Abstract
We conducted a double-blind, placebo-controlled, multicenter, randomized trial to test the hypothesis that 300 mg of tobramycin solution for inhalation administered twice daily for 28 days would be safe and result in a profound decrease in Pseudomonas aeruginosa (Pa) density from the lower airway of young children with cystic fibrosis. Ninety-eight subjects were to be randomized; however, the trial was stopped early because of evidence of a significant microbiological treatment effect. Twenty-one children under age 6 years were randomized (8 active; 13 placebo) and underwent bronchoalveolar lavage at baseline and on Day 28. There was a significant difference between treatment groups in the reduction in Pa density; no Pa was detected on Day 28 in 8 of 8 active group patients compared with 1 of 13 placebo group patients. We observed no differences between treatment groups for clinical indices, markers of inflammation, or incidence of adverse events. No abnormalities in serum creatinine or audiometry and no episodes of significant bronchospasm were observed in association with active treatment. We conclude that 28 days of tobramycin solution for inhalation of 300 mg twice daily is safe and effective for significant reduction of lower airway Pa density in young children with cystic fibrosis.
Morbidity and mortality in patients with cystic fibrosis (CF) are primarily attributable to progressive obstructive pulmonary disease in conjunction with chronic Pseudomonas aeruginosa (Pa) endobronchial infection and an intense neutrophilic inflammatory response (1). Moreover, among young children with CF, increased morbidity and mortality are observed in association with early Pa infection (2, 3). The prevalence of Pa infection increases with age, with positive respiratory tract cultures reported for up to 30% of infants, 30–40% of children 2–10 years of age, and 60–80% of adolescents and adults with CF (4). Chronic Pa infection results in a mucoid bacterial phenotype and may form biofilms that are more refractory to treatment (5–7). Appropriate intravenous or inhaled antibiotic therapy for chronic Pa endobronchial infection results in significant clinical improvement, but the reduction in lower airway Pa density is only 2 log10-fold on average and eradication of Pa is rare (8–10).
In contrast to chronic Pa infection, characteristics of early Pa infection are more favorable for eradication with early intervention. Young children with CF and initial infection have reduced Pa density in the lower airways, lower rates of mucoid Pa isolation, and less antibiotic resistance relative to patients with established infection (11–13).
Aggressive antimicrobial treatment of initial Pa infection in children with CF has achieved eradication of Pa or delayed chronic Pa infection of the upper respiratory tract (14–18). Limitations of these prior studies include the lack of controls or use of historical controls, no evaluation of lower airway microbiology or lower airway inflammation, small sample sizes, and lack of focus on safety. Only one published study of early intervention included a control group (14). Only one uncontrolled study evaluated treatment of initial lower airway Pa infection diagnosed by bronchoalveolar lavage (BAL) (18), despite the poor predictive value of upper airway cultures for identifying lower airway pathogens in CF (19–21). The need for controlled studies is highlighted by reports of transient Pa infection in young children with CF (12).
Safety and emergence of resistant pathogens with aggressive early intervention directed against Pa are a concern for young children with CF, and avoidance of oto- and nephrotoxicity is especially important in this age group. Inhaled antipseudomonal antibiotics, in particular aminoglycosides, can deliver high antibiotic concentrations directly to the lower airways with limited systemic absorption (10, 22, 23). Two Phase III clinical trials demonstrated a 1.6 log10-fold reduction in sputum Pa density and improved FEV1 (12% treatment effect compared with placebo) after 28 days of treatment with tobramycin solution for inhalation (TSI) (TOBI; Chiron, Emeryville, CA), 300 mg twice daily (10). However, these trials did not include children less than 6 years of age, who cannot reliably expectorate sputum or perform lung function tests and for whom other established clinical outcome measures are lacking. We reported that a single 300-mg TSI dose administered to children with CF and younger than age 6 years was safe and produced therapeutic lower airway concentrations of tobramycin (24).
The Cystic Fibrosis Therapeutics Development Network conducted a double-blind, placebo-controlled, multicenter, randomized trial to test the hypothesis that administration of TSI (300 mg twice daily for 28 days) to children with CF and less than 6 years of age would result in a significant antimicrobial effect, including possible eradication of Pa from the lower airway, and would be safe. Some of the results of this study have been previously reported in the form of an abstract (25).
The trial was conducted between February 2000 and May 2001 at nine centers. The study was approved by each center's Institutional Review Board. Written informed consent was obtained for enrolled patients.
Inclusion criteria for enrollment were as follows: age, 6 months or more and less than 6 years; diagnosis of CF; and one historical oropharyngeal (OP) culture positive for Pa within 2 weeks to 12 months before screening (see the online supplement for exclusion/withdrawal criteria). Patients with Pa-positive OP cultures at screening were eligible for bronchoscopy and baseline BAL, performed according to Cystic Fibrosis Therapeutics Development Network standard operating procedures (see the online supplement). Baseline evaluation included physical examination, modified Shwachman score (maximum score of 75, no radiograph score), chest radiograph if none in prior 3 months or clinically indicated before BAL, OP culture, and blood draw for complete blood count with differential, serum creatinine, Pa exotoxin A serology, urea, and cytokines.
Patients with a Pa-positive baseline BAL culture (detection of Pa at any density) were randomized (1:1) to receive TSI (300 mg) or placebo twice daily for 28 days at home. The first dose was administered on Day 0 (less than 10 days after BAL). Randomization was stratified by study center and age (36 months or less, or more than 36 months). Other inhaled antibiotics were restricted until study completion.
Study visits on Days 14, 28, 42, and 56 included clinical evaluation and OP culture. On Day 14, blood was obtained for determination of trough and peak tobramycin concentrations, serum creatinine, complete blood count with differential, and serology. On Day 28 (± 2 days), BAL in the same lobar segment plus baseline tests and procedures were repeated (except radiography).
The primary efficacy end point was change in BAL Pa density from baseline to Day 28. Frequency of lower airway Pa eradication, defined as a Pa-negative lobar BAL culture on Day 28, was also examined (limit of detection, 20 colony-forming units [CFU]/ml). Safety end points included adverse events, changes in renal function and hearing acuity, serum tobramycin concentrations, incidence of bronchospasm or acute respiratory distress, and isolation of tobramycin-resistant lower airway Pa.
Screening OP cultures and qualitative BAL cultures were performed at site clinical laboratories for identification of Pa (for more details concerning laboratory studies, see the online supplement). Subsequent OP cultures, quantitative BAL cultures, and minimal inhibitory concentrations (MICs) for Pa isolates were performed by the network's Microbiology Core (Children's Hospital and Regional Medical Center, Seattle, WA) (26). For BAL isolates, antibiotics tested were amikacin, aztreonam, ceftazidime, ciprofloxacin, gentamicin, imipenem, pipericillin, ticarcillin, tobramycin, and trimethoprim. For OP isolates, tobramycin was tested.
Tobramycin concentrations in Day 28 BAL samples were measured by Chiron (Seattle, WA), using high-pressure liquid chromatography (lower limit of quantitation, 0.4 μg/ml) (27). Concentrations of bioactive tobramycin in Day 28 BAL samples were determined by the Microbiology Core via bioassay procedure (24). A subset of baseline BAL samples served as controls.
An in vitro Pa viability assay mimicking BAL shipping conditions was performed by the Microbiology Core to determine whether negative cultures on Day 28 were due to residual tobramycin in BAL fluid. Pa isolates were genotyped by a random amplified polymorphic DNA polymerase chain reaction technique (28, 29).
BAL cell counts were performed at each center. The network's Cytology Core (Case Western Reserve University, Cleveland, OH) performed differential counts on stained slides. BAL and plasma urea and inflammatory mediator assays were performed by the network's Inflammatory Mediator Core (University of Colorado, Denver, CO). Concentrations reported for BAL samples were corrected for dilution, using urea as a marker (30). Serum exotoxin A titers were tested by the Microbiology Core, using an indirect microtiter enzyme-linked immunosorbent assay (11).
Site clinical laboratories performed complete blood counts with differential and serum creatinine. Serum tobramycin concentrations were measured by TDxFLx assay (Abbott Laboratories, Irving, TX; lower limit of quantitation, 0.2 μg/ml) at Children's Hospital and Regional Medical Center (Seattle, WA) (31). An audiologist performed audiometric evaluation on Days 0 and 28 using standard techniques (see the online supplement) (32, 33).
The design specified randomization of 98 patients to detect a difference of 1.6 log10 CFU/ml between treatment groups in mean change in lower airway Pa density (10). One interim analysis with early stopping for futility or efficacy was planned (34). An additional early interim analysis was initiated because of slow accrual to evaluate the primary end point for futility. At the early analysis the Data Monitoring Committee was also provided unblinded efficacy data; statistical ramifications of this unplanned efficacy analysis were addressed (see the online supplement) (35). Treatment groups were compared under an intent-to-treat analysis. Treatment effect, confidence interval (CI), and p value estimates were adjusted for the efficacy stopping rule (36).
Although the planned sample size was 98 patients, the trial was stopped early by the Data Monitoring Committee after interim analysis of data from the first 21 randomized patients showed a statistically significant microbiological treatment effect. This decision also prevented the exposure of additional patients receiving placebo to the risks of bronchoscopy. At the time the trial was stopped, 113 patients had been screened at 9 participating centers. Among screened patients, 37 qualified for baseline bronchoscopy with BAL, of whom 21 were positive for lower airway Pa and were randomized into the study (13 to placebo and 8 to TSI). All randomized patients completed the study.
Of the 21 randomized patients, 6 were younger than 36 months of age. Treatment groups were fairly similar with regard to patient demographic and clinical characteristics at baseline, as shown in Table 1
Placebo Group (n = 13) n (%) | TSI Group (n = 8) n (%) | |
|---|---|---|
| Female sex | 7 (53.8) | 3 (37.5) |
| CF genotype | ||
| ΔF508 homozygous | 7 (53.8) | 4 (50.0) |
| ΔF508 heterozygous | 4 (30.8) | 3 (37.5) |
| Race | ||
| White | 9 (69.2) | 6 (75.0) |
| Hispanic | 2 (15.4) | 0 (00.0) |
| Black | 1 (07.7) | 0 (00.0) |
| Native American | 0 (00.0) | 1 (12.5) |
| Other | 1 (07.7) | 1 (12.5) |
| Mean (SD) | Mean (SD) | |
| Age, yr | 3.7 (1.6) | 4.0 (1.5) |
| Height, cm | 98.6 (13.3) | 99.0 (9.7) |
| Weight, kg | 14.9 (3.6) | 16.0 (3.0) |
| Oxygen saturation, % | 98.2 (1.7) | 98.0 (1.9) |
| Modified Shwachman score | 67.5 (9.0) | 67.4 (3.9) |
| History subscore | 24.1 (1.4) | 24.3 (1.2) |
| Examination subscore | 21.1 (4.6) | 20.9 (2.3) |
| Nutrition subscore | 22.3 (3.7) | 22.3 (1.9) |
| Sweat chloride*, mEq/L | 106.9 (15.6) | 116.8 (14.2) |
Pa was isolated at similar densities from placebo and TSI group patients at the baseline BAL (Table 2)
Treatment Group and Patient No. | Years since First Pa-positive Culture* | Baseline Log10 cfu/ml† | Day 28 Log10 cfu /ml† | 28-Day Reduction in Pa Density‡ |
|---|---|---|---|---|
| TSI group T1 | 0.15 | 7.87 | 0 | 7.87 |
| T2 | 0.31 | 6.38 | 0 | 6.38 |
| T3 | 3.34 | 6.41 | 0 | 6.41 |
| T4 | 0.11 | 1.61 | 0 | 1.61 |
| T5 | 0.82 | 4.66 | 0 | 4.66 |
| T6§ | 1.07 | 8.17 | 0 | 8.17 |
| T7 | 0.06 | 3.31 | 0 | 3.31 |
| T8§ | 0.67 | 3.53 | 0 | 3.53 |
| Group mean (SD) | 0.8 (1.1) | 5.25 (2.34) | 0.00 (−) | 5.25 (2.34) |
| Placebo group P1 | 0.67 | 1.32 | 5.47 | −4.15 |
| P2 | 3.03 | 7.24 | 6.92 | 0.31 |
| P3 | 0.30 | 4.00 | 5.89 | −1.89 |
| P4 | 4.68 | 4.81 | 2.68 | 2.12 |
| P5§ | 0.16 | 6.34 | 4.64 | 1.70 |
| P6§ | 1.20 | 7.58 | 6.94 | 0.64 |
| P7 | 0.58 | 7.07 | 7.23 | −0.17 |
| P8 | 3.94 | 4.00 | 6.30 | −2.30 |
| P9§ | 1.69 | 2.60 | 0 | 2.60 |
| P10 | 1.26 | NA | 6.31 | NA |
| P11§ | 0.34 | 4.60 | 3.62 | 0.99 |
| P12 | 2.42 | 2.78 | 5.99 | −3.21 |
| P13 | 3.62 | 3.82 | 4.08 | −0.25 |
| Group mean (SD) | 1.8 (1.5) | 4.68 (2.01) | 5.08 (2.06) | −0.30 (2.15) |
The time between date of first positive culture for Pa (determined by medical record review) and date of randomization was longer on average in the placebo group (Table 2), and for a number of patients the study setting did not represent initial Pa isolation. Four placebo group patients and five TSI group patients had mucoid Pa isolated from their baseline BAL.
We had predicted that there would be no detectable lower airway tobramycin 8 hours after the last dose of TSI, based on sputum tobramycin pharmacokinetics in patients with CF and who were 6 years of age and older (37–39). After early termination of the trial by the Data Monitoring Committee, tobramycin concentrations from Day 28 BAL samples were measured to assist in interpretation of cultures with no detectable Pa.
Notably, all Day 28 BAL samples from the TSI group had detectable tobramycin concentrations by two different methods (Figure 1)
Figure 1. BAL tobramycin concentrations versus time interval between last dose of study drug and Day 28 BAL. For each patient the Day 28 BAL tobramycin concentration (μg/ml) measured by high-pressure liquid chromatography (circles) or bioassay (triangles) is plotted against the length of time between last dose of study drug and start of the Day 28 BAL procedure. For TSI group patients, chromatography results are denoted by solid circles, and bioassay results are denoted by open triangles. For placebo group patients, tobramycin was undetectable by both methods; results are denoted by open circles.
[More] [Minimize]Four TSI group patients had their Day 28 BAL performed less than the stipulated 12 hours after the last dose of study drug. However, there was no clear relationship between measured BAL tobramycin concentrations and the length of time between last dose of study drug and start of the Day 28 BAL (Figure 1).
We performed an in vitro experiment to determine whether exposure of Pa isolates to tobramycin concentrations present in Day 28 BAL fluid might have affected the viability of Pa from the TSI group patients during overnight shipping on ice. Among TSI group patients, the median time duration from collection of the Day 28 BAL sample to receipt at the Cystic Fibrosis Therapeutics Development Network Microbiology Core was 23 hours (range, 3 to 28 hours). All distinct morphotypes isolated at baseline from the eight TSI group patients were studied; after 24 hours of incubation at 4°C, there was a less than 1 log10-fold change in total Pa density for all eight patients. After 48 hours of incubation, there was a less than l log10-fold reduction in total Pa density for six of the TSI patients, and a maximum 2 log10-fold reduction was observed for the remaining two patients.
We also examined qualitative culture results from aliquots of Day 28 BAL fluid that were processed at the site clinical laboratory. These specimens were not shipped and therefore were exposed only briefly to study drug before processing. No Pa was detected in 7 of 8 TSI group patients compared with 2 of 13 placebo group patients, which is similar to results obtained for the BAL samples shipped to the network's Microbiology Core.
Figure 2. Pseudomonas aeruginosa (Pa) results of serial OP culture results by treatment group: (A) TSI group and (B) placebo group. Open circles denote a Pa-negative OP culture; solid circles denote a Pa-positive OP culture; NA = sample not available. For comparison, concurrent BAL culture results at baseline and Day 28 are indicated adjacent to OP results as Pa positive (+) or Pa negative (−). Patient P10 had a quantitative BAL culture at baseline that was Pa negative at the network's Microbiology Core, but a qualitative culture performed at the site laboratory was Pa positive, thus allowing the patient to be randomized into the study. Any concomitant antipseudomonal antibiotics received during the study are shown (cipro = ciprofloxacin).
[More] [Minimize]Among patients receiving placebo, four had negative OP results at baseline and/or Day 28 BAL that were discordant with their concurrent Pa-positive lower airway culture results. Five placebo group patients received antipseudomonal antibiotics (oral/inhaled) during the study (Figure 2B, top): one patient (receiving ciprofloxacin before screening and throughout the study) had Pa isolated from all available OP cultures; four patients (started on antipseudomonal antibiotics after the Day 28 BAL) had negative OP cultures on Day 42, but cultures on Day 56 were positive in two of three patients tested. Eight of 13 placebo group patients received no antipseudomonal antibiotics between baseline and Day 56 (Figure 2B, bottom); 5 were Pa positive for all OP cultures and 3 were Pa positive for all but one OP culture.
Eighteen of the 21 randomized patients (7 TSI, 11 placebo) had Pa serology data available at baseline and at one or more follow-up visits. Four of seven TSI group patients had positive exotoxin A titers at baseline; titers for all four patients were reduced on Day 14 and/or Day 28, including two who had negative titers on Day 28 (Figure 3A)
Figure 3. Change in Pa exotoxin A serologic response over time by treatment group: (A) TSI group and (B) placebo group. An exotoxin A titer of 1:200 or more was considered to be positive. Patients with a missing follow-up sample are indicated by an asterisk (*). (A) includes baseline BAL Pa densities (shown in parentheses) for TSI group patients.
[More] [Minimize]Genotyping was performed on each distinct Pa morphotype isolated at baseline (OP or BAL) and at subsequent visits (OP only) for the two patients in the TSI group who had recurrence of upper airway Pa at the Day 42 or Day 56 visit (see Figure E1 in the online supplement). The first patient had four BAL isolates and two OP isolates at baseline, with one OP isolate identified on Day 42 and two OP isolates on Day 56; all isolates were the same genotype. The second patient had one BAL isolate and one OP isolate at baseline, with two OP isolates on Day 56; all isolates were the same genotype.
There was not evidence of a TSI treatment effect on ELF inflammatory markers (Table 3
Placebo Group | TSI Group | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | Mean | SD | n | Mean | SD | |||||
| Log10 white cell count, × 106/ml | ||||||||||
| Baseline | 13 | 1.14 | 0.37 | 7 | 1.36 | 0.32 | ||||
| Day 28 | 13 | 1.11 | 0.48 | 8 | 1.29 | 0.32 | ||||
| Day 28 – baseline | 13 | −0.03 | 0.23 | 7 | −0.06 | 0.23 | ||||
| Log10 neutrophils, × 106/ml | ||||||||||
| Baseline | 13 | 0.71 | 0.49 | 6 | 0.80 | 0.41 | ||||
| Day 28 | 13 | 0.74 | 0.65 | 7 | 0.77 | 0.37 | ||||
| Day 28 – baseline | 13 | 0.03 | 0.42 | 6 | 0.01 | 0.38 | ||||
| Neutrophils, % | ||||||||||
| Baseline | 13 | 42.7 | 20.9 | 7 | 40.4 | 19.3 | ||||
| Day 28 | 13 | 51.8 | 25.1 | 7 | 35.8 | 10.8 | ||||
| Day 28 – baseline | 13 | 9.2 | 24.5 | 7 | −4.6 | 25.2 | ||||
| Log10 IL-8, pg/ml | ||||||||||
| Baseline | 13 | 4.33 | 0.40 | 8 | 4.38 | 0.47 | ||||
| Day 28 | 13 | 4.39 | 0.72 | 8 | 4.24 | 0.46 | ||||
| Day 28 – baseline | 13 | 0.06 | 0.64 | 8 | −0.15 | 0.36 | ||||
| Log10 IL-6, pg/ml | ||||||||||
| Baseline | 13 | 2.54 | 0.39 | 8 | 2.54 | 0.45 | ||||
| Day 28 | 13 | 2.63 | 0.49 | 8 | 2.77 | 0.47 | ||||
| Day 28 – baseline | 13 | 0.09 | 0.31 | 8 | 0.23 | 0.57 | ||||
The 28-day change in peripheral white cell count was –1.7 × 106/ml (SD 1.5) in the TSI group and 0.3 × 106/ml (SD 3.3) in the placebo group, for a mean difference between groups of –2.0 × 106/ml (95% CI, –4.19 to 0.25; p = 0.08). For peripheral neutrophil count, the 28-day change was –1. 4 × 106/ml (SD 1.7) in the TSI group and 0.5 × 106/ml (SD 3.0) in the placebo group for a mean difference of –1.9 × 106/ml (95% CI, –4.09 to 0.30; p = 0.09). Sixty-four percent of the values of plasma interleukin-6 were below the lower limit of detection, and therefore descriptive statistics are not provided.
TSI and placebo groups were similar with regard to change in oxygen saturation, change in weight, and change in total modified Shwachman scores between baseline and Day 56 (see Table E2 in the online supplement). The 56-day difference in the modified Shwachman physical examination subscore was in the direction of treatment benefit (TSI, 2.1 [SD 2.1]; placebo, –1.1 [SD 3.6]), but this finding would not be significant if corrected for multiple comparisons (see Table E3 in the online supplement).
Safety end points included adverse events, incidence of bronchospasm or acute respiratory distress, serum tobramycin concentrations, changes in renal function and hearing acuity, rate of isolation of tobramycin-resistant lower airway Pa, and emergence of new CF pathogens.
There were 107 treatment emergent adverse events, 72 among the 13 placebo group patients, and 35 among the 8 TSI group patients. Treatment emergent adverse events considered to be related to study drug were reported for four patients in the placebo group (six events) and two patients in the TSI group (five events). The rate of occurrence of specific adverse events was similar between treatment groups. Cough was the most common adverse event in both groups, affecting 92% of placebo patients and 88% of TSI patients. There were no episodes of bronchospasm related to study drug.
Two patients had a serious adverse event related to bronchoscopy. One patient had transient vomiting and unsteady gait after lorazepam sedation that required hospitalization and overnight observation. A second patient had an acute episode of laryngospasm and hypoxemia as the bronchoscope was introduced into the airway and required intubation; a chest radiograph showed acute bilateral upper lobe atelectasis. The patient stabilized rapidly but was kept on minimal ventilatory support overnight as a precaution. The patient, who was not randomized to study drug, completed a 14-day course of intravenous antipseudomonal antibiotics and was stable on discharge from the hospital. For the remaining subjects, the most common adverse events related to bronchoscopy were transient cough and fever of mild to moderate intensity. Adverse events reported as definitely or probably related to bronchoscopy occurred in 8 of 13 placebo group patients and in 5 of 8 TSI group patients.
The Day 14 peak (1 hour) serum tobramycin concentration for the TSI group was 1.0 ± 0.4 μg/ml and the trough concentration was 0.4 ± 0.5 (mean ± SD). There was no detectable serum tobramycin among placebo group patients. Serum creatinine levels were within the normal range for both groups at all evaluations (data not shown). There were no changes in auditory threshold in the TSI group.
The distribution of tobramycin MIC concentrations for OP Pa strains was determined at each study visit (data not shown). For placebo patients, MICs for OP Pa isolates were fairly stable over time from the baseline visit through Day 56 (majority less than 2 μg/ml, all less than or equal to 4 μg/ml). For the TSI group, all but one Pa isolate had an MIC ⩽ 2 μg/ml at baseline. For the two TSI group patients with recurrence of positive OP cultures, the Pa isolates had tobramycin MICs ⩽ 0.5 μg/ml.
Tobramycin MICs for lower airway Pa isolates from the baseline BAL were comparable in the two treatment groups, and tobramycin MICs for concurrent OP and BAL Pa isolates were similar (data not shown). There were no tobramycin-resistant isolates among TSI group patients at baseline. The placebo group had similar tobramycin MICs for isolates from baseline and Day 28 BAL cultures (data not shown). One placebo patient had a tobramycin-resistant Pa isolate (MIC = 16 μg/ml) at baseline BAL; on Day 28 that individual had a lower airway isolate with MIC ⩽ 0.5 μg/ml.
Quantitative BAL cultures on Day 28 showed no evidence of emergence of new CF pathogens in either treatment group. Five of eight TSI group patients had coinfection with Staphylococcus aureus at baseline; three of these patients had no detectable S. aureus on Day 28. No patients in the TSI group had any gram-negative organisms on Day 28. Five of 13 placebo group patients had coinfection with S. aureus or Haemophilus influenzae at baseline and Day 28.
We conclude that 28 days of TSI treatment (300 mg twice daily) compared with placebo results in a markedly greater reduction in Pa density in the lower airways and a higher frequency of eradication (as defined by no detectable Pa in culture). The observed antimicrobial treatment effect is more profound than that reported in older, chronically infected patients (10). This is the first placebo-controlled trial showing Pa eradication from the lower airways in patients with CF. This dramatic effect may be due to characteristics of the Pa colonizing younger patients (i.e., more susceptible to tobramycin, less mucoid), lower Pa densities, or slower tobramycin clearance from the lower airways of younger patients.
Interpretation of these results is complicated by the unexpected finding of therapeutic tobramycin concentrations in the Day 28 BAL fluid. However, substantial data corroborate the validity of our primary efficacy end point. First, the bactericidal activity of aminoglycosides depends on active transport (40) and cellular division (41), which are both markedly inhibited at 0–4°C. Our in vitro studies demonstrated that the concentration of tobramycin in the Day 28 BAL fluid did not eradicate any patients' isolates, with Pa densities reduced 2 log10-fold or less under the conditions of specimen shipping. Second, the Day 28 BAL cultures from local site laboratories, with timely dilution and plating, yielded nearly identical results for the frequency of eradication in the two groups. Third, we observed sustained eradication of Pa from OP cultures through Day 56 in six of eight patients in the TSI group.
No significant safety concerns were observed for TSI in these young children with CF based on adverse events, serum tobramycin concentrations, renal function, audiology testing, tobramycin MICs, and lack of emergence of new pathogens. Conclusions related to safety are limited by the small sample size and potentially insensitive measures of renal toxicity (42).
Interpretation of secondary end points was made difficult because of the small sample size resulting from the study being stopped prematurely by the Data Monitoring Committee. The young children enrolled in this study had marked lower airway inflammation at baseline, consistent with prior studies of young, infected patients with CF (12, 13, 43, 44) and stable pediatric and adult patients with mild lung disease (45, 46). Factors that may have contributed to the lack of observed antiinflammatory effects of TSI include the degree and duration of Pa infection at baseline, the short treatment duration and follow-up, and in-fection undetected by lobar BAL (47). Alternatively, the robust inflammatory response in young patients infected with Pa may be, at least in part, intrinsic to the defect in CF (48). The preliminary observation of a greater rate of conversion to negative exotoxin A titers by Day 28 in the TSI group compared with the placebo group requires further study. The 28-day change in peripheral white cell count and neutrophil count, and the 56-day change in the modified Shwachman physical examination subscore were in the direction of a treatment benefit; however, our precision was too low to be highly confident that similar results would be obtained with a larger study. The lack of observed effects for other clinical parameters may have been attributable in part to the insensitivity of the chosen outcome measures.
In patients with CF who are older than 6 years of age, peak sputum tobramycin concentrations occur 10 to 60 minutes after tobramycin inhalation (37, 38) with an estimated half-life in sputum of 2 hours or less (37–39). In contrast, we observed that ELF tobramycin concentrations 4 to 24 hours after the last dose of TSI had no relationship to the interval since last dose, and were comparable to those observed 45 minutes after inhalation of a single dose (24). Possible explanations for the apparent delayed tobramycin clearance from ELF in young children with CF include (1) reduced cough clearance; (2) increased distal deposition and binding to macromolecules (49, 50), possibly due to reduced amounts of central airway purulent secretions; (3) drug accumulation in the lower airways after chronic exposure compared with single dose exposure (37–39); and (4) sampling of distal lower airway secretions rather than sputum. If confirmed, this observation may influence dosing interval for TSI treatment in young children with CF.
Prior studies treating initial Pa infection in CF have varied widely in the age of the patient population, the site of respiratory tract cultures (OP, endolaryngeal, sputum, and BAL), the presence of serum antibodies against Pa at enrollment, the duration of Pa infection before treatment, the treatment regimen, and the study end points (14–18). Previous intervention trials for early Pa infection, although limited by lack of controls, lack of safety data, and focus on upper airway cultures, have consistently demonstrated a microbiological effect. The observed differences in the duration of eradication are likely dependent on patient selection criteria, treatment regimens, and outcome measures. Munck and coworkers treated first Pa (nonmucoid) colonization in young children with intravenous antibiotics for 21 days and inhaled colistin for 2 months and reported transient eradication of Pa for a mean duration of 8 months (17). The majority of these patients were recolonized with a new Pa genotype. We observed eradication of both mucoid and nonmucoid Pa from both BAL and OP cultures at the end of TSI treatment in all eight patients, with recurrent Pa of the same genotype in OP cultures from two patients by Day 56. Potential sources include the sinuses, undetected residual lower airway infection, or reinfection with the same environmental isolate. We have no data on the time to recurrence of Pa infection in the lower airways.
In summary, the accumulating data suggest that early antipseudomonal therapy can eradicate both upper and lower airway Pa colonization. Unfortunately, compelling evidence of a clinical benefit from early intervention is lacking (15, 18, 51). Often these young patients with early Pa infection have minimal symptoms with normal lung function. (12, 16) This highlights the need for better clinical outcome measures for young children, such as raised volume chest tomography to assess early bronchiectasis (52), or raised volume infant lung function tests to detect air trapping and reduced small airway flows (53, 54). Future early intervention trials should focus on safety, determine the minimal amount of antipseudomonal therapy required for sustained Pa eradication, and characterize the host and environmental factors that result in recurrence of Pa infection. Ultimately, we recommend a placebo-controlled trial, with a primary clinical end point, for treatment of first isolation of P. aeruginosa.
The authors thank the patients and families who participated in the clinical trial, and they wish to recognize the efforts of the other members of the Cystic Fibrosis Therapeutics Development Network Study Group (Morty Cohen, Jessica Foster, Charlene Hallmark, Trish Hasbrouck, Jay Hilliard, Kate Hilliard, Lori Ingham, Craig Johnson, Vikki Kociela, Richard Kronmal, Sally Locke, Jean Mundahl, Iris Osberg, Churee Penvari, Jenny Stapp, Sharon Watts, Marcia Wertz, and Judy Williams). A special thanks to Robert Beall, Preston Campbell, and Judy Vaitukaitis for their support and dedication to the Cystic Fibrosis Therapeutics Development Network.
| 1. | Welsh MJ, Ramsey BW, Accurso FJ, Cutting GR. Cystic fibrosis. In: Scriver CR, Beaudet AL, Sly WS, Valle D, editors. The metabolic basis of inherited disease, 8th ed. New York: McGraw-Hill; 2001. p. 521–588. |
| 2. | Kosorok MR, Zeng L, West SHE, Rock MJ, Splaingard ML, Laxova A, Green CG, Collins J, Farrell PM. Acceleration of lung disease in children with cystic fibrosis after Pseudomonas aeruginosa acquisition. Pediatr Pulmonol 2001;32:277–287. |
| 3. | Emerson J, Rosenfeld M, McNamara S, Ramsey B, Gibson RL. Pseudomonas aeruginosa and other predictors of mortality and morbidity in young children with cystic fibrosis. Pediatr Pulmonol 2002;34:91–100. |
| 4. | Cystic Fibrosis Foundation. Patient Registry 2000 [annual report]. Bethesda, MD: Cystic Fibrosis Foundation; 2001. |
| 5. | Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: a common cause of persistent infections. Science 1999;284:1318–1322. |
| 6. | Singh PK, Shaefer AL, Parsek MR, Moninger TO, Welsh MJ, Greenberg EP. Quorum-sensing signals indicate that cystic fibrosis lungs are infected with bacterial biofilms. Nature 2000;407:762–764. |
| 7. | Drenkard E, Ausubel FM. Pseudomonas biofilm formation and antibiotic resistance are linked to phenotypic variation. Nature 2002;416:695–696. |
| 8. | Smith AL, Redding G, Doershuk C, Goldmann D, Gore E, Hilman B, Marks M, Moss R, Ramsey B, Rubio T, et al. Sputum changes associated with therapy for endobronchial exacerbation in cystic fibrosis. J Pediatr 1988;112:547–554. |
| 9. | Regelmann WE, Elliott GR, Warwick WJ, Clawson CC. Reduction of sputum Pseudomonas aeruginosa density by antibiotics improves lung function in cystic fibrosis more than do bronchodilators and chest physiotherapy alone. Am Rev Respir Dis 1990;141:914–921. |
| 10. | Ramsey BW, Pepe MS, Quan JM, Otto KL, Montgomery AB, Williams-Warren J, Vasiljev-K M, Borowitz D, Bowman CM, Marshall BC, Marshall S, Smith AL. Intermittent administration of inhaled tobramycin in patients with cystic fibrosis. N Engl J Med 1999;340:23–30. |
| 11. | Burns JL, Gibson RL, McNamara S, Yim D, Emerson J, Rosenfeld M, Hiatt P, McCoy K, Castile R, Smith AL, et al. Longitudinal assessment of Pseudomonas aeruginosa in infants with cystic fibrosis. J Infect Dis 2001;183:444–452. |
| 12. | Rosenfeld M, Gibson RL, McNamara S, Emerson J, Burns JL, Castile R, Hiatt P, McCoy K, Wilson CB, Inglis A, et al. Early pulmonary infection, inflammation, and clinical outcomes in infants with cystic fibrosis. Pediatr Pulmonol 2001;32:356–366. |
| 13. | Armstrong DS, Grimwood K, Carlin JB, Carzino R, Gutièrrez JP, Hull J, Olinsky A, Phelan EM, Robertson CF, Phelan PD. Lower airway inflammation in infants and young children with cystic fibrosis. Am J Respir Crit Care Med 1997;156:1197–1204. |
| 14. | Wiesemann HG, Steinkamp G, Ratjen F, Bauernfeind A, Przyklenk B, Döoring G, von der Hardt H. Placebo-controlled, double-blind, randomized study of aerosolized tobramycin for early treatment of Pseudomonas aeruginosa colonization in cystic fibrosis. Pediatr Pulmonol 1998;25:88–92. |
| 15. | Frederickson B, Koch C, Hoiby N. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol 1997;23:330–335. |
| 16. | Ratjen F, Doring G, Nikolaizik WH. Effect of inhaled tobramycin on early Pseudomonas aeruginosa colonisation in patients with cystic fibrosis. Lancet 2001;358:983–984. |
| 17. | Munck A, Bonacorsi S, Mariani-Kurkdjian P, Lebourgeois M, Gérardin M, Brahimi N, Navarro J, Bingen E. Genotypic characterization of Pseudomonas aeruginosa strains recovered from patients with cystic fibrosis after initial and subsequent colonization. Pediatr Pulmonol 2001;32:288–292. |
| 18. | Nixon GM, Armstrong DS, Carzino R, Carlin JB, Olinsky A, Robertson CF, Grimwood K. Clinical outcome after early Pseudomonas aeruginosa infection in cystic fibrosis. J Pediatr 2001;138:699–704. |
| 19. | Ramsey BW, Wentz KR, Smith AL, Richardson M, Williams-Warren J, Hedges DL, Gibson R, Redding G, Lent K, Harris K. Predictive value of oropharyngeal cultures for identifying lower airway bacteria in cystic fibrosis patients. Am Rev Respir Dis 1991;144:331–337. |
| 20. | Armstrong DS, Grimwood K, Carlin JB, Carzino R, Olinsky A, Phelan PD. Bronchoalveolar lavage or oropharyngeal cultures to identify lower respiratory pathogens in infants with cystic fibrosis. Pediatr Pulmonol 1996;21:267–275. |
| 21. | Rosenfeld M, Emerson J, Accurso F, Armstrong D, Castile R, Grimwood K, Hiatt P, McCoy K, McNamara S, Ramsey B, et al. Diagnostic accuracy of oropharyngeal cultures in infants and young children with cystic fibrosis. Pediatr Pulmonol 1999;28:321–328. |
| 22. | Ramsey BW, Dorkin HL, Eisenberg JD, Gibson RL, Harwood IR, Kravitz RM, Schidlow DV, Wilmott RW, Astley SJ, McBurnie MA, et al. Efficacy of aerosolized tobramycin in patients with cystic fibrosis. N Engl J Med 1993;328:1740–1746. |
| 23. | Rosenfeld M, Cohen M, Ramsey B. Aerosolized antibiotics for bacterial lower airway infections: principles, efficacy and pitfalls. Clin Pulm Med 1997;4:101–112. |
| 24. | Rosenfeld M, Gibson R, McNamara S, Emerson J, McCoy K, Schell R, Borowitz D, Konstan M, Retsch-Bogart G, Wilmott RW, et al. Serum and lower respiratory tract drug concentrations produced by tobramycin for inhalation in young children with cystic fibrosis. J Pediatr 2001;139:572–577. |
| 25. | Gibson RL, Emerson J, Mcnamara S, Burns JL, Rosenfeld M, Yunker A, Hamblett N, Accurso F, Dovey M, Hiatt P, Moss R, Konstan M, Retsch-Bogart G, Waltz D, Wilmott R, Wagener J, Zeitlin P, Ramsey B. A randomized controlled trial of inhaled tobramycin in young children with cystic fibrosis: eradication of Pseudomonas from the lower airway [abstract 356]. Pediatr Pulmonol 2002(Suppl 24). |
| 26. | Burns JL, Emerson J, Stapp JR, Yim DL, Krzewinski J, Louden L, Ramsey BR, Clausen CR. Microbiology of sputum from patients at cystic fibrosis centers in the United States. Clin Infect Dis 1998;27:158–163. |
| 27. | Barends DM, Zwaan CL, Hulshoff A. Micro-determination of tobramycin in serum by high-performance liquid chromatograph with ultraviolet detection. J Chromatogr 1981;225:417–426. |
| 28. | Mahenthiralingam E, Campbell ME, Foster J, Lam JS, Speert DP. Random amplified polymorphic DNA typing of Pseudomonas aeruginosa isolates recovered from patients with cystic fibrosis. J Clin Microbiol 1996;34:1129–1135. |
| 29. | Krzewinski J, Nguyen CD, Foster JM, Burns JL. Use of random amplified polymorphic DNA PCR to examine epidemiology of Stenotrophomonas maltophilia and Achromobacter (Alcaligenes) xylosoxidans from patients with cystic fibrosis. J Clin Microbiol 2001;39:3597–3602. |
| 30. | Rennard SI, Basset G, Lecossier D, O'Donnell KM, Pinkston P, Martin PG, Crystal RG. Estimation of volume of epithelial lining fluid recovered by lavage using urea as marker of dilution. J Appl Physiol 1986; 60:532–538. |
| 31. | Jolley ME, Stroupe SD, Wang CJ. Fluorescence polarization immunoassay. I. Monitoring aminoglycoside antibiotics in serum and plasma. Clin Chem 1981;27:1190–1197. |
| 32. | Thompson G, Wilson WR. Clinical application of visual reinforcement audiometry. Semin Hearing 1984;5:85–99. |
| 33. | Gravel JS. Behavioral assessment of auditory function. Semin Hearing 1989;10:216–228. |
| 34. | O'Brien PC, Fleming TR. A multiple testing procedure for clinical trials. Biometrics 1979;35:549–556. |
| 35. | Emerson SS. Stopping a clinical trial very early based on unplanned interim analyses: a group sequential approach. Biometrics 1995;51: 1152–1162. |
| 36. | Emerson SS, Fleming TR. Parameter estimation following group sequential hypothesis testing. Biometrika 1990;77:875–892. |
| 37. | Eisenberg J, Pepe M, Williams-Warren J, Vasiliev M, Montgomery AB, Smith AL, Ramsey BW. A comparison of peak sputum tobramycin concentration in patients with cystic fibrosis using jet and ultrasonic nebulizer systems. Chest 1997;111:955–962. |
| 38. | Mukhopadhyay S, Staddon GE, Eastman C, Palmer M, Davies ER, Carswell F. The quantitative distribution of nebulized antibiotic in the lung in cystic fibrosis. Respir Med 1994;88:203–211. |
| 39. | Geller DE, Rosenfeld M, Waltz D, Wilmott R. Efficiency of pulmonary administration of tobramycin solution for inhalation in cystic fibrosis using an improved drug delivery system. Chest 2003;123:28–36. |
| 40. | Bryan LE. Aminoglycoside resistance. In: Bryan LE, editor. Antimicrobial drug resistance. Orlando, FL: Academic Press; 1984. p. 241–251. |
| 41. | Evans DJ, Brown MR, Allison DG, Gilbert P. Susceptibility of bacterial biofilms to tobramycin: role of specific growth rate and phase in the division cycle. J Antimicrob Chemother 1990;25:585–591. |
| 42. | Ring E, Eber E, Erwa W, Zach MS. Urinary N-acetyl-β-d-glucosaminidase activity in patients with cystic fibrosis on long term gentamicin inhalation. Arch Dis Child 1998;78:540–543. |
| 43. | Noah TL, Black HR, Cheng P-W, Wood RE, Leigh MW. Nasal and bronchoalveolar lavage fluid cytokines in early cystic fibrosis. J Infect Dis 1997;175:638–647. |
| 44. | Muhlebach MS, Stewart PW, Leigh MW, Noah TL. Quantitation of inflammatory responses to bacteria in young cystic fibrosis and control patients. Am J Respir Crit Care Med 1999;160:186–191. |
| 45. | Konstan MW, Hilliard KA, Norvell TM, Berger M. Bronchoalveolar lavage findings in cystic fibrosis patients with stable, clinically mild lung disease suggest ongoing infection and inflammation. Am J Respir Crit Care Med 1994;150:448–454. |
| 46. | Sagel SD, Kapsner R, Osberg I, Sontag MK, Accurso FJ. Airway inflammation in children with cystic fibrosis and healthy children assessed by sputum induction. Am J Respir Crit Care Med 2001;164:1425–1431. |
| 47. | Meyer KC, Sharma A. Regional variability of lung inflammation in cystic fibrosis. Am J Respir Crit Care Med 1997;156:1536–1540. |
| 48. | Khan TZ, Wagener JS, Bost T, Martinez J, Accurso FJ, Riches DWH. Early pulmonary inflammation in infants with cystic fibrosis. Am J Respir Crit Care Med 1995;151:1075–1082. |
| 49. | Mendelman PM, Smith AL, Levy J, Weber A, Ramsey B, Davis RL. Aminoglycoside penetration, inactivation, and efficacy in cystic fibrosis sputum. Am Rev Respir Dis 1995;132:761–765. |
| 50. | Ramphal R, Lhermitte M, Filliat M, Roussel P. The binding of anti-pseudomonal antibiotics to macromolecules from cystic fibrosis sputum. J Antimicrob Chemother 1988;22:483–490. |
| 51. | Pamukcu A, Bush A, Buchdahl R. Effects of Pseudomonas aeruginosa colonization on lung function and anthropometric variables in children with cystic fibrosis. Pediatr Pulmonol 1995;19:10–15. |
| 52. | Long FF, Castile RG, Brody AS, Hogan MJ, Flucke RL, Filbrun DA, McCoy KS. Lungs in infants and young children: improved thin section CT with noninvasive controlled-ventilation technique—initial experience. Radiology 1999;212:588–593. |
| 53. | Castile RG, Filbrun D, Flucke R, Frankline W, McCoy K. Adult type pulmonary function tests in infants without respiratory diseases. Pediatr Pulmonol 2000;30:215–227. |
| 54. | Turner DJ, Lanteri CJ, LeSouef PN, Sly PD. Improved detection of abnormal respiratory function using forced expiration from raised lung volume in infants with cystic fibrosis. Eur Respir J 1994;7:1995–1999. |
* Members of the Cystic Fibrosis Therapeutics Development Network Study Group: Morty Cohen, Jessica Foster, Charlene Hallmark, Trish Hasbrouck, Jay Hilliard, Kate Hilliard, Lori Ingham, Craig Johnson, Vikki Kociela, Richard Kronmal, Sally Locke, Jean Mundahl, Iris Osberg, Churee Penvari, Jenny Stapp, Sharon Watts, Marcia Wertz, and Judy Williams.
