Moving the Dial on Airway Inflammation in Response to Trikafta in Adolescents with Cystic Fibrosis

To the Editor:

Cystic fibrosis (CF) lung disease is characterized by excessive and sustained airway inflammation (1). The CFTR (cystic fibrosis transmembrane conductance regulator) highly effective modulator therapy (HEMT) elexacaftor, tezacaftor, and ivacaftor (ETI) demonstrated remarkable pulmonary improvement in patients with CF (2). Although the respiratory benefit reflects a better airway surface hydration and improved mucociliary clearance, data are missing regarding the effect of ETI on inflammation, specifically in patients with relatively mild lung disease. The aim of the current study is to assess the short-term effect of ETI on lung inflammation in adolescents with CF.

Methods

Study design

Patients with CF were enrolled in a multicenter real-life longitudinal study (NCT04301856). Data were collected at initiation (M0) and 1 month (M1). This included antibiotic treatment, pulmonary function tests, sweat tests, inflammatory markers in blood sampling (C-reactive protein [CRP], leukocytes, and polymorphonuclear neutrophil [PMN] counts), sputum cultures, and inflammatory biomarkers. The study was approved by the Centre Institutional Review Board. All parents gave informed consent.

Sputum sample preparation

Sputum was collected by sputum induction. Sputum samples were immediately placed on ice and frozen at −80°C. After thawing, 200 μl of sputum was processed as per laboratory protocol, and supernatant was frozen at −80°C until analysis (3, 4).

Sputum inflammatory biomarkers

Neutrophil elastase (NE) activity was measured using a fluorometric assay (Sigma-Aldrich). IL-1β, IL-4, IL-6, IL-8, IFN-γ, and TNF-α (tumor necrosis factor α) concentrations were measured using V-PLEX Viral Panel 2 Human Kit, and V-PLEX Human IL-8 Kit (Meso Scale Discovery, Inc.). All tests were performed in triplicate.

Statistical analyses

Data were expressed as median (interquartile range). Comparisons were analyzed using the Fisher test, paired Student’s t test, ANOVA, or simple regression analysis. FEV₁ change at M1 was quantified by its relative change from baseline.

Results

Patient population

Seventy-six European White adolescents (33 girls, median [interquartile range] age, 15 [3] yr), were enrolled from seven centers across France. Forty-six patients were p.Phe508del homozygous and had received Orkambi (lumacaftor/ivacaftor) before enrollment; the 30 remaining patients were compound heterozygous for F508del and a minimal function mutation and had not been previously treated with CFTR modulators.

Most of the patients presented with mild lung disease, as assessed by a median predicted FEV₁ (ppFEV₁) of 76% (Table 1). Patients colonized with Pseudomonas aeruginosa (n = 18), Stenotrophomonas maltophilia, or Achromobacter xylosoxidans (n = 17) had a significantly decreased ppFEV₁ (71% [18%] vs. 80% [20%]; P = 0.04). Thirty-one patients received oral or intravenous (IV) antibiotic therapy at initiation of ETI because of bronchial exacerbation.

Table 1.

Change in Clinical, Respiratory, and Bacteriological Endpoints and Inflammatory Biomarkers

Parameters	M0	M1	P Value
Clinical and microbiological endpoints
Weight, kg	50 (11.7)	51.9 (10.2)	<0.0001
Sweat chloride, mEq/L	92.5 (29.5)	44.0 (30)	<0.0001
FEV1, L	2.4 (0.9)	3.1 (1.1)	<0.0001
FEV1, pp	76.0 (21)	96.0 (24)	<0.0001
FVC, L	3.3 (1.1)	3.7 (1.2)	<0.0001
FVC, pp	90.0 (16)	102.0 (18)	<0.0001
FEF25–75, L	2.3 (1.4)	3.3 (1.7)	<0.0001
FEF25–75, pp	58.0 (40.5)	86.5 (46.7)	<0.0001
Bronchial colonization
Staphylococcus aureus	55	49	0.007
Pseudomonas aeruginosa	17	12	<0.0001
Achromobacter xylosoxidans or Stenotrophomonas maltophilia	17	3	0.01
Blood inflammatory biomarkers
Leukocytes, nb/μl	7,465 (3,192)	6,400 (2,595)	<0.0001
Polymorphonuclear neutrophil count, nb/μl	4,035 (2,810)	3,100 (1,846)	<0.0001
CRP, mg/L	0.5 (3.4)	0.0 (0.2)	0.0002
Sputum inflammatory biomarkers
Neutrophil elastase, μg/ml	23.4 (100.8)	2.1 (4.6)	<0.0001
TNF-α, pg/ml	12.4 (36.75)	2.6 (11.4)	0.15
IFNy, pg/ml	3.8 (9.1)	4.0 (8.4)	0.25
IL-1β, pg/ml	132.2 (482.1)	22.9 (76.5)	0.02
IL-4, pg/ml	0.3 (0.5)	0.2 (0.4)	0.75
IL-6, pg/ml	7.0 (28.1)	4.9 (17.4)	0.28
IL-8, pg/ml	25,682.2 (70,448.2)	10,969.9 (28,618.9)	0.006

Definition of abbreviations: CRP = C-reactive protein; FEF_25–75 = forced mid-expiratory flow; M0 = month 0; M1 = month 1; nb = number; pp = percentage predicted; TNF = tumor necrosis factor.

Data are given as median (interquartile range) or bronchial colonization refers to the chronic identification of bacteria in sputum.

Lung function and clinical parameters

Patients presented a significant improvement after 1 month of ETI treatment relative to baseline in both weight (+1.1 [2.9] kg) and FEV₁ (+24% [31%]), and 52 patients demonstrated a decrease in sweat test below 60 mmol/L (Table 1). Five patients with chronic colonization for P. aeruginosa and 14 with A. xylosoxidans or S. maltophilia demonstrated negative cultures 1 month after ETI initiation. The change in ppFEV₁ was not significantly different in patients previously treated with Orkambi (+21% [32%] vs. +29% [23%] for patients naive to CFTR modulators) or those receiving antibiotics at ETI initiation (+21% [24%] vs. +28% [30%] in untreated patients). ppFEV₁ was significantly increased in patients with P. aeruginosa, A. xylosoxidans, or S. maltophilia in sputum at M0 (+32% [28%] vs. +20% [23%] in noncolonized patients; P = 0.03).

Inflammation biomarkers

At baseline, blood concentrations of CRP, leukocytes, and PMN count were within the normal range in contrast to the sputum inflammatory markers, which were elevated (5). ppFEV₁ was negatively correlated with sputum biomarkers, including NE (P = 0.0006) and IL-8 (P = 0.009), as well as with blood CRP (P = 0.00004), leukocytes (P = 0.00004), and PMN counts (P = 0.01).

All blood and sputum inflammatory biomarkers decreased at M1. The difference was significant for NE, IL-1β, and IL-8 in sputum and CRP, leukocytes, and PMN count in blood (Figure 1 and Table 1). FEV₁ change was significantly correlated with that of sputum NE (P = 0.02), blood CRP (P = 0.04), leukocytes (P = 0.0007), and PMN count (P = 0.0002), and nearly at the significant concentration for sputum IL-8 (P = 0.07). Interestingly, some patients did not experience respiratory improvement despite a significant change in inflammatory biomarkers. There was no correlation between changes in sweat test and blood or sputum inflammatory biomarkers. Change in NE at 1 month was not significantly affected by previous treatment with Orkambi (−16 [89] μg/ml vs. −17 [98] μg/ml), antibiotic course at initiation (−16 [80] μg/ml vs. −19 [123] μg/ml). Although baseline NE was more elevated in patients with P. aeruginosa, A. xylosoxidans, or S. maltophilia (66 [111] μg/ml vs. 17 [83] μg/ml; P = 0.1), negativation of sputum culture for these bacteria at M1 did not significantly impact NE change (−52 [136] μg/ml vs. −81 [68] μg/ml; not significant). A similar profile was observed for the other biomarkers.

Figure 1.

Concentration of sputum NE, IL-1β, IL-8, and blood CRP, PMN, and leukocyte counts at baseline (M0) and M1 treatment with elexacaftor, tezacaftor, and ivacaftor. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. CRP = C-reactive protein; M0 = month 0; M1 = month 1; NE = neutrophil elastase; PMN = polymorphonuclear neutrophil.

Discussion

This study in adolescents with relatively mild lung CF disease shows a decrease in sputum and blood inflammatory markers at 1 month after ETI initiation.

There are contradictory reports regarding the effect of HEMT on inflammation (6–8). Our observation that NE, IL-8, and IL-1β decrease significantly in sputum in parallel to CRP and PMN count in blood provides some evidence that ETI blunts neutrophil-derived inflammation in the short term. Once in the CF lung, PMNs acquire a modified phenotype that promotes NE release either free in the airway or bound to PMNs (4). As previously reported for circulating PMNs (9), our results suggest that ETI may also reverse this dysregulation in the lung. Importantly, those changes occurred regardless of whether patients were chronically infected with P. aeruginosa, A. xylosoxidans, or S. maltophilia or had been treated with antibiotics at initiation, suggesting an antiinflammatory effect, independent of infection.

Given the strong association between neutrophilic inflammation and progression of lung damage in early life, our results may suggest that HEMT may prevent or at least attenuate longer-term pulmonary decline (10). An important point is that, at the individual level, the improvement of ppFEV₁ was not directly associated with the variation of the inflammatory markers. This can be the result of irreversible damage. Conversely, patients with remaining significant amounts of NE may require adjunctive anti-elastase therapy. This illustrates the complexity of understanding the exact targets of HEMT on the inflammatory cascade in CF airways and how restoration of CFTR function may affect inflammation and vice versa. This also outlines the necessity of longitudinal studies in the CF population to assess whether this antiinflammatory effect is sustained.