Response to letter to the editor. Again on IMPACT: exacerbation after abrupt discontinuation of ICS and pneumonia in fluticasone furoate-containing FDCs

Abstract

Dear Editors, We read with interest the comments by Barnes et al. [1] on our expert review concerning the effect of triple inhaled corticosteroid (ICS)/long-acting β2 adrenoceptor agonist (LABA)/long-acting muscarinic receptor antagonist (LAMA) therapy in chronic obstructive pulmonary disease (COPD) patients [2]. However, many of the outlined points have already been questioned in several other response letters and editorials, by proposing always the same comments [3–6] that were already fully addressed or discussed in just as many replies [7–9]. It is well known that the IMPACT (InforMing the PAthway of COPD Treatment) study was not designed as an ICS withdrawal trial. In any case, it is unquestionable that ~70% of patients randomized to the LABA/LAMA group were abruptly switched from an ICS-containing therapy, as reported in Table S4 of the Supplementary appendix of the primary publication [10]. The matter of the exacerbation rate with LABA/LAMA was further corroborated by the post-hoc analysis of the IMPACT study concerning the effect of ICS withdrawal and baseline inhaled treatment on exacerbations [11]. The forest plot reported in Figure 2 by Han et al. [11] clearly indicated that, among non-ICS users at screening, no difference was detected between triple therapy and dual bronchodilation with respect to the annual rate of moderate/severe exacerbations (rate ratio 0.88, 95%CI 0.76–1.03; p = 0.115) [11]. In contrast, COPD patients who were abruptly withdrawn from ICS therapy at randomization and assigned to the LABA/LAMA group experienced a higher frequency of moderate/severe exacerbations compared with triple therapy [11]. This evidence clearly indicates that abrupt ICS discontinuation may have deleterious effects on the risk of COPD exacerbations and that the beneficial action of triple therapy could have been overestimated [12,13]. We agree that not only the IMPACT and TRIBUTE (extrafine inhaled triple therapy vs dual bronchodilator therapy in COPD) studies included patients with a past diagnosis of asthma, but also the KRONOS (triple therapy with budesonide/glycopyrrolate/formoterol fumarate with cosuspension delivery technology vs dual therapies in chronic obstructive pulmonary disease) and ETHOS (Efficacy and Safety of Triple Therapy in Obstructive Lung Disease) studies. However, it is important to highlight that, as we previously reported [14], up to 20% of COPD or asthmatic patients may display a mixed picture of overlapping clinical features of both COPD and asthma [15]. A large number of COPD patients enrolled in the IMPACT study presented physiological and/or biological characteristics of asthma at baseline (from 18% to 57%) [10] and even 20% of them presented a very high blood eosinophil level (≥310 cells·μL) [16]. The inclusion of patients with an asthma-like profile may have potentially contributed to the higher exacerbation and mortality rates in the LABA/ LAMA group [14,17,18], and this could also explain the discrepancy with the FLAME (Indacaterol–Glycopyrronium vs Salmeterol–Fluticasone for COPD) study [19], which excluded COPD patients with a history or features of asthma and showed superiority of LABA/LAMA over ICS/ LABA for the prevention of COPD exacerbations. With regard to the differential benefit/risk profile of triple therapies, in our expert review we have already reported the efficacy/safety ratio of triple fixed-dose combinations (FDCs) vs dual bronchodilator FDCs included in the ETHOS, KRONOS, IMPACT, and TRIBUTE studies [2]. Specific details are shown in Table 3 of our publication [2]. Interestingly, it resulted that triple combinations including budesonide (BUD) or beclomethasone dipropionate (BDP) were characterized by a superior efficacy/safety profile than those including fluticasone furoate (FF), as in the IMPACT study [2]. We disagree with the statement that ‘the benefit/risk ratio for all ICSs is very similar’ [1], instead there is the evidence of intra-class differences between ICS compounds, especially concerning the risk of pneumonia. The TRIBUTE and TRILOGY (single inhaler triple therapy vs ICS plus LABA therapy for COPD) studies demonstrated that BDP-containing therapies reduced the rate of moderate to severe exacerbations, by 15% and 23%, respectively, with no difference in the incidence of pneumonia compared with dual bronchodilation [20,21]. In contrast, the IMPACT study showed that FF administered in triple FDC was associated with a 53% greater risk of pneumonia compared