Azithromycin may antagonize inhaled tobramycin when targeting Pseudomonas aeruginosa in cystic fibrosis

Abstract

Rationale: Recent studies of inhaled tobramycin in subjects with cystic fibrosis (CF) find less clinical improvement than previously observed. Nonhuman data suggest that in some strains of Pseudomonas aeruginosa, azithromycin can antagonize tobramycin.

Objectives: We tested the hypothesis that concomitant azithromycin use correlates with less improvement in key outcome measures in subjects receiving inhaled tobramycin while not affecting those receiving a comparative, nonaminoglycoside inhaled antibiotic.

Methods: We studied a cohort of 263 subjects with CF enrolled in a recent clinical trial comparing inhaled tobramycin with aztreonam lysine. We performed a secondary analysis to examine key clinical and microbiologic outcomes based on concomitant, chronic azithromycin use at enrollment.

Measurements and main results: The cohort randomized to inhaled tobramycin and reporting azithromycin use showed a significant decrease in the percent predicted FEV1 after one and three courses of inhaled tobramycin when compared with those not reporting azithromycin use (28 d: -0.51 vs. 3.43%, P < 0.01; 140 d: -1.87 vs. 6.07%, P < 0.01). Combined azithromycin and inhaled tobramycin use was also associated with earlier need for additional antibiotics, lesser improvement in disease-related quality of life, and a trend toward less reduction in sputum P. aeruginosa density. Subjects randomized to inhaled aztreonam lysine had significantly greater improvement in these outcome measures, which were unaffected by concomitant azithromycin use. Outcomes in those not using azithromycin who received inhaled tobramycin were not significantly different from subjects receiving aztreonam lysine. Azithromycin also antagonized tobramycin but not aztreonam lysine in 40% of P. aeruginosa clinical isolates tested in vitro.

Conclusions: Oral azithromycin may antagonize the therapeutic benefits of inhaled tobramycin in subjects with CF with P. aeruginosa airway infection.

Figure 1.

(a) Both absolute and relative change in FEV₁% predicted during the first course of inhaled antibiotic (Day 0–28) are shown for subjects randomized to inhaled tobramycin, with or without azithromycin use. (b) Those groups receiving inhaled aztreonam lysine had similar improvement in FEV₁%, regardless of azithromycin status. Tobramycin-azithromycin (TBRA-AZM) group significantly differed in absolute change in FEV₁% predicted from TBRA group (P = 0.0021), aztreonam lysine (AZTR) group (P < 0.0001), and AZTR-AZM group (P = 0.0002). (c) Relative change in FEV₁% over three courses of inhaled antibiotics (Day 0–140), including all four groups analyzed. TBRA-AZM group significantly different from TBRA group (P = 0.0028), AZTR group (P = 0.0007), and AZTR-AZM group (P = 0.0005). Columns represent mean (SEM). Regression model adjusted for baseline FEV₁.

Figure 2.

Self-report of disease related quality of life using the Cystic Fibrosis Questionnaire-Revised Respiratory Domain Symptom Score (CFQ-R RSS) was included in this study. (a) Change in CFQ-R RSS during first course of inhaled antibiotic. (b) Average change over three courses of inhaled antibiotics. Subjects with combined azithromycin and inhaled tobramycin failed to reach the clinically meaningful threshold of 5.5 points during the first course or averaged over the three courses of inhaled antibiotics. The other three groups had similar improvement in this outcome, beyond 5.5 points after one or averaged over three courses. After controlling for known baseline characteristics, the difference between groups randomized to inhaled tobramycin, based on azithromycin use, was not statistically significant but trended in favor of those subjects without azithromycin use (P < 0.10 over three courses); model covariates include age, sex, baseline FEV₁%; columns represent mean (SEM). AZM = azithromycin; AZTR = aztreonam lysine; TBRA = tobramycin.

Figure 3.

Time to need additional antibiotics for Pseudomonas aeruginosa was analyzed as a surrogate for pulmonary exacerbation. In our analysis, we find that the group with combined azithromycin and inhaled tobramycin required additional antibiotic treatment significantly sooner when compared with all other groups. The tobramycin-azithromycin (TBRA-AZM) group was significantly different from TBRA group (P = 0.0116), AZTR group (P < 0.0001), and aztreonam lysine (AZTR)-AZM group (P = 0.0011). P value from Cox proportional hazards model adjusted for age, sex, and baseline FEV₁%.

Figure 4.

The average reduction in sputum Pseudomonas aeruginosa density reported in cfu log₁₀/g is shown. (a) Mean change while receiving inhaled antibiotics during each of the three courses included in this trial. Virtually no change was observed in the group with combined azithromycin and inhaled tobramycin. The tobramycin-azithromycin (TBRA-AZM) group was significantly different from aztreonam lysine (AZTR)-AZM group (P = 0.008) and AZTR group (P = 0.004) and approached the predefined cutoff for significance when compared with the TBRA group (P = 0.0675). The far right column shows the average reduction in cfu density in subjects reporting azithromycin who crossed over to an extension phase with inhaled aztreonam lysine, after receiving inhaled tobramycin in the primary study. More direct comparisons of only these subjects participating in the crossover extension phase are presented in the online supplement. Regression model adjusted for age, sex, and baseline sputum bacterial density. (b) Group means for sputum P. aeruginosa density across three courses of inhaled antibiotic (from Day 0–140) in subjects randomized to inhaled tobramycin. A significant reduction was observed in the group without azithromycin use, and no change occurred in the group with azithromycin use. The difference in this change, when compared between the two groups and controlled for known covariates, was not statistically significant (P = 0.08).