezepelumab have all been shown to significantly attenuate AHR. However, comparison between studies is inherently difficult given the differences in baselines, and the true DD shift is blunted when participants become mannitol non-responsive. RDR can be used to give further insight as a continuous measure of AHR, providing additional understanding into how flat the dose response curve is compared to that of a non-asthmatic response and thus should be included in future studies. Further direct baseline-matched, head-to-head studies are required to facilitate the comparison between biologics and the placebo, alongside work to establish the value of a normal RDR in response to treatment. Robert Greig: conceptualisation, data curation, formal analysis, writing – original draft. Philipp Suter: conceptualisation, writing – reviewing and editing. Rory Chan: writing – reviewing and editing. Brian Lipworth: conceptualisation, writing – reviewing and editing, supervision. The authors have nothing to report. Dr. Greig reports personal fees (talks) from AstraZeneca. Dr. Suter reports personal fees (talks) from AstraZeneca, personal fees (talks) from GSK, grants from Lung League Fribourg (Switzerland), and grants from Swiss Lung Foundation (Switzerland). Dr. Chan reports institutional grants awarded from Asthma+Lung UK, Chiesi, AstraZeneca and GSK; serving on advisory boards for AstraZeneca and Vitalograph; personal fees (talks and/or drafting educational material) from AstraZeneca, Chiesi, Thorasys and Vitalograph; and support attending meetings from AstraZeneca, Chiesi, NIOX, Sanofi-Regeneron and Vitalograph. Dr. Lipworth reports non-financial support (equipment) from GSK; grants, personal fees (consulting, talks and advisory board), other support (attending ATS and ERS) and from AstraZeneca; personal fees (talks and consulting) from Sanofi, personal fees (consulting, talks and advisory board) from Circassia in relation to the submitted work; grants, personal fees (consulting, talks, advisory board), other support (attending ERS) from Teva, personal fees (talks and consulting), grants and other support (attending ERS and BTS) from Chiesi, personal fees (consulting) from Lupin, personal fees (consulting) from Glenmark, personal fees (consulting) from Dr. Reddy, personal fees (consulting) from Sandoz; grants, personal fees (consulting, talks, advisory board), other support (attending BTS) from Boehringer Ingelheim, grants and personal fees (advisory board and talks) from Mylan outside of the submitted work; and the son of BJL is presently an employee of AstraZeneca. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Greig et al. (Thu,) studied this question.